diff options
Diffstat (limited to 'arch/um')
92 files changed, 1581 insertions, 722 deletions
diff --git a/arch/um/Kconfig b/arch/um/Kconfig index c5292181a66..f945444df49 100644 --- a/arch/um/Kconfig +++ b/arch/um/Kconfig @@ -35,6 +35,11 @@ config GENERIC_CALIBRATE_DELAY bool default y +# Used in kernel/irq/manage.c and include/linux/irq.h +config IRQ_RELEASE_METHOD + bool + default y + menu "UML-specific options" config MODE_TT @@ -69,6 +74,7 @@ config MODE_SKAS option will shrink the UML binary slightly. source "arch/um/Kconfig_arch" +source "mm/Kconfig" config LD_SCRIPT_STATIC bool @@ -122,7 +128,6 @@ config HOSTFS config HPPFS tristate "HoneyPot ProcFS (EXPERIMENTAL)" - depends on BROKEN help hppfs (HoneyPot ProcFS) is a filesystem which allows UML /proc entries to be overridden, removed, or fabricated from the host. @@ -135,8 +140,9 @@ config HPPFS You only need this if you are setting up a UML honeypot. Otherwise, it is safe to say 'N' here. - If you are actively using it, please ask for it to be fixed. In this - moment, it does not work on 2.6 (it works somehow on 2.4). + If you are actively using it, please report any problems, since it's + getting fixed. In this moment, it is experimental on 2.6 (it works on + 2.4). config MCONSOLE bool "Management console" @@ -269,6 +275,8 @@ endmenu source "init/Kconfig" +source "net/Kconfig" + source "drivers/base/Kconfig" source "arch/um/Kconfig_char" @@ -281,7 +289,7 @@ config NETDEVICES source "arch/um/Kconfig_net" -source "net/Kconfig" +source "drivers/net/Kconfig" source "fs/Kconfig" diff --git a/arch/um/Kconfig_i386 b/arch/um/Kconfig_i386 index e41f3748d30..27c18a8d9d1 100644 --- a/arch/um/Kconfig_i386 +++ b/arch/um/Kconfig_i386 @@ -19,6 +19,18 @@ config 3_LEVEL_PGTABLES memory. All the memory that can't be mapped directly will be treated as high memory. +config STUB_CODE + hex + default 0xbfffe000 + +config STUB_DATA + hex + default 0xbffff000 + +config STUB_START + hex + default STUB_CODE + config ARCH_HAS_SC_SIGNALS bool default y diff --git a/arch/um/Kconfig_net b/arch/um/Kconfig_net index 1c2f9a70d91..fa2ab2dd78b 100644 --- a/arch/um/Kconfig_net +++ b/arch/um/Kconfig_net @@ -135,7 +135,7 @@ config UML_NET_MCAST config UML_NET_PCAP bool "pcap transport" - depends on UML_NET && BROKEN + depends on UML_NET && EXPERIMENTAL help The pcap transport makes a pcap packet stream on the host look like an ethernet device inside UML. This is useful for making diff --git a/arch/um/Kconfig_x86_64 b/arch/um/Kconfig_x86_64 index f162f50f0b1..735a047c890 100644 --- a/arch/um/Kconfig_x86_64 +++ b/arch/um/Kconfig_x86_64 @@ -14,6 +14,18 @@ config 3_LEVEL_PGTABLES bool default y +config STUB_CODE + hex + default 0x7fbfffe000 + +config STUB_DATA + hex + default 0x7fbffff000 + +config STUB_START + hex + default STUB_CODE + config ARCH_HAS_SC_SIGNALS bool default n diff --git a/arch/um/Makefile b/arch/um/Makefile index f2a0c40a920..f5a83a72aa7 100644 --- a/arch/um/Makefile +++ b/arch/um/Makefile @@ -51,25 +51,26 @@ MRPROPER_DIRS += $(ARCH_DIR)/include2 endif SYS_DIR := $(ARCH_DIR)/include/sysdep-$(SUBARCH) -include $(srctree)/$(ARCH_DIR)/Makefile-$(SUBARCH) +# -Dvmap=kernel_vmap affects everything, and prevents anything from +# referencing the libpcap.o symbol so named. + +CFLAGS += $(CFLAGS-y) -D__arch_um__ -DSUBARCH=\"$(SUBARCH)\" \ + $(ARCH_INCLUDE) $(MODE_INCLUDE) -Dvmap=kernel_vmap -core-y += $(SUBARCH_CORE) -libs-y += $(SUBARCH_LIBS) +USER_CFLAGS := $(patsubst -I%,,$(CFLAGS)) +USER_CFLAGS := $(patsubst -D__KERNEL__,,$(USER_CFLAGS)) $(ARCH_INCLUDE) \ + $(MODE_INCLUDE) # -Derrno=kernel_errno - This turns all kernel references to errno into # kernel_errno to separate them from the libc errno. This allows -fno-common # in CFLAGS. Otherwise, it would cause ld to complain about the two different # errnos. -CFLAGS += $(CFLAGS-y) -D__arch_um__ -DSUBARCH=\"$(SUBARCH)\" \ - $(ARCH_INCLUDE) $(MODE_INCLUDE) - -USER_CFLAGS := $(patsubst -I%,,$(CFLAGS)) -USER_CFLAGS := $(patsubst -D__KERNEL__,,$(USER_CFLAGS)) $(ARCH_INCLUDE) \ - $(MODE_INCLUDE) $(ARCH_USER_CFLAGS) CFLAGS += -Derrno=kernel_errno -Dsigprocmask=kernel_sigprocmask CFLAGS += $(call cc-option,-fno-unit-at-a-time,) +include $(srctree)/$(ARCH_DIR)/Makefile-$(SUBARCH) + #This will adjust *FLAGS accordingly to the platform. include $(srctree)/$(ARCH_DIR)/Makefile-os-$(OS) @@ -116,18 +117,19 @@ CONFIG_KERNEL_STACK_ORDER ?= 2 STACK_SIZE := $(shell echo $$[ 4096 * (1 << $(CONFIG_KERNEL_STACK_ORDER)) ] ) ifndef START - START = $$(($(TOP_ADDR) - $(SIZE))) + START = $(shell echo $$[ $(TOP_ADDR) - $(SIZE) ] ) endif -CPPFLAGS_vmlinux.lds = $(shell echo -U$(SUBARCH) \ +CPPFLAGS_vmlinux.lds = -U$(SUBARCH) \ -DSTART=$(START) -DELF_ARCH=$(ELF_ARCH) \ - -DELF_FORMAT=\"$(ELF_FORMAT)\" $(CPP_MODE-y) \ - -DKERNEL_STACK_SIZE=$(STACK_SIZE)) + -DELF_FORMAT="$(ELF_FORMAT)" $(CPP_MODE-y) \ + -DKERNEL_STACK_SIZE=$(STACK_SIZE) \ + -DUNMAP_PATH=arch/um/sys-$(SUBARCH)/unmap_fin.o #The wrappers will select whether using "malloc" or the kernel allocator. LINK_WRAPS = -Wl,--wrap,malloc -Wl,--wrap,free -Wl,--wrap,calloc -CFLAGS_vmlinux = $(LINK-y) $(LINK_WRAPS) +CFLAGS_vmlinux := $(LINK-y) $(LINK_WRAPS) define cmd_vmlinux__ $(CC) $(CFLAGS_vmlinux) -o $@ \ -Wl,-T,$(vmlinux-lds) $(vmlinux-init) \ @@ -140,7 +142,8 @@ endef #When cleaning we don't include .config, so we don't include #TT or skas makefiles and don't clean skas_ptregs.h. CLEAN_FILES += linux x.i gmon.out $(ARCH_DIR)/include/uml-config.h \ - $(GEN_HEADERS) $(ARCH_DIR)/include/skas_ptregs.h + $(GEN_HEADERS) $(ARCH_DIR)/include/skas_ptregs.h \ + $(ARCH_DIR)/include/user_constants.h MRPROPER_FILES += $(SYMLINK_HEADERS) $(ARCH_SYMLINKS) \ $(addprefix $(ARCH_DIR)/kernel/,$(KERN_SYMLINKS)) $(ARCH_DIR)/os \ @@ -242,7 +245,7 @@ $(ARCH_DIR)/util: scripts_basic $(SYS_DIR)/sc.h $(ARCH_DIR)/kernel-offsets.h FOR $(ARCH_DIR)/kernel/skas/util: scripts_basic $(ARCH_DIR)/user-offsets.h FORCE $(Q)$(MAKE) $(build)=$@ -$(ARCH_DIR)/os-$(OS)/util: scripts_basic FORCE +$(ARCH_DIR)/os-$(OS)/util: scripts_basic $(ARCH_DIR)/user-offsets.h FORCE $(Q)$(MAKE) $(build)=$@ export SUBARCH USER_CFLAGS OS diff --git a/arch/um/Makefile-i386 b/arch/um/Makefile-i386 index 29e182d5a83..a777e57dbf8 100644 --- a/arch/um/Makefile-i386 +++ b/arch/um/Makefile-i386 @@ -1,4 +1,4 @@ -SUBARCH_CORE := arch/um/sys-i386/ arch/i386/crypto/ +core-y += arch/um/sys-i386/ arch/i386/crypto/ TOP_ADDR := $(CONFIG_TOP_ADDR) @@ -8,21 +8,33 @@ ifeq ($(CONFIG_MODE_SKAS),y) endif endif -CFLAGS += -U__$(SUBARCH)__ -U$(SUBARCH) -ARCH_USER_CFLAGS := +LDFLAGS += -m elf_i386 +ELF_ARCH := $(SUBARCH) +ELF_FORMAT := elf32-$(SUBARCH) +OBJCOPYFLAGS := -O binary -R .note -R .comment -S + +ifeq ("$(origin SUBARCH)", "command line") +ifneq ("$(shell uname -m | sed -e s/i.86/i386/)", "$(SUBARCH)") +CFLAGS += $(call cc-option,-m32) +USER_CFLAGS += $(call cc-option,-m32) +HOSTCFLAGS += $(call cc-option,-m32) +HOSTLDFLAGS += $(call cc-option,-m32) +AFLAGS += $(call cc-option,-m32) +LINK-y += $(call cc-option,-m32) +UML_OBJCOPYFLAGS += -F $(ELF_FORMAT) + +export LDFLAGS HOSTCFLAGS HOSTLDFLAGS UML_OBJCOPYFLAGS +endif +endif + +CFLAGS += -U__$(SUBARCH)__ -U$(SUBARCH) $(STUB_CFLAGS) ifneq ($(CONFIG_GPROF),y) ARCH_CFLAGS += -DUM_FASTCALL endif -ELF_ARCH := $(SUBARCH) -ELF_FORMAT := elf32-$(SUBARCH) - -OBJCOPYFLAGS := -O binary -R .note -R .comment -S - SYS_UTIL_DIR := $(ARCH_DIR)/sys-i386/util - -SYS_HEADERS := $(SYS_DIR)/sc.h $(SYS_DIR)/thread.h +SYS_HEADERS := $(SYS_DIR)/sc.h $(SYS_DIR)/thread.h prepare: $(SYS_HEADERS) diff --git a/arch/um/Makefile-x86_64 b/arch/um/Makefile-x86_64 index 32144562c27..aa2f7174ebc 100644 --- a/arch/um/Makefile-x86_64 +++ b/arch/um/Makefile-x86_64 @@ -1,11 +1,13 @@ # Copyright 2003 - 2004 Pathscale, Inc # Released under the GPL -SUBARCH_LIBS := arch/um/sys-x86_64/ +libs-y += arch/um/sys-x86_64/ START := 0x60000000 -CFLAGS += -U__$(SUBARCH)__ -fno-builtin -ARCH_USER_CFLAGS := -D__x86_64__ +#We #undef __x86_64__ for kernelspace, not for userspace where +#it's needed for headers to work! +CFLAGS += -U__$(SUBARCH)__ -fno-builtin $(STUB_CFLAGS) +USER_CFLAGS += -fno-builtin ELF_ARCH := i386:x86-64 ELF_FORMAT := elf64-x86-64 diff --git a/arch/um/defconfig b/arch/um/defconfig index 4067c3aa5b6..80d30d19d75 100644 --- a/arch/um/defconfig +++ b/arch/um/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.12-rc3-skas3-v9-pre2 -# Sun Apr 24 19:46:10 2005 +# Linux kernel version: 2.6.12-rc6-mm1 +# Tue Jun 14 18:22:21 2005 # CONFIG_GENERIC_HARDIRQS=y CONFIG_UML=y @@ -13,23 +13,32 @@ CONFIG_GENERIC_CALIBRATE_DELAY=y # # UML-specific options # -CONFIG_MODE_TT=y +# CONFIG_MODE_TT is not set +# CONFIG_STATIC_LINK is not set CONFIG_MODE_SKAS=y CONFIG_UML_X86=y # CONFIG_64BIT is not set CONFIG_TOP_ADDR=0xc0000000 # CONFIG_3_LEVEL_PGTABLES is not set +CONFIG_STUB_CODE=0xbfffe000 +CONFIG_STUB_DATA=0xbffff000 +CONFIG_STUB_START=0xbfffe000 CONFIG_ARCH_HAS_SC_SIGNALS=y CONFIG_ARCH_REUSE_HOST_VSYSCALL_AREA=y -CONFIG_LD_SCRIPT_STATIC=y +CONFIG_SELECT_MEMORY_MODEL=y +CONFIG_FLATMEM_MANUAL=y +# CONFIG_DISCONTIGMEM_MANUAL is not set +# CONFIG_SPARSEMEM_MANUAL is not set +CONFIG_FLATMEM=y +CONFIG_FLAT_NODE_MEM_MAP=y +CONFIG_LD_SCRIPT_DYN=y CONFIG_NET=y CONFIG_BINFMT_ELF=y CONFIG_BINFMT_MISC=m -CONFIG_HOSTFS=y +# CONFIG_HOSTFS is not set CONFIG_MCONSOLE=y # CONFIG_MAGIC_SYSRQ is not set # CONFIG_HOST_2G_2G is not set -# CONFIG_SMP is not set CONFIG_NEST_LEVEL=0 CONFIG_KERNEL_HALF_GIGS=1 # CONFIG_HIGHMEM is not set @@ -63,6 +72,8 @@ CONFIG_IKCONFIG_PROC=y CONFIG_KALLSYMS=y # CONFIG_KALLSYMS_ALL is not set CONFIG_KALLSYMS_EXTRA_PASS=y +CONFIG_PRINTK=y +CONFIG_BUG=y CONFIG_BASE_FULL=y CONFIG_FUTEX=y CONFIG_EPOLL=y @@ -81,6 +92,7 @@ CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y # CONFIG_MODULE_FORCE_UNLOAD is not set CONFIG_OBSOLETE_MODPARM=y +# CONFIG_MODVERSIONS is not set # CONFIG_MODULE_SRCVERSION_ALL is not set CONFIG_KMOD=y @@ -115,6 +127,7 @@ CONFIG_UML_SOUND=m CONFIG_SOUND=m CONFIG_HOSTAUDIO=m CONFIG_UML_RANDOM=y +# CONFIG_MMAPPER is not set # # Block devices @@ -176,6 +189,17 @@ CONFIG_INET=y # CONFIG_INET_TUNNEL is not set CONFIG_IP_TCPDIAG=y # CONFIG_IP_TCPDIAG_IPV6 is not set + +# +# TCP congestion control +# +CONFIG_TCP_CONG_BIC=y +CONFIG_TCP_CONG_WESTWOOD=y +CONFIG_TCP_CONG_HTCP=y +# CONFIG_TCP_CONG_HSTCP is not set +# CONFIG_TCP_CONG_HYBLA is not set +# CONFIG_TCP_CONG_VEGAS is not set +# CONFIG_TCP_CONG_SCALABLE is not set # CONFIG_IPV6 is not set # CONFIG_NETFILTER is not set @@ -206,11 +230,15 @@ CONFIG_IP_TCPDIAG=y # Network testing # # CONFIG_NET_PKTGEN is not set +# CONFIG_KGDBOE is not set # CONFIG_NETPOLL is not set +# CONFIG_NETPOLL_RX is not set +# CONFIG_NETPOLL_TRAP is not set # CONFIG_NET_POLL_CONTROLLER is not set # CONFIG_HAMRADIO is not set # CONFIG_IRDA is not set # CONFIG_BT is not set +# CONFIG_IEEE80211 is not set CONFIG_DUMMY=m # CONFIG_BONDING is not set # CONFIG_EQUALIZER is not set @@ -227,6 +255,7 @@ CONFIG_PPP=m # CONFIG_PPP_SYNC_TTY is not set # CONFIG_PPP_DEFLATE is not set # CONFIG_PPP_BSDCOMP is not set +# CONFIG_PPP_MPPE is not set # CONFIG_PPPOE is not set CONFIG_SLIP=m # CONFIG_SLIP_COMPRESSED is not set @@ -240,10 +269,12 @@ CONFIG_SLIP=m # CONFIG_EXT2_FS=y # CONFIG_EXT2_FS_XATTR is not set +# CONFIG_EXT2_FS_XIP is not set CONFIG_EXT3_FS=y # CONFIG_EXT3_FS_XATTR is not set CONFIG_JBD=y # CONFIG_JBD_DEBUG is not set +# CONFIG_REISER4_FS is not set CONFIG_REISERFS_FS=y # CONFIG_REISERFS_CHECK is not set # CONFIG_REISERFS_PROC_INFO is not set @@ -256,6 +287,7 @@ CONFIG_REISERFS_FS=y # CONFIG_XFS_FS is not set # CONFIG_MINIX_FS is not set # CONFIG_ROMFS_FS is not set +CONFIG_INOTIFY=y CONFIG_QUOTA=y # CONFIG_QFMT_V1 is not set # CONFIG_QFMT_V2 is not set @@ -265,6 +297,12 @@ CONFIG_AUTOFS_FS=m CONFIG_AUTOFS4_FS=m # +# Caches +# +# CONFIG_FSCACHE is not set +# CONFIG_FUSE_FS is not set + +# # CD-ROM/DVD Filesystems # CONFIG_ISO9660_FS=m @@ -291,6 +329,8 @@ CONFIG_TMPFS=y # CONFIG_TMPFS_XATTR is not set # CONFIG_HUGETLB_PAGE is not set CONFIG_RAMFS=y +# CONFIG_CONFIGFS_FS is not set +# CONFIG_RELAYFS_FS is not set # # Miscellaneous filesystems @@ -319,6 +359,7 @@ CONFIG_RAMFS=y # CONFIG_NCP_FS is not set # CONFIG_CODA_FS is not set # CONFIG_AFS_FS is not set +# CONFIG_9P_FS is not set # # Partition Types @@ -404,14 +445,15 @@ CONFIG_CRC32=m # CONFIG_PRINTK_TIME is not set CONFIG_DEBUG_KERNEL=y CONFIG_LOG_BUF_SHIFT=14 +CONFIG_DETECT_SOFTLOCKUP=y # CONFIG_SCHEDSTATS is not set -# CONFIG_DEBUG_SLAB is not set +CONFIG_DEBUG_SLAB=y # CONFIG_DEBUG_SPINLOCK is not set # CONFIG_DEBUG_SPINLOCK_SLEEP is not set # CONFIG_DEBUG_KOBJECT is not set CONFIG_DEBUG_INFO=y # CONFIG_DEBUG_FS is not set CONFIG_FRAME_POINTER=y -CONFIG_PT_PROXY=y +# CONFIG_GPROF is not set # CONFIG_GCOV is not set # CONFIG_SYSCALL_DEBUG is not set diff --git a/arch/um/drivers/Makefile b/arch/um/drivers/Makefile index b2de9916c32..de17d4c6e02 100644 --- a/arch/um/drivers/Makefile +++ b/arch/um/drivers/Makefile @@ -10,7 +10,6 @@ slip-objs := slip_kern.o slip_user.o slirp-objs := slirp_kern.o slirp_user.o daemon-objs := daemon_kern.o daemon_user.o mcast-objs := mcast_kern.o mcast_user.o -#pcap-objs := pcap_kern.o pcap_user.o $(PCAP) net-objs := net_kern.o net_user.o mconsole-objs := mconsole_kern.o mconsole_user.o hostaudio-objs := hostaudio_kern.o @@ -18,6 +17,19 @@ ubd-objs := ubd_kern.o ubd_user.o port-objs := port_kern.o port_user.o harddog-objs := harddog_kern.o harddog_user.o +LDFLAGS_pcap.o := -r $(shell $(CC) $(CFLAGS) -print-file-name=libpcap.a) + +targets := pcap_kern.o pcap_user.o + +$(obj)/pcap.o: $(obj)/pcap_kern.o $(obj)/pcap_user.o + $(LD) -r -dp -o $@ $^ $(LDFLAGS) $(LDFLAGS_pcap.o) +#XXX: The call below does not work because the flags are added before the +# object name, so nothing from the library gets linked. +#$(call if_changed,ld) + +# When the above is fixed, don't forget to add this too! +#targets += $(obj)/pcap.o + obj-y := stdio_console.o fd.o chan_kern.o chan_user.o line.o obj-$(CONFIG_SSL) += ssl.o obj-$(CONFIG_STDERR_CONSOLE) += stderr_console.o @@ -26,7 +38,7 @@ obj-$(CONFIG_UML_NET_SLIP) += slip.o slip_common.o obj-$(CONFIG_UML_NET_SLIRP) += slirp.o slip_common.o obj-$(CONFIG_UML_NET_DAEMON) += daemon.o obj-$(CONFIG_UML_NET_MCAST) += mcast.o -#obj-$(CONFIG_UML_NET_PCAP) += pcap.o $(PCAP) +obj-$(CONFIG_UML_NET_PCAP) += pcap.o obj-$(CONFIG_UML_NET) += net.o obj-$(CONFIG_MCONSOLE) += mconsole.o obj-$(CONFIG_MMAPPER) += mmapper_kern.o @@ -41,6 +53,7 @@ obj-$(CONFIG_UML_WATCHDOG) += harddog.o obj-$(CONFIG_BLK_DEV_COW_COMMON) += cow_user.o obj-$(CONFIG_UML_RANDOM) += random.o -USER_OBJS := fd.o null.o pty.o tty.o xterm.o slip_common.o +# pcap_user.o must be added explicitly. +USER_OBJS := fd.o null.o pty.o tty.o xterm.o slip_common.o pcap_user.o include arch/um/scripts/Makefile.rules diff --git a/arch/um/drivers/cow.h b/arch/um/drivers/cow.h index 4fcbe8b1b77..4fcf3a8d13f 100644 --- a/arch/um/drivers/cow.h +++ b/arch/um/drivers/cow.h @@ -3,10 +3,10 @@ #include <asm/types.h> -#if __BYTE_ORDER == __BIG_ENDIAN +#if defined(__BIG_ENDIAN) # define ntohll(x) (x) # define htonll(x) (x) -#elif __BYTE_ORDER == __LITTLE_ENDIAN +#elif defined(__LITTLE_ENDIAN) # define ntohll(x) bswap_64(x) # define htonll(x) bswap_64(x) #else diff --git a/arch/um/drivers/daemon_user.c b/arch/um/drivers/daemon_user.c index cf15b4a8b51..c1b03f7c1da 100644 --- a/arch/um/drivers/daemon_user.c +++ b/arch/um/drivers/daemon_user.c @@ -157,9 +157,9 @@ static void daemon_remove(void *data) os_close_file(pri->fd); os_close_file(pri->control); - if(pri->data_addr != NULL) kfree(pri->data_addr); - if(pri->ctl_addr != NULL) kfree(pri->ctl_addr); - if(pri->local_addr != NULL) kfree(pri->local_addr); + kfree(pri->data_addr); + kfree(pri->ctl_addr); + kfree(pri->local_addr); } int daemon_user_write(int fd, void *buf, int len, struct daemon_data *pri) diff --git a/arch/um/drivers/hostaudio_kern.c b/arch/um/drivers/hostaudio_kern.c index d5742783e19..59602b81b24 100644 --- a/arch/um/drivers/hostaudio_kern.c +++ b/arch/um/drivers/hostaudio_kern.c @@ -57,10 +57,10 @@ __uml_setup("mixer=", set_mixer, "mixer=<mixer device>\n" MIXER_HELP); #else /*MODULE*/ -MODULE_PARM(dsp, "s"); +module_param(dsp, charp, 0644); MODULE_PARM_DESC(dsp, DSP_HELP); -MODULE_PARM(mixer, "s"); +module_param(mixer, charp, 0644); MODULE_PARM_DESC(mixer, MIXER_HELP); #endif diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c index 025d3be8aca..e0fdffa2d54 100644 --- a/arch/um/drivers/line.c +++ b/arch/um/drivers/line.c @@ -406,14 +406,12 @@ void line_disable(struct tty_struct *tty, int current_irq) if(line->driver->read_irq == current_irq) free_irq_later(line->driver->read_irq, tty); else { - free_irq_by_irq_and_dev(line->driver->read_irq, tty); free_irq(line->driver->read_irq, tty); } if(line->driver->write_irq == current_irq) free_irq_later(line->driver->write_irq, tty); else { - free_irq_by_irq_and_dev(line->driver->write_irq, tty); free_irq(line->driver->write_irq, tty); } @@ -604,11 +602,26 @@ int line_get_config(char *name, struct line *lines, unsigned int num, char *str, return n; } -int line_remove(struct line *lines, unsigned int num, char *str) +int line_id(char **str, int *start_out, int *end_out) +{ + char *end; + int n; + + n = simple_strtoul(*str, &end, 0); + if((*end != '\0') || (end == *str)) + return -1; + + *str = end; + *start_out = n; + *end_out = n; + return n; +} + +int line_remove(struct line *lines, unsigned int num, int n) { char config[sizeof("conxxxx=none\0")]; - sprintf(config, "%s=none", str); + sprintf(config, "%d=none", n); return !line_setup(lines, num, config, 0); } @@ -650,11 +663,15 @@ struct tty_driver *line_register_devfs(struct lines *set, return driver; } +static spinlock_t winch_handler_lock; +LIST_HEAD(winch_handlers); + void lines_init(struct line *lines, int nlines) { struct line *line; int i; + spin_lock_init(&winch_handler_lock); for(i = 0; i < nlines; i++){ line = &lines[i]; INIT_LIST_HEAD(&line->chan_list); @@ -711,31 +728,30 @@ irqreturn_t winch_interrupt(int irq, void *data, struct pt_regs *unused) return IRQ_HANDLED; } -DECLARE_MUTEX(winch_handler_sem); -LIST_HEAD(winch_handlers); - void register_winch_irq(int fd, int tty_fd, int pid, struct tty_struct *tty) { struct winch *winch; - down(&winch_handler_sem); winch = kmalloc(sizeof(*winch), GFP_KERNEL); if (winch == NULL) { printk("register_winch_irq - kmalloc failed\n"); - goto out; + return; } + *winch = ((struct winch) { .list = LIST_HEAD_INIT(winch->list), .fd = fd, .tty_fd = tty_fd, .pid = pid, .tty = tty }); + + spin_lock(&winch_handler_lock); list_add(&winch->list, &winch_handlers); + spin_unlock(&winch_handler_lock); + if(um_request_irq(WINCH_IRQ, fd, IRQ_READ, winch_interrupt, SA_INTERRUPT | SA_SHIRQ | SA_SAMPLE_RANDOM, "winch", winch) < 0) printk("register_winch_irq - failed to register IRQ\n"); - out: - up(&winch_handler_sem); } static void unregister_winch(struct tty_struct *tty) @@ -743,7 +759,7 @@ static void unregister_winch(struct tty_struct *tty) struct list_head *ele; struct winch *winch, *found = NULL; - down(&winch_handler_sem); + spin_lock(&winch_handler_lock); list_for_each(ele, &winch_handlers){ winch = list_entry(ele, struct winch, list); if(winch->tty == tty){ @@ -751,21 +767,25 @@ static void unregister_winch(struct tty_struct *tty) break; } } - if(found == NULL) - goto out; + goto err; + + list_del(&winch->list); + spin_unlock(&winch_handler_lock); if(winch->pid != -1) os_kill_process(winch->pid, 1); - free_irq_by_irq_and_dev(WINCH_IRQ, winch); free_irq(WINCH_IRQ, winch); - list_del(&winch->list); kfree(winch); - out: - up(&winch_handler_sem); + + return; +err: + spin_unlock(&winch_handler_lock); } +/* XXX: No lock as it's an exitcall... is this valid? Depending on cleanup + * order... are we sure that nothing else is done on the list? */ static void winch_cleanup(void) { struct list_head *ele; @@ -774,6 +794,9 @@ static void winch_cleanup(void) list_for_each(ele, &winch_handlers){ winch = list_entry(ele, struct winch, list); if(winch->fd != -1){ + /* Why is this different from the above free_irq(), + * which deactivates SIGIO? This searches the FD + * somewhere else and removes it from the list... */ deactivate_fd(winch->fd, WINCH_IRQ); os_close_file(winch->fd); } diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c index d7c7adcc0a6..c190c241419 100644 --- a/arch/um/drivers/mconsole_kern.c +++ b/arch/um/drivers/mconsole_kern.c @@ -419,8 +419,9 @@ void mconsole_config(struct mc_request *req) void mconsole_remove(struct mc_request *req) { struct mc_device *dev; - char *ptr = req->request.data; - int err; + char *ptr = req->request.data, *err_msg = ""; + char error[256]; + int err, start, end, n; ptr += strlen("remove"); while(isspace(*ptr)) ptr++; @@ -429,8 +430,35 @@ void mconsole_remove(struct mc_request *req) mconsole_reply(req, "Bad remove option", 1, 0); return; } - err = (*dev->remove)(&ptr[strlen(dev->name)]); - mconsole_reply(req, "", err, 0); + + ptr = &ptr[strlen(dev->name)]; + + err = 1; + n = (*dev->id)(&ptr, &start, &end); + if(n < 0){ + err_msg = "Couldn't parse device number"; + goto out; + } + else if((n < start) || (n > end)){ + sprintf(error, "Invalid device number - must be between " + "%d and %d", start, end); + err_msg = error; + goto out; + } + + err = (*dev->remove)(n); + switch(err){ + case -ENODEV: + err_msg = "Device doesn't exist"; + break; + case -EBUSY: + err_msg = "Device is currently open"; + break; + default: + break; + } + out: + mconsole_reply(req, err_msg, err, 0); } #ifdef CONFIG_MAGIC_SYSRQ @@ -529,7 +557,7 @@ static int create_proc_mconsole(void) ent = create_proc_entry("mconsole", S_IFREG | 0200, NULL); if(ent == NULL){ - printk("create_proc_mconsole : create_proc_entry failed\n"); + printk(KERN_INFO "create_proc_mconsole : create_proc_entry failed\n"); return(0); } diff --git a/arch/um/drivers/net_kern.c b/arch/um/drivers/net_kern.c index 4eeaf88c1e9..1495007bf6c 100644 --- a/arch/um/drivers/net_kern.c +++ b/arch/um/drivers/net_kern.c @@ -146,7 +146,6 @@ static int uml_net_close(struct net_device *dev) netif_stop_queue(dev); spin_lock(&lp->lock); - free_irq_by_irq_and_dev(dev->irq, dev); free_irq(dev->irq, dev); if(lp->close != NULL) (*lp->close)(lp->fd, &lp->user); @@ -613,25 +612,35 @@ static int net_config(char *str) return(err); } -static int net_remove(char *str) +static int net_id(char **str, int *start_out, int *end_out) +{ + char *end; + int n; + + n = simple_strtoul(*str, &end, 0); + if((*end != '\0') || (end == *str)) + return -1; + + *start_out = n; + *end_out = n; + *str = end; + return n; +} + +static int net_remove(int n) { struct uml_net *device; struct net_device *dev; struct uml_net_private *lp; - char *end; - int n; - - n = simple_strtoul(str, &end, 0); - if((*end != '\0') || (end == str)) - return(-1); device = find_device(n); if(device == NULL) - return(0); + return -ENODEV; dev = device->dev; lp = dev->priv; - if(lp->fd > 0) return(-1); + if(lp->fd > 0) + return -EBUSY; if(lp->remove != NULL) (*lp->remove)(&lp->user); unregister_netdev(dev); platform_device_unregister(&device->pdev); @@ -639,13 +648,14 @@ static int net_remove(char *str) list_del(&device->list); kfree(device); free_netdev(dev); - return(0); + return 0; } static struct mc_device net_mc = { .name = "eth", .config = net_config, .get_config = NULL, + .id = net_id, .remove = net_remove, }; diff --git a/arch/um/drivers/port_kern.c b/arch/um/drivers/port_kern.c index b5ee07472f7..c41efd207fc 100644 --- a/arch/um/drivers/port_kern.c +++ b/arch/um/drivers/port_kern.c @@ -257,7 +257,6 @@ int port_wait(void *data) * connection. Then we loop here throwing out failed * connections until a good one is found. */ - free_irq_by_irq_and_dev(TELNETD_IRQ, conn); free_irq(TELNETD_IRQ, conn); if(conn->fd >= 0) break; diff --git a/arch/um/drivers/ssl.c b/arch/um/drivers/ssl.c index b32a77010fb..62e04ecfada 100644 --- a/arch/um/drivers/ssl.c +++ b/arch/um/drivers/ssl.c @@ -49,7 +49,7 @@ static struct chan_opts opts = { static int ssl_config(char *str); static int ssl_get_config(char *dev, char *str, int size, char **error_out); -static int ssl_remove(char *str); +static int ssl_remove(int n); static struct line_driver driver = { .name = "UML serial line", @@ -69,6 +69,7 @@ static struct line_driver driver = { .name = "ssl", .config = ssl_config, .get_config = ssl_get_config, + .id = line_id, .remove = ssl_remove, }, }; @@ -94,10 +95,10 @@ static int ssl_get_config(char *dev, char *str, int size, char **error_out) str, size, error_out)); } -static int ssl_remove(char *str) +static int ssl_remove(int n) { - return(line_remove(serial_lines, - sizeof(serial_lines)/sizeof(serial_lines[0]), str)); + return line_remove(serial_lines, + sizeof(serial_lines)/sizeof(serial_lines[0]), n); } int ssl_open(struct tty_struct *tty, struct file *filp) diff --git a/arch/um/drivers/stdio_console.c b/arch/um/drivers/stdio_console.c index afbe1e71ed8..005aa6333b6 100644 --- a/arch/um/drivers/stdio_console.c +++ b/arch/um/drivers/stdio_console.c @@ -55,7 +55,7 @@ static struct chan_opts opts = { static int con_config(char *str); static int con_get_config(char *dev, char *str, int size, char **error_out); -static int con_remove(char *str); +static int con_remove(int n); static struct line_driver driver = { .name = "UML console", @@ -75,6 +75,7 @@ static struct line_driver driver = { .name = "con", .config = con_config, .get_config = con_get_config, + .id = line_id, .remove = con_remove, }, }; @@ -99,9 +100,9 @@ static int con_get_config(char *dev, char *str, int size, char **error_out) size, error_out)); } -static int con_remove(char *str) +static int con_remove(int n) { - return(line_remove(vts, sizeof(vts)/sizeof(vts[0]), str)); + return line_remove(vts, sizeof(vts)/sizeof(vts[0]), n); } static int con_open(struct tty_struct *tty, struct file *filp) diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index 2a7f6892c55..344b24d09a7 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -754,24 +754,34 @@ static int ubd_get_config(char *name, char *str, int size, char **error_out) return(len); } -static int ubd_remove(char *str) +static int ubd_id(char **str, int *start_out, int *end_out) +{ + int n; + + n = parse_unit(str); + *start_out = 0; + *end_out = MAX_DEV - 1; + return n; +} + +static int ubd_remove(int n) { struct ubd *dev; - int n, err = -ENODEV; + int err = -ENODEV; - n = parse_unit(&str); + spin_lock(&ubd_lock); - if((n < 0) || (n >= MAX_DEV)) - return(err); + if(ubd_gendisk[n] == NULL) + goto out; dev = &ubd_dev[n]; - if(dev->count > 0) - return(-EBUSY); /* you cannot remove a open disk */ - err = 0; - spin_lock(&ubd_lock); + if(dev->file == NULL) + goto out; - if(ubd_gendisk[n] == NULL) + /* you cannot remove a open disk */ + err = -EBUSY; + if(dev->count > 0) goto out; del_gendisk(ubd_gendisk[n]); @@ -787,15 +797,16 @@ static int ubd_remove(char *str) platform_device_unregister(&dev->pdev); *dev = ((struct ubd) DEFAULT_UBD); err = 0; - out: - spin_unlock(&ubd_lock); - return(err); +out: + spin_unlock(&ubd_lock); + return err; } static struct mc_device ubd_mc = { .name = "ubd", .config = ubd_config, .get_config = ubd_get_config, + .id = ubd_id, .remove = ubd_remove, }; diff --git a/arch/um/drivers/xterm_kern.c b/arch/um/drivers/xterm_kern.c index a4fdf3584ad..d269a80f4b0 100644 --- a/arch/um/drivers/xterm_kern.c +++ b/arch/um/drivers/xterm_kern.c @@ -69,7 +69,6 @@ int xterm_fd(int socket, int *pid_out) * isn't set) this will hang... */ wait_for_completion(&data->ready); - free_irq_by_irq_and_dev(XTERM_IRQ, data); free_irq(XTERM_IRQ, data); ret = data->new_fd; diff --git a/arch/um/include/line.h b/arch/um/include/line.h index 4c5e92c04cc..5323d22a6ca 100644 --- a/arch/um/include/line.h +++ b/arch/um/include/line.h @@ -101,7 +101,8 @@ extern void lines_init(struct line *lines, int nlines); extern void close_lines(struct line *lines, int nlines); extern int line_config(struct line *lines, unsigned int sizeof_lines, char *str); -extern int line_remove(struct line *lines, unsigned int sizeof_lines, char *str); +extern int line_id(char **str, int *start_out, int *end_out); +extern int line_remove(struct line *lines, unsigned int sizeof_lines, int n); extern int line_get_config(char *dev, struct line *lines, unsigned int sizeof_lines, char *str, int size, char **error_out); diff --git a/arch/um/include/mconsole_kern.h b/arch/um/include/mconsole_kern.h index 61c274fcee5..d86ee14260c 100644 --- a/arch/um/include/mconsole_kern.h +++ b/arch/um/include/mconsole_kern.h @@ -20,7 +20,8 @@ struct mc_device { char *name; int (*config)(char *); int (*get_config)(char *, char *, int, char **); - int (*remove)(char *); + int (*id)(char **, int *, int *); + int (*remove)(int); }; #define CONFIG_CHUNK(str, size, current, chunk, end) \ diff --git a/arch/um/include/mem.h b/arch/um/include/mem.h index 10c46c38949..99d3ad4a03e 100644 --- a/arch/um/include/mem.h +++ b/arch/um/include/mem.h @@ -13,6 +13,7 @@ extern int physmem_subst_mapping(void *virt, int fd, __u64 offset, int w); extern int is_remapped(void *virt); extern int physmem_remove_mapping(void *virt); extern void physmem_forget_descriptor(int fd); +extern unsigned long to_phys(void *virt); #endif diff --git a/arch/um/include/registers.h