From e54237053204caebacb3eda3d37363f1a6118707 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 14 Nov 2008 10:38:39 +1100 Subject: CRED: Wrap task credential accesses in the S390 arch Wrap access to task credentials so that they can be separated more easily from the task_struct during the introduction of COW creds. Change most current->(|e|s|fs)[ug]id to current_(|e|s|fs)[ug]id(). Change some task->e?[ug]id to task_e?[ug]id(). In some places it makes more sense to use RCU directly rather than a convenient wrapper; these will be addressed by later patches. Signed-off-by: David Howells Reviewed-by: James Morris Acked-by: Serge Hallyn Cc: Martin Schwidefsky Cc: Heiko Carstens Cc: linux-s390@vger.kernel.org Signed-off-by: James Morris --- arch/s390/hypfs/inode.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c index 36313801cd5..8aadcd7a7cf 100644 --- a/arch/s390/hypfs/inode.c +++ b/arch/s390/hypfs/inode.c @@ -280,8 +280,8 @@ static int hypfs_fill_super(struct super_block *sb, void *data, int silent) if (!sbi) return -ENOMEM; mutex_init(&sbi->lock); - sbi->uid = current->uid; - sbi->gid = current->gid; + sbi->uid = current_uid(); + sbi->gid = current_gid(); sb->s_fs_info = sbi; sb->s_blocksize = PAGE_CACHE_SIZE; sb->s_blocksize_bits = PAGE_CACHE_SHIFT; -- cgit v1.2.3-70-g09d2 From b6dff3ec5e116e3af6f537d4caedcad6b9e5082a Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 14 Nov 2008 10:39:16 +1100 Subject: CRED: Separate task security context from task_struct Separate the task security context from task_struct. At this point, the security data is temporarily embedded in the task_struct with two pointers pointing to it. Note that the Alpha arch is altered as it refers to (E)UID and (E)GID in entry.S via asm-offsets. With comment fixes Signed-off-by: Marc Dionne Signed-off-by: David Howells Acked-by: James Morris Acked-by: Serge Hallyn Signed-off-by: James Morris --- arch/alpha/kernel/asm-offsets.c | 11 +- arch/alpha/kernel/entry.S | 10 +- arch/ia64/ia32/sys_ia32.c | 8 +- arch/mips/kernel/kspd.c | 4 +- arch/s390/kernel/compat_linux.c | 28 ++--- drivers/connector/cn_proc.c | 8 +- fs/binfmt_elf.c | 12 +- fs/binfmt_elf_fdpic.c | 12 +- fs/exec.c | 4 +- fs/fcntl.c | 4 +- fs/file_table.c | 4 +- fs/fuse/dir.c | 12 +- fs/hugetlbfs/inode.c | 4 +- fs/ioprio.c | 12 +- fs/nfsd/auth.c | 22 ++-- fs/nfsd/nfs4recover.c | 12 +- fs/nfsd/nfsfh.c | 6 +- fs/open.c | 17 +-- fs/proc/array.c | 18 +-- fs/proc/base.c | 16 +-- fs/xfs/linux-2.6/xfs_cred.h | 6 +- fs/xfs/linux-2.6/xfs_globals.h | 2 +- fs/xfs/xfs_inode.h | 2 +- fs/xfs/xfs_vnodeops.h | 10 +- include/linux/cred.h | 155 +++++++++++++++++++---- include/linux/init_task.h | 24 ++-- include/linux/sched.h | 52 +------- include/linux/securebits.h | 2 +- ipc/mqueue.c | 2 +- ipc/shm.c | 4 +- kernel/auditsc.c | 52 ++++---- kernel/capability.c | 4 +- kernel/cgroup.c | 4 +- kernel/exit.c | 10 +- kernel/fork.c | 24 ++-- kernel/futex.c | 6 +- kernel/futex_compat.c | 5 +- kernel/ptrace.c | 19 +-- kernel/sched.c | 10 +- kernel/signal.c | 16 +-- kernel/sys.c | 266 ++++++++++++++++++++++----------------- kernel/trace/trace.c | 2 +- kernel/tsacct.c | 4 +- kernel/uid16.c | 28 ++--- kernel/user.c | 4 +- mm/mempolicy.c | 10 +- mm/migrate.c | 10 +- mm/oom_kill.c | 2 +- net/core/scm.c | 10 +- net/sunrpc/auth.c | 2 +- security/commoncap.c | 161 +++++++++++++----------- security/keys/keyctl.c | 25 ++-- security/keys/permission.c | 11 +- security/keys/process_keys.c | 98 ++++++++------- security/keys/request_key.c | 18 +-- security/keys/request_key_auth.c | 12 +- security/selinux/exports.c | 2 +- security/selinux/hooks.c | 116 ++++++++--------- security/selinux/selinuxfs.c | 2 +- security/selinux/xfrm.c | 6 +- security/smack/smack_access.c | 4 +- security/smack/smack_lsm.c | 77 ++++++------ security/smack/smackfs.c | 6 +- 63 files changed, 832 insertions(+), 677 deletions(-) (limited to 'arch/s390') diff --git a/arch/alpha/kernel/asm-offsets.c b/arch/alpha/kernel/asm-offsets.c index 4b18cd94d59..6ff8886e7e2 100644 --- a/arch/alpha/kernel/asm-offsets.c +++ b/arch/alpha/kernel/asm-offsets.c @@ -19,15 +19,18 @@ void foo(void) BLANK(); DEFINE(TASK_BLOCKED, offsetof(struct task_struct, blocked)); - DEFINE(TASK_UID, offsetof(struct task_struct, uid)); - DEFINE(TASK_EUID, offsetof(struct task_struct, euid)); - DEFINE(TASK_GID, offsetof(struct task_struct, gid)); - DEFINE(TASK_EGID, offsetof(struct task_struct, egid)); + DEFINE(TASK_CRED, offsetof(struct task_struct, cred)); DEFINE(TASK_REAL_PARENT, offsetof(struct task_struct, real_parent)); DEFINE(TASK_GROUP_LEADER, offsetof(struct task_struct, group_leader)); DEFINE(TASK_TGID, offsetof(struct task_struct, tgid)); BLANK(); + DEFINE(CRED_UID, offsetof(struct cred, uid)); + DEFINE(CRED_EUID, offsetof(struct cred, euid)); + DEFINE(CRED_GID, offsetof(struct cred, gid)); + DEFINE(CRED_EGID, offsetof(struct cred, egid)); + BLANK(); + DEFINE(SIZEOF_PT_REGS, sizeof(struct pt_regs)); DEFINE(PT_PTRACED, PT_PTRACED); DEFINE(CLONE_VM, CLONE_VM); diff --git a/arch/alpha/kernel/entry.S b/arch/alpha/kernel/entry.S index 5fc61e281ac..f77345bc66a 100644 --- a/arch/alpha/kernel/entry.S +++ b/arch/alpha/kernel/entry.S @@ -850,8 +850,9 @@ osf_getpriority: sys_getxuid: .prologue 0 ldq $2, TI_TASK($8) - ldl $0, TASK_UID($2) - ldl $1, TASK_EUID($2) + ldq $3, TASK_CRED($2) + ldl $0, CRED_UID($3) + ldl $1, CRED_EUID($3) stq $1, 80($sp) ret .end sys_getxuid @@ -862,8 +863,9 @@ sys_getxuid: sys_getxgid: .prologue 0 ldq $2, TI_TASK($8) - ldl $0, TASK_GID($2) - ldl $1, TASK_EGID($2) + ldq $3, TASK_CRED($2) + ldl $0, CRED_GID($3) + ldl $1, CRED_EGID($3) stq $1, 80($sp) ret .end sys_getxgid diff --git a/arch/ia64/ia32/sys_ia32.c b/arch/ia64/ia32/sys_ia32.c index 5e92ae00bdb..2445a9d3488 100644 --- a/arch/ia64/ia32/sys_ia32.c +++ b/arch/ia64/ia32/sys_ia32.c @@ -1772,20 +1772,20 @@ sys32_getgroups16 (int gidsetsize, short __user *grouplist) if (gidsetsize < 0) return -EINVAL; - get_group_info(current->group_info); - i = current->group_info->ngroups; + get_group_info(current->cred->group_info); + i = current->cred->group_info->ngroups; if (gidsetsize) { if (i > gidsetsize) { i = -EINVAL; goto out; } - if (groups16_to_user(grouplist, current->group_info)) { + if (groups16_to_user(grouplist, current->cred->group_info)) { i = -EFAULT; goto out; } } out: - put_group_info(current->group_info); + put_group_info(current->cred->group_info); return i; } diff --git a/arch/mips/kernel/kspd.c b/arch/mips/kernel/kspd.c index b0591ae0ce5..fd6e5122403 100644 --- a/arch/mips/kernel/kspd.c +++ b/arch/mips/kernel/kspd.c @@ -174,8 +174,8 @@ static unsigned int translate_open_flags(int flags) static void sp_setfsuidgid( uid_t uid, gid_t gid) { - current->fsuid = uid; - current->fsgid = gid; + current->cred->fsuid = uid; + current->cred->fsgid = gid; key_fsuid_changed(current); key_fsgid_changed(current); diff --git a/arch/s390/kernel/compat_linux.c b/arch/s390/kernel/compat_linux.c index 4646382af34..6cc87d8c868 100644 --- a/arch/s390/kernel/compat_linux.c +++ b/arch/s390/kernel/compat_linux.c @@ -148,9 +148,9 @@ asmlinkage long sys32_getresuid16(u16 __user *ruid, u16 __user *euid, u16 __user { int retval; - if (!(retval = put_user(high2lowuid(current->uid), ruid)) && - !(retval = put_user(high2lowuid(current->euid), euid))) - retval = put_user(high2lowuid(current->suid), suid); + if (!(retval = put_user(high2lowuid(current->cred->uid), ruid)) && + !(retval = put_user(high2lowuid(current->cred->euid), euid))) + retval = put_user(high2lowuid(current->cred->suid), suid); return retval; } @@ -165,9 +165,9 @@ asmlinkage long sys32_getresgid16(u16 __user *rgid, u16 __user *egid, u16 __user { int retval; - if (!(retval = put_user(high2lowgid(current->gid), rgid)) && - !(retval = put_user(high2lowgid(current->egid), egid))) - retval = put_user(high2lowgid(current->sgid), sgid); + if (!(retval = put_user(high2lowgid(current->cred->gid), rgid)) && + !(retval = put_user(high2lowgid(current->cred->egid), egid))) + retval = put_user(high2lowgid(current->cred->sgid), sgid); return retval; } @@ -217,20 +217,20 @@ asmlinkage long sys32_getgroups16(int gidsetsize, u16 __user *grouplist) if (gidsetsize < 0) return -EINVAL; - get_group_info(current->group_info); - i = current->group_info->ngroups; + get_group_info(current->cred->group_info); + i = current->cred->group_info->ngroups; if (gidsetsize) { if (i > gidsetsize) { i = -EINVAL; goto out; } - if (groups16_to_user(grouplist, current->group_info)) { + if (groups16_to_user(grouplist, current->cred->group_info)) { i = -EFAULT; goto out; } } out: - put_group_info(current->group_info); + put_group_info(current->cred->group_info); return i; } @@ -261,22 +261,22 @@ asmlinkage long sys32_setgroups16(int gidsetsize, u16 __user *grouplist) asmlinkage long sys32_getuid16(void) { - return high2lowuid(current->uid); + return high2lowuid(current->cred->uid); } asmlinkage long sys32_geteuid16(void) { - return high2lowuid(current->euid); + return high2lowuid(current->cred->euid); } asmlinkage long sys32_getgid16(void) { - return high2lowgid(current->gid); + return high2lowgid(current->cred->gid); } asmlinkage long sys32_getegid16(void) { - return high2lowgid(current->egid); + return high2lowgid(current->cred->egid); } /* diff --git a/drivers/connector/cn_proc.c b/drivers/connector/cn_proc.c index 5c9f67f98d1..354c1ff1715 100644 --- a/drivers/connector/cn_proc.c +++ b/drivers/connector/cn_proc.c @@ -116,11 +116,11 @@ void proc_id_connector(struct task_struct *task, int which_id) ev->event_data.id.process_pid = task->pid; ev->event_data.id.process_tgid = task->tgid; if (which_id == PROC_EVENT_UID) { - ev->event_data.id.r.ruid = task->uid; - ev->event_data.id.e.euid = task->euid; + ev->event_data.id.r.ruid = task->cred->uid; + ev->event_data.id.e.euid = task->cred->euid; } else if (which_id == PROC_EVENT_GID) { - ev->event_data.id.r.rgid = task->gid; - ev->event_data.id.e.egid = task->egid; + ev->event_data.id.r.rgid = task->cred->gid; + ev->event_data.id.e.egid = task->cred->egid; } else return; get_seq(&msg->seq, &ev->cpu); diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 8fcfa398d35..7a52477ce49 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -223,10 +223,10 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec, NEW_AUX_ENT(AT_BASE, interp_load_addr); NEW_AUX_ENT(AT_FLAGS, 0); NEW_AUX_ENT(AT_ENTRY, exec->e_entry); - NEW_AUX_ENT(AT_UID, tsk->uid); - NEW_AUX_ENT(AT_EUID, tsk->euid); - NEW_AUX_ENT(AT_GID, tsk->gid); - NEW_AUX_ENT(AT_EGID, tsk->egid); + NEW_AUX_ENT(AT_UID, tsk->cred->uid); + NEW_AUX_ENT(AT_EUID, tsk->cred->euid); + NEW_AUX_ENT(AT_GID, tsk->cred->gid); + NEW_AUX_ENT(AT_EGID, tsk->cred->egid); NEW_AUX_ENT(AT_SECURE, security_bprm_secureexec(bprm)); NEW_AUX_ENT(AT_EXECFN, bprm->exec); if (k_platform) { @@ -1388,8 +1388,8 @@ static int fill_psinfo(struct elf_prpsinfo *psinfo, struct task_struct *p, psinfo->pr_zomb = psinfo->pr_sname == 'Z'; psinfo->pr_nice = task_nice(p); psinfo->pr_flag = p->flags; - SET_UID(psinfo->pr_uid, p->uid); - SET_GID(psinfo->pr_gid, p->gid); + SET_UID(psinfo->pr_uid, p->cred->uid); + SET_GID(psinfo->pr_gid, p->cred->gid); strncpy(psinfo->pr_fname, p->comm, sizeof(psinfo->pr_fname)); return 0; diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 488584c8751..9f67054c2c4 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -623,10 +623,10 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, NEW_AUX_ENT(AT_BASE, interp_params->elfhdr_addr); NEW_AUX_ENT(AT_FLAGS, 0); NEW_AUX_ENT(AT_ENTRY, exec_params->entry_addr); - NEW_AUX_ENT(AT_UID, (elf_addr_t) current_uid()); - NEW_AUX_ENT(AT_EUID, (elf_addr_t) current_euid()); - NEW_AUX_ENT(AT_GID, (elf_addr_t) current_gid()); - NEW_AUX_ENT(AT_EGID, (elf_addr_t) current_egid()); + NEW_AUX_ENT(AT_UID, (elf_addr_t) current->cred->uid); + NEW_AUX_ENT(AT_EUID, (elf_addr_t) current->cred->euid); + NEW_AUX_ENT(AT_GID, (elf_addr_t) current->cred->gid); + NEW_AUX_ENT(AT_EGID, (elf_addr_t) current->cred->egid); NEW_AUX_ENT(AT_SECURE, security_bprm_secureexec(bprm)); NEW_AUX_ENT(AT_EXECFN, bprm->exec); @@ -1440,8 +1440,8 @@ static int fill_psinfo(struct elf_prpsinfo *psinfo, struct task_struct *p, psinfo->pr_zomb = psinfo->pr_sname == 'Z'; psinfo->pr_nice = task_nice(p); psinfo->pr_flag = p->flags; - SET_UID(psinfo->pr_uid, p->uid); - SET_GID(psinfo->pr_gid, p->gid); + SET_UID(psinfo->pr_uid, p->cred->uid); + SET_GID(psinfo->pr_gid, p->cred->gid); strncpy(psinfo->pr_fname, p->comm, sizeof(psinfo->pr_fname)); return 0; diff --git a/fs/exec.c b/fs/exec.c index 604834f3b20..31149e430a8 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1738,7 +1738,7 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs) */ if (get_dumpable(mm) == 2) { /* Setuid core dump mode */ flag = O_EXCL; /* Stop rewrite attacks */ - current->fsuid = 0; /* Dump root private */ + current->cred->fsuid = 0; /* Dump root private */ } retval = coredump_wait(exit_code, &core_state); @@ -1834,7 +1834,7 @@ fail_unlock: if (helper_argv) argv_free(helper_argv); - current->fsuid = fsuid; + current->cred->fsuid = fsuid; coredump_finish(mm); fail: return retval; diff --git a/fs/fcntl.c b/fs/fcntl.c index bf049a805e5..63964d863ad 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -401,8 +401,8 @@ static inline int sigio_perm(struct task_struct *p, struct fown_struct *fown, int sig) { return (((fown->euid == 0) || - (fown->euid == p->suid) || (fown->euid == p->uid) || - (fown->uid == p->suid) || (fown->uid == p->uid)) && + (fown->euid == p->cred->suid) || (fown->euid == p->cred->uid) || + (fown->uid == p->cred->suid) || (fown->uid == p->cred->uid)) && !security_file_send_sigiotask(p, fown, sig)); } diff --git a/fs/file_table.c b/fs/file_table.c index 5ad0eca6eea..3152b53cfab 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -122,8 +122,8 @@ struct file *get_empty_filp(void) INIT_LIST_HEAD(&f->f_u.fu_list); atomic_long_set(&f->f_count, 1); rwlock_init(&f->f_owner.lock); - f->f_uid = tsk->fsuid; - f->f_gid = tsk->fsgid; + f->f_uid = tsk->cred->fsuid; + f->f_gid = tsk->cred->fsgid; eventpoll_init_file(f); /* f->f_version: 0 */ return f; diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index fd03330cade..e97a9898186 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -872,12 +872,12 @@ int fuse_allow_task(struct fuse_conn *fc, struct task_struct *task) if (fc->flags & FUSE_ALLOW_OTHER) return 1; - if (task->euid == fc->user_id && - task->suid == fc->user_id && - task->uid == fc->user_id && - task->egid == fc->group_id && - task->sgid == fc->group_id && - task->gid == fc->group_id) + if (task->cred->euid == fc->user_id && + task->cred->suid == fc->user_id && + task->cred->uid == fc->user_id && + task->cred->egid == fc->group_id && + task->cred->sgid == fc->group_id && + task->cred->gid == fc->group_id) return 1; return 0; diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 08ad76c79b4..870a721b8bd 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -958,7 +958,7 @@ struct file *hugetlb_file_setup(const char *name, size_t size) if (!can_do_hugetlb_shm()) return ERR_PTR(-EPERM); - if (!user_shm_lock(size, current->user)) + if (!user_shm_lock(size, current->cred->user)) return ERR_PTR(-ENOMEM); root = hugetlbfs_vfsmount->mnt_root; @@ -998,7 +998,7 @@ out_inode: out_dentry: dput(dentry); out_shm_unlock: - user_shm_unlock(size, current->user); + user_shm_unlock(size, current->cred->user); return ERR_PTR(error); } diff --git a/fs/ioprio.c b/fs/ioprio.c index 68d2cd80711..bb5210af77c 100644 --- a/fs/ioprio.c +++ b/fs/ioprio.c @@ -32,8 +32,8 @@ static int set_task_ioprio(struct task_struct *task, int ioprio) int err; struct io_context *ioc; - if (task->uid != current_euid() && - task->uid != current_uid() && !capable(CAP_SYS_NICE)) + if (task->cred->uid != current_euid() && + task->cred->uid != current_uid() && !capable(CAP_SYS_NICE)) return -EPERM; err = security_task_setioprio(task, ioprio); @@ -123,7 +123,7 @@ asmlinkage long sys_ioprio_set(int which, int who, int ioprio) break; case IOPRIO_WHO_USER: if (!who) - user = current->user; + user = current->cred->user; else user = find_user(who); @@ -131,7 +131,7 @@ asmlinkage long sys_ioprio_set(int which, int who, int ioprio) break; do_each_thread(g, p) { - if (p->uid != who) + if (p->cred->uid != who) continue; ret = set_task_ioprio(p, ioprio); if (ret) @@ -216,7 +216,7 @@ asmlinkage long sys_ioprio_get(int which, int who) break; case IOPRIO_WHO_USER: if (!who) - user = current->user; + user = current->cred->user; else user = find_user(who); @@ -224,7 +224,7 @@ asmlinkage long sys_ioprio_get(int which, int who) break; do_each_thread(g, p) { - if (p->uid != user->uid) + if (p->cred->uid != user->uid) continue; tmpio = get_task_ioprio(p); if (tmpio < 0) diff --git a/fs/nfsd/auth.c b/fs/nfsd/auth.c index 294992e9bf6..808fc03a6fb 100644 --- a/fs/nfsd/auth.c +++ b/fs/nfsd/auth.c @@ -27,6 +27,7 @@ int nfsexp_flags(struct svc_rqst *rqstp, struct svc_export *exp) int nfsd_setuser(struct svc_rqst *rqstp, struct svc_export *exp) { + struct cred *act_as = current->cred ; struct svc_cred cred = rqstp->rq_cred; int i; int flags = nfsexp_flags(rqstp, exp); @@ -55,25 +56,26 @@ int nfsd_setuser(struct svc_rqst *rqstp, struct svc_export *exp) get_group_info(cred.cr_group_info); if (cred.cr_uid != (uid_t) -1) - current->fsuid = cred.cr_uid; + act_as->fsuid = cred.cr_uid; else - current->fsuid = exp->ex_anon_uid; + act_as->fsuid = exp->ex_anon_uid; if (cred.cr_gid != (gid_t) -1) - current->fsgid = cred.cr_gid; + act_as->fsgid = cred.cr_gid; else - current->fsgid = exp->ex_anon_gid; + act_as->fsgid = exp->ex_anon_gid; if (!cred.cr_group_info) return -ENOMEM; - ret = set_current_groups(cred.cr_group_info); + ret = set_groups(act_as, cred.cr_group_info); put_group_info(cred.cr_group_info); if ((cred.cr_uid)) { - current->cap_effective = - cap_drop_nfsd_set(current->cap_effective); + act_as->cap_effective = + cap_drop_nfsd_set(act_as->cap_effective); } else { - current->cap_effective = - cap_raise_nfsd_set(current->cap_effective, - current->cap_permitted); + act_as->cap_effective = + cap_raise_nfsd_set(act_as->cap_effective, + act_as->cap_permitted); } return ret; } + diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index bb93946ace2..a5e14e8695e 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -57,17 +57,17 @@ static int rec_dir_init = 0; static void nfs4_save_user(uid_t *saveuid, gid_t *savegid) { - *saveuid = current->fsuid; - *savegid = current->fsgid; - current->fsuid = 0; - current->fsgid = 0; + *saveuid = current->cred->fsuid; + *savegid = current->cred->fsgid; + current->cred->fsuid = 0; + current->cred->fsgid = 0; } static void nfs4_reset_user(uid_t saveuid, gid_t savegid) { - current->fsuid = saveuid; - current->fsgid = savegid; + current->cred->fsuid = saveuid; + current->cred->fsgid = savegid; } static void diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index cd25d91895a..e67cfaea086 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -186,9 +186,9 @@ static __be32 nfsd_set_fh_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp) * access control settings being in effect, we cannot * fix that case easily. */ - current->cap_effective = - cap_raise_nfsd_set(current->cap_effective, - current->cap_permitted); + current->cred->cap_effective = + cap_raise_nfsd_set(current->cred->cap_effective, + current->cred->cap_permitted); } else { error = nfsd_setuser_and_check_port(rqstp, exp); if (error) diff --git a/fs/open.c b/fs/open.c index 500cc0c5476..b1238e195e7 100644 --- a/fs/open.c +++ b/fs/open.c @@ -425,6 +425,7 @@ out: */ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) { + struct cred *cred = current->cred; struct path path; struct inode *inode; int old_fsuid, old_fsgid; @@ -434,18 +435,18 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) if (mode & ~S_IRWXO) /* where's F_OK, X_OK, W_OK, R_OK? */ return -EINVAL; - old_fsuid = current->fsuid; - old_fsgid = current->fsgid; + old_fsuid = cred->fsuid; + old_fsgid = cred->fsgid; - current->fsuid = current->uid; - current->fsgid = current->gid; + cred->fsuid = cred->uid; + cred->fsgid = cred->gid; if (!issecure(SECURE_NO_SETUID_FIXUP)) { /* Clear the capabilities if we switch to a non-root user */ - if (current->uid) + if (current->cred->uid) old_cap = cap_set_effective(__cap_empty_set); else - old_cap = cap_set_effective(current->cap_permitted); + old_cap = cap_set_effective(cred->cap_permitted); } res = user_path_at(dfd, filename, LOOKUP_FOLLOW, &path); @@ -484,8 +485,8 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) out_path_release: path_put(&path); out: - current->fsuid = old_fsuid; - current->fsgid = old_fsgid; + cred->fsuid = old_fsuid; + cred->fsgid = old_fsgid; if (!issecure(SECURE_NO_SETUID_FIXUP)) cap_set_effective(old_cap); diff --git a/fs/proc/array.c b/fs/proc/array.c index 6af7fba7abb..62fe9b2009b 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -182,8 +182,8 @@ static inline void task_state(struct seq_file *m, struct pid_namespace *ns, task_tgid_nr_ns(p, ns), pid_nr_ns(pid, ns), ppid, tpid, - p->uid, p->euid, p->suid, p->fsuid, - p->gid, p->egid, p->sgid, p->fsgid); + p->cred->uid, p->cred->euid, p->cred->suid, p->cred->fsuid, + p->cred->gid, p->cred->egid, p->cred->sgid, p->cred->fsgid); task_lock(p); if (p->files) @@ -194,7 +194,7 @@ static inline void task_state(struct seq_file *m, struct pid_namespace *ns, fdt ? fdt->max_fds : 0); rcu_read_unlock(); - group_info = p->group_info; + group_info = p->cred->group_info; get_group_info(group_info); task_unlock(p); @@ -262,7 +262,7 @@ static inline void task_sig(struct seq_file *m, struct task_struct *p) blocked = p->blocked; collect_sigign_sigcatch(p, &ignored, &caught); num_threads = atomic_read(&p->signal->count); - qsize = atomic_read(&p->user->sigpending); + qsize = atomic_read(&p->cred->user->sigpending); qlim = p->signal->rlim[RLIMIT_SIGPENDING].rlim_cur; unlock_task_sighand(p, &flags); } @@ -293,10 +293,12 @@ static void render_cap_t(struct seq_file *m, const char *header, static inline void task_cap(struct seq_file *m, struct task_struct *p) { - render_cap_t(m, "CapInh:\t", &p->cap_inheritable); - render_cap_t(m, "CapPrm:\t", &p->cap_permitted); - render_cap_t(m, "CapEff:\t", &p->cap_effective); - render_cap_t(m, "CapBnd:\t", &p->cap_bset); + struct cred *cred = p->cred; + + render_cap_t(m, "CapInh:\t", &cred->cap_inheritable); + render_cap_t(m, "CapPrm:\t", &cred->cap_permitted); + render_cap_t(m, "CapEff:\t", &cred->cap_effective); + render_cap_t(m, "CapBnd:\t", &cred->cap_bset); } static inline void task_context_switch_counts(struct seq_file *m, diff --git a/fs/proc/base.c b/fs/proc/base.c index 486cf3fe713..6862b360c36 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1428,8 +1428,8 @@ static struct inode *proc_pid_make_inode(struct super_block * sb, struct task_st inode->i_uid = 0; inode->i_gid = 0; if (task_dumpable(task)) { - inode->i_uid = task->euid; - inode->i_gid = task->egid; + inode->i_uid = task->cred->euid; + inode->i_gid = task->cred->egid; } security_task_to_inode(task, inode); @@ -1454,8 +1454,8 @@ static int pid_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat if (task) { if ((inode->i_mode == (S_IFDIR|S_IRUGO|S_IXUGO)) || task_dumpable(task)) { - stat->uid = task->euid; - stat->gid = task->egid; + stat->uid = task->cred->euid; + stat->gid = task->cred->egid; } } rcu_read_unlock(); @@ -1486,8 +1486,8 @@ static int pid_revalidate(struct dentry *dentry, struct nameidata *nd) if (task) { if ((inode->i_mode == (S_IFDIR|S_IRUGO|S_IXUGO)) || task_dumpable(task)) { - inode->i_uid = task->euid; - inode->i_gid = task->egid; + inode->i_uid = task->cred->euid; + inode->i_gid = task->cred->egid; } else { inode->i_uid = 0; inode->i_gid = 0; @@ -1658,8 +1658,8 @@ static int tid_fd_revalidate(struct dentry *dentry, struct nameidata *nd) rcu_read_unlock(); put_files_struct(files); if (task_dumpable(task)) { - inode->i_uid = task->euid; - inode->i_gid = task->egid; + inode->i_uid = task->cred->euid; + inode->i_gid = task->cred->egid; } else { inode->i_uid = 0; inode->i_gid = 0; diff --git a/fs/xfs/linux-2.6/xfs_cred.h b/fs/xfs/linux-2.6/xfs_cred.h index 293043a5573..8c022cd0ad6 100644 --- a/fs/xfs/linux-2.6/xfs_cred.h +++ b/fs/xfs/linux-2.6/xfs_cred.h @@ -23,11 +23,9 @@ /* * Credentials */ -typedef struct cred { - /* EMPTY */ -} cred_t; +typedef const struct cred cred_t; -extern struct cred *sys_cred; +extern cred_t *sys_cred; /* this is a hack.. (assumes sys_cred is the only cred_t in the system) */ static inline int capable_cred(cred_t *cr, int cid) diff --git a/fs/xfs/linux-2.6/xfs_globals.h b/fs/xfs/linux-2.6/xfs_globals.h index 2770b0085ee..6eda8a3eb6f 100644 --- a/fs/xfs/linux-2.6/xfs_globals.h +++ b/fs/xfs/linux-2.6/xfs_globals.h @@ -19,6 +19,6 @@ #define __XFS_GLOBALS_H__ extern uint64_t xfs_panic_mask; /* set to cause more panics */ -extern struct cred *sys_cred; +extern cred_t *sys_cred; #endif /* __XFS_GLOBALS_H__ */ diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 1420c49674d..6be310d41da 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -497,7 +497,7 @@ int xfs_iread(struct xfs_mount *, struct xfs_trans *, xfs_ino_t, xfs_inode_t **, xfs_daddr_t, uint); int xfs_iread_extents(struct xfs_trans *, xfs_inode_t *, int); int xfs_ialloc(struct xfs_trans *, xfs_inode_t *, mode_t, - xfs_nlink_t, xfs_dev_t, struct cred *, xfs_prid_t, + xfs_nlink_t, xfs_dev_t, cred_t *, xfs_prid_t, int, struct xfs_buf **, boolean_t *, xfs_inode_t **); void xfs_dinode_from_disk(struct xfs_icdinode *, struct xfs_dinode_core *); diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h index e932a96bec5..7b0c2ab8833 100644 --- a/fs/xfs/xfs_vnodeops.h +++ b/fs/xfs/xfs_vnodeops.h @@ -16,7 +16,7 @@ struct xfs_iomap; int xfs_open(struct xfs_inode *ip); int xfs_setattr(struct xfs_inode *ip, struct iattr *vap, int flags, - struct cred *credp); + cred_t *credp); #define XFS_ATTR_DMI 0x01 /* invocation from a DMI function */ #define XFS_ATTR_NONBLOCK 0x02 /* return EAGAIN if operation would block */ #define XFS_ATTR_NOLOCK 0x04 /* Don't grab any conflicting locks */ @@ -28,24 +28,24 @@ int xfs_inactive(struct xfs_inode *ip); int xfs_lookup(struct xfs_inode *dp, struct xfs_name *name, struct xfs_inode **ipp, struct xfs_name *ci_name); int xfs_create(struct xfs_inode *dp, struct xfs_name *name, mode_t mode, - xfs_dev_t rdev, struct xfs_inode **ipp, struct cred *credp); + xfs_dev_t rdev, struct xfs_inode **ipp, cred_t *credp); int xfs_remove(struct xfs_inode *dp, struct xfs_name *name, struct xfs_inode *ip); int xfs_link(struct xfs_inode *tdp, struct xfs_inode *sip, struct xfs_name *target_name); int xfs_mkdir(struct xfs_inode *dp, struct xfs_name *dir_name, - mode_t mode, struct xfs_inode **ipp, struct cred *credp); + mode_t mode, struct xfs_inode **ipp, cred_t *credp); int xfs_readdir(struct xfs_inode *dp, void *dirent, size_t bufsize, xfs_off_t *offset, filldir_t filldir); int xfs_symlink(struct xfs_inode *dp, struct xfs_name *link_name, const char *target_path, mode_t mode, struct xfs_inode **ipp, - struct cred *credp); + cred_t *credp); int xfs_inode_flush(struct xfs_inode *ip, int flags); int xfs_set_dmattrs(struct xfs_inode *ip, u_int evmask, u_int16_t state); int xfs_reclaim(struct xfs_inode *ip); int xfs_change_file_space(struct xfs_inode *ip, int cmd, xfs_flock64_t *bf, xfs_off_t offset, - struct cred *credp, int attr_flags); + cred_t *credp, int attr_flags); int xfs_rename(struct xfs_inode *src_dp, struct xfs_name *src_name, struct xfs_inode *src_ip, struct xfs_inode *target_dp, struct xfs_name *target_name, struct xfs_inode *target_ip); diff --git a/include/linux/cred.h b/include/linux/cred.h index b69222cc1fd..3e65587a72e 100644 --- a/include/linux/cred.h +++ b/include/linux/cred.h @@ -12,39 +12,150 @@ #ifndef _LINUX_CRED_H #define _LINUX_CRED_H -#define get_current_user() (get_uid(current->user)) - -#define task_uid(task) ((task)->uid) -#define task_gid(task) ((task)->gid) -#define task_euid(task) ((task)->euid) -#define task_egid(task) ((task)->egid) - -#define current_uid() (current->uid) -#define current_gid() (current->gid) -#define current_euid() (current->euid) -#define current_egid() (current->egid) -#define current_suid() (current->suid) -#define current_sgid() (current->sgid) -#define current_fsuid() (current->fsuid) -#define current_fsgid() (current->fsgid) -#define current_cap() (current->cap_effective) +#include +#include +#include + +struct user_struct; +struct cred; + +/* + * COW Supplementary groups list + */ +#define NGROUPS_SMALL 32 +#define NGROUPS_PER_BLOCK ((unsigned int)(PAGE_SIZE / sizeof(gid_t))) + +struct group_info { + atomic_t usage; + int ngroups; + int nblocks; + gid_t small_block[NGROUPS_SMALL]; + gid_t *blocks[0]; +}; + +/** + * get_group_info - Get a reference to a group info structure + * @group_info: The group info to reference + * + * This must be called with the owning task locked (via task_lock()) when task + * != current. The reason being that the vast majority of callers are looking + * at current->group_info, which can not be changed except by the current task. + * Changing current->group_info requires the task lock, too. + */ +#define get_group_info(group_info) \ +do { \ + atomic_inc(&(group_info)->usage); \ +} while (0) + +/** + * put_group_info - Release a reference to a group info structure + * @group_info: The group info to release + */ +#define put_group_info(group_info) \ +do { \ + if (atomic_dec_and_test(&(group_info)->usage)) \ + groups_free(group_info); \ +} while (0) + +extern struct group_info *groups_alloc(int); +extern void groups_free(struct group_info *); +extern int set_current_groups(struct group_info *); +extern int set_groups(struct cred *, struct group_info *); +extern int groups_search(struct group_info *, gid_t); + +/* access the groups "array" with this macro */ +#define GROUP_AT(gi, i) \ + ((gi)->blocks[(i) / NGROUPS_PER_BLOCK][(i) % NGROUPS_PER_BLOCK]) + +extern int in_group_p(gid_t); +extern int in_egroup_p(gid_t); + +/* + * The security context of a task + * + * The parts of the context break down into two categories: + * + * (1) The objective context of a task. These parts are used when some other + * task is attempting to affect this one. + * + * (2) The subjective context. These details are used when the task is acting + * upon another object, be that a file, a task, a key or whatever. + * + * Note that some members of this structure belong to both categories - the + * LSM security pointer for instance. + * + * A task has two security pointers. task->real_cred points to the objective + * context that defines that task's actual details. The objective part of this + * context is used whenever that task is acted upon. + * + * task->cred points to the subjective context that defines the details of how + * that task is going to act upon another object. This may be overridden + * temporarily to point to another security context, but normally points to the + * same context as task->real_cred. + */ +struct cred { + atomic_t usage; + uid_t uid; /* real UID of the task */ + gid_t gid; /* real GID of the task */ + uid_t suid; /* saved UID of the task */ + gid_t sgid; /* saved GID of the task */ + uid_t euid; /* effective UID of the task */ + gid_t egid; /* effective GID of the task */ + uid_t fsuid; /* UID for VFS ops */ + gid_t fsgid; /* GID for VFS ops */ + unsigned securebits; /* SUID-less security management */ + kernel_cap_t cap_inheritable; /* caps our children can inherit */ + kernel_cap_t cap_permitted; /* caps we're permitted */ + kernel_cap_t cap_effective; /* caps we can actually use */ + kernel_cap_t cap_bset; /* capability bounding set */ +#ifdef CONFIG_KEYS + unsigned char jit_keyring; /* default keyring to attach requested + * keys to */ + struct key *thread_keyring; /* keyring private to this thread */ + struct key *request_key_auth; /* assumed request_key authority */ +#endif +#ifdef CONFIG_SECURITY + void *security; /* subjective LSM security */ +#endif + struct user_struct *user; /* real user ID subscription */ + struct group_info *group_info; /* supplementary groups for euid/fsgid */ + struct rcu_head rcu; /* RCU deletion hook */ + spinlock_t lock; /* lock for pointer changes */ +}; + +#define get_current_user() (get_uid(current->cred->user)) + +#define task_uid(task) ((task)->cred->uid) +#define task_gid(task) ((task)->cred->gid) +#define task_euid(task) ((task)->cred->euid) +#define task_egid(task) ((task)->cred->egid) + +#define current_uid() (current->cred->uid) +#define current_gid() (current->cred->gid) +#define current_euid() (current->cred->euid) +#define current_egid() (current->cred->egid) +#define current_suid() (current->cred->suid) +#define current_sgid() (current->cred->sgid) +#define current_fsuid() (current->cred->fsuid) +#define current_fsgid() (current->cred->fsgid) +#define current_cap() (current->cred->cap_effective) #define current_uid_gid(_uid, _gid) \ do { \ - *(_uid) = current->uid; \ - *(_gid) = current->gid; \ + *(_uid) = current->cred->uid; \ + *(_gid) = current->cred->gid; \ } while(0) #define current_euid_egid(_uid, _gid) \ do { \ - *(_uid) = current->euid; \ - *(_gid) = current->egid; \ + *(_uid) = current->cred->euid; \ + *(_gid) = current->cred->egid; \ } while(0) #define current_fsuid_fsgid(_uid, _gid) \ do { \ - *(_uid) = current->fsuid; \ - *(_gid) = current->fsgid; \ + *(_uid) = current->cred->fsuid; \ + *(_gid) = current->cred->fsgid; \ } while(0) #endif /* _LINUX_CRED_H */ diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 23fd8909b9e..9de41ccd67b 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -113,6 +113,21 @@ extern struct group_info init_groups; # define CAP_INIT_BSET CAP_INIT_EFF_SET #endif +extern struct cred init_cred; + +#define INIT_CRED(p) \ +{ \ + .usage = ATOMIC_INIT(3), \ + .securebits = SECUREBITS_DEFAULT, \ + .cap_inheritable = CAP_INIT_INH_SET, \ + .cap_permitted = CAP_FULL_SET, \ + .cap_effective = CAP_INIT_EFF_SET, \ + .cap_bset = CAP_INIT_BSET, \ + .user = INIT_USER, \ + .group_info = &init_groups, \ + .lock = __SPIN_LOCK_UNLOCKED(p.lock), \ +} + /* * INIT_TASK is used to set up the first task table, touch at * your own risk!. Base=0, limit=0x1fffff (=2MB) @@ -147,13 +162,8 @@ extern struct group_info init_groups; .children = LIST_HEAD_INIT(tsk.children), \ .sibling = LIST_HEAD_INIT(tsk.sibling), \ .group_leader = &tsk, \ - .group_info = &init_groups, \ - .cap_effective = CAP_INIT_EFF_SET, \ - .cap_inheritable = CAP_INIT_INH_SET, \ - .cap_permitted = CAP_FULL_SET, \ - .cap_bset = CAP_INIT_BSET, \ - .securebits = SECUREBITS_DEFAULT, \ - .user = INIT_USER, \ + .__temp_cred = INIT_CRED(tsk.__temp_cred), \ + .cred = &tsk.__temp_cred, \ .comm = "swapper", \ .thread = INIT_THREAD, \ .fs = &init_fs, \ diff --git a/include/linux/sched.h b/include/linux/sched.h index b483f39a711..c8b92502354 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -660,6 +660,7 @@ extern struct user_struct *find_user(uid_t); extern struct user_struct root_user; #define INIT_USER (&root_user) + struct backing_dev_info; struct reclaim_state; @@ -883,38 +884,7 @@ partition_sched_domains(int ndoms_new, cpumask_t *doms_new, #endif /* !CONFIG_SMP */ struct io_context; /* See blkdev.h */ -#define NGROUPS_SMALL 32 -#define NGROUPS_PER_BLOCK ((unsigned int)(PAGE_SIZE / sizeof(gid_t))) -struct group_info { - int ngroups; - atomic_t usage; - gid_t small_block[NGROUPS_SMALL]; - int nblocks; - gid_t *blocks[0]; -}; - -/* - * get_group_info() must be called with the owning task locked (via task_lock()) - * when task != current. The reason being that the vast majority of callers are - * looking at current->group_info, which can not be changed except by the - * current task. Changing current->group_info requires the task lock, too. - */ -#define get_group_info(group_info) do { \ - atomic_inc(&(group_info)->usage); \ -} while (0) -#define put_group_info(group_info) do { \ - if (atomic_dec_and_test(&(group_info)->usage)) \ - groups_free(group_info); \ -} while (0) - -extern struct group_info *groups_alloc(int gidsetsize); -extern void groups_free(struct group_info *group_info); -extern int set_current_groups(struct group_info *group_info); -extern int groups_search(struct group_info *group_info, gid_t grp); -/* access the groups "array" with this macro */ -#define GROUP_AT(gi, i) \ - ((gi)->blocks[(i)/NGROUPS_PER_BLOCK][(i)%NGROUPS_PER_BLOCK]) #ifdef ARCH_HAS_PREFETCH_SWITCH_STACK extern void prefetch_stack(struct task_struct *t); @@ -1181,17 +1151,9 @@ struct task_struct { struct list_head cpu_timers[3]; /* process credentials */ - uid_t uid,euid,suid,fsuid; - gid_t gid,egid,sgid,fsgid; - struct group_info *group_info; - kernel_cap_t cap_effective, cap_inheritable, cap_permitted, cap_bset; - struct user_struct *user; - unsigned securebits; -#ifdef CONFIG_KEYS - unsigned char jit_keyring; /* default keyring to attach requested keys to */ - struct key *request_key_auth; /* assumed request_key authority */ - struct key *thread_keyring; /* keyring private to this thread */ -#endif + struct cred __temp_cred __deprecated; /* temporary credentials to be removed */ + struct cred *cred; /* actual/objective task credentials */ + char comm[TASK_COMM_LEN]; /* executable name excluding path - access with [gs]et_task_comm (which lock it with task_lock()) @@ -1228,9 +1190,6 @@ struct task_struct { int (*notifier)(void *priv); void *notifier_data; sigset_t *notifier_mask; -#ifdef CONFIG_SECURITY - void *security; -#endif struct audit_context *audit_context; #ifdef CONFIG_AUDITSYSCALL uid_t loginuid; @@ -1787,9 +1746,6 @@ extern void wake_up_new_task(struct task_struct *tsk, extern void sched_fork(struct task_struct *p, int clone_flags); extern void sched_dead(struct task_struct *p); -extern int in_group_p(gid_t); -extern int in_egroup_p(gid_t); - extern void proc_caches_init(void); extern void flush_signals(struct task_struct *); extern void ignore_signals(struct task_struct *); diff --git a/include/linux/securebits.h b/include/linux/securebits.h index 92f09bdf117..6d389491bfa 100644 --- a/include/linux/securebits.h +++ b/include/linux/securebits.h @@ -32,7 +32,7 @@ setting is locked or not. A setting which is locked cannot be changed from user-level. */ #define issecure_mask(X) (1 << (X)) -#define issecure(X) (issecure_mask(X) & current->securebits) +#define issecure(X) (issecure_mask(X) & current->cred->securebits) #define SECURE_ALL_BITS (issecure_mask(SECURE_NOROOT) | \ issecure_mask(SECURE_NO_SETUID_FIXUP) | \ diff --git a/ipc/mqueue.c b/ipc/mqueue.c index abda5991d7e..e1885b494ba 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -126,7 +126,7 @@ static struct inode *mqueue_get_inode(struct super_block *sb, int mode, if (S_ISREG(mode)) { struct mqueue_inode_info *info; struct task_struct *p = current; - struct user_struct *u = p->user; + struct user_struct *u = p->cred->user; unsigned long mq_bytes, mq_msg_tblsz; inode->i_fop = &mqueue_file_operations; diff --git a/ipc/shm.c b/ipc/shm.c index 0c3debbe32d..264a9d33c5d 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -366,7 +366,7 @@ static int newseg(struct ipc_namespace *ns, struct ipc_params *params) if (shmflg & SHM_HUGETLB) { /* hugetlb_file_setup takes care of mlock user accounting */ file = hugetlb_file_setup(name, size); - shp->mlock_user = current->user; + shp->mlock_user = current->cred->user; } else { int acctflag = VM_ACCOUNT; /* @@ -767,7 +767,7 @@ asmlinkage long sys_shmctl(int shmid, int cmd, struct shmid_ds __user *buf) goto out_unlock; if(cmd==SHM_LOCK) { - struct user_struct * user = current->user; + struct user_struct *user = current->cred->user; if (!is_file_hugepages(shp->shm_file)) { err = shmem_lock(shp->shm_file, 1, user); if (!err && !(shp->shm_perm.mode & SHM_LOCKED)){ diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 9c7e47ae457..2febf5165fa 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -447,6 +447,7 @@ static int audit_filter_rules(struct task_struct *tsk, struct audit_names *name, enum audit_state *state) { + struct cred *cred = tsk->cred; int i, j, need_sid = 1; u32 sid; @@ -466,28 +467,28 @@ static int audit_filter_rules(struct task_struct *tsk, } break; case AUDIT_UID: - result = audit_comparator(tsk->uid, f->op, f->val); + result = audit_comparator(cred->uid, f->op, f->val); break; case AUDIT_EUID: - result = audit_comparator(tsk->euid, f->op, f->val); + result = audit_comparator(cred->euid, f->op, f->val); break; case AUDIT_SUID: - result = audit_comparator(tsk->suid, f->op, f->val); + result = audit_comparator(cred->suid, f->op, f->val); break; case AUDIT_FSUID: - result = audit_comparator(tsk->fsuid, f->op, f->val); + result = audit_comparator(cred->fsuid, f->op, f->val); break; case AUDIT_GID: - result = audit_comparator(tsk->gid, f->op, f->val); + result = audit_comparator(cred->gid, f->op, f->val); break; case AUDIT_EGID: - result = audit_comparator(tsk->egid, f->op, f->val); + result = audit_comparator(cred->egid, f->op, f->val); break; case AUDIT_SGID: - result = audit_comparator(tsk->sgid, f->op, f->val); + result = audit_comparator(cred->sgid, f->op, f->val); break; case AUDIT_FSGID: - result = audit_comparator(tsk->fsgid, f->op, f->val); + result = audit_comparator(cred->fsgid, f->op, f->val); break; case AUDIT_PERS: result = audit_comparator(tsk->personality, f->op, f->val); @@ -1228,6 +1229,7 @@ static void audit_log_fcaps(struct audit_buffer *ab, struct audit_names *name) static void audit_log_exit(struct audit_context *context, struct task_struct *tsk) { + struct cred *cred = tsk->cred; int i, call_panic = 0; struct audit_buffer *ab; struct audit_aux_data *aux; @@ -1237,14 +1239,14 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts context->pid = tsk->pid; if (!context->ppid) context->ppid = sys_getppid(); - context->uid = tsk->uid; - context->gid = tsk->gid; - context->euid = tsk->euid; - context->suid = tsk->suid; - context->fsuid = tsk->fsuid; - context->egid = tsk->egid; - context->sgid = tsk->sgid; - context->fsgid = tsk->fsgid; + context->uid = cred->uid; + context->gid = cred->gid; + context->euid = cred->euid; + context->suid = cred->suid; + context->fsuid = cred->fsuid; + context->egid = cred->egid; + context->sgid = cred->sgid; + context->fsgid = cred->fsgid; context->personality = tsk->personality; ab = audit_log_start(context, GFP_KERNEL, AUDIT_SYSCALL); @@ -2086,7 +2088,7 @@ int audit_set_loginuid(struct task_struct *task, uid_t loginuid) audit_log_format(ab, "login pid=%d uid=%u " "old auid=%u new auid=%u" " old ses=%u new ses=%u", - task->pid, task->uid, + task->pid, task->cred->uid, task->loginuid, loginuid, task->sessionid, sessionid); audit_log_end(ab); @@ -2469,7 +2471,7 @@ void __audit_ptrace(struct task_struct *t) context->target_pid = t->pid; context->target_auid = audit_get_loginuid(t); - context->target_uid = t->uid; + context->target_uid = t->cred->uid; context->target_sessionid = audit_get_sessionid(t); security_task_getsecid(t, &context->target_sid); memcpy(context->target_comm, t->comm, TASK_COMM_LEN); @@ -2495,7 +2497,7 @@ int __audit_signal_info(int sig, struct task_struct *t) if (tsk->loginuid != -1) audit_sig_uid = tsk->loginuid; else - audit_sig_uid = tsk->uid; + audit_sig_uid = tsk->cred->uid; security_task_getsecid(tsk, &audit_sig_sid); } if (!audit_signals || audit_dummy_context()) @@ -2507,7 +2509,7 @@ int __audit_signal_info(int sig, struct task_struct *t) if (!ctx->target_pid) { ctx->target_pid = t->tgid; ctx->target_auid = audit_get_loginuid(t); - ctx->target_uid = t->uid; + ctx->target_uid = t->cred->uid; ctx->target_sessionid = audit_get_sessionid(t); security_task_getsecid(t, &ctx->target_sid); memcpy(ctx->target_comm, t->comm, TASK_COMM_LEN); @@ -2528,7 +2530,7 @@ int __audit_signal_info(int sig, struct task_struct *t) axp->target_pid[axp->pid_count] = t->tgid; axp->target_auid[axp->pid_count] = audit_get_loginuid(t); - axp->target_uid[axp->pid_count] = t->uid; + axp->target_uid[axp->pid_count] = t->cred->uid; axp->target_sessionid[axp->pid_count] = audit_get_sessionid(t); security_task_getsecid(t, &axp->target_sid[axp->pid_count]); memcpy(axp->target_comm[axp->pid_count], t->comm, TASK_COMM_LEN); @@ -2575,12 +2577,12 @@ void __audit_log_bprm_fcaps(struct linux_binprm *bprm, kernel_cap_t *pP, kernel_ ax->fcap_ver = (vcaps.magic_etc & VFS_CAP_REVISION_MASK) >> VFS_CAP_REVISION_SHIFT; ax->old_pcap.permitted = *pP; - ax->old_pcap.inheritable = current->cap_inheritable; + ax->old_pcap.inheritable = current->cred->cap_inheritable; ax->old_pcap.effective = *pE; - ax->new_pcap.permitted = current->cap_permitted; - ax->new_pcap.inheritable = current->cap_inheritable; - ax->new_pcap.effective = current->cap_effective; + ax->new_pcap.permitted = current->cred->cap_permitted; + ax->new_pcap.inheritable = current->cred->cap_inheritable; + ax->new_pcap.effective = current->cred->cap_effective; } /** diff --git a/kernel/capability.c b/kernel/capability.c index 58b00519624..a404b980b1b 100644 --- a/kernel/capability.c +++ b/kernel/capability.c @@ -171,8 +171,8 @@ kernel_cap_t cap_set_effective(const kernel_cap_t pE_new) spin_lock(&task_capability_lock); - pE_old = current->cap_effective; - current->cap_effective = pE_new; + pE_old = current->cred->cap_effective; + current->cred->cap_effective = pE_new; spin_unlock(&task_capability_lock); diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 78f9b310c4f..e210526e640 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -1293,7 +1293,9 @@ static int attach_task_by_pid(struct cgroup *cgrp, u64 pid) rcu_read_unlock(); euid = current_euid(); - if (euid && euid != tsk->uid && euid != tsk->suid) { + if (euid && + euid != tsk->cred->uid && + euid != tsk->cred->suid) { put_task_struct(tsk); return -EACCES; } diff --git a/kernel/exit.c b/kernel/exit.c index 80137a5d946..e0f6e1892fb 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -160,7 +160,7 @@ void release_task(struct task_struct * p) int zap_leader; repeat: tracehook_prepare_release_task(p); - atomic_dec(&p->user->processes); + atomic_dec(&p->cred->user->processes); proc_flush_task(p); write_lock_irq(&tasklist_lock); tracehook_finish_release_task(p); @@ -1272,7 +1272,7 @@ static int wait_task_zombie(struct task_struct *p, int options, return 0; if (unlikely(options & WNOWAIT)) { - uid_t uid = p->uid; + uid_t uid = p->cred->uid; int exit_code = p->exit_code; int why, status; @@ -1393,7 +1393,7 @@ static int wait_task_zombie(struct task_struct *p, int options, if (!retval && infop) retval = put_user(pid, &infop->si_pid); if (!retval && infop) - retval = put_user(p->uid, &infop->si_uid); + retval = put_user(p->cred->uid, &infop->si_uid); if (!retval) retval = pid; @@ -1458,7 +1458,7 @@ static int wait_task_stopped(int ptrace, struct task_struct *p, if (!unlikely(options & WNOWAIT)) p->exit_code = 0; - uid = p->uid; + uid = p->cred->uid; unlock_sig: spin_unlock_irq(&p->sighand->siglock); if (!exit_code) @@ -1535,7 +1535,7 @@ static int wait_task_continued(struct task_struct *p, int options, spin_unlock_irq(&p->sighand->siglock); pid = task_pid_vnr(p); - uid = p->uid; + uid = p->cred->uid; get_task_struct(p); read_unlock(&tasklist_lock); diff --git a/kernel/fork.c b/kernel/fork.c index f6083561dfe..81fdc773390 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -147,8 +147,8 @@ void __put_task_struct(struct task_struct *tsk) WARN_ON(tsk == current); security_task_free(tsk); - free_uid(tsk->user); - put_group_info(tsk->group_info); + free_uid(tsk->__temp_cred.user); + put_group_info(tsk->__temp_cred.group_info); delayacct_tsk_free(tsk); if (!profile_handoff_task(tsk)) @@ -969,17 +969,18 @@ static struct task_struct *copy_process(unsigned long clone_flags, DEBUG_LOCKS_WARN_ON(!p->hardirqs_enabled); DEBUG_LOCKS_WARN_ON(!p->softirqs_enabled); #endif + p->cred = &p->__temp_cred; retval = -EAGAIN; - if (atomic_read(&p->user->processes) >= + if (atomic_read(&p->cred->user->processes) >= p->signal->rlim[RLIMIT_NPROC].rlim_cur) { if (!capable(CAP_SYS_ADMIN) && !capable(CAP_SYS_RESOURCE) && - p->user != current->nsproxy->user_ns->root_user) + p->cred->user != current->nsproxy->user_ns->root_user) goto bad_fork_free; } - atomic_inc(&p->user->__count); - atomic_inc(&p->user->processes); - get_group_info(p->group_info); + atomic_inc(&p->cred->user->__count); + atomic_inc(&p->cred->user->processes); + get_group_info(p->cred->group_info); /* * If multiple threads are within copy_process(), then this check @@ -1035,9 +1036,8 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->real_start_time = p->start_time; monotonic_to_bootbased(&p->real_start_time); #ifdef CONFIG_SECURITY - p->security = NULL; + p->cred->security = NULL; #endif - p->cap_bset = current->cap_bset; p->io_context = NULL; p->audit_context = NULL; cgroup_fork(p); @@ -1298,9 +1298,9 @@ bad_fork_cleanup_cgroup: bad_fork_cleanup_put_domain: module_put(task_thread_info(p)->exec_domain->module); bad_fork_cleanup_count: - put_group_info(p->group_info); - atomic_dec(&p->user->processes); - free_uid(p->user); + put_group_info(p->cred->group_info); + atomic_dec(&p->cred->user->processes); + free_uid(p->cred->user); bad_fork_free: free_task(p); fork_out: diff --git a/kernel/futex.c b/kernel/futex.c index e06962132aa..28421d8210b 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -443,7 +443,8 @@ static struct task_struct * futex_find_get_task(pid_t pid) rcu_read_lock(); p = find_task_by_vpid(pid); - if (!p || (euid != p->euid && euid != p->uid)) + if (!p || (euid != p->cred->euid && + euid != p->cred->uid)) p = ERR_PTR(-ESRCH); else get_task_struct(p); @@ -1846,7 +1847,8 @@ sys_get_robust_list(int pid, struct robust_list_head __user * __user *head_ptr, if (!p) goto err_unlock; ret = -EPERM; - if (euid != p->euid && euid != p->uid && + if (euid != p->cred->euid && + euid != p->cred->uid && !capable(CAP_SYS_PTRACE)) goto err_unlock; head = p->robust_list; diff --git a/kernel/futex_compat.c b/kernel/futex_compat.c index 3254d4e41e8..2c3fd5ed34f 100644 --- a/kernel/futex_compat.c +++ b/kernel/futex_compat.c @@ -151,8 +151,9 @@ compat_sys_get_robust_list(int pid, compat_uptr_t __user *head_ptr, if (!p) goto err_unlock; ret = -EPERM; - if (euid != p->euid && euid != p->uid && - !capable(CAP_SYS_PTRACE)) + if (euid != p->cred->euid && + euid != p->cred->uid && + !capable(CAP_SYS_PTRACE)) goto err_unlock; head = p->compat_robust_list; read_unlock(&tasklist_lock); diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 937f6b5b200..49849d12dd1 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -115,6 +115,8 @@ int ptrace_check_attach(struct task_struct *child, int kill) int __ptrace_may_access(struct task_struct *task, unsigned int mode) { + struct cred *cred = current->cred, *tcred = task->cred; + /* May we inspect the given task? * This check is used both for attaching with ptrace * and for allowing access to sensitive information in /proc. @@ -123,19 +125,18 @@ int __ptrace_may_access(struct task_struct *task, unsigned int mode) * because setting up the necessary parent/child relationship * or halting the specified task is impossible. */ - uid_t uid; - gid_t gid; + uid_t uid = cred->uid; + gid_t gid = cred->gid; int dumpable = 0; /* Don't let security modules deny introspection */ if (task == current) return 0; - current_uid_gid(&uid, &gid); - if ((uid != task->euid || - uid != task->suid || - uid != task->uid || - gid != task->egid || - gid != task->sgid || - gid != task->gid) && !capable(CAP_SYS_PTRACE)) + if ((uid != tcred->euid || + uid != tcred->suid || + uid != tcred->uid || + gid != tcred->egid || + gid != tcred->sgid || + gid != tcred->gid) && !capable(CAP_SYS_PTRACE)) return -EPERM; smp_rmb(); if (task->mm) diff --git a/kernel/sched.c b/kernel/sched.c index c3b8b1fcde0..733c59e645a 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -345,7 +345,7 @@ static inline struct task_group *task_group(struct task_struct *p) struct task_group *tg; #ifdef CONFIG_USER_SCHED - tg = p->user->tg; + tg = p->cred->user->tg; #elif defined(CONFIG_CGROUP_SCHED) tg = container_of(task_subsys_state(p, cpu_cgroup_subsys_id), struct task_group, css); @@ -5182,8 +5182,8 @@ recheck: /* can't change other user's priorities */ euid = current_euid(); - if (euid != p->euid && - euid != p->uid) + if (euid != p->cred->euid && + euid != p->cred->uid) return -EPERM; } @@ -5417,7 +5417,9 @@ long sched_setaffinity(pid_t pid, const cpumask_t *in_mask) euid = current_euid(); retval = -EPERM; - if (euid != p->euid && euid != p->uid && !capable(CAP_SYS_NICE)) + if (euid != p->cred->euid && + euid != p->cred->uid && + !capable(CAP_SYS_NICE)) goto out_unlock; retval = security_task_setscheduler(p, 0, NULL); diff --git a/kernel/signal.c b/kernel/signal.c index 167b535fe1a..80e8a6489f9 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -187,7 +187,7 @@ static struct sigqueue *__sigqueue_alloc(struct task_struct *t, gfp_t flags, * In order to avoid problems with "switch_user()", we want to make * sure that the compiler doesn't re-load "t->user" */ - user = t->user; + user = t->cred->user; barrier(); atomic_inc(&user->sigpending); if (override_rlimit || @@ -582,8 +582,8 @@ static int check_kill_permission(int sig, struct siginfo *info, uid = current_uid(); euid = current_euid(); - if ((euid ^ t->suid) && (euid ^ t->uid) && - (uid ^ t->suid) && (uid ^ t->uid) && + if ((euid ^ t->cred->suid) && (euid ^ t->cred->uid) && + (uid ^ t->cred->suid) && (uid ^ t->cred->uid) && !capable(CAP_KILL)) { switch (sig) { case SIGCONT: @@ -1100,8 +1100,8 @@ int kill_pid_info_as_uid(int sig, struct siginfo *info, struct pid *pid, goto out_unlock; } if ((info == SEND_SIG_NOINFO || (!is_si_special(info) && SI_FROMUSER(info))) - && (euid != p->suid) && (euid != p->uid) - && (uid != p->suid) && (uid != p->uid)) { + && (euid != p->cred->suid) && (euid != p->cred->uid) + && (uid != p->cred->suid) && (uid != p->cred->uid)) { ret = -EPERM; goto out_unlock; } @@ -1374,7 +1374,7 @@ int do_notify_parent(struct task_struct *tsk, int sig) info.si_pid = task_pid_nr_ns(tsk, tsk->parent->nsproxy->pid_ns); rcu_read_unlock(); - info.si_uid = tsk->uid; + info.si_uid = tsk->cred->uid; thread_group_cputime(tsk, &cputime); info.si_utime = cputime_to_jiffies(cputime.utime); @@ -1445,7 +1445,7 @@ static void do_notify_parent_cldstop(struct task_struct *tsk, int why) info.si_pid = task_pid_nr_ns(tsk, tsk->parent->nsproxy->pid_ns); rcu_read_unlock(); - info.si_uid = tsk->uid; + info.si_uid = tsk->cred->uid; info.si_utime = cputime_to_clock_t(tsk->utime); info.si_stime = cputime_to_clock_t(tsk->stime); @@ -1713,7 +1713,7 @@ static int ptrace_signal(int signr, siginfo_t *info, info->si_errno = 0; info->si_code = SI_USER; info->si_pid = task_pid_vnr(current->parent); - info->si_uid = current->parent->uid; + info->si_uid = current->parent->cred->uid; } /* If the (new) signal is now blocked, requeue it. */ diff --git a/kernel/sys.c b/kernel/sys.c index ed5c29c748a..5d81f07c015 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -117,7 +117,9 @@ static int set_one_prio(struct task_struct *p, int niceval, int error) uid_t euid = current_euid(); int no_nice; - if (p->uid != euid && p->euid != euid && !capable(CAP_SYS_NICE)) { + if (p->cred->uid != euid && + p->cred->euid != euid && + !capable(CAP_SYS_NICE)) { error = -EPERM; goto out; } @@ -174,7 +176,7 @@ asmlinkage long sys_setpriority(int which, int who, int niceval) } while_each_pid_thread(pgrp, PIDTYPE_PGID, p); break; case PRIO_USER: - user = current->user; + user = current->cred->user; if (!who) who = current_uid(); else @@ -182,7 +184,7 @@ asmlinkage long sys_setpriority(int which, int who, int niceval) goto out_unlock; /* No processes for this user */ do_each_thread(g, p) - if (p->uid == who) + if (p->cred->uid == who) error = set_one_prio(p, niceval, error); while_each_thread(g, p); if (who != current_uid()) @@ -236,7 +238,7 @@ asmlinkage long sys_getpriority(int which, int who) } while_each_pid_thread(pgrp, PIDTYPE_PGID, p); break; case PRIO_USER: - user = current->user; + user = current->cred->user; if (!who) who = current_uid(); else @@ -244,7 +246,7 @@ asmlinkage long sys_getpriority(int which, int who) goto out_unlock; /* No processes for this user */ do_each_thread(g, p) - if (p->uid == who) { + if (p->cred->uid == who) { niceval = 20 - task_nice(p); if (niceval > retval) retval = niceval; @@ -472,8 +474,9 @@ void ctrl_alt_del(void) */ asmlinkage long sys_setregid(gid_t rgid, gid_t egid) { - int old_rgid = current->gid; - int old_egid = current->egid; + struct cred *cred = current->cred; + int old_rgid = cred->gid; + int old_egid = cred->egid; int new_rgid = old_rgid; int new_egid = old_egid; int retval; @@ -484,7 +487,7 @@ asmlinkage long sys_setregid(gid_t rgid, gid_t egid) if (rgid != (gid_t) -1) { if ((old_rgid == rgid) || - (current->egid==rgid) || + (cred->egid == rgid) || capable(CAP_SETGID)) new_rgid = rgid; else @@ -492,8 +495,8 @@ asmlinkage long sys_setregid(gid_t rgid, gid_t egid) } if (egid != (gid_t) -1) { if ((old_rgid == egid) || - (current->egid == egid) || - (current->sgid == egid) || + (cred->egid == egid) || + (cred->sgid == egid) || capable(CAP_SETGID)) new_egid = egid; else @@ -505,10 +508,10 @@ asmlinkage long sys_setregid(gid_t rgid, gid_t egid) } if (rgid != (gid_t) -1 || (egid != (gid_t) -1 && egid != old_rgid)) - current->sgid = new_egid; - current->fsgid = new_egid; - current->egid = new_egid; - current->gid = new_rgid; + cred->sgid = new_egid; + cred->fsgid = new_egid; + cred->egid = new_egid; + cred->gid = new_rgid; key_fsgid_changed(current); proc_id_connector(current, PROC_EVENT_GID); return 0; @@ -521,7 +524,8 @@ asmlinkage long sys_setregid(gid_t rgid, gid_t egid) */ asmlinkage long sys_setgid(gid_t gid) { - int old_egid = current->egid; + struct cred *cred = current->cred; + int old_egid = cred->egid; int retval; retval = security_task_setgid(gid, (gid_t)-1, (gid_t)-1, LSM_SETID_ID); @@ -533,13 +537,13 @@ asmlinkage long sys_setgid(gid_t gid) set_dumpable(current->mm, suid_dumpable); smp_wmb(); } - current->gid = current->egid = current->sgid = current->fsgid = gid; - } else if ((gid == current->gid) || (gid == current->sgid)) { + cred->gid = cred->egid = cred->sgid = cred->fsgid = gid; + } else if ((gid == cred->gid) || (gid == cred->sgid)) { if (old_egid != gid) { set_dumpable(current->mm, suid_dumpable); smp_wmb(); } - current->egid = current->fsgid = gid; + cred->egid = cred->fsgid = gid; } else return -EPERM; @@ -570,7 +574,7 @@ static int set_user(uid_t new_ruid, int dumpclear) set_dumpable(current->mm, suid_dumpable); smp_wmb(); } - current->uid = new_ruid; + current->cred->uid = new_ruid; return 0; } @@ -591,6 +595,7 @@ static int set_user(uid_t new_ruid, int dumpclear) */ asmlinkage long sys_setreuid(uid_t ruid, uid_t euid) { + struct cred *cred = current->cred; int old_ruid, old_euid, old_suid, new_ruid, new_euid; int retval; @@ -598,14 +603,14 @@ asmlinkage long sys_setreuid(uid_t ruid, uid_t euid) if (retval) return retval; - new_ruid = old_ruid = current->uid; - new_euid = old_euid = current->euid; - old_suid = current->suid; + new_ruid = old_ruid = cred->uid; + new_euid = old_euid = cred->euid; + old_suid = cred->suid; if (ruid != (uid_t) -1) { new_ruid = ruid; if ((old_ruid != ruid) && - (current->euid != ruid) && + (cred->euid != ruid) && !capable(CAP_SETUID)) return -EPERM; } @@ -613,8 +618,8 @@ asmlinkage long sys_setreuid(uid_t ruid, uid_t euid) if (euid != (uid_t) -1) { new_euid = euid; if ((old_ruid != euid) && - (current->euid != euid) && - (current->suid != euid) && + (cred->euid != euid) && + (cred->suid != euid) && !capable(CAP_SETUID)) return -EPERM; } @@ -626,11 +631,11 @@ asmlinkage long sys_setreuid(uid_t ruid, uid_t euid) set_dumpable(current->mm, suid_dumpable); smp_wmb(); } - current->fsuid = current->euid = new_euid; + cred->fsuid = cred->euid = new_euid; if (ruid != (uid_t) -1 || (euid != (uid_t) -1 && euid != old_ruid)) - current->suid = current->euid; - current->fsuid = current->euid; + cred->suid = cred->euid; + cred->fsuid = cred->euid; key_fsuid_changed(current); proc_id_connector(current, PROC_EVENT_UID); @@ -653,7 +658,8 @@ asmlinkage long sys_setreuid(uid_t ruid, uid_t euid) */ asmlinkage long sys_setuid(uid_t uid) { - int old_euid = current->euid; + struct cred *cred = current->cred; + int old_euid = cred->euid; int old_ruid, old_suid, new_suid; int retval; @@ -661,23 +667,23 @@ asmlinkage long sys_setuid(uid_t uid) if (retval) return retval; - old_ruid = current->uid; - old_suid = current->suid; + old_ruid = cred->uid; + old_suid = cred->suid; new_suid = old_suid; if (capable(CAP_SETUID)) { if (uid != old_ruid && set_user(uid, old_euid != uid) < 0) return -EAGAIN; new_suid = uid; - } else if ((uid != current->uid) && (uid != new_suid)) + } else if ((uid != cred->uid) && (uid != new_suid)) return -EPERM; if (old_euid != uid) { set_dumpable(current->mm, suid_dumpable); smp_wmb(); } - current->fsuid = current->euid = uid; - current->suid = new_suid; + cred->fsuid = cred->euid = uid; + cred->suid = new_suid; key_fsuid_changed(current); proc_id_connector(current, PROC_EVENT_UID); @@ -692,9 +698,10 @@ asmlinkage long sys_setuid(uid_t uid) */ asmlinkage long sys_setresuid(uid_t ruid, uid_t euid, uid_t suid) { - int old_ruid = current->uid; - int old_euid = current->euid; - int old_suid = current->suid; + struct cred *cred = current->cred; + int old_ruid = cred->uid; + int old_euid = cred->euid; + int old_suid = cred->suid; int retval; retval = security_task_setuid(ruid, euid, suid, LSM_SETID_RES); @@ -702,30 +709,31 @@ asmlinkage long sys_setresuid(uid_t ruid, uid_t euid, uid_t suid) return retval; if (!capable(CAP_SETUID)) { - if ((ruid != (uid_t) -1) && (ruid != current->uid) && - (ruid != current->euid) && (ruid != current->suid)) + if ((ruid != (uid_t) -1) && (ruid != cred->uid) && + (ruid != cred->euid) && (ruid != cred->suid)) return -EPERM; - if ((euid != (uid_t) -1) && (euid != current->uid) && - (euid != current->euid) && (euid != current->suid)) + if ((euid != (uid_t) -1) && (euid != cred->uid) && + (euid != cred->euid) && (euid != cred->suid)) return -EPERM; - if ((suid != (uid_t) -1) && (suid != current->uid) && - (suid != current->euid) && (suid != current->suid)) + if ((suid != (uid_t) -1) && (suid != cred->uid) && + (suid != cred->euid) && (suid != cred->suid)) return -EPERM; } if (ruid != (uid_t) -1) { - if (ruid != current->uid && set_user(ruid, euid != current->euid) < 0) + if (ruid != cred->uid && + set_user(ruid, euid != cred->euid) < 0) return -EAGAIN; } if (euid != (uid_t) -1) { - if (euid != current->euid) { + if (euid != cred->euid) { set_dumpable(current->mm, suid_dumpable); smp_wmb(); } - current->euid = euid; + cred->euid = euid; } - current->fsuid = current->euid; + cred->fsuid = cred->euid; if (suid != (uid_t) -1) - current->suid = suid; + cred->suid = suid; key_fsuid_changed(current); proc_id_connector(current, PROC_EVENT_UID); @@ -735,11 +743,12 @@ asmlinkage long sys_setresuid(uid_t ruid, uid_t euid, uid_t suid) asmlinkage long sys_getresuid(uid_t __user *ruid, uid_t __user *euid, uid_t __user *suid) { + struct cred *cred = current->cred; int retval; - if (!(retval = put_user(current->uid, ruid)) && - !(retval = put_user(current->euid, euid))) - retval = put_user(current->suid, suid); + if (!(retval = put_user(cred->uid, ruid)) && + !(retval = put_user(cred->euid, euid))) + retval = put_user(cred->suid, suid); return retval; } @@ -749,6 +758,7 @@ asmlinkage long sys_getresuid(uid_t __user *ruid, uid_t __user *euid, uid_t __us */ asmlinkage long sys_setresgid(gid_t rgid, gid_t egid, gid_t sgid) { + struct cred *cred = current->cred; int retval; retval = security_task_setgid(rgid, egid, sgid, LSM_SETID_RES); @@ -756,28 +766,28 @@ asmlinkage long sys_setresgid(gid_t rgid, gid_t egid, gid_t sgid) return retval; if (!capable(CAP_SETGID)) { - if ((rgid != (gid_t) -1) && (rgid != current->gid) && - (rgid != current->egid) && (rgid != current->sgid)) + if ((rgid != (gid_t) -1) && (rgid != cred->gid) && + (rgid != cred->egid) && (rgid != cred->sgid)) return -EPERM; - if ((egid != (gid_t) -1) && (egid != current->gid) && - (egid != current->egid) && (egid != current->sgid)) + if ((egid != (gid_t) -1) && (egid != cred->gid) && + (egid != cred->egid) && (egid != cred->sgid)) return -EPERM; - if ((sgid != (gid_t) -1) && (sgid != current->gid) && - (sgid != current->egid) && (sgid != current->sgid)) + if ((sgid != (gid_t) -1) && (sgid != cred->gid) && + (sgid != cred->egid) && (sgid != cred->sgid)) return -EPERM; } if (egid != (gid_t) -1) { - if (egid != current->egid) { + if (egid != cred->egid) { set_dumpable(current->mm, suid_dumpable); smp_wmb(); } - current->egid = egid; + cred->egid = egid; } - current->fsgid = current->egid; + cred->fsgid = cred->egid; if (rgid != (gid_t) -1) - current->gid = rgid; + cred->gid = rgid; if (sgid != (gid_t) -1) - current->sgid = sgid; + cred->sgid = sgid; key_fsgid_changed(current); proc_id_connector(current, PROC_EVENT_GID); @@ -786,11 +796,12 @@ asmlinkage long sys_setresgid(gid_t rgid, gid_t egid, gid_t sgid) asmlinkage long sys_getresgid(gid_t __user *rgid, gid_t __user *egid, gid_t __user *sgid) { + struct cred *cred = current->cred; int retval; - if (!(retval = put_user(current->gid, rgid)) && - !(retval = put_user(current->egid, egid))) - retval = put_user(current->sgid, sgid); + if (!(retval = put_user(cred->gid, rgid)) && + !(retval = put_user(cred->egid, egid))) + retval = put_user(cred->sgid, sgid); return retval; } @@ -804,20 +815,21 @@ asmlinkage long sys_getresgid(gid_t __user *rgid, gid_t __user *egid, gid_t __us */ asmlinkage long sys_setfsuid(uid_t uid) { + struct cred *cred = current->cred; int old_fsuid; - old_fsuid = current->fsuid; + old_fsuid = cred->fsuid; if (security_task_setuid(uid, (uid_t)-1, (uid_t)-1, LSM_SETID_FS)) return old_fsuid; - if (uid == current->uid || uid == current->euid || - uid == current->suid || uid == current->fsuid || + if (uid == cred->uid || uid == cred->euid || + uid == cred->suid || uid == cred->fsuid || capable(CAP_SETUID)) { if (uid != old_fsuid) { set_dumpable(current->mm, suid_dumpable); smp_wmb(); } - current->fsuid = uid; + cred->fsuid = uid; } key_fsuid_changed(current); @@ -833,20 +845,21 @@ asmlinkage long sys_setfsuid(uid_t uid) */ asmlinkage long sys_setfsgid(gid_t gid) { + struct cred *cred = current->cred; int old_fsgid; - old_fsgid = current->fsgid; + old_fsgid = cred->fsgid; if (security_task_setgid(gid, (gid_t)-1, (gid_t)-1, LSM_SETID_FS)) return old_fsgid; - if (gid == current->gid || gid == current->egid || - gid == current->sgid || gid == current->fsgid || + if (gid == cred->gid || gid == cred->egid || + gid == cred->sgid || gid == cred->fsgid || capable(CAP_SETGID)) { if (gid != old_fsgid) { set_dumpable(current->mm, suid_dumpable); smp_wmb(); } - current->fsgid = gid; + cred->fsgid = gid; key_fsgid_changed(current); proc_id_connector(current, PROC_EVENT_GID); } @@ -1208,8 +1221,15 @@ int groups_search(struct group_info *group_info, gid_t grp) return 0; } -/* validate and set current->group_info */ -int set_current_groups(struct group_info *group_info) +/** + * set_groups - Change a group subscription in a security record + * @sec: The security record to alter + * @group_info: The group list to impose + * + * Validate a group subscription and, if valid, impose it upon a task security + * record. + */ +int set_groups(struct cred *cred, struct group_info *group_info) { int retval; struct group_info *old_info; @@ -1221,20 +1241,34 @@ int set_current_groups(struct group_info *group_info) groups_sort(group_info); get_group_info(group_info); - task_lock(current); - old_info = current->group_info; - current->group_info = group_info; - task_unlock(current); + spin_lock(&cred->lock); + old_info = cred->group_info; + cred->group_info = group_info; + spin_unlock(&cred->lock); put_group_info(old_info); - return 0; } +EXPORT_SYMBOL(set_groups); + +/** + * set_current_groups - Change current's group subscription + * @group_info: The group list to impose + * + * Validate a group subscription and, if valid, impose it upon current's task + * security record. + */ +int set_current_groups(struct group_info *group_info) +{ + return set_groups(current->cred, group_info); +} + EXPORT_SYMBOL(set_current_groups); asmlinkage long sys_getgroups(int gidsetsize, gid_t __user *grouplist) { + struct cred *cred = current->cred; int i = 0; /* @@ -1246,13 +1280,13 @@ asmlinkage long sys_getgroups(int gidsetsize, gid_t __user *grouplist) return -EINVAL; /* no need to grab task_lock here; it cannot change */ - i = current->group_info->ngroups; + i = cred->group_info->ngroups; if (gidsetsize) { if (i > gidsetsize) { i = -EINVAL; goto out; } - if (groups_to_user(grouplist, current->group_info)) { + if (groups_to_user(grouplist, cred->group_info)) { i = -EFAULT; goto out; } @@ -1296,9 +1330,10 @@ asmlinkage long sys_setgroups(int gidsetsize, gid_t __user *grouplist) */ int in_group_p(gid_t grp) { + struct cred *cred = current->cred; int retval = 1; - if (grp != current->fsgid) - retval = groups_search(current->group_info, grp); + if (grp != cred->fsgid) + retval = groups_search(cred->group_info, grp); return retval; } @@ -1306,9 +1341,10 @@ EXPORT_SYMBOL(in_group_p); int in_egroup_p(gid_t grp) { + struct cred *cred = current->cred; int retval = 1; - if (grp != current->egid) - retval = groups_search(current->group_info, grp); + if (grp != cred->egid) + retval = groups_search(cred->group_info, grp); return retval; } @@ -1624,7 +1660,9 @@ asmlinkage long sys_umask(int mask) asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3, unsigned long arg4, unsigned long arg5) { - long error = 0; + struct task_struct *me = current; + unsigned char comm[sizeof(me->comm)]; + long error; if (security_task_prctl(option, arg2, arg3, arg4, arg5, &error)) return error; @@ -1635,39 +1673,41 @@ asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3, error = -EINVAL; break; } - current->pdeath_signal = arg2; + me->pdeath_signal = arg2; + error = 0; break; case PR_GET_PDEATHSIG: - error = put_user(current->pdeath_signal, (int __user *)arg2); + error = put_user(me->pdeath_signal, (int __user *)arg2); break; case PR_GET_DUMPABLE: - error = get_dumpable(current->mm); + error = get_dumpable(me->mm); break; case PR_SET_DUMPABLE: if (arg2 < 0 || arg2 > 1) { error = -EINVAL; break; } - set_dumpable(current->mm, arg2); + set_dumpable(me->mm, arg2); + error = 0; break; case PR_SET_UNALIGN: - error = SET_UNALIGN_CTL(current, arg2); + error = SET_UNALIGN_CTL(me, arg2); break; case PR_GET_UNALIGN: - error = GET_UNALIGN_CTL(current, arg2); + error = GET_UNALIGN_CTL(me, arg2); break; case PR_SET_FPEMU: - error = SET_FPEMU_CTL(current, arg2); + error = SET_FPEMU_CTL(me, arg2); break; case PR_GET_FPEMU: - error = GET_FPEMU_CTL(current, arg2); + error = GET_FPEMU_CTL(me, arg2); break; case PR_SET_FPEXC: - error = SET_FPEXC_CTL(current, arg2); + error = SET_FPEXC_CTL(me, arg2); break; case PR_GET_FPEXC: - error = GET_FPEXC_CTL(current, arg2); + error = GET_FPEXC_CTL(me, arg2); break; case PR_GET_TIMING: error = PR_TIMING_STATISTICAL; @@ -1675,33 +1715,28 @@ asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3, case PR_SET_TIMING: if (arg2 != PR_TIMING_STATISTICAL) error = -EINVAL; + else + error = 0; break; - case PR_SET_NAME: { - struct task_struct *me = current; - unsigned char ncomm[sizeof(me->comm)]; - - ncomm[sizeof(me->comm)-1] = 0; - if (strncpy_from_user(ncomm, (char __user *)arg2, - sizeof(me->comm)-1) < 0) + case PR_SET_NAME: + comm[sizeof(me->comm)-1] = 0; + if (strncpy_from_user(comm, (char __user *)arg2, + sizeof(me->comm) - 1) < 0) return -EFAULT; - set_task_comm(me, ncomm); + set_task_comm(me, comm); return 0; - } - case PR_GET_NAME: { - struct task_struct *me = current; - unsigned char tcomm[sizeof(me->comm)]; - - get_task_comm(tcomm, me); - if (copy_to_user((char __user *)arg2, tcomm, sizeof(tcomm))) + case PR_GET_NAME: + get_task_comm(comm, me); + if (copy_to_user((char __user *)arg2, comm, + sizeof(comm))) return -EFAULT; return 0; - } case PR_GET_ENDIAN: - error = GET_ENDIAN(current, arg2); + error = GET_ENDIAN(me, arg2); break; case PR_SET_ENDIAN: - error = SET_ENDIAN(current, arg2); + error = SET_ENDIAN(me, arg2); break; case PR_GET_SECCOMP: @@ -1725,6 +1760,7 @@ asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3, current->default_timer_slack_ns; else current->timer_slack_ns = arg2; + error = 0; break; default: error = -EINVAL; diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 9f3b478f917..5c97c5b4ea8 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -246,7 +246,7 @@ __update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu) memcpy(data->comm, tsk->comm, TASK_COMM_LEN); data->pid = tsk->pid; - data->uid = tsk->uid; + data->uid = task_uid(tsk); data->nice = tsk->static_prio - 20 - MAX_RT_PRIO; data->policy = tsk->policy; data->rt_priority = tsk->rt_priority; diff --git a/kernel/tsacct.c b/kernel/tsacct.c index 8ebcd8532df..6d1ed07bf31 100644 --- a/kernel/tsacct.c +++ b/kernel/tsacct.c @@ -53,8 +53,8 @@ void bacct_add_tsk(struct taskstats *stats, struct task_struct *tsk) stats->ac_flag |= AXSIG; stats->ac_nice = task_nice(tsk); stats->ac_sched = tsk->policy; - stats->ac_uid = tsk->uid; - stats->ac_gid = tsk->gid; + stats->ac_uid = tsk->cred->uid; + stats->ac_gid = tsk->cred->gid; stats->ac_pid = tsk->pid; rcu_read_lock(); stats->ac_ppid = pid_alive(tsk) ? diff --git a/kernel/uid16.c b/kernel/uid16.c index 3e41c1673e2..71f07fc39fe 100644 --- a/kernel/uid16.c +++ b/kernel/uid16.c @@ -86,9 +86,9 @@ asmlinkage long sys_getresuid16(old_uid_t __user *ruid, old_uid_t __user *euid, { int retval; - if (!(retval = put_user(high2lowuid(current->uid), ruid)) && - !(retval = put_user(high2lowuid(current->euid), euid))) - retval = put_user(high2lowuid(current->suid), suid); + if (!(retval = put_user(high2lowuid(current->cred->uid), ruid)) && + !(retval = put_user(high2lowuid(current->cred->euid), euid))) + retval = put_user(high2lowuid(current->cred->suid), suid); return retval; } @@ -106,9 +106,9 @@ asmlinkage long sys_getresgid16(old_gid_t __user *rgid, old_gid_t __user *egid, { int retval; - if (!(retval = put_user(high2lowgid(current->gid), rgid)) && - !(retval = put_user(high2lowgid(current->egid), egid))) - retval = put_user(high2lowgid(current->sgid), sgid); + if (!(retval = put_user(high2lowgid(current->cred->gid), rgid)) && + !(retval = put_user(high2lowgid(current->cred->egid), egid))) + retval = put_user(high2lowgid(current->cred->sgid), sgid); return retval; } @@ -166,20 +166,20 @@ asmlinkage long sys_getgroups16(int gidsetsize, old_gid_t __user *grouplist) if (gidsetsize < 0) return -EINVAL; - get_group_info(current->group_info); - i = current->group_info->ngroups; + get_group_info(current->cred->group_info); + i = current->cred->group_info->ngroups; if (gidsetsize) { if (i > gidsetsize) { i = -EINVAL; goto out; } - if (groups16_to_user(grouplist, current->group_info)) { + if (groups16_to_user(grouplist, current->cred->group_info)) { i = -EFAULT; goto out; } } out: - put_group_info(current->group_info); + put_group_info(current->cred->group_info); return i; } @@ -210,20 +210,20 @@ asmlinkage long sys_setgroups16(int gidsetsize, old_gid_t __user *grouplist) asmlinkage long sys_getuid16(void) { - return high2lowuid(current->uid); + return high2lowuid(current->cred->uid); } asmlinkage long sys_geteuid16(void) { - return high2lowuid(current->euid); + return high2lowuid(current->cred->euid); } asmlinkage long sys_getgid16(void) { - return high2lowgid(current->gid); + return high2lowgid(current->cred->gid); } asmlinkage long sys_getegid16(void) { - return high2lowgid(current->egid); + return high2lowgid(current->cred->egid); } diff --git a/kernel/user.c b/kernel/user.c index 39d6159fae4..104d22ac84d 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -457,11 +457,11 @@ void switch_uid(struct user_struct *new_user) * cheaply with the new uid cache, so if it matters * we should be checking for it. -DaveM */ - old_user = current->user; + old_user = current->cred->user; atomic_inc(&new_user->processes); atomic_dec(&old_user->processes); switch_uid_keyring(new_user); - current->user = new_user; + current->cred->user = new_user; sched_switch_user(current); /* diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 07a96474077..b23492ee3e5 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -1110,12 +1110,12 @@ asmlinkage long sys_migrate_pages(pid_t pid, unsigned long maxnode, const unsigned long __user *old_nodes, const unsigned long __user *new_nodes) { + struct cred *cred, *tcred; struct mm_struct *mm; struct task_struct *task; nodemask_t old; nodemask_t new; nodemask_t task_nodes; - uid_t uid, euid; int err; err = get_nodes(&old, old_nodes, maxnode); @@ -1145,10 +1145,10 @@ asmlinkage long sys_migrate_pages(pid_t pid, unsigned long maxnode, * capabilities, superuser privileges or the same * userid as the target process. */ - uid = current_uid(); - euid = current_euid(); - if (euid != task->suid && euid != task->uid && - uid != task->suid && uid != task->uid && + cred = current->cred; + tcred = task->cred; + if (cred->euid != tcred->suid && cred->euid != tcred->uid && + cred->uid != tcred->suid && cred->uid != tcred->uid && !capable(CAP_SYS_NICE)) { err = -EPERM; goto out; diff --git a/mm/migrate.c b/mm/migrate.c index 6263c24c4af..794443da1b4 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1045,10 +1045,10 @@ asmlinkage long sys_move_pages(pid_t pid, unsigned long nr_pages, const int __user *nodes, int __user *status, int flags) { + struct cred *cred, *tcred; struct task_struct *task; struct mm_struct *mm; int err; - uid_t uid, euid; /* Check flags */ if (flags & ~(MPOL_MF_MOVE|MPOL_MF_MOVE_ALL)) @@ -1076,10 +1076,10 @@ asmlinkage long sys_move_pages(pid_t pid, unsigned long nr_pages, * capabilities, superuser privileges or the same * userid as the target process. */ - uid = current_uid(); - euid = current_euid(); - if (euid != task->suid && euid != task->uid && - uid != task->suid && uid != task->uid && + cred = current->cred; + tcred = task->cred; + if (cred->euid != tcred->suid && cred->euid != tcred->uid && + cred->uid != tcred->suid && cred->uid != tcred->uid && !capable(CAP_SYS_NICE)) { err = -EPERM; goto out; diff --git a/mm/oom_kill.c b/mm/oom_kill.c index 34a458aa799..3af787ba207 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -298,7 +298,7 @@ static void dump_tasks(const struct mem_cgroup *mem) task_lock(p); printk(KERN_INFO "[%5d] %5d %5d %8lu %8lu %3d %3d %s\n", - p->pid, p->uid, p->tgid, p->mm->total_vm, + p->pid, p->cred->uid, p->tgid, p->mm->total_vm, get_mm_rss(p->mm), (int)task_cpu(p), p->oomkilladj, p->comm); task_unlock(p); diff --git a/net/core/scm.c b/net/core/scm.c index 4681d8f9b45..c28ca32a7d9 100644 --- a/net/core/scm.c +++ b/net/core/scm.c @@ -44,11 +44,13 @@ static __inline__ int scm_check_creds(struct ucred *creds) { + struct cred *cred = current->cred; + if ((creds->pid == task_tgid_vnr(current) || capable(CAP_SYS_ADMIN)) && - ((creds->uid == current_uid() || creds->uid == current_euid() || - creds->uid == current_suid()) || capable(CAP_SETUID)) && - ((creds->gid == current_gid() || creds->gid == current_egid() || - creds->gid == current_sgid()) || capable(CAP_SETGID))) { + ((creds->uid == cred->uid || creds->uid == cred->euid || + creds->uid == cred->suid) || capable(CAP_SETUID)) && + ((creds->gid == cred->gid || creds->gid == cred->egid || + creds->gid == cred->sgid) || capable(CAP_SETGID))) { return 0; } return -EPERM; diff --git a/net/sunrpc/auth.c b/net/sunrpc/auth.c index 8fc38057880..c7954321260 100644 --- a/net/sunrpc/auth.c +++ b/net/sunrpc/auth.c @@ -353,7 +353,7 @@ rpcauth_lookupcred(struct rpc_auth *auth, int flags) struct auth_cred acred = { .uid = current_fsuid(), .gid = current_fsgid(), - .group_info = current->group_info, + .group_info = current->cred->group_info, }; struct rpc_cred *ret; diff --git a/security/commoncap.c b/security/commoncap.c index fb4e240720d..fa61679f8c7 100644 --- a/security/commoncap.c +++ b/security/commoncap.c @@ -30,7 +30,7 @@ int cap_netlink_send(struct sock *sk, struct sk_buff *skb) { - NETLINK_CB(skb).eff_cap = current->cap_effective; + NETLINK_CB(skb).eff_cap = current_cap(); return 0; } @@ -52,7 +52,7 @@ EXPORT_SYMBOL(cap_netlink_recv); int cap_capable(struct task_struct *tsk, int cap, int audit) { /* Derived from include/linux/sched.h:capable. */ - if (cap_raised(tsk->cap_effective, cap)) + if (cap_raised(tsk->cred->cap_effective, cap)) return 0; return -EPERM; } @@ -67,7 +67,8 @@ int cap_settime(struct timespec *ts, struct timezone *tz) int cap_ptrace_may_access(struct task_struct *child, unsigned int mode) { /* Derived from arch/i386/kernel/ptrace.c:sys_ptrace. */ - if (cap_issubset(child->cap_permitted, current->cap_permitted)) + if (cap_issubset(child->cred->cap_permitted, + current->cred->cap_permitted)) return 0; if (capable(CAP_SYS_PTRACE)) return 0; @@ -76,8 +77,8 @@ int cap_ptrace_may_access(struct task_struct *child, unsigned int mode) int cap_ptrace_traceme(struct task_struct *parent) { - /* Derived from arch/i386/kernel/ptrace.c:sys_ptrace. */ - if (cap_issubset(current->cap_permitted, parent->cap_permitted)) + if (cap_issubset(current->cred->cap_permitted, + parent->cred->cap_permitted)) return 0; if (has_capability(parent, CAP_SYS_PTRACE)) return 0; @@ -87,10 +88,12 @@ int cap_ptrace_traceme(struct task_struct *parent) int cap_capget (struct task_struct *target, kernel_cap_t *effective, kernel_cap_t *inheritable, kernel_cap_t *permitted) { + struct cred *cred = target->cred; + /* Derived from kernel/capability.c:sys_capget. */ - *effective = target->cap_effective; - *inheritable = target->cap_inheritable; - *permitted = target->cap_permitted; + *effective = cred->cap_effective; + *inheritable = cred->cap_inheritable; + *permitted = cred->cap_permitted; return 0; } @@ -122,24 +125,26 @@ int cap_capset_check(const kernel_cap_t *effective, const kernel_cap_t *inheritable, const kernel_cap_t *permitted) { + const struct cred *cred = current->cred; + if (cap_inh_is_capped() && !cap_issubset(*inheritable, - cap_combine(current->cap_inheritable, - current->cap_permitted))) { + cap_combine(cred->cap_inheritable, + cred->cap_permitted))) { /* incapable of using this inheritable set */ return -EPERM; } if (!cap_issubset(*inheritable, - cap_combine(current->cap_inheritable, - current->cap_bset))) { + cap_combine(cred->cap_inheritable, + cred->cap_bset))) { /* no new pI capabilities outside bounding set */ return -EPERM; } /* verify restrictions on target's new Permitted set */ if (!cap_issubset (*permitted, - cap_combine (current->cap_permitted, - current->cap_permitted))) { + cap_combine (cred->cap_permitted, + cred->cap_permitted))) { return -EPERM; } @@ -155,9 +160,11 @@ void cap_capset_set(const kernel_cap_t *effective, const kernel_cap_t *inheritable, const kernel_cap_t *permitted) { - current->cap_effective = *effective; - current->cap_inheritable = *inheritable; - current->cap_permitted = *permitted; + struct cred *cred = current->cred; + + cred->cap_effective = *effective; + cred->cap_inheritable = *inheritable; + cred->cap_permitted = *permitted; } static inline void bprm_clear_caps(struct linux_binprm *bprm) @@ -211,8 +218,8 @@ static inline int bprm_caps_from_vfs_caps(struct cpu_vfs_cap_data *caps, * pP' = (X & fP) | (pI & fI) */ bprm->cap_post_exec_permitted.cap[i] = - (current->cap_bset.cap[i] & permitted) | - (current->cap_inheritable.cap[i] & inheritable); + (current->cred->cap_bset.cap[i] & permitted) | + (current->cred->cap_inheritable.cap[i] & inheritable); if (permitted & ~bprm->cap_post_exec_permitted.cap[i]) { /* @@ -354,8 +361,8 @@ int cap_bprm_set_security (struct linux_binprm *bprm) if (bprm->e_uid == 0 || current_uid() == 0) { /* pP' = (cap_bset & ~0) | (pI & ~0) */ bprm->cap_post_exec_permitted = cap_combine( - current->cap_bset, current->cap_inheritable - ); + current->cred->cap_bset, + current->cred->cap_inheritable); bprm->cap_effective = (bprm->e_uid == 0); ret = 0; } @@ -366,44 +373,39 @@ int cap_bprm_set_security (struct linux_binprm *bprm) void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe) { - kernel_cap_t pP = current->cap_permitted; - kernel_cap_t pE = current->cap_effective; - uid_t uid; - gid_t gid; + struct cred *cred = current->cred; - current_uid_gid(&uid, &gid); - - if (bprm->e_uid != uid || bprm->e_gid != gid || + if (bprm->e_uid != cred->uid || bprm->e_gid != cred->gid || !cap_issubset(bprm->cap_post_exec_permitted, - current->cap_permitted)) { + cred->cap_permitted)) { set_dumpable(current->mm, suid_dumpable); current->pdeath_signal = 0; if (unsafe & ~LSM_UNSAFE_PTRACE_CAP) { if (!capable(CAP_SETUID)) { - bprm->e_uid = uid; - bprm->e_gid = gid; + bprm->e_uid = cred->uid; + bprm->e_gid = cred->gid; } if (cap_limit_ptraced_target()) { bprm->cap_post_exec_permitted = cap_intersect( bprm->cap_post_exec_permitted, - current->cap_permitted); + cred->cap_permitted); } } } - current->suid = current->euid = current->fsuid = bprm->e_uid; - current->sgid = current->egid = current->fsgid = bprm->e_gid; + cred->suid = cred->euid = cred->fsuid = bprm->e_uid; + cred->sgid = cred->egid = cred->fsgid = bprm->e_gid; /* For init, we want to retain the capabilities set * in the init_task struct. Thus we skip the usual * capability rules */ if (!is_global_init(current)) { - current->cap_permitted = bprm->cap_post_exec_permitted; + cred->cap_permitted = bprm->cap_post_exec_permitted; if (bprm->cap_effective) - current->cap_effective = bprm->cap_post_exec_permitted; + cred->cap_effective = bprm->cap_post_exec_permitted; else - cap_clear(current->cap_effective); + cap_clear(cred->cap_effective); } /* @@ -418,27 +420,30 @@ void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe) * Number 1 above might fail if you don't have a full bset, but I think * that is interesting information to audit. */ - if (!cap_isclear(current->cap_effective)) { - if (!cap_issubset(CAP_FULL_SET, current->cap_effective) || - (bprm->e_uid != 0) || (current->uid != 0) || + if (!cap_isclear(cred->cap_effective)) { + if (!cap_issubset(CAP_FULL_SET, cred->cap_effective) || + (bprm->e_uid != 0) || (cred->uid != 0) || issecure(SECURE_NOROOT)) - audit_log_bprm_fcaps(bprm, &pP, &pE); + audit_log_bprm_fcaps(bprm, &cred->cap_permitted, + &cred->cap_effective); } - current->securebits &= ~issecure_mask(SECURE_KEEP_CAPS); + cred->securebits &= ~issecure_mask(SECURE_KEEP_CAPS); } int cap_bprm_secureexec (struct linux_binprm *bprm) { - if (current_uid() != 0) { + const struct cred *cred = current->cred; + + if (cred->uid != 0) { if (bprm->cap_effective) return 1; if (!cap_isclear(bprm->cap_post_exec_permitted)) return 1; } - return (current_euid() != current_uid() || - current_egid() != current_gid()); + return (cred->euid != cred->uid || + cred->egid != cred->gid); } int cap_inode_setxattr(struct dentry *dentry, const char *name, @@ -501,25 +506,27 @@ int cap_inode_removexattr(struct dentry *dentry, const char *name) static inline void cap_emulate_setxuid (int old_ruid, int old_euid, int old_suid) { - uid_t euid = current_euid(); + struct cred *cred = current->cred; if ((old_ruid == 0 || old_euid == 0 || old_suid == 0) && - (current_uid() != 0 && euid != 0 && current_suid() != 0) && + (cred->uid != 0 && cred->euid != 0 && cred->suid != 0) && !issecure(SECURE_KEEP_CAPS)) { - cap_clear (current->cap_permitted); - cap_clear (current->cap_effective); + cap_clear (cred->cap_permitted); + cap_clear (cred->cap_effective); } - if (old_euid == 0 && euid != 0) { - cap_clear (current->cap_effective); + if (old_euid == 0 && cred->euid != 0) { + cap_clear (cred->cap_effective); } - if (old_euid != 0 && euid == 0) { - current->cap_effective = current->cap_permitted; + if (old_euid != 0 && cred->euid == 0) { + cred->cap_effective = cred->cap_permitted; } } int cap_task_post_setuid (uid_t old_ruid, uid_t old_euid, uid_t old_suid, int flags) { + struct cred *cred = current->cred; + switch (flags) { case LSM_SETID_RE: case LSM_SETID_ID: @@ -541,16 +548,16 @@ int cap_task_post_setuid (uid_t old_ruid, uid_t old_euid, uid_t old_suid, */ if (!issecure (SECURE_NO_SETUID_FIXUP)) { - if (old_fsuid == 0 && current_fsuid() != 0) { - current->cap_effective = + if (old_fsuid == 0 && cred->fsuid != 0) { + cred->cap_effective = cap_drop_fs_set( - current->cap_effective); + cred->cap_effective); } - if (old_fsuid != 0 && current_fsuid() == 0) { - current->cap_effective = + if (old_fsuid != 0 && cred->fsuid == 0) { + cred->cap_effective = cap_raise_fs_set( - current->cap_effective, - current->cap_permitted); + cred->cap_effective, + cred->cap_permitted); } } break; @@ -575,7 +582,8 @@ int cap_task_post_setuid (uid_t old_ruid, uid_t old_euid, uid_t old_suid, */ static int cap_safe_nice(struct task_struct *p) { - if (!cap_issubset(p->cap_permitted, current->cap_permitted) && + if (!cap_issubset(p->cred->cap_permitted, + current->cred->cap_permitted) && !capable(CAP_SYS_NICE)) return -EPERM; return 0; @@ -610,7 +618,7 @@ static long cap_prctl_drop(unsigned long cap) return -EPERM; if (!cap_valid(cap)) return -EINVAL; - cap_lower(current->cap_bset, cap); + cap_lower(current->cred->cap_bset, cap); return 0; } @@ -633,6 +641,7 @@ int cap_task_setnice (struct task_struct *p, int nice) int cap_task_prctl(int option, unsigned long arg2, unsigned long arg3, unsigned long arg4, unsigned long arg5, long *rc_p) { + struct cred *cred = current->cred; long error = 0; switch (option) { @@ -640,7 +649,7 @@ int cap_task_prctl(int option, unsigned long arg2, unsigned long arg3, if (!cap_valid(arg2)) error = -EINVAL; else - error = !!cap_raised(current->cap_bset, arg2); + error = !!cap_raised(cred->cap_bset, arg2); break; #ifdef CONFIG_SECURITY_FILE_CAPABILITIES case PR_CAPBSET_DROP: @@ -667,9 +676,9 @@ int cap_task_prctl(int option, unsigned long arg2, unsigned long arg3, * capability-based-privilege environment. */ case PR_SET_SECUREBITS: - if ((((current->securebits & SECURE_ALL_LOCKS) >> 1) - & (current->securebits ^ arg2)) /*[1]*/ - || ((current->securebits & SECURE_ALL_LOCKS + if ((((cred->securebits & SECURE_ALL_LOCKS) >> 1) + & (cred->securebits ^ arg2)) /*[1]*/ + || ((cred->securebits & SECURE_ALL_LOCKS & ~arg2)) /*[2]*/ || (arg2 & ~(SECURE_ALL_LOCKS | SECURE_ALL_BITS)) /*[3]*/ || (cap_capable(current, CAP_SETPCAP, SECURITY_CAP_AUDIT) != 0)) { /*[4]*/ @@ -682,11 +691,11 @@ int cap_task_prctl(int option, unsigned long arg2, unsigned long arg3, */ error = -EPERM; /* cannot change a locked bit */ } else { - current->securebits = arg2; + cred->securebits = arg2; } break; case PR_GET_SECUREBITS: - error = current->securebits; + error = cred->securebits; break; #endif /* def CONFIG_SECURITY_FILE_CAPABILITIES */ @@ -701,10 +710,9 @@ int cap_task_prctl(int option, unsigned long arg2, unsigned long arg3, else if (issecure(SECURE_KEEP_CAPS_LOCKED)) error = -EPERM; else if (arg2) - current->securebits |= issecure_mask(SECURE_KEEP_CAPS); + cred->securebits |= issecure_mask(SECURE_KEEP_CAPS); else - current->securebits &= - ~issecure_mask(SECURE_KEEP_CAPS); + cred->securebits &= ~issecure_mask(SECURE_KEEP_CAPS); break; default: @@ -719,11 +727,12 @@ int cap_task_prctl(int option, unsigned long arg2, unsigned long arg3, void cap_task_reparent_to_init (struct task_struct *p) { - cap_set_init_eff(p->cap_effective); - cap_clear(p->cap_inheritable); - cap_set_full(p->cap_permitted); - p->securebits = SECUREBITS_DEFAULT; - return; + struct cred *cred = p->cred; + + cap_set_init_eff(cred->cap_effective); + cap_clear(cred->cap_inheritable); + cap_set_full(cred->cap_permitted); + p->cred->securebits = SECUREBITS_DEFAULT; } int cap_syslog (int type) diff --git a/security/keys/keyctl.c b/security/keys/keyctl.c index fcce331eca7..8833b447ade 100644 --- a/security/keys/keyctl.c +++ b/security/keys/keyctl.c @@ -889,7 +889,7 @@ long keyctl_instantiate_key(key_serial_t id, /* the appropriate instantiation authorisation key must have been * assumed before calling this */ ret = -EPERM; - instkey = current->request_key_auth; + instkey = current->cred->request_key_auth; if (!instkey) goto error; @@ -932,8 +932,8 @@ long keyctl_instantiate_key(key_serial_t id, /* discard the assumed authority if it's just been disabled by * instantiation of the key */ if (ret == 0) { - key_put(current->request_key_auth); - current->request_key_auth = NULL; + key_put(current->cred->request_key_auth); + current->cred->request_key_auth = NULL; } error2: @@ -960,7 +960,7 @@ long keyctl_negate_key(key_serial_t id, unsigned timeout, key_serial_t ringid) /* the appropriate instantiation authorisation key must have been * assumed before calling this */ ret = -EPERM; - instkey = current->request_key_auth; + instkey = current->cred->request_key_auth; if (!instkey) goto error; @@ -983,8 +983,8 @@ long keyctl_negate_key(key_serial_t id, unsigned timeout, key_serial_t ringid) /* discard the assumed authority if it's just been disabled by * instantiation of the key */ if (ret == 0) { - key_put(current->request_key_auth); - current->request_key_auth = NULL; + key_put(current->cred->request_key_auth); + current->cred->request_key_auth = NULL; } error: @@ -999,6 +999,7 @@ error: */ long keyctl_set_reqkey_keyring(int reqkey_defl) { + struct cred *cred = current->cred; int ret; switch (reqkey_defl) { @@ -1018,10 +1019,10 @@ long keyctl_set_reqkey_keyring(int reqkey_defl) case KEY_REQKEY_DEFL_USER_KEYRING: case KEY_REQKEY_DEFL_USER_SESSION_KEYRING: set: - current->jit_keyring = reqkey_defl; + cred->jit_keyring = reqkey_defl; case KEY_REQKEY_DEFL_NO_CHANGE: - return current->jit_keyring; + return cred->jit_keyring; case KEY_REQKEY_DEFL_GROUP_KEYRING: default: @@ -1086,8 +1087,8 @@ long keyctl_assume_authority(key_serial_t id) /* we divest ourselves of authority if given an ID of 0 */ if (id == 0) { - key_put(current->request_key_auth); - current->request_key_auth = NULL; + key_put(current->cred->request_key_auth); + current->cred->request_key_auth = NULL; ret = 0; goto error; } @@ -1103,8 +1104,8 @@ long keyctl_assume_authority(key_serial_t id) goto error; } - key_put(current->request_key_auth); - current->request_key_auth = authkey; + key_put(current->cred->request_key_auth); + current->cred->request_key_auth = authkey; ret = authkey->serial; error: diff --git a/security/keys/permission.c b/security/keys/permission.c index 3b41f9b5253..baf3d5f31e7 100644 --- a/security/keys/permission.c +++ b/security/keys/permission.c @@ -22,6 +22,7 @@ int key_task_permission(const key_ref_t key_ref, struct task_struct *context, key_perm_t perm) { + struct cred *cred = context->cred; struct key *key; key_perm_t kperm; int ret; @@ -29,7 +30,7 @@ int key_task_permission(const key_ref_t key_ref, key = key_ref_to_ptr(key_ref); /* use the second 8-bits of permissions for keys the caller owns */ - if (key->uid == context->fsuid) { + if (key->uid == cred->fsuid) { kperm = key->perm >> 16; goto use_these_perms; } @@ -37,14 +38,14 @@ int key_task_permission(const key_ref_t key_ref, /* use the third 8-bits of permissions for keys the caller has a group * membership in common with */ if (key->gid != -1 && key->perm & KEY_GRP_ALL) { - if (key->gid == context->fsgid) { + if (key->gid == cred->fsgid) { kperm = key->perm >> 8; goto use_these_perms; } - task_lock(context); - ret = groups_search(context->group_info, key->gid); - task_unlock(context); + spin_lock(&cred->lock); + ret = groups_search(cred->group_info, key->gid); + spin_unlock(&cred->lock); if (ret) { kperm = key->perm >> 8; diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c index 1c793b7090a..b0904cdda2e 100644 --- a/security/keys/process_keys.c +++ b/security/keys/process_keys.c @@ -42,7 +42,7 @@ struct key_user root_key_user = { */ int install_user_keyrings(void) { - struct user_struct *user = current->user; + struct user_struct *user = current->cred->user; struct key *uid_keyring, *session_keyring; char buf[20]; int ret; @@ -156,7 +156,7 @@ int install_thread_keyring(void) sprintf(buf, "_tid.%u", tsk->pid); - keyring = keyring_alloc(buf, tsk->uid, tsk->gid, tsk, + keyring = keyring_alloc(buf, tsk->cred->uid, tsk->cred->gid, tsk, KEY_ALLOC_QUOTA_OVERRUN, NULL); if (IS_ERR(keyring)) { ret = PTR_ERR(keyring); @@ -164,8 +164,8 @@ int install_thread_keyring(void) } task_lock(tsk); - old = tsk->thread_keyring; - tsk->thread_keyring = keyring; + old = tsk->cred->thread_keyring; + tsk->cred->thread_keyring = keyring; task_unlock(tsk); ret = 0; @@ -192,7 +192,7 @@ int install_process_keyring(void) if (!tsk->signal->process_keyring) { sprintf(buf, "_pid.