From c186afb4dbd0050a537b96c7fbee2dba3b57fc38 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 2 Feb 2014 21:16:54 -0500 Subject: switch ->is_partially_uptodate() to saner arguments Signed-off-by: Al Viro --- mm/filemap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm/filemap.c') diff --git a/mm/filemap.c b/mm/filemap.c index 7a13f6ac5421..46e98019af6c 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1148,7 +1148,7 @@ find_page: if (!page->mapping) goto page_not_up_to_date_locked; if (!mapping->a_ops->is_partially_uptodate(page, - desc, offset)) + offset, desc->count)) goto page_not_up_to_date_locked; unlock_page(page); } -- cgit v1.2.3 From 9e8c2af96e0d2d5fe298dd796fb6bc16e888a48d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 2 Feb 2014 22:10:25 -0500 Subject: callers of iov_copy_from_user_atomic() don't need pagecache_disable() ... it does that itself (via kmap_atomic()) Signed-off-by: Al Viro --- fs/btrfs/file.c | 5 ----- fs/fuse/file.c | 2 -- mm/filemap.c | 3 --- 3 files changed, 10 deletions(-) (limited to 'mm/filemap.c') diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 0165b8672f09..34e096201da1 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -425,13 +425,8 @@ static noinline int btrfs_copy_from_user(loff_t pos, int num_pages, struct page *page = prepared_pages[pg]; /* * Copy data from userspace to the current page - * - * Disable pagefault to avoid recursive lock since - * the pages are already locked */ - pagefault_disable(); copied = iov_iter_copy_from_user_atomic(page, i, offset, count); - pagefault_enable(); /* Flush processor's dcache for this page */ flush_dcache_page(page); diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 77bcc303c3ae..a91d3b4d32f3 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -1003,9 +1003,7 @@ static ssize_t fuse_fill_write_pages(struct fuse_req *req, if (mapping_writably_mapped(mapping)) flush_dcache_page(page); - pagefault_disable(); tmp = iov_iter_copy_from_user_atomic(page, ii, offset, bytes); - pagefault_enable(); flush_dcache_page(page); mark_page_accessed(page); diff --git a/mm/filemap.c b/mm/filemap.c index 46e98019af6c..bfb7a97d6d0f 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1974,7 +1974,6 @@ size_t iov_iter_copy_from_user_atomic(struct page *page, char *kaddr; size_t copied; - BUG_ON(!in_atomic()); kaddr = kmap_atomic(page); if (likely(i->nr_segs == 1)) { int left; @@ -2348,9 +2347,7 @@ again: if (mapping_writably_mapped(mapping)) flush_dcache_page(page); - pagefault_disable(); copied = iov_iter_copy_from_user_atomic(page, i, offset, bytes); - pagefault_enable(); flush_dcache_page(page); mark_page_accessed(page); -- cgit v1.2.3 From 6e58e79db8a16222b31fc8da1ca2ac2dccfc4237 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 3 Feb 2014 17:07:03 -0500 Subject: introduce copy_page_to_iter, kill loop over iovec in generic_file_aio_read() generic_file_aio_read() was looping over the target iovec, with loop over (source) pages nested inside that. Just set an iov_iter up and pass *that* to do_generic_file_aio_read(). With copy_page_to_iter() doing all work of mapping and copying a page to iovec and advancing iov_iter. Switch shmem_file_aio_read() to the same and kill file_read_actor(), while we are at it. Signed-off-by: Al Viro --- include/linux/fs.h | 1 - include/linux/uio.h | 2 + mm/filemap.c | 202 ++++++++++++++++++++++++++++------------------------ mm/shmem.c | 77 +++++++------------- 4 files changed, 138 insertions(+), 144 deletions(-) (limited to 'mm/filemap.c') diff --git a/include/linux/fs.h b/include/linux/fs.h index 2261ac8f0534..2a5b1744f80a 