b/arch/um/include/registers.h index 8744abb5224..0a35e6d0baa 100644 --- a/arch/um/include/registers.h +++ b/arch/um/include/registers.h @@ -14,6 +14,7 @@ extern int restore_fp_registers(int pid, unsigned long *fp_regs); extern void save_registers(int pid, union uml_pt_regs *regs); extern void restore_registers(int pid, union uml_pt_regs *regs); extern void init_registers(int pid); +extern void get_safe_registers(unsigned long * regs); #endif diff --git a/arch/um/include/sysdep-i386/ptrace_user.h b/arch/um/include/sysdep-i386/ptrace_user.h index eca8066e7a4..899aa4b2a78 100644 --- a/arch/um/include/sysdep-i386/ptrace_user.h +++ b/arch/um/include/sysdep-i386/ptrace_user.h @@ -20,11 +20,24 @@ #define PT_SYSCALL_ARG3_OFFSET PT_OFFSET(EDX) #define PT_SYSCALL_ARG4_OFFSET PT_OFFSET(ESI) #define PT_SYSCALL_ARG5_OFFSET PT_OFFSET(EDI) +#define PT_SYSCALL_ARG6_OFFSET PT_OFFSET(EBP) #define PT_SYSCALL_RET_OFFSET PT_OFFSET(EAX) +#define REGS_SYSCALL_NR EAX /* This is used before a system call */ +#define REGS_SYSCALL_ARG1 EBX +#define REGS_SYSCALL_ARG2 ECX +#define REGS_SYSCALL_ARG3 EDX +#define REGS_SYSCALL_ARG4 ESI +#define REGS_SYSCALL_ARG5 EDI +#define REGS_SYSCALL_ARG6 EBP + +#define REGS_IP_INDEX EIP +#define REGS_SP_INDEX UESP + #define PT_IP_OFFSET PT_OFFSET(EIP) #define PT_IP(regs) ((regs)[EIP]) +#define PT_SP_OFFSET PT_OFFSET(UESP) #define PT_SP(regs) ((regs)[UESP]) #ifndef FRAME_SIZE diff --git a/arch/um/include/sysdep-i386/stub.h b/arch/um/include/sysdep-i386/stub.h new file mode 100644 index 00000000000..d3699fe1c61 --- /dev/null +++ b/arch/um/include/sysdep-i386/stub.h @@ -0,0 +1,65 @@ +/* + * Copyright (C) 2004 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#ifndef __SYSDEP_STUB_H +#define __SYSDEP_STUB_H + +#include <asm/ptrace.h> +#include <asm/unistd.h> + +extern void stub_segv_handler(int sig); +extern void stub_clone_handler(void); + +#define STUB_SYSCALL_RET EAX +#define STUB_MMAP_NR __NR_mmap2 +#define MMAP_OFFSET(o) ((o) >> PAGE_SHIFT) + +static inline long stub_syscall2(long syscall, long arg1, long arg2) +{ + long ret; + + __asm__("movl %0, %%ecx; " : : "g" (arg2) : "%ecx"); + __asm__("movl %0, %%ebx; " : : "g" (arg1) : "%ebx"); + __asm__("movl %0, %%eax; " : : "g" (syscall) : "%eax"); + __asm__("int $0x80;" : : : "%eax"); + __asm__ __volatile__("movl %%eax, %0; " : "=g" (ret) :); + return(ret); +} + +static inline long stub_syscall3(long syscall, long arg1, long arg2, long arg3) +{ + __asm__("movl %0, %%edx; " : : "g" (arg3) : "%edx"); + return(stub_syscall2(syscall, arg1, arg2)); +} + +static inline long stub_syscall4(long syscall, long arg1, long arg2, long arg3, + long arg4) +{ + __asm__("movl %0, %%esi; " : : "g" (arg4) : "%esi"); + return(stub_syscall3(syscall, arg1, arg2, arg3)); +} + +static inline long stub_syscall6(long syscall, long arg1, long arg2, long arg3, + long arg4, long arg5, long arg6) +{ + long ret; + __asm__("movl %0, %%eax; " : : "g" (syscall) : "%eax"); + __asm__("movl %0, %%ebx; " : : "g" (arg1) : "%ebx"); + __asm__("movl %0, %%ecx; " : : "g" (arg2) : "%ecx"); + __asm__("movl %0, %%edx; " : : "g" (arg3) : "%edx"); + __asm__("movl %0, %%esi; " : : "g" (arg4) : "%esi"); + __asm__("movl %0, %%edi; " : : "g" (arg5) : "%edi"); + __asm__ __volatile__("pushl %%ebp ; movl %1, %%ebp; " + "int $0x80; popl %%ebp ; " + "movl %%eax, %0; " : "=g" (ret) : "g" (arg6) : "%eax"); + return(ret); +} + +static inline void trap_myself(void) +{ + __asm("int3"); +} + +#endif diff --git a/arch/um/include/sysdep-x86_64/ptrace_user.h b/arch/um/include/sysdep-x86_64/ptrace_user.h index 31729973fb1..128faf02736 100644 --- a/arch/um/include/sysdep-x86_64/ptrace_user.h +++ b/arch/um/include/sysdep-x86_64/ptrace_user.h @@ -55,6 +55,20 @@ #define PTRACE_OLDSETOPTIONS 21 #endif +/* These are before the system call, so the the system call number is RAX + * rather than ORIG_RAX, and arg4 is R10 rather than RCX + */ +#define REGS_SYSCALL_NR PT_INDEX(RAX) +#define REGS_SYSCALL_ARG1 PT_INDEX(RDI) +#define REGS_SYSCALL_ARG2 PT_INDEX(RSI) +#define REGS_SYSCALL_ARG3 PT_INDEX(RDX) +#define REGS_SYSCALL_ARG4 PT_INDEX(R10) +#define REGS_SYSCALL_ARG5 PT_INDEX(R8) +#define REGS_SYSCALL_ARG6 PT_INDEX(R9) + +#define REGS_IP_INDEX PT_INDEX(RIP) +#define REGS_SP_INDEX PT_INDEX(RSP) + #endif /* diff --git a/arch/um/include/sysdep-x86_64/stub.h b/arch/um/include/sysdep-x86_64/stub.h new file mode 100644 index 00000000000..f599058d826 --- /dev/null +++ b/arch/um/include/sysdep-x86_64/stub.h @@ -0,0 +1,58 @@ +/* + * Copyright (C) 2004 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#ifndef __SYSDEP_STUB_H +#define __SYSDEP_STUB_H + +#include <asm/ptrace.h> +#include <asm/unistd.h> +#include <sysdep/ptrace_user.h> + +extern void stub_segv_handler(int sig); +extern void stub_clone_handler(void); + +#define STUB_SYSCALL_RET PT_INDEX(RAX) +#define STUB_MMAP_NR __NR_mmap +#define MMAP_OFFSET(o) (o) + +static inline long stub_syscall2(long syscall, long arg1, long arg2) +{ + long ret; + + __asm__("movq %0, %%rsi; " : : "g" (arg2) : "%rsi"); + __asm__("movq %0, %%rdi; " : : "g" (arg1) : "%rdi"); + __asm__("movq %0, %%rax; " : : "g" (syscall) : "%rax"); + __asm__("syscall;" : : : "%rax", "%r11", "%rcx"); + __asm__ __volatile__("movq %%rax, %0; " : "=g" (ret) :); + return(ret); +} + +static inline long stub_syscall3(long syscall, long arg1, long arg2, long arg3) +{ + __asm__("movq %0, %%rdx; " : : "g" (arg3) : "%rdx"); + return(stub_syscall2(syscall, arg1, arg2)); +} + +static inline long stub_syscall4(long syscall, long arg1, long arg2, long arg3, + long arg4) +{ + __asm__("movq %0, %%r10; " : : "g" (arg4) : "%r10"); + return(stub_syscall3(syscall, arg1, arg2, arg3)); +} + +static inline long stub_syscall6(long syscall, long arg1, long arg2, long arg3, + long arg4, long arg5, long arg6) +{ + __asm__("movq %0, %%r9; " : : "g" (arg6) : "%r9"); + __asm__("movq %0, %%r8; " : : "g" (arg5) : "%r8"); + return(stub_syscall4(syscall, arg1, arg2, arg3, arg4)); +} + +static inline void trap_myself(void) +{ + __asm("int3"); +} + +#endif diff --git a/arch/um/include/time_user.h b/arch/um/include/time_user.h index 6793a2fcd0a..17d7ef2141f 100644 --- a/arch/um/include/time_user.h +++ b/arch/um/include/time_user.h @@ -8,11 +8,12 @@ extern void timer(void); extern void switch_timers(int to_real); -extern void set_interval(int timer_type); extern void idle_sleep(int secs); extern void enable_timer(void); +extern void prepare_timer(void * ptr); extern void disable_timer(void); extern unsigned long time_lock(void); extern void time_unlock(unsigned long); +extern void user_time_init(void); #endif diff --git a/arch/um/include/tlb.h b/arch/um/include/tlb.h index da1097285b8..c6f9628f39b 100644 --- a/arch/um/include/tlb.h +++ b/arch/um/include/tlb.h @@ -37,31 +37,25 @@ struct host_vm_op { extern void mprotect_kernel_vm(int w); extern void force_flush_all(void); extern void fix_range_common(struct mm_struct *mm, unsigned long start_addr, - unsigned long end_addr, int force, int data, - void (*do_ops)(int, struct host_vm_op *, int)); + unsigned long end_addr, int force, + void (*do_ops)(union mm_context *, + struct host_vm_op *, int)); extern int flush_tlb_kernel_range_common(unsigned long start, unsigned long end); extern int add_mmap(unsigned long virt, unsigned long phys, unsigned long len, int r, int w, int x, struct host_vm_op *ops, int index, - int last_filled, int data, - void (*do_ops)(int, struct host_vm_op *, int)); + int last_filled, union mm_context *mmu, + void (*do_ops)(union mm_context *, struct host_vm_op *, + int)); extern int add_munmap(unsigned long addr, unsigned long len, struct host_vm_op *ops, int index, int last_filled, - int data, void (*do_ops)(int, struct host_vm_op *, int)); + union mm_context *mmu, + void (*do_ops)(union mm_context *, struct host_vm_op *, + int)); extern int add_mprotect(unsigned long addr, unsigned long len, int r, int w, int x, struct host_vm_op *ops, int index, - int last_filled, int data, - void (*do_ops)(int, struct host_vm_op *, int)); + int last_filled, union mm_context *mmu, + void (*do_ops)(union mm_context *, struct host_vm_op *, + int)); #endif - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/kernel/dyn.lds.S b/arch/um/kernel/dyn.lds.S index 715b0838a68..3942a5f245d 100644 --- a/arch/um/kernel/dyn.lds.S +++ b/arch/um/kernel/dyn.lds.S @@ -67,6 +67,12 @@ SECTIONS *(.stub .text.* .gnu.linkonce.t.*) /* .gnu.warning sections are handled specially by elf32.em. */ *(.gnu.warning) + + . = ALIGN(4096); + __syscall_stub_start = .; + *(.__syscall_stub*) + __syscall_stub_end = .; + . = ALIGN(4096); } =0x90909090 .fini : { KEEP (*(.fini)) diff --git a/arch/um/kernel/exitcode.c b/arch/um/kernel/exitcode.c index 0ea87f24b36..d21ebad666b 100644 --- a/arch/um/kernel/exitcode.c +++ b/arch/um/kernel/exitcode.c @@ -48,7 +48,7 @@ static int make_proc_exitcode(void) ent = create_proc_entry("exitcode", 0600, &proc_root); if(ent == NULL){ - printk("make_proc_exitcode : Failed to register " + printk(KERN_WARNING "make_proc_exitcode : Failed to register " "/proc/exitcode\n"); return(0); } diff --git a/arch/um/kernel/helper.c b/arch/um/kernel/helper.c index 13b1f5c2f7e..f83e1e8e239 100644 --- a/arch/um/kernel/helper.c +++ b/arch/um/kernel/helper.c @@ -13,6 +13,7 @@ #include "user.h" #include "kern_util.h" #include "user_util.h" +#include "helper.h" #include "os.h" struct helper_data { @@ -149,7 +150,7 @@ int run_helper_thread(int (*proc)(void *), void *arg, unsigned int flags, return(pid); } -int helper_wait(int pid, int block) +int helper_wait(int pid) { int ret; @@ -160,14 +161,3 @@ int helper_wait(int pid, int block) } return(ret); } - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/kernel/irq.c b/arch/um/kernel/irq.c index d44fb528254..9f18061ef4c 100644 --- a/arch/um/kernel/irq.c +++ b/arch/um/kernel/irq.c @@ -124,14 +124,16 @@ void irq_unlock(unsigned long flags) spin_unlock_irqrestore(&irq_spinlock, flags); } -/* presently hw_interrupt_type must define (startup || enable) && - * disable && end */ +/* hw_interrupt_type must define (startup || enable) && + * (shutdown || disable) && end */ static void dummy(unsigned int irq) { } -static struct hw_interrupt_type SIGIO_irq_type = { +/* This is used for everything else than the timer. */ +static struct hw_interrupt_type normal_irq_type = { .typename = "SIGIO", + .release = free_irq_by_irq_and_dev, .disable = dummy, .enable = dummy, .ack = dummy, @@ -140,6 +142,7 @@ static struct hw_interrupt_type SIGIO_irq_type = { static struct hw_interrupt_type SIGVTALRM_irq_type = { .typename = "SIGVTALRM", + .release = free_irq_by_irq_and_dev, .shutdown = dummy, /* never called */ .disable = dummy, .enable = dummy, @@ -160,7 +163,7 @@ void __init init_IRQ(void) irq_desc[i].status = IRQ_DISABLED; irq_desc[i].action = NULL; irq_desc[i].depth = 1; - irq_desc[i].handler = &SIGIO_irq_type; + irq_desc[i].handler = &normal_irq_type; enable_irq(i); } } diff --git a/arch/um/kernel/irq_user.c b/arch/um/kernel/irq_user.c index b3074cbaa47..c3ccaf24f3e 100644 --- a/arch/um/kernel/irq_user.c +++ b/arch/um/kernel/irq_user.c @@ -85,8 +85,6 @@ void sigio_handler(int sig, union uml_pt_regs *regs) next = irq_fd->next; if(irq_fd->freed){ free_irq(irq_fd->irq, irq_fd->id); - free_irq_by_irq_and_dev(irq_fd->irq, - irq_fd->id); } } } diff --git a/arch/um/kernel/main.c b/arch/um/kernel/main.c index e59f5815267..1e1a87f1c51 100644 --- a/arch/um/kernel/main.c +++ b/arch/um/kernel/main.c @@ -69,7 +69,6 @@ static __init void do_uml_initcalls(void) static void last_ditch_exit(int sig) { - kmalloc_ok = 0; signal(SIGINT, SIG_DFL); signal(SIGTERM, SIG_DFL); signal(SIGHUP, SIG_DFL); diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c index c22825f13e4..5597bd39e6b 100644 --- a/arch/um/kernel/mem.c +++ b/arch/um/kernel/mem.c @@ -53,7 +53,6 @@ static void setup_highmem(unsigned long highmem_start, for(i = 0; i < highmem_len >> PAGE_SHIFT; i++){ page = &mem_map[highmem_pfn + i]; ClearPageReserved(page); - set_bit(PG_highmem, &page->flags); set_page_count(page, 1); __free_page(page); } diff --git a/arch/um/kernel/physmem.c b/arch/um/kernel/physmem.c index 420e6d51fa0..a24e3b7f4bf 100644 --- a/arch/um/kernel/physmem.c +++ b/arch/um/kernel/physmem.c @@ -353,6 +353,8 @@ void map_memory(unsigned long virt, unsigned long phys, unsigned long len, #define PFN_UP(x) (((x) + PAGE_SIZE-1) >> PAGE_SHIFT) +extern int __syscall_stub_start, __binary_start; + void setup_physmem(unsigned long start, unsigned long reserve_end, unsigned long len, unsigned long highmem) { @@ -371,6 +373,12 @@ void setup_physmem(unsigned long start, unsigned long reserve_end, exit(1); } + /* Special kludge - This page will be mapped in to userspace processes + * from physmem_fd, so it needs to be written out there. + */ + os_seek_file(physmem_fd, __pa(&__syscall_stub_start)); + os_write_file(physmem_fd, &__syscall_stub_start, PAGE_SIZE); + bootmap_size = init_bootmem(pfn, pfn + delta); free_bootmem(__pa(reserve_end) + bootmap_size, len - bootmap_size - reserve); diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c index 1b5ef3e96c7..67acd92c532 100644 --- a/arch/um/kernel/process.c +++ b/arch/um/kernel/process.c @@ -32,6 +32,7 @@ #include "uml-config.h" #include "choose-mode.h" #include "mode.h" +#include "tempfile.h" #ifdef UML_CONFIG_MODE_SKAS #include "skas.h" #include "skas_ptrace.h" @@ -130,7 +131,7 @@ int start_fork_tramp(void *thread_arg, unsigned long temp_stack, return(arg.pid); } -static int ptrace_child(void) +static int ptrace_child(void *arg) { int ret; int pid = os_getpid(), ppid = getppid(); @@ -159,16 +160,20 @@ static int ptrace_child(void) _exit(ret); } -static int start_ptraced_child(void) +static int start_ptraced_child(void **stack_out) { + void *stack; + unsigned long sp; int pid, n, status; - pid = fork(); - if(pid == 0) - ptrace_child(); - + stack = mmap(NULL, PAGE_SIZE, PROT_READ | PROT_WRITE | PROT_EXEC, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + if(stack == MAP_FAILED) + panic("check_ptrace : mmap failed, errno = %d", errno); + sp = (unsigned long) stack + PAGE_SIZE - sizeof(void *); + pid = clone(ptrace_child, (void *) sp, SIGCHLD, NULL); if(pid < 0) - panic("check_ptrace : fork failed, errno = %d", errno); + panic("check_ptrace : clone failed, errno = %d", errno); CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED)); if(n < 0) panic("check_ptrace : wait failed, errno = %d", errno); @@ -176,6 +181,7 @@ static int start_ptraced_child(void) panic("check_ptrace : expected SIGSTOP, got status = %d", status); + *stack_out = stack; return(pid); } @@ -183,12 +189,12 @@ static int start_ptraced_child(void) * just avoid using sysemu, not panic, but only if SYSEMU features are broken. * So only for SYSEMU features we test mustpanic, while normal host features * must work anyway!