%u", tsk->tgid); - keyring = keyring_alloc(buf, tsk->uid, tsk->gid, tsk, + keyring = keyring_alloc(buf, tsk->cred->uid, tsk->cred->gid, tsk, KEY_ALLOC_QUOTA_OVERRUN, NULL); if (IS_ERR(keyring)) { ret = PTR_ERR(keyring); @@ -238,7 +238,7 @@ static int install_session_keyring(struct key *keyring) if (tsk->signal->session_keyring) flags = KEY_ALLOC_IN_QUOTA; - keyring = keyring_alloc(buf, tsk->uid, tsk->gid, tsk, + keyring = keyring_alloc(buf, tsk->cred->uid, tsk->cred->gid, tsk, flags, NULL); if (IS_ERR(keyring)) return PTR_ERR(keyring); @@ -292,14 +292,14 @@ int copy_thread_group_keys(struct task_struct *tsk) */ int copy_keys(unsigned long clone_flags, struct task_struct *tsk) { - key_check(tsk->thread_keyring); - key_check(tsk->request_key_auth); + key_check(tsk->cred->thread_keyring); + key_check(tsk->cred->request_key_auth); /* no thread keyring yet */ - tsk->thread_keyring = NULL; + tsk->cred->thread_keyring = NULL; /* copy the request_key() authorisation for this thread */ - key_get(tsk->request_key_auth); + key_get(tsk->cred->request_key_auth); return 0; @@ -322,8 +322,8 @@ void exit_thread_group_keys(struct signal_struct *tg) */ void exit_keys(struct task_struct *tsk) { - key_put(tsk->thread_keyring); - key_put(tsk->request_key_auth); + key_put(tsk->cred->thread_keyring); + key_put(tsk->cred->request_key_auth); } /* end exit_keys() */ @@ -337,8 +337,8 @@ int exec_keys(struct task_struct *tsk) /* newly exec'd tasks don't get a thread keyring */ task_lock(tsk); - old = tsk->thread_keyring; - tsk->thread_keyring = NULL; + old = tsk->cred->thread_keyring; + tsk->cred->thread_keyring = NULL; task_unlock(tsk); key_put(old); @@ -373,10 +373,11 @@ int suid_keys(struct task_struct *tsk) void key_fsuid_changed(struct task_struct *tsk) { /* update the ownership of the thread keyring */ - if (tsk->thread_keyring) { - down_write(&tsk->thread_keyring->sem); - tsk->thread_keyring->uid = tsk->fsuid; - up_write(&tsk->thread_keyring->sem); + BUG_ON(!tsk->cred); + if (tsk->cred->thread_keyring) { + down_write(&tsk->cred->thread_keyring->sem); + tsk->cred->thread_keyring->uid = tsk->cred->fsuid; + up_write(&tsk->cred->thread_keyring->sem); } } /* end key_fsuid_changed() */ @@ -388,10 +389,11 @@ void key_fsuid_changed(struct task_struct *tsk) void key_fsgid_changed(struct task_struct *tsk) { /* update the ownership of the thread keyring */ - if (tsk->thread_keyring) { - down_write(&tsk->thread_keyring->sem); - tsk->thread_keyring->gid = tsk->fsgid; - up_write(&tsk->thread_keyring->sem); + BUG_ON(!tsk->cred); + if (tsk->cred->thread_keyring) { + down_write(&tsk->cred->thread_keyring->sem); + tsk->cred->thread_keyring->gid = tsk->cred->fsgid; + up_write(&tsk->cred->thread_keyring->sem); } } /* end key_fsgid_changed() */ @@ -426,9 +428,9 @@ key_ref_t search_process_keyrings(struct key_type *type, err = ERR_PTR(-EAGAIN); /* search the thread keyring first */ - if (context->thread_keyring) { + if (context->cred->thread_keyring) { key_ref = keyring_search_aux( - make_key_ref(context->thread_keyring, 1), + make_key_ref(context->cred->thread_keyring, 1), context, type, description, match); if (!IS_ERR(key_ref)) goto found; @@ -493,9 +495,9 @@ key_ref_t search_process_keyrings(struct key_type *type, } } /* or search the user-session keyring */ - else if (context->user->session_keyring) { + else if (context->cred->user->session_keyring) { key_ref = keyring_search_aux( - make_key_ref(context->user->session_keyring, 1), + make_key_ref(context->cred->user->session_keyring, 1), context, type, description, match); if (!IS_ERR(key_ref)) goto found; @@ -517,20 +519,20 @@ key_ref_t search_process_keyrings(struct key_type *type, * search the keyrings of the process mentioned there * - we don't permit access to request_key auth keys via this method */ - if (context->request_key_auth && + if (context->cred->request_key_auth && context == current && type != &key_type_request_key_auth ) { /* defend against the auth key being revoked */ - down_read(&context->request_key_auth->sem); + down_read(&context->cred->request_key_auth->sem); - if (key_validate(context->request_key_auth) == 0) { - rka = context->request_key_auth->payload.data; + if (key_validate(context->cred->request_key_auth) == 0) { + rka = context->cred->request_key_auth->payload.data; key_ref = search_process_keyrings(type, description, match, rka->context); - up_read(&context->request_key_auth->sem); + up_read(&context->cred->request_key_auth->sem); if (!IS_ERR(key_ref)) goto found; @@ -547,7 +549,7 @@ key_ref_t search_process_keyrings(struct key_type *type, break; } } else { - up_read(&context->request_key_auth->sem); + up_read(&context->cred->request_key_auth->sem); } } @@ -580,15 +582,16 @@ key_ref_t lookup_user_key(key_serial_t id, int create, int partial, { struct request_key_auth *rka; struct task_struct *t = current; - key_ref_t key_ref, skey_ref; + struct cred *cred = t->cred; struct key *key; + key_ref_t key_ref, skey_ref; int ret; key_ref = ERR_PTR(-ENOKEY); switch (id) { case KEY_SPEC_THREAD_KEYRING: - if (!t->thread_keyring) { + if (!cred->thread_keyring) { if (!create) goto error; @@ -599,7 +602,7 @@ key_ref_t lookup_user_key(key_serial_t id, int create, int partial, } } - key = t->thread_keyring; + key = cred->thread_keyring; atomic_inc(&key->usage); key_ref = make_key_ref(key, 1); break; @@ -628,7 +631,8 @@ key_ref_t lookup_user_key(key_serial_t id, int create, int partial, ret = install_user_keyrings(); if (ret < 0) goto error; - ret = install_session_keyring(t->user->session_keyring); + ret = install_session_keyring( + cred->user->session_keyring); if (ret < 0) goto error; } @@ -641,25 +645,25 @@ key_ref_t lookup_user_key(key_serial_t id, int create, int partial, break; case KEY_SPEC_USER_KEYRING: - if (!t->user->uid_keyring) { + if (!cred->user->uid_keyring) { ret = install_user_keyrings(); if (ret < 0) goto error; } - key = t->user->uid_keyring; + key = cred->user->uid_keyring; atomic_inc(&key->usage); key_ref = make_key_ref(key, 1); break; case KEY_SPEC_USER_SESSION_KEYRING: - if (!t->user->session_keyring) { + if (!cred->user->session_keyring) { ret = install_user_keyrings(); if (ret < 0) goto error; } - key = t->user->session_keyring; + key = cred->user->session_keyring; atomic_inc(&key->usage); key_ref = make_key_ref(key, 1); break; @@ -670,7 +674,7 @@ key_ref_t lookup_user_key(key_serial_t id, int create, int partial, goto error; case KEY_SPEC_REQKEY_AUTH_KEY: - key = t->request_key_auth; + key = cred->request_key_auth; if (!key) goto error; @@ -679,19 +683,19 @@ key_ref_t lookup_user_key(key_serial_t id, int create, int partial, break; case KEY_SPEC_REQUESTOR_KEYRING: - if (!t->request_key_auth) + if (!cred->request_key_auth) goto error; - down_read(&t->request_key_auth->sem); - if (t->request_key_auth->flags & KEY_FLAG_REVOKED) { + down_read(&cred->request_key_auth->sem); + if (cred->request_key_auth->flags & KEY_FLAG_REVOKED) { key_ref = ERR_PTR(-EKEYREVOKED); key = NULL; } else { - rka = t->request_key_auth->payload.data; + rka = cred->request_key_auth->payload.data; key = rka->dest_keyring; atomic_inc(&key->usage); } - up_read(&t->request_key_auth->sem); + up_read(&cred->request_key_auth->sem); if (!key) goto error; key_ref = make_key_ref(key, 1); @@ -791,7 +795,7 @@ long join_session_keyring(const char *name) keyring = find_keyring_by_name(name, false); if (PTR_ERR(keyring) == -ENOKEY) { /* not found - try and create a new one */ - keyring = keyring_alloc(name, tsk->uid, tsk->gid, tsk, + keyring = keyring_alloc(name, tsk->cred->uid, tsk->cred->gid, tsk, KEY_ALLOC_IN_QUOTA, NULL); if (IS_ERR(keyring)) { ret = PTR_ERR(keyring); diff --git a/security/keys/request_key.c b/security/keys/request_key.c index 8e9d93b4a40..3e9b9eb1dd2 100644 --- a/security/keys/request_key.c +++ b/security/keys/request_key.c @@ -104,7 +104,8 @@ static int call_sbin_request_key(struct key_construction *cons, /* we specify the process's default keyrings */ sprintf(keyring_str[0], "%d", - tsk->thread_keyring ? tsk->thread_keyring->serial : 0); + tsk->cred->thread_keyring ? + tsk->cred->thread_keyring->serial : 0); prkey = 0; if (tsk->signal->process_keyring) @@ -117,7 +118,7 @@ static int call_sbin_request_key(struct key_construction *cons, sskey = rcu_dereference(tsk->signal->session_keyring)->serial; rcu_read_unlock(); } else { - sskey = tsk->user->session_keyring->serial; + sskey = tsk->cred->user->session_keyring->serial; } sprintf(keyring_str[2], "%d", sskey); @@ -232,11 +233,11 @@ static void construct_get_dest_keyring(struct key **_dest_keyring) } else { /* use a default keyring; falling through the cases until we * find one that we actually have */ - switch (tsk->jit_keyring) { + switch (tsk->cred->jit_keyring) { case KEY_REQKEY_DEFL_DEFAULT: case KEY_REQKEY_DEFL_REQUESTOR_KEYRING: - if (tsk->request_key_auth) { - authkey = tsk->request_key_auth; + if (tsk->cred->request_key_auth) { + authkey = tsk->cred->request_key_auth; down_read(&authkey->sem); rka = authkey->payload.data; if (!test_bit(KEY_FLAG_REVOKED, @@ -249,7 +250,7 @@ static void construct_get_dest_keyring(struct key **_dest_keyring) } case KEY_REQKEY_DEFL_THREAD_KEYRING: - dest_keyring = key_get(tsk->thread_keyring); + dest_keyring = key_get(tsk->cred->thread_keyring); if (dest_keyring) break; @@ -268,11 +269,12 @@ static void construct_get_dest_keyring(struct key **_dest_keyring) break; case KEY_REQKEY_DEFL_USER_SESSION_KEYRING: - dest_keyring = key_get(tsk->user->session_keyring); + dest_keyring = + key_get(tsk->cred->user->session_keyring); break; case KEY_REQKEY_DEFL_USER_KEYRING: - dest_keyring = key_get(tsk->user->uid_keyring); + dest_keyring = key_get(tsk->cred->user->uid_keyring); break; case KEY_REQKEY_DEFL_GROUP_KEYRING: diff --git a/security/keys/request_key_auth.c b/security/keys/request_key_auth.c index 1762d44711d..2125579d5d7 100644 --- a/security/keys/request_key_auth.c +++ b/security/keys/request_key_auth.c @@ -164,22 +164,22 @@ struct key *request_key_auth_new(struct key *target, const void *callout_info, /* see if the calling process is already servicing the key request of * another process */ - if (current->request_key_auth) { + if (current->cred->request_key_auth) { /* it is - use that instantiation context here too */ - down_read(¤t->request_key_auth->sem); + down_read(¤t->cred->request_key_auth->sem); /* if the auth key has been revoked, then the key we're * servicing is already instantiated */ if (test_bit(KEY_FLAG_REVOKED, - ¤t->request_key_auth->flags)) + ¤t->cred->request_key_auth->flags)) goto auth_key_revoked; - irka = current->request_key_auth->payload.data; + irka = current->cred->request_key_auth->payload.data; rka->context = irka->context; rka->pid = irka->pid; get_task_struct(rka->context); - up_read(¤t->request_key_auth->sem); + up_read(¤t->cred->request_key_auth->sem); } else { /* it isn't - use this process as the context */ @@ -214,7 +214,7 @@ struct key *request_key_auth_new(struct key *target, const void *callout_info, return authkey; auth_key_revoked: - up_read(¤t->request_key_auth->sem); + up_read(¤t->cred->request_key_auth->sem); kfree(rka->callout_info); kfree(rka); kleave("= -EKEYREVOKED"); diff --git a/security/selinux/exports.c b/security/selinux/exports.c index 64af2d3409e..cf02490cd1e 100644 --- a/security/selinux/exports.c +++ b/security/selinux/exports.c @@ -39,7 +39,7 @@ EXPORT_SYMBOL_GPL(selinux_string_to_sid); int selinux_secmark_relabel_packet_permission(u32 sid) { if (selinux_enabled) { - struct task_security_struct *tsec = current->security; + struct task_security_struct *tsec = current->cred->security; return avc_has_perm(tsec->sid, sid, SECCLASS_PACKET, PACKET__RELABELTO, NULL); diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 9f6da154cc8..328308f2882 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -167,21 +167,21 @@ static int task_alloc_security(struct task_struct *task) return -ENOMEM; tsec->osid = tsec->sid = SECINITSID_UNLABELED; - task->security = tsec; + task->cred->security = tsec; return 0; } static void task_free_security(struct task_struct *task) { - struct task_security_struct *tsec = task->security; - task->security = NULL; + struct task_security_struct *tsec = task->cred->security; + task->cred->security = NULL; kfree(tsec); } static int inode_alloc_security(struct inode *inode) { - struct task_security_struct *tsec = current->security; + struct task_security_struct *tsec = current->cred->security; struct inode_security_struct *isec; isec = kmem_cache_zalloc(sel_inode_cache, GFP_NOFS); @@ -215,7 +215,7 @@ static void inode_free_security(struct inode *inode) static int file_alloc_security(struct file *file) { - struct task_security_struct *tsec = current->security; + struct task_security_struct *tsec = current->cred->security; struct file_security_struct *fsec; fsec = kzalloc(sizeof(struct file_security_struct), GFP_KERNEL); @@ -554,7 +554,7 @@ static int selinux_set_mnt_opts(struct super_block *sb, struct security_mnt_opts *opts) { int rc = 0, i; - struct task_security_struct *tsec = current->security; + struct task_security_struct *tsec = current->cred->security; struct superblock_security_struct *sbsec = sb->s_security; const char *name = sb->s_type->name; struct inode *inode = sbsec->sb->s_root->d_inode; @@ -1353,8 +1353,8 @@ static int task_has_perm(struct task_struct *tsk1, { struct task_security_struct *tsec1, *tsec2; - tsec1 = tsk1->security; - tsec2 = tsk2->security; + tsec1 = tsk1->cred->security; + tsec2 = tsk2->cred->security; return avc_has_perm(tsec1->sid, tsec2->sid, SECCLASS_PROCESS, perms, NULL); } @@ -1374,7 +1374,7 @@ static int task_has_capability(struct task_struct *tsk, u32 av = CAP_TO_MASK(cap); int rc; - tsec = tsk->security; + tsec = tsk->cred->security; AVC_AUDIT_DATA_INIT(&ad, CAP); ad.tsk = tsk; @@ -1405,7 +1405,7 @@ static int task_has_system(struct task_struct *tsk, { struct task_security_struct *tsec; - tsec = tsk->security; + tsec = tsk->cred->security; return avc_has_perm(tsec->sid, SECINITSID_KERNEL, SECCLASS_SYSTEM, perms, NULL); @@ -1426,7 +1426,7 @@ static int inode_has_perm(struct task_struct *tsk, if (unlikely(IS_PRIVATE(inode))) return 0; - tsec = tsk->security; + tsec = tsk->cred->security; isec = inode->i_security; if (!adp) { @@ -1466,7 +1466,7 @@ static int file_has_perm(struct task_struct *tsk, struct file *file, u32 av) { - struct task_security_struct *tsec = tsk->security; + struct task_security_struct *tsec = tsk->cred->security; struct file_security_struct *fsec = file->f_security; struct inode *inode = file->f_path.dentry->d_inode; struct avc_audit_data ad; @@ -1503,7 +1503,7 @@ static int may_create(struct inode *dir, struct avc_audit_data ad; int rc; - tsec = current->security; + tsec = current->cred->security; dsec = dir->i_security; sbsec = dir->i_sb->s_security; @@ -1540,7 +1540,7 @@ static int may_create_key(u32 ksid, { struct task_security_struct *tsec; - tsec = ctx->security; + tsec = ctx->cred->security; return avc_has_perm(tsec->sid, ksid, SECCLASS_KEY, KEY__CREATE, NULL); } @@ -1561,7 +1561,7 @@ static int may_link(struct inode *dir, u32 av; int rc; - tsec = current->security; + tsec = current->cred->security; dsec = dir->i_security; isec = dentry->d_inode->i_security; @@ -1606,7 +1606,7 @@ static inline int may_rename(struct inode *old_dir, int old_is_dir, new_is_dir; int rc; - tsec = current->security; + tsec = current->cred->security; old_dsec = old_dir->i_security; old_isec = old_dentry->d_inode->i_security; old_is_dir = S_ISDIR(old_dentry->d_inode->i_mode); @@ -1659,7 +1659,7 @@ static int superblock_has_perm(struct task_struct *tsk, struct task_security_struct *tsec; struct superblock_security_struct *sbsec; - tsec = tsk->security; + tsec = tsk->cred->security; sbsec = sb->s_security; return avc_has_perm(tsec->sid, sbsec->sid, SECCLASS_FILESYSTEM, perms, ad); @@ -1758,8 +1758,8 @@ static int selinux_ptrace_may_access(struct task_struct *child, return rc; if (mode == PTRACE_MODE_READ) { - struct task_security_struct *tsec = current->security; - struct task_security_struct *csec = child->security; + struct task_security_struct *tsec = current->cred->security; + struct task_security_struct *csec = child->cred->security; return avc_has_perm(tsec->sid, csec->sid, SECCLASS_FILE, FILE__READ, NULL); } @@ -1874,7 +1874,7 @@ static int selinux_sysctl(ctl_table *table, int op) if (rc) return rc; - tsec = current->security; + tsec = current->cred->security; rc = selinux_sysctl_get_sid(table, (op == 0001) ? SECCLASS_DIR : SECCLASS_FILE, &tsid); @@ -2025,7 +2025,7 @@ static int selinux_bprm_set_security(struct linux_binprm *bprm) if (bsec->set) return 0; - tsec = current->security; + tsec = current->cred->security; isec = inode->i_security; /* Default to the current task SID. */ @@ -2090,7 +2090,7 @@ static int selinux_bprm_check_security(struct linux_binprm *bprm) static int selinux_bprm_secureexec(struct linux_binprm *bprm) { - struct task_security_struct *tsec = current->security; + struct task_security_struct *tsec = current->cred->security; int atsecure = 0; if (tsec->osid != tsec->sid) { @@ -2214,7 +2214,7 @@ static void selinux_bprm_apply_creds(struct linux_binprm *bprm, int unsafe) secondary_ops->bprm_apply_creds(bprm, unsafe); - tsec = current->security; + tsec = current->cred->security; bsec = bprm->security; sid = bsec->sid; @@ -2243,7 +2243,7 @@ static void selinux_bprm_apply_creds(struct linux_binprm *bprm, int unsafe) rcu_read_lock(); tracer = tracehook_tracer_task(current); if (likely(tracer != NULL)) { - sec = tracer->security; + sec = tracer->cred->security; ptsid = sec->sid; } rcu_read_unlock(); @@ -2274,7 +2274,7 @@ static void selinux_bprm_post_apply_creds(struct linux_binprm *bprm) int rc, i; unsigned long flags; - tsec = current->security; + tsec = current->cred->security; bsec = bprm->security; if (bsec->unsafe) { @@ -2521,7 +2521,7 @@ static int selinux_inode_init_security(struct inode *inode, struct inode *dir, int rc; char *namep = NULL, *context; - tsec = current->security; + tsec = current->cred->security; dsec = dir->i_security; sbsec = dir->i_sb->s_security; @@ -2706,7 +2706,7 @@ static int selinux_inode_setotherxattr(struct dentry *dentry, const char *name) static int selinux_inode_setxattr(struct dentry *dentry, const char *name, const void *value, size_t size, int flags) { - struct task_security_struct *tsec = current->security; + struct task_security_struct *tsec = current->cred->security; struct inode *inode = dentry->d_inode; struct inode_security_struct *isec = inode->i_security; struct superblock_security_struct *sbsec; @@ -2918,7 +2918,7 @@ static int selinux_revalidate_file_permission(struct file *file, int mask) static int selinux_file_permission(struct file *file, int mask) { struct inode *inode = file->f_path.dentry->d_inode; - struct task_security_struct *tsec = current->security; + struct task_security_struct *tsec = current->cred->security; struct file_security_struct *fsec = file->f_security; struct inode_security_struct *isec = inode->i_security; @@ -2995,7 +2995,8 @@ static int selinux_file_mmap(struct file *file, unsigned long reqprot, unsigned long addr, unsigned long addr_only) { int rc = 0; - u32 sid = ((struct task_security_struct *)(current->security))->sid; + u32 sid = ((struct task_security_struct *) + (current->cred->security))->sid; if (addr < mmap_min_addr) rc = avc_has_perm(sid, sid, SECCLASS_MEMPROTECT, @@ -3107,7 +3108,7 @@ static int selinux_file_set_fowner(struct file *file) struct task_security_struct *tsec; struct file_security_struct *fsec; - tsec = current->security; + tsec = current->cred->security; fsec = file->f_security; fsec->fown_sid = tsec->sid; @@ -3125,7 +3126,7 @@ static int selinux_file_send_sigiotask(struct task_struct *tsk, /* struct fown_struct is never outside the context of a struct file */ file = container_of(fown, struct file, f_owner); - tsec = tsk->security; + tsec = tsk->cred->security; fsec = file->f_security; if (!signum) @@ -3188,12 +3189,12 @@ static int selinux_task_alloc_security(struct task_struct *tsk) struct task_security_struct *tsec1, *tsec2; int rc; - tsec1 = current->security; + tsec1 = current->cred->security; rc = task_alloc_security(tsk); if (rc) return rc; - tsec2 = tsk->security; + tsec2 = tsk->cred->security; tsec2->osid = tsec1->osid; tsec2->sid = tsec1->sid; @@ -3251,7 +3252,7 @@ static int selinux_task_getsid(struct task_struct *p) static void selinux_task_getsecid(struct task_struct *p, u32 *secid) { - struct task_security_struct *tsec = p->security; + struct task_security_struct *tsec = p->cred->security; *secid = tsec->sid; } @@ -3343,7 +3344,7 @@ static int selinux_task_kill(struct task_struct *p, struct siginfo *info, perm = PROCESS__SIGNULL; /* null signal; existence test */ else perm = signal_to_av(sig); - tsec = p->security; + tsec = p->cred->security; if (secid) rc = avc_has_perm(secid, tsec->sid, SECCLASS_PROCESS, perm, NULL); else @@ -3375,7 +3376,7 @@ static void selinux_task_reparent_to_init(struct task_struct *p) secondary_ops->task_reparent_to_init(p); - tsec = p->security; + tsec = p->cred->security; tsec->osid = tsec->sid; tsec->sid = SECINITSID_KERNEL; return; @@ -3384,7 +3385,7 @@ static void selinux_task_reparent_to_init(struct task_struct *p) static void selinux_task_to_inode(struct task_struct *p, struct inode *inode) { - struct task_security_struct *tsec = p->security; + struct task_security_struct *tsec = p->cred->security; struct inode_security_struct *isec = inode->i_security; isec->sid = tsec->sid; @@ -3632,7 +3633,7 @@ static int socket_has_perm(struct task_struct *task, struct socket *sock, struct avc_audit_data ad; int err = 0; - tsec = task->security; + tsec = task->cred->security; isec = SOCK_INODE(sock)->i_security; if (isec->sid == SECINITSID_KERNEL) @@ -3656,7 +3657,7 @@ static int selinux_socket_create(int family, int type, if (kern) goto out; - tsec = current->security; + tsec = current->cred->security; newsid = tsec->sockcreate_sid ? : tsec->sid; err = avc_has_perm(tsec->sid, newsid, socket_type_to_security_class(family, type, @@ -3677,7 +3678,7 @@ static int selinux_socket_post_create(struct socket *sock, int family, isec = SOCK_INODE(sock)->i_security; - tsec = current->security; + tsec = current->cred->security; newsid = tsec->sockcreate_sid ? : tsec->sid; isec->sclass = socket_type_to_security_class(family, type, protocol); isec->sid = kern ? SECINITSID_KERNEL : newsid; @@ -3723,7 +3724,7 @@ static int selinux_socket_bind(struct socket *sock, struct sockaddr *address, in struct sock *sk = sock->sk; u32 sid, node_perm; - tsec = current->security; + tsec = current->cred->security; isec = SOCK_INODE(sock)->i_security; if (family == PF_INET) { @@ -4764,7 +4765,7 @@ static int ipc_alloc_security(struct task_struct *task, struct kern_ipc_perm *perm, u16 sclass) { - struct task_security_struct *tsec = task->security; + struct task_security_struct *tsec = task->cred->security; struct ipc_security_struct *isec; isec = kzalloc(sizeof(struct ipc_security_struct), GFP_KERNEL); @@ -4814,7 +4815,7 @@ static int ipc_has_perm(struct kern_ipc_perm *ipc_perms, struct ipc_security_struct *isec; struct avc_audit_data ad; - tsec = current->security; + tsec = current->cred->security; isec = ipc_perms->security; AVC_AUDIT_DATA_INIT(&ad, IPC); @@ -4845,7 +4846,7 @@ static int selinux_msg_queue_alloc_security(struct msg_queue *msq) if (rc) return rc; - tsec = current->security; + tsec = current->cred->security; isec = msq->q_perm.security; AVC_AUDIT_DATA_INIT(&ad, IPC); @@ -4871,7 +4872,7 @@ static int selinux_msg_queue_associate(struct msg_queue *msq, int msqflg) struct ipc_security_struct *isec; struct avc_audit_data ad; - tsec = current->security; + tsec = current->cred->security; isec = msq->q_perm.security; AVC_AUDIT_DATA_INIT(&ad, IPC); @@ -4917,7 +4918,7 @@ static int selinux_msg_queue_msgsnd(struct msg_queue *msq, struct msg_msg *msg, struct avc_audit_data ad; int rc; - tsec = current->security; + tsec = current->cred->security; isec = msq->q_perm.security; msec = msg->security; @@ -4965,7 +4966,7 @@ static int selinux_msg_queue_msgrcv(struct msg_queue *msq, struct msg_msg *msg, struct avc_audit_data ad; int rc; - tsec = target->security; + tsec = target->cred->security; isec = msq->q_perm.security; msec = msg->security; @@ -4992,7 +4993,7 @@ static int selinux_shm_alloc_security(struct shmid_kernel *shp) if (rc) return rc; - tsec = current->security; + tsec = current->cred->security; isec = shp->shm_perm.security; AVC_AUDIT_DATA_INIT(&ad, IPC); @@ -5018,7 +5019,7 @@ static int selinux_shm_associate(struct shmid_kernel *shp, int shmflg) struct ipc_security_struct *isec; struct avc_audit_data ad; - tsec = current->security; + tsec = current->cred->security; isec = shp->shm_perm.security; AVC_AUDIT_DATA_INIT(&ad, IPC); @@ -5091,7 +5092,7 @@ static int selinux_sem_alloc_security(struct sem_array *sma) if (rc) return rc; - tsec = current->security; + tsec = current->cred->security; isec = sma->sem_perm.security; AVC_AUDIT_DATA_INIT(&ad, IPC); @@ -5117,7 +5118,7 @@ static int selinux_sem_associate(struct sem_array *sma, int semflg) struct ipc_security_struct *isec; struct avc_audit_data ad; - tsec = current->security; + tsec = current->cred->security; isec = sma->sem_perm.security; AVC_AUDIT_DATA_INIT(&ad, IPC); @@ -5224,7 +5225,7 @@ static int selinux_getprocattr(struct task_struct *p, return error; } - tsec = p->security; + tsec = p->cred->security; if (!strcmp(name, "current")) sid = tsec->sid; @@ -5308,7 +5309,7 @@ static int selinux_setprocattr(struct task_struct *p, operation. See selinux_bprm_set_security for the execve checks and may_create for the file creation checks. The operation will then fail if the context is not permitted. */ - tsec = p->security; + tsec = p->cred->security; if (!strcmp(name, "exec")) tsec->exec_sid = sid; else if (!strcmp(name, "fscreate")) @@ -5361,7 +5362,8 @@ boundary_ok: rcu_read_lock(); tracer = tracehook_tracer_task(p); if (tracer != NULL) { - struct task_security_struct *ptsec = tracer->security; + struct task_security_struct *ptsec = + tracer->cred->security; u32 ptsid = ptsec->sid; rcu_read_unlock(); error = avc_has_perm_noaudit(ptsid, sid, @@ -5405,7 +5407,7 @@ static void selinux_release_secctx(char *secdata, u32 seclen) static int selinux_key_alloc(struct key *k, struct task_struct *tsk, unsigned long flags) { - struct task_security_struct *tsec = tsk->security; + struct task_security_struct *tsec = tsk->cred->security; struct key_security_struct *ksec; ksec = kzalloc(sizeof(struct key_security_struct), GFP_KERNEL); @@ -5439,7 +5441,7 @@ static int selinux_key_permission(key_ref_t key_ref, key = key_ref_to_ptr(key_ref); - tsec = ctx->security; + tsec = ctx->cred->security; ksec = key->security; /* if no specific permissions are requested, we skip the @@ -5683,7 +5685,7 @@ static __init int selinux_init(void) /* Set the security state for the initial task. */ if (task_alloc_security(current)) panic("SELinux: Failed to initialize initial task.\n"); - tsec = current->security; + tsec = current->cred->security; tsec->osid = tsec->sid = SECINITSID_KERNEL; sel_inode_cache = kmem_cache_create("selinux_inode_security", diff --git a/security/selinux/selinuxfs.c b/security/selinux/selinuxfs.c index 69c9dccc8cf..10715d1330b 100644 --- a/security/selinux/selinuxfs.c +++ b/security/selinux/selinuxfs.c @@ -97,7 +97,7 @@ static int task_has_security(struct task_struct *tsk, { struct task_security_struct *tsec; - tsec = tsk->security; + tsec = tsk->cred->security; if (!tsec) return -EACCES; diff --git a/security/selinux/xfrm.c b/security/selinux/xfrm.c index 8f17f542a11..d7db76617b0 100644 --- a/security/selinux/xfrm.c +++ b/security/selinux/xfrm.c @@ -197,7 +197,7 @@ static int selinux_xfrm_sec_ctx_alloc(struct xfrm_sec_ctx **ctxp, struct xfrm_user_sec_ctx *uctx, u32 sid) { int rc = 0; - struct task_security_struct *tsec = current->security; + struct task_security_struct *tsec = current->cred->security; struct xfrm_sec_ctx *ctx = NULL; char *ctx_str = NULL; u32 str_len; @@ -333,7 +333,7 @@ void selinux_xfrm_policy_free(struct xfrm_sec_ctx *ctx) */ int selinux_xfrm_policy_delete(struct xfrm_sec_ctx *ctx) { - struct task_security_struct *tsec = current->security; + struct task_security_struct *tsec = current->cred->security; int rc = 0; if (ctx) { @@ -378,7 +378,7 @@ void selinux_xfrm_state_free(struct xfrm_state *x) */ int selinux_xfrm_state_delete(struct xfrm_state *x) { - struct task_security_struct *tsec = current->security; + struct task_security_struct *tsec = current->cred->security; struct xfrm_sec_ctx *ctx = x->security; int rc = 0; diff --git a/security/smack/smack_access.c b/security/smack/smack_access.c index 79ff21ed4c3..b6dd4fc0fb0 100644 --- a/security/smack/smack_access.c +++ b/security/smack/smack_access.c @@ -164,7 +164,7 @@ int smk_curacc(char *obj_label, u32 mode) { int rc; - rc = smk_access(current->security, obj_label, mode); + rc = smk_access(current->cred->security, obj_label, mode); if (rc == 0) return 0; @@ -173,7 +173,7 @@ int smk_curacc(char *obj_label, u32 mode) * only one that gets privilege and current does not * have that label. */ - if (smack_onlycap != NULL && smack_onlycap != current->security) + if (smack_onlycap != NULL && smack_onlycap != current->cred->security) return rc; if (capable(CAP_MAC_OVERRIDE)) diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c index 6e2dc0bab70..791da238d04 100644 --- a/security/smack/smack_lsm.c +++ b/security/smack/smack_lsm.c @@ -102,7 +102,8 @@ static int smack_ptrace_may_access(struct task_struct *ctp, unsigned int mode) if (rc != 0) return rc; - rc = smk_access(current->security, ctp->security, MAY_READWRITE); + rc = smk_access(current->cred->security, ctp->cred->security, + MAY_READWRITE); if (rc != 0 && capable(CAP_MAC_OVERRIDE)) return 0; return rc; @@ -124,7 +125,8 @@ static int smack_ptrace_traceme(struct task_struct *ptp) if (rc != 0) return rc; - rc = smk_access(ptp->security, current->security, MAY_READWRITE); + rc = smk_access(ptp->cred->security, current->cred->security, + MAY_READWRITE); if (rc != 0 && has_capability(ptp, CAP_MAC_OVERRIDE)) return 0; return rc; @@ -141,7 +143,7 @@ static int smack_ptrace_traceme(struct task_struct *ptp) static int smack_syslog(int type) { int rc; - char *sp = current->security; + char *sp = current->cred->security; rc = cap_syslog(type); if (rc != 0) @@ -373,7 +375,7 @@ static int smack_sb_umount(struct vfsmount *mnt, int flags) */ static int smack_inode_alloc_security(struct inode *inode) { - inode->i_security = new_inode_smack(current->security); + inode->i_security = new_inode_smack(current->cred->security); if (inode->i_security == NULL) return -ENOMEM; return 0; @@ -818,7 +820,7 @@ static int smack_file_permission(struct file *file, int mask) */ static int smack_file_alloc_security(struct file *file) { - file->f_security = current->security; + file->f_security = current->cred->security; return 0; } @@ -916,7 +918,7 @@ static int smack_file_fcntl(struct file *file, unsigned int cmd, */ static int smack_file_set_fowner(struct file *file) { - file->f_security = current->security; + file->f_security = current->cred->security; return 0; } @@ -941,7 +943,7 @@ static int smack_file_send_sigiotask(struct task_struct *tsk, * struct fown_struct is never outside the context of a struct file */ file = container_of(fown, struct file, f_owner); - rc = smk_access(file->f_security, tsk->security, MAY_WRITE); + rc = smk_access(file->f_security, tsk->cred->security, MAY_WRITE); if (rc != 0 && has_capability(tsk, CAP_MAC_OVERRIDE)) return 0; return rc; @@ -984,7 +986,7 @@ static int smack_file_receive(struct file *file) */ static int smack_task_alloc_security(struct task_struct *tsk) { - tsk->security = current->security; + tsk->cred->security = current->cred->security; return 0; } @@ -999,7 +1001,7 @@ static int smack_task_alloc_security(struct task_struct *tsk) */ static void smack_task_free_security(struct task_struct *task) { - task->security = NULL; + task->cred->security = NULL; } /** @@ -1011,7 +1013,7 @@ static void smack_task_free_security(struct task_struct *task) */ static int smack_task_setpgid(struct task_struct *p, pid_t pgid) { - return smk_curacc(p->security, MAY_WRITE); + return smk_curacc(p->cred->security, MAY_WRITE); } /** @@ -1022,7 +1024,7 @@ static int smack_task_setpgid(struct task_struct *p, pid_t pgid) */ static int smack_task_getpgid(struct task_struct *p) { - return smk_curacc(p->security, MAY_READ); + return smk_curacc(p->cred->security, MAY_READ); } /** @@ -1033,7 +1035,7 @@ static int smack_task_getpgid(struct task_struct *p) */ static int smack_task_getsid(struct task_struct *p) { - return smk_curacc(p->security, MAY_READ); + return smk_curacc(p->cred->security, MAY_READ); } /** @@ -1045,7 +1047,7 @@ static int smack_task_getsid(struct task_struct *p) */ static void smack_task_getsecid(struct task_struct *p, u32 *secid) { - *secid = smack_to_secid(p->security); + *secid = smack_to_secid(p->cred->security); } /** @@ -1061,7 +1063,7 @@ static int smack_task_setnice(struct task_struct *p, int nice) rc = cap_task_setnice(p, nice); if (rc == 0) - rc = smk_curacc(p->security, MAY_WRITE); + rc = smk_curacc(p->cred->security, MAY_WRITE); return rc; } @@ -1078,7 +1080,7 @@ static int smack_task_setioprio(struct task_struct *p, int ioprio) rc = cap_task_setioprio(p, ioprio); if (rc == 0) - rc = smk_curacc(p->security, MAY_WRITE); + rc = smk_curacc(p->cred->security, MAY_WRITE); return rc; } @@ -1090,7 +1092,7 @@ static int smack_task_setioprio(struct task_struct *p, int ioprio) */ static int smack_task_getioprio(struct task_struct *p) { - return smk_curacc(p->security, MAY_READ); + return smk_curacc(p->cred->security, MAY_READ); } /** @@ -1108,7 +1110,7 @@ static int smack_task_setscheduler(struct task_struct *p, int policy, rc = cap_task_setscheduler(p, policy, lp); if (rc == 0) - rc = smk_curacc(p->security, MAY_WRITE); + rc = smk_curacc(p->cred->security, MAY_WRITE); return rc; } @@ -1120,7 +1122,7 @@ static int smack_task_setscheduler(struct task_struct *p, int policy, */ static int smack_task_getscheduler(struct task_struct *p) { - return smk_curacc(p->security, MAY_READ); + return smk_curacc(p->cred->security, MAY_READ); } /** @@ -1131,7 +1133,7 @@ static int smack_task_getscheduler(struct task_struct *p) */ static int smack_task_movememory(struct task_struct *p) { - return smk_curacc(p->security, MAY_WRITE); + return smk_curacc(p->cred->security, MAY_WRITE); } /** @@ -1154,13 +1156,13 @@ static int smack_task_kill(struct task_struct *p, struct siginfo *info, * can write the receiver. */ if (secid == 0) - return smk_curacc(p->security, MAY_WRITE); + return smk_curacc(p->cred->security, MAY_WRITE); /* * If the secid isn't 0 we're dealing with some USB IO * specific behavior. This is not clean. For one thing * we can't take privilege into account. */ - return smk_access(smack_from_secid(secid), p->security, MAY_WRITE); + return smk_access(smack_from_secid(secid), p->cred->security, MAY_WRITE); } /** @@ -1173,7 +1175,7 @@ static int smack_task_wait(struct task_struct *p) { int rc; - rc = smk_access(current->security, p->security, MAY_WRITE); + rc = smk_access(current->cred->security, p->cred->security, MAY_WRITE); if (rc == 0) return 0; @@ -1204,7 +1206,7 @@ static int smack_task_wait(struct task_struct *p) static void smack_task_to_inode(struct task_struct *p, struct inode *inode) { struct inode_smack *isp = inode->i_security; - isp->smk_inode = p->security; + isp->smk_inode = p->cred->security; } /* @@ -1223,7 +1225,7 @@ static void smack_task_to_inode(struct task_struct *p, struct inode *inode) */ static int smack_sk_alloc_security(struct sock *sk, int family, gfp_t gfp_flags) { - char *csp = current->security; + char *csp = current->cred->security; struct socket_smack *ssp; ssp = kzalloc(sizeof(struct socket_smack), gfp_flags); @@ -1448,7 +1450,7 @@ static int smack_flags_to_may(int flags) */ static int smack_msg_msg_alloc_security(struct msg_msg *msg) { - msg->security = current->security; + msg->security = current->cred->security; return 0; } @@ -1484,7 +1486,7 @@ static int smack_shm_alloc_security(struct shmid_kernel *shp) { struct kern_ipc_perm *isp = &shp->shm_perm; - isp->security = current->security; + isp->security = current->cred->security; return 0; } @@ -1593,7 +1595,7 @@ static int smack_sem_alloc_security(struct sem_array *sma) { struct kern_ipc_perm *isp = &sma->sem_perm; - isp->security = current->security; + isp->security = current->cred->security; return 0; } @@ -1697,7 +1699,7 @@ static int smack_msg_queue_alloc_security(struct msg_queue *msq) { struct kern_ipc_perm *kisp = &msq->q_perm; - kisp->security = current->security; + kisp->security = current->cred->security; return 0; } @@ -1852,7 +1854,7 @@ static void smack_d_instantiate(struct dentry *opt_dentry, struct inode *inode) struct super_block *sbp; struct superblock_smack *sbsp; struct inode_smack *isp; - char *csp = current->security; + char *csp = current->cred->security; char *fetched; char *final; struct dentry *dp; @@ -2009,7 +2011,7 @@ static int smack_getprocattr(struct task_struct *p, char *name, char **value) if (strcmp(name, "current") != 0) return -EINVAL; - cp = kstrdup(p->security, GFP_KERNEL); + cp = kstrdup(p->cred->security, GFP_KERNEL); if (cp == NULL) return -ENOMEM; @@ -2055,7 +2057,7 @@ static int smack_setprocattr(struct task_struct *p, char *name, if (newsmack == NULL) return -EINVAL; - p->security = newsmack; + p->cred->security = newsmack; return size; } @@ -2288,8 +2290,8 @@ static void smack_sock_graft(struct sock *sk, struct socket *parent) return; ssp = sk->sk_security; - ssp->smk_in = current->security; - ssp->smk_out = current->security; + ssp->smk_in = current->cred->security; + ssp->smk_out = current->cred->security; ssp->smk_packet[0] = '\0'; rc = smack_netlabel(sk); @@ -2362,7 +2364,7 @@ static int smack_inet_conn_request(struct sock *sk, struct sk_buff *skb, static int smack_key_alloc(struct key *key, struct task_struct *tsk, unsigned long flags) { - key->security = tsk->security; + key->security = tsk->cred->security; return 0; } @@ -2403,10 +2405,11 @@ static int smack_key_permission(key_ref_t key_ref, /* * This should not occur */ - if (context->security == NULL) + if (context->cred->security == NULL) return -EACCES; - return smk_access(context->security, keyp->security, MAY_READWRITE); + return smk_access(context->cred->security, keyp->security, + MAY_READWRITE); } #endif /* CONFIG_KEYS */ @@ -2726,7 +2729,7 @@ static __init int smack_init(void) /* * Set the security state for the initial task. */ - current->security = &smack_known_floor.smk_known; + current->cred->security = &smack_known_floor.smk_known; /* * Initialize locks diff --git a/security/smack/smackfs.c b/security/smack/smackfs.c index c21d8c8bf0c..c5ca279e050 100644 --- a/security/smack/smackfs.c +++ b/security/smack/smackfs.c @@ -336,7 +336,7 @@ static void smk_cipso_doi(void) audit_info.loginuid = audit_get_loginuid(current); audit_info.sessionid = audit_get_sessionid(current); - audit_info.secid = smack_to_secid(current->security); + audit_info.secid = smack_to_secid(current->cred->security); rc = netlbl_cfg_map_del(NULL, &audit_info); if (rc != 0) @@ -371,7 +371,7 @@ static void smk_unlbl_ambient(char *oldambient) audit_info.loginuid = audit_get_loginuid(current); audit_info.sessionid = audit_get_sessionid(current); - audit_info.secid = smack_to_secid(current->security); + audit_info.secid = smack_to_secid(current->cred->security); if (oldambient != NULL) { rc = netlbl_cfg_map_del(oldambient, &audit_info); @@ -843,7 +843,7 @@ static ssize_t smk_write_onlycap(struct file *file, const char __user *buf, size_t count, loff_t *ppos) { char in[SMK_LABELLEN]; - char *sp = current->security; + char *sp = current->cred->security; if (!capable(CAP_MAC_ADMIN)) return -EPERM; -- cgit v1.2.3-70-g09d2 From eeda3fd64f75bcbfaa70ce946513abaf3f23b8e0 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 19 Nov 2008 21:40:23 -0800 Subject: netdev: introduce dev_get_stats() In order for the network device ops get_stats call to be immutable, the handling of the default internal network device stats block has to be changed. Add a new helper function which replaces the old use of internal_get_stats. Note: change return code to make it clear that the caller should not go changing the returned statistics. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- arch/s390/appldata/appldata_net_sum.c | 4 ++-- drivers/net/bonding/bond_main.c | 5 +++-- drivers/net/sfc/ethtool.c | 2 +- drivers/parisc/led.c | 4 ++-- include/linux/netdevice.h | 4 +++- net/core/dev.c | 23 ++++++++++++++++++----- net/core/net-sysfs.c | 3 +-- net/core/rtnetlink.c | 6 +++--- 8 files changed, 33 insertions(+), 18 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/appldata/appldata_net_sum.c b/arch/s390/appldata/appldata_net_sum.c index 3b746556e1a..fa741f84c5b 100644 --- a/arch/s390/appldata/appldata_net_sum.c +++ b/arch/s390/appldata/appldata_net_sum.c @@ -67,7 +67,6 @@ static void appldata_get_net_sum_data(void *data) int i; struct appldata_net_sum_data *net_data; struct net_device *dev; - struct net_device_stats *stats; unsigned long rx_packets, tx_packets, rx_bytes, tx_bytes, rx_errors, tx_errors, rx_dropped, tx_dropped, collisions; @@ -86,7 +85,8 @@ static void appldata_get_net_sum_data(void *data) collisions = 0; read_lock(&dev_base_lock); for_each_netdev(&init_net, dev) { - stats = dev->get_stats(dev); + const struct net_device_stats *stats = dev_get_stats(dev); + rx_packets += stats->rx_packets; tx_packets += stats->tx_packets; rx_bytes += stats->rx_bytes; diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index a08ea480805..db5f5c24a25 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -3899,7 +3899,7 @@ static int bond_close(struct net_device *bond_dev) static struct net_device_stats *bond_get_stats(struct net_device *bond_dev) { struct bonding *bond = netdev_priv(bond_dev); - struct net_device_stats *stats = &(bond->stats), *sstats; + struct net_device_stats *stats = &bond->stats; struct net_device_stats local_stats; struct slave *slave; int i; @@ -3909,7 +3909,8 @@ static struct net_device_stats *bond_get_stats(struct net_device *bond_dev) read_lock_bh(&bond->lock); bond_for_each_slave(bond, slave, i) { - sstats = slave->dev->get_stats(slave->dev); + const struct net_device_stats *sstats = dev_get_stats(slave->dev); + local_stats.rx_packets += sstats->rx_packets; local_stats.rx_bytes += sstats->rx_bytes; local_stats.rx_errors += sstats->rx_errors; diff --git a/drivers/net/sfc/ethtool.c b/drivers/net/sfc/ethtool.c index abd8fcd6e62..cd92c4d8dbc 100644 --- a/drivers/net/sfc/ethtool.c +++ b/drivers/net/sfc/ethtool.c @@ -426,7 +426,7 @@ static void efx_ethtool_get_stats(struct net_device *net_dev, EFX_BUG_ON_PARANOID(stats->n_stats != EFX_ETHTOOL_NUM_STATS); /* Update MAC and NIC statistics */ - net_dev->get_stats(net_dev); + dev_get_stats(net_dev); /* Fill detailed statistics buffer */ for (i = 0; i < EFX_ETHTOOL_NUM_STATS; i++) { diff --git a/drivers/parisc/led.c b/drivers/parisc/led.c index f9b12664f9f..454b6532e40 100644 --- a/drivers/parisc/led.c +++ b/drivers/parisc/led.c @@ -360,13 +360,13 @@ static __inline__ int led_get_net_activity(void) read_lock(&dev_base_lock); rcu_read_lock(); for_each_netdev(&init_net, dev) { - struct net_device_stats *stats; + const struct net_device_stats *stats; struct in_device *in_dev = __in_dev_get_rcu(dev); if (!in_dev || !in_dev->ifa_list) continue; if (ipv4_is_loopback(in_dev->ifa_list->ifa_local)) continue; - stats = dev->get_stats(dev); + stats = dev_get_stats(dev); rx_total += stats->rx_packets; tx_total += stats->tx_packets; } diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 9060f5f3517..981a089d514 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -864,9 +864,9 @@ struct net_device unsigned short vid); #ifdef CONFIG_NET_POLL_CONTROLLER void (*poll_controller)(struct net_device *dev); -#endif #endif }; +#endif }; #define to_net_dev(d) container_of(d, struct net_device, dev) @@ -1780,6 +1780,8 @@ extern void netdev_features_change(struct net_device *dev); /* Load a device via the kmod */ extern void dev_load(struct net *net, const char *name); extern void dev_mcast_init(void); +extern const struct net_device_stats *dev_get_stats(struct net_device *dev); + extern int netdev_max_backlog; extern int weight_p; extern int netdev_set_master(struct net_device *dev, struct net_device *master); diff --git a/net/core/dev.c b/net/core/dev.c index ca14ab407b3..8843f4e3f5e 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2620,7 +2620,7 @@ void dev_seq_stop(struct seq_file *seq, void *v) static void dev_seq_printf_stats(struct seq_file *seq, struct net_device *dev) { - struct net_device_stats *stats = dev->get_stats(dev); + const struct net_device_stats *stats = dev_get_stats(dev); seq_printf(seq, "%6s:%8lu %7lu %4lu %4lu %4lu %5lu %10lu %9lu " "%8lu %7lu %4lu %4lu %4lu %5lu %7lu %10lu\n", @@ -4288,10 +4288,24 @@ void netdev_run_todo(void) } } -static struct net_device_stats *internal_stats(struct net_device *dev) -{ - return &dev->stats; +/** + * dev_get_stats - get network device statistics + * @dev: device to get statistics from + * + * Get network statistics from device. The device driver may provide + * its own method by setting dev->netdev_ops->get_stats; otherwise + * the internal statistics structure is used. + */ +const struct net_device_stats *dev_get_stats(struct net_device *dev) + { + const struct net_device_ops *ops = dev->netdev_ops; + + if (ops->ndo_get_stats) + return ops->ndo_get_stats(dev); + else + return &dev->stats; } +EXPORT_SYMBOL(dev_get_stats); static void netdev_init_one_queue(struct net_device *dev, struct netdev_queue *queue, @@ -4370,7 +4384,6 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, netdev_init_queues(dev); - dev->get_stats = internal_stats; netpoll_netdev_init(dev); setup(dev); strcpy(dev->name, name); diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 146dcfeb060..afd42d71732 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -270,7 +270,6 @@ static ssize_t netstat_show(const struct device *d, unsigned long offset) { struct net_device *dev = to_net_dev(d); - struct net_device_stats *stats; ssize_t ret = -EINVAL; WARN_ON(offset > sizeof(struct net_device_stats) || @@ -278,7 +277,7 @@ static ssize_t netstat_show(const struct device *d, read_lock(&dev_base_lock); if (dev_isalive(dev)) { - stats = dev->get_stats(dev); + const struct net_device_stats *stats = dev_get_stats(dev); ret = sprintf(buf, fmt_ulong, *(unsigned long *)(((u8 *) stats) + offset)); } diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 6f8e0778e56..790dd205bb5 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -551,7 +551,7 @@ static void set_operstate(struct net_device *dev, unsigned char transition) } static void copy_rtnl_link_stats(struct rtnl_link_stats *a, - struct net_device_stats *b) + const struct net_device_stats *b) { a->rx_packets = b->rx_packets; a->tx_packets = b->tx_packets; @@ -609,7 +609,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, struct netdev_queue *txq; struct ifinfomsg *ifm; struct nlmsghdr *nlh; - struct net_device_stats *stats; + const struct net_device_stats *stats; struct nlattr *attr; nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ifm), flags); @@ -666,7 +666,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, if (attr == NULL) goto nla_put_failure; - stats = dev->get_stats(dev); + stats = dev_get_stats(dev); copy_rtnl_link_stats(nla_data(attr), stats); if (dev->rtnl_link_ops) { -- cgit v1.2.3-70-g09d2 From 5439050f9f1aa92381ab9beccbea6ddf0c687c2b Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 25 Dec 2008 13:37:57 +0100 Subject: [S390] cpu topology: fix cpu_core_map initialization Common code doesn't call arch_update_cpu_topology() anymore on cpu hotplug. But our architecture backend relied on that in order to update the cpu_core_map. For machines without cpu topology support this leads uninitialized cpu_core_maps for later on added cpus. To solve this just initialize the maps with cpu_possible_map, since that will be always valid for machines without topology support. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/topology.