From 70524490ee2ea1bbf6cee6c106597b3ac25a3fc2 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 11 Apr 2006 15:51:17 +0200 Subject: [PATCH] splice: add support for sys_tee() Basically an in-kernel implementation of tee, which uses splice and the pipe buffers as an intelligent way to pass data around by reference. Where the user space tee consumes the input and produces a stdout and file output, this syscall merely duplicates the data inside a pipe to another pipe. No data is copied, the output just grabs a reference to the input pipe data. Signed-off-by: Jens Axboe --- include/asm-ia64/unistd.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/asm-ia64') diff --git a/include/asm-ia64/unistd.h b/include/asm-ia64/unistd.h index 1c749acca02..a40ebec6aee 100644 --- a/include/asm-ia64/unistd.h +++ b/include/asm-ia64/unistd.h @@ -289,12 +289,13 @@ #define __NR_set_robust_list 1298 #define __NR_get_robust_list 1299 #define __NR_sync_file_range 1300 +#define __NR_tee 1301 #ifdef __KERNEL__ #include -#define NR_syscalls 277 /* length of syscall table */ +#define NR_syscalls 278 /* length of syscall table */ #define __ARCH_WANT_SYS_RT_SIGACTION -- cgit v1.2.3-70-g09d2 From 308a878210cde6ab19df9f392c24db53ad6f56bf Mon Sep 17 00:00:00 2001 From: Russ Anderson Date: Tue, 18 Apr 2006 11:26:34 -0500 Subject: [IA64] Remove unused variable in sn_sal.h cnodeid was being set but not used. The dead code was left over from a previous version that grabbed a per node lock. Signed-off-by: Russ Anderson (rja@sgi.com) Signed-off-by: Tony Luck --- include/asm-ia64/sn/sn_sal.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'include/asm-ia64') diff --git a/include/asm-ia64/sn/sn_sal.h b/include/asm-ia64/sn/sn_sal.h index bf4cc867a69..7ddce80e191 100644 --- a/include/asm-ia64/sn/sn_sal.h +++ b/include/asm-ia64/sn/sn_sal.h @@ -8,7 +8,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All rights reserved. + * Copyright (c) 2000-2006 Silicon Graphics, Inc. All rights reserved. */ @@ -705,10 +705,8 @@ static inline int sn_change_memprotect(u64 paddr, u64 len, u64 perms, u64 *nasid_array) { struct ia64_sal_retval ret_stuff; - int cnodeid; unsigned long irq_flags; - cnodeid = nasid_to_cnodeid(get_node_number(paddr)); local_irq_save(irq_flags); ia64_sal_oemcall_nolock(&ret_stuff, SN_SAL_MEMPROTECT, paddr, len, (u64)nasid_array, perms, 0, 0, 0); -- cgit v1.2.3-70-g09d2 From 0d9adec525b87d8ab7e64efeabffb5b3f293056e Mon Sep 17 00:00:00 2001 From: Jack Steiner Date: Tue, 18 Apr 2006 15:00:45 -0500 Subject: [IA64] - Fix MAX_PXM_DOMAINS for systems with > 256 nodes Correctly size the PXM-related arrays for systems that have more than 256 nodes. Signed-off-by: Jack Steiner Signed-off-by: Tony Luck --- include/asm-ia64/acpi.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'include/asm-ia64') diff --git a/include/asm-ia64/acpi.h b/include/asm-ia64/acpi.h index d734585a23c..09a5dd0e44a 100644 --- a/include/asm-ia64/acpi.h +++ b/include/asm-ia64/acpi.h @@ -110,9 +110,8 @@ extern void prefill_possible_map(void); extern int additional_cpus; #ifdef CONFIG_ACPI_NUMA -/* Proximity bitmap length; _PXM is at most 255 (8 bit)*/ -#ifdef CONFIG_IA64_NR_NODES -#define MAX_PXM_DOMAINS CONFIG_IA64_NR_NODES +#if MAX_NUMNODES > 256 +#define MAX_PXM_DOMAINS MAX_NUMNODES #else #define MAX_PXM_DOMAINS (256) #endif -- cgit v1.2.3-70-g09d2 From 86db2f4239e2556cd37b853c2307aa9d43041458 Mon Sep 17 00:00:00 2001 From: Russ Anderson Date: Thu, 20 Apr 2006 17:05:43 -0700 Subject: [IA64-SGI] SN SAL call to inject