*/ -static int stop_ptraced_child(int pid, int exitcode, int mustexit) +static int stop_ptraced_child(int pid, void *stack, int exitcode, int mustpanic) { int status, n, ret = 0; if(ptrace(PTRACE_CONT, pid, 0, 0) < 0) - panic("stop_ptraced_child : ptrace failed, errno = %d", errno); + panic("check_ptrace : ptrace failed, errno = %d", errno); CATCH_EINTR(n = waitpid(pid, &status, 0)); if(!WIFEXITED(status) || (WEXITSTATUS(status) != exitcode)) { int exit_with = WEXITSTATUS(status); @@ -199,24 +205,40 @@ static int stop_ptraced_child(int pid, int exitcode, int mustexit) printk("check_ptrace : child exited with exitcode %d, while " "expecting %d; status 0x%x", exit_with, exitcode, status); - if (mustexit) + if (mustpanic) panic("\n"); else printk("\n"); ret = -1; } + if(munmap(stack, PAGE_SIZE) < 0) + panic("check_ptrace : munmap failed, errno = %d", errno); return ret; } static int force_sysemu_disabled = 0; +int ptrace_faultinfo = 1; +int proc_mm = 1; + +static int __init skas0_cmd_param(char *str, int* add) +{ + ptrace_faultinfo = proc_mm = 0; + return 0; +} + static int __init nosysemu_cmd_param(char *str, int* add) { force_sysemu_disabled = 1; return 0; } +__uml_setup("skas0", skas0_cmd_param, + "skas0\n" + " Disables SKAS3 usage, so that SKAS0 is used, unless you \n" + " specify mode=tt.\n\n"); + __uml_setup("nosysemu", nosysemu_cmd_param, "nosysemu\n" " Turns off syscall emulation patch for ptrace (SYSEMU) on.\n" @@ -227,11 +249,12 @@ __uml_setup("nosysemu", nosysemu_cmd_param, static void __init check_sysemu(void) { + void *stack; int pid, syscall, n, status, count=0; printk("Checking syscall emulation patch for ptrace..."); sysemu_supported = 0; - pid = start_ptraced_child(); + pid = start_ptraced_child(&stack); if(ptrace(PTRACE_SYSEMU, pid, 0, 0) < 0) goto fail; @@ -249,7 +272,7 @@ static void __init check_sysemu(void) panic("check_sysemu : failed to modify system " "call return, errno = %d", errno); - if (stop_ptraced_child(pid, 0, 0) < 0) + if (stop_ptraced_child(pid, stack, 0, 0) < 0) goto fail_stopped; sysemu_supported = 1; @@ -257,7 +280,7 @@ static void __init check_sysemu(void) set_using_sysemu(!force_sysemu_disabled); printk("Checking advanced syscall emulation patch for ptrace..."); - pid = start_ptraced_child(); + pid = start_ptraced_child(&stack); while(1){ count++; if(ptrace(PTRACE_SYSEMU_SINGLESTEP, pid, 0, 0) < 0) @@ -282,7 +305,7 @@ static void __init check_sysemu(void) break; } } - if (stop_ptraced_child(pid, 0, 0) < 0) + if (stop_ptraced_child(pid, stack, 0, 0) < 0) goto fail_stopped; sysemu_supported = 2; @@ -293,17 +316,18 @@ static void __init check_sysemu(void) return; fail: - stop_ptraced_child(pid, 1, 0); + stop_ptraced_child(pid, stack, 1, 0); fail_stopped: printk("missing\n"); } void __init check_ptrace(void) { + void *stack; int pid, syscall, n, status; printk("Checking that ptrace can change system call numbers..."); - pid = start_ptraced_child(); + pid = start_ptraced_child(&stack); if (ptrace(PTRACE_OLDSETOPTIONS, pid, 0, (void *)PTRACE_O_TRACESYSGOOD) < 0) panic("check_ptrace: PTRACE_SETOPTIONS failed, errno = %d", errno); @@ -330,7 +354,7 @@ void __init check_ptrace(void) break; } } - stop_ptraced_child(pid, 0, 1); + stop_ptraced_child(pid, stack, 0, 1); printk("OK\n"); check_sysemu(); } @@ -358,49 +382,54 @@ void forward_pending_sigio(int target) kill(target, SIGIO); } +extern void *__syscall_stub_start, __syscall_stub_end; + #ifdef UML_CONFIG_MODE_SKAS -static inline int check_skas3_ptrace_support(void) + +static inline void check_skas3_ptrace_support(void) { struct ptrace_faultinfo fi; - int pid, n, ret = 1; + void *stack; + int pid, n; printf("Checking for the skas3 patch in the host..."); - pid = start_ptraced_child(); + pid = start_ptraced_child(&stack); n = ptrace(PTRACE_FAULTINFO, pid, 0, &fi); if (n < 0) { + ptrace_faultinfo = 0; if(errno == EIO) printf("not found\n"); else { perror("not found"); } - ret = 0; - } else { - printf("found\n"); + } + else { + if (!ptrace_faultinfo) + printf("found but disabled on command line\n"); + else + printf("found\n"); } init_registers(pid); - stop_ptraced_child(pid, 1, 1); - - return(ret); + stop_ptraced_child(pid, stack, 1, 1); } int can_do_skas(void) { - int ret = 1; - printf("Checking for /proc/mm..."); if (os_access("/proc/mm", OS_ACC_W_OK) < 0) { + proc_mm = 0; printf("not found\n"); - ret = 0; - goto out; } else { - printf("found\n"); + if (!proc_mm) + printf("found but disabled on command line\n"); + else + printf("found\n"); } - ret = check_skas3_ptrace_support(); -out: - return ret; + check_skas3_ptrace_support(); + return 1; } #else int can_do_skas(void) diff --git a/arch/um/kernel/process_kern.c b/arch/um/kernel/process_kern.c index 804c6bbdf67..c23d8a08d0f 100644 --- a/arch/um/kernel/process_kern.c +++ b/arch/um/kernel/process_kern.c @@ -8,6 +8,7 @@ #include "linux/kernel.h" #include "linux/sched.h" #include "linux/interrupt.h" +#include "linux/string.h" #include "linux/mm.h" #include "linux/slab.h" #include "linux/utsname.h" @@ -95,8 +96,8 @@ int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags) current->thread.request.u.thread.proc = fn; current->thread.request.u.thread.arg = arg; - pid = do_fork(CLONE_VM | CLONE_UNTRACED | flags, 0, NULL, 0, NULL, - NULL); + pid = do_fork(CLONE_VM | CLONE_UNTRACED | flags, 0, + ¤t->thread.regs, 0, NULL, NULL); if(pid < 0) panic("do_fork failed in kernel_thread, errno = %d", pid); return(pid); @@ -168,7 +169,7 @@ int current_pid(void) void default_idle(void) { - uml_idle_timer(); + CHOOSE_MODE(uml_idle_timer(), (void) 0); atomic_inc(&init_mm.mm_count); current->mm = &init_mm; @@ -322,12 +323,7 @@ void do_uml_exitcalls(void) char *uml_strdup(char *string) { - char *new; - - new = kmalloc(strlen(string) + 1, GFP_KERNEL); - if(new == NULL) return(NULL); - strcpy(new, string); - return(new); + return kstrdup(string, GFP_KERNEL); } int copy_to_user_proc(void __user *to, void *from, int size) @@ -416,7 +412,7 @@ int __init make_proc_sysemu(void) if (ent == NULL) { - printk("Failed to register /proc/sysemu\n"); + printk(KERN_WARNING "Failed to register /proc/sysemu\n"); return(0); } diff --git a/arch/um/kernel/reboot.c b/arch/um/kernel/reboot.c index 207f89d7490..a637e885c58 100644 --- a/arch/um/kernel/reboot.c +++ b/arch/um/kernel/reboot.c @@ -38,35 +38,28 @@ static void kill_off_processes(void) void uml_cleanup(void) { - kill_off_processes(); + kmalloc_ok = 0; do_uml_exitcalls(); + kill_off_processes(); } void machine_restart(char * __unused) { - do_uml_exitcalls(); - kill_off_processes(); + uml_cleanup(); CHOOSE_MODE(reboot_tt(), reboot_skas()); } -EXPORT_SYMBOL(machine_restart); - void machine_power_off(void) { - do_uml_exitcalls(); - kill_off_processes(); + uml_cleanup(); CHOOSE_MODE(halt_tt(), halt_skas()); } -EXPORT_SYMBOL(machine_power_off); - void machine_halt(void) { machine_power_off(); } -EXPORT_SYMBOL(machine_halt); - /* * Overrides for Emacs so that we follow Linus's tabbing style. * Emacs will notice this stuff at the end of the file and automatically diff --git a/arch/um/kernel/skas/Makefile b/arch/um/kernel/skas/Makefile index d37d1bfcd6f..d296d55ade4 100644 --- a/arch/um/kernel/skas/Makefile +++ b/arch/um/kernel/skas/Makefile @@ -3,11 +3,14 @@ # Licensed under the GPL # -obj-y := exec_kern.o mem.o mem_user.o mmu.o process.o process_kern.o \ - syscall_kern.o syscall_user.o time.o tlb.o trap_user.o uaccess.o \ +obj-y := clone.o exec_kern.o mem.o mem_user.o mmu.o process.o process_kern.o \ + syscall_kern.o syscall_user.o tlb.o trap_user.o uaccess.o \ subdir- := util -USER_OBJS := process.o time.o +USER_OBJS := process.o clone.o include arch/um/scripts/Makefile.rules + +# clone.o is in the stub, so it can't be built with profiling +$(obj)/clone.o : c_flags = -Wp,-MD,$(depfile) $(call unprofile,$(USER_CFLAGS)) diff --git a/arch/um/kernel/skas/clone.c b/arch/um/kernel/skas/clone.c new file mode 100644 index 00000000000..4dc55f10cd1 --- /dev/null +++ b/arch/um/kernel/skas/clone.c @@ -0,0 +1,44 @@ +#include <sched.h> +#include <signal.h> +#include <sys/mman.h> +#include <sys/time.h> +#include <asm/unistd.h> +#include <asm/page.h> +#include "ptrace_user.h" +#include "skas.h" +#include "stub-data.h" +#include "uml-config.h" +#include "sysdep/stub.h" + +/* This is in a separate file because it needs to be compiled with any + * extraneous gcc flags (-pg, -fprofile-arcs, -ftest-coverage) disabled + */ +void __attribute__ ((__section__ (".__syscall_stub"))) +stub_clone_handler(void) +{ + long err; + struct stub_data *from = (struct stub_data *) UML_CONFIG_STUB_DATA; + + err = stub_syscall2(__NR_clone, CLONE_PARENT | CLONE_FILES | SIGCHLD, + UML_CONFIG_STUB_DATA + PAGE_SIZE / 2 - + sizeof(void *)); + if(err != 0) + goto out; + + err = stub_syscall4(__NR_ptrace, PTRACE_TRACEME, 0, 0, 0); + if(err) + goto out; + + err = stub_syscall3(__NR_setitimer, ITIMER_VIRTUAL, + (long) &from->timer, 0); + if(err) + goto out; + + err = stub_syscall6(STUB_MMAP_NR, UML_CONFIG_STUB_DATA, PAGE_SIZE, + PROT_READ | PROT_WRITE, MAP_FIXED | MAP_SHARED, + from->fd, from->offset); + out: + /* save current result. Parent: pid; child: retcode of mmap */ + from->err = err; + trap_myself(); +} diff --git a/arch/um/kernel/skas/exec_kern.c b/arch/um/kernel/skas/exec_kern.c index c6b4d5dba78..77ed7bbab21 100644 --- a/arch/um/kernel/skas/exec_kern.c +++ b/arch/um/kernel/skas/exec_kern.c @@ -18,7 +18,7 @@ void flush_thread_skas(void) { force_flush_all(); - switch_mm_skas(current->mm->context.skas.mm_fd); + switch_mm_skas(¤t->mm->context.skas.id); } void start_thread_skas(struct pt_regs *regs, unsigned long eip, diff --git a/arch/um/kernel/skas/include/mm_id.h b/arch/um/kernel/skas/include/mm_id.h new file mode 100644 index 00000000000..48dd0989dda --- /dev/null +++ b/arch/um/kernel/skas/include/mm_id.h @@ -0,0 +1,17 @@ +/* + * Copyright (C) 2005 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __MM_ID_H +#define __MM_ID_H + +struct mm_id { + union { + int mm_fd; + int pid; + } u; + unsigned long stack; +}; + +#endif diff --git a/arch/um/kernel/skas/include/mmu-skas.h b/arch/um/kernel/skas/include/mmu-skas.h index 4cd60d7213f..278b72f1d9a 100644 --- a/arch/um/kernel/skas/include/mmu-skas.h +++ b/arch/um/kernel/skas/include/mmu-skas.h @@ -6,10 +6,15 @@ #ifndef __SKAS_MMU_H #define __SKAS_MMU_H +#include "mm_id.h" + struct mmu_context_skas { - int mm_fd; + struct mm_id id; + unsigned long last_page_table; }; +extern void switch_mm_skas(struct mm_id * mm_idp); + #endif /* diff --git a/arch/um/kernel/skas/include/mode-skas.h b/arch/um/kernel/skas/include/mode-skas.h index c1e33bd788d..bcd26a6a388 100644 --- a/arch/um/kernel/skas/include/mode-skas.h +++ b/arch/um/kernel/skas/include/mode-skas.h @@ -13,7 +13,6 @@ extern unsigned long exec_fp_regs[]; extern unsigned long exec_fpx_regs[]; extern int have_fpx_regs; -extern void user_time_init_skas(void); extern void sig_handler_common_skas(int sig, void *sc_ptr); extern void halt_skas(void); extern void reboot_skas(void); diff --git a/arch/um/kernel/skas/include/skas.h b/arch/um/kernel/skas/include/skas.h index 96b51dba347..d983ea84254 100644 --- a/arch/um/kernel/skas/include/skas.h +++ b/arch/um/kernel/skas/include/skas.h @@ -6,9 +6,11 @@ #ifndef __SKAS_H #define __SKAS_H +#include "mm_id.h" #include "sysdep/ptrace.h" extern int userspace_pid[]; +extern int proc_mm, ptrace_faultinfo; extern void switch_threads(void *me, void *next); extern void thread_wait(void *sw, void *fb); @@ -22,16 +24,18 @@ extern void new_thread_proc(void *stack, void (*handler)(int sig)); extern void remove_sigstack(void); extern void new_thread_handler(int sig); extern void handle_syscall(union uml_pt_regs *regs); -extern void map(int fd, unsigned long virt, unsigned long len, int r, int w, - int x, int phys_fd, unsigned long long offset); -extern int unmap(int fd, void *addr, unsigned long len); -extern int protect(int fd, unsigned long addr, unsigned long len, - int r, int w, int x); +extern int map(struct mm_id * mm_idp, unsigned long virt, unsigned long len, + int r, int w, int x, int phys_fd, unsigned long long offset); +extern int unmap(struct mm_id * mm_idp, void *addr, unsigned long len); +extern int protect(struct mm_id * mm_idp, unsigned long addr, + unsigned long len, int r, int w, int x); extern void user_signal(int sig, union uml_pt_regs *regs, int pid); extern int new_mm(int from); -extern void start_userspace(int cpu); +extern int start_userspace(unsigned long stub_stack); +extern int copy_context_skas0(unsigned long stack, int pid); extern void get_skas_faultinfo(int pid, struct faultinfo * fi); extern long execute_syscall_skas(void *r); +extern unsigned long current_stub_stack(void); #endif diff --git a/arch/um/kernel/skas/include/stub-data.h b/arch/um/kernel/skas/include/stub-data.h new file mode 100644 index 00000000000..f6ed92c3727 --- /dev/null +++ b/arch/um/kernel/skas/include/stub-data.h @@ -0,0 +1,18 @@ +/* + * Copyright (C) 2005 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __STUB_DATA_H +#define __STUB_DATA_H + +#include <sys/time.h> + +struct stub_data { + long offset; + int fd; + struct itimerval timer; + long err; +}; + +#endif diff --git a/arch/um/kernel/skas/mem.c b/arch/um/kernel/skas/mem.c index 438db2f4345..147466d7ff4 100644 --- a/arch/um/kernel/skas/mem.c +++ b/arch/um/kernel/skas/mem.c @@ -5,7 +5,9 @@ #include "linux/config.h" #include "linux/mm.h" +#include "asm/pgtable.h" #include "mem_user.h" +#include "skas.h" unsigned long set_task_sizes_skas(int arg, unsigned long *host_size_out, unsigned long *task_size_out) @@ -18,7 +20,9 @@ unsigned long set_task_sizes_skas(int arg, unsigned long *host_size_out, *task_size_out = CONFIG_HOST_TASK_SIZE; #else *host_size_out = top; - *task_size_out = top; + if (proc_mm && ptrace_faultinfo) + *task_size_out = top; + else *task_size_out = CONFIG_STUB_START & PGDIR_MASK; #endif return(((unsigned long) set_task_sizes_skas) & ~0xffffff); } diff --git a/arch/um/kernel/skas/mem_user.c b/arch/um/kernel/skas/mem_user.c index 1310bf1e88d..b0980ff3bd9 100644 --- a/arch/um/kernel/skas/mem_user.c +++ b/arch/um/kernel/skas/mem_user.c @@ -3,100 +3,171 @@ * Licensed under the GPL */ +#include <signal.h> #include <errno.h> #include <sys/mman.h> +#include <sys/wait.h> +#include <asm/page.h> +#include <asm/unistd.h> #include "mem_user.h" #include "mem.h" +#include "mm_id.h" #include "user.h" #include "os.h" #include "proc_mm.h" - -void map(int fd, unsigned long virt, unsigned long len, int r, int w, - int x, int phys_fd, unsigned long long offset) +#include "ptrace_user.h" +#include "user_util.h" +#include "kern_util.h" +#include "task.h" +#include "registers.h" +#include "uml-config.h" +#include "sysdep/ptrace.h" +#include "sysdep/stub.h" +#include "skas.h" + +extern unsigned long syscall_stub, __syscall_stub_start; + +extern void wait_stub_done(int pid, int sig, char * fname); + +static long run_syscall_stub(struct mm_id * mm_idp, int syscall, + unsigned long *args) { - struct proc_mm_op map; - int prot, n; - - prot = (r ? PROT_READ : 0) | (w ? PROT_WRITE : 0) | - (x ? PROT_EXEC : 0); - - map = ((struct proc_mm_op) { .op = MM_MMAP, - .u = - { .mmap = - { .addr = virt, - .len = len, - .prot = prot, - .flags = MAP_SHARED | - MAP_FIXED, - .fd = phys_fd, - .offset = offset - } } } ); - n = os_write_file(fd, &map, sizeof(map)); - if(n != sizeof(map)) - printk("map : /proc/mm map failed, err = %d\n", -n); + int n, pid = mm_idp->u.pid; + unsigned long regs[MAX_REG_NR]; + + get_safe_registers(regs); + regs[REGS_IP_INDEX] = UML_CONFIG_STUB_CODE + + ((unsigned long) &syscall_stub - + (unsigned long) &__syscall_stub_start); + /* XXX Don't have a define for starting a syscall */ + regs[REGS_SYSCALL_NR] = syscall; + regs[REGS_SYSCALL_ARG1] = args[0]; + regs[REGS_SYSCALL_ARG2] = args[1]; + regs[REGS_SYSCALL_ARG3] = args[2]; + regs[REGS_SYSCALL_ARG4] = args[3]; + regs[REGS_SYSCALL_ARG5] = args[4]; + regs[REGS_SYSCALL_ARG6] = args[5]; + n = ptrace_setregs(pid, regs); + if(n < 0){ + printk("run_syscall_stub : PTRACE_SETREGS failed, " + "errno = %d\n", n); + return(n); + } + + wait_stub_done(pid, 0, "run_syscall_stub"); + + return(*((unsigned long *) mm_idp->stack)); } -int unmap(int fd, void *addr, unsigned long len) +int map(struct mm_id *mm_idp, unsigned long virt, unsigned long len, + int r, int w, int x, int phys_fd, unsigned long long offset) { - struct proc_mm_op unmap; - int n; - - unmap = ((struct proc_mm_op) { .op = MM_MUNMAP, - .u = - { .munmap = - { .addr = (unsigned long) addr, - .len = len } } } ); - n = os_write_file(fd, &unmap, sizeof(unmap)); - if(n != sizeof(unmap)) { - if(n < 0) - return(n); - else if(n > 0) - return(-EIO); - } - - return(0); + int prot, n; + + prot = (r ? PROT_READ : 0) | (w ? PROT_WRITE : 0) | + (x ? PROT_EXEC : 0); + + if(proc_mm){ + struct proc_mm_op map; + int fd = mm_idp->u.mm_fd; + map = ((struct proc_mm_op) { .op = MM_MMAP, + .u = + { .mmap = + { .addr = virt, + .len = len, + .prot = prot, + .flags = MAP_SHARED | + MAP_FIXED, + .fd = phys_fd, + .offset= offset + } } } ); + n = os_write_file(fd, &map, sizeof(map)); + if(n != sizeof(map)) + printk("map : /proc/mm map failed, err = %d\n", -n); + } + else { + long res; + unsigned long args[] = { virt, len, prot, + MAP_SHARED | MAP_FIXED, phys_fd, + MMAP_OFFSET(offset) }; + + res = run_syscall_stub(mm_idp, STUB_MMAP_NR, args); + if((void *) res == MAP_FAILED) + printk("mmap stub failed, errno = %d\n", res); + } + + return 0; } -int protect(int fd, unsigned long addr, unsigned long len, int r, int w, - int x, int must_succeed) +int unmap(struct mm_id *mm_idp, void *addr, unsigned long len) { - struct proc_mm_op protect; - int prot, n; - - prot = (r ? PROT_READ : 0) | (w ? PROT_WRITE : 0) | - (x ? PROT_EXEC : 0); - - protect = ((struct proc_mm_op) { .op = MM_MPROTECT, - .u = - { .mprotect = - { .addr = (unsigned long) addr, - .len = len, - .prot = prot } } } ); - - n = os_write_file(fd, &protect, sizeof(protect)); - if(n != sizeof(protect)) { - if(n == 0) return(0); - - if(must_succeed) - panic("protect failed, err = %d", -n); - - return(-EIO); - } + int n; + + if(proc_mm){ + struct proc_mm_op unmap; + int fd = mm_idp->u.mm_fd; + unmap = ((struct proc_mm_op) { .op = MM_MUNMAP, + .u = + { .munmap = + { .addr = + (unsigned long) addr, + .len = len } } } ); + n = os_write_file(fd, &unmap, sizeof(unmap)); + if(n != sizeof(unmap)) { + if(n < 0) + return(n); + else if(n > 0) + return(-EIO); + } + } + else { + int res; + unsigned long args[] = { (unsigned long) addr, len, 0, 0, 0, + 0 }; + + res = run_syscall_stub(mm_idp, __NR_munmap, args); + if(res < 0) + printk("munmap stub failed, errno = %d\n", res); + } + + return(0); +} - return(0); +int protect(struct mm_id *mm_idp, unsigned long addr, unsigned long len, + int r, int w, int x) +{ + struct proc_mm_op protect; + int prot, n; + + prot = (r ? PROT_READ : 0) | (w ? PROT_WRITE : 0) | + (x ? PROT_EXEC : 0); + + if(proc_mm){ + int fd = mm_idp->u.mm_fd; + protect = ((struct proc_mm_op) { .op = MM_MPROTECT, + .u = + { .mprotect = + { .addr = + (unsigned long) addr, + .len = len, + .prot = prot } } } ); + + n = os_write_file(fd, &protect, sizeof(protect)); + if(n != sizeof(protect)) + panic("protect failed, err = %d", -n); + } + else { + int res; + unsigned long args[] = { addr, len, prot, 0, 0, 0 }; + + res = run_syscall_stub(mm_idp, __NR_mprotect, args); + if(res < 0) + panic("mprotect stub failed, errno = %d\n", res); + } + + return(0); } void before_mem_skas(unsigned long unused) { } - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/kernel/skas/mmu.c b/arch/um/kernel/skas/mmu.c index 6cb9a6d028a..d232daa42c3 100644 --- a/arch/um/kernel/skas/mmu.c +++ b/arch/um/kernel/skas/mmu.c @@ -3,46 +3,143 @@ * Licensed under the GPL */ +#include "linux/config.h" #include "linux/sched.h" #include "linux/list.h" #include "linux/spinlock.h" #include "linux/slab.h" +#include "linux/errno.h" +#include "linux/mm.h" #include "asm/current.h" #include "asm/segment.h" #include "asm/mmu.h" +#include "asm/pgalloc.h" +#include "asm/pgtable.h" #include "os.h" #include "skas.h" +extern int __syscall_stub_start; + +static int init_stub_pte(struct mm_struct *mm, unsigned long proc, + unsigned long kernel) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + + spin_lock(&mm->page_table_lock); + pgd = pgd_offset(mm, proc); + pud = pud_alloc(mm, pgd, proc); + if (!pud) + goto out; + + pmd = pmd_alloc(mm, pud, proc); + if (!pmd) + goto out_pmd; + + pte = pte_alloc_map(mm, pmd, proc); + if (!pte) + goto out_pte; + + /* There's an interaction between the skas0 stub pages, stack + * randomization, and the BUG at the end of exit_mmap. exit_mmap + * checks that the number of page tables freed is the same as had + * been allocated. If the stack is on the last page table page, + * then the stack pte page will be freed, and if not, it won't. To + * avoid having to know where the stack is, or if the process mapped + * something at the top of its address space for some other reason, + * we set TASK_SIZE to end at the start of the last page table. + * This keeps exit_mmap off the last page, but introduces a leak + * of that page. So, we hang onto it here and free it in + * destroy_context_skas. + */ + + mm->context.skas.last_page_table = pmd_page_kernel(*pmd); + + *pte = mk_pte(virt_to_page(kernel), __pgprot(_PAGE_PRESENT)); + *pte = pte_mkexec(*pte); + *pte = pte_wrprotect(*pte); + spin_unlock(&mm->page_table_lock); + return(0); + + out_pmd: + pud_free(pud); + out_pte: + pmd_free(pmd); + out: + spin_unlock(&mm->page_table_lock); + return(-ENOMEM); +} + int init_new_context_skas(struct task_struct *task, struct mm_struct *mm) { - int from; + struct mm_struct *cur_mm = current->mm; + struct mm_id *cur_mm_id = &cur_mm->context.skas.id; + struct mm_id *mm_id = &mm->context.skas.id; + unsigned long stack; + int from, ret; - if((current->mm != NULL) && (current->mm != &init_mm)) - from = current->mm->context.skas.mm_fd; - else from = -1; + if(proc_mm){ + if((cur_mm != NULL) && (cur_mm != &init_mm)) + from = cur_mm->context.skas.id.u.mm_fd; + else from = -1; - mm->context.skas.mm_fd = new_mm(from); - if(mm->context.skas.mm_fd < 0){ - printk("init_new_context_skas - new_mm failed, errno = %d\n", - mm->context.skas.mm_fd); - return(mm->context.skas.mm_fd); + ret = new_mm(from); + if(ret < 0){ + printk("init_new_context_skas - new_mm failed, " + "errno = %d\n", ret); + return ret; + } + mm_id->u.mm_fd = ret; } + else { + /* This zeros the entry that pgd_alloc didn't, needed since + * we are about to reinitialize it, and want mm.nr_ptes to + * be accurate. + */ + mm->pgd[USER_PTRS_PER_PGD] = __pgd(0); - return(0); + ret = init_stub_pte(mm, CONFIG_STUB_CODE, + (unsigned long) &__syscall_stub_start); + if(ret) + goto out; + + ret = -ENOMEM; + stack = get_zeroed_page(GFP_KERNEL); + if(stack == 0) + goto out; + mm_id->stack = stack; + + ret = init_stub_pte(mm, CONFIG_STUB_DATA, stack); + if(ret) + goto out_free; + + mm->nr_ptes--; + + if((cur_mm != NULL) && (cur_mm != &init_mm)) + mm_id->u.pid = copy_context_skas0(stack, + cur_mm_id->u.pid); + else mm_id->u.pid = start_userspace(stack); + } + + return 0; + + out_free: + free_page(mm_id->stack); + out: + return ret; } void destroy_context_skas(struct mm_struct *mm) { - os_close_file(mm->context.skas.mm_fd); -} + struct mmu_context_skas *mmu = &mm->context.skas; -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ + if(proc_mm) + os_close_file(mmu->id.u.mm_fd); + else { + os_kill_ptraced_process(mmu->id.u.pid, 1); + free_page(mmu->id.stack); + free_page(mmu->last_page_table); + } +} diff --git a/arch/um/kernel/skas/process.c b/arch/um/kernel/skas/process.c index 773cd2b525f..6dd9e5bf18e 100644 --- a/arch/um/kernel/skas/process.c +++ b/arch/um/kernel/skas/process.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2002- 2004 Jeff Dike (jdike@addtoit.com) * Licensed under the GPL */ @@ -13,7 +13,9 @@ #include <sys/wait.h> #include <sys/mman.h> #include <sys/user.h> +#include <sys/time.h> #include <asm/unistd.h> +#include <asm/types.h> #include "user.h" #include "ptrace_user.h" #include "time_user.h" @@ -21,13 +23,18 @@ #include "user_util.h" #include "kern_util.h" #include "skas.h" +#include "stub-data.h" +#include "mm_id.h" #include "sysdep/sigcontext.h" +#include "sysdep/stub.h" #include "os.h" #include "proc_mm.h" #include "skas_ptrace.h" #include "chan_user.h" #include "signal_user.h" #include "registers.h" +#include "mem.h" +#include "uml-config.h" #include "process.h" int is_skas_winch(int pid, int fd, void *data) @@ -39,20 +46,55 @@ int is_skas_winch(int pid, int fd, void *data) return(1); } -void get_skas_faultinfo(int pid, struct faultinfo * fi) +void wait_stub_done(int pid, int sig, char * fname) { - int err; - - err = ptrace(PTRACE_FAULTINFO, pid, 0, fi); - if(err) - panic("get_skas_faultinfo - PTRACE_FAULTINFO failed, " - "errno = %d\n", errno); + int n, status, err; + + do { + if ( sig != -1 ) { + err = ptrace(PTRACE_CONT, pid, 0, sig); + if(err) + panic("%s : continue failed, errno = %d\n", + fname, errno); + } + sig = 0; + + CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED)); + } while((n >= 0) && WIFSTOPPED(status) && + (WSTOPSIG(status) == SIGVTALRM)); + + if((n < 0) || !WIFSTOPPED(status) || + (WSTOPSIG(status) != SIGUSR1 && WSTOPSIG(status) != SIGTRAP)){ + panic("%s : failed to wait for SIGUSR1/SIGTRAP, " + "pid = %d, n = %d, errno = %d, status = 0x%x\n", + fname, pid, n, errno, status); + } +} - /* Special handling for i386, which has different structs */ - if (sizeof(struct ptrace_faultinfo) < sizeof(struct faultinfo)) - memset((char *)fi + sizeof(struct ptrace_faultinfo), 0, - sizeof(struct faultinfo) - - sizeof(struct ptrace_faultinfo)); +void get_skas_faultinfo(int pid, struct faultinfo * fi) +{ + int err; + + if(ptrace_faultinfo){ + err = ptrace(PTRACE_FAULTINFO, pid, 0, fi); + if(err) + panic("get_skas_faultinfo - PTRACE_FAULTINFO failed, " + "errno = %d\n", errno); + + /* Special handling for i386, which has different structs */ + if (sizeof(struct ptrace_faultinfo) < sizeof(struct faultinfo)) + memset((char *)fi + sizeof(struct ptrace_faultinfo), 0, + sizeof(struct faultinfo) - + sizeof(struct ptrace_faultinfo)); + } + else { + wait_stub_done(pid, SIGSEGV, "get_skas_faultinfo"); + + /* faultinfo is prepared by the stub-segv-handler at start of + * the stub stack page. We just have to copy it. + */ + memcpy(fi, (void *)current_stub_stack(), sizeof(*fi)); + } } static void handle_segv(int pid, union uml_pt_regs * regs) @@ -91,11 +133,56 @@ static void handle_trap(int pid, union uml_pt_regs *regs, int local_using_sysemu handle_syscall(regs); } -static int userspace_tramp(void *arg) +extern int __syscall_stub_start; + +static int userspace_tramp(void *stack) { - init_new_thread_signals(0); - enable_timer(); + void *addr; + ptrace(PTRACE_TRACEME, 0, 0, 0); + + init_new_thread_signals(1); + enable_timer(); + + if(!proc_mm){ + /* This has a pte, but it can't be mapped in with the usual + * tlb_flush mechanism because this is part of that mechanism + */ + int fd; + __u64 offset; + + fd = phys_mapping(to_phys(&__syscall_stub_start), &offset); + addr = mmap64((void *) UML_CONFIG_STUB_CODE, page_size(), + PROT_EXEC, MAP_FIXED | MAP_PRIVATE, fd, offset); + if(addr == MAP_FAILED){ + printk("mapping mmap stub failed, errno = %d\n", + errno); + exit(1); + } + + if(stack != NULL){ + fd = phys_mapping(to_phys(stack), &offset); + addr = mmap((void *) UML_CONFIG_STUB_DATA, page_size(), + PROT_READ | PROT_WRITE, + MAP_FIXED | MAP_SHARED, fd, offset); + if(addr == MAP_FAILED){ + printk("mapping segfault stack failed, " + "errno = %d\n", errno); + exit(1); + } + } + } + if(!ptrace_faultinfo && (stack != NULL)){ + unsigned long v = UML_CONFIG_STUB_CODE + + (unsigned long) stub_segv_handler - + (unsigned long) &__syscall_stub_start; + + set_sigstack((void *) UML_CONFIG_STUB_DATA, page_size()); + set_handler(SIGSEGV, (void *) v, SA_ONSTACK, + SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, + SIGUSR1, -1); + } + os_stop_process(os_getpid()); return(0); } @@ -105,11 +192,11 @@ static int userspace_tramp(void *arg) #define NR_CPUS 1 int userspace_pid[NR_CPUS]; -void start_userspace(int cpu) +int start_userspace(unsigned long stub_stack) { void *stack; unsigned long sp; - int pid, status, n; + int pid, status, n, flags; stack = mmap(NULL, PAGE_SIZE, PROT_READ | PROT_WRITE | PROT_EXEC, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); @@ -117,8 +204,9 @@ void start_userspace(int cpu) panic("start_userspace : mmap failed, errno = %d", errno); sp = (unsigned long) stack + PAGE_SIZE - sizeof(void *); - pid = clone(userspace_tramp, (void *) sp, - CLONE_FILES | CLONE_VM | SIGCHLD, NULL); + flags = CLONE_FILES | SIGCHLD; + if(proc_mm) flags |= CLONE_VM; + pid = clone(userspace_tramp, (void *) sp, flags, (void *) stub_stack); if(pid < 0) panic("start_userspace : clone failed, errno = %d", errno); @@ -140,7 +228,7 @@ void start_userspace(int cpu) if(munmap(stack, PAGE_SIZE) < 