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c index a947899dcba..586c5e33b3f 100644 --- a/arch/s390/kernel/topology.c +++ b/arch/s390/kernel/topology.c @@ -78,7 +78,7 @@ cpumask_t cpu_coregroup_map(unsigned int cpu) cpus_clear(mask); if (!machine_has_topology) - return cpu_present_map; + return cpu_possible_map; spin_lock_irqsave(&topology_lock, flags); while (core) { if (cpu_isset(cpu, core->mask)) { @@ -168,7 +168,7 @@ static void topology_update_polarization_simple(void) int cpu; mutex_lock(&smp_cpu_state_mutex); - for_each_present_cpu(cpu) + for_each_possible_cpu(cpu) smp_cpu_polarization[cpu] = POLARIZATION_HRZ; mutex_unlock(&smp_cpu_state_mutex); } @@ -199,7 +199,7 @@ int topology_set_cpu_management(int fc) rc = ptf(PTF_HORIZONTAL); if (rc) return -EBUSY; - for_each_present_cpu(cpu) + for_each_possible_cpu(cpu) smp_cpu_polarization[cpu] = POLARIZATION_UNKNWN; return rc; } @@ -208,7 +208,7 @@ static void update_cpu_core_map(void) { int cpu; - for_each_present_cpu(cpu) + for_each_possible_cpu(cpu) cpu_core_map[cpu] = cpu_coregroup_map(cpu); } -- cgit v1.2.3-70-g09d2 From 8f2961c39eacd5d450048d8a3e91675486f3a015 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 25 Dec 2008 13:37:58 +0100 Subject: [S390] audit: get s390 ret_from_fork in sync with other architectures On s390 we have ret_from_fork jump not to the "do all work we normally do on return from syscall" as on x86, ppc, etc., but to the "do all such work except audit". Historical reasons - the codepath triggered when we have AUDIT process flag set is separated from the normall one and they converge at sysc_return, which is the common part of post-syscall work. And does not include calling audit_syscall_exit() - that's done in the end of sysc_tracesys path, just before that path jumps to sysc_return. IOW, the child returning from fork()/clone()/vfork() doesn't call audit_syscall_exit() at all, so no matter what we do with its audit context, we are not going to see the audit entry. The fix is simple: have ret_from_fork go to the point just past the call of sys_.... in the 'we have AUDIT flag set' path. There we have (64bit variant; for 31bit the situation is the same): sysc_tracenogo: tm __TI_flags+7(%r9),(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT) jz sysc_return la %r2,SP_PTREGS(%r15) # load pt_regs larl %r14,sysc_return # return point is sysc_return jg do_syscall_trace_exit which is precisely what we need - check the flag, bugger off to sysc_return if not set, otherwise call do_syscall_trace_exit() and bugger off to sysc_return. r9 has just been properly set by ret_from_fork itself, so we are fine. Tested on s390x, seems to work fine. WARNING: it's been about 16 years since my last contact with 3X0 assembler[1], so additional review would be very welcome. I don't think I've managed to screw it up, but... [1] that *was* in another country and besides, the box is dead... Signed-off-by: Al Viro Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/entry.S | 2 +- arch/s390/kernel/entry64.S | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 198ea18a534..c1cfc7e39ec 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -436,7 +436,7 @@ ret_from_fork: basr %r14,%r1 TRACE_IRQS_ON stosm __SF_EMPTY(%r15),0x03 # reenable interrupts - b BASED(sysc_return) + b BASED(sysc_tracenogo) # # kernel_execve function needs to deal with pt_regs that is not diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S index 89c121ae633..ea90d9a676f 100644 --- a/arch/s390/kernel/entry64.S +++ b/arch/s390/kernel/entry64.S @@ -417,7 +417,7 @@ ret_from_fork: 0: brasl %r14,schedule_tail TRACE_IRQS_ON stosm 24(%r15),0x03 # reenable interrupts - j sysc_return + j sysc_tracenogo # # kernel_execve function needs to deal with pt_regs that is not -- cgit v1.2.3-70-g09d2 From f414f5f15376764d68a31dc568d9e814d3fcb58a Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 25 Dec 2008 13:37:59 +0100 Subject: [S390] cpu topology: dont destroy cpu sets on topology change Call rebuild_sched_domains instead of arch_reinit_sched_domains if cpu topology changes. This leaves cpu sets alone which otherwise would be destroyed. If and how it makes sense to define cpu sets on a virtualized architecture is another question. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/topology.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c index 586c5e33b3f..36faac50e77 100644 --- a/arch/s390/kernel/topology.c +++ b/arch/s390/kernel/topology.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include @@ -234,7 +235,7 @@ void arch_update_cpu_topology(void) static void topology_work_fn(struct work_struct *work) { - arch_reinit_sched_domains(); + rebuild_sched_domains(); } void topology_schedule_update(void) -- cgit v1.2.3-70-g09d2 From b020632e40c3ed5e8c0c066d022672907e8401cf Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 25 Dec 2008 13:38:36 +0100 Subject: [S390] introduce vdso on s390 Add a vdso to speed up gettimeofday and clock_getres/clock_gettime for CLOCK_REALTIME/CLOCK_MONOTONIC. Signed-off-by: Martin Schwidefsky --- arch/s390/Kconfig | 3 + arch/s390/include/asm/auxvec.h | 2 + arch/s390/include/asm/elf.h | 16 +++ arch/s390/include/asm/mmu.h | 1 + arch/s390/include/asm/page.h | 2 + arch/s390/include/asm/vdso.h | 39 ++++++ arch/s390/kernel/Makefile | 7 +- arch/s390/kernel/asm-offsets.c | 15 ++ arch/s390/kernel/time.c | 31 ++++ arch/s390/kernel/vdso.c | 234 +++++++++++++++++++++++++++++++ arch/s390/kernel/vdso32/Makefile | 55 ++++++++ arch/s390/kernel/vdso32/clock_getres.S | 39 ++++++ arch/s390/kernel/vdso32/clock_gettime.S | 128 +++++++++++++++++ arch/s390/kernel/vdso32/gettimeofday.S | 82 +++++++++++ arch/s390/kernel/vdso32/note.S | 12 ++ arch/s390/kernel/vdso32/vdso32.lds.S | 138 ++++++++++++++++++ arch/s390/kernel/vdso32/vdso32_wrapper.S | 13 ++ arch/s390/kernel/vdso64/Makefile | 55 ++++++++ arch/s390/kernel/vdso64/clock_getres.S | 39 ++++++ arch/s390/kernel/vdso64/clock_gettime.S | 89 ++++++++++++ arch/s390/kernel/vdso64/gettimeofday.S | 56 ++++++++ arch/s390/kernel/vdso64/note.S | 12 ++ arch/s390/kernel/vdso64/vdso64.lds.S | 138 ++++++++++++++++++ arch/s390/kernel/vdso64/vdso64_wrapper.S | 13 ++ 24 files changed, 1218 insertions(+), 1 deletion(-) create mode 100644 arch/s390/include/asm/vdso.h create mode 100644 arch/s390/kernel/vdso.c create mode 100644 arch/s390/kernel/vdso32/Makefile create mode 100644 arch/s390/kernel/vdso32/clock_getres.S create mode 100644 arch/s390/kernel/vdso32/clock_gettime.S create mode 100644 arch/s390/kernel/vdso32/gettimeofday.S create mode 100644 arch/s390/kernel/vdso32/note.S create mode 100644 arch/s390/kernel/vdso32/vdso32.lds.S create mode 100644 arch/s390/kernel/vdso32/vdso32_wrapper.S create mode 100644 arch/s390/kernel/vdso64/Makefile create mode 100644 arch/s390/kernel/vdso64/clock_getres.S create mode 100644 arch/s390/kernel/vdso64/clock_gettime.S create mode 100644 arch/s390/kernel/vdso64/gettimeofday.S create mode 100644 arch/s390/kernel/vdso64/note.S create mode 100644 arch/s390/kernel/vdso64/vdso64.lds.S create mode 100644 arch/s390/kernel/vdso64/vdso64_wrapper.S (limited to 'arch/s390') diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 8116a3328a1..37bb37334de 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -43,6 +43,9 @@ config GENERIC_HWEIGHT config GENERIC_TIME def_bool y +config GENERIC_TIME_VSYSCALL + def_bool y + config GENERIC_CLOCKEVENTS def_bool y diff --git a/arch/s390/include/asm/auxvec.h b/arch/s390/include/asm/auxvec.h index 0d340720fd9..a1f153e8913 100644 --- a/arch/s390/include/asm/auxvec.h +++ b/arch/s390/include/asm/auxvec.h @@ -1,4 +1,6 @@ #ifndef __ASMS390_AUXVEC_H #define __ASMS390_AUXVEC_H +#define AT_SYSINFO_EHDR 33 + #endif diff --git a/arch/s390/include/asm/elf.h b/arch/s390/include/asm/elf.h index 261785ab5b2..d480f39d65e 100644 --- a/arch/s390/include/asm/elf.h +++ b/arch/s390/include/asm/elf.h @@ -120,6 +120,10 @@ typedef s390_compat_regs compat_elf_gregset_t; #include /* for save_access_regs */ #include +#include + +extern unsigned int vdso_enabled; + /* * This is used to ensure we don't load something for the wrong architecture. */ @@ -191,4 +195,16 @@ do { \ current->mm->context.noexec == 0; \ }) +#define ARCH_DLINFO \ +do { \ + if (vdso_enabled) \ + NEW_AUX_ENT(AT_SYSINFO_EHDR, \ + (unsigned long)current->mm->context.vdso_base); \ +} while (0) + +struct linux_binprm; + +#define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1 +int arch_setup_additional_pages(struct linux_binprm *, int); + #endif diff --git a/arch/s390/include/asm/mmu.h b/arch/s390/include/asm/mmu.h index d2b4ff83147..3b59216e628 100644 --- a/arch/s390/include/asm/mmu.h +++ b/arch/s390/include/asm/mmu.h @@ -6,6 +6,7 @@ typedef struct { struct list_head pgtable_list; unsigned long asce_bits; unsigned long asce_limit; + unsigned long vdso_base; int noexec; int has_pgste; /* The mmu context has extended page tables */ int alloc_pgste; /* cloned contexts will have extended page tables */ diff --git a/arch/s390/include/asm/page.h b/arch/s390/include/asm/page.h index 991ba939408..32e8f6aa438 100644 --- a/arch/s390/include/asm/page.h +++ b/arch/s390/include/asm/page.h @@ -152,4 +152,6 @@ void arch_alloc_page(struct page *page, int order); #include #include +#define __HAVE_ARCH_GATE_AREA 1 + #endif /* _S390_PAGE_H */ diff --git a/arch/s390/include/asm/vdso.h b/arch/s390/include/asm/vdso.h new file mode 100644 index 00000000000..a44f4fe16a3 --- /dev/null +++ b/arch/s390/include/asm/vdso.h @@ -0,0 +1,39 @@ +#ifndef __S390_VDSO_H__ +#define __S390_VDSO_H__ + +#ifdef __KERNEL__ + +/* Default link addresses for the vDSOs */ +#define VDSO32_LBASE 0 +#define VDSO64_LBASE 0 + +#define VDSO_VERSION_STRING LINUX_2.6.26 + +#ifndef __ASSEMBLY__ + +/* + * Note about this structure: + * + * NEVER USE THIS IN USERSPACE CODE DIRECTLY. The layout of this + * structure is supposed to be known only to the function in the vdso + * itself and may change without notice. + */ + +struct vdso_data { + __u64 tb_update_count; /* Timebase atomicity ctr 0x00 */ + __u64 xtime_tod_stamp; /* TOD clock for xtime 0x08 */ + __u64 xtime_clock_sec; /* Kernel time 0x10 */ + __u64 xtime_clock_nsec; /* 0x18 */ + __u64 wtom_clock_sec; /* Wall to monotonic clock 0x20 */ + __u64 wtom_clock_nsec; /* 0x28 */ + __u32 tz_minuteswest; /* Minutes west of Greenwich 0x30 */ + __u32 tz_dsttime; /* Type of dst correction 0x34 */ +}; + +extern struct vdso_data *vdso_data; + +#endif /* __ASSEMBLY__ */ + +#endif /* __KERNEL__ */ + +#endif /* __S390_VDSO_H__ */ diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile index 50f657e7734..235b9484a4d 100644 --- a/arch/s390/kernel/Makefile +++ b/arch/s390/kernel/Makefile @@ -14,7 +14,8 @@ CFLAGS_ptrace.o += -DUTS_MACHINE='"$(UTS_MACHINE)"' obj-y := bitmap.o traps.o time.o process.o base.o early.o \ setup.o sys_s390.o ptrace.o signal.o cpcmd.o ebcdic.o \ - s390_ext.o debug.o irq.o ipl.o dis.o diag.o mem_detect.o + s390_ext.o debug.o irq.o ipl.o dis.o diag.o mem_detect.o \ + vdso.o obj-y += $(if $(CONFIG_64BIT),entry64.o,entry.o) obj-y += $(if $(CONFIG_64BIT),reipl64.o,reipl.o) @@ -39,3 +40,7 @@ S390_KEXEC_OBJS := machine_kexec.o crash.o S390_KEXEC_OBJS += $(if $(CONFIG_64BIT),relocate_kernel64.o,relocate_kernel.o) obj-$(CONFIG_KEXEC) += $(S390_KEXEC_OBJS) +# vdso +obj-$(CONFIG_64BIT) += vdso64/ +obj-$(CONFIG_32BIT) += vdso32/ +obj-$(CONFIG_COMPAT) += vdso32/ diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c index 3d144e6020c..e641f60bac9 100644 --- a/arch/s390/kernel/asm-offsets.c +++ b/arch/s390/kernel/asm-offsets.c @@ -6,6 +6,7 @@ #include #include +#include int main(void) { @@ -38,5 +39,19 @@ int main(void) DEFINE(__SF_BACKCHAIN, offsetof(struct stack_frame, back_chain)); DEFINE(__SF_GPRS, offsetof(struct stack_frame, gprs)); DEFINE(__SF_EMPTY, offsetof(struct stack_frame, empty1)); + BLANK(); + /* timeval/timezone offsets for use by vdso */ + DEFINE(__VDSO_UPD_COUNT, offsetof(struct vdso_data, tb_update_count)); + DEFINE(__VDSO_XTIME_STAMP, offsetof(struct vdso_data, xtime_tod_stamp)); + DEFINE(__VDSO_XTIME_SEC, offsetof(struct vdso_data, xtime_clock_sec)); + DEFINE(__VDSO_XTIME_NSEC, offsetof(struct vdso_data, xtime_clock_nsec)); + DEFINE(__VDSO_WTOM_SEC, offsetof(struct vdso_data, wtom_clock_sec)); + DEFINE(__VDSO_WTOM_NSEC, offsetof(struct vdso_data, wtom_clock_nsec)); + DEFINE(__VDSO_TIMEZONE, offsetof(struct vdso_data, tz_minuteswest)); + /* constants used by the vdso */ + DEFINE(CLOCK_REALTIME, CLOCK_REALTIME); + DEFINE(CLOCK_MONOTONIC, CLOCK_MONOTONIC); + DEFINE(CLOCK_REALTIME_RES, MONOTONIC_RES_NSEC); + return 0; } diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index eccefbbff88..b73bbf31f43 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -36,6 +36,7 @@ #include #include #include +#include #include #include #include @@ -223,6 +224,36 @@ static struct clocksource clocksource_tod = { }; +void update_vsyscall(struct timespec *wall_time, struct clocksource *clock) +{ + if (clock != &clocksource_tod) + return; + + /* Make userspace gettimeofday spin until we're done. */ + ++vdso_data->tb_update_count; + smp_wmb(); + vdso_data->xtime_tod_stamp = clock->cycle_last; + vdso_data->xtime_clock_sec = xtime.tv_sec; + vdso_data->xtime_clock_nsec = xtime.tv_nsec; + vdso_data->wtom_clock_sec = wall_to_monotonic.tv_sec; + vdso_data->wtom_clock_nsec = wall_to_monotonic.tv_nsec; + smp_wmb(); + ++vdso_data->tb_update_count; +} + +extern struct timezone sys_tz; + +void update_vsyscall_tz(void) +{ + /* Make userspace gettimeofday spin until we're done. */ + ++vdso_data->tb_update_count; + smp_wmb(); + vdso_data->tz_minuteswest = sys_tz.tz_minuteswest; + vdso_data->tz_dsttime = sys_tz.tz_dsttime; + smp_wmb(); + ++vdso_data->tb_update_count; +} + /* * Initialize the TOD clock and the CPU timer of * the boot cpu. diff --git a/arch/s390/kernel/vdso.c b/arch/s390/kernel/vdso.c new file mode 100644 index 00000000000..10a6ccef441 --- /dev/null +++ b/arch/s390/kernel/vdso.c @@ -0,0 +1,234 @@ +/* + * vdso setup for s390 + * + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include + +/* Max supported size for symbol names */ +#define MAX_SYMNAME 64 + +#if defined(CONFIG_32BIT) || defined(CONFIG_COMPAT) +extern char vdso32_start, vdso32_end; +static void *vdso32_kbase = &vdso32_start; +static unsigned int vdso32_pages; +static struct page **vdso32_pagelist; +#endif + +#ifdef CONFIG_64BIT +extern char vdso64_start, vdso64_end; +static void *vdso64_kbase = &vdso64_start; +static unsigned int vdso64_pages; +static struct page **vdso64_pagelist; +#endif /* CONFIG_64BIT */ + +/* + * Should the kernel map a VDSO page into processes and pass its + * address down to glibc upon exec()? + */ +unsigned int __read_mostly vdso_enabled = 1; + +static int __init vdso_setup(char *s) +{ + vdso_enabled = simple_strtoul(s, NULL, 0); + return 1; +} +__setup("vdso=", vdso_setup); + +/* + * The vdso data page + */ +static union { + struct vdso_data data; + u8 page[PAGE_SIZE]; +} vdso_data_store __attribute__((__section__(".data.page_aligned"))); +struct vdso_data *vdso_data = &vdso_data_store.data; + +/* + * This is called from binfmt_elf, we create the special vma for the + * vDSO and insert it into the mm struct tree + */ +int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp) +{ + struct mm_struct *mm = current->mm; + struct page **vdso_pagelist; + unsigned long vdso_pages; + unsigned long vdso_base; + int rc; + + if (!vdso_enabled) + return 0; + /* + * Only map the vdso for dynamically linked elf binaries. + */ + if (!uses_interp) + return 0; + + vdso_base = mm->mmap_base; +#ifdef CONFIG_64BIT + vdso_pagelist = vdso64_pagelist; + vdso_pages = vdso64_pages; +#ifdef CONFIG_COMPAT + if (test_thread_flag(TIF_31BIT)) { + vdso_pagelist = vdso32_pagelist; + vdso_pages = vdso32_pages; + } +#endif +#else + vdso_pagelist = vdso32_pagelist; + vdso_pages = vdso32_pages; +#endif + + /* + * vDSO has a problem and was disabled, just don't "enable" it for + * the process + */ + if (vdso_pages == 0) + return 0; + + current->mm->context.vdso_base = 0; + + /* + * pick a base address for the vDSO in process space. We try to put + * it at vdso_base which is the "natural" base for it, but we might + * fail and end up putting it elsewhere. + */ + down_write(&mm->mmap_sem); + vdso_base = get_unmapped_area(NULL, vdso_base, + vdso_pages << PAGE_SHIFT, 0, 0); + if (IS_ERR_VALUE(vdso_base)) { + rc = vdso_base; + goto out_up; + } + + /* + * our vma flags don't have VM_WRITE so by default, the process + * isn't allowed to write those pages. + * gdb can break that with ptrace interface, and thus trigger COW + * on those pages but it's then your responsibility to never do that + * on the "data" page of the vDSO or you'll stop getting kernel + * updates and your nice userland gettimeofday will be totally dead. + * It's fine to use that for setting breakpoints in the vDSO code + * pages though + * + * Make sure the vDSO gets into every core dump. + * Dumping its contents makes post-mortem fully interpretable later + * without matching up the same kernel and hardware config to see + * what PC values meant. + */ + rc = install_special_mapping(mm, vdso_base, vdso_pages << PAGE_SHIFT, + VM_READ|VM_EXEC| + VM_MAYREAD|VM_MAYWRITE|VM_MAYEXEC| + VM_ALWAYSDUMP, + vdso_pagelist); + if (rc) + goto out_up; + + /* Put vDSO base into mm struct */ + current->mm->context.vdso_base = vdso_base; + + up_write(&mm->mmap_sem); + return 0; + +out_up: + up_write(&mm->mmap_sem); + return rc; +} + +const char *arch_vma_name(struct vm_area_struct *vma) +{ + if (vma->vm_mm && vma->vm_start == vma->vm_mm->context.vdso_base) + return "[vdso]"; + return NULL; +} + +static int __init vdso_init(void) +{ + int i; + +#if defined(CONFIG_32BIT) || defined(CONFIG_COMPAT) + /* Calculate the size of the 32 bit vDSO */ + vdso32_pages = ((&vdso32_end - &vdso32_start + + PAGE_SIZE - 1) >> PAGE_SHIFT) + 1; + + /* Make sure pages are in the correct state */ + vdso32_pagelist = kzalloc(sizeof(struct page *) * (vdso32_pages + 1), + GFP_KERNEL); + BUG_ON(vdso32_pagelist == NULL); + for (i = 0; i < vdso32_pages - 1; i++) { + struct page *pg = virt_to_page(vdso32_kbase + i*PAGE_SIZE); + ClearPageReserved(pg); + get_page(pg); + vdso32_pagelist[i] = pg; + } + vdso32_pagelist[vdso32_pages - 1] = virt_to_page(vdso_data); + vdso32_pagelist[vdso32_pages] = NULL; +#endif + +#ifdef CONFIG_64BIT + /* Calculate the size of the 64 bit vDSO */ + vdso64_pages = ((&vdso64_end - &vdso64_start + + PAGE_SIZE - 1) >> PAGE_SHIFT) + 1; + + /* Make sure pages are in the correct state */ + vdso64_pagelist = kzalloc(sizeof(struct page *) * (vdso64_pages + 1), + GFP_KERNEL); + BUG_ON(vdso64_pagelist == NULL); + for (i = 0; i < vdso64_pages - 1; i++) { + struct page *pg = virt_to_page(vdso64_kbase + i*PAGE_SIZE); + ClearPageReserved(pg); + get_page(pg); + vdso64_pagelist[i] = pg; + } + vdso64_pagelist[vdso64_pages - 1] = virt_to_page(vdso_data); + vdso64_pagelist[vdso64_pages] = NULL; +#endif /* CONFIG_64BIT */ + + get_page(virt_to_page(vdso_data)); + + smp_wmb(); + + return 0; +} +arch_initcall(vdso_init); + +int in_gate_area_no_task(unsigned long addr) +{ + return 0; +} + +int in_gate_area(struct task_struct *task, unsigned long addr) +{ + return 0; +} + +struct vm_area_struct *get_gate_vma(struct task_struct *tsk) +{ + return NULL; +} diff --git a/arch/s390/kernel/vdso32/Makefile b/arch/s390/kernel/vdso32/Makefile new file mode 100644 index 00000000000..ca78ad60ba2 --- /dev/null +++ b/arch/s390/kernel/vdso32/Makefile @@ -0,0 +1,55 @@ +# List of files in the vdso, has to be asm only for now + +obj-vdso32 = gettimeofday.o clock_getres.o clock_gettime.o note.o + +# Build rules + +targets := $(obj-vdso32) vdso32.so vdso32.so.dbg +obj-vdso32 := $(addprefix $(obj)/, $(obj-vdso32)) + +KBUILD_AFLAGS_31 := $(filter-out -m64,$(KBUILD_AFLAGS)) +KBUILD_AFLAGS_31 += -m31 -s + +KBUILD_CFLAGS_31 := $(filter-out -m64,$(KBUILD_CFLAGS)) +KBUILD_CFLAGS_31 += -m31 -fPIC -shared -fno-common -fno-builtin +KBUILD_CFLAGS_31 += -nostdlib -Wl,-soname=linux-vdso32.so.1 \ + $(call ld-option, -Wl$(comma)--hash-style=sysv) + +$(targets:%=$(obj)/%.dbg): KBUILD_CFLAGS = $(KBUILD_CFLAGS_31) +$(targets:%=$(obj)/%.dbg): KBUILD_AFLAGS = $(KBUILD_AFLAGS_31) + +obj-y += vdso32_wrapper.o +extra-y += vdso32.lds +CPPFLAGS_vdso32.lds += -P -C -U$(ARCH) + +# Force dependency (incbin is bad) +$(obj)/vdso32_wrapper.o : $(obj)/vdso32.so + +# link rule for the .so file, .lds has to be first +$(obj)/vdso32.so.dbg: $(src)/vdso32.lds $(obj-vdso32) + $(call if_changed,vdso32ld) + +# strip rule for the .so file +$(obj)/%.so: OBJCOPYFLAGS := -S +$(obj)/%.so: $(obj)/%.so.dbg FORCE + $(call if_changed,objcopy) + +# assembly rules for the .S files +$(obj-vdso32): %.o: %.S + $(call if_changed_dep,vdso32as) + +# actual build commands +quiet_cmd_vdso32ld = VDSO32L $@ + cmd_vdso32ld = $(CC) $(c_flags) -Wl,-T $^ -o $@ +quiet_cmd_vdso32as = VDSO32A $@ + cmd_vdso32as = $(CC) $(a_flags) -c -o $@ $< + +# install commands for the unstripped file +quiet_cmd_vdso_install = INSTALL $@ + cmd_vdso_install = cp $(obj)/$@.dbg $(MODLIB)/vdso/$@ + +vdso32.so: $(obj)/vdso32.so.dbg + @mkdir -p $(MODLIB)/vdso + $(call cmd,vdso_install) + +vdso_install: vdso32.so diff --git a/arch/s390/kernel/vdso32/clock_getres.S b/arch/s390/kernel/vdso32/clock_getres.S new file mode 100644 index 00000000000..9532c4e6a9d --- /dev/null +++ b/arch/s390/kernel/vdso32/clock_getres.S @@ -0,0 +1,39 @@ +/* + * Userland implementation of clock_getres() for 32 bits processes in a + * s390 kernel for use in the vDSO + * + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + */ +#include +#include +#include + + .text + .align 4 + .globl __kernel_clock_getres + .type __kernel_clock_getres,@function +__kernel_clock_getres: + .cfi_startproc + chi %r2,CLOCK_REALTIME + je 0f + chi %r2,CLOCK_MONOTONIC + jne 3f +0: ltr %r3,%r3 + jz 2f /* res == NULL */ + basr %r1,0 +1: l %r0,4f-1b(%r1) + xc 0(4,%r3),0(%r3) /* set tp->tv_sec to zero */ + st %r0,4(%r3) /* store tp->tv_usec */ +2: lhi %r2,0 + br %r14 +3: lhi %r1,__NR_clock_getres /* fallback to svc */ + svc 0 + br %r14 +4: .long CLOCK_REALTIME_RES + .cfi_endproc + .size __kernel_clock_getres,.-__kernel_clock_getres diff --git a/arch/s390/kernel/vdso32/clock_gettime.S b/arch/s390/kernel/vdso32/clock_gettime.S new file mode 100644 index 00000000000..4a98909a831 --- /dev/null +++ b/arch/s390/kernel/vdso32/clock_gettime.S @@ -0,0 +1,128 @@ +/* + * Userland implementation of clock_gettime() for 32 bits processes in a + * s390 kernel for use in the vDSO + * + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + */ +#include +#include +#include + + .text + .align 4 + .globl __kernel_clock_gettime + .type __kernel_clock_gettime,@function +__kernel_clock_gettime: + .cfi_startproc + basr %r5,0 +0: al %r5,21f-0b(%r5) /* get &_vdso_data */ + chi %r2,CLOCK_REALTIME + je 10f + chi %r2,CLOCK_MONOTONIC + jne 19f + + /* CLOCK_MONOTONIC */ + ltr %r3,%r3 + jz 9f /* tp == NULL */ +1: l %r4,__VDSO_UPD_COUNT+4(%r5) /* load update counter */ + tml %r4,0x0001 /* pending update ? loop */ + jnz 1b + stck 24(%r15) /* Store TOD clock */ + lm %r0,%r1,24(%r15) + s %r0,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */ + sl %r1,__VDSO_XTIME_STAMP+4(%r5) + brc 3,2f + ahi %r0,-1 +2: mhi %r0,1000 /* cyc2ns(clock,cycle_delta) */ + lr %r2,%r0 + lhi %r0,1000 + ltr %r1,%r1 + mr %r0,%r0 + jnm 3f + ahi %r0,1000 +3: alr %r0,%r2 + srdl %r0,12 + al %r0,__VDSO_XTIME_NSEC(%r5) /* + xtime */ + al %r1,__VDSO_XTIME_NSEC+4(%r5) + brc 12,4f + ahi %r0,1 +4: l %r2,__VDSO_XTIME_SEC+4(%r5) + al %r0,__VDSO_WTOM_NSEC(%r5) /* + wall_to_monotonic */ + al %r1,__VDSO_WTOM_NSEC+4(%r5) + brc 12,5f + ahi %r0,1 +5: al %r2,__VDSO_WTOM_SEC+4(%r5) + cl %r4,__VDSO_UPD_COUNT+4(%r5) /* check update counter */ + jne 1b + basr %r5,0 +6: ltr %r0,%r0 + jnz 7f + cl %r1,20f-6b(%r5) + jl 8f +7: ahi %r2,1 + sl %r1,20f-6b(%r5) + brc 3,6b + ahi %r0,-1 + j 6b +8: st %r2,0(%r3) /* store tp->tv_sec */ + st %r1,4(%r3) /* store tp->tv_nsec */ +9: lhi %r2,0 + br %r14 + + /* CLOCK_REALTIME */ +10: ltr %r3,%r3 /* tp == NULL */ + jz 18f +11: l %r4,__VDSO_UPD_COUNT+4(%r5) /* load update counter */ + tml %r4,0x0001 /* pending update ? loop */ + jnz 11b + stck 24(%r15) /* Store TOD clock */ + lm %r0,%r1,24(%r15) + s %r0,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */ + sl %r1,__VDSO_XTIME_STAMP+4(%r5) + brc 3,12f + ahi %r0,-1 +12: mhi %r0,1000 /* cyc2ns(clock,cycle_delta) */ + lr %r2,%r0 + lhi %r0,1000 + ltr %r1,%r1 + mr %r0,%r0 + jnm 13f + ahi %r0,1000 +13: alr %r0,%r2 + srdl %r0,12 + al %r0,__VDSO_XTIME_NSEC(%r5) /* + xtime */ + al %r1,__VDSO_XTIME_NSEC+4(%r5) + brc 12,14f + ahi %r0,1 +14: l %r2,__VDSO_XTIME_SEC+4(%r5) + cl %r4,__VDSO_UPD_COUNT+4(%r5) /* check update counter */ + jne 11b + basr %r5,0 +15: ltr %r0,%r0 + jnz 16f + cl %r1,20f-15b(%r5) + jl 17f +16: ahi %r2,1 + sl %r1,20f-15b(%r5) + brc 3,15b + ahi %r0,-1 + j 15b +17: st %r2,0(%r3) /* store tp->tv_sec */ + st %r1,4(%r3) /* store tp->tv_nsec */ +18: lhi %r2,0 + br %r14 + + /* Fallback to system call */ +19: lhi %r1,__NR_clock_gettime + svc 0 + br %r14 + +20: .long 1000000000 +21: .long _vdso_data - 0b + .cfi_endproc + .size __kernel_clock_gettime,.-__kernel_clock_gettime diff --git a/arch/s390/kernel/vdso32/gettimeofday.S b/arch/s390/kernel/vdso32/gettimeofday.S new file mode 100644 index 00000000000..c32f29c3d70 --- /dev/null +++ b/arch/s390/kernel/vdso32/gettimeofday.S @@ -0,0 +1,82 @@ +/* + * Userland implementation of gettimeofday() for 32 bits processes in a + * s390 kernel for use in the vDSO + * + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + */ +#include +#include +#include + +#include +#include +#include + + .text + .align 4 + .globl __kernel_gettimeofday + .type __kernel_gettimeofday,@function +__kernel_gettimeofday: + .cfi_startproc + basr %r5,0 +0: al %r5,13f-0b(%r5) /* get &_vdso_data */ +1: ltr %r3,%r3 /* check if tz is NULL */ + je 2f + mvc 0(8,%r3),__VDSO_TIMEZONE(%r5) +2: ltr %r2,%r2 /* check if tv is NULL */ + je 10f + l %r4,__VDSO_UPD_COUNT+4(%r5) /* load update counter */ + tml %r4,0x0001 /* pending update ? loop */ + jnz 1b + stck 24(%r15) /* Store TOD clock */ + lm %r0,%r1,24(%r15) + s %r0,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */ + sl %r1,__VDSO_XTIME_STAMP+4(%r5) + brc 3,3f + ahi %r0,-1 +3: mhi %r0,1000 /* cyc2ns(clock,cycle_delta) */ + st %r0,24(%r15) + lhi %r0,1000 + ltr %r1,%r1 + mr %r0,%r0 + jnm 4f + ahi %r0,1000 +4: al %r0,24(%r15) + srdl %r0,12 + al %r0,__VDSO_XTIME_NSEC(%r5) /* + xtime */ + al %r1,__VDSO_XTIME_NSEC+4(%r5) + brc 12,5f + ahi %r0,1 +5: mvc 24(4,%r15),__VDSO_XTIME_SEC+4(%r5) + cl %r4,__VDSO_UPD_COUNT+4(%r5) /* check update counter */ + jne 1b + l %r4,24(%r15) /* get tv_sec from stack */ + basr %r5,0 +6: ltr %r0,%r0 + jnz 7f + cl %r1,11f-6b(%r5) + jl 8f +7: ahi %r4,1 + sl %r1,11f-6b(%r5) + brc 3,6b + ahi %r0,-1 + j 6b +8: st %r4,0(%r2) /* store tv->tv_sec */ + ltr %r1,%r1 + m %r0,12f-6b(%r5) + jnm 9f + al %r0,12f-6b(%r5) +9: srl %r0,6 + st %r0,4(%r2) /* store tv->tv_usec */ +10: slr %r2,%r2 + br %r14 +11: .long 1000000000 +12: .long 274877907 +13: .long _vdso_data - 0b + .cfi_endproc + .size __kernel_gettimeofday,.-__kernel_gettimeofday diff --git a/arch/s390/kernel/vdso32/note.S b/arch/s390/kernel/vdso32/note.S new file mode 100644 index 00000000000..79a071e4357 --- /dev/null +++ b/arch/s390/kernel/vdso32/note.S @@ -0,0 +1,12 @@ +/* + * This supplies .note.* sections to go into the PT_NOTE inside the vDSO text. + * Here we can supply some information useful to userland. + */ + +#include +#include +#include + +ELFNOTE_START(Linux, 0, "a") + .long LINUX_VERSION_CODE +ELFNOTE_END diff --git a/arch/s390/kernel/vdso32/vdso32.lds.S b/arch/s390/kernel/vdso32/vdso32.lds.S new file mode 100644 index 00000000000..a8c379fa124 --- /dev/null +++ b/arch/s390/kernel/vdso32/vdso32.lds.S @@ -0,0 +1,138 @@ +/* + * This is the infamous ld script for the 32 bits vdso + * library + */ +#include + +OUTPUT_FORMAT("elf32-s390", "elf32-s390", "elf32-s390") +OUTPUT_ARCH(s390:31-bit) +ENTRY(_start) + +SECTIONS +{ + . = VDSO32_LBASE + SIZEOF_HEADERS; + + .hash : { *(.hash) } :text + .gnu.hash : { *(.gnu.hash) } + .dynsym : { *(.dynsym) } + .dynstr : { *(.dynstr) } + .gnu.version : { *(.gnu.version) } + .gnu.version_d : { *(.gnu.version_d) } + .gnu.version_r : { *(.gnu.version_r) } + + .note : { *(.note.*) } :text :note + + . = ALIGN(16); + .text : { + *(.text .stub .text.* .gnu.linkonce.t.*) + } :text + PROVIDE(__etext = .); + PROVIDE(_etext = .); + PROVIDE(etext = .); + + /* + * Other stuff is appended to the text segment: + */ + .rodata : { *(.rodata .rodata.* .gnu.linkonce.r.*) } + .rodata1 : { *(.rodata1) } + + .dynamic : { *(.dynamic) } :text :dynamic + + .eh_frame_hdr : { *(.eh_frame_hdr) } :text :eh_frame_hdr + .eh_frame : { KEEP (*(.eh_frame)) } :text + .gcc_except_table : { *(.gcc_except_table .gcc_except_table.*) } + + .rela.dyn ALIGN(8) : { *(.rela.dyn) } + .got ALIGN(8) : { *(.got .toc) } + + _end = .; + PROVIDE(end = .); + + /* + * Stabs debugging sections are here too. + */ + .stab 0 : { *(.stab) } + .stabstr 0 : { *(.stabstr) } + .stab.excl 0 : { *(.stab.excl) } + .stab.exclstr 0 : { *(.stab.exclstr) } + .stab.index 0 : { *(.stab.index) } + .stab.indexstr 0 : { *(.stab.indexstr) } + .comment 0 : { *(.comment) } + + /* + * DWARF debug sections. + * Symbols in the DWARF debugging sections are relative to the + * beginning of the section so we begin them at 0. + */ + /* DWARF 1 */ + .debug 0 : { *(.debug) } + .line 0 : { *(.line) } + /* GNU DWARF 1 extensions */ + .debug_srcinfo 0 : { *(.debug_srcinfo) } + .debug_sfnames 0 : { *(.debug_sfnames) } + /* DWARF 1.1 and DWARF 2 */ + .debug_aranges 0 : { *(.debug_aranges) } + .debug_pubnames 0 : { *(.debug_pubnames) } + /* DWARF 2 */ + .debug_info 0 : { *(.debug_info .gnu.linkonce.wi.*) } + .debug_abbrev 0 : { *(.debug_abbrev) } + .debug_line 0 : { *(.debug_line) } + .debug_frame 0 : { *(.debug_frame) } + .debug_str 0 : { *(.debug_str) } + .debug_loc 0 : { *(.debug_loc) } + .debug_macinfo 0 : { *(.debug_macinfo) } + /* SGI/MIPS DWARF 2 extensions */ + .debug_weaknames 0 : { *(.debug_weaknames) } + .debug_funcnames 0 : { *(.debug_funcnames) } + .debug_typenames 0 : { *(.debug_typenames) } + .debug_varnames 0 : { *(.debug_varnames) } + /* DWARF 3 */ + .debug_pubtypes 0 : { *(.debug_pubtypes) } + .debug_ranges 0 : { *(.debug_ranges) } + .gnu.attributes 0 : { KEEP (*(.gnu.attributes)) } + + . = ALIGN(4096); + PROVIDE(_vdso_data = .); + + /DISCARD/ : { + *(.note.GNU-stack) + *(.branch_lt) + *(.data .data.* .gnu.linkonce.d.* .sdata*) + *(.bss .sbss .dynbss .dynsbss) + } +} + +/* + * Very old versions of ld do not recognize this name token; use the constant. + */ +#define PT_GNU_EH_FRAME 0x6474e550 + +/* + * We must supply the ELF program headers explicitly to get just one + * PT_LOAD segment, and set the flags explicitly to make segments read-only. + */ +PHDRS +{ + text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */ + dynamic PT_DYNAMIC FLAGS(4); /* PF_R */ + note PT_NOTE FLAGS(4); /* PF_R */ + eh_frame_hdr PT_GNU_EH_FRAME; +} + +/* + * This controls what symbols we export from the DSO. + */ +VERSION +{ + VDSO_VERSION_STRING { + global: + /* + * Has to be there for the kernel to find + */ + __kernel_gettimeofday; + __kernel_clock_gettime; + __kernel_clock_getres; + + local: *; + }; +} diff --git a/arch/s390/kernel/vdso32/vdso32_wrapper.S b/arch/s390/kernel/vdso32/vdso32_wrapper.S new file mode 100644 index 00000000000..61639a89e70 --- /dev/null +++ b/arch/s390/kernel/vdso32/vdso32_wrapper.S @@ -0,0 +1,13 @@ +#include +#include + + .section ".data.page_aligned" + + .globl vdso32_start, vdso32_end + .balign PAGE_SIZE +vdso32_start: + .incbin "arch/s390/kernel/vdso32/vdso32.so" + .balign PAGE_SIZE +vdso32_end: + + .previous diff --git a/arch/s390/kernel/vdso64/Makefile b/arch/s390/kernel/vdso64/Makefile new file mode 100644 index 00000000000..6fc8e829258 --- /dev/null +++ b/arch/s390/kernel/vdso64/Makefile @@ -0,0 +1,55 @@ +# List of files in the vdso, has to be asm only for now + +obj-vdso64 = gettimeofday.o clock_getres.o clock_gettime.o note.o + +# Build rules + +targets := $(obj-vdso64) vdso64.so vdso64.so.dbg +obj-vdso64 := $(addprefix $(obj)/, $(obj-vdso64)) + +KBUILD_AFLAGS_64 := $(filter-out -m64,$(KBUILD_AFLAGS)) +KBUILD_AFLAGS_64 += -m64 -s + +KBUILD_CFLAGS_64 := $(filter-out -m64,$(KBUILD_CFLAGS)) +KBUILD_CFLAGS_64 += -m64 -fPIC -shared -fno-common -fno-builtin +KBUILD_CFLAGS_64 += -nostdlib -Wl,-soname=linux-vdso64.so.1 \ + $(call ld-option, -Wl$(comma)--hash-style=sysv) + +$(targets:%=$(obj)/%.dbg): KBUILD_CFLAGS = $(KBUILD_CFLAGS_64) +$(targets:%=$(obj)/%.dbg): KBUILD_AFLAGS = $(KBUILD_AFLAGS_64) + +obj-y += vdso64_wrapper.o +extra-y += vdso64.lds +CPPFLAGS_vdso64.lds += -P -C -U$(ARCH) + +# Force dependency (incbin is bad) +$(obj)/vdso64_wrapper.o : $(obj)/vdso64.so + +# link rule for the .so file, .lds has to be first +$(obj)/vdso64.so.dbg: $(src)/vdso64.lds $(obj-vdso64) + $(call if_changed,vdso64ld) + +# strip rule for the .so file +$(obj)/%.so: OBJCOPYFLAGS := -S +$(obj)/%.so: $(obj)/%.so.dbg FORCE + $(call if_changed,objcopy) + +# assembly rules for the .S files +$(obj-vdso64): %.o: %.S + $(call if_changed_dep,vdso64as) + +# actual build commands +quiet_cmd_vdso64ld = VDSO64L $@ + cmd_vdso64ld = $(CC) $(c_flags) -Wl,-T $^ -o $@ +quiet_cmd_vdso64as = VDSO64A $@ + cmd_vdso64as = $(CC) $(a_flags) -c -o $@ $< + +# install commands for the unstripped file +quiet_cmd_vdso_install = INSTALL $@ + cmd_vdso_install = cp $(obj)/$@.dbg $(MODLIB)/vdso/$@ + +vdso64.so: $(obj)/vdso64.so.dbg + @mkdir -p $(MODLIB)/vdso + $(call cmd,vdso_install) + +vdso_install: vdso64.so diff --git a/arch/s390/kernel/vdso64/clock_getres.S b/arch/s390/kernel/vdso64/clock_getres.S new file mode 100644 index 00000000000..488e31a3c0e --- /dev/null +++ b/arch/s390/kernel/vdso64/clock_getres.S @@ -0,0 +1,39 @@ +/* + * Userland implementation of clock_getres() for 64 bits processes in a + * s390 kernel for use in the vDSO + * + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + */ +#include +#include +#include + + .text + .align 4 + .globl __kernel_clock_getres + .type __kernel_clock_getres,@function +__kernel_clock_getres: + .cfi_startproc + cghi %r2,CLOCK_REALTIME + je 0f + cghi %r2,CLOCK_MONOTONIC + jne 2f +0: ltgr %r3,%r3 + jz 1f /* res == NULL */ + larl %r1,3f + lg %r0,0(%r1) + xc 0(8,%r3),0(%r3) /* set tp->tv_sec to zero */ + stg %r0,8(%r3) /* store tp->tv_usec */ +1: lghi %r2,0 + br %r14 +2: lghi %r1,__NR_clock_getres /* fallback to svc */ + svc 0 + br %r14 +3: .quad CLOCK_REALTIME_RES + .cfi_endproc + .size __kernel_clock_getres,.-__kernel_clock_getres diff --git a/arch/s390/kernel/vdso64/clock_gettime.S b/arch/s390/kernel/vdso64/clock_gettime.S new file mode 100644 index 00000000000..738a410b7eb --- /dev/null +++ b/arch/s390/kernel/vdso64/clock_gettime.S @@ -0,0 +1,89 @@ +/* + * Userland implementation of clock_gettime() for 64 bits processes in a + * s390 kernel for use in the vDSO + * + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + */ +#include +#include +#include + + .text + .align 4 + .globl __kernel_clock_gettime + .type __kernel_clock_gettime,@function +__kernel_clock_gettime: + .cfi_startproc + larl %r5,_vdso_data + cghi %r2,CLOCK_REALTIME + je 4f + cghi %r2,CLOCK_MONOTONIC + jne 9f + + /* CLOCK_MONOTONIC */ + ltgr %r3,%r3 + jz 3f /* tp == NULL */ +0: lg %r4,__VDSO_UPD_COUNT(%r5) /* load update counter */ + tmll %r4,0x0001 /* pending update ? loop */ + jnz 0b + stck 48(%r15) /* Store TOD clock */ + lg %r1,48(%r15) + sg %r1,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */ + mghi %r1,1000 + srlg %r1,%r1,12 /* cyc2ns(clock,cycle_delta) */ + alg %r1,__VDSO_XTIME_NSEC(%r5) /* + xtime */ + lg %r0,__VDSO_XTIME_SEC(%r5) + alg %r1,__VDSO_WTOM_NSEC(%r5) /* + wall_to_monotonic */ + alg %r0,__VDSO_WTOM_SEC(%r5) + clg %r4,__VDSO_UPD_COUNT(%r5) /* check update counter */ + jne 0b + larl %r5,10f +1: clg %r1,0(%r5) + jl 2f + slg %r1,0(%r5) + aghi %r0,1 + j 1b +2: stg %r0,0(%r3) /* store tp->tv_sec */ + stg %r1,8(%r3) /* store tp->tv_nsec */ +3: lghi %r2,0 + br %r14 + + /* CLOCK_REALTIME */ +4: ltr %r3,%r3 /* tp == NULL */ + jz 8f +5: lg %r4,__VDSO_UPD_COUNT(%r5) /* load update counter */ + tmll %r4,0x0001 /* pending update ? loop */ + jnz 5b + stck 48(%r15) /* Store TOD clock */ + lg %r1,48(%r15) + sg %r1,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */ + mghi %r1,1000 + srlg %r1,%r1,12 /* cyc2ns(clock,cycle_delta) */ + alg %r1,__VDSO_XTIME_NSEC(%r5) /* + xtime */ + lg %r0,__VDSO_XTIME_SEC(%r5) + clg %r4,__VDSO_UPD_COUNT(%r5) /* check update counter */ + jne 5b + larl %r5,10f +6: clg %r1,0(%r5) + jl 7f + slg %r1,0(%r5) + aghi %r0,1 + j 6b +7: stg %r0,0(%r3) /* store tp->tv_sec */ + stg %r1,8(%r3) /* store tp->tv_nsec */ +8: lghi %r2,0 + br %r14 + + /* Fallback to system call */ +9: lghi %r1,__NR_clock_gettime + svc 0 + br %r14 + +10: .quad 1000000000 + .cfi_endproc + .size __kernel_clock_gettime,.-__kernel_clock_gettime diff --git a/arch/s390/kernel/vdso64/gettimeofday.S b/arch/s390/kernel/vdso64/gettimeofday.S new file mode 100644 index 00000000000..f873e75634e --- /dev/null +++ b/arch/s390/kernel/vdso64/gettimeofday.S @@ -0,0 +1,56 @@ +/* + * Userland implementation of gettimeofday() for 64 bits processes in a + * s390 kernel for use in the vDSO + * + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + */ +#include +#include +#include + + .text + .align 4 + .globl __kernel_gettimeofday + .type __kernel_gettimeofday,@function +__kernel_gettimeofday: + .cfi_startproc + larl %r5,_vdso_data +0: ltgr %r3,%r3 /* check if tz is NULL */ + je 1f + mvc 0(8,%r3),__VDSO_TIMEZONE(%r5) +1: ltgr %r2,%r2 /* check if tv is NULL */ + je 4f + lg %r4,__VDSO_UPD_COUNT(%r5) /* load update counter */ + tmll %r4,0x0001 /* pending update ? loop */ + jnz 0b + stck 48(%r15) /* Store TOD clock */ + lg %r1,48(%r15) + sg %r1,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */ + mghi %r1,1000 + srlg %r1,%r1,12 /* cyc2ns(clock,cycle_delta) */ + alg %r1,__VDSO_XTIME_NSEC(%r5) /* + xtime.tv_nsec */ + lg %r0,__VDSO_XTIME_SEC(%r5) /* xtime.tv_sec */ + clg %r4,__VDSO_UPD_COUNT(%r5) /* check update counter */ + jne 0b + larl %r5,5f +2: clg %r1,0(%r5) + jl 3f + slg %r1,0(%r5) + aghi %r0,1 + j 2b +3: stg %r0,0(%r2) /* store tv->tv_sec */ + slgr %r0,%r0 /* tv_nsec -> tv_usec */ + ml %r0,8(%r5) + srlg %r0,%r0,6 + stg %r0,8(%r2) /* store tv->tv_usec */ +4: lghi %r2,0 + br %r14 +5: .quad 1000000000 + .long 274877907 + .cfi_endproc + .size __kernel_gettimeofday,.-__kernel_gettimeofday diff --git a/arch/s390/kernel/vdso64/note.S b/arch/s390/kernel/vdso64/note.S new file mode 100644 index 00000000000..79a071e4357 --- /dev/null +++ b/arch/s390/kernel/vdso64/note.S @@ -0,0 +1,12 @@ +/* + * This supplies .note.* sections to go into the PT_NOTE inside the vDSO text. + * Here we can supply some information useful to userland. + */ + +#include +#include +#include + +ELFNOTE_START(Linux, 0, "a") + .long LINUX_VERSION_CODE +ELFNOTE_END diff --git a/arch/s390/kernel/vdso64/vdso64.lds.S b/arch/s390/kernel/vdso64/vdso64.lds.S new file mode 100644 index 00000000000..9f5979d102a --- /dev/null +++ b/arch/s390/kernel/vdso64/vdso64.lds.S @@ -0,0 +1,138 @@ +/* + * This is the infamous ld script for the 64 bits vdso + * library + */ +#include + +OUTPUT_FORMAT("elf64-s390", "elf64-s390", "elf64-s390") +OUTPUT_ARCH(s390:64-bit) +ENTRY(_start) + +SECTIONS +{ + . = VDSO64_LBASE + SIZEOF_HEADERS; + + .hash : { *(.hash) } :text + .gnu.hash : { *(.gnu.hash) } + .dynsym : { *(.dynsym) } + .dynstr : { *(.dynstr) } + .gnu.version : { *(.gnu.version) } + .gnu.version_d : { *(.gnu.version_d) } + .gnu.version_r : { *(.gnu.version_r) } + + .note : { *(.note.*) } :text :note + + . = ALIGN(16); + .text : { + *(.text .stub .text.* .gnu.linkonce.t.*) + } :text + PROVIDE(__etext = .); + PROVIDE(_etext = .); + PROVIDE(etext = .); + + /* + * Other stuff is appended to the text segment: + */ + .rodata : { *(.rodata .rodata.* .gnu.linkonce.r.*) } + .rodata1 : { *(.rodata1) } + + .dynamic : { *(.dynamic) } :text :dynamic + + .eh_frame_hdr : { *(.eh_frame_hdr) } :text :eh_frame_hdr + .eh_frame : { KEEP (*(.eh_frame)) } :text + .gcc_except_table : { *(.gcc_except_table .gcc_except_table.*) } + + .rela.dyn ALIGN(8) : { *(.rela.dyn) } + .got ALIGN(8) : { *(.got .toc) } + + _end = .; + PROVIDE(end = .); + + /* + * Stabs debugging sections are here too. + */ + .stab 0 : { *(.stab) } + .stabstr 0 : { *(.stabstr) } + .stab.excl 0 : { *(.stab.excl) } + .stab.exclstr 0 : { *(.stab.exclstr) } + .stab.index 0 : { *(.stab.index) } + .stab.indexstr 0 : { *(.stab.indexstr) } + .comment 0 : { *(.comment) } + + /* + * DWARF debug sections. + * Symbols in the DWARF debugging sections are relative to the + * beginning of the section so we begin them at 0. + */ + /* DWARF 1 */ + .debug 0 : { *(.debug) } + .line 0 : { *(.line) } + /* GNU DWARF 1 extensions */ + .debug_srcinfo 0 : { *(.debug_srcinfo) } + .debug_sfnames 0 : { *(.debug_sfnames) } + /* DWARF 1.1 and DWARF 2 */ + .debug_aranges 0 : { *(.debug_aranges) } + .debug_pubnames 0 : { *(.debug_pubnames) } + /* DWARF 2 */ + .debug_info 0 : { *(.debug_info .gnu.linkonce.wi.*) } + .debug_abbrev 0 : { *(.debug_abbrev) } + .debug_line 0 : { *(.debug_line) } + .debug_frame 0 : { *(.debug_frame) } + .debug_str 0 : { *(.debug_str) } + .debug_loc 0 : { *(.debug_loc) } + .debug_macinfo 0 : { *(.debug_macinfo) } + /* SGI/MIPS DWARF 2 extensions */ + .debug_weaknames 0 : { *(.debug_weaknames) } + .debug_funcnames 0 : { *(.debug_funcnames) } + .debug_typenames 0 : { *(.debug_typenames) } + .debug_varnames 0 : { *(.debug_varnames) } + /* DWARF 3 */ + .debug_pubtypes 0 : { *(.debug_pubtypes) } + .debug_ranges 0 : { *(.debug_ranges) } + .gnu.attributes 0 : { KEEP (*(.gnu.attributes)) } + + . = ALIGN(4096); + PROVIDE(_vdso_data = .); + + /DISCARD/ : { + *(.note.GNU-stack) + *(.branch_lt) + *(.data .data.* .gnu.linkonce.d.* .sdata*) + *(.bss .sbss .dynbss .dynsbss) + } +} + +/* + * Very old versions of ld do not recognize this name token; use the constant. + */ +#define PT_GNU_EH_FRAME 0x6474e550 + +/* + * We must supply the ELF program headers explicitly to get just one + * PT_LOAD segment, and set the flags explicitly to make segments read-only. + */ +PHDRS +{ + text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */ + dynamic PT_DYNAMIC FLAGS(4); /* PF_R */ + note PT_NOTE FLAGS(4); /* PF_R */ + eh_frame_hdr PT_GNU_EH_FRAME; +} + +/* + * This controls what symbols we export from the DSO. + */ +VERSION +{ + VDSO_VERSION_STRING { + global: + /* + * Has to be there for the kernel to find + */ + __kernel_gettimeofday; + __kernel_clock_gettime; + __kernel_clock_getres; + + local: *; + }; +} diff --git a/arch/s390/kernel/vdso64/vdso64_wrapper.S b/arch/s390/kernel/vdso64/vdso64_wrapper.S new file mode 100644 index 00000000000..d8e2ac14d56 --- /dev/null +++ b/arch/s390/kernel/vdso64/vdso64_wrapper.S @@ -0,0 +1,13 @@ +#include +#include + + .section ".data.page_aligned" + + .globl vdso64_start, vdso64_end + .balign PAGE_SIZE +vdso64_start: + .incbin "arch/s390/kernel/vdso64/vdso64.so" + .balign PAGE_SIZE +vdso64_end: + + .previous -- cgit v1.2.3-70-g09d2 From 750887dedc088d28198b170bcae83695247797d1 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 25 Dec 2008 13:38:37 +0100 Subject: [S390] convert etr/stp to stop_machine interface This converts the etr and stp code to the new stop_machine interface which allows to synchronize all cpus without allocating any memory. This way we get rid of the only reason why we haven't converted s390 to the generic IPI interface yet. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/time.c | 212 ++++++++++++++++++++++++++++++------------------ 1 file changed, 131 insertions(+), 81 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index b73bbf31f43..6e09bc285ba 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -20,6 +20,8 @@ #include #include #include +#include +#include #include #include #include @@ -391,6 +393,15 @@ static void enable_sync_clock(void) atomic_set_mask(0x80000000, sw_ptr); } +/* Single threaded workqueue used for etr and stp sync events */ +static struct workqueue_struct *time_sync_wq; + +static void __init time_init_wq(void) +{ + if (!time_sync_wq) + time_sync_wq = create_singlethread_workqueue("timesync"); +} + /* * External Time Reference (ETR) code. */ @@ -483,17 +494,18 @@ static int __init etr_init(void) if (!test_bit(CLOCK_SYNC_HAS_ETR, &clock_sync_flags)) return 0; + time_init_wq(); /* Check if this machine has the steai instruction. */ if (etr_steai(&aib, ETR_STEAI_STEPPING_PORT) == 0) etr_steai_available = 1; setup_timer(&etr_timer, etr_timeout, 0UL); if (etr_port0_online) { set_bit(ETR_EVENT_PORT0_CHANGE, &etr_events); - schedule_work(&etr_work); + queue_work(time_sync_wq, &etr_work); } if (etr_port1_online) { set_bit(ETR_EVENT_PORT1_CHANGE, &etr_events); - schedule_work(&etr_work); + queue_work(time_sync_wq, &etr_work); } return 0; } @@ -520,7 +532,7 @@ void etr_switch_to_local(void) if (test_bit(CLOCK_SYNC_ETR, &clock_sync_flags)) disable_sync_clock(NULL); set_bit(ETR_EVENT_SWITCH_LOCAL, &etr_events); - schedule_work(&etr_work); + queue_work(time_sync_wq, &etr_work); } /* @@ -536,7 +548,7 @@ void etr_sync_check(void) if (test_bit(CLOCK_SYNC_ETR, &clock_sync_flags)) disable_sync_clock(NULL); set_bit(ETR_EVENT_SYNC_CHECK, &etr_events); - schedule_work(&etr_work); + queue_work(time_sync_wq, &etr_work); } /* @@ -560,13 +572,13 @@ static void etr_timing_alert(struct etr_irq_parm *intparm) * Both ports are not up-to-date now. */ set_bit(ETR_EVENT_PORT_ALERT, &etr_events); - schedule_work(&etr_work); + queue_work(time_sync_wq, &etr_work); } static void etr_timeout(unsigned long dummy) { set_bit(ETR_EVENT_UPDATE, &etr_events); - schedule_work(&etr_work); + queue_work(time_sync_wq, &etr_work); } /* @@ -673,14 +685,16 @@ static int etr_aib_follows(struct etr_aib *a1, struct etr_aib *a2, int p) } struct clock_sync_data { + atomic_t cpus; int in_sync; unsigned long long fixup_cc; + int etr_port; + struct etr_aib *etr_aib; }; -static void clock_sync_cpu_start(void *dummy) +static void clock_sync_cpu(struct clock_sync_data *sync) { - struct clock_sync_data *sync = dummy; - + atomic_dec(&sync->cpus); enable_sync_clock(); /* * This looks like a busy wait loop but it isn't. etr_sync_cpus @@ -706,39 +720,35 @@ static void clock_sync_cpu_start(void *dummy) fixup_clock_comparator(sync->fixup_cc); } -static void clock_sync_cpu_end(void *dummy) -{ -} - /* * Sync the TOD clock using the port refered to by aibp. This port * has to be enabled and the other port has to be disabled. The * last eacr update has to be more than 1.6 seconds in the past. */ -static int etr_sync_clock(struct etr_aib *aib, int port) +static int etr_sync_clock(void *data) { - struct etr_aib *sync_port; - struct clock_sync_data etr_sync; + static int first; unsigned long long clock, old_clock, delay, delta; - int follows; + struct clock_sync_data *etr_sync; + struct etr_aib *sync_port, *aib; + int port; int rc; - /* Check if the current aib is adjacent to the sync port aib. */ - sync_port = (port == 0) ? &etr_port0 : &etr_port1; - follows = etr_aib_follows(sync_port, aib, port); - memcpy(sync_port, aib, sizeof(*aib)); - if (!follows) - return -EAGAIN; + etr_sync = data; - /* - * Catch all other cpus and make them wait until we have - * successfully synced the clock. smp_call_function will - * return after all other cpus are in etr_sync_cpu_start. - */ - memset(&etr_sync, 0, sizeof(etr_sync)); - preempt_disable(); - smp_call_function(clock_sync_cpu_start, &etr_sync, 0); - local_irq_disable(); + if (xchg(&first, 1) == 1) { + /* Slave */ + clock_sync_cpu(etr_sync); + return 0; + } + + /* Wait until all other cpus entered the sync function. */ + while (atomic_read(&etr_sync->cpus) != 0) + cpu_relax(); + + port = etr_sync->etr_port; + aib = etr_sync->etr_aib; + sync_port = (port == 0) ? &etr_port0 : &etr_port1; enable_sync_clock(); /* Set clock to next OTE. */ @@ -755,16 +765,16 @@ static int etr_sync_clock(struct etr_aib *aib, int port) delay = (unsigned long long) (aib->edf2.etv - sync_port->edf2.etv) << 32; delta = adjust_time(old_clock, clock, delay); - etr_sync.fixup_cc = delta; + etr_sync->fixup_cc = delta; fixup_clock_comparator(delta); /* Verify that the clock is properly set. */ if (!etr_aib_follows(sync_port, aib, port)) { /* Didn't work. */ disable_sync_clock(NULL); - etr_sync.in_sync = -EAGAIN; + etr_sync->in_sync = -EAGAIN; rc = -EAGAIN; } else { - etr_sync.in_sync = 1; + etr_sync->in_sync = 1; rc = 0; } } else { @@ -772,12 +782,33 @@ static int etr_sync_clock(struct etr_aib *aib, int port) __ctl_clear_bit(0, 29); __ctl_clear_bit(14, 21); disable_sync_clock(NULL); - etr_sync.in_sync = -EAGAIN; + etr_sync->in_sync = -EAGAIN; rc = -EAGAIN; } - local_irq_enable(); - smp_call_function(clock_sync_cpu_end, NULL, 0); - preempt_enable(); + xchg(&first, 0); + return rc; +} + +static int etr_sync_clock_stop(struct etr_aib *aib, int port) +{ + struct clock_sync_data etr_sync; + struct etr_aib *sync_port; + int follows; + int rc; + + /* Check if the current aib is adjacent to the sync port aib. */ + sync_port = (port == 0) ? &etr_port0 : &etr_port1; + follows = etr_aib_follows(sync_port, aib, port); + memcpy(sync_port, aib, sizeof(*aib)); + if (!follows) + return -EAGAIN; + memset(&etr_sync, 0, sizeof(etr_sync)); + etr_sync.etr_aib = aib; + etr_sync.etr_port = port; + get_online_cpus(); + atomic_set(&etr_sync.cpus, num_online_cpus() - 1); + rc = stop_machine(etr_sync_clock, &etr_sync, &cpu_online_map); + put_online_cpus(); return rc; } @@ -934,7 +965,7 @@ static void etr_update_eacr(struct etr_eacr eacr) } /* - * ETR tasklet. In this function you'll find the main logic. In + * ETR work. In this function you'll find the main logic. In * particular this is the only function that calls etr_update_eacr(), * it "controls" the etr control register. */ @@ -1067,7 +1098,7 @@ static void etr_work_fn(struct work_struct *work) etr_update_eacr(eacr); set_bit(CLOCK_SYNC_ETR, &clock_sync_flags); if (now < etr_tolec + (1600000 << 12) || - etr_sync_clock(&aib, sync_port) != 0) { + etr_sync_clock_stop(&aib, sync_port) != 0) { /* Sync failed. Try again in 1/2 second. */ eacr.es = 0; etr_update_eacr(eacr); @@ -1156,13 +1187,13 @@ static ssize_t etr_online_store(struct sys_device *dev, return count; /* Nothing to do. */ etr_port0_online = value; set_bit(ETR_EVENT_PORT0_CHANGE, &etr_events); - schedule_work(&etr_work); + queue_work(time_sync_wq, &etr_work); } else { if (etr_port1_online == value) return count; /* Nothing to do. */ etr_port1_online = value; set_bit(ETR_EVENT_PORT1_CHANGE, &etr_events); - schedule_work(&etr_work); + queue_work(time_sync_wq, &etr_work); } return count; } @@ -1396,8 +1427,12 @@ static void __init stp_reset(void) static int __init stp_init(void) { - if (test_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags) && stp_online) - schedule_work(&stp_work); + if (!test_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags)) + return 0; + time_init_wq(); + if (!stp_online) + return 0; + queue_work(time_sync_wq, &stp_work); return 0; } @@ -1414,7 +1449,7 @@ arch_initcall(stp_init); static void stp_timing_alert(struct stp_irq_parm *intparm) { if (intparm->tsc || intparm->lac || intparm->tcpc) - schedule_work(&stp_work); + queue_work(time_sync_wq, &stp_work); } /* @@ -1428,7 +1463,7 @@ void stp_sync_check(void) if (!test_bit(CLOCK_SYNC_STP, &clock_sync_flags)) return; disable_sync_clock(NULL); - schedule_work(&stp_work); + queue_work(time_sync_wq, &stp_work); } /* @@ -1442,46 +1477,34 @@ void stp_island_check(void) if (!test_bit(CLOCK_SYNC_STP, &clock_sync_flags)) return; disable_sync_clock(NULL); - schedule_work(&stp_work); + queue_work(time_sync_wq, &stp_work); } -/* - * STP tasklet. Check for the STP state and take over the clock - * synchronization if the STP clock source is usable. - */ -static void stp_work_fn(struct work_struct *work) + +static int stp_sync_clock(void *data) { - struct clock_sync_data stp_sync; + static int first; unsigned long long old_clock, delta; + struct clock_sync_data *stp_sync; int rc; - if (!stp_online) { - chsc_sstpc(stp_page, STP_OP_CTRL, 0x0000); - return; - } + stp_sync = data; - rc = chsc_sstpc(stp_page, STP_OP_CTRL, 0xb0e0); - if (rc) - return; + if (xchg(&first, 1) == 1) { + /* Slave */ + clock_sync_cpu(stp_sync); + return 0; + } - rc = chsc_sstpi(stp_page, &stp_info, sizeof(struct stp_sstpi)); - if (rc || stp_info.c == 0) - return; + /* Wait until all other cpus entered the sync function. */ + while (atomic_read(&stp_sync->cpus) != 0) + cpu_relax(); - /* - * Catch all other cpus and make them wait until we have - * successfully synced the clock. smp_call_function will - * return after all other cpus are in clock_sync_cpu_start. - */ - memset(&stp_sync, 0, sizeof(stp_sync)); - preempt_disable(); - smp_call_function(clock_sync_cpu_start, &stp_sync, 0); - local_irq_disable(); enable_sync_clock(); set_bit(CLOCK_SYNC_STP, &clock_sync_flags); if (test_and_clear_bit(CLOCK_SYNC_ETR, &clock_sync_flags)) - schedule_work(&etr_work); + queue_work(time_sync_wq, &etr_work); rc = 0; if (stp_info.todoff[0] || stp_info.todoff[1] || @@ -1500,16 +1523,43 @@ static void stp_work_fn(struct work_struct *work) } if (rc) { disable_sync_clock(NULL); - stp_sync.in_sync = -EAGAIN; + stp_sync->in_sync = -EAGAIN; clear_bit(CLOCK_SYNC_STP, &clock_sync_flags); if (etr_port0_online || etr_port1_online) - schedule_work(&etr_work); + queue_work(time_sync_wq, &etr_work); } else - stp_sync.in_sync = 1; + stp_sync->in_sync = 1; + xchg(&first, 0); + return 0; +} - local_irq_enable(); - smp_call_function(clock_sync_cpu_end, NULL, 0); - preempt_enable(); +/* + * STP work. Check for the STP state and take over the clock + * synchronization if the STP clock source is usable. + */ +static void stp_work_fn(struct work_struct *work) +{ + struct clock_sync_data stp_sync; + int rc; + + if (!stp_online) { + chsc_sstpc(stp_page, STP_OP_CTRL, 0x0000); + return; + } + + rc = chsc_sstpc(stp_page, STP_OP_CTRL, 0xb0e0); + if (rc) + return; + + rc = chsc_sstpi(stp_page, &stp_info, sizeof(struct stp_sstpi)); + if (rc || stp_info.c == 0) + return; + + memset(&stp_sync, 0, sizeof(stp_sync)); + get_online_cpus(); + atomic_set(&stp_sync.cpus, num_online_cpus() - 1); + stop_machine(stp_sync_clock, &stp_sync, &cpu_online_map); + put_online_cpus(); } /* @@ -1618,7 +1668,7 @@ static ssize_t stp_online_store(struct sysdev_class *class, if (!test_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags)) return -EOPNOTSUPP; stp_online = value; - schedule_work(&stp_work); + queue_work(time_sync_wq, &stp_work); return count; } -- cgit v1.2.3-70-g09d2 From 0b3016b781abeabc502042c942cbc611e31250c7 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 25 Dec 2008 13:38:38 +0100 Subject: [S390] serialize stp/etr work The work function dispatched with schedule_work() can be run twice on different cpus because run_workqueue clears the WORK_STRUCT_PENDING bit and then executes the function. Another cpu can call schedule_work() again and run the work function a second time before the first call is completed. This patch serialized the etr and stp work function with a mutex. Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/time.c | 23 ++++++++++++++++++----- 1 file changed, 18 insertions(+), 5 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index 6e09bc285ba..49652856124 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -467,6 +467,7 @@ static struct timer_list etr_timer; static void etr_timeout(unsigned long dummy); static void etr_work_fn(struct work_struct *work); +static DEFINE_MUTEX(etr_work_mutex); static DECLARE_WORK(etr_work, etr_work_fn); /* @@ -976,6 +977,9 @@ static void etr_work_fn(struct work_struct *work) struct etr_aib aib; int sync_port; + /* prevent multiple execution. */ + mutex_lock(&etr_work_mutex); + /* Create working copy of etr_eacr. */ eacr = etr_eacr; @@ -991,7 +995,7 @@ static void etr_work_fn(struct work_struct *work) del_timer_sync(&etr_timer); etr_update_eacr(eacr); clear_bit(CLOCK_SYNC_ETR, &clock_sync_flags); - return; + goto out_unlock; } /* Store aib to get the current ETR status word. */ @@ -1078,7 +1082,7 @@ static void etr_work_fn(struct work_struct *work) eacr.es || sync_port < 0) { etr_update_eacr(eacr); etr_set_tolec_timeout(now); - return; + goto out_unlock; } /* @@ -1106,6 +1110,8 @@ static void etr_work_fn(struct work_struct *work) etr_set_sync_timeout(); } else etr_set_tolec_timeout(now); +out_unlock: + mutex_unlock(&etr_work_mutex); } /* @@ -1394,6 +1400,7 @@ static struct stp_sstpi stp_info; static void *stp_page; static void stp_work_fn(struct work_struct *work); +static DEFINE_MUTEX(stp_work_mutex); static DECLARE_WORK(stp_work, stp_work_fn); static int __init early_parse_stp(char *p) @@ -1542,24 +1549,30 @@ static void stp_work_fn(struct work_struct *work) struct clock_sync_data stp_sync; int rc; + /* prevent multiple execution. */ + mutex_lock(&stp_work_mutex); + if (!stp_online) { chsc_sstpc(stp_page, STP_OP_CTRL, 0x0000); - return; + goto out_unlock; } rc = chsc_sstpc(stp_page, STP_OP_CTRL, 0xb0e0); if (rc) - return; + goto out_unlock; rc = chsc_sstpi(stp_page, &stp_info, sizeof(struct stp_sstpi)); if (rc || stp_info.c == 0) - return; + goto out_unlock; memset(&stp_sync, 0, sizeof(stp_sync)); get_online_cpus(); atomic_set(&stp_sync.cpus, num_online_cpus() - 1); stop_machine(stp_sync_clock, &stp_sync, &cpu_online_map); put_online_cpus(); + +out_unlock: + mutex_unlock(&stp_work_mutex); } /* -- cgit v1.2.3-70-g09d2 From ca9fc75a68ee98812bb6d212405fea039421910b Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 25 Dec 2008 13:38:39 +0100 Subject: [S390] convert s390 to generic IPI infrastructure Since etr/stp don't need the old smp_call_function semantics anymore we can convert s390 to the generic IPI infrastructure. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/Kconfig | 1 + arch/s390/include/asm/sigp.h | 1 + arch/s390/include/asm/smp.h | 5 +- arch/s390/kernel/smp.c | 175 +++++-------------------------------------- 4 files changed, 24 insertions(+), 158 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 37bb37334de..7899f0bce65 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -73,6 +73,7 @@ mainmenu "Linux Kernel Configuration" config S390 def_bool y + select USE_GENERIC_SMP_HELPERS if SMP select HAVE_OPROFILE select HAVE_KPROBES select HAVE_KRETPROBES diff --git a/arch/s390/include/asm/sigp.h b/arch/s390/include/asm/sigp.h index e16d56f8dfe..ec403d4304f 100644 --- a/arch/s390/include/asm/sigp.h +++ b/arch/s390/include/asm/sigp.h @@ -61,6 +61,7 @@ typedef enum { ec_schedule=0, ec_call_function, + ec_call_function_single, ec_bit_last } ec_bit_sig; diff --git a/arch/s390/include/asm/smp.h b/arch/s390/include/asm/smp.h index ae89cf2478f..024b91e0623 100644 --- a/arch/s390/include/asm/smp.h +++ b/arch/s390/include/asm/smp.h @@ -91,8 +91,9 @@ extern int __cpu_up (unsigned int cpu); extern struct mutex smp_cpu_state_mutex; extern int smp_cpu_polarization[]; -extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *), - void *info, int wait); +extern void arch_send_call_function_single_ipi(int cpu); +extern void arch_send_call_function_ipi(cpumask_t mask); + #endif #ifndef CONFIG_SMP diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index b5595688a47..176a43e5b8b 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -77,159 +77,6 @@ static DEFINE_PER_CPU(struct cpu, cpu_devices); static void smp_ext_bitcall(int, ec_bit_sig); -/* - * Structure and data for __smp_call_function_map(). This is designed to - * minimise static memory requirements. It also looks cleaner. - */ -static DEFINE_SPINLOCK(call_lock); - -struct call_data_struct { - void (*func) (void *info); - void *info; - cpumask_t started; - cpumask_t finished; - int wait; -}; - -static struct call_data_struct *call_data; - -/* - * 'Call function' interrupt callback - */ -static void do_call_function(void) -{ - void (*func) (void *info) = call_data->func; - void *info = call_data->info; - int wait = call_data->wait; - - cpu_set(smp_processor_id(), call_data->started); - (*func)(info); - if (wait) - cpu_set(smp_processor_id(), call_data->finished);; -} - -static void __smp_call_function_map(void (*func) (void *info), void *info, - int wait, cpumask_t map) -{ - struct call_data_struct data; - int cpu, local = 0; - - /* - * Can deadlock when interrupts are disabled or if in wrong context. - */ - WARN_ON(irqs_disabled() || in_irq()); - - /* - * Check for local function call. We have to have the same call order - * as in on_each_cpu() because of machine_restart_smp(). - */ - if (cpu_isset(smp_processor_id(), map)) { - local = 1; - cpu_clear(smp_processor_id(), map); - } - - cpus_and(map, map, cpu_online_map); - if (cpus_empty(map)) - goto out; - - data.func = func; - data.info = info; - data.started = CPU_MASK_NONE; - data.wait = wait; - if (wait) - data.finished = CPU_MASK_NONE; - - call_data = &data; - - for_each_cpu_mask(cpu, map) - smp_ext_bitcall(cpu, ec_call_function); - - /* Wait for response */ - while (!cpus_equal(map, data.started)) - cpu_relax(); - if (wait) - while (!cpus_equal(map, data.finished)) - cpu_relax(); -out: - if (local) { - local_irq_disable(); - func(info); - local_irq_enable(); - } -} - -/* - * smp_call_function: - * @func: the function to run; this must be fast and non-blocking - * @info: an arbitrary pointer to pass to the function - * @wait: if true, wait (atomically) until function has completed on other CPUs - * - * Run a function on all other CPUs. - * - * You must not call this function with disabled interrupts, from a - * hardware interrupt handler or from a bottom half. - */ -int smp_call_function(void (*func) (void *info), void *info, int wait) -{ - cpumask_t map; - - spin_lock(&call_lock); - map = cpu_online_map; - cpu_clear(smp_processor_id(), map); - __smp_call_function_map(func, info, wait, map); - spin_unlock(&call_lock); - return 0; -} -EXPORT_SYMBOL(smp_call_function); - -/* - * smp_call_function_single: - * @cpu: the CPU where func should run - * @func: the function to run; this must be fast and non-blocking - * @info: an arbitrary pointer to pass to the function - * @wait: if true, wait (atomically) until function has completed on other CPUs - * - * Run a function on one processor. - * - * You must not call this function with disabled interrupts, from a - * hardware interrupt handler or from a bottom half. - */ -int smp_call_function_single(int cpu, void (*func) (void *info), void *info, - int wait) -{ - spin_lock(&call_lock); - __smp_call_function_map(func, info, wait, cpumask_of_cpu(cpu)); - spin_unlock(&call_lock); - return 0; -} -EXPORT_SYMBOL(smp_call_function_single); - -/** - * smp_call_function_mask(): Run a function on a set of other CPUs. - * @mask: The set of cpus to run on. Must not include the current cpu. - * @func: The function to run. This must be fast and non-blocking. - * @info: An arbitrary pointer to pass to the function. - * @wait: If true, wait (atomically) until function has completed on other CPUs. - * - * Returns 0 on success, else a negative status code. - * - * If @wait is true, then returns once @func has returned; otherwise - * it returns just before the target cpu calls @func. - * - * You must not call this function with disabled interrupts or from a - * hardware interrupt handler or from a bottom half handler. - */ -int smp_call_function_mask(cpumask_t mask, void (*func)(void *), void *info, - int wait) -{ - spin_lock(&call_lock); - cpu_clear(smp_processor_id(), mask); - __smp_call_function_map(func, info, wait, mask); - spin_unlock(&call_lock); - return 0; -} -EXPORT_SYMBOL(smp_call_function_mask); - void smp_send_stop(void) { int cpu, rc; @@ -271,7 +118,10 @@ static void do_ext_call_interrupt(__u16 code) bits = xchg(&S390_lowcore.ext_call_fast, 0); if (test_bit(ec_call_function, &bits)) - do_call_function(); + generic_smp_call_function_interrupt(); + + if (test_bit(ec_call_function_single, &bits)) + generic_smp_call_function_single_interrupt(); } /* @@ -288,6 +138,19 @@ static void smp_ext_bitcall(int cpu, ec_bit_sig sig) udelay(10); } +void arch_send_call_function_ipi(cpumask_t mask) +{ + int cpu; + + for_each_cpu_mask(cpu, mask) + smp_ext_bitcall(cpu, ec_call_function); +} + +void arch_send_call_function_single_ipi(int cpu) +{ + smp_ext_bitcall(cpu, ec_call_function_single); +} + #ifndef CONFIG_64BIT /* * this function sends a 'purge tlb' signal to another CPU. @@ -588,9 +451,9 @@ int __cpuinit start_secondary(void *cpuvoid) /* call cpu notifiers */ notify_cpu_starting(smp_processor_id()); /* Mark this cpu as online */ - spin_lock(&call_lock); + ipi_call_lock(); cpu_set(smp_processor_id(), cpu_online_map); - spin_unlock(&call_lock); + ipi_call_unlock(); /* Switch on interrupts */ local_irq_enable(); /* Print info about this processor */ -- cgit v1.2.3-70-g09d2 From 320c04c068c62b71fe9ea55e06e4968b4edc9e48 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 25 Dec 2008 13:38:40 +0100 Subject: [S390] Move stfle to header file. stfle will be needed by the ap_bus module to figure out wether the AP queue adapter interruption facility is installed. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/system.h | 20 ++++++++++++++++++-- arch/s390/kernel/setup.c | 17 ----------------- 2 files changed, 18 insertions(+), 19 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/include/asm/system.h b/arch/s390/include/asm/system.h index 819e7d99ca0..c17a3cfbe24 100644 --- a/arch/s390/include/asm/system.h +++ b/arch/s390/include/asm/system.h @@ -12,6 +12,7 @@ #define __ASM_SYSTEM_H #include +#include #include #include #include @@ -413,8 +414,6 @@ __set_psw_mask(unsigned long mask) #define local_mcck_enable() __set_psw_mask(psw_kernel_bits) #define local_mcck_disable() __set_psw_mask(psw_kernel_bits & ~PSW_MASK_MCHECK) -int stfle(unsigned long long *list, int doublewords); - #ifdef CONFIG_SMP extern void smp_ctl_set_bit(int cr, int bit); @@ -438,6 +437,23 @@ static inline unsigned int stfl(void) return S390_lowcore.stfl_fac_list; } +static inline int __stfle(unsigned long long *list, int doublewords) +{ + typedef struct { unsigned long long _[doublewords]; } addrtype; + register unsigned long __nr asm("0") = doublewords - 1; + + asm volatile(".insn s,0xb2b00000,%0" /* stfle */ + : "=m" (*(addrtype *) list), "+d" (__nr) : : "cc"); + return __nr + 1; +} + +static inline int stfle(unsigned long long *list, int doublewords) +{ + if (!(stfl() & (1UL << 24))) + return -EOPNOTSUPP; + return __stfle(list, doublewords); +} + static inline unsigned short stap(void) { unsigned short cpu_address; diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 400b040df7f..a0d44bc0c38 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -651,23 +651,6 @@ setup_memory(void) #endif } -static int __init __stfle(unsigned long long *list, int doublewords) -{ - typedef struct { unsigned long long _[doublewords]; } addrtype; - register unsigned long __nr asm("0") = doublewords - 1; - - asm volatile(".insn s,0xb2b00000,%0" /* stfle */ - : "=m" (*(addrtype *) list), "+d" (__nr) : : "cc"); - return __nr + 1; -} - -int __init stfle(unsigned long long *list, int doublewords) -{ - if (!(stfl() & (1UL << 24))) - return -EOPNOTSUPP; - return __stfle(list, doublewords); -} - /* * Setup hardware capabilities. */ -- cgit v1.2.3-70-g09d2 From cb17a6364a29b4dfe5bbb00696032fb63d780157 Mon Sep 17 00:00:00 2001 From: Felix Beck Date: Thu, 25 Dec 2008 13:38:41 +0100 Subject: [S390] zcrypt: Use of Thin Interrupts When the machine supports AP adapter interrupts polling will be switched off at module initialization and the driver will work in interrupt mode. Signed-off-by: Felix Beck Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/isc.h | 1 + drivers/s390/crypto/ap_bus.c | 174 ++++++++++++++++++++++++++++++++++++++++++- drivers/s390/crypto/ap_bus.h | 6 +- 3 files changed, 177 insertions(+), 4 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/include/asm/isc.h b/arch/s390/include/asm/isc.h index 34bb8916db4..1420a111594 100644 --- a/arch/s390/include/asm/isc.h +++ b/arch/s390/include/asm/isc.h @@ -17,6 +17,7 @@ #define CHSC_SCH_ISC 7 /* CHSC subchannels */ /* Adapter interrupts. */ #define QDIO_AIRQ_ISC IO_SCH_ISC /* I/O subchannel in qdio mode */ +#define AP_ISC 6 /* adjunct processor (crypto) devices */ /* Functions for registration of I/O interruption subclasses */ void isc_register(unsigned int isc); diff --git a/drivers/s390/crypto/ap_bus.c b/drivers/s390/crypto/ap_bus.c index e3fe6838293..2335be5f9bf 100644 --- a/drivers/s390/crypto/ap_bus.c +++ b/drivers/s390/crypto/ap_bus.c @@ -5,6 +5,7 @@ * Author(s): Cornelia Huck * Martin Schwidefsky * Ralph Wuerthner + * Felix Beck * * Adjunct processor bus. * @@ -34,6 +35,10 @@ #include #include #include +#include +#include +#include +#include #include #include @@ -46,6 +51,7 @@ static enum hrtimer_restart ap_poll_timeout(struct hrtimer *); static int ap_poll_thread_start(void); static void ap_poll_thread_stop(void); static void ap_request_timeout(unsigned long); +static inline void ap_schedule_poll_timer(void); /* * Module description. @@ -80,18 +86,28 @@ static int ap_config_time = AP_CONFIG_TIME; static DECLARE_WORK(ap_config_work, ap_scan_bus); /* - * Tasklet & timer for AP request polling. + * Tasklet & timer for AP request polling and interrupts */ static DECLARE_TASKLET(ap_tasklet, ap_poll_all, 0); static atomic_t ap_poll_requests = ATOMIC_INIT(0); static DECLARE_WAIT_QUEUE_HEAD(ap_poll_wait); static struct task_struct *ap_poll_kthread = NULL; static DEFINE_MUTEX(ap_poll_thread_mutex); +static void *ap_interrupt_indicator; static struct hrtimer ap_poll_timer; /* In LPAR poll with 4kHz frequency. Poll every 250000 nanoseconds. * If z/VM change to 1500000 nanoseconds to adjust to z/VM polling.*/ static unsigned long long poll_timeout = 250000; +/** + * ap_using_interrupts() - Returns non-zero if interrupt support is + * available. + */ +static inline int ap_using_interrupts(void) +{ + return ap_interrupt_indicator != NULL; +} + /** * ap_intructions_available() - Test if AP instructions are available. * @@ -112,6 +128,23 @@ static inline int ap_instructions_available(void) return reg1; } +/** + * ap_interrupts_available(): Test if AP interrupts are available. + * + * Returns 1 if AP interrupts are available. + */ +static int ap_interrupts_available(void) +{ + unsigned long long facility_bits[2]; + + if (stfle(facility_bits, 2) <= 1) + return 0; + if (!(facility_bits[0] & (1ULL << 61)) || + !(facility_bits[1] & (1ULL << 62))) + return 0; + return 1; +} + /** * ap_test_queue(): Test adjunct processor queue. * @qid: The AP queue number @@ -152,6 +185,80 @@ static inline struct ap_queue_status ap_reset_queue(ap_qid_t qid) return reg1; } +#ifdef CONFIG_64BIT +/** + * ap_queue_interruption_control(): Enable interruption for a specific AP. + * @qid: The AP queue number + * @ind: The notification indicator byte + * + * Returns AP queue status. + */ +static inline struct ap_queue_status +ap_queue_interruption_control(ap_qid_t qid, void *ind) +{ + register unsigned long reg0 asm ("0") = qid | 0x03000000UL; + register unsigned long reg1_in asm ("1") = 0x0000800000000000UL | AP_ISC; + register struct ap_queue_status reg1_out asm ("1"); + register void *reg2 asm ("2") = ind; + asm volatile( + ".long 0xb2af0000" /* PQAP(RAPQ) */ + : "+d" (reg0), "+d" (reg1_in), "=d" (reg1_out), "+d" (reg2) + : + : "cc" ); + return reg1_out; +} +#endif + +/** + * ap_queue_enable_interruption(): Enable interruption on an AP. + * @qid: The AP queue number + * @ind: the notification indicator byte + * + * Enables interruption on AP queue via ap_queue_interruption_control(). Based + * on the return value it waits a while and tests the AP queue if interrupts + * have been switched on using ap_test_queue(). + */ +static int ap_queue_enable_interruption(ap_qid_t qid, void *ind) +{ +#ifdef CONFIG_64BIT + struct ap_queue_status status; + int t_depth, t_device_type, rc, i; + + rc = -EBUSY; + status = ap_queue_interruption_control(qid, ind); + + for (i = 0; i < AP_MAX_RESET; i++) { + switch (status.response_code) { + case AP_RESPONSE_NORMAL: + if (status.int_enabled) + return 0; + break; + case AP_RESPONSE_RESET_IN_PROGRESS: + case AP_RESPONSE_BUSY: + break; + case AP_RESPONSE_Q_NOT_AVAIL: + case AP_RESPONSE_DECONFIGURED: + case AP_RESPONSE_CHECKSTOPPED: + case AP_RESPONSE_INVALID_ADDRESS: + return -ENODEV; + case AP_RESPONSE_OTHERWISE_CHANGED: + if (status.int_enabled) + return 0; + break; + default: + break; + } + if (i < AP_MAX_RESET - 1) { + udelay(5); + status = ap_test_queue(qid, &t_depth, &t_device_type); + } + } + return rc; +#else + return -EINVAL; +#endif +} + /** * __ap_send(): Send message to adjunct processor queue. * @qid: The AP queue number @@ -295,6 +402,11 @@ static int ap_query_queue(ap_qid_t qid, int *queue_depth, int *device_type) case AP_RESPONSE_CHECKSTOPPED: rc = -ENODEV; break; + case AP_RESPONSE_INVALID_ADDRESS: + rc = -ENODEV; + break; + case AP_RESPONSE_OTHERWISE_CHANGED: + break; case AP_RESPONSE_BUSY: break; default: @@ -345,6 +457,15 @@ static int ap_init_queue(ap_qid_t qid) status = ap_test_queue(qid, &dummy, &dummy); } } + if (rc == 0 && ap_using_interrupts()) { + rc = ap_queue_enable_interruption(qid, ap_interrupt_indicator); + /* If interruption mode is supported by the machine, + * but an AP can not be enabled for interruption then + * the AP will be discarded. */ + if (rc) + pr_err("Registering adapter interrupts for " + "AP %d failed\n", AP_QID_DEVICE(qid)); + } return rc; } @@ -599,6 +720,14 @@ static ssize_t ap_config_time_show(struct bus_type *bus, char *buf) return snprintf(buf, PAGE_SIZE, "%d\n", ap_config_time); } +static ssize_t ap_interrupts_show(struct bus_type *bus, char *buf) +{ + return snprintf(buf, PAGE_SIZE, "%d\n", + ap_using_interrupts() ? 1 : 0); +} + +static BUS_ATTR(ap_interrupts, 0444, ap_interrupts_show, NULL); + static ssize_t ap_config_time_store(struct bus_type *bus, const char *buf, size_t count) { @@ -653,7 +782,8 @@ static ssize_t poll_timeout_store(struct bus_type *bus, const char *buf, ktime_t hr_time; /* 120 seconds = maximum poll interval */ - if (sscanf(buf, "%llu\n", &time) != 1 || time < 1 || time > 120000000000) + if (sscanf(buf, "%llu\n", &time) != 1 || time < 1 || + time > 120000000000ULL) return -EINVAL; poll_timeout = time; hr_time = ktime_set(0, poll_timeout); @@ -672,6 +802,7 @@ static struct bus_attribute *const ap_bus_attrs[] = { &bus_attr_ap_domain, &bus_attr_config_time, &bus_attr_poll_thread, + &bus_attr_ap_interrupts, &bus_attr_poll_timeout, NULL, }; @@ -814,6 +945,11 @@ out: return rc; } +static void ap_interrupt_handler(void *unused1, void *unused2) +{ + tasklet_schedule(&ap_tasklet); +} + /** * __ap_scan_bus(): Scan the AP bus. * @dev: Pointer to device @@ -928,6 +1064,8 @@ ap_config_timeout(unsigned long ptr) */ static inline void ap_schedule_poll_timer(void) { + if (ap_using_interrupts()) + return; if (hrtimer_is_queued(&ap_poll_timer)) return; hrtimer_start(&ap_poll_timer, ktime_set(0, poll_timeout), @@ -1207,6 +1345,12 @@ static void ap_poll_all(unsigned long dummy) unsigned long flags; struct ap_device *ap_dev; + /* Reset the indicator if interrupts are used. Thus new interrupts can + * be received. Doing it in the beginning of the tasklet is therefor + * important that no requests on any AP get lost. + */ + if (ap_using_interrupts()) + xchg((u8 *)ap_interrupt_indicator, 0); do { flags = 0; spin_lock(&ap_device_lock); @@ -1268,6 +1412,8 @@ static int ap_poll_thread_start(void) { int rc; + if (ap_using_interrupts()) + return 0; mutex_lock(&ap_poll_thread_mutex); if (!ap_poll_kthread) { ap_poll_kthread = kthread_run(ap_poll_thread, NULL, "appoll"); @@ -1301,8 +1447,12 @@ static void ap_request_timeout(unsigned long data) { struct ap_device *ap_dev = (struct ap_device *) data; - if (ap_dev->reset == AP_RESET_ARMED) + if (ap_dev->reset == AP_RESET_ARMED) { ap_dev->reset = AP_RESET_DO; + + if (ap_using_interrupts()) + tasklet_schedule(&ap_tasklet); + } } static void ap_reset_domain(void) @@ -1345,6 +1495,16 @@ int __init ap_module_init(void) printk(KERN_WARNING "AP instructions not installed.\n"); return -ENODEV; } + if (ap_interrupts_available()) { + isc_register(AP_ISC); + ap_interrupt_indicator = s390_register_adapter_interrupt( + &ap_interrupt_handler, NULL, AP_ISC); + if (IS_ERR(ap_interrupt_indicator)) { + ap_interrupt_indicator = NULL; + isc_unregister(AP_ISC); + } + } + register_reset_call(&ap_reset_call); /* Create /sys/bus/ap. */ @@ -1408,6 +1568,10 @@ out_bus: bus_unregister(&ap_bus_type); out: unregister_reset_call(&ap_reset_call); + if (ap_using_interrupts()) { + s390_unregister_adapter_interrupt(ap_interrupt_indicator, AP_ISC); + isc_unregister(AP_ISC); + } return rc; } @@ -1443,6 +1607,10 @@ void ap_module_exit(void) bus_remove_file(&ap_bus_type, ap_bus_attrs[i]); bus_unregister(&ap_bus_type); unregister_reset_call(&ap_reset_call); + if (ap_using_interrupts()) { + s390_unregister_adapter_interrupt(ap_interrupt_indicator, AP_ISC); + isc_unregister(AP_ISC); + } } #ifndef CONFIG_ZCRYPT_MONOLITHIC diff --git a/drivers/s390/crypto/ap_bus.h b/drivers/s390/crypto/ap_bus.h index 446378b308f..a3536224180 100644 --- a/drivers/s390/crypto/ap_bus.h +++ b/drivers/s390/crypto/ap_bus.h @@ -5,6 +5,7 @@ * Author(s): Cornelia Huck * Martin Schwidefsky * Ralph Wuerthner + * Felix Beck * * Adjunct processor bus header file. * @@ -67,7 +68,8 @@ struct ap_queue_status { unsigned int queue_empty : 1; unsigned int replies_waiting : 1; unsigned int queue_full : 1; - unsigned int pad1 : 5; + unsigned int pad1 : 4; + unsigned int int_enabled : 1; unsigned int response_code : 8; unsigned int pad2 : 16; }; @@ -78,6 +80,8 @@ struct ap_queue_status { #define AP_RESPONSE_DECONFIGURED 0x03 #define AP_RESPONSE_CHECKSTOPPED 0x04 #define AP_RESPONSE_BUSY 0x05 +#define AP_RESPONSE_INVALID_ADDRESS 0x06 +#define AP_RESPONSE_OTHERWISE_CHANGED 0x07 #define AP_RESPONSE_Q_FULL 0x10 #define AP_RESPONSE_NO_PENDING_REPLY 0x10 #define AP_RESPONSE_INDEX_TOO_BIG 0x11 -- cgit v1.2.3-70-g09d2 From bbd50e172f75b1d12ef9b1bcf593b51a44199016 Mon Sep 17 00:00:00 2001 From: Jan Glauber Date: Thu, 25 Dec 2008 13:38:43 +0100 Subject: [S390] qdio: fix qeth port count detection qeth needs to get the port count information before qdio has allocated a page for the chsc operation. Extend qdio_get_ssqd_desc() to store the data in the specified structure. Signed-off-by: Jan Glauber Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/qdio.h | 16 ++++++++-------- drivers/s390/cio/qdio.h | 3 +++ drivers/s390/cio/qdio_main.c | 18 +++++++++--------- drivers/s390/cio/qdio_setup.c | 33 ++++++++++++++++++++++++--------- drivers/s390/net/qeth_core_main.c | 15 +++++++++++---- 5 files changed, 55 insertions(+), 30 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/include/asm/qdio.h b/arch/s390/include/asm/qdio.h index 4734c3f0535..27fc1746de1 100644 --- a/arch/s390/include/asm/qdio.h +++ b/arch/s390/include/asm/qdio.h @@ -373,16 +373,16 @@ struct qdio_initialize { #define QDIO_FLAG_SYNC_OUTPUT 0x02 #define QDIO_FLAG_PCI_OUT 0x10 -extern int qdio_initialize(struct qdio_initialize *init_data); -extern int qdio_allocate(struct qdio_initialize *init_data); -extern int qdio_establish(struct qdio_initialize *init_data); +extern int qdio_initialize(struct qdio_initialize *); +extern int qdio_allocate(struct qdio_initialize *); +extern int qdio_establish(struct qdio_initialize *); extern int qdio_activate(struct ccw_device *); -extern int do_QDIO(struct ccw_device*, unsigned int flags, - int q_nr, int qidx, int count); -extern int qdio_cleanup(struct ccw_device*, int how); -extern int qdio_shutdown(struct ccw_device*, int how); +extern int do_QDIO(struct ccw_device *cdev, unsigned int callflags, + int q_nr, int bufnr, int count); +extern int qdio_cleanup(struct ccw_device*, int); +extern int qdio_shutdown(struct ccw_device*, int); extern int qdio_free(struct ccw_device *); -extern struct qdio_ssqd_desc *qdio_get_ssqd_desc(struct ccw_device *cdev); +extern int qdio_get_ssqd_desc(struct ccw_device *dev, struct qdio_ssqd_desc*); #endif /* __QDIO_H__ */ diff --git a/drivers/s390/cio/qdio.h b/drivers/s390/cio/qdio.h index e3ea1d5f281..e4e33839c7f 100644 --- a/drivers/s390/cio/qdio.h +++ b/drivers/s390/cio/qdio.h @@ -378,6 +378,9 @@ void qdio_int_handler(struct ccw_device *cdev, unsigned long intparm, int qdio_allocate_qs(struct qdio_irq *irq_ptr, int nr_input_qs, int nr_output_qs); void qdio_setup_ssqd_info(struct qdio_irq *irq_ptr); +int qdio_setup_get_ssqd(struct qdio_irq *irq_ptr, + struct subchannel_id *schid, + struct qdio_ssqd_desc *data); int qdio_setup_irq(struct qdio_initialize *init_data); void qdio_print_subchannel_info(struct qdio_irq *irq_ptr, struct ccw_device *cdev); diff --git a/drivers/s390/cio/qdio_main.c b/drivers/s390/cio/qdio_main.c index 7c865915199..7572a005762 100644 --- a/drivers/s390/cio/qdio_main.c +++ b/drivers/s390/cio/qdio_main.c @@ -1129,23 +1129,23 @@ void qdio_int_handler(struct ccw_device *cdev, unsigned long intparm, /** * qdio_get_ssqd_desc - get qdio subchannel description * @cdev: ccw device to get description for + * @data: where to store the ssqd * - * Returns a pointer to the saved qdio subchannel description, - * or NULL for not setup qdio devices. + * Returns 0 or an error code. The results of the chsc are stored in the + * specified structure. */ -struct qdio_ssqd_desc *qdio_get_ssqd_desc(struct ccw_device *cdev) +int qdio_get_ssqd_desc(struct ccw_device *cdev, + struct qdio_ssqd_desc *data) { - struct qdio_irq *irq_ptr; char dbf_text[15]; + if (!cdev || !cdev->private) + return -EINVAL; + sprintf(dbf_text, "qssq%4x", cdev->private->schid.sch_no); QDIO_DBF_TEXT0(0, setup, dbf_text); - irq_ptr = cdev->private->qdio_data; - if (!irq_ptr) - return NULL; - - return &irq_ptr->ssqd_desc; + return qdio_setup_get_ssqd(NULL, &cdev->private->schid, data); } EXPORT_SYMBOL_GPL(qdio_get_ssqd_desc); diff --git a/drivers/s390/cio/qdio_setup.c b/drivers/s390/cio/qdio_setup.c index a0b6b46e746..22ee0272a10 100644 --- a/drivers/s390/cio/qdio_setup.c +++ b/drivers/s390/cio/qdio_setup.c @@ -243,22 +243,31 @@ no_qebsm: QDIO_DBF_TEXT0(0, setup, "noV=V"); } -static int __get_ssqd_info(struct qdio_irq *irq_ptr) +/* + * If there is a qdio_irq we use the chsc_page and store the information + * in the qdio_irq, otherwise we copy it to the specified structure. + */ +int qdio_setup_get_ssqd(struct qdio_irq *irq_ptr, + struct subchannel_id *schid, + struct qdio_ssqd_desc *data) { struct chsc_ssqd_area *ssqd; int rc; QDIO_DBF_TEXT0(0, setup, "getssqd"); - ssqd = (struct chsc_ssqd_area *)irq_ptr->chsc_page; + if (irq_ptr != NULL) + ssqd = (struct chsc_ssqd_area *)irq_ptr->chsc_page; + else + ssqd = (struct chsc_ssqd_area *)__get_free_page(GFP_KERNEL); memset(ssqd, 0, PAGE_SIZE); ssqd->request = (struct chsc_header) { .length = 0x0010, .code = 0x0024, }; - ssqd->first_sch = irq_ptr->schid.sch_no; - ssqd->last_sch = irq_ptr->schid.sch_no; - ssqd->ssid = irq_ptr->schid.ssid; + ssqd->first_sch = schid->sch_no; + ssqd->last_sch = schid->sch_no; + ssqd->ssid = schid->ssid; if (chsc(ssqd)) return -EIO; @@ -268,11 +277,17 @@ static int __get_ssqd_info(struct qdio_irq *irq_ptr) if (!(ssqd->qdio_ssqd.flags & CHSC_FLAG_QDIO_CAPABILITY) || !(ssqd->qdio_ssqd.flags & CHSC_FLAG_VALIDITY) || - (ssqd->qdio_ssqd.sch != irq_ptr->schid.sch_no)) + (ssqd->qdio_ssqd.sch != schid->sch_no)) return -EINVAL; - memcpy(&irq_ptr->ssqd_desc, &ssqd->qdio_ssqd, - sizeof(struct qdio_ssqd_desc)); + if (irq_ptr != NULL) + memcpy(&irq_ptr->ssqd_desc, &ssqd->qdio_ssqd, + sizeof(struct qdio_ssqd_desc)); + else { + memcpy(data, &ssqd->qdio_ssqd, + sizeof(struct qdio_ssqd_desc)); + free_page((unsigned long)ssqd); + } return 0; } @@ -282,7 +297,7 @@ void qdio_setup_ssqd_info(struct qdio_irq *irq_ptr) char dbf_text[15]; int rc; - rc = __get_ssqd_info(irq_ptr); + rc = qdio_setup_get_ssqd(irq_ptr, &irq_ptr->schid, NULL); if (rc) { QDIO_DBF_TEXT2(0, setup, "ssqdasig"); sprintf(dbf_text, "schn%4x", irq_ptr->schid.sch_no); diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index 52d26592c72..ebdc549da53 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -3757,7 +3757,7 @@ static int qeth_core_driver_group(const char *buf, struct device *root_dev, int qeth_core_hardsetup_card(struct qeth_card *card) { - struct qdio_ssqd_desc *qdio_ssqd; + struct qdio_ssqd_desc *ssqd; int retries = 3; int mpno = 0; int rc; @@ -3792,9 +3792,16 @@ retry: return rc; } - qdio_ssqd = qdio_get_ssqd_desc(CARD_DDEV(card)); - if (qdio_ssqd) - mpno = qdio_ssqd->pcnt; + ssqd = kmalloc(sizeof(struct qdio_ssqd_desc), GFP_KERNEL); + if (!ssqd) { + rc = -ENOMEM; + goto out; + } + rc = qdio_get_ssqd_desc(CARD_DDEV(card), ssqd); + if (rc == 0) + mpno = ssqd->pcnt; + kfree(ssqd); + if (mpno) mpno = min(mpno - 1, QETH_MAX_PORTNO); if (card->info.portno > mpno) { -- cgit v1.2.3-70-g09d2 From 22f9934767f49012ffbae753b28b8055bd28348f Mon Sep 17 00:00:00 2001 From: Jan Glauber Date: Thu, 25 Dec 2008 13:38:46 +0100 Subject: [S390] qdio: rework debug feature logging - make qdio_trace a per device view - remove s390dbf exceptions - remove CONFIG_QDIO_DEBUG, not needed anymore if we check for the level before calling sprintf - use snprintf for dbf entries - add start markers to see if the dbf view wrapped - add a global error view for all queues Signed-off-by: Jan Glauber Signed-off-by: Martin Schwidefsky --- arch/s390/Kconfig | 10 -- drivers/s390/cio/qdio.h | 3 + drivers/s390/cio/qdio_debug.c | 97 ++++++------ drivers/s390/cio/qdio_debug.h | 112 +++++++------- drivers/s390/cio/qdio_main.c | 324 ++++++++++++---------------------------- drivers/s390/cio/qdio_setup.c | 110 +++++--------- drivers/s390/cio/qdio_thinint.c | 25 +--- 7 files changed, 238 insertions(+), 443 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 7899f0bce65..90ba5d35cf3 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -347,16 +347,6 @@ config QDIO If unsure, say Y. -config QDIO_DEBUG - bool "Extended debugging information" - depends on QDIO - help - Say Y here to get extended debugging output in - /sys/kernel/debug/s390dbf/qdio... - Warning: this option reduces the performance of the QDIO module. - - If unsure, say N. - config CHSC_SCH tristate "Support for CHSC subchannels" help diff --git a/drivers/s390/cio/qdio.h b/drivers/s390/cio/qdio.h index e4e33839c7f..7b50882577d 100644 --- a/drivers/s390/cio/qdio.h +++ b/drivers/s390/cio/qdio.h @@ -10,6 +10,7 @@ #include #include +#include #include "chsc.h" #define QDIO_BUSY_BIT_PATIENCE 100 /* 100 microseconds */ @@ -300,11 +301,13 @@ struct qdio_irq { struct qdio_q *input_qs[QDIO_MAX_QUEUES_PER_IRQ]; struct qdio_q *output_qs[QDIO_MAX_QUEUES_PER_IRQ]; + debug_info_t *debug_area; struct mutex setup_mutex; }; /* helper functions */ #define queue_type(q) q->irq_ptr->qib.qfmt +#define SCH_NO(q) (q->irq_ptr->schid.sch_no) #define is_thinint_irq(irq) \ (irq->qib.qfmt == QDIO_IQDIO_QFMT || \ diff --git a/drivers/s390/cio/qdio_debug.c b/drivers/s390/cio/qdio_debug.c index f05590355be..0add07c1b5d 100644 --- a/drivers/s390/cio/qdio_debug.c +++ b/drivers/s390/cio/qdio_debug.c @@ -14,7 +14,7 @@ #include "qdio.h" debug_info_t *qdio_dbf_setup; -debug_info_t *qdio_dbf_trace; +debug_info_t *qdio_dbf_error; static struct dentry *debugfs_root; #define MAX_DEBUGFS_QUEUES 32 @@ -22,59 +22,33 @@ static struct dentry *debugfs_queues[MAX_DEBUGFS_QUEUES] = { NULL }; static DEFINE_MUTEX(debugfs_mutex); #define QDIO_DEBUGFS_NAME_LEN 40 -void qdio_allocate_do_dbf(struct qdio_initialize *init_data) +void qdio_allocate_dbf(struct qdio_initialize *init_data, + struct qdio_irq *irq_ptr) { - char dbf_text[20]; - - sprintf(dbf_text, "qfmt:%x", init_data->q_format); - QDIO_DBF_TEXT0(0, setup, dbf_text); - QDIO_DBF_HEX0(0, setup, init_data->adapter_name, 8); - sprintf(dbf_text, "qpff%4x", init_data->qib_param_field_format); - QDIO_DBF_TEXT0(0, setup, dbf_text); - QDIO_DBF_HEX0(0, setup, &init_data->qib_param_field, sizeof(void *)); - QDIO_DBF_HEX0(0, setup, &init_data->input_slib_elements, sizeof(void *)); - QDIO_DBF_HEX0(0, setup, &init_data->output_slib_elements, sizeof(void *)); - sprintf(dbf_text, "niq:%4x", init_data->no_input_qs); - QDIO_DBF_TEXT0(0, setup, dbf_text); - sprintf(dbf_text, "noq:%4x", init_data->no_output_qs); - QDIO_DBF_TEXT0(0, setup, dbf_text); - QDIO_DBF_HEX0(0, setup, &init_data->input_handler, sizeof(void *)); - QDIO_DBF_HEX0(0, setup, &init_data->output_handler, sizeof(void *)); - QDIO_DBF_HEX0(0, setup, &init_data->int_parm, sizeof(long)); - QDIO_DBF_HEX0(0, setup, &init_data->flags, sizeof(long)); - QDIO_DBF_HEX0(0, setup, &init_data->input_sbal_addr_array, sizeof(void *)); - QDIO_DBF_HEX0(0, setup, &init_data->output_sbal_addr_array, sizeof(void *)); -} - -static void qdio_unregister_dbf_views(void) -{ - if (qdio_dbf_setup) - debug_unregister(qdio_dbf_setup); - if (qdio_dbf_trace) - debug_unregister(qdio_dbf_trace); -} - -static int qdio_register_dbf_views(void) -{ - qdio_dbf_setup = debug_register("qdio_setup", QDIO_DBF_SETUP_PAGES, - QDIO_DBF_SETUP_NR_AREAS, - QDIO_DBF_SETUP_LEN); - if (!qdio_dbf_setup) - goto oom; - debug_register_view(qdio_dbf_setup, &debug_hex_ascii_view); - debug_set_level(qdio_dbf_setup, QDIO_DBF_SETUP_LEVEL); - - qdio_dbf_trace = debug_register("qdio_trace", QDIO_DBF_TRACE_PAGES, - QDIO_DBF_TRACE_NR_AREAS, - QDIO_DBF_TRACE_LEN); - if (!qdio_dbf_trace) - goto oom; - debug_register_view(qdio_dbf_trace, &debug_hex_ascii_view); - debug_set_level(qdio_dbf_trace, QDIO_DBF_TRACE_LEVEL); - return 0; -oom: - qdio_unregister_dbf_views(); - return -ENOMEM; + char text[20]; + + DBF_EVENT("qfmt:%1d", init_data->q_format); + DBF_HEX(init_data->adapter_name, 8); + DBF_EVENT("qpff%4x", init_data->qib_param_field_format); + DBF_HEX(&init_data->qib_param_field, sizeof(void *)); + DBF_HEX(&init_data->input_slib_elements, sizeof(void *)); + DBF_HEX(&init_data->output_slib_elements, sizeof(void *)); + DBF_EVENT("niq:%1d noq:%1d", init_data->no_input_qs, + init_data->no_output_qs); + DBF_HEX(&init_data->input_handler, sizeof(void *)); + DBF_HEX(&init_data->output_handler, sizeof(void *)); + DBF_HEX(&init_data->int_parm, sizeof(long)); + DBF_HEX(&init_data->flags, sizeof(long)); + DBF_HEX(&init_data->input_sbal_addr_array, sizeof(void *)); + DBF_HEX(&init_data->output_sbal_addr_array, sizeof(void *)); + DBF_EVENT("irq:%8lx", (unsigned long)irq_ptr); + + /* allocate trace view for the interface */ + snprintf(text, 20, "qdio_%s", dev_name(&init_data->cdev->dev)); + irq_ptr->debug_area = debug_register(text, 2, 1, 16); + debug_register_view(irq_ptr->debug_area, &debug_hex_ascii_view); + debug_set_level(irq_ptr->debug_area, DBF_WARN); + DBF_DEV_EVENT(DBF_ERR, irq_ptr, "dbf created"); } static int qstat_show(struct seq_file *m, void *v) @@ -223,11 +197,24 @@ void qdio_shutdown_debug_entries(struct qdio_irq *irq_ptr, struct ccw_device *cd int __init qdio_debug_init(void) { debugfs_root = debugfs_create_dir("qdio_queues", NULL); - return qdio_register_dbf_views(); + + qdio_dbf_setup = debug_register("qdio_setup", 16, 1, 16); + debug_register_view(qdio_dbf_setup, &debug_hex_ascii_view); + debug_set_level(qdio_dbf_setup, DBF_INFO); + DBF_EVENT("dbf created\n"); + + qdio_dbf_error = debug_register("qdio_error", 4, 1, 16); + debug_register_view(qdio_dbf_error, &debug_hex_ascii_view); + debug_set_level(qdio_dbf_error, DBF_INFO); + DBF_ERROR("dbf created\n"); + return 0; } void qdio_debug_exit(void) { debugfs_remove(debugfs_root); - qdio_unregister_dbf_views(); + if (qdio_dbf_setup) + debug_unregister(qdio_dbf_setup); + if (qdio_dbf_error) + debug_unregister(qdio_dbf_error); } diff --git a/drivers/s390/cio/qdio_debug.h b/drivers/s390/cio/qdio_debug.h index 5a4d85b829a..5d70bd162ae 100644 --- a/drivers/s390/cio/qdio_debug.h +++ b/drivers/s390/cio/qdio_debug.h @@ -12,80 +12,72 @@ #include #include "qdio.h" -#define QDIO_DBF_HEX(ex, name, level, addr, len) \ +/* that gives us 15 characters in the text event views */ +#define QDIO_DBF_LEN 16 + +extern debug_info_t *qdio_dbf_setup; +extern debug_info_t *qdio_dbf_error; + +/* sort out low debug levels early to avoid wasted sprints */ +static inline int qdio_dbf_passes(debug_info_t *dbf_grp, int level) +{ + return (level <= dbf_grp->level); +} + +#define DBF_ERR 3 /* error conditions */ +#define DBF_WARN 4 /* warning conditions */ +#define DBF_INFO 6 /* informational */ + +#undef DBF_EVENT +#undef DBF_ERROR +#undef DBF_DEV_EVENT + +#define DBF_EVENT(text...) \ do { \ - if (ex) \ - debug_exception(qdio_dbf_##name, level, (void *)(addr), len); \ - else \ - debug_event(qdio_dbf_##name, level, (void *)(addr), len); \ + char debug_buffer[QDIO_DBF_LEN]; \ + snprintf(debug_buffer, QDIO_DBF_LEN, text); \ + debug_text_event(qdio_dbf_setup, DBF_ERR, debug_buffer); \ } while (0) -#define QDIO_DBF_TEXT(ex, name, level, text) \ + +#define DBF_HEX(addr, len) \ do { \ - if (ex) \ - debug_text_exception(qdio_dbf_##name, level, text); \ - else \ - debug_text_event(qdio_dbf_##name, level, text); \ + debug_event(qdio_dbf_setup, DBF_ERR, (void*)(addr), len); \ } while (0) -#define QDIO_DBF_HEX0(ex, name, addr, len) QDIO_DBF_HEX(ex, name, 0, addr, len) -#define QDIO_DBF_HEX1(ex, name, addr, len) QDIO_DBF_HEX(ex, name, 1, addr, len) -#define QDIO_DBF_HEX2(ex, name, addr, len) QDIO_DBF_HEX(ex, name, 2, addr, len) - -#ifdef CONFIG_QDIO_DEBUG -#define QDIO_DBF_HEX3(ex, name, addr, len) QDIO_DBF_HEX(ex, name, 3, addr, len) -#define QDIO_DBF_HEX4(ex, name, addr, len) QDIO_DBF_HEX(ex, name, 4, addr, len) -#define QDIO_DBF_HEX5(ex, name, addr, len) QDIO_DBF_HEX(ex, name, 5, addr, len) -#define QDIO_DBF_HEX6(ex, name, addr, len) QDIO_DBF_HEX(ex, name, 6, addr, len) -#else -#define QDIO_DBF_HEX3(ex, name, addr, len) do {} while (0) -#define QDIO_DBF_HEX4(ex, name, addr, len) do {} while (0) -#define QDIO_DBF_HEX5(ex, name, addr, len) do {} while (0) -#define QDIO_DBF_HEX6(ex, name, addr, len) do {} while (0) -#endif /* CONFIG_QDIO_DEBUG */ - -#define QDIO_DBF_TEXT0(ex, name, text) QDIO_DBF_TEXT(ex, name, 0, text) -#define QDIO_DBF_TEXT1(ex, name, text) QDIO_DBF_TEXT(ex, name, 1, text) -#define QDIO_DBF_TEXT2(ex, name, text) QDIO_DBF_TEXT(ex, name, 2, text) - -#ifdef CONFIG_QDIO_DEBUG -#define QDIO_DBF_TEXT3(ex, name, text) QDIO_DBF_TEXT(ex, name, 3, text) -#define QDIO_DBF_TEXT4(ex, name, text) QDIO_DBF_TEXT(ex, name, 4, text) -#define QDIO_DBF_TEXT5(ex, name, text) QDIO_DBF_TEXT(ex, name, 5, text) -#define QDIO_DBF_TEXT6(ex, name, text) QDIO_DBF_TEXT(ex, name, 6, text) -#else -#define QDIO_DBF_TEXT3(ex, name, text) do {} while (0) -#define QDIO_DBF_TEXT4(ex, name, text) do {} while (0) -#define QDIO_DBF_TEXT5(ex, name, text) do {} while (0) -#define QDIO_DBF_TEXT6(ex, name, text) do {} while (0) -#endif /* CONFIG_QDIO_DEBUG */ +#define DBF_ERROR(text...) \ + do { \ + char debug_buffer[QDIO_DBF_LEN]; \ + snprintf(debug_buffer, QDIO_DBF_LEN, text); \ + debug_text_event(qdio_dbf_error, DBF_ERR, debug_buffer); \ + } while (0) -/* s390dbf views */ -#define QDIO_DBF_SETUP_LEN 8 -#define QDIO_DBF_SETUP_PAGES 8 -#define QDIO_DBF_SETUP_NR_AREAS 1 +#define DBF_ERROR_HEX(addr, len) \ + do { \ + debug_event(qdio_dbf_error, DBF_ERR, (void*)(addr), len); \ + } while (0) -#define QDIO_DBF_TRACE_LEN 8 -#define QDIO_DBF_TRACE_NR_AREAS 2 -#ifdef CONFIG_QDIO_DEBUG -#define QDIO_DBF_TRACE_PAGES 32 -#define QDIO_DBF_SETUP_LEVEL 6 -#define QDIO_DBF_TRACE_LEVEL 4 -#else /* !CONFIG_QDIO_DEBUG */ -#define QDIO_DBF_TRACE_PAGES 8 -#define QDIO_DBF_SETUP_LEVEL 2 -#define QDIO_DBF_TRACE_LEVEL 2 -#endif /* CONFIG_QDIO_DEBUG */ +#define DBF_DEV_EVENT(level, device, text...) \ + do { \ + char debug_buffer[QDIO_DBF_LEN]; \ + if (qdio_dbf_passes(device->debug_area, level)) { \ + snprintf(debug_buffer, QDIO_DBF_LEN, text); \ + debug_text_event(device->debug_area, level, debug_buffer); \ + } \ + } while (0) -extern debug_info_t *qdio_dbf_setup; -extern debug_info_t *qdio_dbf_trace; +#define DBF_DEV_HEX(level, device, addr, len) \ + do { \ + debug_event(device->debug_area, level, (void*)(addr), len); \ + } while (0) -void qdio_allocate_do_dbf(struct qdio_initialize *init_data); -void debug_print_bstat(struct qdio_q *q); +void qdio_allocate_dbf(struct qdio_initialize *init_data, + struct qdio_irq *irq_ptr); void qdio_setup_debug_entries(struct qdio_irq *irq_ptr, struct ccw_device *cdev); void qdio_shutdown_debug_entries(struct qdio_irq *irq_ptr, struct ccw_device *cdev); int qdio_debug_init(void); void qdio_debug_exit(void); + #endif diff --git a/drivers/s390/cio/qdio_main.c b/drivers/s390/cio/qdio_main.c index 9e6ab41e5a3..c810214d3d0 100644 --- a/drivers/s390/cio/qdio_main.c +++ b/drivers/s390/cio/qdio_main.c @@ -95,8 +95,6 @@ static inline int do_siga_output(unsigned long schid, unsigned long mask, static inline int qdio_check_ccq(struct qdio_q *q, unsigned int ccq) { - char dbf_text[15]; - /* all done or next buffer state different */ if (ccq == 0 || ccq == 32) return 0; @@ -104,8 +102,7 @@ static inline int qdio_check_ccq(struct qdio_q *q, unsigned int ccq) if (ccq == 96 || ccq == 97) return 1; /* notify devices immediately */ - sprintf(dbf_text, "%d", ccq); - QDIO_DBF_TEXT2(1, trace, dbf_text); + DBF_ERROR("%4x ccq:%3d", SCH_NO(q), ccq); return -EIO; } @@ -126,7 +123,6 @@ static int qdio_do_eqbs(struct qdio_q *q, unsigned char *state, int tmp_count = count, tmp_start = start; int nr = q->nr; int rc; - char dbf_text[15]; BUG_ON(!q->irq_ptr->sch_token); qdio_perf_stat_inc(&perf_stats.debug_eqbs_all); @@ -144,15 +140,15 @@ again: qdio_perf_stat_inc(&perf_stats.debug_eqbs_incomplete); return (count - tmp_count); } + if (rc == 1) { - QDIO_DBF_TEXT5(1, trace, "eqAGAIN"); + DBF_DEV_EVENT(DBF_WARN, q->irq_ptr, "EQBS again:%2d", ccq); goto again; } if (rc < 0) { - QDIO_DBF_TEXT2(1, trace, "eqberr"); - sprintf(dbf_text, "%2x,%2x,%d,%d", count, tmp_count, ccq, nr); - QDIO_DBF_TEXT2(1, trace, dbf_text); + DBF_ERROR("%4x EQBS ERROR", SCH_NO(q)); + DBF_ERROR("%3d%3d%2d", count, tmp_count, nr); q->handler(q->irq_ptr->cdev, QDIO_ERROR_ACTIVATE_CHECK_CONDITION, 0, -1, -1, q->irq_ptr->int_parm); @@ -179,7 +175,6 @@ static int qdio_do_sqbs(struct qdio_q *q, unsigned char state, int start, int tmp_count = count, tmp_start = start; int nr = q->nr; int rc; - char dbf_text[15]; BUG_ON(!q->irq_ptr->sch_token); qdio_perf_stat_inc(&perf_stats.debug_sqbs_all); @@ -190,17 +185,13 @@ again: ccq = do_sqbs(q->irq_ptr->sch_token, state, nr, &tmp_start, &tmp_count); rc = qdio_check_ccq(q, ccq); if (rc == 1) { - QDIO_DBF_TEXT5(1, trace, "sqAGAIN"); + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "SQBS again:%2d", ccq); qdio_perf_stat_inc(&perf_stats.debug_sqbs_incomplete); goto again; } if (rc < 0) { - QDIO_DBF_TEXT3(1, trace, "sqberr"); - sprintf(dbf_text, "%2x,%2x", count, tmp_count); - QDIO_DBF_TEXT3(1, trace, dbf_text); - sprintf(dbf_text, "%d,%d", ccq, nr); - QDIO_DBF_TEXT3(1, trace, dbf_text); - + DBF_ERROR("%4x SQBS ERROR", SCH_NO(q)); + DBF_ERROR("%3d%3d%2d", count, tmp_count, nr); q->handler(q->irq_ptr->cdev, QDIO_ERROR_ACTIVATE_CHECK_CONDITION, 0, -1, -1, q->irq_ptr->int_parm); @@ -287,14 +278,13 @@ static int qdio_siga_sync(struct qdio_q *q, unsigned int output, if (!need_siga_sync(q)) return 0; + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "siga-s:"); + DBF_DEV_HEX(DBF_INFO, q->irq_ptr, q, sizeof(void *)); qdio_perf_stat_inc(&perf_stats.siga_sync); cc = do_siga_sync(q->irq_ptr->schid, output, input); - if (cc) { - QDIO_DBF_TEXT4(0, trace, "sigasync"); - QDIO_DBF_HEX4(0, trace, &q, sizeof(void *)); - QDIO_DBF_HEX3(0, trace, &cc, sizeof(int *)); - } + if (cc) + DBF_ERROR("%4x SIGA-S:%2d", SCH_NO(q), cc); return cc; } @@ -338,17 +328,13 @@ static int qdio_siga_output(struct qdio_q *q) int cc; u32 busy_bit; u64 start_time = 0; - char dbf_text[15]; - - QDIO_DBF_TEXT5(0, trace, "sigaout"); - QDIO_DBF_HEX5(0, trace, &q, sizeof(void *)); + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "siga-w:%1d", q->nr); qdio_perf_stat_inc(&perf_stats.siga_out); again: cc = qdio_do_siga_output(q, &busy_bit); if (queue_type(q) == QDIO_IQDIO_QFMT && cc == 2 && busy_bit) { - sprintf(dbf_text, "bb%4x%2x", q->irq_ptr->schid.sch_no, q->nr); - QDIO_DBF_TEXT3(0, trace, dbf_text); + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "siga-w bb:%2d", q->nr); if (!start_time) start_time = get_usecs(); @@ -359,7 +345,7 @@ again: if (cc == 2 && busy_bit) cc |= QDIO_ERROR_SIGA_BUSY; if (cc) - QDIO_DBF_HEX3(0, trace, &cc, sizeof(int *)); + DBF_ERROR("%4x SIGA-W:%2d", SCH_NO(q), cc); return cc; } @@ -367,14 +353,12 @@ static inline int qdio_siga_input(struct qdio_q *q) { int cc; - QDIO_DBF_TEXT4(0, trace, "sigain"); - QDIO_DBF_HEX4(0, trace, &q, sizeof(void *)); - + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "siga-r:%1d", q->nr); qdio_perf_stat_inc(&perf_stats.siga_in); cc = do_siga_input(q->irq_ptr->schid, q->mask); if (cc) - QDIO_DBF_HEX3(0, trace, &cc, sizeof(int *)); + DBF_ERROR("%4x SIGA-R:%2d", SCH_NO(q), cc); return cc; } @@ -407,18 +391,12 @@ inline void qdio_stop_polling(struct qdio_q *q) static void announce_buffer_error(struct qdio_q *q) { - char dbf_text[15]; - - if (q->is_input_q) - QDIO_DBF_TEXT3(1, trace, "inperr"); - else - QDIO_DBF_TEXT3(0, trace, "outperr"); - - sprintf(dbf_text, "%x-%x-%x", q->first_to_check, - q->sbal[q->first_to_check]->element[14].flags, - q->sbal[q->first_to_check]->element[15].flags); - QDIO_DBF_TEXT3(1, trace, dbf_text); - QDIO_DBF_HEX2(1, trace, q->sbal[q->first_to_check], 256); + DBF_ERROR("%4x BUF ERROR", SCH_NO(q)); + DBF_ERROR((q->is_input_q) ? "IN:%2d" : "OUT:%2d", q->nr); + DBF_ERROR("FTC:%3d", q->first_to_check); + DBF_ERROR("F14:%2x F15:%2x", + q->sbal[q->first_to_check]->element[14].flags & 0xff, + q->sbal[q->first_to_check]->element[15].flags & 0xff); q->qdio_error = QDIO_ERROR_SLSB_STATE; } @@ -461,7 +439,7 @@ check_next: switch (state) { case SLSB_P_INPUT_PRIMED: - QDIO_DBF_TEXT5(0, trace, "inptprim"); + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "in prim: %3d", count); /* * Only ACK the first buffer. The ACK will be removed in @@ -500,13 +478,12 @@ check_next: case SLSB_CU_INPUT_EMPTY: case SLSB_P_INPUT_NOT_INIT: case SLSB_P_INPUT_ACK: - QDIO_DBF_TEXT5(0, trace, "inpnipro"); + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "in nop"); break; default: BUG(); } out: - QDIO_DBF_HEX4(0, trace, &q->first_to_check, sizeof(int)); return q->first_to_check; } @@ -520,8 +497,7 @@ int qdio_inbound_q_moved(struct qdio_q *q) if (!need_siga_sync(q) && !pci_out_supported(q)) q->u.in.timestamp = get_usecs(); - QDIO_DBF_TEXT4(0, trace, "inhasmvd"); - QDIO_DBF_HEX4(0, trace, &q, sizeof(void *)); + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "in moved"); return 1; } else return 0; @@ -530,9 +506,6 @@ int qdio_inbound_q_moved(struct qdio_q *q) static int qdio_inbound_q_done(struct qdio_q *q) { unsigned char state = 0; -#ifdef CONFIG_QDIO_DEBUG - char dbf_text[15]; -#endif if (!atomic_read(&q->nr_buf_used)) return 1; @@ -557,20 +530,12 @@ static int qdio_inbound_q_done(struct qdio_q *q) * has (probably) not moved (see qdio_inbound_processing). */ if (get_usecs() > q->u.in.timestamp + QDIO_INPUT_THRESHOLD) { -#ifdef CONFIG_QDIO_DEBUG - QDIO_DBF_TEXT4(0, trace, "inqisdon"); - QDIO_DBF_HEX4(0, trace, &q, sizeof(void *)); - sprintf(dbf_text, "pf%02x", q->first_to_check); - QDIO_DBF_TEXT4(0, trace, dbf_text); -#endif /* CONFIG_QDIO_DEBUG */ + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "in done:%3d", + q->first_to_check); return 1; } else { -#ifdef CONFIG_QDIO_DEBUG - QDIO_DBF_TEXT4(0, trace, "inqisntd"); - QDIO_DBF_HEX4(0, trace, &q, sizeof(void *)); - sprintf(dbf_text, "pf%02x", q->first_to_check); - QDIO_DBF_TEXT4(0, trace, dbf_text); -#endif /* CONFIG_QDIO_DEBUG */ + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "in notd:%3d", + q->first_to_check); return 0; } } @@ -578,9 +543,6 @@ static int qdio_inbound_q_done(struct qdio_q *q) void qdio_kick_inbound_handler(struct qdio_q *q) { int count, start, end; -#ifdef CONFIG_QDIO_DEBUG - char dbf_text[15]; -#endif qdio_perf_stat_inc(&perf_stats.inbound_handler); @@ -591,10 +553,7 @@ void qdio_kick_inbound_handler(struct qdio_q *q) else count = end + QDIO_MAX_BUFFERS_PER_Q - start; -#ifdef CONFIG_QDIO_DEBUG - sprintf(dbf_text, "s=%2xc=%2x", start, count); - QDIO_DBF_TEXT4(0, trace, dbf_text); -#endif /* CONFIG_QDIO_DEBUG */ + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "kih s:%3d c:%3d", start, count); if (unlikely(q->irq_ptr->state != QDIO_IRQ_STATE_ACTIVE)) return; @@ -667,7 +626,7 @@ check_next: switch (state) { case SLSB_P_OUTPUT_EMPTY: /* the adapter got it */ - QDIO_DBF_TEXT5(0, trace, "outpempt"); + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "out empty:%1d %3d", q->nr, count); atomic_sub(count, &q->nr_buf_used); q->first_to_check = add_buf(q->first_to_check, count); @@ -686,7 +645,7 @@ check_next: break; case SLSB_CU_OUTPUT_PRIMED: /* the adapter has not fetched the output yet */ - QDIO_DBF_TEXT5(0, trace, "outpprim"); + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "out primed:%1d", q->nr); break; case SLSB_P_OUTPUT_NOT_INIT: case SLSB_P_OUTPUT_HALTED: @@ -711,8 +670,7 @@ static inline int qdio_outbound_q_moved(struct qdio_q *q) if ((bufnr != q->last_move_ftc) || q->qdio_error) { q->last_move_ftc = bufnr; - QDIO_DBF_TEXT4(0, trace, "oqhasmvd"); - QDIO_DBF_HEX4(0, trace, &q, sizeof(void *)); + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "out moved:%1d", q->nr); return 1; } else return 0; @@ -747,12 +705,8 @@ static inline int qdio_outbound_q_moved(struct qdio_q *q) static void qdio_kick_outbound_q(struct qdio_q *q) { int rc; -#ifdef CONFIG_QDIO_DEBUG - char dbf_text[15]; - QDIO_DBF_TEXT5(0, trace, "kickoutq"); - QDIO_DBF_HEX5(0, trace, &q, sizeof(void *)); -#endif /* CONFIG_QDIO_DEBUG */ + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "kickoutq:%1d", q->nr); if (!need_siga_out(q)) return; @@ -761,15 +715,9 @@ static void qdio_kick_outbound_q(struct qdio_q *q) switch (rc) { case 0: /* TODO: improve error handling for CC=0 case */ -#ifdef CONFIG_QDIO_DEBUG - if (q->u.out.timestamp) { - QDIO_DBF_TEXT3(0, trace, "cc2reslv"); - sprintf(dbf_text, "%4x%2x%2x", q->irq_ptr->schid.sch_no, - q->nr, - atomic_read(&q->u.out.busy_siga_counter)); - QDIO_DBF_TEXT3(0, trace, dbf_text); - } -#endif /* CONFIG_QDIO_DEBUG */ + if (q->u.out.timestamp) + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "cc2 rslv:%4x", + atomic_read(&q->u.out.busy_siga_counter)); /* went smooth this time, reset timestamp */ q->u.out.timestamp = 0; break; @@ -786,12 +734,7 @@ static void qdio_kick_outbound_q(struct qdio_q *q) tasklet_schedule(&q->tasklet); break; } - QDIO_DBF_TEXT2(0, trace, "cc2REPRT"); -#ifdef CONFIG_QDIO_DEBUG - sprintf(dbf_text, "%4x%2x%2x", q->irq_ptr->schid.sch_no, q->nr, - atomic_read(&q->u.out.busy_siga_counter)); - QDIO_DBF_TEXT3(0, trace, dbf_text); -#endif /* CONFIG_QDIO_DEBUG */ + DBF_ERROR("%4x cc2 REP:%1d", SCH_NO(q), q->nr); default: /* for plain cc=1, 2 or 3 */ q->qdio_error = rc; @@ -801,9 +744,6 @@ static void qdio_kick_outbound_q(struct qdio_q *q) static void qdio_kick_outbound_handler(struct qdio_q *q) { int start, end, count; -#ifdef CONFIG_QDIO_DEBUG - char dbf_text[15]; -#endif start = q->first_to_kick; end = q->last_move_ftc; @@ -812,13 +752,8 @@ static void qdio_kick_outbound_handler(struct qdio_q *q) else count = end + QDIO_MAX_BUFFERS_PER_Q - start; -#ifdef CONFIG_QDIO_DEBUG - QDIO_DBF_TEXT4(0, trace, "kickouth"); - QDIO_DBF_HEX4(0, trace, &q, sizeof(void *)); - - sprintf(dbf_text, "s=%2xc=%2x", start, count); - QDIO_DBF_TEXT4(0, trace, dbf_text); -#endif /* CONFIG_QDIO_DEBUG */ + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "kickouth: %1d", q->nr); + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "s:%3d c:%3d", start, count); if (unlikely(q->irq_ptr->state != QDIO_IRQ_STATE_ACTIVE)) return; @@ -913,27 +848,18 @@ void qdio_check_outbound_after_thinint(struct qdio_q *q) static inline void qdio_set_state(struct qdio_irq *irq_ptr, enum qdio_irq_states state) { -#ifdef CONFIG_QDIO_DEBUG - char dbf_text[15]; - - QDIO_DBF_TEXT5(0, trace, "newstate"); - sprintf(dbf_text, "%4x%4x", irq_ptr->schid.sch_no, state); - QDIO_DBF_TEXT5(0, trace, dbf_text); -#endif /* CONFIG_QDIO_DEBUG */ + DBF_DEV_EVENT(DBF_INFO, irq_ptr, "newstate: %1d", state); irq_ptr->state = state; mb(); } -static void qdio_irq_check_sense(struct subchannel_id schid, struct irb *irb) +static void qdio_irq_check_sense(struct qdio_irq *irq_ptr, struct irb *irb) { - char dbf_text[15]; - if (irb->esw.esw0.erw.cons) { - sprintf(dbf_text, "sens%4x", schid.sch_no); - QDIO_DBF_TEXT2(1, trace, dbf_text); - QDIO_DBF_HEX0(0, trace, irb, 64); - QDIO_DBF_HEX0(0, trace, irb->ecw, 64); + DBF_ERROR("%4x sense:", irq_ptr->schid.sch_no); + DBF_ERROR_HEX(irb, 64); + DBF_ERROR_HEX(irb->ecw, 64); } } @@ -967,14 +893,10 @@ static void qdio_handle_activate_check(struct ccw_device *cdev, { struct qdio_irq *irq_ptr = cdev->private->qdio_data; struct qdio_q *q; - char dbf_text[15]; - QDIO_DBF_TEXT2(1, trace, "ick2"); - sprintf(dbf_text, "%s", dev_name(&cdev->dev)); - QDIO_DBF_TEXT2(1, trace, dbf_text); - QDIO_DBF_HEX2(0, trace, &intparm, sizeof(int)); - QDIO_DBF_HEX2(0, trace, &dstat, sizeof(int)); - QDIO_DBF_HEX2(0, trace, &cstat, sizeof(int)); + DBF_ERROR("%4x ACT CHECK", irq_ptr->schid.sch_no); + DBF_ERROR("intp :%lx", intparm); + DBF_ERROR("ds: %2x cs:%2x", dstat, cstat); if (irq_ptr->nr_input_qs) { q = irq_ptr->input_qs[0]; @@ -1027,28 +949,29 @@ static void qdio_int_error(struct ccw_device *cdev) } static int qdio_establish_check_errors(struct ccw_device *cdev, int cstat, - int dstat) + int dstat) { struct qdio_irq *irq_ptr = cdev->private->qdio_data; if (cstat || (dstat & ~(DEV_STAT_CHN_END | DEV_STAT_DEV_END))) { - QDIO_DBF_TEXT2(1, setup, "eq:ckcon"); + DBF_ERROR("EQ:ck con"); goto error; } if (!(dstat & DEV_STAT_DEV_END)) { - QDIO_DBF_TEXT2(1, setup, "eq:no de"); + DBF_ERROR("EQ:no dev"); goto error; } if (dstat & ~(DEV_STAT_CHN_END | DEV_STAT_DEV_END)) { - QDIO_DBF_TEXT2(1, setup, "eq:badio"); + DBF_ERROR("EQ: bad io"); goto error; } return 0; error: - QDIO_DBF_HEX2(0, trace, &cstat, sizeof(int)); - QDIO_DBF_HEX2(0, trace, &dstat, sizeof(int)); + DBF_ERROR("%4x EQ:error", irq_ptr->schid.sch_no); + DBF_ERROR("ds: %2x cs:%2x", dstat, cstat); + qdio_set_state(irq_ptr, QDIO_IRQ_STATE_ERR); return 1; } @@ -1057,12 +980,8 @@ static void qdio_establish_handle_irq(struct ccw_device *cdev, int cstat, int dstat) { struct qdio_irq *irq_ptr = cdev->private->qdio_data; - char dbf_text[15]; - - sprintf(dbf_text, "qehi%4x", cdev->private->schid.sch_no); - QDIO_DBF_TEXT0(0, setup, dbf_text); - QDIO_DBF_TEXT0(0, trace, dbf_text); + DBF_DEV_EVENT(DBF_INFO, irq_ptr, "qest irq"); if (!qdio_establish_check_errors(cdev, cstat, dstat)) qdio_set_state(irq_ptr, QDIO_IRQ_STATE_ESTABLISHED); } @@ -1073,25 +992,21 @@ void qdio_int_handler(struct ccw_device *cdev, unsigned long intparm, { struct qdio_irq *irq_ptr = cdev->private->qdio_data; int cstat, dstat; - char dbf_text[15]; qdio_perf_stat_inc(&perf_stats.qdio_int); if (!intparm || !irq_ptr) { - sprintf(dbf_text, "qihd%4x", cdev->private->schid.sch_no); - QDIO_DBF_TEXT2(1, setup, dbf_text); + DBF_ERROR("qint:%4x", cdev->private->schid.sch_no); return; } if (IS_ERR(irb)) { switch (PTR_ERR(irb)) { case -EIO: - sprintf(dbf_text, "ierr%4x", irq_ptr->schid.sch_no); - QDIO_DBF_TEXT2(1, setup, dbf_text); + DBF_ERROR("%4x IO error", irq_ptr->schid.sch_no); return; case -ETIMEDOUT: - sprintf(dbf_text, "qtoh%4x", irq_ptr->schid.sch_no); - QDIO_DBF_TEXT2(1, setup, dbf_text); + DBF_ERROR("%4x IO timeout", irq_ptr->schid.sch_no); qdio_int_error(cdev); return; default: @@ -1099,7 +1014,7 @@ void qdio_int_handler(struct ccw_device *cdev, unsigned long intparm, return; } } - qdio_irq_check_sense(irq_ptr->schid, irb); + qdio_irq_check_sense(irq_ptr, irb); cstat = irb->scsw.cmd.cstat; dstat = irb->scsw.cmd.dstat; @@ -1142,14 +1057,11 @@ void qdio_int_handler(struct ccw_device *cdev, unsigned long intparm, int qdio_get_ssqd_desc(struct ccw_device *cdev, struct qdio_ssqd_desc *data) { - char dbf_text[15]; if (!cdev || !cdev->private) return -EINVAL; - sprintf(dbf_text, "qssq%4x", cdev->private->schid.sch_no); - QDIO_DBF_TEXT0(0, setup, dbf_text); - + DBF_EVENT("get ssqd:%4x", cdev->private->schid.sch_no); return qdio_setup_get_ssqd(NULL, &cdev->private->schid, data); } EXPORT_SYMBOL_GPL(qdio_get_ssqd_desc); @@ -1164,14 +1076,9 @@ EXPORT_SYMBOL_GPL(qdio_get_ssqd_desc); */ int qdio_cleanup(struct ccw_device *cdev, int how) { - struct qdio_irq *irq_ptr; - char dbf_text[15]; + struct qdio_irq *irq_ptr = cdev->private->qdio_data; int rc; - sprintf(dbf_text, "qcln%4x", cdev->private->schid.sch_no); - QDIO_DBF_TEXT0(0, setup, dbf_text); - - irq_ptr = cdev->private->qdio_data; if (!irq_ptr) return -ENODEV; @@ -1204,18 +1111,15 @@ static void qdio_shutdown_queues(struct ccw_device *cdev) */ int qdio_shutdown(struct ccw_device *cdev, int how) { - struct qdio_irq *irq_ptr; + struct qdio_irq *irq_ptr = cdev->private->qdio_data; int rc; unsigned long flags; - char dbf_text[15]; - sprintf(dbf_text, "qshu%4x", cdev->private->schid.sch_no); - QDIO_DBF_TEXT0(0, setup, dbf_text); - - irq_ptr = cdev->private->qdio_data; if (!irq_ptr) return -ENODEV; + DBF_EVENT("qshutdown:%4x", cdev->private->schid.sch_no); + mutex_lock(&irq_ptr->setup_mutex); /* * Subchannel was already shot down. We cannot prevent being called @@ -1239,10 +1143,8 @@ int qdio_shutdown(struct ccw_device *cdev, int how) /* default behaviour is halt */ rc = ccw_device_halt(cdev, QDIO_DOING_CLEANUP); if (rc) { - sprintf(dbf_text, "sher%4x", irq_ptr->schid.sch_no); - QDIO_DBF_TEXT0(0, setup, dbf_text); - sprintf(dbf_text, "rc=%d", rc); - QDIO_DBF_TEXT0(0, setup, dbf_text); + DBF_ERROR("%4x SHUTD ERR", irq_ptr->schid.sch_no); + DBF_ERROR("rc:%4d", rc); goto no_cleanup; } @@ -1276,17 +1178,18 @@ EXPORT_SYMBOL_GPL(qdio_shutdown); */ int qdio_free(struct ccw_device *cdev) { - struct qdio_irq *irq_ptr; - char dbf_text[15]; - - sprintf(dbf_text, "qfre%4x", cdev->private->schid.sch_no); - QDIO_DBF_TEXT0(0, setup, dbf_text); + struct qdio_irq *irq_ptr = cdev->private->qdio_data; - irq_ptr = cdev->private->qdio_data; if (!irq_ptr) return -ENODEV; + DBF_EVENT("qfree:%4x", cdev->private->schid.sch_no); mutex_lock(&irq_ptr->setup_mutex); + + if (irq_ptr->debug_area != NULL) { + debug_unregister(irq_ptr->debug_area); + irq_ptr->debug_area = NULL; + } cdev->private->qdio_data = NULL; mutex_unlock(&irq_ptr->setup_mutex); @@ -1305,10 +1208,6 @@ EXPORT_SYMBOL_GPL(qdio_free); int qdio_initialize(struct qdio_initialize *init_data) { int rc; - char dbf_text[15]; - - sprintf(dbf_text, "qini%4x", init_data->cdev->private->schid.sch_no); - QDIO_DBF_TEXT0(0, setup, dbf_text); rc = qdio_allocate(init_data); if (rc) @@ -1328,10 +1227,8 @@ EXPORT_SYMBOL_GPL(qdio_initialize); int qdio_allocate(struct qdio_initialize *init_data) { struct qdio_irq *irq_ptr; - char dbf_text[15]; - sprintf(dbf_text, "qalc%4x", init_data->cdev->private->schid.sch_no); - QDIO_DBF_TEXT0(0, setup, dbf_text); + DBF_EVENT("qallocate:%4x", init_data->cdev->private->schid.sch_no); if ((init_data->no_input_qs && !init_data->input_handler) || (init_data->no_output_qs && !init_data->output_handler)) @@ -1345,16 +1242,13 @@ int qdio_allocate(struct qdio_initialize *init_data) (!init_data->output_sbal_addr_array)) return -EINVAL; - qdio_allocate_do_dbf(init_data); - /* irq_ptr must be in GFP_DMA since it contains ccw1.cda */ irq_ptr = (void *) get_zeroed_page(GFP_KERNEL | GFP_DMA); if (!irq_ptr) goto out_err; - QDIO_DBF_TEXT0(0, setup, "irq_ptr:"); - QDIO_DBF_HEX0(0, setup, &irq_ptr, sizeof(void *)); mutex_init(&irq_ptr->setup_mutex); + qdio_allocate_dbf(init_data, irq_ptr); /* * Allocate a page for the chsc calls in qdio_establish. @@ -1372,9 +1266,6 @@ int qdio_allocate(struct qdio_initialize *init_data) goto out_rel; WARN_ON((unsigned long)irq_ptr->qdr & 0xfff); - QDIO_DBF_TEXT0(0, setup, "qdr:"); - QDIO_DBF_HEX0(0, setup, &irq_ptr->qdr, sizeof(void *)); - if (qdio_allocate_qs(irq_ptr, init_data->no_input_qs, init_data->no_output_qs)) goto out_rel; @@ -1395,14 +1286,12 @@ EXPORT_SYMBOL_GPL(qdio_allocate); */ int qdio_establish(struct qdio_initialize *init_data) { - char dbf_text[20]; struct qdio_irq *irq_ptr; struct ccw_device *cdev = init_data->cdev; unsigned long saveflags; int rc; - sprintf(dbf_text, "qest%4x", cdev->private->schid.sch_no); - QDIO_DBF_TEXT0(0, setup, dbf_text); + DBF_EVENT("qestablish:%4x", cdev->private->schid.sch_no); irq_ptr = cdev->private->qdio_data; if (!irq_ptr) @@ -1432,10 +1321,8 @@ int qdio_establish(struct qdio_initialize *init_data) rc = ccw_device_start(cdev, &irq_ptr->ccw, QDIO_DOING_ESTABLISH, 0, 0); if (rc) { - sprintf(dbf_text, "eq:io%4x", irq_ptr->schid.sch_no); - QDIO_DBF_TEXT2(1, setup, dbf_text); - sprintf(dbf_text, "eq:rc%4x", rc); - QDIO_DBF_TEXT2(1, setup, dbf_text); + DBF_ERROR("%4x est IO ERR", irq_ptr->schid.sch_no); + DBF_ERROR("rc:%4x", rc); } spin_unlock_irqrestore(get_ccwdev_lock(cdev), saveflags); @@ -1456,10 +1343,8 @@ int qdio_establish(struct qdio_initialize *init_data) } qdio_setup_ssqd_info(irq_ptr); - sprintf(dbf_text, "qDmmwc%2x", irq_ptr->ssqd_desc.mmwc); - QDIO_DBF_TEXT2(0, setup, dbf_text); - sprintf(dbf_text, "qib ac%2x", irq_ptr->qib.ac); - QDIO_DBF_TEXT2(0, setup, dbf_text); + DBF_EVENT("qDmmwc:%2x", irq_ptr->ssqd_desc.mmwc); + DBF_EVENT("qib ac:%4x", irq_ptr->qib.ac); /* qebsm is now setup if available, initialize buffer states */ qdio_init_buf_states(irq_ptr); @@ -1480,10 +1365,8 @@ int qdio_activate(struct ccw_device *cdev) struct qdio_irq *irq_ptr; int rc; unsigned long saveflags; - char dbf_text[20]; - sprintf(dbf_text, "qact%4x", cdev->private->schid.sch_no); - QDIO_DBF_TEXT0(0, setup, dbf_text); + DBF_EVENT("qactivate:%4x", cdev->private->schid.sch_no); irq_ptr = cdev->private->qdio_data; if (!irq_ptr) @@ -1509,10 +1392,8 @@ int qdio_activate(struct ccw_device *cdev) rc = ccw_device_start(cdev, &irq_ptr->ccw, QDIO_DOING_ACTIVATE, 0, DOIO_DENY_PREFETCH); if (rc) { - sprintf(dbf_text, "aq:io%4x", irq_ptr->schid.sch_no); - QDIO_DBF_TEXT2(1, setup, dbf_text); - sprintf(dbf_text, "aq:rc%4x", rc); - QDIO_DBF_TEXT2(1, setup, dbf_text); + DBF_ERROR("%4x act IO ERR", irq_ptr->schid.sch_no); + DBF_ERROR("rc:%4x", rc); } spin_unlock_irqrestore(get_ccwdev_lock(cdev), saveflags); @@ -1658,7 +1539,7 @@ static void handle_outbound(struct qdio_q *q, unsigned int callflags, if (state != SLSB_CU_OUTPUT_PRIMED) qdio_kick_outbound_q(q); else { - QDIO_DBF_TEXT5(0, trace, "fast-req"); + DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "fast-req"); qdio_perf_stat_inc(&perf_stats.fast_requeue); } out: @@ -1678,12 +1559,6 @@ int do_QDIO(struct ccw_device *cdev, unsigned int callflags, int q_nr, int bufnr, int count) { struct qdio_irq *irq_ptr; -#ifdef CONFIG_QDIO_DEBUG - char dbf_text[20]; - - sprintf(dbf_text, "doQD%4x", cdev->private->schid.sch_no); - QDIO_DBF_TEXT3(0, trace, dbf_text); -#endif /* CONFIG_QDIO_DEBUG */ if ((bufnr > QDIO_MAX_BUFFERS_PER_Q) || (count > QDIO_MAX_BUFFERS_PER_Q) || @@ -1697,33 +1572,24 @@ int do_QDIO(struct ccw_device *cdev, unsigned int callflags, if (!irq_ptr) return -ENODEV; -#ifdef CONFIG_QDIO_DEBUG if (callflags & QDIO_FLAG_SYNC_INPUT) - QDIO_DBF_HEX3(0, trace, &irq_ptr->input_qs[q_nr], - sizeof(void *)); + DBF_DEV_EVENT(DBF_INFO, irq_ptr, "doQDIO input"); else - QDIO_DBF_HEX3(0, trace, &irq_ptr->output_qs[q_nr], - sizeof(void *)); - - sprintf(dbf_text, "flag%04x", callflags); - QDIO_DBF_TEXT3(0, trace, dbf_text); - sprintf(dbf_text, "qi%02xct%02x", bufnr, count); - QDIO_DBF_TEXT3(0, trace, dbf_text); -#endif /* CONFIG_QDIO_DEBUG */ + DBF_DEV_EVENT(DBF_INFO, irq_ptr, "doQDIO output"); + DBF_DEV_EVENT(DBF_INFO, irq_ptr, "q:%1d flag:%4x", q_nr, callflags); + DBF_DEV_EVENT(DBF_INFO, irq_ptr, "buf:%2d cnt:%3d", bufnr, count); if (irq_ptr->state != QDIO_IRQ_STATE_ACTIVE) return -EBUSY; if (callflags & QDIO_FLAG_SYNC_INPUT) - handle_inbound(irq_ptr->input_qs[q_nr], - callflags, bufnr, count); + handle_inbound(irq_ptr->input_qs[q_nr], callflags, bufnr, + count); else if (callflags & QDIO_FLAG_SYNC_OUTPUT) - handle_outbound(irq_ptr->output_qs[q_nr], - callflags, bufnr, count); - else { - QDIO_DBF_TEXT3(1, trace, "doQD:inv"); + handle_outbound(irq_ptr->output_qs[q_nr], callflags, bufnr, + count); + else return -EINVAL; - } return 0; } EXPORT_SYMBOL_GPL(do_QDIO); diff --git a/drivers/s390/cio/qdio_setup.c b/drivers/s390/cio/qdio_setup.c index 22ee0272a10..7323799b2ce 100644 --- a/drivers/s390/cio/qdio_setup.c +++ b/drivers/s390/cio/qdio_setup.c @@ -120,14 +120,12 @@ static void setup_queues_misc(struct qdio_q *q, struct qdio_irq *irq_ptr, } static void setup_storage_lists(struct qdio_q *q, struct qdio_irq *irq_ptr, - void **sbals_array, char *dbf_text, int i) + void **sbals_array, int i) { struct qdio_q *prev; int j; - QDIO_DBF_TEXT0(0, setup, dbf_text); - QDIO_DBF_HEX0(0, setup, &q, sizeof(void *)); - + DBF_HEX(&q, sizeof(void *)); q->sl = (struct sl *)((char *)q->slib + PAGE_SIZE / 2); /* fill in sbal */ @@ -150,31 +148,27 @@ static void setup_storage_lists(struct qdio_q *q, struct qdio_irq *irq_ptr, for (j = 0; j < QDIO_MAX_BUFFERS_PER_Q; j++) q->sl->element[j].sbal = (unsigned long)q->sbal[j]; - QDIO_DBF_TEXT2(0, setup, "sl-sb-b0"); - QDIO_DBF_HEX2(0, setup, q->sl, sizeof(void *)); - QDIO_DBF_HEX2(0, setup, &q->slsb, sizeof(void *)); - QDIO_DBF_HEX2(0, setup, q->sbal, sizeof(void *)); + DBF_EVENT("sl-slsb-sbal"); + DBF_HEX(q->sl, sizeof(void *)); + DBF_HEX(&q->slsb, sizeof(void *)); + DBF_HEX(q->sbal, sizeof(void *)); } static void setup_queues(struct qdio_irq *irq_ptr, struct qdio_initialize *qdio_init) { - char dbf_text[20]; struct qdio_q *q; void **input_sbal_array = qdio_init->input_sbal_addr_array; void **output_sbal_array = qdio_init->output_sbal_addr_array; int i; - sprintf(dbf_text, "qset%4x", qdio_init->cdev->private->schid.sch_no); - QDIO_DBF_TEXT0(0, setup, dbf_text); - for_each_input_queue(irq_ptr, q, i) { - sprintf(dbf_text, "in-q%4x", i); + DBF_EVENT("in-q:%1d", i); setup_queues_misc(q, irq_ptr, qdio_init->input_handler, i); q->is_input_q = 1; spin_lock_init(&q->u.in.lock); - setup_storage_lists(q, irq_ptr, input_sbal_array, dbf_text, i); + setup_storage_lists(q, irq_ptr, input_sbal_array, i); input_sbal_array += QDIO_MAX_BUFFERS_PER_Q; if (is_thinint_irq(irq_ptr)) @@ -186,12 +180,11 @@ static void setup_queues(struct qdio_irq *irq_ptr, } for_each_output_queue(irq_ptr, q, i) { - sprintf(dbf_text, "outq%4x", i); + DBF_EVENT("outq:%1d", i); setup_queues_misc(q, irq_ptr, qdio_init->output_handler, i); q->is_input_q = 0; - setup_storage_lists(q, irq_ptr, output_sbal_array, - dbf_text, i); + setup_storage_lists(q, irq_ptr, output_sbal_array, i); output_sbal_array += QDIO_MAX_BUFFERS_PER_Q; tasklet_init(&q->tasklet, qdio_outbound_processing, @@ -222,8 +215,6 @@ static void process_ac_flags(struct qdio_irq *irq_ptr, unsigned char qdioac) static void check_and_setup_qebsm(struct qdio_irq *irq_ptr, unsigned char qdioac, unsigned long token) { - char dbf_text[15]; - if (!(irq_ptr->qib.rflags & QIB_RFLAGS_ENABLE_QEBSM)) goto no_qebsm; if (!(qdioac & AC1_SC_QEBSM_AVAILABLE) || @@ -232,15 +223,14 @@ static void check_and_setup_qebsm(struct qdio_irq *irq_ptr, irq_ptr->sch_token = token; - QDIO_DBF_TEXT0(0, setup, "V=V:1"); - sprintf(dbf_text, "%8lx", irq_ptr->sch_token); - QDIO_DBF_TEXT0(0, setup, dbf_text); + DBF_EVENT("V=V:1"); + DBF_EVENT("%8lx", irq_ptr->sch_token); return; no_qebsm: irq_ptr->sch_token = 0; irq_ptr->qib.rflags &= ~QIB_RFLAGS_ENABLE_QEBSM; - QDIO_DBF_TEXT0(0, setup, "noV=V"); + DBF_EVENT("noV=V"); } /* @@ -254,7 +244,7 @@ int qdio_setup_get_ssqd(struct qdio_irq *irq_ptr, struct chsc_ssqd_area *ssqd; int rc; - QDIO_DBF_TEXT0(0, setup, "getssqd"); + DBF_EVENT("getssqd:%4x", schid->sch_no); if (irq_ptr != NULL) ssqd = (struct chsc_ssqd_area *)irq_ptr->chsc_page; else @@ -294,16 +284,12 @@ int qdio_setup_get_ssqd(struct qdio_irq *irq_ptr, void qdio_setup_ssqd_info(struct qdio_irq *irq_ptr) { unsigned char qdioac; - char dbf_text[15]; int rc; rc = qdio_setup_get_ssqd(irq_ptr, &irq_ptr->schid, NULL); if (rc) { - QDIO_DBF_TEXT2(0, setup, "ssqdasig"); - sprintf(dbf_text, "schn%4x", irq_ptr->schid.sch_no); - QDIO_DBF_TEXT2(0, setup, dbf_text); - sprintf(dbf_text, "rc:%d", rc); - QDIO_DBF_TEXT2(0, setup, dbf_text); + DBF_ERROR("%4x ssqd ERR", irq_ptr->schid.sch_no); + DBF_ERROR("rc:%x", rc); /* all flags set, worst case */ qdioac = AC1_SIGA_INPUT_NEEDED | AC1_SIGA_OUTPUT_NEEDED | AC1_SIGA_SYNC_NEEDED; @@ -312,9 +298,7 @@ void qdio_setup_ssqd_info(struct qdio_irq *irq_ptr) check_and_setup_qebsm(irq_ptr, qdioac, irq_ptr->ssqd_desc.sch_token); process_ac_flags(irq_ptr, qdioac); - - sprintf(dbf_text, "qdioac%2x", qdioac); - QDIO_DBF_TEXT2(0, setup, dbf_text); + DBF_EVENT("qdioac:%4x", qdioac); } void qdio_release_memory(struct qdio_irq *irq_ptr) @@ -434,7 +418,7 @@ int qdio_setup_irq(struct qdio_initialize *init_data) /* get qdio commands */ ciw = ccw_device_get_ciw(init_data->cdev, CIW_TYPE_EQUEUE); if (!ciw) { - QDIO_DBF_TEXT2(1, setup, "no eq"); + DBF_ERROR("%4x NO EQ", irq_ptr->schid.sch_no); rc = -EINVAL; goto out_err; } @@ -442,7 +426,7 @@ int qdio_setup_irq(struct qdio_initialize *init_data) ciw = ccw_device_get_ciw(init_data->cdev, CIW_TYPE_AQUEUE); if (!ciw) { - QDIO_DBF_TEXT2(1, setup, "no aq"); + DBF_ERROR("%4x NO AQ", irq_ptr->schid.sch_no); rc = -EINVAL; goto out_err; } @@ -462,56 +446,38 @@ void qdio_print_subchannel_info(struct qdio_irq *irq_ptr, { char s[80]; - sprintf(s, "qdio: %s ", dev_name(&cdev->dev)); - switch (irq_ptr->qib.qfmt) { - case QDIO_QETH_QFMT: - sprintf(s + strlen(s), "OSA "); - break; - case QDIO_ZFCP_QFMT: - sprintf(s + strlen(s), "ZFCP "); - break; - case QDIO_IQDIO_QFMT: - sprintf(s + strlen(s), "HS "); - break; - } - sprintf(s + strlen(s), "on SC %x using ", irq_ptr->schid.sch_no); - sprintf(s + strlen(s), "AI:%d ", is_thinint_irq(irq_ptr)); - sprintf(s + strlen(s), "QEBSM:%d ", (irq_ptr->sch_token) ? 1 : 0); - sprintf(s + strlen(s), "PCI:%d ", - (irq_ptr->qib.ac & QIB_AC_OUTBOUND_PCI_SUPPORTED) ? 1 : 0); - sprintf(s + strlen(s), "TDD:%d ", css_general_characteristics.aif_tdd); - sprintf(s + strlen(s), "SIGA:"); - sprintf(s + strlen(s), "%s", (irq_ptr->siga_flag.input) ? "R" : " "); - sprintf(s + strlen(s), "%s", (irq_ptr->siga_flag.output) ? "W" : " "); - sprintf(s + strlen(s), "%s", (irq_ptr->siga_flag.sync) ? "S" : " "); - sprintf(s + strlen(s), "%s", - (!irq_ptr->siga_flag.no_sync_ti) ? "A" : " "); - sprintf(s + strlen(s), "%s", - (!irq_ptr->siga_flag.no_sync_out_ti) ? "O" : " "); - sprintf(s + strlen(s), "%s", - (!irq_ptr->siga_flag.no_sync_out_pci) ? "P" : " "); - sprintf(s + strlen(s), "\n"); + snprintf(s, 80, "qdio: %s %s on SC %x using " + "AI:%d QEBSM:%d PCI:%d TDD:%d SIGA:%s%s%s%s%s%s\n", + dev_name(&cdev->dev), + (irq_ptr->qib.qfmt == QDIO_QETH_QFMT) ? "OSA" : + ((irq_ptr->qib.qfmt == QDIO_ZFCP_QFMT) ? "ZFCP" : "HS"), + irq_ptr->schid.sch_no, + is_thinint_irq(irq_ptr), + (irq_ptr->sch_token) ? 1 : 0, + (irq_ptr->qib.ac & QIB_AC_OUTBOUND_PCI_SUPPORTED) ? 1 : 0, + css_general_characteristics.aif_tdd, + (irq_ptr->siga_flag.input) ? "R" : " ", + (irq_ptr->siga_flag.output) ? "W" : " ", + (irq_ptr->siga_flag.sync) ? "S" : " ", + (!irq_ptr->siga_flag.no_sync_ti) ? "A" : " ", + (!irq_ptr->siga_flag.no_sync_out_ti) ? "O" : " ", + (!irq_ptr->siga_flag.no_sync_out_pci) ? "P" : " "); printk(KERN_INFO "%s", s); } int __init qdio_setup_init(void) { - char dbf_text[15]; - qdio_q_cache = kmem_cache_create("qdio_q", sizeof(struct qdio_q), 256, 0, NULL); if (!qdio_q_cache) return -ENOMEM; /* Check for OSA/FCP thin interrupts (bit 67). */ - sprintf(dbf_text, "thini%1x", - (css_general_characteristics.aif_osa) ? 1 : 0); - QDIO_DBF_TEXT0(0, setup, dbf_text); + DBF_EVENT("thinint:%1d", + (css_general_characteristics.aif_osa) ? 1 : 0); /* Check for QEBSM support in general (bit 58). */ - sprintf(dbf_text, "cssQBS:%1x", - (qebsm_possible()) ? 1 : 0); - QDIO_DBF_TEXT0(0, setup, dbf_text); + DBF_EVENT("cssQEBSM:%1d", (qebsm_possible()) ? 1 : 0); return 0; } diff --git a/drivers/s390/cio/qdio_thinint.c b/drivers/s390/cio/qdio_thinint.c index c47f6699a43..47ee741d65b 100644 --- a/drivers/s390/cio/qdio_thinint.c +++ b/drivers/s390/cio/qdio_thinint.c @@ -258,8 +258,6 @@ static void tiqdio_thinint_handler(void *ind, void *drv_data) static int set_subchannel_ind(struct qdio_irq *irq_ptr, int reset) { struct scssc_area *scssc_area; - char dbf_text[15]; - void *ptr; int rc; scssc_area = (struct scssc_area *)irq_ptr->chsc_page; @@ -294,19 +292,15 @@ static int set_subchannel_ind(struct qdio_irq *irq_ptr, int reset) rc = chsc_error_from_response(scssc_area->response.code); if (rc) { - sprintf(dbf_text, "sidR%4x", scssc_area->response.code); - QDIO_DBF_TEXT1(0, trace, dbf_text); - QDIO_DBF_TEXT1(0, setup, dbf_text); - ptr = &scssc_area->response; - QDIO_DBF_HEX2(1, setup, &ptr, QDIO_DBF_SETUP_LEN); + DBF_ERROR("%4x SSI r:%4x", irq_ptr->schid.sch_no, + scssc_area->response.code); + DBF_ERROR_HEX(&scssc_area->response, sizeof(void *)); return rc; } - QDIO_DBF_TEXT2(0, setup, "setscind"); - QDIO_DBF_HEX2(0, setup, &scssc_area->summary_indicator_addr, - sizeof(unsigned long)); - QDIO_DBF_HEX2(0, setup, &scssc_area->subchannel_indicator_addr, - sizeof(unsigned long)); + DBF_EVENT("setscind"); + DBF_HEX(&scssc_area->summary_indicator_addr, sizeof(unsigned long)); + DBF_HEX(&scssc_area->subchannel_indicator_addr, sizeof(unsigned long)); return 0; } @@ -327,14 +321,11 @@ void tiqdio_free_memory(void) int __init tiqdio_register_thinints(void) { - char dbf_text[20]; - isc_register(QDIO_AIRQ_ISC); tiqdio_alsi = s390_register_adapter_interrupt(&tiqdio_thinint_handler, NULL, QDIO_AIRQ_ISC); if (IS_ERR(tiqdio_alsi)) { - sprintf(dbf_text, "regthn%lx", PTR_ERR(tiqdio_alsi)); - QDIO_DBF_TEXT0(0, setup, dbf_text); + DBF_EVENT("RTI:%lx", PTR_ERR(tiqdio_alsi)); tiqdio_alsi = NULL; isc_unregister(QDIO_AIRQ_ISC); return -ENOMEM; @@ -360,7 +351,7 @@ void qdio_setup_thinint(struct qdio_irq *irq_ptr) if (!is_thinint_irq(irq_ptr)) return; irq_ptr->dsci = get_indicator(); - QDIO_DBF_HEX1(0, setup, &irq_ptr->dsci, sizeof(void *)); + DBF_HEX(&irq_ptr->dsci, sizeof(void *)); } void qdio_shutdown_thinint(struct qdio_irq *irq_ptr) -- cgit v1.2.3-70-g09d2 From 6bcac508fbebdca52f5a55d69a4316997ecb5391 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 25 Dec 2008 13:38:49 +0100 Subject: [S390] service level interface. Add a new proc interface /proc/service_levels that allows any code to report a relevant service level, e.g. the microcode level of devices, the service level of the hypervisor, etc. Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/sysinfo.h | 11 ++++ drivers/s390/net/qeth_core.h | 2 + drivers/s390/net/qeth_core_main.c | 11 ++++ drivers/s390/sysinfo.c | 127 +++++++++++++++++++++++++++++++++++++- 4 files changed, 149 insertions(+), 2 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/include/asm/sysinfo.h b/arch/s390/include/asm/sysinfo.h index 79d01343f8b..ad93212d9e1 100644 --- a/arch/s390/include/asm/sysinfo.h +++ b/arch/s390/include/asm/sysinfo.h @@ -118,4 +118,15 @@ static inline int stsi(void *sysinfo, int fc, int sel1, int sel2) return r0; } +/* + * Service level reporting interface. + */ +struct service_level { + struct list_head list; + void (*seq_print)(struct seq_file *, struct service_level *); +}; + +int register_service_level(struct service_level *); +int unregister_service_level(struct service_level *); + #endif /* __ASM_S390_SYSINFO_H */ diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h index af6d6045851..1414d3df69d 100644 --- a/drivers/s390/net/qeth_core.h +++ b/drivers/s390/net/qeth_core.h @@ -31,6 +31,7 @@ #include #include #include +#include #include "qeth_core_mpc.h" @@ -733,6 +734,7 @@ struct qeth_card { struct qeth_osn_info osn_info; struct qeth_discipline discipline; atomic_t force_alloc_skb; + struct service_level qeth_service_level; }; struct qeth_card_list_struct { diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index ebdc549da53..3dc752c29c7 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -1138,6 +1138,14 @@ static int qeth_setup_card(struct qeth_card *card) return 0; } +static void qeth_core_sl_print(struct seq_file *m, struct service_level *slr) +{ + struct qeth_card *card = container_of(slr, struct qeth_card, + qeth_service_level); + seq_printf(m, "qeth: %s firmware level %s\n", CARD_BUS_ID(card), + card->info.mcl_level); +} + static struct qeth_card *qeth_alloc_card(void) { struct qeth_card *card; @@ -1157,6 +1165,8 @@ static struct qeth_card *qeth_alloc_card(void) return NULL; } card->options.layer2 = -1; + card->qeth_service_level.seq_print = qeth_core_sl_print; + register_service_level(&card->qeth_service_level); return card; } @@ -3730,6 +3740,7 @@ static void qeth_core_free_card(struct qeth_card *card) free_netdev(card->dev); kfree(card->ip_tbd_list); qeth_free_qdio_buffers(card); + unregister_service_level(&card->qeth_service_level); kfree(card); } diff --git a/drivers/s390/sysinfo.c b/drivers/s390/sysinfo.c index c3e4ab07b9c..0eea9078138 100644 --- a/drivers/s390/sysinfo.c +++ b/drivers/s390/sysinfo.c @@ -1,17 +1,21 @@ /* * drivers/s390/sysinfo.c * - * Copyright (C) 2001 IBM Deutschland Entwicklung GmbH, IBM Corporation - * Author(s): Ulrich Weigand (Ulrich.Weigand@de.ibm.com) + * Copyright IBM Corp. 2001, 2008 + * Author(s): Ulrich Weigand (Ulrich.Weigand@de.ibm.com) + * Martin Schwidefsky */ #include #include #include +#include #include #include +#include #include #include +#include /* Sigh, math-emu. Don't ask. */ #include @@ -271,6 +275,125 @@ static __init int create_proc_sysinfo(void) __initcall(create_proc_sysinfo); +/* + * Service levels interface. + */ + +static DECLARE_RWSEM(service_level_sem); +static LIST_HEAD(service_level_list); + +int register_service_level(struct service_level *slr) +{ + struct service_level *ptr; + + down_write(&service_level_sem); + list_for_each_entry(ptr, &service_level_list, list) + if (ptr == slr) { + up_write(&service_level_sem); + return -EEXIST; + } + list_add_tail(&slr->list, &service_level_list); + up_write(&service_level_sem); + return 0; +} +EXPORT_SYMBOL(register_service_level); + +int unregister_service_level(struct service_level *slr) +{ + struct service_level *ptr, *next; + int rc = -ENOENT; + + down_write(&service_level_sem); + list_for_each_entry_safe(ptr, next, &service_level_list, list) { + if (ptr != slr) + continue; + list_del(&ptr->list); + rc = 0; + break; + } + up_write(&service_level_sem); + return rc; +} +EXPORT_SYMBOL(unregister_service_level); + +static void *service_level_start(struct seq_file *m, loff_t *pos) +{ + down_read(&service_level_sem); + return seq_list_start(&service_level_list, *pos); +} + +static void *service_level_next(struct seq_file *m, void *p, loff_t *pos) +{ + return seq_list_next(p, &service_level_list, pos); +} + +static void service_level_stop(struct seq_file *m, void *p) +{ + up_read(&service_level_sem); +} + +static int service_level_show(struct seq_file *m, void *p) +{ + struct service_level *slr; + + slr = list_entry(p, struct service_level, list); + slr->seq_print(m, slr); + return 0; +} + +static const struct seq_operations service_level_seq_ops = { + .start = service_level_start, + .next = service_level_next, + .stop = service_level_stop, + .show = service_level_show +}; + +static int service_level_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &service_level_seq_ops); +} + +static const struct file_operations service_level_ops = { + .open = service_level_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release +}; + +static void service_level_vm_print(struct seq_file *m, + struct service_level *slr) +{ + char *query_buffer, *str; + + query_buffer = kmalloc(1024, GFP_KERNEL | GFP_DMA); + if (!query_buffer) + return; + cpcmd("QUERY CPLEVEL", query_buffer, 1024, NULL); + str = strchr(query_buffer, '\n'); + if (str) + *str = 0; + seq_printf(m, "VM: %s\n", query_buffer); + kfree(query_buffer); +} + +static struct service_level service_level_vm = { + .seq_print = service_level_vm_print +}; + +static __init int create_proc_service_level(void) +{ + proc_create("service_levels", 0, NULL, &service_level_ops); + if (MACHINE_IS_VM) + register_service_level(&service_level_vm); + return 0; +} + +subsys_initcall(create_proc_service_level); + +/* + * Bogomips calculation based on cpu capability. + */ + int get_cpu_capability(unsigned int *capability) { struct sysinfo_1_2_2 *info; -- cgit v1.2.3-70-g09d2 From 5d360a75f87f288e9a25d56cca503a7c7939e490 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 25 Dec 2008 13:38:56 +0100 Subject: [S390] ftrace: function tracer backend for s390 This implements just the basic function tracer (_mcount) backend for s390. The dynamic variant will come later. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/Kconfig | 1 + arch/s390/include/asm/ftrace.h | 8 ++++++ arch/s390/kernel/Makefile | 6 +++++ arch/s390/kernel/mcount.S | 56 ++++++++++++++++++++++++++++++++++++++++++ arch/s390/kernel/s390_ksyms.c | 5 ++++ 5 files changed, 76 insertions(+) create mode 100644 arch/s390/include/asm/ftrace.h create mode 100644 arch/s390/kernel/mcount.S (limited to 'arch/s390') diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 90ba5d35cf3..e6efc1277b0 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -74,6 +74,7 @@ mainmenu "Linux Kernel Configuration" config S390 def_bool y select USE_GENERIC_SMP_HELPERS if SMP + select HAVE_FUNCTION_TRACER select HAVE_OPROFILE select HAVE_KPROBES select HAVE_KRETPROBES diff --git a/arch/s390/include/asm/ftrace.h b/arch/s390/include/asm/ftrace.h new file mode 100644 index 00000000000..5a5bc75e19d --- /dev/null +++ b/arch/s390/include/asm/ftrace.h @@ -0,0 +1,8 @@ +#ifndef _ASM_S390_FTRACE_H +#define _ASM_S390_FTRACE_H + +#ifndef __ASSEMBLY__ +extern void _mcount(void); +#endif + +#endif /* _ASM_S390_FTRACE_H */ diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile index 235b9484a4d..92730cc3cdb 100644 --- a/arch/s390/kernel/Makefile +++ b/arch/s390/kernel/Makefile @@ -2,6 +2,11 @@ # Makefile for the linux kernel. # +ifdef CONFIG_FUNCTION_TRACER +# Do not trace early boot code +CFLAGS_REMOVE_early.o = -pg +endif + # # Passing null pointers is ok for smp code, since we access the lowcore here. # @@ -34,6 +39,7 @@ obj-$(CONFIG_COMPAT) += compat_linux.o compat_signal.o \ obj-$(CONFIG_VIRT_TIMER) += vtime.o obj-$(CONFIG_STACKTRACE) += stacktrace.o obj-$(CONFIG_KPROBES) += kprobes.o +obj-$(CONFIG_FUNCTION_TRACER) += mcount.o # Kexec part S390_KEXEC_OBJS := machine_kexec.o crash.o diff --git a/arch/s390/kernel/mcount.S b/arch/s390/kernel/mcount.S new file mode 100644 index 00000000000..397d131a345 --- /dev/null +++ b/arch/s390/kernel/mcount.S @@ -0,0 +1,56 @@ +/* + * Copyright IBM Corp. 2008 + * + * Author(s): Heiko Carstens , + * + */ + +#ifndef CONFIG_64BIT +.globl _mcount +_mcount: + stm %r0,%r5,8(%r15) + st %r14,56(%r15) + lr %r1,%r15 + ahi %r15,-96 + l %r3,100(%r15) + la %r2,0(%r14) + st %r1,0(%r15) + la %r3,0(%r3) + bras %r14,0f + .long ftrace_trace_function +0: l %r14,0(%r14) + l %r14,0(%r14) + basr %r14,%r14 + ahi %r15,96 + lm %r0,%r5,8(%r15) + l %r14,56(%r15) + br %r14 + +.globl ftrace_stub +ftrace_stub: + br %r14 + +#else /* CONFIG_64BIT */ + +.globl _mcount +_mcount: + stmg %r0,%r5,16(%r15) + stg %r14,112(%r15) + lgr %r1,%r15 + aghi %r15,-160 + stg %r1,0(%r15) + lgr %r2,%r14 + lg %r3,168(%r15) + larl %r14,ftrace_trace_function + lg %r14,0(%r14) + basr %r14,%r14 + aghi %r15,160 + lmg %r0,%r5,16(%r15) + lg %r14,112(%r15) + br %r14 + +.globl ftrace_stub +ftrace_stub: + br %r14 + +#endif /* CONFIG_64BIT */ diff --git a/arch/s390/kernel/s390_ksyms.c b/arch/s390/kernel/s390_ksyms.c index 48238a114ce..46b90cb0370 100644 --- a/arch/s390/kernel/s390_ksyms.c +++ b/arch/s390/kernel/s390_ksyms.c @@ -14,6 +14,7 @@ #include #include #include +#include #ifdef CONFIG_IP_MULTICAST #include #endif @@ -43,3 +44,7 @@ EXPORT_SYMBOL(csum_fold); EXPORT_SYMBOL(console_mode); EXPORT_SYMBOL(console_devno); EXPORT_SYMBOL(console_irq); + +#ifdef CONFIG_FUNCTION_TRACER +EXPORT_SYMBOL(_mcount); +#endif -- cgit v1.2.3-70-g09d2 From 547e3cec4f7a4846707db19cba43aa58eb07f48f Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 25 Dec 2008 13:39:00 +0100 Subject: [S390] remove ptrace warning on 31 bit. A kernel compile on 31 bit gives the following warnings in ptrace.c: arch/s390/kernel/ptrace.c: In function 'peek_user': arch/s390/kernel/ptrace.c:207: warning: unused variable 'dummy' arch/s390/kernel/ptrace.c: In function 'poke_user': arch/s390/kernel/ptrace.c:315: warning: unused variable 'dummy' Getting rid of the dummy variables removes the warnings. Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/ptrace.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c index 38ff2bce120..75c496f4f16 100644 --- a/arch/s390/kernel/ptrace.c +++ b/arch/s390/kernel/ptrace.c @@ -204,7 +204,6 @@ static unsigned long __peek_user(struct task_struct *child, addr_t addr) static int peek_user(struct task_struct *child, addr_t addr, addr_t data) { - struct user *dummy = NULL; addr_t tmp, mask; /* @@ -213,8 +212,8 @@ peek_user(struct task_struct *child, addr_t addr, addr_t data) */ mask = __ADDR_MASK; #ifdef CONFIG_64BIT - if (addr >= (addr_t) &dummy->regs.acrs && - addr < (addr_t) &dummy->regs.orig_gpr2) + if (addr >= (addr_t) &((struct user *) NULL)->regs.acrs && + addr < (addr_t) &((struct user *) NULL)->regs.orig_gpr2) mask = 3; #endif if ((addr & mask) || addr > sizeof(struct user) - __ADDR_MASK) @@ -312,7 +311,6 @@ static int __poke_user(struct task_struct *child, addr_t addr, addr_t data) static int poke_user(struct task_struct *child, addr_t addr, addr_t data) { - struct user *dummy = NULL; addr_t mask; /* @@ -321,8 +319,8 @@ poke_user(struct task_struct *child, addr_t addr, addr_t data) */ mask = __ADDR_MASK; #ifdef CONFIG_64BIT - if (addr >= (addr_t) &dummy->regs.acrs && - addr < (addr_t) &dummy->regs.orig_gpr2) + if (addr >= (addr_t) &((struct user *) NULL)->regs.acrs && + addr < (addr_t) &((struct user *) NULL)->regs.orig_gpr2) mask = 3; #endif if ((addr & mask) || addr > sizeof(struct user) - __ADDR_MASK) -- cgit v1.2.3-70-g09d2 From d7b604891b7ce7de93e3758ec6b3b7b7964e1884 Mon Sep 17 00:00:00 2001 From: Peter Oberparleiter Date: Thu, 25 Dec 2008 13:39:02 +0100 Subject: [S390] cio: update sac values Values for the sac field have changed - update code accordingly. Signed-off-by: Peter Oberparleiter Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/fcx.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/include/asm/fcx.h b/arch/s390/include/asm/fcx.h index 8be1f3a5804..ef617099507 100644 --- a/arch/s390/include/asm/fcx.h +++ b/arch/s390/include/asm/fcx.h @@ -248,8 +248,8 @@ struct dcw { #define TCCB_MAX_SIZE (sizeof(struct tccb_tcah) + \ TCCB_MAX_DCW * sizeof(struct dcw) + \ sizeof(struct tccb_tcat)) -#define TCCB_SAC_DEFAULT 0xf901 -#define TCCB_SAC_INTRG 0xf902 +#define TCCB_SAC_DEFAULT 0x1ffe +#define TCCB_SAC_INTRG 0x1fff /** * struct tccb_tcah - Transport-Command-Area Header (TCAH) -- cgit v1.2.3-70-g09d2 From edd5378740fe7717b1377ee04e593cfad8d29650 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 25 Dec 2008 13:39:16 +0100 Subject: [S390] mark disabled_wait as noreturn function disabled_wait() won't return, so add an __attribute__((noreturn)). This will remove a false positive finding which our internal code checker reports. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/processor.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/s390') diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h index 4af80af2a88..066b99502e0 100644 --- a/arch/s390/include/asm/processor.h +++ b/arch/s390/include/asm/processor.h @@ -13,6 +13,7 @@ #ifndef __ASM_S390_PROCESSOR_H #define __ASM_S390_PROCESSOR_H +#include #include #ifdef __KERNEL__ @@ -258,7 +259,7 @@ static inline void enabled_wait(void) * Function to drop a processor into disabled wait state */ -static inline void disabled_wait(unsigned long code) +static inline void ATTRIB_NORET disabled_wait(unsigned long code) { unsigned long ctl_buf; psw_t dw_psw; @@ -322,6 +323,7 @@ static inline void disabled_wait(unsigned long code) : "=m" (ctl_buf) : "a" (&dw_psw), "a" (&ctl_buf), "m" (dw_psw) : "cc", "0"); #endif /* __s390x__ */ + while (1); } /* -- cgit v1.2.3-70-g09d2 From 2d6cd2a5908adecd06c8cee2a73814463ed71493 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 25 Dec 2008 13:39:17 +0100 Subject: [S390] remove warnings with functions ending in BUG Functions which end in a BUG() statement and skip the return statement cause compile warnings on s390, e.g.: mm/bootmem.c: In function 'mark_bootmem': mm/bootmem.c:321: warning: control reaches end of non-void function To avoid the warning add an endless loop to the BUG() macro. Signed-off-by: Martin Schwidefsky Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/bug.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'arch/s390') diff --git a/arch/s390/include/asm/bug.h b/arch/s390/include/asm/bug.h index 384e3621e34..7efd0abe888 100644 --- a/arch/s390/include/asm/bug.h +++ b/arch/s390/include/asm/bug.h @@ -47,7 +47,10 @@ #endif /* CONFIG_DEBUG_BUGVERBOSE */ -#define BUG() __EMIT_BUG(0) +#define BUG() do { \ + __EMIT_BUG(0); \ + for (;;); \ +} while (0) #define WARN_ON(x) ({ \ int __ret_warn_on = !!