memory errors The SGI Altix SAL provides an interface for modifying the ECC on memory to create memory errors. The SAL call can be used to inject memory errors for testing MCA recovery code. Signed-off-by: Russ Anderson (rja@sgi.com) Signed-off-by: Tony Luck --- include/asm-ia64/sn/sn_sal.h | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'include/asm-ia64') diff --git a/include/asm-ia64/sn/sn_sal.h b/include/asm-ia64/sn/sn_sal.h index 7ddce80e191..51aca022cf3 100644 --- a/include/asm-ia64/sn/sn_sal.h +++ b/include/asm-ia64/sn/sn_sal.h @@ -85,6 +85,7 @@ #define SN_SAL_GET_PROM_FEATURE_SET 0x02000065 #define SN_SAL_SET_OS_FEATURE_SET 0x02000066 +#define SN_SAL_INJECT_ERROR 0x02000067 /* * Service-specific constants @@ -1138,4 +1139,16 @@ ia64_sn_set_os_feature(int feature) return rv.status; } +static inline int +sn_inject_error(u64 paddr, u64 *data, u64 *ecc) +{ + struct ia64_sal_retval ret_stuff; + unsigned long irq_flags; + + local_irq_save(irq_flags); + ia64_sal_oemcall_nolock(&ret_stuff, SN_SAL_INJECT_ERROR, paddr, (u64)data, + (u64)ecc, 0, 0, 0, 0); + local_irq_restore(irq_flags); + return ret_stuff.status; +} #endif /* _ASM_IA64_SN_SN_SAL_H */ -- cgit v1.2.3-70-g09d2 From a72391e42f0a13116995045b3d492d660f96697d Mon Sep 17 00:00:00 2001 From: Satoru Takeuchi Date: Thu, 20 Apr 2006 18:49:48 +0900 Subject: [IA64] eliminate compile time warnings This patch removes following compile time warnings: drivers/pci/pci-sysfs.c: In function `pci_read_legacy_io': drivers/pci/pci-sysfs.c:257: warning: implicit declaration of function `ia64_pci_legacy_read' drivers/pci/pci-sysfs.c: In function `pci_write_legacy_io': drivers/pci/pci-sysfs.c:280: warning: implicit declaration of function `ia64_pci_legacy_write' It also fixes wrong definition of ia64_pci_legacy_write (type of `bus' is not `pci_dev', but `pci_bus'). Signed-Off-By: Satoru Takeuchi Signed-off-by: Tony Luck --- arch/ia64/pci/pci.c | 3 +-- include/asm-ia64/machvec.h | 2 ++ 2 files changed, 3 insertions(+), 2 deletions(-) (limited to 'include/asm-ia64') diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c index 9ba32b2d96d..ab829a22f8a 100644 --- a/arch/ia64/pci/pci.c +++ b/arch/ia64/pci/pci.c @@ -31,7 +31,6 @@ #include #include - /* * Low-level SAL-based PCI configuration access functions. Note that SAL * calls are already serialized (via sal_lock), so we don't need another @@ -707,7 +706,7 @@ int ia64_pci_legacy_read(struct pci_bus *bus, u16 port, u32 *val, u8 size) * * Simply writes @size bytes of @val to @port. */ -int ia64_pci_legacy_write(struct pci_dev *bus, u16 port, u32 val, u8 size) +int ia64_pci_legacy_write(struct pci_bus *bus, u16 port, u32 val, u8 size) { int ret = size; diff --git a/include/asm-ia64/machvec.h b/include/asm-ia64/machvec.h index c3e4ed8a3e1..a9c995a86c2 100644 --- a/include/asm-ia64/machvec.h +++ b/include/asm-ia64/machvec.h @@ -347,9 +347,11 @@ extern ia64_mv_dma_supported swiotlb_dma_supported; #endif #ifndef platform_pci_legacy_read # define platform_pci_legacy_read ia64_pci_legacy_read +extern int ia64_pci_legacy_read(struct pci_bus *bus, u16 port, u32 *val, u8 size); #endif #ifndef platform_pci_legacy_write # define platform_pci_legacy_write ia64_pci_legacy_write +extern int ia64_pci_legacy_write(struct pci_bus *bus, u16 port, u32 val, u8 size); #endif #ifndef platform_inb # define platform_inb __ia64_inb -- cgit v1.2.3-70-g09d2 From e5ecc192dfc5e0b325dd8c99ce4c755714c9acbf Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Thu, 13 Apr 2006 18:23:53 -0700 Subject: [IA64] Setup an IA64 specific reclaim distance RECLAIM_DISTANCE is checked on bootup against the SLIT table distances. Zone reclaim is important for system that have higher latencies but not for systems that have multiple nodes on one motherboard and therefore low latencies. We found that on motherboard latencies are typically 1 to 1.4 of local memory access speed whereas multinode systems which benefit from zone reclaim have usually more than 1.5 times the latency of a local access. Set the reclaim distance for IA64 to 1.5 times. Signed-off-by: Christoph Lameter Signed-off-by: Tony Luck --- include/asm-ia64/topology.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/asm-ia64') diff --git a/include/asm-ia64/topology.h b/include/asm-ia64/topology.h index 3ee19dfa46d..616b5ed2aa7 100644 --- a/include/asm-ia64/topology.h +++ b/include/asm-ia64/topology.h @@ -22,6 +22,11 @@ /* Nodes w/o CPUs are preferred for memory allocations, see build_zonelists */ #define PENALTY_FOR_NODE_WITH_CPUS 255 +/* + * Distance above which we begin to use zone reclaim + */ +#define RECLAIM_DISTANCE 15 + /* * Returns the number of the node containing CPU 'cpu' */ -- cgit v1.2.3-70-g09d2 From 912d35f86781e64d73be1ef358f703c08905ac37 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Wed, 26 Apr 2006 10:59:21 +0200 Subject: [PATCH] Add support for the sys_vmsplice syscall sys_splice() moves data to/from pipes with a file input/output. sys_vmsplice() moves data to a pipe, with the input being a user address range instead. This uses an approach suggested by Linus, where we can hold partial ranges inside the pages[] map. Hopefully this will be useful for network receive support as well. Signed-off-by: Jens Axboe --- arch/ia64/kernel/entry.S | 1 + arch/powerpc/kernel/systbl.S | 1 + arch/powerpc/platforms/cell/spu_callbacks.c | 1 + fs/splice.c | 292 ++++++++++++++++++++++++---- include/asm-i386/unistd.h | 3 +- include/asm-ia64/unistd.h | 3 +- include/asm-powerpc/unistd.h | 3 +- include/asm-x86_64/unistd.h | 4 +- include/linux/syscalls.h | 3 + 9 files changed, 268 insertions(+), 43 deletions(-) (limited to 'include/asm-ia64') diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S index e3079881121..bcb80ca5cf4 100644 --- a/arch/ia64/kernel/entry.S +++ b/arch/ia64/kernel/entry.S @@ -1610,5 +1610,6 @@ sys_call_table: data8 sys_get_robust_list data8 sys_sync_file_range // 1300 data8 sys_tee + data8 sys_vmsplice .org sys_call_table + 8*NR_syscalls // guard against failures to increase NR_syscalls diff --git a/arch/powerpc/kernel/systbl.S b/arch/powerpc/kernel/systbl.S index 8d152269050..0b98eea73c5 100644 --- a/arch/powerpc/kernel/systbl.S +++ b/arch/powerpc/kernel/systbl.S @@ -324,6 +324,7 @@ COMPAT_SYS(ppoll) SYSCALL(unshare) SYSCALL(splice) SYSCALL(tee) +SYSCALL(vmsplice) /* * please add new calls to arch/powerpc/platforms/cell/spu_callbacks.c diff --git a/arch/powerpc/platforms/cell/spu_callbacks.c b/arch/powerpc/platforms/cell/spu_callbacks.c index deb3afb9448..b283380a2a1 100644 --- a/arch/powerpc/platforms/cell/spu_callbacks.c +++ b/arch/powerpc/platforms/cell/spu_callbacks.c @@ -318,6 +318,7 @@ void *spu_syscall_table[] = { [__NR_unshare] sys_unshare, [__NR_splice] sys_splice, [__NR_tee] sys_tee, + [__NR_vmsplice] sys_vmsplice, }; long spu_sys_callback(struct spu_syscall_block *s) diff --git a/fs/splice.c b/fs/splice.c index 8c6030c762e..0b2c1f060ca 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -27,6 +27,7 @@ #include #include #include +#include /* * Passed to the actors @@ -38,6 +39,22 @@ struct splice_desc { loff_t pos; /* file position */ }; +struct partial_page { + unsigned int offset; + unsigned int len; +}; + +/* + * Passed to move_to_pipe + */ +struct splice_pipe_desc { + struct page **pages; /* page map */ + struct partial_page *partial; /* pages[] may not be contig */ + int nr_pages; /* number of pages in map */ + unsigned int flags; /* splice flags */ + struct pipe_buf_operations *ops;/* ops associated with output pipe */ +}; + /* * Attempt to steal a page from a pipe buffer. This should perhaps go into * a vm helper function, it's already simplified quite a bit by the @@ -128,6 +145,19 @@ static void page_cache_pipe_buf_unmap(struct pipe_inode_info *info, kunmap(buf->page); } +static void *user_page_pipe_buf_map(struct file *file, + struct pipe_inode_info *pipe, + struct pipe_buffer *buf) +{ + return kmap(buf->page); +} + +static void user_page_pipe_buf_unmap(struct pipe_inode_info *pipe, + struct pipe_buffer *buf) +{ + kunmap(buf->page); +} + static void page_cache_pipe_buf_get(struct pipe_inode_info *info, struct pipe_buffer *buf) { @@ -143,19 +173,33 @@ static struct pipe_buf_operations page_cache_pipe_buf_ops = { .get = page_cache_pipe_buf_get, }; +static int user_page_pipe_buf_steal(struct pipe_inode_info *pipe, + struct pipe_buffer *buf) +{ + return 1; +} + +static struct pipe_buf_operations user_page_pipe_buf_ops = { + .can_merge = 0, + .map = user_page_pipe_buf_map, + .unmap = user_page_pipe_buf_unmap, + .release = page_cache_pipe_buf_release, + .steal = user_page_pipe_buf_steal, + .get = page_cache_pipe_buf_get, +}; + /* * Pipe output worker. This sets up our pipe format with the page cache * pipe buffer operations. Otherwise very similar to the regular pipe_writev(). */ -static ssize_t move_to_pipe(struct pipe_inode_info *pipe, struct page **pages, - int nr_pages, unsigned long len, - unsigned int offset, unsigned int flags) +static ssize_t move_to_pipe(struct pipe_inode_info *pipe, + struct splice_pipe_desc *spd) { - int ret, do_wakeup, i; + int ret, do_wakeup, page_nr; ret = 0; do_wakeup = 0; - i = 0; + page_nr = 0; if (pipe->inode) mutex_lock(&pipe->inode->i_mutex); @@ -171,27 +215,19 @@ static ssize_t move_to_pipe(struct pipe_inode_info *pipe, struct page **pages, if (pipe->nrbufs < PIPE_BUFFERS) { int newbuf = (pipe->curbuf + pipe->nrbufs) & (PIPE_BUFFERS - 1); struct pipe_buffer *buf = pipe->bufs + newbuf; - struct page *page = pages[i++]; - unsigned long this_len; - this_len = PAGE_CACHE_SIZE - offset; - if (this_len > len) - this_len = len; - - buf->page = page; - buf->offset = offset; - buf->len = this_len; - buf->ops = &page_cache_pipe_buf_ops; + buf->page = spd->pages[page_nr]; + buf->offset = spd->partial[page_nr].offset; + buf->len = spd->partial[page_nr].len; + buf->ops = spd->ops; pipe->nrbufs++; + page_nr++; + ret += buf->len; + if (pipe->inode) do_wakeup = 1; - ret += this_len; - len -= this_len; - offset = 0; - if (!--nr_pages) - break; - if (!len) + if (!--spd->nr_pages) break; if (pipe->nrbufs < PIPE_BUFFERS) continue; @@ -199,7 +235,7 @@ static ssize_t move_to_pipe(struct pipe_inode_info *pipe, struct page **pages, break; } - if (flags & SPLICE_F_NONBLOCK) { + if (spd->flags & SPLICE_F_NONBLOCK) { if (!ret) ret = -EAGAIN; break; @@ -234,8 +270,8 @@ static ssize_t move_to_pipe(struct pipe_inode_info *pipe, struct page **pages, kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN); } - while (i < nr_pages) - page_cache_release(pages[i++]); + while (page_nr < spd->nr_pages) + page_cache_release(spd->pages[page_nr++]); return ret; } @@ -246,17 +282,24 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, unsigned int flags) { struct address_space *mapping = in->f_mapping; - unsigned int loff, offset, nr_pages; + unsigned int loff, nr_pages; struct page *pages[PIPE_BUFFERS]; + struct partial_page partial[PIPE_BUFFERS]; struct page *page; pgoff_t index, end_index; loff_t isize; - size_t bytes; - int i, error; + size_t total_len; + int error; + struct splice_pipe_desc spd = { + .pages = pages, + .partial = partial, + .flags = flags, + .ops = &page_cache_pipe_buf_ops, + }; index = *ppos >> PAGE_CACHE_SHIFT; - loff = offset = *ppos & ~PAGE_CACHE_MASK; - nr_pages = (len + offset + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + loff = *ppos & ~PAGE_CACHE_MASK; + nr_pages = (len + loff + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; if (nr_pages > PIPE_BUFFERS) nr_pages = PIPE_BUFFERS; @@ -266,15 +309,15 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, * read-ahead if this is a non-zero offset (we are likely doing small * chunk splice and the page is already there) for a single page. */ - if (!offset || nr_pages > 1) - do_page_cache_readahead(mapping, in, index, nr_pages); + if (!loff || spd.nr_pages > 1) + do_page_cache_readahead(mapping, in, index, spd.nr_pages); /* * Now fill in the holes: */ error = 0; - bytes = 0; - for (i = 0; i < nr_pages; i++, index++) { + total_len = 0; + for (spd.nr_pages = 0; spd.nr_pages < nr_pages; spd.nr_pages++, index++) { unsigned int this_len; if (!len) @@ -367,26 +410,29 @@ readpage: */ if (end_index == index) { loff = PAGE_CACHE_SIZE - (isize & ~PAGE_CACHE_MASK); - if (bytes + loff > isize) { + if (total_len + loff > isize) { page_cache_release(page); break; } /* * force quit after adding this page */ - nr_pages = i; + nr_pages = spd.nr_pages; this_len = min(this_len, loff); + loff = 0; } } fill_it: - pages[i] = page; - bytes += this_len; + pages[spd.nr_pages] = page; + partial[spd.nr_pages].offset = loff; + partial[spd.nr_pages].len = this_len; len -= this_len; + total_len += this_len; loff = 0; } - if (i) - return move_to_pipe(pipe, pages, i, bytes, offset, flags); + if (spd.nr_pages) + return move_to_pipe(pipe, &spd); return error; } @@ -1018,6 +1064,174 @@ static long do_splice(struct file *in, loff_t __user *off_in, return -EINVAL; } +/* + * Map an iov into an array of pages and offset/length tupples. With the + * partial_page structure, we can map several non-contiguous ranges into + * our ones pages[] map instead of splitting that operation into pieces. + * Could easily be exported as a generic helper for other users, in which + * case one would probably want to add a 'max_nr_pages' parameter as well. + */ +static int get_iovec_page_array(const struct iovec __user *iov, + unsigned int nr_vecs, struct page **pages, + struct partial_page *partial) +{ + int buffers = 0, error = 0; + + /* + * It's ok to take the mmap_sem for reading, even + * across a "get_user()". + */ + down_read(¤t->mm->mmap_sem); + + while (nr_vecs) { + unsigned long off, npages; + void __user *base; + size_t len; + int i; + + /* + * Get user address base and length for this iovec. + */ + error = get_user(base, &iov->iov_base); + if (unlikely(error)) + break; + error = get_user(len, &iov->iov_len); + if (unlikely(error)) + break; + + /* + * Sanity check this iovec. 