0) panic("start_userspace : munmap failed, errno = %d\n", errno); - userspace_pid[cpu] = pid; + return(pid); } void userspace(union uml_pt_regs *regs) @@ -174,7 +262,9 @@ void userspace(union uml_pt_regs *regs) if(WIFSTOPPED(status)){ switch(WSTOPSIG(status)){ case SIGSEGV: - handle_segv(pid, regs); + if(PTRACE_FULL_FAULTINFO || !ptrace_faultinfo) + user_signal(SIGSEGV, regs, pid); + else handle_segv(pid, regs); break; case SIGTRAP + 0x80: handle_trap(pid, regs, local_using_sysemu); @@ -194,6 +284,7 @@ void userspace(union uml_pt_regs *regs) printk("userspace - child stopped with signal " "%d\n", WSTOPSIG(status)); } + pid = userspace_pid[0]; interrupt_end(); /* Avoid -ERESTARTSYS handling in host */ @@ -207,6 +298,67 @@ void userspace(union uml_pt_regs *regs) #define INIT_JMP_HALT 3 #define INIT_JMP_REBOOT 4 + +int copy_context_skas0(unsigned long new_stack, int pid) +{ + int err; + unsigned long regs[MAX_REG_NR]; + unsigned long current_stack = current_stub_stack(); + struct stub_data *data = (struct stub_data *) current_stack; + struct stub_data *child_data = (struct stub_data *) new_stack; + __u64 new_offset; + int new_fd = phys_mapping(to_phys((void *)new_stack), &new_offset); + + /* prepare offset and fd of child's stack as argument for parent's + * and child's mmap2 calls + */ + *data = ((struct stub_data) { .offset = MMAP_OFFSET(new_offset), + .fd = new_fd, + .timer = ((struct itimerval) + { { 0, 1000000 / hz() }, + { 0, 1000000 / hz() }})}); + get_safe_registers(regs); + + /* Set parent's instruction pointer to start of clone-stub */ + regs[REGS_IP_INDEX] = UML_CONFIG_STUB_CODE + + (unsigned long) stub_clone_handler - + (unsigned long) &__syscall_stub_start; + regs[REGS_SP_INDEX] = UML_CONFIG_STUB_DATA + PAGE_SIZE - + sizeof(void *); + err = ptrace_setregs(pid, regs); + if(err < 0) + panic("copy_context_skas0 : PTRACE_SETREGS failed, " + "pid = %d, errno = %d\n", pid, errno); + + /* set a well known return code for detection of child write failure */ + child_data->err = 12345678; + + /* Wait, until parent has finished its work: read child's pid from + * parent's stack, and check, if bad result. + */ + wait_stub_done(pid, 0, "copy_context_skas0"); + + pid = data->err; + if(pid < 0) + panic("copy_context_skas0 - stub-parent reports error %d\n", + pid); + + /* Wait, until child has finished too: read child's result from + * child's stack and check it. + */ + wait_stub_done(pid, -1, "copy_context_skas0"); + if (child_data->err != UML_CONFIG_STUB_DATA) + panic("copy_context_skas0 - stub-child reports error %d\n", + child_data->err); + + if (ptrace(PTRACE_OLDSETOPTIONS, pid, NULL, + (void *)PTRACE_O_TRACESYSGOOD) < 0) + panic("copy_context_skas0 : PTRACE_SETOPTIONS failed, " + "errno = %d\n", errno); + + return pid; +} + void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr, void (*handler)(int)) { @@ -334,21 +486,19 @@ void reboot_skas(void) siglongjmp(initial_jmpbuf, INIT_JMP_REBOOT); } -void switch_mm_skas(int mm_fd) +void switch_mm_skas(struct mm_id *mm_idp) { int err; #warning need cpu pid in switch_mm_skas - err = ptrace(PTRACE_SWITCH_MM, userspace_pid[0], 0, mm_fd); - if(err) - panic("switch_mm_skas - PTRACE_SWITCH_MM failed, errno = %d\n", - errno); -} - -void kill_off_processes_skas(void) -{ -#warning need to loop over userspace_pids in kill_off_processes_skas - os_kill_ptraced_process(userspace_pid[0], 1); + if(proc_mm){ + err = ptrace(PTRACE_SWITCH_MM, userspace_pid[0], 0, + mm_idp->u.mm_fd); + if(err) + panic("switch_mm_skas - PTRACE_SWITCH_MM failed, " + "errno = %d\n", errno); + } + else userspace_pid[0] = mm_idp->u.pid; } /* diff --git a/arch/um/kernel/skas/process_kern.c b/arch/um/kernel/skas/process_kern.c index fc71ef29578..cbabab104ac 100644 --- a/arch/um/kernel/skas/process_kern.c +++ b/arch/um/kernel/skas/process_kern.c @@ -111,8 +111,7 @@ int copy_thread_skas(int nr, unsigned long clone_flags, unsigned long sp, void (*handler)(int); if(current->thread.forking){ - memcpy(&p->thread.regs.regs.skas, - ¤t->thread.regs.regs.skas, + memcpy(&p->thread.regs.regs.skas, ®s->regs.skas, sizeof(p->thread.regs.regs.skas)); REGS_SET_SYSCALL_RETURN(p->thread.regs.regs.skas.regs, 0); if(sp != 0) REGS_SP(p->thread.regs.regs.skas.regs) = sp; @@ -176,12 +175,14 @@ static int start_kernel_proc(void *unused) return(0); } +extern int userspace_pid[]; + int start_uml_skas(void) { - start_userspace(0); + if(proc_mm) + userspace_pid[0] = start_userspace(0); init_new_thread_signals(1); - uml_idle_timer(); init_task.thread.request.u.thread.proc = start_kernel_proc; init_task.thread.request.u.thread.arg = NULL; @@ -202,13 +203,30 @@ int thread_pid_skas(struct task_struct *task) return(userspace_pid[0]); } -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ +void kill_off_processes_skas(void) +{ + if(proc_mm) +#warning need to loop over userspace_pids in kill_off_processes_skas + os_kill_ptraced_process(userspace_pid[0], 1); + else { + struct task_struct *p; + int pid, me; + + me = os_getpid(); + for_each_process(p){ + if(p->mm == NULL) + continue; + + pid = p->mm->context.skas.id.u.pid; + os_kill_ptraced_process(pid, 1); + } + } +} + +unsigned long current_stub_stack(void) +{ + if(current->mm == NULL) + return(0); + + return(current->mm->context.skas.id.stack); +} diff --git a/arch/um/kernel/skas/syscall_user.c b/arch/um/kernel/skas/syscall_user.c index 2828e6e3772..6b066497014 100644 --- a/arch/um/kernel/skas/syscall_user.c +++ b/arch/um/kernel/skas/syscall_user.c @@ -15,7 +15,7 @@ void handle_syscall(union uml_pt_regs *regs) { long result; -#if UML_CONFIG_SYSCALL_DEBUG +#ifdef UML_CONFIG_SYSCALL_DEBUG int index; index = record_syscall_start(UPT_SYSCALL_NR(regs)); @@ -27,7 +27,7 @@ void handle_syscall(union uml_pt_regs *regs) REGS_SET_SYSCALL_RETURN(regs->skas.regs, result); syscall_trace(regs, 1); -#if UML_CONFIG_SYSCALL_DEBUG +#ifdef UML_CONFIG_SYSCALL_DEBUG record_syscall_end(index, result); #endif } diff --git a/arch/um/kernel/skas/time.c b/arch/um/kernel/skas/time.c deleted file mode 100644 index 98091494b89..00000000000 --- a/arch/um/kernel/skas/time.c +++ /dev/null @@ -1,30 +0,0 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#include <sys/signal.h> -#include <sys/time.h> -#include "time_user.h" -#include "process.h" -#include "user.h" - -void user_time_init_skas(void) -{ - if(signal(SIGALRM, (__sighandler_t) alarm_handler) == SIG_ERR) - panic("Couldn't set SIGALRM handler"); - if(signal(SIGVTALRM, (__sighandler_t) alarm_handler) == SIG_ERR) - panic("Couldn't set SIGVTALRM handler"); - set_interval(ITIMER_VIRTUAL); -} - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/kernel/skas/tlb.c b/arch/um/kernel/skas/tlb.c index b8c5e71763d..6230999c672 100644 --- a/arch/um/kernel/skas/tlb.c +++ b/arch/um/kernel/skas/tlb.c @@ -6,6 +6,7 @@ #include "linux/stddef.h" #include "linux/sched.h" +#include "linux/config.h" #include "linux/mm.h" #include "asm/page.h" #include "asm/pgtable.h" @@ -17,7 +18,7 @@ #include "os.h" #include "tlb.h" -static void do_ops(int fd, struct host_vm_op *ops, int last) +static void do_ops(union mm_context *mmu, struct host_vm_op *ops, int last) { struct host_vm_op *op; int i; @@ -26,18 +27,18 @@ static void do_ops(int fd, struct host_vm_op *ops, int last) op = &ops[i]; switch(op->type){ case MMAP: - map(fd, op->u.mmap.addr, op->u.mmap.len, + map(&mmu->skas.id, op->u.mmap.addr, op->u.mmap.len, op->u.mmap.r, op->u.mmap.w, op->u.mmap.x, op->u.mmap.fd, op->u.mmap.offset); break; case MUNMAP: - unmap(fd, (void *) op->u.munmap.addr, + unmap(&mmu->skas.id, (void *) op->u.munmap.addr, op->u.munmap.len); break; case MPROTECT: - protect(fd, op->u.mprotect.addr, op->u.mprotect.len, - op->u.mprotect.r, op->u.mprotect.w, - op->u.mprotect.x); + protect(&mmu->skas.id, op->u.mprotect.addr, + op->u.mprotect.len, op->u.mprotect.r, + op->u.mprotect.w, op->u.mprotect.x); break; default: printk("Unknown op type %d in do_ops\n", op->type); @@ -46,12 +47,15 @@ static void do_ops(int fd, struct host_vm_op *ops, int last) } } +extern int proc_mm; + static void fix_range(struct mm_struct *mm, unsigned long start_addr, unsigned long end_addr, int force) { - int fd = mm->context.skas.mm_fd; + if(!proc_mm && (end_addr > CONFIG_STUB_START)) + end_addr = CONFIG_STUB_START; - fix_range_common(mm, start_addr, end_addr, force, fd, do_ops); + fix_range_common(mm, start_addr, end_addr, force, do_ops); } void __flush_tlb_one_skas(unsigned long addr) @@ -69,17 +73,20 @@ void flush_tlb_range_skas(struct vm_area_struct *vma, unsigned long start, void flush_tlb_mm_skas(struct mm_struct *mm) { + unsigned long end; + /* Don't bother flushing if this address space is about to be * destroyed. */ if(atomic_read(&mm->mm_users) == 0) return; - fix_range(mm, 0, host_task_size, 0); - flush_tlb_kernel_range_common(start_vm, end_vm); + end = proc_mm ? task_size : CONFIG_STUB_START; + fix_range(mm, 0, end, 0); } void force_flush_all_skas(void) { - fix_range(current->mm, 0, host_task_size, 1); + unsigned long end = proc_mm ? task_size : CONFIG_STUB_START; + fix_range(current->mm, 0, end, 1); } diff --git a/arch/um/kernel/skas/trap_user.c b/arch/um/kernel/skas/trap_user.c index 0dee1d95c80..9950a6716fe 100644 --- a/arch/um/kernel/skas/trap_user.c +++ b/arch/um/kernel/skas/trap_user.c @@ -58,7 +58,6 @@ void user_signal(int sig, union uml_pt_regs *regs, int pid) int segv = ((sig == SIGFPE) || (sig == SIGSEGV) || (sig == SIGBUS) || (sig == SIGILL) || (sig == SIGTRAP)); - regs->skas.is_user = 1; if (segv) get_skas_faultinfo(pid, ®s->skas.faultinfo); info = &sig_info[sig]; diff --git a/arch/um/kernel/syscall_kern.c b/arch/um/kernel/syscall_kern.c index b7a55251e89..8e1a3501ff4 100644 --- a/arch/um/kernel/syscall_kern.c +++ b/arch/um/kernel/syscall_kern.c @@ -31,7 +31,8 @@ long sys_fork(void) long ret; current->thread.forking = 1; - ret = do_fork(SIGCHLD, 0, NULL, 0, NULL, NULL); + ret = do_fork(SIGCHLD, UPT_SP(¤t->thread.regs.regs), + ¤t->thread.regs, 0, NULL, NULL); current->thread.forking = 0; return(ret); } @@ -41,8 +42,9 @@ long sys_vfork(void) long ret; current->thread.forking = 1; - ret = do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, 0, NULL, 0, NULL, - NULL); + ret = do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, + UPT_SP(¤t->thread.regs.regs), + ¤t->thread.regs, 0, NULL, NULL); current->thread.forking = 0; return(ret); } @@ -162,14 +164,3 @@ int next_syscall_index(int limit) spin_unlock(&syscall_lock); return(ret); } - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c index c40c86a3f91..c40b611e3d9 100644 --- a/arch/um/kernel/time.c +++ b/arch/um/kernel/time.c @@ -33,7 +33,7 @@ void timer(void) timeradd(&xtime, &local_offset, &xtime); } -void set_interval(int timer_type) +static void set_interval(int timer_type) { int usec = 1000000/hz(); struct itimerval interval = ((struct itimerval) { { 0, usec }, @@ -45,12 +45,14 @@ void set_interval(int timer_type) void enable_timer(void) { + set_interval(ITIMER_VIRTUAL); +} + +void prepare_timer(void * ptr) +{ int usec = 1000000/hz(); - struct itimerval enable = ((struct itimerval) { { 0, usec }, - { 0, usec }}); - if(setitimer(ITIMER_VIRTUAL, &enable, NULL)) - printk("enable_timer - setitimer failed, errno = %d\n", - errno); + *(struct itimerval *)ptr = ((struct itimerval) { { 0, usec }, + { 0, usec }}); } void disable_timer(void) @@ -155,13 +157,15 @@ void idle_sleep(int secs) nanosleep(&ts, NULL); } -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ +/* XXX This partly duplicates init_irq_signals */ + +void user_time_init(void) +{ + set_handler(SIGVTALRM, (__sighandler_t) alarm_handler, + SA_ONSTACK | SA_RESTART, SIGUSR1, SIGIO, SIGWINCH, + SIGALRM, SIGUSR2, -1); + set_handler(SIGALRM, (__sighandler_t) alarm_handler, + SA_ONSTACK | SA_RESTART, SIGUSR1, SIGIO, SIGWINCH, + SIGVTALRM, SIGUSR2, -1); + set_interval(ITIMER_VIRTUAL); +} diff --git a/arch/um/kernel/time_kern.c b/arch/um/kernel/time_kern.c index 6516fc52afe..4e08f7545d6 100644 --- a/arch/um/kernel/time_kern.c +++ b/arch/um/kernel/time_kern.c @@ -137,7 +137,10 @@ long um_stime(int __user *tptr) void timer_handler(int sig, union uml_pt_regs *regs) { local_irq_disable(); - update_process_times(CHOOSE_MODE(user_context(UPT_SP(regs)), (regs)->skas.is_user)); + irq_enter(); + update_process_times(CHOOSE_MODE(user_context(UPT_SP(regs)), + (regs)->skas.is_user)); + irq_exit(); local_irq_enable(); if(current_thread->cpu == 0) timer_irq(regs); @@ -162,7 +165,7 @@ int __init timer_init(void) { int err; - CHOOSE_MODE(user_time_init_tt(), user_time_init_skas()); + user_time_init(); err = request_irq(TIMER_IRQ, um_timer, SA_INTERRUPT, "timer", NULL); if(err != 0) printk(KERN_ERR "timer_init : request_irq failed - " diff --git a/arch/um/kernel/tlb.c b/arch/um/kernel/tlb.c index eda477edfdf..83ec8d4747f 100644 --- a/arch/um/kernel/tlb.c +++ b/arch/um/kernel/tlb.c @@ -18,13 +18,15 @@ #define ADD_ROUND(n, inc) (((n) + (inc)) & ~((inc) - 1)) void fix_range_common(struct mm_struct *mm, unsigned long start_addr, - unsigned long end_addr, int force, int data, - void (*do_ops)(int, struct host_vm_op *, int)) + unsigned long end_addr, int force, + void (*do_ops)(union mm_context *, struct host_vm_op *, + int)) { pgd_t *npgd; pud_t *npud; pmd_t *npmd; pte_t *npte; + union mm_context *mmu = &mm->context; unsigned long addr, end; int r, w, x; struct host_vm_op ops[16]; @@ -40,7 +42,7 @@ void fix_range_common(struct mm_struct *mm, unsigned long start_addr, end = end_addr; if(force || pgd_newpage(*npgd)){ op_index = add_munmap(addr, end - addr, ops, - op_index, last_op, data, + op_index, last_op, mmu, do_ops); pgd_mkuptodate(*npgd); } @@ -55,7 +57,7 @@ void fix_range_common(struct mm_struct *mm, unsigned long start_addr, end = end_addr; if(force || pud_newpage(*npud)){ op_index = add_munmap(addr, end - addr, ops, - op_index, last_op, data, + op_index, last_op, mmu, do_ops); pud_mkuptodate(*npud); } @@ -70,7 +72,7 @@ void fix_range_common(struct mm_struct *mm, unsigned long start_addr, end = end_addr; if(force || pmd_newpage(*npmd)){ op_index = add_munmap(addr, end - addr, ops, - op_index, last_op, data, + op_index, last_op, mmu, do_ops); pmd_mkuptodate(*npmd); } @@ -93,21 +95,21 @@ void fix_range_common(struct mm_struct *mm, unsigned long start_addr, op_index = add_mmap(addr, pte_val(*npte) & PAGE_MASK, PAGE_SIZE, r, w, x, ops, - op_index, last_op, data, + op_index, last_op, mmu, do_ops); else op_index = add_munmap(addr, PAGE_SIZE, ops, - op_index, last_op, data, + op_index, last_op, mmu, do_ops); } else if(pte_newprot(*npte)) op_index = add_mprotect(addr, PAGE_SIZE, r, w, x, ops, - op_index, last_op, data, + op_index, last_op, mmu, do_ops); *npte = pte_mkuptodate(*npte); addr += PAGE_SIZE; } - (*do_ops)(data, ops, op_index); + (*do_ops)(mmu, ops, op_index); } int flush_tlb_kernel_range_common(unsigned long start, unsigned long end) @@ -195,51 +197,6 @@ int flush_tlb_kernel_range_common(unsigned long start, unsigned long end) return(updated); } -void flush_tlb_page(struct vm_area_struct *vma, unsigned long address) -{ - address &= PAGE_MASK; - flush_tlb_range(vma, address, address + PAGE_SIZE); -} - -void flush_tlb_all(void) -{ - flush_tlb_mm(current->mm); -} - -void flush_tlb_kernel_range(unsigned long start, unsigned long end) -{ - CHOOSE_MODE_PROC(flush_tlb_kernel_range_tt, - flush_tlb_kernel_range_common, start, end); -} - -void flush_tlb_kernel_vm(void) -{ - CHOOSE_MODE(flush_tlb_kernel_vm_tt(), - flush_tlb_kernel_range_common(start_vm, end_vm)); -} - -void __flush_tlb_one(unsigned long addr) -{ - CHOOSE_MODE_PROC(__flush_tlb_one_tt, __flush_tlb_one_skas, addr); -} - -void flush_tlb_range(struct vm_area_struct *vma, unsigned long start, - unsigned long end) -{ - CHOOSE_MODE_PROC(flush_tlb_range_tt, flush_tlb_range_skas, vma, start, - end); -} - -void flush_tlb_mm(struct mm_struct *mm) -{ - CHOOSE_MODE_PROC(flush_tlb_mm_tt, flush_tlb_mm_skas, mm); -} - -void force_flush_all(void) -{ - CHOOSE_MODE(force_flush_all_tt(), force_flush_all_skas()); -} - pgd_t *pgd_offset_proc(struct mm_struct *mm, unsigned long address) { return(pgd_offset(mm, address)); @@ -270,9 +227,9 @@ pte_t *addr_pte(struct task_struct *task, unsigned long addr) } int add_mmap(unsigned long virt, unsigned long phys, unsigned long len, - int r, int w, int x, struct host_vm_op *ops, int index, - int last_filled, int data, - void (*do_ops)(int, struct host_vm_op *, int)) + int r, int w, int x, struct host_vm_op *ops, int index, + int last_filled, union mm_context *mmu, + void (*do_ops)(union mm_context *, struct host_vm_op *, int)) { __u64 offset; struct host_vm_op *last; @@ -292,7 +249,7 @@ int add_mmap(unsigned long virt, unsigned long phys, unsigned long len, } if(index == last_filled){ - (*do_ops)(data, ops, last_filled); + (*do_ops)(mmu, ops, last_filled); index = -1; } @@ -310,8 +267,8 @@ int add_mmap(unsigned long virt, unsigned long phys, unsigned long len, } int add_munmap(unsigned long addr, unsigned long len, struct host_vm_op *ops, - int index, int last_filled, int data, - void (*do_ops)(int, struct host_vm_op *, int)) + int index, int last_filled, union mm_context *mmu, + void (*do_ops)(union mm_context *, struct host_vm_op *, int)) { struct host_vm_op *last; @@ -325,7 +282,7 @@ int add_munmap(unsigned long addr, unsigned long len, struct host_vm_op *ops, } if(index == last_filled){ - (*do_ops)(data, ops, last_filled); + (*do_ops)(mmu, ops, last_filled); index = -1; } @@ -337,8 +294,9 @@ int add_munmap(unsigned long addr, unsigned long len, struct host_vm_op *ops, } int add_mprotect(unsigned long addr, unsigned long len, int r, int w, int x, - struct host_vm_op *ops, int index, int last_filled, int data, - void (*do_ops)(int, struct host_vm_op *, int)) + struct host_vm_op *ops, int index, int last_filled, + union mm_context *mmu, + void (*do_ops)(union mm_context *, struct host_vm_op *, int)) { struct host_vm_op *last; @@ -354,7 +312,7 @@ int add_mprotect(unsigned long addr, unsigned long len, int r, int w, int x, } if(index == last_filled){ - (*do_ops)(data, ops, last_filled); + (*do_ops)(mmu, ops, last_filled); index = -1; } @@ -367,3 +325,49 @@ int add_mprotect(unsigned long addr, unsigned long len, int r, int w, int x, .x = x } } }); return(index); } + +void flush_tlb_page(struct vm_area_struct *vma, unsigned long address) +{ + address &= PAGE_MASK; + flush_tlb_range(vma, address, address + PAGE_SIZE); +} + +void flush_tlb_all(void) +{ + flush_tlb_mm(current->mm); +} + +void flush_tlb_kernel_range(unsigned long start, unsigned long end) +{ + CHOOSE_MODE_PROC(flush_tlb_kernel_range_tt, + flush_tlb_kernel_range_common, start, end); +} + +void flush_tlb_kernel_vm(void) +{ + CHOOSE_MODE(flush_tlb_kernel_vm_tt(), + flush_tlb_kernel_range_common(start_vm, end_vm)); +} + +void __flush_tlb_one(unsigned long addr) +{ + CHOOSE_MODE_PROC(__flush_tlb_one_tt, __flush_tlb_one_skas, addr); +} + +void flush_tlb_range(struct vm_area_struct *vma, unsigned long start, + unsigned long end) +{ + CHOOSE_MODE_PROC(flush_tlb_range_tt, flush_tlb_range_skas, vma, start, + end); +} + +void flush_tlb_mm(struct mm_struct *mm) +{ + CHOOSE_MODE_PROC(flush_tlb_mm_tt, flush_tlb_mm_skas, mm); +} + +void force_flush_all(void) +{ + CHOOSE_MODE(force_flush_all_tt(), force_flush_all_skas()); +} + diff --git a/arch/um/kernel/tt/Makefile b/arch/um/kernel/tt/Makefile index c3faea21a99..6939e5af847 100644 --- a/arch/um/kernel/tt/Makefile +++ b/arch/um/kernel/tt/Makefile @@ -3,27 +3,12 @@ # Licensed under the GPL # -extra-y := unmap_fin.o -targets := unmap.o -clean-files := unmap_tmp.o - obj-y = exec_kern.o exec_user.o gdb.o ksyms.o mem.o mem_user.o process_kern.o \ - syscall_kern.o syscall_user.o time.o tlb.o tracer.o trap_user.o \ + syscall_kern.o syscall_user.o tlb.o tracer.o trap_user.o \ uaccess.o uaccess_user.o obj-$(CONFIG_PT_PROXY) += gdb_kern.o ptproxy/ -USER_OBJS := gdb.o time.o tracer.o +USER_OBJS := gdb.o tracer.o include arch/um/scripts/Makefile.rules - -UNMAP_CFLAGS := $(patsubst -pg -DPROFILING,,$(USER_CFLAGS)) -UNMAP_CFLAGS := $(patsubst -fprofile-arcs -ftest-coverage,,$(UNMAP_CFLAGS)) - -#XXX: partially copied from arch/um/scripts/Makefile.rules -$(obj)/unmap.o: c_flags = -Wp,-MD,$(depfile) $(UNMAP_CFLAGS) - -$(obj)/unmap_fin.o : $(obj)/unmap.o - $(LD) -r -o $(obj)/unmap_tmp.o $< $(shell $(CC) -print-file-name=libc.a) - $(OBJCOPY) $(obj)/unmap_tmp.o $@ -G switcheroo - diff --git a/arch/um/kernel/tt/gdb.c b/arch/um/kernel/tt/gdb.c index 19a0ad7b35b..37e22d71a0d 100644 --- a/arch/um/kernel/tt/gdb.c +++ b/arch/um/kernel/tt/gdb.c @@ -153,10 +153,10 @@ void remove_gdb_cb(void *unused) exit_debugger_cb(NULL); } -int gdb_remove(char *unused) +int gdb_remove(int unused) { initial_thread_cb(remove_gdb_cb, NULL); - return(0); + return 0; } void signal_usr1(int sig) diff --git a/arch/um/kernel/tt/gdb_kern.c b/arch/um/kernel/tt/gdb_kern.c index 93fb121f86a..26506388a6a 100644 --- a/arch/um/kernel/tt/gdb_kern.c +++ b/arch/um/kernel/tt/gdb_kern.c @@ -10,7 +10,7 @@ #ifdef CONFIG_MCONSOLE extern int gdb_config(char *str); -extern int gdb_remove(char *unused); +extern int gdb_remove(int n); static struct mc_device gdb_mc = { .name = "gdb", diff --git a/arch/um/kernel/tt/include/debug.h b/arch/um/kernel/tt/include/debug.h index 8eff674107c..738435461e1 100644 --- a/arch/um/kernel/tt/include/debug.h +++ b/arch/um/kernel/tt/include/debug.h @@ -4,8 +4,8 @@ * Licensed under the GPL */ -#ifndef __DEBUG_H -#define __DEBUG_H +#ifndef __UML_TT_DEBUG_H +#define __UML_TT_DEBUG_H extern int debugger_proxy(int status, pid_t pid); extern void child_proxy(pid_t pid, int status); @@ -13,17 +13,6 @@ extern void init_proxy (pid_t pid, int waiting, int status); extern int start_debugger(char *prog, int startup, int stop, int *debugger_fd); extern void fake_child_exit(void); extern int gdb_config(char *str); -extern int gdb_remove(char *unused); +extern int gdb_remove(int unused); #endif - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/kernel/tt/include/mode-tt.h b/arch/um/kernel/tt/include/mode-tt.h index efe46201906..e171e15fead 100644 --- a/arch/um/kernel/tt/include/mode-tt.h +++ b/arch/um/kernel/tt/include/mode-tt.h @@ -13,7 +13,6 @@ enum { OP_NONE, OP_EXEC, OP_FORK, OP_TRACE_ON, OP_REBOOT, OP_HALT, OP_CB }; extern int tracing_pid; extern int tracer(int (*init_proc)(void *), void *sp); -extern void user_time_init_tt(void); extern void sig_handler_common_tt(int sig, void *sc); extern void syscall_handler_tt(int sig, union uml_pt_regs *regs); extern void reboot_tt(void); diff --git a/arch/um/kernel/tt/process_kern.c b/arch/um/kernel/tt/process_kern.c index 776310fd5b8..a189a2b9293 100644 --- a/arch/um/kernel/tt/process_kern.c +++ b/arch/um/kernel/tt/process_kern.c @@ -266,10 +266,10 @@ int copy_thread_tt(int nr, unsigned long clone_flags, unsigned long sp, } if(current->thread.forking){ - sc_to_sc(UPT_SC(&p->thread.regs.regs), - UPT_SC(¤t->thread.regs.regs)); + sc_to_sc(UPT_SC(&p->thread.regs.regs), UPT_SC(®s->regs)); SC_SET_SYSCALL_RETURN(UPT_SC(&p->thread.regs.regs), 0); - if(sp != 0) SC_SP(UPT_SC(&p->thread.regs.regs)) = sp; + if(sp != 0) + SC_SP(UPT_SC(&p->thread.regs.regs)) = sp; } p->thread.mode.tt.extern_pid = new_pid; @@ -459,14 +459,3 @@ int is_valid_pid(int pid) read_unlock(&tasklist_lock); return(0); } - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/kernel/tt/time.c b/arch/um/kernel/tt/time.c deleted file mode 100644 index 8565b71b07c..00000000000 --- a/arch/um/kernel/tt/time.c +++ /dev/null @@ -1,28 +0,0 @@ -/* - * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#include <signal.h> -#include <sys/time.h> -#include <time_user.h> -#include "process.h" -#include "user.h" - -void user_time_init_tt(void) -{ - if(signal(SIGVTALRM, (__sighandler_t) alarm_handler) == SIG_ERR) - panic("Couldn't set SIGVTALRM handler"); - set_interval(ITIMER_VIRTUAL); -} - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/kernel/tt/tlb.c b/arch/um/kernel/tt/tlb.c index 203216ad86f..2eefb43bc9c 100644 --- a/arch/um/kernel/tt/tlb.c +++ b/arch/um/kernel/tt/tlb.c @@ -17,7 +17,7 @@ #include "os.h" #include "tlb.h" -static void do_ops(int unused, struct host_vm_op *ops, int last) +static void do_ops(union mm_context *mmu, struct host_vm_op *ops, int last) { struct host_vm_op *op; int i; @@ -55,7 +55,7 @@ static void fix_range(struct mm_struct *mm, unsigned long start_addr, panic("fix_range fixing wrong address space, current = 0x%p", current); - fix_range_common(mm, start_addr, end_addr, force, 0, do_ops); + fix_range_common(mm, start_addr, end_addr, force, do_ops); } atomic_t vmchange_seq = ATOMIC_INIT(1); diff --git a/arch/um/kernel/tt/unmap.c b/arch/um/kernel/tt/unmap.c deleted file mode 100644 index 3f7aecdbe53..00000000000 --- a/arch/um/kernel/tt/unmap.c +++ /dev/null @@ -1,31 +0,0 @@ -/* - * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#include <sys/mman.h> - -int switcheroo(int fd, int prot, void *from, void *to, int size) -{ - if(munmap(to, size) < 0){ - return(-1); - } - if(mmap(to, size, prot, MAP_SHARED | MAP_FIXED, fd, 0) != to){ - return(-1); - } - if(munmap(from, size) < 0){ - return(-1); - } - return(0); -} - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c index 8736d098f0e..ca2bb6f09a7 100644 --- a/arch/um/kernel/um_arch.c +++ b/arch/um/kernel/um_arch.c @@ -38,6 +38,9 @@ #include "choose-mode.h" #include "mode_kern.h" #include "mode.h" +#ifdef UML_CONFIG_MODE_SKAS +#include "skas.h" +#endif #define DEFAULT_COMMAND_LINE "root=98:0" @@ -318,6 +321,7 @@ int linux_main(int argc, char **argv) unsigned long avail, diff; unsigned long virtmem_size, max_physmem; unsigned int i, add; + char * mode; for (i = 1; i < argc; i++){ if((i == 1) && (argv[i][0] == ' ')) continue; @@ -338,6 +342,21 @@ int linux_main(int argc, char **argv) exit(1); } #endif + +#ifndef CONFIG_MODE_SKAS + mode = "TT"; +#else + /* Show to the user the result of selection */ + if (mode_tt) + mode = "TT"; + else if (proc_mm && ptrace_faultinfo) + mode = "SKAS3"; + else + mode = "SKAS0"; +#endif + + printf("UML running in %s mode\n", mode); + uml_start = CHOOSE_MODE_PROC(set_task_sizes_tt, set_task_sizes_skas, 0, &host_task_size, &task_size); diff --git a/arch/um/kernel/uml.lds.S b/arch/um/kernel/uml.lds.S index dd5355500bd..b03326d391c 100644 --- a/arch/um/kernel/uml.lds.S +++ b/arch/um/kernel/uml.lds.S @@ -14,19 +14,10 @@ SECTIONS /* Used in arch/um/kernel/mem.c. Any memory between START and __binary_start * is remapped.*/ __binary_start = .; -#ifdef MODE_TT - .thread_private : { - __start_thread_private = .; - errno = .; - . += 4; - arch/um/kernel/tt/unmap_fin.o (.data) - __end_thread_private = .; - } - . = ALIGN(4096); - .remap : { arch/um/kernel/tt/unmap_fin.o (.text) } - /* We want it only if we are in MODE_TT. In both cases, however, when MODE_TT - * is off the resulting binary segfaults.*/ +#ifdef MODE_TT + .remap_data : { UNMAP_PATH (.data .bss) } + .remap : { UNMAP_PATH (.text) } . = ALIGN(4096); /* Init code and data */ #endif @@ -39,6 +30,7 @@ SECTIONS _einittext = .; } . = ALIGN(4096); + .text : { *(.text) @@ -48,6 +40,12 @@ SECTIONS /* .gnu.warning sections are handled specially by elf32.em. */ *(.gnu.warning) *(.gnu.linkonce.t*) + + . = ALIGN(4096); + __syscall_stub_start = .; + *(.