(x); \ -- cgit v1.2.3-70-g09d2 From e37f50e1811d68170e4d58a087cf634b2bf1cef9 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 25 Dec 2008 13:39:19 +0100 Subject: [S390] Add processor type march=z10 and a processor type safety check. This patch adds the code generation option for IBM System z10 and adds a check in head[31,64].S to prevents the execution of a kernel compiled for a new processor type on an old machine. Signed-off-by: Martin Schwidefsky --- arch/s390/Kconfig | 8 ++++++++ arch/s390/Makefile | 1 + arch/s390/kernel/head.S | 49 +++++++++++++++++++++++++++++++++++++++++++++++ arch/s390/kernel/head31.S | 27 +++----------------------- arch/s390/kernel/head64.S | 23 ---------------------- 5 files changed, 61 insertions(+), 47 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index e6efc1277b0..7a18d2380e2 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -230,6 +230,14 @@ config MARCH_Z9_109 Class (z9 BC). The kernel will be slightly faster but will not work on older machines such as the z990, z890, z900, and z800. +config MARCH_Z10 + bool "IBM System z10" + help + Select this to enable optimizations for IBM System z10. The + kernel will be slightly faster but will not work on older + machines such as the z990, z890, z900, z800, z9-109, z9-ec + and z9-bc. + endchoice config PACK_STACK diff --git a/arch/s390/Makefile b/arch/s390/Makefile index 792a4e7743c..578c61f15a4 100644 --- a/arch/s390/Makefile +++ b/arch/s390/Makefile @@ -34,6 +34,7 @@ cflags-$(CONFIG_MARCH_G5) += $(call cc-option,-march=g5) cflags-$(CONFIG_MARCH_Z900) += $(call cc-option,-march=z900) cflags-$(CONFIG_MARCH_Z990) += $(call cc-option,-march=z990) cflags-$(CONFIG_MARCH_Z9_109) += $(call cc-option,-march=z9-109) +cflags-$(CONFIG_MARCH_Z10) += $(call cc-option,-march=z10) #KBUILD_IMAGE is necessary for make rpm KBUILD_IMAGE :=arch/s390/boot/image diff --git a/arch/s390/kernel/head.S b/arch/s390/kernel/head.S index 83477c7dc74..ec7e35f6055 100644 --- a/arch/s390/kernel/head.S +++ b/arch/s390/kernel/head.S @@ -461,6 +461,55 @@ start: .byte 0xf0,0xf1,0xf2,0xf3,0xf4,0xf5,0xf6,0xf7 .byte 0xf8,0xf9,0xfa,0xfb,0xfc,0xfd,0xfe,0xff +# +# startup-code at 0x10000, running in absolute addressing mode +# this is called either by the ipl loader or directly by PSW restart +# or linload or SALIPL +# + .org 0x10000 +startup:basr %r13,0 # get base +.LPG0: + +#ifndef CONFIG_MARCH_G5 + # check processor version against MARCH_{G5,Z900,Z990,Z9_109,Z10} + stidp __LC_CPUID # store cpuid + lhi %r0,(3f-2f) / 2 + la %r1,2f-.LPG0(%r13) +0: clc __LC_CPUID+4(2),0(%r1) + jne 3f + lpsw 1f-.LPG0(13) # machine type not good enough, crash + .align 16 +1: .long 0x000a0000,0x00000000 +2: +#if defined(CONFIG_MARCH_Z10) + .short 0x9672, 0x2064, 0x2066, 0x2084, 0x2086, 0x2094, 0x2096 +#elif defined(CONFIG_MARCH_Z9_109) + .short 0x9672, 0x2064, 0x2066, 0x2084, 0x2086 +#elif defined(CONFIG_MARCH_Z990) + .short 0x9672, 0x2064, 0x2066 +#elif defined(CONFIG_MARCH_Z900) + .short 0x9672 +#endif +3: la %r1,2(%r1) + brct %r0,0b +#endif + + l %r13,0f-.LPG0(%r13) + b 0(%r13) +0: .long startup_continue + +# +# params at 10400 (setup.h) +# + .org PARMAREA + .long 0,0 # IPL_DEVICE + .long 0,0 # INITRD_START + .long 0,0 # INITRD_SIZE + + .org COMMAND_LINE + .byte "root=/dev/ram0 ro" + .byte 0 + #ifdef CONFIG_64BIT #include "head64.S" #else diff --git a/arch/s390/kernel/head31.S b/arch/s390/kernel/head31.S index a816e2de32b..7add8454773 100644 --- a/arch/s390/kernel/head31.S +++ b/arch/s390/kernel/head31.S @@ -10,34 +10,13 @@ * */ -# -# startup-code at 0x10000, running in absolute addressing mode -# this is called either by the ipl loader or directly by PSW restart -# or linload or SALIPL -# - .org 0x10000 -startup:basr %r13,0 # get base -.LPG0: l %r13,0f-.LPG0(%r13) - b 0(%r13) -0: .long startup_continue - -# -# params at 10400 (setup.h) -# - .org PARMAREA - .long 0,0 # IPL_DEVICE - .long 0,0 # INITRD_START - .long 0,0 # INITRD_SIZE - - .org COMMAND_LINE - .byte "root=/dev/ram0 ro" - .byte 0 - .org 0x11000 startup_continue: basr %r13,0 # get base -.LPG1: mvi __LC_AR_MODE_ID,0 # set ESA flag (mode 0) +.LPG1: + + mvi __LC_AR_MODE_ID,0 # set ESA flag (mode 0) lctl %c0,%c15,.Lctl-.LPG1(%r13) # load control registers l %r12,.Lparmaddr-.LPG1(%r13) # pointer to parameter area # move IPL device to lowcore diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S index 1d06961e87b..b1a8ceb522b 100644 --- a/arch/s390/kernel/head64.S +++ b/arch/s390/kernel/head64.S @@ -10,29 +10,6 @@ * */ -# -# startup-code at 0x10000, running in absolute addressing mode -# this is called either by the ipl loader or directly by PSW restart -# or linload or SALIPL -# - .org 0x10000 -startup:basr %r13,0 # get base -.LPG0: l %r13,0f-.LPG0(%r13) - b 0(%r13) -0: .long startup_continue - -# -# params at 10400 (setup.h) -# - .org PARMAREA - .quad 0 # IPL_DEVICE - .quad 0 # INITRD_START - .quad 0 # INITRD_SIZE - - .org COMMAND_LINE - .byte "root=/dev/ram0 ro" - .byte 0 - .org 0x11000 startup_continue: -- cgit v1.2.3-70-g09d2 From 64253acbf189085fb0ceb9a3fbe43b40f4a0a932 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Thu, 25 Dec 2008 13:39:20 +0100 Subject: [S390] s390: use the new byteorder headers Signed-off-by: Harvey Harrison Signed-off-by: Andrew Morton Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/byteorder.h | 72 ++++++++++++--------------------------- 1 file changed, 21 insertions(+), 51 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/include/asm/byteorder.h b/arch/s390/include/asm/byteorder.h index 1fe2492baa8..8bcf277c846 100644 --- a/arch/s390/include/asm/byteorder.h +++ b/arch/s390/include/asm/byteorder.h @@ -11,32 +11,39 @@ #include -#ifdef __GNUC__ +#define __BIG_ENDIAN + +#ifndef __s390x__ +# define __SWAB_64_THRU_32__ +#endif #ifdef __s390x__ -static inline __u64 ___arch__swab64p(const __u64 *x) +static inline __u64 __arch_swab64p(const __u64 *x) { __u64 result; asm volatile("lrvg %0,%1" : "=d" (result) : "m" (*x)); return result; } +#define __arch_swab64p __arch_swab64p -static inline __u64 ___arch__swab64(__u64 x) +static inline __u64 __arch_swab64(__u64 x) { __u64 result; asm volatile("lrvgr %0,%1" : "=d" (result) : "d" (x)); return result; } +#define __arch_swab64 __arch_swab64 -static inline void ___arch__swab64s(__u64 *x) +static inline void __arch_swab64s(__u64 *x) { - *x = ___arch__swab64p(x); + *x = __arch_swab64p(x); } +#define __arch_swab64s __arch_swab64s #endif /* __s390x__ */ -static inline __u32 ___arch__swab32p(const __u32 *x) +static inline __u32 __arch_swab32p(const __u32 *x) { __u32 result; @@ -53,25 +60,20 @@ static inline __u32 ___arch__swab32p(const __u32 *x) #endif /* __s390x__ */ return result; } +#define __arch_swab32p __arch_swab32p -static inline __u32 ___arch__swab32(__u32 x) +#ifdef __s390x__ +static inline __u32 __arch_swab32(__u32 x) { -#ifndef __s390x__ - return ___arch__swab32p(&x); -#else /* __s390x__ */ __u32 result; asm volatile("lrvr %0,%1" : "=d" (result) : "d" (x)); return result; -#endif /* __s390x__ */ -} - -static __inline__ void ___arch__swab32s(__u32 *x) -{ - *x = ___arch__swab32p(x); } +#define __arch_swab32 __arch_swab32 +#endif /* __s390x__ */ -static __inline__ __u16 ___arch__swab16p(const __u16 *x) +static inline __u16 __arch_swab16p(const __u16 *x) { __u16 result; @@ -86,40 +88,8 @@ static __inline__ __u16 ___arch__swab16p(const __u16 *x) #endif /* __s390x__ */ return result; } +#define __arch_swab16p __arch_swab16p -static __inline__ __u16 ___arch__swab16(__u16 x) -{ - return ___arch__swab16p(&x); -} - -static __inline__ void ___arch__swab16s(__u16 *x) -{ - *x = ___arch__swab16p(x); -} - -#ifdef __s390x__ -#define __arch__swab64(x) ___arch__swab64(x) -#define __arch__swab64p(x) ___arch__swab64p(x) -#define __arch__swab64s(x) ___arch__swab64s(x) -#endif /* __s390x__ */ -#define __arch__swab32(x) ___arch__swab32(x) -#define __arch__swab16(x) ___arch__swab16(x) -#define __arch__swab32p(x) ___arch__swab32p(x) -#define __arch__swab16p(x) ___arch__swab16p(x) -#define __arch__swab32s(x) ___arch__swab32s(x) -#define __arch__swab16s(x) ___arch__swab16s(x) - -#ifndef __s390x__ -#if !defined(__STRICT_ANSI__) || defined(__KERNEL__) -# define __BYTEORDER_HAS_U64__ -# define __SWAB_64_THRU_32__ -#endif -#else /* __s390x__ */ -#define __BYTEORDER_HAS_U64__ -#endif /* __s390x__ */ - -#endif /* __GNUC__ */ - -#include +#include #endif /* _S390_BYTEORDER_H */ -- cgit v1.2.3-70-g09d2 From c58d92b233a41b81fbbea6add4ba1b512784f837 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 25 Dec 2008 13:39:21 +0100 Subject: [S390] Remove initial kernel stack backchain initialization. Early init code clears the backchain of the initial kernel stack frame. This is not necessary since it is pre initialized with zeros. Plus it was broken on 64 bit since it cleared only four of eight bytes. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/head31.S | 1 - arch/s390/kernel/head64.S | 1 - 2 files changed, 2 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/head31.S b/arch/s390/kernel/head31.S index 7add8454773..db476d114ca 100644 --- a/arch/s390/kernel/head31.S +++ b/arch/s390/kernel/head31.S @@ -29,7 +29,6 @@ startup_continue: ahi %r15,1<<(PAGE_SHIFT+THREAD_ORDER) # init_task_union+THREAD_SIZE st %r15,__LC_KERNEL_STACK # set end of kernel stack ahi %r15,-96 - xc __SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15) # clear backchain # # Save ipl parameters, clear bss memory, initialize storage key for kernel pages, # and create a kernel NSS if the SAVESYS= parm is defined diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S index b1a8ceb522b..3ccd36b24b8 100644 --- a/arch/s390/kernel/head64.S +++ b/arch/s390/kernel/head64.S @@ -96,7 +96,6 @@ startup_continue: aghi %r15,1<<(PAGE_SHIFT+THREAD_ORDER) # init_task_union + THREAD_SIZE stg %r15,__LC_KERNEL_STACK # set end of kernel stack aghi %r15,-160 - xc __SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15) # clear backchain # # Save ipl parameters, clear bss memory, initialize storage key for kernel pages, # and create a kernel NSS if the SAVESYS= parm is defined -- cgit v1.2.3-70-g09d2 From 9fee8db222a344b9ee407fad9c652ed823690788 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 25 Dec 2008 13:39:22 +0100 Subject: [S390] add new machine types to setup_hwcaps. Add the machine types for z9-bc, z10-ec and z10-bc to the elf_platform detection in setup_hwcaps. Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/setup.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'arch/s390') diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index a0d44bc0c38..323dedecdf5 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -722,8 +722,13 @@ static void __init setup_hwcaps(void) strcpy(elf_platform, "z990"); break; case 0x2094: + case 0x2096: strcpy(elf_platform, "z9-109"); break; + case 0x2097: + case 0x2098: + strcpy(elf_platform, "z10"); + break; } } -- cgit v1.2.3-70-g09d2 From 2b1a61f0a8c714c96277bf16a823a84bafa1397d Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 25 Dec 2008 13:39:23 +0100 Subject: [S390] cpu topology: introduce kernel parameter Introduce a topology=[on|off] kernel parameter which allows to switch cpu topology on/off. Default will be off, since it looks like that for some workloards this doesn't behave very well (on s390). Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- Documentation/kernel-parameters.txt | 8 ++++++++ arch/s390/kernel/topology.c | 12 +++++++++++- 2 files changed, 19 insertions(+), 1 deletion(-) (limited to 'arch/s390') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index c9115c1b672..09ede6c90ad 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2252,6 +2252,14 @@ and is between 256 and 4096 characters. It is defined in the file See comment before function dc390_setup() in drivers/scsi/tmscsim.c. + topology= [S390] + Format: {off | on} + Specify if the kernel should make use of the cpu + topology informations if the hardware supports these. + The scheduler will make use of these informations and + e.g. base its process migration decisions on it. + Default is off. + tp720= [HW,PS2] trix= [HW,OSS] MediaTrix AudioTrix Pro diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c index 36faac50e77..71e6f56bfea 100644 --- a/arch/s390/kernel/topology.c +++ b/arch/s390/kernel/topology.c @@ -58,6 +58,7 @@ struct core_info { cpumask_t mask; }; +static int topology_enabled; static void topology_work_fn(struct work_struct *work); static struct tl_info *tl_info; static struct core_info core_info; @@ -78,7 +79,7 @@ cpumask_t cpu_coregroup_map(unsigned int cpu) cpumask_t mask; cpus_clear(mask); - if (!machine_has_topology) + if (!topology_enabled || !machine_has_topology) return cpu_possible_map; spin_lock_irqsave(&topology_lock, flags); while (core) { @@ -263,6 +264,15 @@ static void topology_interrupt(__u16 code) schedule_work(&topology_work); } +static int __init early_parse_topology(char *p) +{ + if (strncmp(p, "on", 2)) + return 0; + topology_enabled = 1; + return 0; +} +early_param("topology", early_parse_topology); + static int __init init_topology_update(void) { int rc; -- cgit v1.2.3-70-g09d2 From 349f1b671a4b2612c1355612bedadc81f86d26f1 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Thu, 25 Dec 2008 13:39:24 +0100 Subject: [S390] cpu topology: remove dead code Interrupts haven't been implemented. So remove the dead code. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/topology.c | 19 +------------------ 1 file changed, 1 insertion(+), 18 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c index 71e6f56bfea..b3aac53fd6d 100644 --- a/arch/s390/kernel/topology.c +++ b/arch/s390/kernel/topology.c @@ -63,7 +63,6 @@ static void topology_work_fn(struct work_struct *work); static struct tl_info *tl_info; static struct core_info core_info; static int machine_has_topology; -static int machine_has_topology_irq; static struct timer_list topology_timer; static void set_topology_timer(void); static DECLARE_WORK(topology_work, topology_work_fn); @@ -259,11 +258,6 @@ static void set_topology_timer(void) add_timer(&topology_timer); } -static void topology_interrupt(__u16 code) -{ - schedule_work(&topology_work); -} - static int __init early_parse_topology(char *p) { if (strncmp(p, "on", 2)) @@ -283,14 +277,7 @@ static int __init init_topology_update(void) goto out; } init_timer_deferrable(&topology_timer); - if (machine_has_topology_irq) { - rc = register_external_interrupt(0x2005, topology_interrupt); - if (rc) - goto out; - ctl_set_bit(0, 8); - } - else - set_topology_timer(); + set_topology_timer(); out: update_cpu_core_map(); return rc; @@ -311,9 +298,6 @@ void __init s390_init_cpu_topology(void) return; machine_has_topology = 1; - if (facility_bits & (1ULL << 51)) - machine_has_topology_irq = 1; - tl_info = alloc_bootmem_pages(PAGE_SIZE); info = tl_info; stsi(info, 15, 1, 2); @@ -337,5 +321,4 @@ void __init s390_init_cpu_topology(void) return; error: machine_has_topology = 0; - machine_has_topology_irq = 0; } -- cgit v1.2.3-70-g09d2 From c185b783b0993c294ca22ad0a55e6cdf8df4f9d3 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 25 Dec 2008 13:39:25 +0100 Subject: [S390] Remove config options. On s390 we always want to run with precise cputime accounting. Remove the config options VIRT_TIMER and VIRT_CPU_ACCOUNTING. Signed-off-by: Martin Schwidefsky --- arch/s390/Kconfig | 18 +++----------- arch/s390/include/asm/system.h | 4 --- arch/s390/include/asm/timer.h | 9 ------- arch/s390/kernel/Makefile | 3 +-- arch/s390/kernel/entry.S | 56 ++++++------------------------------------ arch/s390/kernel/entry64.S | 56 ++++++------------------------------------ arch/s390/kernel/smp.c | 2 -- arch/s390/kernel/time.c | 4 +-- arch/s390/kernel/vtime.c | 11 --------- 9 files changed, 20 insertions(+), 143 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 7a18d2380e2..8152fefc97b 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -69,6 +69,9 @@ config PGSTE bool default y if KVM +config VIRT_CPU_ACCOUNTING + def_bool y + mainmenu "Linux Kernel Configuration" config S390 @@ -469,22 +472,9 @@ config PAGE_STATES hypervisor. The ESSA instruction is used to do the states changes between a page that has content and the unused state. -config VIRT_TIMER - bool "Virtual CPU timer support" - help - This provides a kernel interface for virtual CPU timers. - Default is disabled. - -config VIRT_CPU_ACCOUNTING - bool "Base user process accounting on virtual cpu timer" - depends on VIRT_TIMER - help - Select this option to use CPU timer deltas to do user - process accounting. - config APPLDATA_BASE bool "Linux - VM Monitor Stream, base infrastructure" - depends on PROC_FS && VIRT_TIMER=y + depends on PROC_FS help This provides a kernel interface for creating and updating z/VM APPLDATA monitor records. The monitor records are updated at certain time diff --git a/arch/s390/include/asm/system.h b/arch/s390/include/asm/system.h index c17a3cfbe24..024ef42ed6d 100644 --- a/arch/s390/include/asm/system.h +++ b/arch/s390/include/asm/system.h @@ -99,13 +99,9 @@ static inline void restore_access_regs(unsigned int *acrs) prev = __switch_to(prev,next); \ } while (0) -#ifdef CONFIG_VIRT_CPU_ACCOUNTING extern void account_vtime(struct task_struct *); extern void account_tick_vtime(struct task_struct *); extern void account_system_vtime(struct task_struct *); -#else -#define account_vtime(x) do { /* empty */ } while (0) -#endif #ifdef CONFIG_PFAULT extern void pfault_irq_init(void); diff --git a/arch/s390/include/asm/timer.h b/arch/s390/include/asm/timer.h index d98d79e35cd..61705d60f99 100644 --- a/arch/s390/include/asm/timer.h +++ b/arch/s390/include/asm/timer.h @@ -48,18 +48,9 @@ extern int del_virt_timer(struct vtimer_list *timer); extern void init_cpu_vtimer(void); extern void vtime_init(void); -#ifdef CONFIG_VIRT_TIMER - extern void vtime_start_cpu_timer(void); extern void vtime_stop_cpu_timer(void); -#else - -static inline void vtime_start_cpu_timer(void) { } -static inline void vtime_stop_cpu_timer(void) { } - -#endif /* CONFIG_VIRT_TIMER */ - #endif /* __KERNEL__ */ #endif /* _ASM_S390_TIMER_H */ diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile index 92730cc3cdb..90dad65c20e 100644 --- a/arch/s390/kernel/Makefile +++ b/arch/s390/kernel/Makefile @@ -20,7 +20,7 @@ CFLAGS_ptrace.o += -DUTS_MACHINE='"$(UTS_MACHINE)"' obj-y := bitmap.o traps.o time.o process.o base.o early.o \ setup.o sys_s390.o ptrace.o signal.o cpcmd.o ebcdic.o \ s390_ext.o debug.o irq.o ipl.o dis.o diag.o mem_detect.o \ - vdso.o + vdso.o vtime.o obj-y += $(if $(CONFIG_64BIT),entry64.o,entry.o) obj-y += $(if $(CONFIG_64BIT),reipl64.o,reipl.o) @@ -36,7 +36,6 @@ obj-$(CONFIG_COMPAT) += compat_linux.o compat_signal.o \ compat_wrapper.o compat_exec_domain.o \ $(compat-obj-y) -obj-$(CONFIG_VIRT_TIMER) += vtime.o obj-$(CONFIG_STACKTRACE) += stacktrace.o obj-$(CONFIG_KPROBES) += kprobes.o obj-$(CONFIG_FUNCTION_TRACER) += mcount.o diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index c1cfc7e39ec..55de521aef7 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -109,13 +109,6 @@ STACK_SIZE = 1 << STACK_SHIFT * R15 - kernel stack pointer */ - .macro STORE_TIMER lc_offset -#ifdef CONFIG_VIRT_CPU_ACCOUNTING - stpt \lc_offset -#endif - .endm - -#ifdef CONFIG_VIRT_CPU_ACCOUNTING .macro UPDATE_VTIME lc_from,lc_to,lc_sum lm %r10,%r11,\lc_from sl %r10,\lc_to @@ -128,7 +121,6 @@ STACK_SIZE = 1 << STACK_SHIFT al %r10,BASED(.Lc_1) 1: stm %r10,%r11,\lc_sum .endm -#endif .macro SAVE_ALL_BASE savearea stm %r12,%r15,\savearea @@ -198,7 +190,7 @@ STACK_SIZE = 1 << STACK_SHIFT ni \psworg+1,0xfd # clear wait state bit .endif lm %r0,%r15,SP_R0(%r15) # load gprs 0-15 of user - STORE_TIMER __LC_EXIT_TIMER + stpt __LC_EXIT_TIMER lpsw \psworg # back to caller .endm @@ -247,20 +239,18 @@ __critical_start: .globl system_call system_call: - STORE_TIMER __LC_SYNC_ENTER_TIMER + stpt __LC_SYNC_ENTER_TIMER sysc_saveall: SAVE_ALL_BASE __LC_SAVE_AREA SAVE_ALL_SVC __LC_SVC_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA lh %r7,0x8a # get svc number from lowcore -#ifdef CONFIG_VIRT_CPU_ACCOUNTING sysc_vtime: UPDATE_VTIME __LC_EXIT_TIMER,__LC_SYNC_ENTER_TIMER,__LC_USER_TIMER sysc_stime: UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER sysc_update: mvc __LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER -#endif sysc_do_svc: l %r9,__LC_THREAD_INFO # load pointer to thread_info struct ltr %r7,%r7 # test for svc 0 @@ -490,20 +480,18 @@ pgm_check_handler: * we just ignore the PER event (FIXME: is there anything we have to do * for LPSW?). */ - STORE_TIMER __LC_SYNC_ENTER_TIMER + stpt __LC_SYNC_ENTER_TIMER SAVE_ALL_BASE __LC_SAVE_AREA tm __LC_PGM_INT_CODE+1,0x80 # check whether we got a per exception bnz BASED(pgm_per) # got per exception -> special case SAVE_ALL_SYNC __LC_PGM_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_PGM_OLD_PSW,__LC_SAVE_AREA -#ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? bz BASED(pgm_no_vtime) UPDATE_VTIME __LC_EXIT_TIMER,__LC_SYNC_ENTER_TIMER,__LC_USER_TIMER UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER mvc __LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER pgm_no_vtime: -#endif l %r9,__LC_THREAD_INFO # load pointer to thread_info struct TRACE_IRQS_OFF l %r3,__LC_PGM_ILC # load program interruption code @@ -536,14 +524,12 @@ pgm_per: pgm_per_std: SAVE_ALL_SYNC __LC_PGM_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_PGM_OLD_PSW,__LC_SAVE_AREA -#ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? bz BASED(pgm_no_vtime2) UPDATE_VTIME __LC_EXIT_TIMER,__LC_SYNC_ENTER_TIMER,__LC_USER_TIMER UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER mvc __LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER pgm_no_vtime2: -#endif l %r9,__LC_THREAD_INFO # load pointer to thread_info struct TRACE_IRQS_OFF l %r1,__TI_task(%r9) @@ -565,11 +551,9 @@ pgm_no_vtime2: pgm_svcper: SAVE_ALL_SYNC __LC_SVC_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA -#ifdef CONFIG_VIRT_CPU_ACCOUNTING UPDATE_VTIME __LC_EXIT_TIMER,__LC_SYNC_ENTER_TIMER,__LC_USER_TIMER UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER mvc __LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER -#endif lh %r7,0x8a # get svc number from lowcore l %r9,__LC_THREAD_INFO # load pointer to thread_info struct TRACE_IRQS_OFF @@ -599,19 +583,17 @@ kernel_per: .globl io_int_handler io_int_handler: - STORE_TIMER __LC_ASYNC_ENTER_TIMER + stpt __LC_ASYNC_ENTER_TIMER stck __LC_INT_CLOCK SAVE_ALL_BASE __LC_SAVE_AREA+16 SAVE_ALL_ASYNC __LC_IO_OLD_PSW,__LC_SAVE_AREA+16 CREATE_STACK_FRAME __LC_IO_OLD_PSW,__LC_SAVE_AREA+16 -#ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? bz BASED(io_no_vtime) UPDATE_VTIME __LC_EXIT_TIMER,__LC_ASYNC_ENTER_TIMER,__LC_USER_TIMER UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER mvc __LC_LAST_UPDATE_TIMER(8),__LC_ASYNC_ENTER_TIMER io_no_vtime: -#endif l %r9,__LC_THREAD_INFO # load pointer to thread_info struct TRACE_IRQS_OFF l %r1,BASED(.Ldo_IRQ) # load address of do_IRQ @@ -741,19 +723,17 @@ io_notify_resume: .globl ext_int_handler ext_int_handler: - STORE_TIMER __LC_ASYNC_ENTER_TIMER + stpt __LC_ASYNC_ENTER_TIMER stck __LC_INT_CLOCK SAVE_ALL_BASE __LC_SAVE_AREA+16 SAVE_ALL_ASYNC __LC_EXT_OLD_PSW,__LC_SAVE_AREA+16 CREATE_STACK_FRAME __LC_EXT_OLD_PSW,__LC_SAVE_AREA+16 -#ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? bz BASED(ext_no_vtime) UPDATE_VTIME __LC_EXIT_TIMER,__LC_ASYNC_ENTER_TIMER,__LC_USER_TIMER UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER mvc __LC_LAST_UPDATE_TIMER(8),__LC_ASYNC_ENTER_TIMER ext_no_vtime: -#endif l %r9,__LC_THREAD_INFO # load pointer to thread_info struct TRACE_IRQS_OFF la %r2,SP_PTREGS(%r15) # address of register-save area @@ -776,7 +756,6 @@ mcck_int_handler: la %r12,__LC_MCK_OLD_PSW tm __LC_MCCK_CODE,0x80 # system damage? bo BASED(mcck_int_main) # yes -> rest of mcck code invalid -#ifdef CONFIG_VIRT_CPU_ACCOUNTING mvc __LC_SAVE_AREA+52(8),__LC_ASYNC_ENTER_TIMER mvc __LC_ASYNC_ENTER_TIMER(8),__LC_CPU_TIMER_SAVE_AREA tm __LC_MCCK_CODE+5,0x02 # stored cpu timer value valid? @@ -793,9 +772,7 @@ mcck_int_handler: la %r14,__LC_LAST_UPDATE_TIMER 0: spt 0(%r14) mvc __LC_ASYNC_ENTER_TIMER(8),0(%r14) -1: -#endif - tm __LC_MCCK_CODE+2,0x09 # mwp + ia of old psw valid? +1: tm __LC_MCCK_CODE+2,0x09 # mwp + ia of old psw valid? bno BASED(mcck_int_main) # no -> skip cleanup critical tm __LC_MCK_OLD_PSW+1,0x01 # test problem state bit bnz BASED(mcck_int_main) # from user -> load async stack @@ -812,7 +789,6 @@ mcck_int_main: be BASED(0f) l %r15,__LC_PANIC_STACK # load panic stack 0: CREATE_STACK_FRAME __LC_MCK_OLD_PSW,__LC_SAVE_AREA+32 -#ifdef CONFIG_VIRT_CPU_ACCOUNTING tm __LC_MCCK_CODE+2,0x08 # mwp of old psw valid? bno BASED(mcck_no_vtime) # no -> skip cleanup critical tm SP_PSW+1(%r15),0x01 # interrupting from user ? @@ -821,7 +797,6 @@ mcck_int_main: UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER mvc __LC_LAST_UPDATE_TIMER(8),__LC_ASYNC_ENTER_TIMER mcck_no_vtime: -#endif l %r9,__LC_THREAD_INFO # load pointer to thread_info struct la %r2,SP_PTREGS(%r15) # load pt_regs l %r1,BASED(.Ls390_mcck) @@ -843,16 +818,13 @@ mcck_no_vtime: mcck_return: mvc __LC_RETURN_MCCK_PSW(8),SP_PSW(%r15) # move return PSW ni __LC_RETURN_MCCK_PSW+1,0xfd # clear wait state bit -#ifdef CONFIG_VIRT_CPU_ACCOUNTING mvc __LC_ASYNC_ENTER_TIMER(8),__LC_SAVE_AREA+52 tm __LC_RETURN_MCCK_PSW+1,0x01 # returning to user ? bno BASED(0f) lm %r0,%r15,SP_R0(%r15) # load gprs 0-15 stpt __LC_EXIT_TIMER lpsw __LC_RETURN_MCCK_PSW # back to caller -0: -#endif - lm %r0,%r15,SP_R0(%r15) # load gprs 0-15 +0: lm %r0,%r15,SP_R0(%r15) # load gprs 0-15 lpsw __LC_RETURN_MCCK_PSW # back to caller RESTORE_ALL __LC_RETURN_MCCK_PSW,0 @@ -976,13 +948,11 @@ cleanup_system_call: b BASED(1f) 0: la %r12,__LC_SAVE_AREA+32 1: -#ifdef CONFIG_VIRT_CPU_ACCOUNTING clc __LC_RETURN_PSW+4(4),BASED(cleanup_system_call_insn+4) bh BASED(0f) mvc __LC_SYNC_ENTER_TIMER(8),__LC_ASYNC_ENTER_TIMER 0: clc __LC_RETURN_PSW+4(4),BASED(cleanup_system_call_insn+8) bhe BASED(cleanup_vtime) -#endif clc __LC_RETURN_PSW+4(4),BASED(cleanup_system_call_insn) bh BASED(0f) mvc __LC_SAVE_AREA(16),0(%r12) @@ -993,7 +963,6 @@ cleanup_system_call: l %r12,__LC_SAVE_AREA+48 # argh st %r15,12(%r12) lh %r7,0x8a -#ifdef CONFIG_VIRT_CPU_ACCOUNTING cleanup_vtime: clc __LC_RETURN_PSW+4(4),BASED(cleanup_system_call_insn+12) bhe BASED(cleanup_stime) @@ -1004,18 +973,15 @@ cleanup_stime: UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER cleanup_update: mvc __LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER -#endif mvc __LC_RETURN_PSW+4(4),BASED(cleanup_table_system_call+4) la %r12,__LC_RETURN_PSW br %r14 cleanup_system_call_insn: .long sysc_saveall + 0x80000000 -#ifdef CONFIG_VIRT_CPU_ACCOUNTING .long system_call + 0x80000000 .long sysc_vtime + 0x80000000 .long sysc_stime + 0x80000000 .long sysc_update + 0x80000000 -#endif cleanup_sysc_return: mvc __LC_RETURN_PSW(4),0(%r12) @@ -1026,11 +992,9 @@ cleanup_sysc_return: cleanup_sysc_leave: clc 4(4,%r12),BASED(cleanup_sysc_leave_insn) be BASED(2f) -#ifdef CONFIG_VIRT_CPU_ACCOUNTING mvc __LC_EXIT_TIMER(8),__LC_ASYNC_ENTER_TIMER clc 4(4,%r12),BASED(cleanup_sysc_leave_insn+4) be BASED(2f) -#endif mvc __LC_RETURN_PSW(8),SP_PSW(%r15) c %r12,BASED(.Lmck_old_psw) bne BASED(0f) @@ -1043,9 +1007,7 @@ cleanup_sysc_leave: br %r14 cleanup_sysc_leave_insn: .long sysc_done - 4 + 0x80000000 -#ifdef CONFIG_VIRT_CPU_ACCOUNTING .long sysc_done - 8 + 0x80000000 -#endif cleanup_io_return: mvc __LC_RETURN_PSW(4),0(%r12) @@ -1056,11 +1018,9 @@ cleanup_io_return: cleanup_io_leave: clc 4(4,%r12),BASED(cleanup_io_leave_insn) be BASED(2f) -#ifdef CONFIG_VIRT_CPU_ACCOUNTING mvc __LC_EXIT_TIMER(8),__LC_ASYNC_ENTER_TIMER clc 4(4,%r12),BASED(cleanup_io_leave_insn+4) be BASED(2f) -#endif mvc __LC_RETURN_PSW(8),SP_PSW(%r15) c %r12,BASED(.Lmck_old_psw) bne BASED(0f) @@ -1073,9 +1033,7 @@ cleanup_io_leave: br %r14 cleanup_io_leave_insn: .long io_done - 4 + 0x80000000 -#ifdef CONFIG_VIRT_CPU_ACCOUNTING .long io_done - 8 + 0x80000000 -#endif /* * Integer constants diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S index ea90d9a676f..16bb4fd1a40 100644 --- a/arch/s390/kernel/entry64.S +++ b/arch/s390/kernel/entry64.S @@ -96,20 +96,12 @@ _TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \ #define LOCKDEP_SYS_EXIT #endif - .macro STORE_TIMER lc_offset -#ifdef CONFIG_VIRT_CPU_ACCOUNTING - stpt \lc_offset -#endif - .endm - -#ifdef CONFIG_VIRT_CPU_ACCOUNTING .macro UPDATE_VTIME lc_from,lc_to,lc_sum lg %r10,\lc_from slg %r10,\lc_to alg %r10,\lc_sum stg %r10,\lc_sum .endm -#endif /* * Register usage in interrupt handlers: @@ -186,7 +178,7 @@ _TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \ ni \psworg+1,0xfd # clear wait state bit .endif lmg %r0,%r15,SP_R0(%r15) # load gprs 0-15 of user - STORE_TIMER __LC_EXIT_TIMER + stpt __LC_EXIT_TIMER lpswe \psworg # back to caller .endm @@ -233,20 +225,18 @@ __critical_start: .globl system_call system_call: - STORE_TIMER __LC_SYNC_ENTER_TIMER + stpt __LC_SYNC_ENTER_TIMER sysc_saveall: SAVE_ALL_BASE __LC_SAVE_AREA SAVE_ALL_SVC __LC_SVC_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA llgh %r7,__LC_SVC_INT_CODE # get svc number from lowcore -#ifdef CONFIG_VIRT_CPU_ACCOUNTING sysc_vtime: UPDATE_VTIME __LC_EXIT_TIMER,__LC_SYNC_ENTER_TIMER,__LC_USER_TIMER sysc_stime: UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER sysc_update: mvc __LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER -#endif sysc_do_svc: lg %r9,__LC_THREAD_INFO # load pointer to thread_info struct ltgr %r7,%r7 # test for svc 0 @@ -469,20 +459,18 @@ pgm_check_handler: * we just ignore the PER event (FIXME: is there anything we have to do * for LPSW?). */ - STORE_TIMER __LC_SYNC_ENTER_TIMER + stpt __LC_SYNC_ENTER_TIMER SAVE_ALL_BASE __LC_SAVE_AREA tm __LC_PGM_INT_CODE+1,0x80 # check whether we got a per exception jnz pgm_per # got per exception -> special case SAVE_ALL_SYNC __LC_PGM_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_PGM_OLD_PSW,__LC_SAVE_AREA -#ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? jz pgm_no_vtime UPDATE_VTIME __LC_EXIT_TIMER,__LC_SYNC_ENTER_TIMER,__LC_USER_TIMER UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER mvc __LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER pgm_no_vtime: -#endif lg %r9,__LC_THREAD_INFO # load pointer to thread_info struct mvc SP_ARGS(8,%r15),__LC_LAST_BREAK TRACE_IRQS_OFF @@ -516,14 +504,12 @@ pgm_per: pgm_per_std: SAVE_ALL_SYNC __LC_PGM_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_PGM_OLD_PSW,__LC_SAVE_AREA -#ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? jz pgm_no_vtime2 UPDATE_VTIME __LC_EXIT_TIMER,__LC_SYNC_ENTER_TIMER,__LC_USER_TIMER UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER mvc __LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER pgm_no_vtime2: -#endif lg %r9,__LC_THREAD_INFO # load pointer to thread_info struct TRACE_IRQS_OFF lg %r1,__TI_task(%r9) @@ -545,11 +531,9 @@ pgm_no_vtime2: pgm_svcper: SAVE_ALL_SYNC __LC_SVC_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA -#ifdef CONFIG_VIRT_CPU_ACCOUNTING UPDATE_VTIME __LC_EXIT_TIMER,__LC_SYNC_ENTER_TIMER,__LC_USER_TIMER UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER mvc __LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER -#endif llgh %r7,__LC_SVC_INT_CODE # get svc number from lowcore lg %r9,__LC_THREAD_INFO # load pointer to thread_info struct lg %r1,__TI_task(%r9) @@ -575,19 +559,17 @@ kernel_per: */ .globl io_int_handler io_int_handler: - STORE_TIMER __LC_ASYNC_ENTER_TIMER + stpt __LC_ASYNC_ENTER_TIMER stck __LC_INT_CLOCK SAVE_ALL_BASE __LC_SAVE_AREA+32 SAVE_ALL_ASYNC __LC_IO_OLD_PSW,__LC_SAVE_AREA+32 CREATE_STACK_FRAME __LC_IO_OLD_PSW,__LC_SAVE_AREA+32 -#ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? jz io_no_vtime UPDATE_VTIME __LC_EXIT_TIMER,__LC_ASYNC_ENTER_TIMER,__LC_USER_TIMER UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER mvc __LC_LAST_UPDATE_TIMER(8),__LC_ASYNC_ENTER_TIMER io_no_vtime: -#endif lg %r9,__LC_THREAD_INFO # load pointer to thread_info struct TRACE_IRQS_OFF la %r2,SP_PTREGS(%r15) # address of register-save area @@ -739,19 +721,17 @@ io_notify_resume: */ .globl ext_int_handler ext_int_handler: - STORE_TIMER __LC_ASYNC_ENTER_TIMER + stpt __LC_ASYNC_ENTER_TIMER stck __LC_INT_CLOCK SAVE_ALL_BASE __LC_SAVE_AREA+32 SAVE_ALL_ASYNC __LC_EXT_OLD_PSW,__LC_SAVE_AREA+32 CREATE_STACK_FRAME __LC_EXT_OLD_PSW,__LC_SAVE_AREA+32 -#ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? jz ext_no_vtime UPDATE_VTIME __LC_EXIT_TIMER,__LC_ASYNC_ENTER_TIMER,__LC_USER_TIMER UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER mvc __LC_LAST_UPDATE_TIMER(8),__LC_ASYNC_ENTER_TIMER ext_no_vtime: -#endif lg %r9,__LC_THREAD_INFO # load pointer to thread_info struct TRACE_IRQS_OFF la %r2,SP_PTREGS(%r15) # address of register-save area @@ -773,7 +753,6 @@ mcck_int_handler: la %r12,__LC_MCK_OLD_PSW tm __LC_MCCK_CODE,0x80 # system damage? jo mcck_int_main # yes -> rest of mcck code invalid -#ifdef CONFIG_VIRT_CPU_ACCOUNTING la %r14,4095 mvc __LC_SAVE_AREA+104(8),__LC_ASYNC_ENTER_TIMER mvc __LC_ASYNC_ENTER_TIMER(8),__LC_CPU_TIMER_SAVE_AREA-4095(%r14) @@ -791,9 +770,7 @@ mcck_int_handler: la %r14,__LC_LAST_UPDATE_TIMER 0: spt 0(%r14) mvc __LC_ASYNC_ENTER_TIMER(8),0(%r14) -1: -#endif - tm __LC_MCCK_CODE+2,0x09 # mwp + ia of old psw valid? +1: tm __LC_MCCK_CODE+2,0x09 # mwp + ia of old psw valid? jno mcck_int_main # no -> skip cleanup critical tm __LC_MCK_OLD_PSW+1,0x01 # test problem state bit jnz mcck_int_main # from user -> load kernel stack @@ -809,7 +786,6 @@ mcck_int_main: jz 0f lg %r15,__LC_PANIC_STACK # load panic stack 0: CREATE_STACK_FRAME __LC_MCK_OLD_PSW,__LC_SAVE_AREA+64 -#ifdef CONFIG_VIRT_CPU_ACCOUNTING tm __LC_MCCK_CODE+2,0x08 # mwp of old psw valid? jno mcck_no_vtime # no -> no timer update tm SP_PSW+1(%r15),0x01 # interrupting from user ? @@ -818,7 +794,6 @@ mcck_int_main: UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER mvc __LC_LAST_UPDATE_TIMER(8),__LC_ASYNC_ENTER_TIMER mcck_no_vtime: -#endif lg %r9,__LC_THREAD_INFO # load pointer to thread_info struct la %r2,SP_PTREGS(%r15) # load pt_regs brasl %r14,s390_do_machine_check @@ -839,14 +814,11 @@ mcck_return: mvc __LC_RETURN_MCCK_PSW(16),SP_PSW(%r15) # move return PSW ni __LC_RETURN_MCCK_PSW+1,0xfd # clear wait state bit lmg %r0,%r15,SP_R0(%r15) # load gprs 0-15 -#ifdef CONFIG_VIRT_CPU_ACCOUNTING mvc __LC_ASYNC_ENTER_TIMER(8),__LC_SAVE_AREA+104 tm __LC_RETURN_MCCK_PSW+1,0x01 # returning to user ? jno 0f stpt __LC_EXIT_TIMER -0: -#endif - lpswe __LC_RETURN_MCCK_PSW # back to caller +0: lpswe __LC_RETURN_MCCK_PSW # back to caller /* * Restart interruption handler, kick starter for additional CPUs @@ -964,13 +936,11 @@ cleanup_system_call: j 1f 0: la %r12,__LC_SAVE_AREA+64 1: -#ifdef CONFIG_VIRT_CPU_ACCOUNTING clc __LC_RETURN_PSW+8(8),BASED(cleanup_system_call_insn+8) jh 0f mvc __LC_SYNC_ENTER_TIMER(8),__LC_ASYNC_ENTER_TIMER 0: clc __LC_RETURN_PSW+8(8),BASED(cleanup_system_call_insn+16) jhe cleanup_vtime -#endif clc __LC_RETURN_PSW+8(8),BASED(cleanup_system_call_insn) jh 0f mvc __LC_SAVE_AREA(32),0(%r12) @@ -981,7 +951,6 @@ cleanup_system_call: lg %r12,__LC_SAVE_AREA+96 # argh stg %r15,24(%r12) llgh %r7,__LC_SVC_INT_CODE -#ifdef CONFIG_VIRT_CPU_ACCOUNTING cleanup_vtime: clc __LC_RETURN_PSW+8(8),BASED(cleanup_system_call_insn+24) jhe cleanup_stime @@ -992,18 +961,15 @@ cleanup_stime: UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER cleanup_update: mvc __LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER -#endif mvc __LC_RETURN_PSW+8(8),BASED(cleanup_table_system_call+8) la %r12,__LC_RETURN_PSW br %r14 cleanup_system_call_insn: .quad sysc_saveall -#ifdef CONFIG_VIRT_CPU_ACCOUNTING .quad system_call .quad sysc_vtime .quad sysc_stime .quad sysc_update -#endif cleanup_sysc_return: mvc __LC_RETURN_PSW(8),0(%r12) @@ -1014,11 +980,9 @@ cleanup_sysc_return: cleanup_sysc_leave: clc 8(8,%r12),BASED(cleanup_sysc_leave_insn) je 2f -#ifdef CONFIG_VIRT_CPU_ACCOUNTING mvc __LC_EXIT_TIMER(8),__LC_ASYNC_ENTER_TIMER clc 8(8,%r12),BASED(cleanup_sysc_leave_insn+8) je 2f -#endif mvc __LC_RETURN_PSW(16),SP_PSW(%r15) cghi %r12,__LC_MCK_OLD_PSW jne 0f @@ -1031,9 +995,7 @@ cleanup_sysc_leave: br %r14 cleanup_sysc_leave_insn: .quad sysc_done - 4 -#ifdef CONFIG_VIRT_CPU_ACCOUNTING .quad sysc_done - 8 -#endif cleanup_io_return: mvc __LC_RETURN_PSW(8),0(%r12) @@ -1044,11 +1006,9 @@ cleanup_io_return: cleanup_io_leave: clc 8(8,%r12),BASED(cleanup_io_leave_insn) je 2f -#ifdef CONFIG_VIRT_CPU_ACCOUNTING mvc __LC_EXIT_TIMER(8),__LC_ASYNC_ENTER_TIMER clc 8(8,%r12),BASED(cleanup_io_leave_insn+8) je 2f -#endif mvc __LC_RETURN_PSW(16),SP_PSW(%r15) cghi %r12,__LC_MCK_OLD_PSW jne 0f @@ -1061,9 +1021,7 @@ cleanup_io_leave: br %r14 cleanup_io_leave_insn: .quad io_done - 4 -#ifdef CONFIG_VIRT_CPU_ACCOUNTING .quad io_done - 8 -#endif /* * Integer constants diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 176a43e5b8b..bf170c698f0 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -441,10 +441,8 @@ int __cpuinit start_secondary(void *cpuvoid) preempt_disable(); /* Enable TOD clock interrupts on the secondary cpu. */ init_cpu_timer(); -#ifdef CONFIG_VIRT_TIMER /* Enable cpu timer interrupts on the secondary cpu. */ init_cpu_vtimer(); -#endif /* Enable pfault pseudo page faults on this cpu. */ pfault_init(); diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index 49652856124..95620296e1a 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -286,10 +286,8 @@ void __init time_init(void) /* Enable TOD clock interrupts on the boot cpu. */ init_cpu_timer(); - -#ifdef CONFIG_VIRT_TIMER + /* Enable cpu timer interrupts on the boot cpu. */ vtime_init(); -#endif } /* diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c index 0fa5dc5d68e..75a6e62ea97 100644 --- a/arch/s390/kernel/vtime.c +++ b/arch/s390/kernel/vtime.c @@ -27,7 +27,6 @@ static ext_int_info_t ext_int_info_timer; static DEFINE_PER_CPU(struct vtimer_queue, virt_cpu_timer); -#ifdef CONFIG_VIRT_CPU_ACCOUNTING /* * Update process times based on virtual cpu times stored by entry.S * to the lowcore fields user_timer, system_timer & steal_clock. @@ -125,16 +124,6 @@ static inline void set_vtimer(__u64 expires) /* store expire time for this CPU timer */ __get_cpu_var(virt_cpu_timer).to_expire = expires; } -#else -static inline void set_vtimer(__u64 expires) -{ - S390_lowcore.last_update_timer = expires; - asm volatile ("SPT %0" : : "m" (S390_lowcore.last_update_timer)); - - /* store expire time for this CPU timer */ - __get_cpu_var(virt_cpu_timer).to_expire = expires; -} -#endif void vtime_start_cpu_timer(void) { -- cgit v1.2.3-70-g09d2 From 4f7e90d6d61fa0e56821787521c12f8a626c4037 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 25 Dec 2008 13:39:26 +0100 Subject: [S390] clear_table inline assembly contraints Tell the compile that the clear_table inline assembly writes to the memory referenced by *s. Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/pgalloc.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'arch/s390') diff --git a/arch/s390/include/asm/pgalloc.h b/arch/s390/include/asm/pgalloc.h index f5b2bf3d7c1..b2658b9220f 100644 --- a/arch/s390/include/asm/pgalloc.h +++ b/arch/s390/include/asm/pgalloc.h @@ -28,6 +28,8 @@ void disable_noexec(struct mm_struct *, struct task_struct *); static inline void clear_table(unsigned long *s, unsigned long val, size_t n) { + typedef struct { char _[n]; } addrtype; + *s = val; n = (n / 256) - 1; asm volatile( @@ -39,7 +41,8 @@ static inline void clear_table(unsigned long *s, unsigned long val, size_t n) "0: mvc 256(256,%0),0(%0)\n" " la %0,256(%0)\n" " brct %1,0b\n" - : "+a" (s), "+d" (n)); + : "+a" (s), "+d" (n), "=m" (*(addrtype *) s) + : "m" (*(addrtype *) s)); } static inline void crst_table_init(unsigned long *crst, unsigned long entry) -- cgit v1.2.3-70-g09d2 From 33b1d09ef317d64c58b135c2e811bb55b3f15b19 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 25 Dec 2008 13:39:27 +0100 Subject: [S390] panic_stack leak in smp_alloc_lowcore Fix freeing of the panic_stack if the allocation of async_stack failed. Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/smp.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index bf170c698f0..1e7db1ab745 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -500,18 +500,15 @@ static int __cpuinit smp_alloc_lowcore(int cpu) save_area = get_zeroed_page(GFP_KERNEL); if (!save_area) - goto out_save_area; + goto out; lowcore->extended_save_area_addr = (u32) save_area; } #endif lowcore_ptr[cpu] = lowcore; return 0; -#ifndef CONFIG_64BIT -out_save_area: - free_page(panic_stack); -#endif out: + free_page(panic_stack); free_pages(async_stack, ASYNC_ORDER); free_pages((unsigned long) lowcore, lc_order); return -ENOMEM; -- cgit v1.2.3-70-g09d2 From 0946100f566d04cc93e86654f3ba588ee94b3f4c Mon Sep 17 00:00:00 2001 From: Hendrik Brueckner Date: Thu, 25 Dec 2008 13:39:29 +0100 Subject: [S390] s390/setup: set default preferred console device "ttyS" This patch sets the default console device for s390. The console= kernel parameter can be still used to switch the preferred console to some other device. In that case, console messages are also printed on the default console device (ttyS0). Signed-off-by: Hendrik Brueckner Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/setup.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'arch/s390') diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 323dedecdf5..bed9dcb34fd 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -740,6 +740,9 @@ static void __init setup_hwcaps(void) void __init setup_arch(char **cmdline_p) { + /* set up preferred console */ + add_preferred_console("ttyS", 0, NULL); + /* * print what head.S has found out about the machine */ -- cgit v1.2.3-70-g09d2 From acfa922c5a00eca0abac0e2b6d6666c606c27345 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Thu, 25 Dec 2008 13:39:30 +0100 Subject: [S390] s390: Remove redundant test The loop above the modified code only terminates when rc is a valid pointer. A simplified version of the semantic patch that makes this change is as follows: (http://www.emn.fr/x-info/coccinelle/) // @r exists@ local idexpression x; expression E; position p1,p2; @@ if (x@p1 == NULL || ...) { ... when forall return ...; } ... when != \(x=E\|x--\|x++\|--x\|++x\|x-=E\|x+=E\|x|=E\|x&=E\|&x\) ( x@p2 == NULL | x@p2 != NULL ) // another path to the test that is not through p1? @s exists@ local idexpression r.x; position r.p1,r.p2; @@ ... when != x@p1 ( x@p2 == NULL | x@p2 != NULL ) @fix depends on !s@ position r.p1,r.p2; expression x,E; statement S1,S2; @@ ( - if ((x@p2 != NULL) || ...) S1 | - if ((x@p2 == NULL) && ...) S1 | - BUG_ON(x@p2 == NULL); ) // Signed-off-by: Julia Lawall Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/debug.c | 2 +- drivers/s390/net/qeth_core_main.c | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c index d80fcd4a7fe..21c90374a58 100644 --- a/arch/s390/kernel/debug.c +++ b/arch/s390/kernel/debug.c @@ -388,7 +388,7 @@ debug_info_copy(debug_info_t* in, int mode) debug_info_free(rc); } while (1); - if(!rc || (mode == NO_AREAS)) + if (mode == NO_AREAS) goto out; for(i = 0; i < in->nr_areas; i++){ diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index 3dc752c29c7..bacd58c5061 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -2335,7 +2335,6 @@ static int qeth_init_input_buffer(struct qeth_card *card, * the QETH_IN_BUF_REQUEUE_THRESHOLD we should never run out off * buffers */ - BUG_ON(!pool_entry); buf->pool_entry = pool_entry; for (i = 0; i < QETH_MAX_BUFFER_ELEMENTS(card); ++i) { -- cgit v1.2.3-70-g09d2 From 2f526e5acbf1179d30bc9fa1d20b69ec4c0bb689 Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Thu, 25 Dec 2008 13:39:34 +0100 Subject: [S390] convert cpcmd printks to pr_xxx macros. Signed-off-by: Christian Borntraeger Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/cpcmd.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/cpcmd.c b/arch/s390/kernel/cpcmd.c index d8c1131e081..3e8b8816f30 100644 --- a/arch/s390/kernel/cpcmd.c +++ b/arch/s390/kernel/cpcmd.c @@ -7,6 +7,9 @@ * Christian Borntraeger (cborntra@de.ibm.com), */ +#define KMSG_COMPONENT "cpcmd" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + #include #include #include @@ -104,8 +107,8 @@ int cpcmd(const char *cmd, char *response, int rlen, int *response_code) (((unsigned long)response + rlen) >> 31)) { lowbuf = kmalloc(rlen, GFP_KERNEL | GFP_DMA); if (!lowbuf) { - printk(KERN_WARNING - "cpcmd: could not allocate response buffer\n"); + pr_warning("The cpcmd kernel function failed to " + "allocate a response buffer\n"); return -ENOMEM; } spin_lock_irqsave(&cpcmd_lock, flags); -- cgit v1.2.3-70-g09d2 From 39f09392498d8ee876bea3ad34f26c447924c818 Mon Sep 17 00:00:00 2001 From: Jan Glauber Date: Thu, 25 Dec 2008 13:39:37 +0100 Subject: [S390] convert cpacf printks to pr_xxx macros. Signed-off-by: Jan Glauber Signed-off-by: Martin Schwidefsky --- arch/s390/crypto/aes_s390.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/crypto/aes_s390.c b/arch/s390/crypto/aes_s390.c index e33f32b54c0..c42cd898f68 100644 --- a/arch/s390/crypto/aes_s390.c +++ b/arch/s390/crypto/aes_s390.c @@ -17,6 +17,9 @@ * */ +#define KMSG_COMPONENT "aes_s390" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + #include #include #include @@ -169,7 +172,8 @@ static int fallback_init_cip(struct crypto_tfm *tfm) CRYPTO_ALG_ASYNC | CRYPTO_ALG_NEED_FALLBACK); if (IS_ERR(sctx->fallback.cip)) { - printk(KERN_ERR "Error allocating fallback algo %s\n", name); + pr_err("Allocating AES fallback algorithm %s failed\n", + name); return PTR_ERR(sctx->fallback.blk); } @@ -349,7 +353,8 @@ static int fallback_init_blk(struct crypto_tfm *tfm) CRYPTO_ALG_ASYNC | CRYPTO_ALG_NEED_FALLBACK); if (IS_ERR(sctx->fallback.blk)) { - printk(KERN_ERR "Error allocating fallback algo %s\n", name); + pr_err("Allocating AES fallback algorithm %s failed\n", + name); return PTR_ERR(sctx->fallback.blk); } @@ -515,9 +520,8 @@ static int __init aes_s390_init(void) /* z9 109 and z9 BC/EC only support 128 bit key length */ if (keylen_flag == AES_KEYLEN_128) - printk(KERN_INFO - "aes_s390: hardware acceleration only available for " - "128 bit keys\n"); + pr_info("AES hardware acceleration is only available for" + " 128-bit keys\n"); ret = crypto_register_alg(&aes_alg); if (ret) -- cgit v1.2.3-70-g09d2 From feab6501d84f7e2f75cf0f5c7d43341431904374 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 25 Dec 2008 13:39:38 +0100 Subject: [S390] convert time printks to pr_xxx macros. Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/time.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index 95620296e1a..5be981a36c3 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -12,6 +12,9 @@ * Copyright (C) 1991, 1992, 1995 Linus Torvalds */ +#define KMSG_COMPONENT "time" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + #include #include #include @@ -319,8 +322,8 @@ static unsigned long long adjust_time(unsigned long long old, } sched_clock_base_cc += delta; if (adjust.offset != 0) { - printk(KERN_NOTICE "etr: time adjusted by %li micro-seconds\n", - adjust.offset); + pr_notice("The ETR interface has adjusted the clock " + "by %li microseconds\n", adjust.offset); adjust.modes = ADJ_OFFSET_SINGLESHOT; do_adjtimex(&adjust); } @@ -481,8 +484,8 @@ static void etr_reset(void) etr_tolec = get_clock(); set_bit(CLOCK_SYNC_HAS_ETR, &clock_sync_flags); } else if (etr_port0_online || etr_port1_online) { - printk(KERN_WARNING "Running on non ETR capable " - "machine, only local mode available.\n"); + pr_warning("The real or virtual hardware system does " + "not provide an ETR interface\n"); etr_port0_online = etr_port1_online = 0; } } @@ -1423,7 +1426,8 @@ static void __init stp_reset(void) if (rc == 0) set_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags); else if (stp_online) { - printk(KERN_WARNING "Running on non STP capable machine.\n"); + pr_warning("The real or virtual hardware system does " + "not provide an STP interface\n"); free_bootmem((unsigned long) stp_page, PAGE_SIZE); stp_page = NULL; stp_online = 0; -- cgit v1.2.3-70-g09d2 From f55495ba1a56adba57ea7d3164a766d83a631c58 Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Thu, 25 Dec 2008 13:39:39 +0100 Subject: [S390] convert hypfs printks to pr_xxx macros. Signed-off-by: Michael Holzheu Signed-off-by: Martin Schwidefsky --- arch/s390/hypfs/hypfs_diag.c | 10 +++++++--- arch/s390/hypfs/inode.c | 14 ++++++++------ 2 files changed, 15 insertions(+), 9 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/hypfs/hypfs_diag.c b/arch/s390/hypfs/hypfs_diag.c index b9a1ce1f28e..b1e892a4381 100644 --- a/arch/s390/hypfs/hypfs_diag.c +++ b/arch/s390/hypfs/hypfs_diag.c @@ -3,10 +3,13 @@ * Hypervisor filesystem for Linux on s390. Diag 204 and 224 * implementation. * - * Copyright (C) IBM Corp. 2006 + * Copyright IBM Corp. 2006, 2008 * Author(s): Michael Holzheu */ +#define KMSG_COMPONENT "hypfs" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + #include #include #include @@ -527,13 +530,14 @@ __init int hypfs_diag_init(void) int rc; if (diag204_probe()) { - printk(KERN_ERR "hypfs: diag 204 not working."); + pr_err("The hardware system does not support hypfs\n"); return -ENODATA; } rc = diag224_get_name_table(); if (rc) { diag204_free_buffer(); - printk(KERN_ERR "hypfs: could not get name table.\n"); + pr_err("The hardware system does not provide all " + "functions required by hypfs\n"); } return rc; } diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c index 36313801cd5..65e93c7ce78 100644 --- a/arch/s390/hypfs/inode.c +++ b/arch/s390/hypfs/inode.c @@ -2,10 +2,13 @@ * arch/s390/hypfs/inode.c * Hypervisor filesystem for Linux on s390. * - * Copyright (C) IBM Corp. 2006 + * Copyright IBM Corp. 2006, 2008 * Author(s): Michael Holzheu */ +#define KMSG_COMPONENT "hypfs" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + #include #include #include @@ -200,7 +203,7 @@ static ssize_t hypfs_aio_write(struct kiocb *iocb, const struct iovec *iov, else rc = hypfs_diag_create_files(sb, sb->s_root); if (rc) { - printk(KERN_ERR "hypfs: Update failed\n"); + pr_err("Updating the hypfs tree failed\n"); hypfs_delete_tree(sb->s_root); goto out; } @@ -252,8 +255,7 @@ static int hypfs_parse_options(char *options, struct super_block *sb) break; case opt_err: default: - printk(KERN_ERR "hypfs: Unrecognized mount option " - "\"%s\" or missing value\n", str); + pr_err("%s is not a valid mount option\n", str); return -EINVAL; } } @@ -317,7 +319,7 @@ static int hypfs_fill_super(struct super_block *sb, void *data, int silent) } hypfs_update_update(sb); sb->s_root = root_dentry; - printk(KERN_INFO "hypfs: Hypervisor filesystem mounted\n"); + pr_info("Hypervisor filesystem mounted\n"); return 0; err_tree: @@ -513,7 +515,7 @@ fail_sysfs: if (!MACHINE_IS_VM) hypfs_diag_exit(); fail_diag: - printk(KERN_ERR "hypfs: Initialization failed with rc = %i.