0 read succeeds. + */ + if (unlikely(!len)) + break; + error = -EFAULT; + if (unlikely(!base)) + break; + + /* + * Get this base offset and number of pages, then map + * in the user pages. + */ + off = (unsigned long) base & ~PAGE_MASK; + npages = (off + len + PAGE_SIZE - 1) >> PAGE_SHIFT; + if (npages > PIPE_BUFFERS - buffers) + npages = PIPE_BUFFERS - buffers; + + error = get_user_pages(current, current->mm, + (unsigned long) base, npages, 0, 0, + &pages[buffers], NULL); + + if (unlikely(error <= 0)) + break; + + /* + * Fill this contiguous range into the partial page map. + */ + for (i = 0; i < error; i++) { + const int plen = min_t(size_t, len, PAGE_SIZE) - off; + + partial[buffers].offset = off; + partial[buffers].len = plen; + + off = 0; + len -= plen; + buffers++; + } + + /* + * We didn't complete this iov, stop here since it probably + * means we have to move some of this into a pipe to + * be able to continue. + */ + if (len) + break; + + /* + * Don't continue if we mapped fewer pages than we asked for, + * or if we mapped the max number of pages that we have + * room for. + */ + if (error < npages || buffers == PIPE_BUFFERS) + break; + + nr_vecs--; + iov++; + } + + up_read(¤t->mm->mmap_sem); + + if (buffers) + return buffers; + + return error; +} + +/* + * vmsplice splices a user address range into a pipe. It can be thought of + * as splice-from-memory, where the regular splice is splice-from-file (or + * to file). In both cases the output is a pipe, naturally. + * + * Note that vmsplice only supports splicing _from_ user memory to a pipe, + * not the other way around. Splicing from user memory is a simple operation + * that can be supported without any funky alignment restrictions or nasty + * vm tricks. We simply map in the user memory and fill them into a pipe. + * The reverse isn't quite as easy, though. There are two possible solutions + * for that: + * + * - memcpy() the data internally, at which point we might as well just + * do a regular read() on the buffer anyway. + * - Lots of nasty vm tricks, that are neither fast nor flexible (it + * has restriction limitations on both ends of the pipe). + * + * Alas, it isn't here. + * + */ +static long do_vmsplice(struct file *file, const struct iovec __user *iov, + unsigned long nr_segs, unsigned int flags) +{ + struct pipe_inode_info *pipe = file->f_dentry->d_inode->i_pipe; + struct page *pages[PIPE_BUFFERS]; + struct partial_page partial[PIPE_BUFFERS]; + struct splice_pipe_desc spd = { + .pages = pages, + .partial = partial, + .flags = flags, + .ops = &user_page_pipe_buf_ops, + }; + + if (unlikely(!pipe)) + return -EBADF; + if (unlikely(nr_segs > UIO_MAXIOV)) + return -EINVAL; + else if (unlikely(!nr_segs)) + return 0; + + spd.nr_pages = get_iovec_page_array(iov, nr_segs, pages, partial); + if (spd.nr_pages <= 0) + return spd.nr_pages; + + return move_to_pipe(pipe, &spd); +} + +asmlinkage long sys_vmsplice(int fd, const struct iovec __user *iov, + unsigned long nr_segs, unsigned int flags) +{ + struct file *file; + long error; + int fput; + + error = -EBADF; + file = fget_light(fd, &fput); + if (file) { + if (file->f_mode & FMODE_WRITE) + error = do_vmsplice(file, iov, nr_segs, flags); + + fput_light(file, fput); + } + + return error; +} + asmlinkage long sys_splice(int fd_in, loff_t __user *off_in, int fd_out, loff_t __user *off_out, size_t len, unsigned int flags) diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h index d81d6cfc1bb..eb4b152c82f 100644 --- a/include/asm-i386/unistd.h +++ b/include/asm-i386/unistd.h @@ -321,8 +321,9 @@ #define __NR_splice 313 #define __NR_sync_file_range 314 #define __NR_tee 315 +#define __NR_vmsplice 316 -#define NR_syscalls 316 +#define NR_syscalls 317 /* * user-visible error numbers are in the range -1 - -128: see diff --git a/include/asm-ia64/unistd.h b/include/asm-ia64/unistd.h index a40ebec6aee..7107763168b 100644 --- a/include/asm-ia64/unistd.h +++ b/include/asm-ia64/unistd.h @@ -290,12 +290,13 @@ #define __NR_get_robust_list 1299 #define __NR_sync_file_range 1300 #define __NR_tee 1301 +#define __NR_vmsplice 1302 #ifdef __KERNEL__ #include -#define NR_syscalls 278 /* length of syscall table */ +#define NR_syscalls 279 /* length of syscall table */ #define __ARCH_WANT_SYS_RT_SIGACTION diff --git a/include/asm-powerpc/unistd.h b/include/asm-powerpc/unistd.h index c612f1a6277..34325e29259 100644 --- a/include/asm-powerpc/unistd.h +++ b/include/asm-powerpc/unistd.h @@ -303,8 +303,9 @@ #define __NR_unshare 282 #define __NR_splice 283 #define __NR_tee 284 +#define __NR_vmsplice 285 -#define __NR_syscalls 285 +#define __NR_syscalls 286 #ifdef __KERNEL__ #define __NR__exit __NR_exit diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h index 98c36eae567..feb77cb8c04 100644 --- a/include/asm-x86_64/unistd.h +++ b/include/asm-x86_64/unistd.h @@ -615,8 +615,10 @@ __SYSCALL(__NR_splice, sys_splice) __SYSCALL(__NR_tee, sys_tee) #define __NR_sync_file_range 277 __SYSCALL(__NR_sync_file_range, sys_sync_file_range) +#define __NR_vmsplice 278 +__SYSCALL(__NR_vmsplice, sys_vmsplice) -#define __NR_syscall_max __NR_sync_file_range +#define __NR_syscall_max __NR_vmsplice #ifndef __NO_STUBS diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index d3ebc0e68b2..3996960fc56 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -574,6 +574,9 @@ asmlinkage long sys_splice(int fd_in, loff_t __user *off_in, int fd_out, loff_t __user *off_out, size_t len, unsigned int flags); +asmlinkage long sys_vmsplice(int fd, const struct iovec __user *iov, + unsigned long nr_segs, unsigned int flags); + asmlinkage long sys_tee(int fdin, int fdout, size_t len, unsigned int flags); asmlinkage long sys_sync_file_range(int fd, loff_t offset, loff_t nbytes, -- cgit v1.2.3-70-g09d2 From f0fe253c4719faf76d40f581cdc0e8aef77273bb Mon Sep 17 00:00:00 2001 From: Jack Steiner Date: Sat, 22 Apr 2006 09:36:07 -0500 Subject: [IA64-SGI] - Fix discover of nearest cpu node to IO node Fix a bug that causes discovery of the nearest node/cpu to a TIO (IO node) to fail. Signed-off-by: Jack Steiner Signed-off-by: Tony Luck --- arch/ia64/sn/kernel/sn2/sn_hwperf.c | 4 ++-- include/asm-ia64/sn/sn2/sn_hwperf.h | 6 +++++- 2 files changed, 7 insertions(+), 3 deletions(-) (limited to 'include/asm-ia64') diff --git a/arch/ia64/sn/kernel/sn2/sn_hwperf.c b/arch/ia64/sn/kernel/sn2/sn_hwperf.c index d917afa30b2..7ec65bc0ccf 100644 --- a/arch/ia64/sn/kernel/sn2/sn_hwperf.c +++ b/arch/ia64/sn/kernel/sn2/sn_hwperf.c @@ -284,6 +284,8 @@ static int sn_hwperf_get_nearest_node_objdata(struct sn_hwperf_object_info *objb /* find nearest node with cpus and nearest memory */ for (router=NULL, j=0; j < op->ports; j++) { dest = sn_hwperf_findobj_id(objbuf, nobj, ptdata[j].conn_id); + if (dest && SN_HWPERF_IS_ROUTER(dest)) + router = dest; if (!dest || SN_HWPERF_FOREIGN(dest) || !SN_HWPERF_IS_NODE(dest) || SN_HWPERF_IS_IONODE(dest)) { continue; @@ -299,8 +301,6 @@ static int sn_hwperf_get_nearest_node_objdata(struct sn_hwperf_object_info *objb *near_mem_node = c; found_mem++; } - if (SN_HWPERF_IS_ROUTER(dest)) - router = dest; } if (router && (!found_cpu || !found_mem)) { diff --git a/include/asm-ia64/sn/sn2/sn_hwperf.h