__syscall_stub*) + __syscall_stub_end = .; + . = ALIGN(4096); } #include "asm/common.lds.S" diff --git a/arch/um/os-Linux/elf_aux.c b/arch/um/os-Linux/elf_aux.c index f0d6060e3e5..9416e1c2992 100644 --- a/arch/um/os-Linux/elf_aux.c +++ b/arch/um/os-Linux/elf_aux.c @@ -9,8 +9,10 @@ */ #include <elf.h> #include <stddef.h> +#include <asm/elf.h> #include "init.h" #include "elf_user.h" +#include "mem_user.h" #if ELF_CLASS == ELFCLASS32 typedef Elf32_auxv_t elf_auxv_t; @@ -40,6 +42,9 @@ __init void scan_elf_aux( char **envp) break; case AT_SYSINFO_EHDR: vsyscall_ehdr = auxv->a_un.a_val; + /* See if the page is under TASK_SIZE */ + if (vsyscall_ehdr < (unsigned long) envp) + vsyscall_ehdr = 0; break; case AT_HWCAP: elf_aux_hwcap = auxv->a_un.a_val; diff --git a/arch/um/os-Linux/sys-i386/registers.c b/arch/um/os-Linux/sys-i386/registers.c index 9a0ad094d92..3125d320722 100644 --- a/arch/um/os-Linux/sys-i386/registers.c +++ b/arch/um/os-Linux/sys-i386/registers.c @@ -121,6 +121,11 @@ void init_registers(int pid) err); } +void get_safe_registers(unsigned long *regs) +{ + memcpy(regs, exec_regs, HOST_FRAME_SIZE * sizeof(unsigned long)); +} + /* * Overrides for Emacs so that we follow Linus's tabbing style. * Emacs will notice this stuff at the end of the file and automatically diff --git a/arch/um/os-Linux/sys-x86_64/registers.c b/arch/um/os-Linux/sys-x86_64/registers.c index 6286c974bbe..44438d15c3d 100644 --- a/arch/um/os-Linux/sys-x86_64/registers.c +++ b/arch/um/os-Linux/sys-x86_64/registers.c @@ -69,6 +69,11 @@ void init_registers(int pid) err); } +void get_safe_registers(unsigned long *regs) +{ + memcpy(regs, exec_regs, HOST_FRAME_SIZE * sizeof(unsigned long)); +} + /* * Overrides for Emacs so that we follow Linus's tabbing style. * Emacs will notice this stuff at the end of the file and automatically diff --git a/arch/um/os-Linux/user_syms.c b/arch/um/os-Linux/user_syms.c index 75d7af9ae1d..56d3f870926 100644 --- a/arch/um/os-Linux/user_syms.c +++ b/arch/um/os-Linux/user_syms.c @@ -83,6 +83,9 @@ EXPORT_SYMBOL_PROTO(statfs64); EXPORT_SYMBOL_PROTO(getuid); +EXPORT_SYMBOL_PROTO(fsync); +EXPORT_SYMBOL_PROTO(fdatasync); + /* * Overrides for Emacs so that we follow Linus's tabbing style. * Emacs will notice this stuff at the end of the file and automatically diff --git a/arch/um/scripts/Makefile.rules b/arch/um/scripts/Makefile.rules index 98346c71149..17f305b6bad 100644 --- a/arch/um/scripts/Makefile.rules +++ b/arch/um/scripts/Makefile.rules @@ -10,6 +10,17 @@ USER_OBJS := $(foreach file,$(USER_OBJS),$(obj)/$(file)) $(USER_OBJS) : c_flags = -Wp,-MD,$(depfile) $(USER_CFLAGS) \ $(CFLAGS_$(notdir $@)) +# The stubs and unmap.o can't try to call mcount or update basic block data +define unprofile + $(patsubst -pg,,$(patsubst -fprofile-arcs -ftest-coverage,,$(1))) +endef + + +# The stubs and unmap.o can't try to call mcount or update basic block data +define unprofile + $(patsubst -pg,,$(patsubst -fprofile-arcs -ftest-coverage,,$(1))) +endef + quiet_cmd_make_link = SYMLINK $@ cmd_make_link = ln -sf $(srctree)/arch/$(SUBARCH)/$($(notdir $@)-dir)/$(notdir $@) $@ diff --git a/arch/um/scripts/Makefile.unmap b/arch/um/scripts/Makefile.unmap new file mode 100644 index 00000000000..802d027a1e1 --- /dev/null +++ b/arch/um/scripts/Makefile.unmap @@ -0,0 +1,22 @@ +clean-files += unmap_tmp.o unmap_fin.o unmap.o + +ifdef CONFIG_MODE_TT + +#Always build unmap_fin.o +extra-y += unmap_fin.o +#Do dependency tracking for unmap.o (it will be always built, but won't get the tracking unless we use this). +targets += unmap.o + +#XXX: partially copied from arch/um/scripts/Makefile.rules +$(obj)/unmap.o: _c_flags = $(call unprofile,$(CFLAGS)) + +quiet_cmd_wrapld = LD $@ +define cmd_wrapld + $(LD) $(LDFLAGS) -r -o $(obj)/unmap_tmp.o $< $(shell $(CC) $(CFLAGS) -print-file-name=libc.a); \ + $(OBJCOPY) $(UML_OBJCOPYFLAGS) $(obj)/unmap_tmp.o $@ -G switcheroo +endef + +$(obj)/unmap_fin.o : $(obj)/unmap.o FORCE + $(call if_changed,wrapld) + +endif diff --git a/arch/um/sys-i386/Makefile b/arch/um/sys-i386/Makefile index 4351e560550..77c3c4d29f5 100644 --- a/arch/um/sys-i386/Makefile +++ b/arch/um/sys-i386/Makefile @@ -1,6 +1,6 @@ obj-y = bitops.o bugs.o checksum.o delay.o fault.o ksyms.o ldt.o ptrace.o \ - ptrace_user.o semaphore.o signal.o sigcontext.o syscalls.o sysrq.o \ - sys_call_table.o + ptrace_user.o semaphore.o signal.o sigcontext.o stub.o stub_segv.o \ + syscalls.o sysrq.o sys_call_table.o obj-$(CONFIG_HIGHMEM) += highmem.o obj-$(CONFIG_MODULES) += module.o @@ -16,4 +16,14 @@ semaphore.c-dir = kernel highmem.c-dir = mm module.c-dir = kernel +STUB_CFLAGS = -Wp,-MD,$(depfile) $(call unprofile,$(USER_CFLAGS)) + +# _cflags works with kernel files, not with userspace ones, but c_flags does, +# why ask why? +$(obj)/stub_segv.o : c_flags = $(STUB_CFLAGS) + +$(obj)/stub.o : a_flags = $(STUB_CFLAGS) + subdir- := util + +include arch/um/scripts/Makefile.unmap diff --git a/arch/um/sys-i386/ldt.c b/arch/um/sys-i386/ldt.c index dc755b0b9db..bd3c34aa52e 100644 --- a/arch/um/sys-i386/ldt.c +++ b/arch/um/sys-i386/ldt.c @@ -4,96 +4,106 @@ */ #include "linux/config.h" +#include "linux/sched.h" #include "linux/slab.h" +#include "linux/types.h" #include "asm/uaccess.h" #include "asm/ptrace.h" +#include "asm/smp.h" +#include "asm/ldt.h" #include "choose-mode.h" #include "kern.h" +#include "mode_kern.h" #ifdef CONFIG_MODE_TT -extern int modify_ldt(int func, void *ptr, unsigned long bytecount); -/* XXX this needs copy_to_user and copy_from_user */ +extern int modify_ldt(int func, void *ptr, unsigned long bytecount); -int sys_modify_ldt_tt(int func, void __user *ptr, unsigned long bytecount) +static int do_modify_ldt_tt(int func, void *ptr, unsigned long bytecount) { - if (!access_ok(VERIFY_READ, ptr, bytecount)) - return -EFAULT; - return modify_ldt(func, ptr, bytecount); } + #endif #ifdef CONFIG_MODE_SKAS -extern int userspace_pid[]; +#include "skas.h" #include "skas_ptrace.h" -int sys_modify_ldt_skas(int func, void __user *ptr, unsigned long bytecount) +static int do_modify_ldt_skas(int func, void *ptr, unsigned long bytecount) { struct ptrace_ldt ldt; - void *buf; - int res, n; + u32 cpu; + int res; - buf = kmalloc(bytecount, GFP_KERNEL); - if(buf == NULL) - return(-ENOMEM); + ldt = ((struct ptrace_ldt) { .func = func, + .ptr = ptr, + .bytecount = bytecount }); - res = 0; + cpu = get_cpu(); + res = ptrace(PTRACE_LDT, userspace_pid[cpu], 0, (unsigned long) &ldt); + put_cpu(); + + return res; +} +#endif + +int sys_modify_ldt(int func, void __user *ptr, unsigned long bytecount) +{ + struct user_desc info; + int res = 0; + void *buf = NULL; + void *p = NULL; /* What we pass to host. */ switch(func){ case 1: - case 0x11: - res = copy_from_user(buf, ptr, bytecount); - break; - } + case 0x11: /* write_ldt */ + /* Do this check now to avoid overflows. */ + if (bytecount != sizeof(struct user_desc)) { + res = -EINVAL; + goto out; + } + + if(copy_from_user(&info, ptr, sizeof(info))) { + res = -EFAULT; + goto out; + } - if(res != 0){ - res = -EFAULT; + p = &info; + break; + case 0: + case 2: /* read_ldt */ + + /* The use of info avoids kmalloc on the write case, not on the + * read one. */ + buf = kmalloc(bytecount, GFP_KERNEL); + if (!buf) { + res = -ENOMEM; + goto out; + } + p = buf; + default: + res = -ENOSYS; goto out; } - ldt = ((struct ptrace_ldt) { .func = func, - .ptr = buf, - .bytecount = bytecount }); -#warning Need to look up userspace_pid by cpu - res = ptrace(PTRACE_LDT, userspace_pid[0], 0, (unsigned long) &ldt); + res = CHOOSE_MODE_PROC(do_modify_ldt_tt, do_modify_ldt_skas, func, + p, bytecount); if(res < 0) goto out; switch(func){ case 0: case 2: - n = res; - res = copy_to_user(ptr, buf, n); - if(res != 0) + /* Modify_ldt was for reading and returned the number of read + * bytes.*/ + if(copy_to_user(ptr, p, res)) res = -EFAULT; - else - res = n; break; } - out: +out: kfree(buf); - return(res); -} -#endif - -int sys_modify_ldt(int func, void __user *ptr, unsigned long bytecount) -{ - return(CHOOSE_MODE_PROC(sys_modify_ldt_tt, sys_modify_ldt_skas, func, - ptr, bytecount)); + return res; } - - - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/sys-i386/signal.c b/arch/um/sys-i386/signal.c index 03913ca5d25..4efc69a039d 100644 --- a/arch/um/sys-i386/signal.c +++ b/arch/um/sys-i386/signal.c @@ -312,7 +312,7 @@ long sys_sigreturn(struct pt_regs regs) unsigned long __user *extramask = frame->extramask; int sig_size = (_NSIG_WORDS - 1) * sizeof(unsigned long); - if(copy_from_user(&set.sig[0], oldmask, sizeof(&set.sig[0])) || + if(copy_from_user(&set.sig[0], oldmask, sizeof(set.sig[0])) || copy_from_user(&set.sig[1], extramask, sig_size)) goto segfault; diff --git a/arch/um/sys-i386/stub.S b/arch/um/sys-i386/stub.S new file mode 100644 index 00000000000..2f2c70a8f04 --- /dev/null +++ b/arch/um/sys-i386/stub.S @@ -0,0 +1,8 @@ +#include "uml-config.h" + + .globl syscall_stub +.section .__syscall_stub, "x" +syscall_stub: + int $0x80 + mov %eax, UML_CONFIG_STUB_DATA + int3 diff --git a/arch/um/sys-i386/stub_segv.c b/arch/um/sys-i386/stub_segv.c new file mode 100644 index 00000000000..68aeabe3a65 --- /dev/null +++ b/arch/um/sys-i386/stub_segv.c @@ -0,0 +1,30 @@ +/* + * Copyright (C) 2004 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include <signal.h> +#include <asm/sigcontext.h> +#include <asm/unistd.h> +#include "uml-config.h" +#include "sysdep/sigcontext.h" +#include "sysdep/faultinfo.h" + +void __attribute__ ((__section__ (".__syscall_stub"))) +stub_segv_handler(int sig) +{ + struct sigcontext *sc = (struct sigcontext *) (&sig + 1); + + GET_FAULTINFO_FROM_SC(*((struct faultinfo *) UML_CONFIG_STUB_DATA), + sc); + + __asm__("movl %0, %%eax ; int $0x80": : "g" (__NR_getpid)); + __asm__("movl %%eax, %%ebx ; movl %0, %%eax ; movl %1, %%ecx ;" + "int $0x80": : "g" (__NR_kill), "g" (SIGUSR1)); + /* Load pointer to sigcontext into esp, since we need to leave + * the stack in its original form when we do the sigreturn here, by + * hand. + */ + __asm__("mov %0,%%esp ; movl %1, %%eax ; " + "int $0x80" : : "a" (sc), "g" (__NR_sigreturn)); +} diff --git a/arch/um/sys-i386/syscalls.c b/arch/um/sys-i386/syscalls.c index 335e2d89504..83e9be820a8 100644 --- a/arch/um/sys-i386/syscalls.c +++ b/arch/um/sys-i386/syscalls.c @@ -69,15 +69,11 @@ long sys_clone(unsigned long clone_flags, unsigned long newsp, { long ret; - /* XXX: normal arch do here this pass, and also pass the regs to - * do_fork, instead of NULL. Currently the arch-independent code - * ignores these values, while the UML code (actually it's - * copy_thread) does the right thing. But this should change, - probably. */ - /*if (!newsp) - newsp = UPT_SP(current->thread.regs);*/ + if (!newsp) + newsp = UPT_SP(¤t->thread.regs.regs); current->thread.forking = 1; - ret = do_fork(clone_flags, newsp, NULL, 0, parent_tid, child_tid); + ret = do_fork(clone_flags, newsp, ¤t->thread.regs, 0, parent_tid, + child_tid); current->thread.forking = 0; return(ret); } @@ -197,14 +193,3 @@ long sys_sigaction(int sig, const struct old_sigaction __user *act, return ret; } - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/sys-i386/unmap.c b/arch/um/sys-i386/unmap.c new file mode 100644 index 00000000000..1b0ad0e4adc --- /dev/null +++ b/arch/um/sys-i386/unmap.c @@ -0,0 +1,25 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <linux/mman.h> +#include <asm/unistd.h> + +static int errno; + +static inline _syscall2(int,munmap,void *,start,size_t,len) +static inline _syscall6(void *,mmap2,void *,addr,size_t,len,int,prot,int,flags,int,fd,off_t,offset) +int switcheroo(int fd, int prot, void *from, void *to, int size) +{ + if(munmap(to, size) < 0){ + return(-1); + } + if(mmap2(to, size, prot, MAP_SHARED | MAP_FIXED, fd, 0) == (void*) -1 ){ + return(-1); + } + if(munmap(from, size) < 0){ + return(-1); + } + return(0); +} diff --git a/arch/um/sys-x86_64/Makefile b/arch/um/sys-x86_64/Makefile index 608466ad6b2..7488206ce6f 100644 --- a/arch/um/sys-x86_64/Makefile +++ b/arch/um/sys-x86_64/Makefile @@ -6,8 +6,8 @@ #XXX: why into lib-y? lib-y = bitops.o bugs.o csum-partial.o delay.o fault.o mem.o memcpy.o \ - ptrace.o ptrace_user.o semaphore.o sigcontext.o signal.o \ - syscalls.o sysrq.o thunk.o syscall_table.o + ptrace.o ptrace_user.o semaphore.o sigcontext.o signal.o stub.o \ + stub_segv.o syscalls.o syscall_table.o sysrq.o thunk.o obj-y := ksyms.o obj-$(CONFIG_MODULES) += module.o um_module.o @@ -28,4 +28,14 @@ semaphore.c-dir = kernel thunk.S-dir = lib module.c-dir = kernel +STUB_CFLAGS = -Wp,-MD,$(depfile) $(call unprofile,$(USER_CFLAGS)) + +# _cflags works with kernel files, not with userspace ones, but c_flags does, +# why ask why? +$(obj)/stub_segv.o : c_flags = $(STUB_CFLAGS) + +$(obj)/stub.o : a_flags = $(STUB_CFLAGS) + subdir- := util + +include arch/um/scripts/Makefile.unmap diff --git a/arch/um/sys-x86_64/signal.c b/arch/um/sys-x86_64/signal.c index 73a7926f737..8fdaed06c10 100644 --- a/arch/um/sys-x86_64/signal.c +++ b/arch/um/sys-x86_64/signal.c @@ -168,7 +168,7 @@ int setup_signal_stack_si(unsigned long stack_top, int sig, frame = (struct rt_sigframe __user *) round_down(stack_top - sizeof(struct rt_sigframe), 16) - 8; - ((unsigned char *) frame) -= 128; + frame = (struct rt_sigframe *) ((unsigned long) frame - 128); if (!access_ok(VERIFY_WRITE, fp, sizeof(struct _fpstate))) goto out; diff --git a/arch/um/sys-x86_64/stub.S b/arch/um/sys-x86_64/stub.S new file mode 100644 index 00000000000..31c14925716 --- /dev/null +++ b/arch/um/sys-x86_64/stub.S @@ -0,0 +1,15 @@ +#include "uml-config.h" + + .globl syscall_stub +.section .__syscall_stub, "x" +syscall_stub: + syscall + /* We don't have 64-bit constants, so this constructs the address + * we need. + */ + movq $(UML_CONFIG_STUB_DATA >> 32), %rbx + salq $32, %rbx + movq $(UML_CONFIG_STUB_DATA & 0xffffffff), %rcx + or %rcx, %rbx + movq %rax, (%rbx) + int3 diff --git a/arch/um/sys-x86_64/stub_segv.c b/arch/um/sys-x86_64/stub_segv.c new file mode 100644 index 00000000000..161d1fe9c03 --- /dev/null +++ b/arch/um/sys-x86_64/stub_segv.c @@ -0,0 +1,31 @@ +/* + * Copyright (C) 2004 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include <signal.h> +#include <linux/compiler.h> +#include <asm/unistd.h> +#include "uml-config.h" +#include "sysdep/sigcontext.h" +#include "sysdep/faultinfo.h" + +void __attribute__ ((__section__ (".__syscall_stub"))) +stub_segv_handler(int sig) +{ + struct ucontext *uc; + + __asm__("movq %%rdx, %0" : "=g" (uc) :); + GET_FAULTINFO_FROM_SC(*((struct faultinfo *) UML_CONFIG_STUB_DATA), + &uc->uc_mcontext); + + __asm__("movq %0, %%rax ; syscall": : "g" (__NR_getpid)); + __asm__("movq %%rax, %%rdi ; movq %0, %%rax ; movq %1, %%rsi ;" + "syscall": : "g" (__NR_kill), "g" (SIGUSR1)); + /* Two popqs to restore the stack to the state just before entering + * the handler, one pops the return address, the other pops the frame + * pointer. + */ + __asm__("popq %%rax ; popq %%rax ; movq %0, %%rax ; syscall" : : "g" + (__NR_rt_sigreturn)); +} diff --git a/arch/um/sys-x86_64/syscalls.c b/arch/um/sys-x86_64/syscalls.c index 6f44f40204e..3259a4db453 100644 --- a/arch/um/sys-x86_64/syscalls.c +++ b/arch/um/sys-x86_64/syscalls.c @@ -174,26 +174,11 @@ long sys_clone(unsigned long clone_flags, unsigned long newsp, { long ret; - /* XXX: normal arch do here this pass, and also pass the regs to - * do_fork, instead of NULL. Currently the arch-independent code - * ignores these values, while the UML code (actually it's - * copy_thread) does the right thing. But this should change, - probably. */ - /*if (!newsp) - newsp = UPT_SP(current->thread.regs);*/ + if (!newsp) + newsp = UPT_SP(¤t->thread.regs.regs); current->thread.forking = 1; - ret = do_fork(clone_flags, newsp, NULL, 0, parent_tid, child_tid); + ret = do_fork(clone_flags, newsp, ¤t->thread.regs, 0, parent_tid, + child_tid); current->thread.forking = 0; return(ret); } - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/sys-x86_64/unmap.c b/arch/um/sys-x86_64/unmap.c new file mode 100644 index 00000000000..f4a4bffd8a1 --- /dev/null +++ b/arch/um/sys-x86_64/unmap.c @@ -0,0 +1,25 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <linux/mman.h> +#include <asm/unistd.h> + +static int errno; + +static inline _syscall2(int,munmap,void *,start,size_t,len) +static inline _syscall6(void *,mmap,void *,addr,size_t,len,int,prot,int,flags,int,fd,off_t,offset) +int switcheroo(int fd, int prot, void *from, void *to, int size) +{ + if(munmap(to, size) < 0){ + return(-1); + } + if(mmap(to, size, prot, MAP_SHARED | MAP_FIXED, fd, 0) == (void*) -1){ + return(-1); + } + if(munmap(from, size) < 0){ + return(-1); + } + return(0); +} |