\n", rc); + pr_err("Initialization of hypfs failed with rc=%i\n", rc); return rc; } -- cgit v1.2.3-70-g09d2 From 3b6ed4ab48be2110aa93d9828783bfe86193331d Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 25 Dec 2008 13:39:40 +0100 Subject: [S390] convert setup printks to pr_xxx macros. Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/setup.c | 72 +++++++++++++++++++++++++++++++----------------- 1 file changed, 46 insertions(+), 26 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index bed9dcb34fd..ba724d180be 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -14,6 +14,9 @@ * This file handles the architecture-dependent parts of initialization */ +#define KMSG_COMPONENT "setup" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + #include #include #include @@ -291,8 +294,8 @@ unsigned int switch_amode = 0; #endif EXPORT_SYMBOL_GPL(switch_amode); -static void set_amode_and_uaccess(unsigned long user_amode, - unsigned long user32_amode) +static int set_amode_and_uaccess(unsigned long user_amode, + unsigned long user32_amode) { psw_user_bits = PSW_BASE_BITS | PSW_MASK_DAT | user_amode | PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK | @@ -309,11 +312,11 @@ static void set_amode_and_uaccess(unsigned long user_amode, PSW_MASK_MCHECK | PSW_DEFAULT_KEY; if (MACHINE_HAS_MVCOS) { - printk("mvcos available.\n"); memcpy(&uaccess, &uaccess_mvcos_switch, sizeof(uaccess)); + return 1; } else { - printk("mvcos not available.\n"); memcpy(&uaccess, &uaccess_pt, sizeof(uaccess)); + return 0; } } @@ -328,9 +331,10 @@ static int __init early_parse_switch_amode(char *p) early_param("switch_amode", early_parse_switch_amode); #else /* CONFIG_S390_SWITCH_AMODE */ -static inline void set_amode_and_uaccess(unsigned long user_amode, - unsigned long user32_amode) +static inline int set_amode_and_uaccess(unsigned long user_amode, + unsigned long user32_amode) { + return 0; } #endif /* CONFIG_S390_SWITCH_AMODE */ @@ -355,11 +359,20 @@ early_param("noexec", early_parse_noexec); static void setup_addressing_mode(void) { if (s390_noexec) { - printk("S390 execute protection active, "); - set_amode_and_uaccess(PSW_ASC_SECONDARY, PSW32_ASC_SECONDARY); + if (set_amode_and_uaccess(PSW_ASC_SECONDARY, + PSW32_ASC_SECONDARY)) + pr_info("Execute protection active, " + "mvcos available\n"); + else + pr_info("Execute protection active, " + "mvcos not available\n"); } else if (switch_amode) { - printk("S390 address spaces switched, "); - set_amode_and_uaccess(PSW_ASC_PRIMARY, PSW32_ASC_PRIMARY); + if (set_amode_and_uaccess(PSW_ASC_PRIMARY, PSW32_ASC_PRIMARY)) + pr_info("Address spaces switched, " + "mvcos available\n"); + else + pr_info("Address spaces switched, " + "mvcos not available\n"); } #ifdef CONFIG_TRACE_IRQFLAGS sysc_restore_trace_psw.mask = psw_kernel_bits & ~PSW_MASK_MCHECK; @@ -572,15 +585,15 @@ setup_memory(void) start = PFN_PHYS(start_pfn) + bmap_size + PAGE_SIZE; if (start + INITRD_SIZE > memory_end) { - printk("initrd extends beyond end of memory " - "(0x%08lx > 0x%08lx)\n" + pr_err("initrd extends beyond end of " + "memory (0x%08lx > 0x%08lx) " "disabling initrd\n", start + INITRD_SIZE, memory_end); INITRD_START = INITRD_SIZE = 0; } else { - printk("Moving initrd (0x%08lx -> 0x%08lx, " - "size: %ld)\n", - INITRD_START, start, INITRD_SIZE); + pr_info("Moving initrd (0x%08lx -> " + "0x%08lx, size: %ld)\n", + INITRD_START, start, INITRD_SIZE); memmove((void *) start, (void *) INITRD_START, INITRD_SIZE); INITRD_START = start; @@ -642,8 +655,9 @@ setup_memory(void) initrd_start = INITRD_START; initrd_end = initrd_start + INITRD_SIZE; } else { - printk("initrd extends beyond end of memory " - "(0x%08lx > 0x%08lx)\ndisabling initrd\n", + pr_err("initrd extends beyond end of " + "memory (0x%08lx > 0x%08lx) " + "disabling initrd\n", initrd_start + INITRD_SIZE, memory_end); initrd_start = initrd_end = 0; } @@ -747,21 +761,27 @@ setup_arch(char **cmdline_p) * print what head.S has found out about the machine */ #ifndef CONFIG_64BIT - printk((MACHINE_IS_VM) ? - "We are running under VM (31 bit mode)\n" : - "We are running native (31 bit mode)\n"); - printk((MACHINE_HAS_IEEE) ? - "This machine has an IEEE fpu\n" : - "This machine has no IEEE fpu\n"); + if (MACHINE_IS_VM) + pr_info("Linux is running as a z/VM " + "guest operating system in 31-bit mode\n"); + else + pr_info("Linux is running natively in 31-bit mode\n"); + if (MACHINE_HAS_IEEE) + pr_info("The hardware system has IEEE compatible " + "floating point units\n"); + else + pr_info("The hardware system has no IEEE compatible " + "floating point units\n"); #else /* CONFIG_64BIT */ if (MACHINE_IS_VM) - printk("We are running under VM (64 bit mode)\n"); + pr_info("Linux is running as a z/VM " + "guest operating system in 64-bit mode\n"); else if (MACHINE_IS_KVM) { - printk("We are running under KVM (64 bit mode)\n"); + pr_info("Linux is running under KVM in 64-bit mode\n"); add_preferred_console("hvc", 0, NULL); s390_virtio_console_init(); } else - printk("We are running native (64 bit mode)\n"); + pr_info("Linux is running natively in 64-bit mode\n"); #endif /* CONFIG_64BIT */ /* Have one command line that is parsed and saved in /proc/cmdline */ -- cgit v1.2.3-70-g09d2 From e7534b0ec9a28c53a7bcc874e818a24d3b49680f Mon Sep 17 00:00:00 2001 From: Gerald Schaefer Date: Thu, 25 Dec 2008 13:39:41 +0100 Subject: [S390] convert appldata printks to pr_xxx macros. Signed-off-by: Gerald Schaefer Signed-off-by: Martin Schwidefsky --- arch/s390/appldata/appldata.h | 4 ---- arch/s390/appldata/appldata_base.c | 12 +++++++----- arch/s390/appldata/appldata_os.c | 21 +++++++++------------ 3 files changed, 16 insertions(+), 21 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/appldata/appldata.h b/arch/s390/appldata/appldata.h index 17a2636fec0..f0b23fc759b 100644 --- a/arch/s390/appldata/appldata.h +++ b/arch/s390/appldata/appldata.h @@ -26,10 +26,6 @@ #define CTL_APPLDATA_NET_SUM 2125 #define CTL_APPLDATA_PROC 2126 -#define P_INFO(x...) printk(KERN_INFO MY_PRINT_NAME " info: " x) -#define P_ERROR(x...) printk(KERN_ERR MY_PRINT_NAME " error: " x) -#define P_WARNING(x...) printk(KERN_WARNING MY_PRINT_NAME " status: " x) - struct appldata_ops { struct list_head list; struct ctl_table_header *sysctl_header; diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c index a06a47cdd5e..27b70d8a359 100644 --- a/arch/s390/appldata/appldata_base.c +++ b/arch/s390/appldata/appldata_base.c @@ -10,6 +10,9 @@ * Author: Gerald Schaefer */ +#define KMSG_COMPONENT "appldata" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + #include #include #include @@ -32,7 +35,6 @@ #include "appldata.h" -#define MY_PRINT_NAME "appldata" /* for debug messages, etc. */ #define APPLDATA_CPU_INTERVAL 10000 /* default (CPU) time for sampling interval in milliseconds */ @@ -390,8 +392,8 @@ appldata_generic_handler(ctl_table *ctl, int write, struct file *filp, (unsigned long) ops->data, ops->size, ops->mod_lvl); if (rc != 0) { - P_ERROR("START DIAG 0xDC for %s failed, " - "return code: %d\n", ops->name, rc); + pr_err("Starting the data collection for %s " + "failed with rc=%d\n", ops->name, rc); module_put(ops->owner); } else ops->active = 1; @@ -401,8 +403,8 @@ appldata_generic_handler(ctl_table *ctl, int write, struct file *filp, (unsigned long) ops->data, ops->size, ops->mod_lvl); if (rc != 0) - P_ERROR("STOP DIAG 0xDC for %s failed, " - "return code: %d\n", ops->name, rc); + pr_err("Stopping the data collection for %s " + "failed with rc=%d\n", ops->name, rc); module_put(ops->owner); } spin_unlock(&appldata_ops_lock); diff --git a/arch/s390/appldata/appldata_os.c b/arch/s390/appldata/appldata_os.c index eb44f9f8ab9..55c80ffd42b 100644 --- a/arch/s390/appldata/appldata_os.c +++ b/arch/s390/appldata/appldata_os.c @@ -9,6 +9,9 @@ * Author: Gerald Schaefer */ +#define KMSG_COMPONENT "appldata" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + #include #include #include @@ -22,7 +25,6 @@ #include "appldata.h" -#define MY_PRINT_NAME "appldata_os" /* for debug messages, etc. */ #define LOAD_INT(x) ((x) >> FSHIFT) #define LOAD_FRAC(x) LOAD_INT(((x) & (FIXED_1-1)) * 100) @@ -143,21 +145,16 @@ static void appldata_get_os_data(void *data) (unsigned long) ops.data, new_size, ops.mod_lvl); if (rc != 0) - P_ERROR("os: START NEW DIAG 0xDC failed, " - "return code: %d, new size = %i\n", rc, - new_size); + pr_err("Starting a new OS data collection " + "failed with rc=%d\n", rc); rc = appldata_diag(APPLDATA_RECORD_OS_ID, APPLDATA_STOP_REC, (unsigned long) ops.data, ops.size, ops.mod_lvl); if (rc != 0) - P_ERROR("os: STOP OLD DIAG 0xDC failed, " - "return code: %d, old size = %i\n", rc, - ops.size); - else - P_INFO("os: old record size = %i stopped\n", - ops.size); + pr_err("Stopping a faulty OS data " + "collection failed with rc=%d\n", rc); } ops.size = new_size; } @@ -178,8 +175,8 @@ static int __init appldata_os_init(void) max_size = sizeof(struct appldata_os_data) + (NR_CPUS * sizeof(struct appldata_os_per_cpu)); if (max_size > APPLDATA_MAX_REC_SIZE) { - P_ERROR("Max. size of OS record = %i, bigger than maximum " - "record size (%i)\n", max_size, APPLDATA_MAX_REC_SIZE); + pr_err("Maximum OS record size %i exceeds the maximum " + "record size %i\n", max_size, APPLDATA_MAX_REC_SIZE); rc = -ENOMEM; goto out; } -- cgit v1.2.3-70-g09d2 From c5612c1956597687e4bdf59f2649d20d60f5a2b6 Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Thu, 25 Dec 2008 13:39:43 +0100 Subject: [S390] convert s390 debug feature printks to pr_xxx macros. Signed-off-by: Michael Holzheu Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/debug.c | 37 +++++++++++++++++++------------------ 1 file changed, 19 insertions(+), 18 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c index 21c90374a58..ba03fc0a3a5 100644 --- a/arch/s390/kernel/debug.c +++ b/arch/s390/kernel/debug.c @@ -10,6 +10,9 @@ * Bugreports to: */ +#define KMSG_COMPONENT "s390dbf" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + #include #include #include @@ -693,8 +696,8 @@ debug_info_t *debug_register_mode(const char *name, int pages_per_area, /* Since debugfs currently does not support uid/gid other than root, */ /* we do not allow gid/uid != 0 until we get support for that. */ if ((uid != 0) || (gid != 0)) - printk(KERN_WARNING "debug: Warning - Currently only uid/gid " - "= 0 are supported. Using root as owner now!"); + pr_warning("Root becomes the owner of all s390dbf files " + "in sysfs\n"); if (!initialized) BUG(); mutex_lock(&debug_mutex); @@ -709,7 +712,7 @@ debug_info_t *debug_register_mode(const char *name, int pages_per_area, debug_register_view(rc, &debug_pages_view); out: if (!rc){ - printk(KERN_ERR "debug: debug_register failed for %s\n",name); + pr_err("Registering debug feature %s failed\n", name); } mutex_unlock(&debug_mutex); return rc; @@ -763,8 +766,8 @@ debug_set_size(debug_info_t* id, int nr_areas, int pages_per_area) if(pages_per_area > 0){ new_areas = debug_areas_alloc(pages_per_area, nr_areas); if(!new_areas) { - printk(KERN_WARNING "debug: could not allocate memory "\ - "for pagenumber: %i\n",pages_per_area); + pr_info("Allocating memory for %i pages failed\n", + pages_per_area); rc = -ENOMEM; goto out; } @@ -780,8 +783,7 @@ debug_set_size(debug_info_t* id, int nr_areas, int pages_per_area) memset(id->active_entries,0,sizeof(int)*id->nr_areas); memset(id->active_pages, 0, sizeof(int)*id->nr_areas); spin_unlock_irqrestore(&id->lock,flags); - printk(KERN_INFO "debug: %s: set new size (%i pages)\n"\ - ,id->name, pages_per_area); + pr_info("%s: set new size (%i pages)\n" ,id->name, pages_per_area); out: return rc; } @@ -800,10 +802,9 @@ debug_set_level(debug_info_t* id, int new_level) spin_lock_irqsave(&id->lock,flags); if(new_level == DEBUG_OFF_LEVEL){ id->level = DEBUG_OFF_LEVEL; - printk(KERN_INFO "debug: %s: switched off\n",id->name); + pr_info("%s: switched off\n",id->name); } else if ((new_level > DEBUG_MAX_LEVEL) || (new_level < 0)) { - printk(KERN_INFO - "debug: %s: level %i is out of range (%i - %i)\n", + pr_info("%s: level %i is out of range (%i - %i)\n", id->name, new_level, 0, DEBUG_MAX_LEVEL); } else { id->level = new_level; @@ -1108,8 +1109,8 @@ debug_register_view(debug_info_t * id, struct debug_view *view) pde = debugfs_create_file(view->name, mode, id->debugfs_root_entry, id , &debug_file_ops); if (!pde){ - printk(KERN_WARNING "debug: debugfs_create_file() failed!"\ - " Cannot register view %s/%s\n", id->name,view->name); + pr_err("Registering view %s/%s failed due to out of " + "memory\n", id->name,view->name); rc = -1; goto out; } @@ -1119,10 +1120,8 @@ debug_register_view(debug_info_t * id, struct debug_view *view) break; } if (i == DEBUG_MAX_VIEWS) { - printk(KERN_WARNING "debug: cannot register view %s/%s\n", - id->name,view->name); - printk(KERN_WARNING - "debug: maximum number of views reached (%i)!\n", i); + pr_err("Registering view %s/%s would exceed the maximum " + "number of views %i\n", id->name, view->name, i); debugfs_remove(pde); rc = -1; } else { @@ -1303,7 +1302,8 @@ debug_input_level_fn(debug_info_t * id, struct debug_view *view, new_level = debug_get_uint(str); } if(new_level < 0) { - printk(KERN_INFO "debug: level `%s` is not valid\n", str); + pr_warning("%s is not a valid level for a debug " + "feature\n", str); rc = -EINVAL; } else { debug_set_level(id, new_level); @@ -1380,7 +1380,8 @@ debug_input_flush_fn(debug_info_t * id, struct debug_view *view, goto out; } - printk(KERN_INFO "debug: area `%c` is not valid\n", input_buf[0]); + pr_info("Flushing debug data failed because %c is not a valid " + "area\n", input_buf[0]); out: *offset += user_len; -- cgit v1.2.3-70-g09d2 From 93098bf01578763bb84b627fe893f219cc44b995 Mon Sep 17 00:00:00 2001 From: Hongjie Yang Date: Thu, 25 Dec 2008 13:39:45 +0100 Subject: [S390] convert dcssblk and extmem printks messages to pr_xxx macros. Signed-off-by: Hongjie Yang Signed-off-by: Martin Schwidefsky --- arch/s390/mm/extmem.c | 106 ++++++++++++++++++------------------------- drivers/s390/block/dcssblk.c | 72 ++++++++++++++--------------- 2 files changed, 76 insertions(+), 102 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/mm/extmem.c b/arch/s390/mm/extmem.c index 580fc64cc73..5c845712960 100644 --- a/arch/s390/mm/extmem.c +++ b/arch/s390/mm/extmem.c @@ -7,6 +7,9 @@ * (C) IBM Corporation 2002-2004 */ +#define KMSG_COMPONENT "extmem" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + #include #include #include @@ -24,19 +27,6 @@ #include #include -#define DCSS_DEBUG /* Debug messages on/off */ - -#define DCSS_NAME "extmem" -#ifdef DCSS_DEBUG -#define PRINT_DEBUG(x...) printk(KERN_DEBUG DCSS_NAME " debug:" x) -#else -#define PRINT_DEBUG(x...) do {} while (0) -#endif -#define PRINT_INFO(x...) printk(KERN_INFO DCSS_NAME " info:" x) -#define PRINT_WARN(x...) printk(KERN_WARNING DCSS_NAME " warning:" x) -#define PRINT_ERR(x...) printk(KERN_ERR DCSS_NAME " error:" x) - - #define DCSS_LOADSHR 0x00 #define DCSS_LOADNSR 0x04 #define DCSS_PURGESEG 0x08 @@ -286,7 +276,7 @@ query_segment_type (struct dcss_segment *seg) goto out_free; } if (diag_cc > 1) { - PRINT_WARN ("segment_type: diag returned error %ld\n", vmrc); + pr_warning("Querying a DCSS type failed with rc=%ld\n", vmrc); rc = dcss_diag_translate_rc (vmrc); goto out_free; } @@ -368,7 +358,6 @@ query_segment_type (struct dcss_segment *seg) * -EIO : could not perform query diagnose * -ENOENT : no such segment * -ENOTSUPP: multi-part segment cannot be used with linux - * -ENOSPC : segment cannot be used (overlaps with storage) * -ENOMEM : out of memory * 0 .. 6 : type of segment as defined in include/asm-s390/extmem.h */ @@ -480,9 +469,8 @@ __segment_load (char *name, int do_nonshared, unsigned long *addr, unsigned long goto out_resource; } if (diag_cc > 1) { - PRINT_WARN ("segment_load: could not load segment %s - " - "diag returned error (%ld)\n", - name, end_addr); + pr_warning("Loading DCSS %s failed with rc=%ld\n", name, + end_addr); rc = dcss_diag_translate_rc(end_addr); dcss_diag(&purgeseg_scode, seg->dcss_name, &dummy, &dummy); @@ -496,15 +484,13 @@ __segment_load (char *name, int do_nonshared, unsigned long *addr, unsigned long *addr = seg->start_addr; *end = seg->end; if (do_nonshared) - PRINT_INFO ("segment_load: loaded segment %s range %p .. %p " - "type %s in non-shared mode\n", name, - (void*)seg->start_addr, (void*)seg->end, - segtype_string[seg->vm_segtype]); + pr_info("DCSS %s of range %p to %p and type %s loaded as " + "exclusive-writable\n", name, (void*) seg->start_addr, + (void*) seg->end, segtype_string[seg->vm_segtype]); else { - PRINT_INFO ("segment_load: loaded segment %s range %p .. %p " - "type %s in shared mode\n", name, - (void*)seg->start_addr, (void*)seg->end, - segtype_string[seg->vm_segtype]); + pr_info("DCSS %s of range %p to %p and type %s loaded in " + "shared access mode\n", name, (void*) seg->start_addr, + (void*) seg->end, segtype_string[seg->vm_segtype]); } goto out; out_resource: @@ -593,14 +579,14 @@ segment_modify_shared (char *name, int do_nonshared) goto out_unlock; } if (do_nonshared == seg->do_nonshared) { - PRINT_INFO ("segment_modify_shared: not reloading segment %s" - " - already in requested mode\n",name); + pr_info("DCSS %s is already in the requested access " + "mode\n", name); rc = 0; goto out_unlock; } if (atomic_read (&seg->ref_count) != 1) { - PRINT_WARN ("segment_modify_shared: not reloading segment %s - " - "segment is in use by other driver(s)\n",name); + pr_warning("DCSS %s is in use and cannot be reloaded\n", + name); rc = -EAGAIN; goto out_unlock; } @@ -613,8 +599,8 @@ segment_modify_shared (char *name, int do_nonshared) seg->res->flags |= IORESOURCE_READONLY; if (request_resource(&iomem_resource, seg->res)) { - PRINT_WARN("segment_modify_shared: could not reload segment %s" - " - overlapping resources\n", name); + pr_warning("DCSS %s overlaps with used memory resources " + "and cannot be reloaded\n", name); rc = -EBUSY; kfree(seg->res); goto out_del_mem; @@ -632,9 +618,8 @@ segment_modify_shared (char *name, int do_nonshared) goto out_del_res; } if (diag_cc > 1) { - PRINT_WARN ("segment_modify_shared: could not reload segment %s" - " - diag returned error (%ld)\n", - name, end_addr); + pr_warning("Reloading DCSS %s failed with rc=%ld\n", name, + end_addr); rc = dcss_diag_translate_rc(end_addr); goto out_del_res; } @@ -673,8 +658,7 @@ segment_unload(char *name) mutex_lock(&dcss_lock); seg = segment_by_name (name); if (seg == NULL) { - PRINT_ERR ("could not find segment %s in segment_unload, " - "please report to linux390@de.ibm.com\n",name); + pr_err("Unloading unknown DCSS %s failed\n", name); goto out_unlock; } if (atomic_dec_return(&seg->ref_count) != 0) @@ -709,8 +693,7 @@ segment_save(char *name) seg = segment_by_name (name); if (seg == NULL) { - PRINT_ERR("could not find segment %s in segment_save, please " - "report to linux390@de.ibm.com\n", name); + pr_err("Saving unknown DCSS %s failed\n", name); goto out; } @@ -727,14 +710,14 @@ segment_save(char *name) response = 0; cpcmd(cmd1, NULL, 0, &response); if (response) { - PRINT_ERR("segment_save: DEFSEG failed with response code %i\n", - response); + pr_err("Saving a DCSS failed with DEFSEG response code " + "%i\n", response); goto out; } cpcmd(cmd2, NULL, 0, &response); if (response) { - PRINT_ERR("segment_save: SAVESEG failed with response code %i\n", - response); + pr_err("Saving a DCSS failed with SAVESEG response code " + "%i\n", response); goto out; } out: @@ -749,44 +732,41 @@ void segment_warning(int rc, char *seg_name) { switch (rc) { case -ENOENT: - PRINT_WARN("cannot load/query segment %s, " - "does not exist\n", seg_name); + pr_err("DCSS %s cannot be loaded or queried\n", seg_name); break; case -ENOSYS: - PRINT_WARN("cannot load/query segment %s, " - "not running on VM\n", seg_name); + pr_err("DCSS %s cannot be loaded or queried without " + "z/VM\n", seg_name); break; case -EIO: - PRINT_WARN("cannot load/query segment %s, " - "hardware error\n", seg_name); + pr_err("Loading or querying DCSS %s resulted in a " + "hardware error\n", seg_name); break; case -ENOTSUPP: - PRINT_WARN("cannot load/query segment %s, " - "is a multi-part segment\n", seg_name); + pr_err("DCSS %s has multiple page ranges and cannot be " + "loaded or queried\n", seg_name); break; case -ENOSPC: - PRINT_WARN("cannot load/query segment %s, " - "overlaps with storage\n", seg_name); + pr_err("DCSS %s overlaps with used storage and cannot " + "be loaded\n", seg_name); break; case -EBUSY: - PRINT_WARN("cannot load/query segment %s, " - "overlaps with already loaded dcss\n", seg_name); + pr_err("%s needs used memory resources and cannot be " + "loaded or queried\n", seg_name); break; case -EPERM: - PRINT_WARN("cannot load/query segment %s, " - "already loaded in incompatible mode\n", seg_name); + pr_err("DCSS %s is already loaded in a different access " + "mode\n", seg_name); break; case -ENOMEM: - PRINT_WARN("cannot load/query segment %s, " - "out of memory\n", seg_name); + pr_err("There is not enough memory to load or query " + "DCSS %s\n", seg_name); break; case -ERANGE: - PRINT_WARN("cannot load/query segment %s, " - "exceeds kernel mapping range\n", seg_name); + pr_err("DCSS %s exceeds the kernel mapping range (%lu) " + "and cannot be loaded\n", seg_name, VMEM_MAX_PHYS); break; default: - PRINT_WARN("cannot load/query segment %s, " - "return value %i\n", seg_name, rc); break; } } diff --git a/drivers/s390/block/dcssblk.c b/drivers/s390/block/dcssblk.c index b1b204c448e..26ffc6ab441 100644 --- a/drivers/s390/block/dcssblk.c +++ b/drivers/s390/block/dcssblk.c @@ -4,6 +4,9 @@ * Authors: Carsten Otte, Stefan Weinhuber, Gerald Schaefer */ +#define KMSG_COMPONENT "dcssblk" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + #include #include #include @@ -17,21 +20,11 @@ #include #include -//#define DCSSBLK_DEBUG /* Debug messages on/off */ #define DCSSBLK_NAME "dcssblk" #define DCSSBLK_MINORS_PER_DISK 1 #define DCSSBLK_PARM_LEN 400 #define DCSS_BUS_ID_SIZE 20 -#ifdef DCSSBLK_DEBUG -#define PRINT_DEBUG(x...) printk(KERN_DEBUG DCSSBLK_NAME " debug: " x) -#else -#define PRINT_DEBUG(x...) do {} while (0) -#endif -#define PRINT_INFO(x...) printk(KERN_INFO DCSSBLK_NAME " info: " x) -#define PRINT_WARN(x...) printk(KERN_WARNING DCSSBLK_NAME " warning: " x) -#define PRINT_ERR(x...) printk(KERN_ERR DCSSBLK_NAME " error: " x) - static int dcssblk_open(struct block_device *bdev, fmode_t mode); static int dcssblk_release(struct gendisk *disk, fmode_t mode); static int dcssblk_make_request(struct request_queue *q, struct bio *bio); @@ -262,10 +255,9 @@ dcssblk_is_continuous(struct dcssblk_dev_info *dev_info) /* check continuity */ for (i = 0; i < dev_info->num_of_segments - 1; i++) { if ((sort_list[i].end + 1) != sort_list[i+1].start) { - PRINT_ERR("Segment %s is not contiguous with " - "segment %s\n", - sort_list[i].segment_name, - sort_list[i+1].segment_name); + pr_err("Adjacent DCSSs %s and %s are not " + "contiguous\n", sort_list[i].segment_name, + sort_list[i+1].segment_name); rc = -EINVAL; goto out; } @@ -276,10 +268,10 @@ dcssblk_is_continuous(struct dcssblk_dev_info *dev_info) !(sort_list[i+1].segment_type & SEGMENT_EXCLUSIVE) || (sort_list[i+1].segment_type == SEG_TYPE_ER)) { - PRINT_ERR("Segment %s has different type from " - "segment %s\n", - sort_list[i].segment_name, - sort_list[i+1].segment_name); + pr_err("DCSS %s and DCSS %s have " + "incompatible types\n", + sort_list[i].segment_name, + sort_list[i+1].segment_name); rc = -EINVAL; goto out; } @@ -381,8 +373,9 @@ dcssblk_shared_store(struct device *dev, struct device_attribute *attr, const ch } else if (inbuf[0] == '0') { /* reload segments in exclusive mode */ if (dev_info->segment_type == SEG_TYPE_SC) { - PRINT_ERR("Segment type SC (%s) cannot be loaded in " - "non-shared mode\n", dev_info->segment_name); + pr_err("DCSS %s is of type SC and cannot be " + "loaded as exclusive-writable\n", + dev_info->segment_name); rc = -EINVAL; goto out; } @@ -405,9 +398,8 @@ dcssblk_shared_store(struct device *dev, struct device_attribute *attr, const ch goto out; removeseg: - PRINT_ERR("Could not reload segment(s) of the device %s, removing " - "segment(s) now!\n", - dev_info->segment_name); + pr_err("DCSS device %s is removed after a failed access mode " + "change\n", dev_info->segment_name); temp = entry; list_for_each_entry(entry, &dev_info->seg_list, lh) { if (entry != temp) @@ -455,17 +447,17 @@ dcssblk_save_store(struct device *dev, struct device_attribute *attr, const char if (inbuf[0] == '1') { if (atomic_read(&dev_info->use_count) == 0) { // device is idle => we save immediately - PRINT_INFO("Saving segment(s) of the device %s\n", - dev_info->segment_name); + pr_info("All DCSSs that map to device %s are " + "saved\n", dev_info->segment_name); list_for_each_entry(entry, &dev_info->seg_list, lh) { segment_save(entry->segment_name); } } else { // device is busy => we save it when it becomes // idle in dcssblk_release - PRINT_INFO("Device %s is currently busy, segment(s) " - "will be saved when it becomes idle...\n", - dev_info->segment_name); + pr_info("Device %s is in use, its DCSSs will be " + "saved when it becomes idle\n", + dev_info->segment_name); dev_info->save_pending = 1; } } else if (inbuf[0] == '0') { @@ -473,9 +465,9 @@ dcssblk_save_store(struct device *dev, struct device_attribute *attr, const char // device is busy & the user wants to undo his save // request dev_info->save_pending = 0; - PRINT_INFO("Pending save for segment(s) of the device " - "%s deactivated\n", - dev_info->segment_name); + pr_info("A pending save request for device %s " + "has been canceled\n", + dev_info->segment_name); } } else { up_write(&dcssblk_devices_sem); @@ -615,9 +607,8 @@ dcssblk_add_store(struct device *dev, struct device_attribute *attr, const char seg_byte_size = (dev_info->end - dev_info->start + 1); set_capacity(dev_info->gd, seg_byte_size >> 9); // size in sectors - PRINT_INFO("Loaded segment(s) %s, size = %lu Byte, " - "capacity = %lu (512 Byte) sectors\n", local_buf, - seg_byte_size, seg_byte_size >> 9); + pr_info("Loaded %s with total size %lu bytes and capacity %lu " + "sectors\n", local_buf, seg_byte_size, seg_byte_size >> 9); dev_info->save_pending = 0; dev_info->is_shared = 1; @@ -745,13 +736,15 @@ dcssblk_remove_store(struct device *dev, struct device_attribute *attr, const ch dev_info = dcssblk_get_device_by_name(local_buf); if (dev_info == NULL) { up_write(&dcssblk_devices_sem); - PRINT_WARN("Device %s is not loaded!\n", local_buf); + pr_warning("Device %s cannot be removed because it is not a " + "known device\n", local_buf); rc = -ENODEV; goto out_buf; } if (atomic_read(&dev_info->use_count) != 0) { up_write(&dcssblk_devices_sem); - PRINT_WARN("Device %s is in use!\n", local_buf); + pr_warning("Device %s cannot be removed while it is in " + "use\n", local_buf); rc = -EBUSY; goto out_buf; } @@ -808,8 +801,8 @@ dcssblk_release(struct gendisk *disk, fmode_t mode) down_write(&dcssblk_devices_sem); if (atomic_dec_and_test(&dev_info->use_count) && (dev_info->save_pending)) { - PRINT_INFO("Device %s became idle and is being saved now\n", - dev_info->segment_name); + pr_info("Device %s has become idle and is being saved " + "now\n", dev_info->segment_name); list_for_each_entry(entry, &dev_info->seg_list, lh) { segment_save(entry->segment_name); } @@ -852,7 +845,8 @@ dcssblk_make_request(struct request_queue *q, struct bio *bio) case SEG_TYPE_SC: /* cannot write to these segments */ if (bio_data_dir(bio) == WRITE) { - PRINT_WARN("rejecting write to ro device %s\n", + pr_warning("Writing to %s failed because it " + "is a read-only device\n", dev_name(&dev_info->dev)); goto fail; } -- cgit v1.2.3-70-g09d2 From 395d31d40cc38270dd7c024691404e2eddf0678d Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Thu, 25 Dec 2008 13:39:50 +0100 Subject: [S390] convert cpu related printks to pr_xxx macros. Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/Makefile | 4 +- arch/s390/kernel/processor.c | 98 ++++++++++++++++++++++++++++++++++++++++++++ arch/s390/kernel/setup.c | 88 --------------------------------------- arch/s390/kernel/smp.c | 17 ++++---- arch/s390/kernel/topology.c | 5 ++- 5 files changed, 112 insertions(+), 100 deletions(-) create mode 100644 arch/s390/kernel/processor.c (limited to 'arch/s390') diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile index 90dad65c20e..3edc6c6f258 100644 --- a/arch/s390/kernel/Makefile +++ b/arch/s390/kernel/Makefile @@ -17,8 +17,8 @@ CFLAGS_smp.o := -Wno-nonnull # CFLAGS_ptrace.o += -DUTS_MACHINE='"$(UTS_MACHINE)"' -obj-y := bitmap.o traps.o time.o process.o base.o early.o \ - setup.o sys_s390.o ptrace.o signal.o cpcmd.o ebcdic.o \ +obj-y := bitmap.o traps.o time.o process.o base.o early.o setup.o \ + processor.o sys_s390.o ptrace.o signal.o cpcmd.o ebcdic.o \ s390_ext.o debug.o irq.o ipl.o dis.o diag.o mem_detect.o \ vdso.o vtime.o diff --git a/arch/s390/kernel/processor.c b/arch/s390/kernel/processor.c new file mode 100644 index 00000000000..82c1872cfe8 --- /dev/null +++ b/arch/s390/kernel/processor.c @@ -0,0 +1,98 @@ +/* + * arch/s390/kernel/processor.c + * + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + */ + +#define KMSG_COMPONENT "cpu" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include +#include +#include +#include +#include + +#include +#include +#include + +void __cpuinit print_cpu_info(struct cpuinfo_S390 *cpuinfo) +{ + pr_info("Processor %d started, address %d, identification %06X\n", + cpuinfo->cpu_nr, cpuinfo->cpu_addr, cpuinfo->cpu_id.ident); +} + +/* + * show_cpuinfo - Get information on one CPU for use by procfs. + */ + +static int show_cpuinfo(struct seq_file *m, void *v) +{ + static const char *hwcap_str[8] = { + "esan3", "zarch", "stfle", "msa", "ldisp", "eimm", "dfp", + "edat" + }; + struct cpuinfo_S390 *cpuinfo; + unsigned long n = (unsigned long) v - 1; + int i; + + s390_adjust_jiffies(); + preempt_disable(); + if (!n) { + seq_printf(m, "vendor_id : IBM/S390\n" + "# processors : %i\n" + "bogomips per cpu: %lu.%02lu\n", + num_online_cpus(), loops_per_jiffy/(500000/HZ), + (loops_per_jiffy/(5000/HZ))%100); + seq_puts(m, "features\t: "); + for (i = 0; i < 8; i++) + if (hwcap_str[i] && (elf_hwcap & (1UL << i))) + seq_printf(m, "%s ", hwcap_str[i]); + seq_puts(m, "\n"); + } + + if (cpu_online(n)) { +#ifdef CONFIG_SMP + if (smp_processor_id() == n) + cpuinfo = &S390_lowcore.cpu_data; + else + cpuinfo = &lowcore_ptr[n]->cpu_data; +#else + cpuinfo = &S390_lowcore.cpu_data; +#endif + seq_printf(m, "processor %li: " + "version = %02X, " + "identification = %06X, " + "machine = %04X\n", + n, cpuinfo->cpu_id.version, + cpuinfo->cpu_id.ident, + cpuinfo->cpu_id.machine); + } + preempt_enable(); + return 0; +} + +static void *c_start(struct seq_file *m, loff_t *pos) +{ + return *pos < NR_CPUS ? (void *)((unsigned long) *pos + 1) : NULL; +} + +static void *c_next(struct seq_file *m, void *v, loff_t *pos) +{ + ++*pos; + return c_start(m, pos); +} + +static void c_stop(struct seq_file *m, void *v) +{ +} + +const struct seq_operations cpuinfo_op = { + .start = c_start, + .next = c_next, + .stop = c_stop, + .show = show_cpuinfo, +}; + diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index ba724d180be..b7a1efd5522 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -35,7 +35,6 @@ #include #include #include -#include #include #include #include @@ -829,90 +828,3 @@ setup_arch(char **cmdline_p) /* Setup zfcpdump support */ setup_zfcpdump(console_devno); } - -void __cpuinit print_cpu_info(struct cpuinfo_S390 *cpuinfo) -{ - printk(KERN_INFO "cpu %d " -#ifdef CONFIG_SMP - "phys_idx=%d " -#endif - "vers=%02X ident=%06X machine=%04X unused=%04X\n", - cpuinfo->cpu_nr, -#ifdef CONFIG_SMP - cpuinfo->cpu_addr, -#endif - cpuinfo->cpu_id.version, - cpuinfo->cpu_id.ident, - cpuinfo->cpu_id.machine, - cpuinfo->cpu_id.unused); -} - -/* - * show_cpuinfo - Get information on one CPU for use by procfs. - */ - -static int show_cpuinfo(struct seq_file *m, void *v) -{ - static const char *hwcap_str[8] = { - "esan3", "zarch", "stfle", "msa", "ldisp", "eimm", "dfp", - "edat" - }; - struct cpuinfo_S390 *cpuinfo; - unsigned long n = (unsigned long) v - 1; - int i; - - s390_adjust_jiffies(); - preempt_disable(); - if (!n) { - seq_printf(m, "vendor_id : IBM/S390\n" - "# processors : %i\n" - "bogomips per cpu: %lu.%02lu\n", - num_online_cpus(), loops_per_jiffy/(500000/HZ), - (loops_per_jiffy/(5000/HZ))%100); - seq_puts(m, "features\t: "); - for (i = 0; i < 8; i++) - if (hwcap_str[i] && (elf_hwcap & (1UL << i))) - seq_printf(m, "%s ", hwcap_str[i]); - seq_puts(m, "\n"); - } - - if (cpu_online(n)) { -#ifdef CONFIG_SMP - if (smp_processor_id() == n) - cpuinfo = &S390_lowcore.cpu_data; - else - cpuinfo = &lowcore_ptr[n]->cpu_data; -#else - cpuinfo = &S390_lowcore.cpu_data; -#endif - seq_printf(m, "processor %li: " - "version = %02X, " - "identification = %06X, " - "machine = %04X\n", - n, cpuinfo->cpu_id.version, - cpuinfo->cpu_id.ident, - cpuinfo->cpu_id.machine); - } - preempt_enable(); - return 0; -} - -static void *c_start(struct seq_file *m, loff_t *pos) -{ - return *pos < NR_CPUS ? (void *)((unsigned long) *pos + 1) : NULL; -} -static void *c_next(struct seq_file *m, void *v, loff_t *pos) -{ - ++*pos; - return c_start(m, pos); -} -static void c_stop(struct seq_file *m, void *v) -{ -} -const struct seq_operations cpuinfo_op = { - .start = c_start, - .next = c_next, - .stop = c_stop, - .show = show_cpuinfo, -}; - diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 1e7db1ab745..6fc78541dc5 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -20,6 +20,9 @@ * cpu_number_map in other architectures. */ +#define KMSG_COMPONENT "cpu" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + #include #include #include @@ -251,8 +254,8 @@ static void __init smp_get_save_area(unsigned int cpu, unsigned int phy_cpu) if (ipl_info.type != IPL_TYPE_FCP_DUMP) return; if (cpu >= NR_CPUS) { - printk(KERN_WARNING "Registers for cpu %i not saved since dump " - "kernel was compiled with NR_CPUS=%i\n", cpu, NR_CPUS); + pr_warning("CPU %i exceeds the maximum %i and is excluded from " + "the dump\n", cpu, NR_CPUS - 1); return; } zfcpdump_save_areas[cpu] = kmalloc(sizeof(union save_area), GFP_KERNEL); @@ -425,7 +428,7 @@ static void __init smp_detect_cpus(void) } out: kfree(info); - printk(KERN_INFO "CPUs: %d configured, %d standby\n", c_cpus, s_cpus); + pr_info("%d configured CPUs, %d standby CPUs\n", c_cpus, s_cpus); get_online_cpus(); __smp_rescan_cpus(); put_online_cpus(); @@ -548,12 +551,8 @@ int __cpuinit __cpu_up(unsigned int cpu) ccode = signal_processor_p((__u32)(unsigned long)(lowcore_ptr[cpu]), cpu, sigp_set_prefix); - if (ccode) { - printk("sigp_set_prefix failed for cpu %d " - "with condition code %d\n", - (int) cpu, (int) ccode); + if (ccode) return -EIO; - } idle = current_set[cpu]; cpu_lowcore = lowcore_ptr[cpu]; @@ -636,7 +635,7 @@ void __cpu_die(unsigned int cpu) while (!smp_cpu_not_running(cpu)) cpu_relax(); smp_free_lowcore(cpu); - printk(KERN_INFO "Processor %d spun down\n", cpu); + pr_info("Processor %d stopped\n", cpu); } void cpu_die(void) diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c index b3aac53fd6d..83579ed6293 100644 --- a/arch/s390/kernel/topology.c +++ b/arch/s390/kernel/topology.c @@ -3,6 +3,9 @@ * Author(s): Heiko Carstens */ +#define KMSG_COMPONENT "cpu" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + #include #include #include @@ -306,7 +309,7 @@ void __init s390_init_cpu_topology(void) for (i = 0; i < info->mnest - 2; i++) nr_cores *= info->mag[NR_MAG - 3 - i]; - printk(KERN_INFO "CPU topology:"); + pr_info("The CPU configuration topology of the machine is:"); for (i = 0; i < NR_MAG; i++) printk(" %d", info->mag[i]); printk(" / %d\n", info->mnest); -- cgit v1.2.3-70-g09d2 From 9be3eec2c83848a1ca57ebad13c63c95d0df01e2 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Fri, 26 Dec 2008 22:23:42 +1030 Subject: cpumask: cpu_coregroup_mask(): s390 Like cpu_coregroup_map, but returns a (const) pointer. Compile-tested on s390 (defconfig). Signed-off-by: Rusty Russell Signed-off-by: Mike Travis --- arch/s390/include/asm/topology.h | 1 + arch/s390/kernel/topology.c | 5 +++++ 2 files changed, 6 insertions(+) (limited to 'arch/s390') diff --git a/arch/s390/include/asm/topology.h b/arch/s390/include/asm/topology.h index d96c9164345..fff4a86c602 100644 --- a/arch/s390/include/asm/topology.h +++ b/arch/s390/include/asm/topology.h @@ -6,6 +6,7 @@ #define mc_capable() (1) cpumask_t cpu_coregroup_map(unsigned int cpu); +const struct cpumask *cpu_coregroup_mask(unsigned int cpu); extern cpumask_t cpu_core_map[NR_CPUS]; diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c index a947899dcba..0601cd3231e 100644 --- a/arch/s390/kernel/topology.c +++ b/arch/s390/kernel/topology.c @@ -93,6 +93,11 @@ cpumask_t cpu_coregroup_map(unsigned int cpu) return mask; } +const struct cpumask *cpu_coregroup_mask(unsigned int cpu) +{ + return &cpu_core_map[cpu]; +} + static void add_cpus_to_core(struct tl_cpu *tl_cpu, struct core_info *core) { unsigned int cpu; -- cgit v1.2.3-70-g09d2 From abf137dd7712132ee56d5b3143c2ff61a72a5faa Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 9 Dec 2008 08:11:22 +0100 Subject: aio: make the lookup_ioctx() lockless The mm->ioctx_list is currently protected by a reader-writer lock, so we always grab that lock on the read side for doing ioctx lookups. As the workload is extremely reader biased, turn this into an rcu hlist so we can make lookup_ioctx() lockless. Get rid of the rwlock and use a spinlock for providing update side exclusion. There's usually only 1 entry on this list, so it doesn't make sense to look into fancier data structures. Reviewed-by: Jeff Moyer Signed-off-by: Jens Axboe --- arch/s390/mm/pgtable.c | 4 +- fs/aio.c | 100 ++++++++++++++++++++++++++--------------------- include/linux/aio.h | 5 ++- include/linux/mm_types.h | 5 ++- kernel/fork.c | 4 +- 5 files changed, 67 insertions(+), 51 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/mm/pgtable.c b/arch/s390/mm/pgtable.c index ef3635b52fc..0767827540b 100644 --- a/arch/s390/mm/pgtable.c +++ b/arch/s390/mm/pgtable.c @@ -263,7 +263,7 @@ int s390_enable_sie(void) /* lets check if we are allowed to replace the mm */ task_lock(tsk); if (!tsk->mm || atomic_read(&tsk->mm->mm_users) > 1 || - tsk->mm != tsk->active_mm || tsk->mm->ioctx_list) { + tsk->mm != tsk->active_mm || !hlist_empty(&tsk->mm->ioctx_list)) { task_unlock(tsk); return -EINVAL; } @@ -279,7 +279,7 @@ int s390_enable_sie(void) /* Now lets check again if something happened */ task_lock(tsk); if (!tsk->mm || atomic_read(&tsk->mm->mm_users) > 1 || - tsk->mm != tsk->active_mm || tsk->mm->ioctx_list) { + tsk->mm != tsk->active_mm || !hlist_empty(&tsk->mm->ioctx_list)) { mmput(mm); task_unlock(tsk); return -EINVAL; diff --git a/fs/aio.c b/fs/aio.c index f658441d566..d6f89d3c15e 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -191,6 +191,20 @@ static int aio_setup_ring(struct kioctx *ctx) kunmap_atomic((void *)((unsigned long)__event & PAGE_MASK), km); \ } while(0) +static void ctx_rcu_free(struct rcu_head *head) +{ + struct kioctx *ctx = container_of(head, struct kioctx, rcu_head); + unsigned nr_events = ctx->max_reqs; + + kmem_cache_free(kioctx_cachep, ctx); + + if (nr_events) { + spin_lock(&aio_nr_lock); + BUG_ON(aio_nr - nr_events > aio_nr); + aio_nr -= nr_events; + spin_unlock(&aio_nr_lock); + } +} /* __put_ioctx * Called when the last user of an aio context has gone away, @@ -198,8 +212,6 @@ static int aio_setup_ring(struct kioctx *ctx) */ static void __put_ioctx(struct kioctx *ctx) { - unsigned nr_events = ctx->max_reqs; - BUG_ON(ctx->reqs_active); cancel_delayed_work(&ctx->wq); @@ -208,14 +220,7 @@ static void __put_ioctx(struct kioctx *ctx) mmdrop(ctx->mm); ctx->mm = NULL; pr_debug("__put_ioctx: freeing %p\n", ctx); - kmem_cache_free(kioctx_cachep, ctx); - - if (nr_events) { - spin_lock(&aio_nr_lock); - BUG_ON(aio_nr - nr_events > aio_nr); - aio_nr -= nr_events; - spin_unlock(&aio_nr_lock); - } + call_rcu(&ctx->rcu_head, ctx_rcu_free); } #define get_ioctx(kioctx) do { \ @@ -235,6 +240,7 @@ static struct kioctx *ioctx_alloc(unsigned nr_events) { struct mm_struct *mm; struct kioctx *ctx; + int did_sync = 0; /* Prevent overflows */ if ((nr_events > (0x10000000U / sizeof(struct io_event))) || @@ -267,21 +273,30 @@ static struct kioctx *ioctx_alloc(unsigned nr_events) goto out_freectx; /* limit the number of system wide aios */ - spin_lock(&aio_nr_lock); - if (aio_nr + ctx->max_reqs > aio_max_nr || - aio_nr + ctx->max_reqs < aio_nr) - ctx->max_reqs = 0; - else - aio_nr += ctx->max_reqs; - spin_unlock(&aio_nr_lock); + do { + spin_lock_bh(&aio_nr_lock); + if (aio_nr + nr_events > aio_max_nr || + aio_nr + nr_events < aio_nr) + ctx->max_reqs = 0; + else + aio_nr += ctx->max_reqs; + spin_unlock_bh(&aio_nr_lock); + if (ctx->max_reqs || did_sync) + break; + + /* wait for rcu callbacks to have completed before giving up */ + synchronize_rcu(); + did_sync = 1; + ctx->max_reqs = nr_events; + } while (1); + if (ctx->max_reqs == 0) goto out_cleanup; /* now link into global list. */ - write_lock(&mm->ioctx_list_lock); - ctx->next = mm->ioctx_list; - mm->ioctx_list = ctx; - write_unlock(&mm->ioctx_list_lock); + spin_lock(&mm->ioctx_lock); + hlist_add_head_rcu(&ctx->list, &mm->ioctx_list); + spin_unlock(&mm->ioctx_lock); dprintk("aio: allocated ioctx %p[%ld]: mm=%p mask=0x%x\n", ctx, ctx->user_id, current->mm, ctx->ring_info.nr); @@ -375,11 +390,12 @@ ssize_t wait_on_sync_kiocb(struct kiocb *iocb) */ void exit_aio(struct mm_struct *mm) { - struct kioctx *ctx = mm->ioctx_list; - mm->ioctx_list = NULL; - while (ctx) { - struct kioctx *next = ctx->next; - ctx->next = NULL; + struct kioctx *ctx; + + while (!hlist_empty(&mm->ioctx_list)) { + ctx = hlist_entry(mm->ioctx_list.first, struct kioctx, list); + hlist_del_rcu(&ctx->list); + aio_cancel_all(ctx); wait_for_all_aios(ctx); @@ -394,7 +410,6 @@ void exit_aio(struct mm_struct *mm) atomic_read(&ctx->users), ctx->dead, ctx->reqs_active); put_ioctx(ctx); - ctx = next; } } @@ -555,19 +570,21 @@ int aio_put_req(struct kiocb *req) static struct kioctx *lookup_ioctx(unsigned long ctx_id) { - struct kioctx *ioctx; - struct mm_struct *mm; + struct mm_struct *mm = current->mm; + struct kioctx *ctx = NULL; + struct hlist_node *n; - mm = current->mm; - read_lock(&mm->ioctx_list_lock); - for (ioctx = mm->ioctx_list; ioctx; ioctx = ioctx->next) - if (likely(ioctx->user_id == ctx_id && !ioctx->dead)) { - get_ioctx(ioctx); + rcu_read_lock(); + + hlist_for_each_entry_rcu(ctx, n, &mm->ioctx_list, list) { + if (ctx->user_id == ctx_id && !ctx->dead) { + get_ioctx(ctx); break; } - read_unlock(&mm->ioctx_list_lock); + } - return ioctx; + rcu_read_unlock(); + return ctx; } /* @@ -1215,19 +1232,14 @@ out: static void io_destroy(struct kioctx *ioctx) { struct mm_struct *mm = current->mm; - struct kioctx **tmp; int was_dead; /* delete the entry from the list is someone else hasn't already */ - write_lock(&mm->ioctx_list_lock); + spin_lock(&mm->ioctx_lock); was_dead = ioctx->dead; ioctx->dead = 1; - for (tmp = &mm->ioctx_list; *tmp && *tmp != ioctx; - tmp = &(*tmp)->next) - ; - if (*tmp) - *tmp = ioctx->next; - write_unlock(&mm->ioctx_list_lock); + hlist_del_rcu(&ioctx->list); + spin_unlock(&mm->ioctx_lock); dprintk("aio_release(%p)\n", ioctx); if (likely(!was_dead)) diff --git a/include/linux/aio.h b/include/linux/aio.h index f6b8cf99b59..b16a957030f 100644 --- a/include/linux/aio.h +++ b/include/linux/aio.h @@ -5,6 +5,7 @@ #include #include #include +#include #include @@ -183,7 +184,7 @@ struct kioctx { /* This needs improving */ unsigned long user_id; - struct kioctx *next; + struct hlist_node list; wait_queue_head_t wait; @@ -199,6 +200,8 @@ struct kioctx { struct aio_ring_info ring_info; struct delayed_work wq; + + struct rcu_head rcu_head; }; /* prototypes */ diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index fe825471d5a..9cfc9b627fd 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -232,8 +232,9 @@ struct mm_struct { struct core_state *core_state; /* coredumping support */ /* aio bits */ - rwlock_t ioctx_list_lock; /* aio lock */ - struct kioctx *ioctx_list; + spinlock_t ioctx_lock; + struct hlist_head ioctx_list; + #ifdef CONFIG_MM_OWNER /* * "owner" points to a task that is regarded as the canonical diff --git a/kernel/fork.c b/kernel/fork.c index 6144b36cd89..43cbf30669e 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -415,8 +415,8 @@ static struct mm_struct * mm_init(struct mm_struct * mm, struct task_struct *p) set_mm_counter(mm, file_rss, 0); set_mm_counter(mm, anon_rss, 0); spin_lock_init(&mm->page_table_lock); - rwlock_init(&mm->ioctx_list_lock); - mm->ioctx_list = NULL; + spin_lock_init(&mm->ioctx_lock); + INIT_HLIST_HEAD(&mm->ioctx_list); mm->free_area_cache = TASK_UNMAPPED_BASE; mm->cached_hole_size = ~0UL; mm_init_owner(mm, p); -- cgit v1.2.3-70-g09d2 From db40598863e8cbbd11053ad3c8bae89000f603f9 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Tue, 30 Dec 2008 09:26:02 -0600 Subject: virtio: use KVM_S390_VIRTIO_RING_ALIGN instead of relying on pagesize This doesn't really matter, since s390 pagesize is 4k anyway. Signed-off-by: Rusty Russell Acked-by: Christian Borntraeger --- arch/s390/include/asm/kvm_virtio.h | 4 ++++ drivers/s390/kvm/kvm_virtio.c | 9 ++++++--- 2 files changed, 10 insertions(+), 3 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/include/asm/kvm_virtio.h b/arch/s390/include/asm/kvm_virtio.h index c13568b9351..0503936f101 100644 --- a/arch/s390/include/asm/kvm_virtio.h +++ b/arch/s390/include/asm/kvm_virtio.h @@ -50,6 +50,10 @@ struct kvm_vqconfig { #define KVM_S390_VIRTIO_RESET 1 #define KVM_S390_VIRTIO_SET_STATUS 2 +/* The alignment to use between consumer and producer parts of vring. + * This is pagesize for historical reasons. */ +#define KVM_S390_VIRTIO_RING_ALIGN 4096 + #ifdef __KERNEL__ /* early virtio console setup */ #ifdef CONFIG_S390_GUEST diff --git a/drivers/s390/kvm/kvm_virtio.c b/drivers/s390/kvm/kvm_virtio.c index 3d442444c61..f5a2dbe7557 100644 --- a/drivers/s390/kvm/kvm_virtio.c +++ b/drivers/s390/kvm/kvm_virtio.c @@ -188,7 +188,8 @@ static struct virtqueue *kvm_find_vq(struct virtio_device *vdev, config = kvm_vq_config(kdev->desc)+index; err = vmem_add_mapping(config->address, - vring_size(config->num, PAGE_SIZE)); + vring_size(config->num, + KVM_S390_VIRTIO_RING_ALIGN)); if (err) goto out; @@ -209,7 +210,8 @@ static struct virtqueue *kvm_find_vq(struct virtio_device *vdev, return vq; unmap: vmem_remove_mapping(config->address, - vring_size(config->num, PAGE_SIZE)); + vring_size(config->num, + KVM_S390_VIRTIO_RING_ALIGN)); out: return ERR_PTR(err); } @@ -220,7 +222,8 @@ static void kvm_del_vq(struct virtqueue *vq) vring_del_virtqueue(vq); vmem_remove_mapping(config->address, - vring_size(config->num, PAGE_SIZE)); + vring_size(config->num, + KVM_S390_VIRTIO_RING_ALIGN)); } /* -- cgit v1.2.3-70-g09d2 From 2bb23a63f22f0e2d91fee93ff5ca9c29e180b146 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 1 Jan 2009 10:12:20 +1030 Subject: cpumask: Introduce topology_core_cpumask()/topology_thread_cpumask(): s390 Impact: New API The old topology_core_siblings() and topology_thread_siblings() return a cpumask_t; these new ones return a (const) struct cpumask *. Signed-off-by: Rusty Russell Signed-off-by: Mike Travis --- arch/s390/include/asm/topology.h | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/s390') diff --git a/arch/s390/include/asm/topology.h b/arch/s390/include/asm/topology.h index fff4a86c602..c93eb50e1d0 100644 --- a/arch/s390/include/asm/topology.h +++ b/arch/s390/include/asm/topology.h @@ -11,6 +11,7 @@ const struct cpumask *cpu_coregroup_mask(unsigned int cpu); extern cpumask_t cpu_core_map[NR_CPUS]; #define topology_core_siblings(cpu) (cpu_core_map[cpu]) +#define topology_core_cpumask(cpu) (&cpu_core_map[cpu]) int topology_set_cpu_management(int fc); void topology_schedule_update(void); -- cgit v1.2.3-70-g09d2