b/include/asm-ia64/sn/sn2/sn_hwperf.h index 291ef3d69da..e61ebac38cd 100644 --- a/include/asm-ia64/sn/sn2/sn_hwperf.h +++ b/include/asm-ia64/sn/sn2/sn_hwperf.h @@ -45,8 +45,12 @@ struct sn_hwperf_object_info { #define SN_HWPERF_IS_NODE(x) ((x) && strstr((x)->name, "SHub")) #define SN_HWPERF_IS_NODE_SHUB2(x) ((x) && strstr((x)->name, "SHub 2.")) #define SN_HWPERF_IS_IONODE(x) ((x) && strstr((x)->name, "TIO")) -#define SN_HWPERF_IS_ROUTER(x) ((x) && strstr((x)->name, "Router")) #define SN_HWPERF_IS_NL3ROUTER(x) ((x) && strstr((x)->name, "NL3Router")) +#define SN_HWPERF_IS_NL4ROUTER(x) ((x) && strstr((x)->name, "NL4Router")) +#define SN_HWPERF_IS_OLDROUTER(x) ((x) && strstr((x)->name, "Router")) +#define SN_HWPERF_IS_ROUTER(x) (SN_HWPERF_IS_NL3ROUTER(x) || \ + SN_HWPERF_IS_NL4ROUTER(x) || \ + SN_HWPERF_IS_OLDROUTER(x)) #define SN_HWPERF_FOREIGN(x) ((x) && !(x)->sn_hwp_this_part && !(x)->sn_hwp_is_shared) #define SN_HWPERF_SAME_OBJTYPE(x,y) ((SN_HWPERF_IS_NODE(x) && SN_HWPERF_IS_NODE(y)) ||\ (SN_HWPERF_IS_IONODE(x) && SN_HWPERF_IS_IONODE(y)) ||\ -- cgit v1.2.3-70-g09d2 From 1df57c0c21c92a6d4fcfe5304c84151ed9beb7a2 Mon Sep 17 00:00:00 2001 From: Cliff Wickman Date: Tue, 25 Apr 2006 10:47:48 -0500 Subject: [IA64] enable dumps to capture second page of kernel stack In SLES10 (2.6.16) crash dumping (in my experience, LKCD) is unable to capture the second page of the 2-page task/stack allocation. This is particularly troublesome for dump analysis, as the stack traceback cannot be done. (A similar convention is probably needed throughout the kernel to make kernel multi-page allocations detectable for dumping) Multi-page kernel allocations are represented by the single page structure associated with the first page of the allocation. The page structures associated with the other pages are unintialized. If the dumper is selecting only kernel pages it has no way to identify any but the first page of the allocation. The fix is to make the task/stack allocation a compound page. Signed-off-by: Cliff Wickman Signed-off-by: Tony Luck --- include/asm-ia64/thread_info.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/asm-ia64') diff --git a/include/asm-ia64/thread_info.h b/include/asm-ia64/thread_info.h index 56394a2c705..e5392c4d30c 100644 --- a/include/asm-ia64/thread_info.h +++ b/include/asm-ia64/thread_info.h @@ -67,7 +67,7 @@ struct thread_info { #define end_of_stack(p) (unsigned long *)((void *)(p) + IA64_RBS_OFFSET) #define __HAVE_ARCH_TASK_STRUCT_ALLOCATOR -#define alloc_task_struct() ((task_t *)__get_free_pages(GFP_KERNEL, KERNEL_STACK_SIZE_ORDER)) +#define alloc_task_struct() ((task_t *)__get_free_pages(GFP_KERNEL | __GFP_COMP, KERNEL_STACK_SIZE_ORDER)) #define free_task_struct(tsk) free_pages((unsigned long) (tsk), KERNEL_STACK_SIZE_ORDER) #endif /* !__ASSEMBLY */ -- cgit v1.2.3-70-g09d2 From 913ed41eb5c948d2f8b5deffd29c2638eceef3d7 Mon Sep 17 00:00:00 2001 From: Jon Mason Date: Wed, 3 May 2006 17:26:58 -0500 Subject: [IA64] remove asm-ia64/bitops.h self-inclusion asm-ia64/bitops.h includes itself. The #ifndef _ASM_IA64_BITOPS_H prevents this from being an issue, but it should still be removed. Signed-off-by: Jon Mason Signed-off-by: Tony Luck --- include/asm-ia64/bitops.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/asm-ia64') diff --git a/include/asm-ia64/bitops.h b/include/asm-ia64/bitops.h index 90921e16279..6cc517e212a 100644 --- a/include/asm-ia64/bitops.h +++ b/include/asm-ia64/bitops.h @@ -11,7 +11,6 @@ #include #include -#include #include /** -- cgit v1.2.3-70-g09d2