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2390,7 +2390,6 @@ extern int generic_file_mmap(struct file *, struct vm_area_struct *); extern int generic_file_readonly_mmap(struct file *, struct vm_area_struct *); extern int generic_file_remap_pages(struct vm_area_struct *, unsigned long addr, unsigned long size, pgoff_t pgoff); -extern int file_read_actor(read_descriptor_t * desc, struct page *page, unsigned long offset, unsigned long size); int generic_write_checks(struct file *file, loff_t *pos, size_t *count, int isblk); extern ssize_t generic_file_aio_read(struct kiocb *, const struct iovec *, unsigned long, loff_t); extern ssize_t __generic_file_aio_write(struct kiocb *, const struct iovec *, unsigned long, diff --git a/include/linux/uio.h b/include/linux/uio.h index 347d70ce098e..199bcc34241b 100644 --- a/include/linux/uio.h +++ b/include/linux/uio.h @@ -67,6 +67,8 @@ size_t iov_iter_copy_from_user(struct page *page, void iov_iter_advance(struct iov_iter *i, size_t bytes); int iov_iter_fault_in_readable(struct iov_iter *i, size_t bytes); size_t iov_iter_single_seg_count(const struct iov_iter *i); +size_t copy_page_to_iter(struct page *page, size_t offset, size_t bytes, + struct iov_iter *i); static inline void iov_iter_init(struct iov_iter *i, const struct iovec *iov, unsigned long nr_segs, diff --git a/mm/filemap.c b/mm/filemap.c index bfb7a97d6d0f..a16eb2c4f316 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1085,11 +1085,90 @@ static void shrink_readahead_size_eio(struct file *filp, ra->ra_pages /= 4; } +size_t copy_page_to_iter(struct page *page, size_t offset, size_t bytes, + struct iov_iter *i) +{ + size_t skip, copy, left, wanted; + const struct iovec *iov; + char __user *buf; + void *kaddr, *from; + + if (unlikely(bytes > i->count)) + bytes = i->count; + + if (unlikely(!bytes)) + return 0; + + wanted = bytes; + iov = i->iov; + skip = i->iov_offset; + buf = iov->iov_base + skip; + copy = min(bytes, iov->iov_len - skip); + + if (!fault_in_pages_writeable(buf, copy)) { + kaddr = kmap_atomic(page); + from = kaddr + offset; + + /* first chunk, usually the only one */ + left = __copy_to_user_inatomic(buf, from, copy); + copy -= left; + skip += copy; + from += copy; + bytes -= copy; + + while (unlikely(!left && bytes)) { + iov++; + buf = iov->iov_base; + copy = min(bytes, iov->iov_len); + left = __copy_to_user_inatomic(buf, from, copy); + copy -= left; + skip = copy; + from += copy; + bytes -= copy; + } + if (likely(!bytes)) { + kunmap_atomic(kaddr); + goto done; + } + offset = from - kaddr; + buf += copy; + kunmap_atomic(kaddr); + copy = min(bytes, iov->iov_len - skip); + } + /* Too bad - revert to non-atomic kmap */ + kaddr = kmap(page); + from = kaddr + offset; + left = __copy_to_user(buf, from, copy); + copy -= left; + skip += copy; + from += copy; + bytes -= copy; + while (unlikely(!left && bytes)) { + iov++; + buf = iov->iov_base; + copy = min(bytes, iov->iov_len); + left = __copy_to_user(buf, from, copy); + copy -= left; + skip = copy; + from += copy; + bytes -= copy; + } + kunmap(page); +done: + i->count -= wanted - bytes; + i->nr_segs -= iov - i->iov; + i->iov = iov; + i->iov_offset = skip; + return wanted - bytes; +} +EXPORT_SYMBOL(copy_page_to_iter); + /** * do_generic_file_read - generic file read routine * @filp: the file to read * @ppos: current file position - * @desc: read_descriptor + * @iter: data destination + * @written: already copied * * This is a generic file read routine, and uses the * mapping->a_ops->readpage() function for the actual low-level stuff. @@ -1097,8 +1176,8 @@ static void shrink_readahead_size_eio(struct file *filp, * This is really ugly. But the goto's actually try to clarify some * of the logic when it comes to error handling etc. */ -static void do_generic_file_read(struct file *filp, loff_t *ppos, - read_descriptor_t *desc) +static ssize_t do_generic_file_read(struct file *filp, loff_t *ppos, + struct iov_iter *iter, ssize_t written) { struct address_space *mapping = filp->f_mapping; struct inode *inode = mapping->host; @@ -1108,12 +1187,12 @@ static void do_generic_file_read(struct file *filp, loff_t *ppos, pgoff_t prev_index; unsigned long offset; /* offset into pagecache page */ unsigned int prev_offset; - int error; + int error = 0; index = *ppos >> PAGE_CACHE_SHIFT; prev_index = ra->prev_pos >> PAGE_CACHE_SHIFT; prev_offset = ra->prev_pos & (PAGE_CACHE_SIZE-1); - last_index = (*ppos + desc->count + PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT; + last_index = (*ppos + iter->count + PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT; offset = *ppos & ~PAGE_CACHE_MASK; for (;;) { @@ -1148,7 +1227,7 @@ find_page: if (!page->mapping) goto page_not_up_to_date_locked; if (!mapping->a_ops->is_partially_uptodate(page, - offset, desc->count)) + offset, iter->count)) goto page_not_up_to_date_locked; unlock_page(page); } @@ -1198,24 +1277,23 @@ page_ok: /* * Ok, we have the page, and it's up-to-date, so * now we can copy it to user space... - * - * The file_read_actor routine returns how many bytes were - * actually used.. - * NOTE! This may not be the same as how much of a user buffer - * we filled up (we may be padding etc), so we can only update - * "pos" here (the actor routine has to update the user buffer - * pointers and the remaining count). */ - ret = file_read_actor(desc, page, offset, nr); + + ret = copy_page_to_iter(page, offset, nr, iter); offset += ret; index += offset >> PAGE_CACHE_SHIFT; offset &= ~PAGE_CACHE_MASK; prev_offset = offset; page_cache_release(page); - if (ret == nr && desc->count) - continue; - goto out; + written += ret; + if (!iov_iter_count(iter)) + goto out; + if (ret < nr) { + error = -EFAULT; + goto out; + } + continue; page_not_up_to_date: /* Get exclusive access to the page ... */ @@ -1250,6 +1328,7 @@ readpage: if (unlikely(error)) { if (error == AOP_TRUNCATED_PAGE) { page_cache_release(page); + error = 0; goto find_page; } goto readpage_error; @@ -1280,7 +1359,6 @@ readpage: readpage_error: /* UHHUH! A synchronous read error occurred. Report it */ - desc->error = error; page_cache_release(page); goto out; @@ -1291,16 +1369,17 @@ no_cached_page: */ page = page_cache_alloc_cold(mapping); if (!page) { - desc->error = -ENOMEM; + error = -ENOMEM; goto out; } error = add_to_page_cache_lru(page, mapping, index, GFP_KERNEL); if (error) { page_cache_release(page); - if (error == -EEXIST) + if (error == -EEXIST) { + error = 0; goto find_page; - desc->error = error; + } goto out; } goto readpage; @@ -1313,44 +1392,7 @@ out: *ppos = ((loff_t)index << PAGE_CACHE_SHIFT) + offset; file_accessed(filp); -} - -int file_read_actor(read_descriptor_t *desc, struct page *page, - unsigned long offset, unsigned long size) -{ - char *kaddr; - unsigned long left, count = desc->count; - - if (size > count) - size = count; - - /* - * Faults on the destination of a read are common, so do it before - * taking the kmap. - */ - if (!fault_in_pages_writeable(desc->arg.buf, size)) { - kaddr = kmap_atomic(page); - left = __copy_to_user_inatomic(desc->arg.buf, - kaddr + offset, size); - kunmap_atomic(kaddr); - if (left == 0) - goto success; - } - - /* Do it the slow way */ - kaddr = kmap(page); - left = __copy_to_user(desc->arg.buf, kaddr + offset, size); - kunmap(page); - - if (left) { - size -= left; - desc->error = -EFAULT; - } -success: - desc->count = count - size; - desc->written += size; - desc->arg.buf += size; - return size; + return written ? written : error; } /* @@ -1408,14 +1450,15 @@ generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov, { struct file *filp = iocb->ki_filp; ssize_t retval; - unsigned long seg = 0; size_t count; loff_t *ppos = &iocb->ki_pos; + struct iov_iter i; count = 0; retval = generic_segment_checks(iov, &nr_segs, &count, VERIFY_WRITE); if (retval) return retval; + iov_iter_init(&i, iov, nr_segs, count, 0); /* coalesce the iovecs and go direct-to-BIO for O_DIRECT */ if (filp->f_flags & O_DIRECT) { @@ -1437,6 +1480,11 @@ generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov, if (retval > 0) { *ppos = pos + retval; count -= retval; + /* + * If we did a short DIO read we need to skip the + * section of the iov that we've already read data into. + */ + iov_iter_advance(&i, retval); } /* @@ -1453,39 +1501,7 @@ generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov, } } - count = retval; - for (seg = 0; seg < nr_segs; seg++) { - read_descriptor_t desc; - loff_t offset = 0; - - /* - * If we did a short DIO read we need to skip the section of the - * iov that we've already read data into. - */ - if (count) { - if (count > iov[seg].iov_len) { - count -= iov[seg].iov_len; - continue; - } - offset = count; - count = 0; - } - - desc.written = 0; - desc.arg.buf = iov[seg].iov_base + offset; - desc.count = iov[seg].iov_len - offset; - if (desc.count == 0) - continue; - desc.error = 0; - do_generic_file_read(filp, ppos, &desc); - retval += desc.written; - if (desc.error) { - retval = retval ?: desc.error; - break; - } - if (desc.count > 0) - break; - } + retval = do_generic_file_read(filp, ppos, &i, retval); out: return retval; } diff --git a/mm/shmem.c b/mm/shmem.c index 9398e6cd48cb..17d3799d04bd 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -1462,13 +1462,25 @@ shmem_write_end(struct file *file, struct address_space *mapping, return copied; } -static void do_shmem_file_read(struct file *filp, loff_t *ppos, read_descriptor_t *desc) +static ssize_t shmem_file_aio_read(struct kiocb *iocb, + const struct iovec *iov, unsigned long nr_segs, loff_t pos) { - struct inode *inode = file_inode(filp); + struct file *file = iocb->ki_filp; + struct inode *inode = file_inode(file); struct address_space *mapping = inode->i_mapping; pgoff_t index; unsigned long offset; enum sgp_type sgp = SGP_READ; + int error; + ssize_t retval; + size_t count; + loff_t *ppos = &iocb->ki_pos; + struct iov_iter iter; + + retval = generic_segment_checks(iov, &nr_segs, &count, VERIFY_WRITE); + if (retval) + return retval; + iov_iter_init(&iter, iov, nr_segs, count, 0); /* * Might this read be for a stacking filesystem? Then when reading @@ -1496,10 +1508,10 @@ static void do_shmem_file_read(struct file *filp, loff_t *ppos, read_descriptor_ break; } - desc->error = shmem_getpage(inode, index, &page, sgp, NULL); - if (desc->error) { - if (desc->error == -EINVAL) - desc->error = 0; + error = shmem_getpage(inode, index, &page, sgp, NULL); + if (error) { + if (error == -EINVAL) + error = 0; break; } if (page) @@ -1543,61 +1555,26 @@ static void do_shmem_file_read(struct file *filp, loff_t *ppos, read_descriptor_ /* * Ok, we have the page, and it's up-to-date, so * now we can copy it to user space... - * - * The actor routine returns how many bytes were actually used.. - * NOTE! This may not be the same as how much of a user buffer - * we filled up (we may be padding etc), so we can only update - * "pos" here (the actor routine has to update the user buffer - * pointers and the remaining count). */ - ret = file_read_actor(desc, page, offset, nr); + ret = copy_page_to_iter(page, offset, nr, &iter); + retval += ret; offset += ret; index += offset >> PAGE_CACHE_SHIFT; offset &= ~PAGE_CACHE_MASK; page_cache_release(page); - if (ret != nr || !desc->count) + if (!iov_iter_count(&iter)) break; - + if (ret < nr) { + error = -EFAULT; + break; + } cond_resched(); } *ppos = ((loff_t) index << PAGE_CACHE_SHIFT) + offset; - file_accessed(filp); -} - -static ssize_t shmem_file_aio_read(struct kiocb *iocb, - const struct iovec *iov, unsigned long nr_segs, loff_t pos) -{ - struct file *filp = iocb->ki_filp; - ssize_t retval; - unsigned long seg; - size_t count; - loff_t *ppos = &iocb->ki_pos; - - retval = generic_segment_checks(iov, &nr_segs, &count, VERIFY_WRITE); - if (retval) - return retval; - - for (seg = 0; seg < nr_segs; seg++) { - read_descriptor_t desc; - - desc.written = 0; - desc.arg.buf = iov[seg].iov_base; - desc.count = iov[seg].iov_len; - if (desc.count == 0) - continue; - desc.error = 0; - do_shmem_file_read(filp, ppos, &desc); - retval += desc.written; - if (desc.error) { - retval = retval ?: desc.error; - break; - } - if (desc.count > 0) - break; - } - return retval; + file_accessed(file); + return retval ? retval : error; } static ssize_t shmem_file_splice_read(struct file *in, loff_t *ppos, -- cgit v1.2.3 From 4f18cd317a118c28482f97303600a2fe2ada6c79 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 5 Feb 2014 19:11:33 -0500 Subject: take iov_iter stuff to mm/iov_iter.c Signed-off-by: Al Viro --- mm/Makefile | 2 +- mm/filemap.c | 221 --------------------------------------------------------- mm/iov_iter.c | 224 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 225 insertions(+), 222 deletions(-) create mode 100644 mm/iov_iter.c (limited to 'mm/filemap.c') diff --git a/mm/Makefile b/mm/Makefile index 310c90a09264..178a43406b0c 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -17,7 +17,7 @@ obj-y := filemap.o mempool.o oom_kill.o fadvise.o \ util.o mmzone.o vmstat.o backing-dev.o \ mm_init.o mmu_context.o percpu.o slab_common.o \ compaction.o balloon_compaction.o \ - interval_tree.o list_lru.o $(mmu-y) + interval_tree.o list_lru.o iov_iter.o $(mmu-y) obj-y += init-mm.o diff --git a/mm/filemap.c b/mm/filemap.c index a16eb2c4f316..c4730efa5d9e 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1085,84 +1085,6 @@ static void shrink_readahead_size_eio(struct file *filp, ra->ra_pages /= 4; } -size_t copy_page_to_iter(struct page *page, size_t offset, size_t bytes, - struct iov_iter *i) -{ - size_t skip, copy, left, wanted; - const struct iovec *iov; - char __user *buf; - void *kaddr, *from; - - if (unlikely(bytes > i->count)) - bytes = i->count; - - if (unlikely(!bytes)) - return 0; - - wanted = bytes; - iov = i->iov; - skip = i->iov_offset; - buf = iov->iov_base + skip; - copy = min(bytes, iov->iov_len - skip); - - if (!fault_in_pages_writeable(buf, copy)) { - kaddr = kmap_atomic(page); - from = kaddr + offset; - - /* first chunk, usually the only one */ - left = __copy_to_user_inatomic(buf, from, copy); - copy -= left; - skip += copy; - from += copy; - bytes -= copy; - - while (unlikely(!left && bytes)) { - iov++; - buf = iov->iov_base; - copy = min(bytes, iov->iov_len); - left = __copy_to_user_inatomic(buf, from, copy); - copy -= left; - skip = copy; - from += copy; - bytes -= copy; - } - if (likely(!bytes)) { - kunmap_atomic(kaddr); - goto done; - } - offset = from - kaddr; - buf += copy; - kunmap_atomic(kaddr); - copy = min(bytes, iov->iov_len - skip); - } - /* Too bad - revert to non-atomic kmap */ - kaddr = kmap(page); - from = kaddr + offset; - left = __copy_to_user(buf, from, copy); - copy -= left; - skip += copy; - from += copy; - bytes -= copy; - while (unlikely(!left && bytes)) { - iov++; - buf = iov->iov_base; - copy = min(bytes, iov->iov_len); - left = __copy_to_user(buf, from, copy); - copy -= left; - skip = copy; - from += copy; - bytes -= copy; - } - kunmap(page); -done: - i->count -= wanted - bytes; - i->nr_segs -= iov - i->iov; - i->iov = iov; - i->iov_offset = skip; - return wanted - bytes; -} -EXPORT_SYMBOL(copy_page_to_iter); - /** * do_generic_file_read - generic file read routine * @filp: the file to read @@ -1957,149 +1879,6 @@ struct page *read_cache_page(struct address_space *mapping, } EXPORT_SYMBOL(read_cache_page); -static size_t __iovec_copy_from_user_inatomic(char *vaddr, - const struct iovec *iov, size_t base, size_t bytes) -{ - size_t copied = 0, left = 0; - - while (bytes) { - char __user *buf = iov->iov_base + base; - int copy = min(bytes, iov->iov_len - base); - - base = 0; - left = __copy_from_user_inatomic(vaddr, buf, copy); - copied += copy; - bytes -= copy; - vaddr += copy; - iov++; - - if (unlikely(left)) - break; - } - return copied - left; -} - -/* - * Copy as much as we can into the page and return the number of bytes which - * were successfully copied. If a fault is encountered then return the number of - * bytes which were copied. - */ -size_t iov_iter_copy_from_user_atomic(struct page *page, - struct iov_iter *i, unsigned long offset, size_t bytes) -{ - char *kaddr; - size_t copied; - - kaddr = kmap_atomic(page); - if (likely(i->nr_segs == 1)) { - int left; - char __user *buf = i->iov->iov_base + i->iov_offset; - left = __copy_from_user_inatomic(kaddr + offset, buf, bytes); - copied = bytes - left; - } else { - copied = __iovec_copy_from_user_inatomic(kaddr + offset, - i->iov, i->iov_offset, bytes); - } - kunmap_atomic(kaddr); - - return copied; -} -EXPORT_SYMBOL(iov_iter_copy_from_user_atomic); - -/* - * This has the same sideeffects and return value as - * iov_iter_copy_from_user_atomic(). - * The difference is that it attempts to resolve faults. - * Page must not be locked. - */ -size_t iov_iter_copy_from_user(struct page *page, - struct iov_iter *i, unsigned long offset, size_t bytes) -{ - char *kaddr; - size_t copied; - - kaddr = kmap(page); - if (likely(i->nr_segs == 1)) { - int left; - char __user *buf = i->iov->iov_base + i->iov_offset; - left = __copy_from_user(kaddr + offset, buf, bytes); - copied = bytes - left; - } else { - copied = __iovec_copy_from_user_inatomic(kaddr + offset, - i->iov, i->iov_offset, bytes); - } - kunmap(page); - return copied; -} -EXPORT_SYMBOL(iov_iter_copy_from_user); - -void iov_iter_advance(struct iov_iter *i, size_t bytes) -{ - BUG_ON(i->count < bytes); - - if (likely(i->nr_segs == 1)) { - i->iov_offset += bytes; - i->count -= bytes; - } else { - const struct iovec *iov = i->iov; - size_t base = i->iov_offset; - unsigned long nr_segs = i->nr_segs; - - /* - * The !iov->iov_len check ensures we skip over unlikely - * zero-length segments (without overruning the iovec). - */ - while (bytes || unlikely(i->count && !iov->iov_len)) { - int copy; - - copy = min(bytes, iov->iov_len - base); - BUG_ON(!i->count || i->count < copy); - i->count -= copy; - bytes -= copy; - base += copy; - if (iov->iov_len == base) { - iov++; - nr_segs--; - base = 0; - } - } - i->iov = iov; - i->iov_offset = base; - i->nr_segs = nr_segs; - } -} -EXPORT_SYMBOL(iov_iter_advance); - -/* - * Fault in the first iovec of the given iov_iter, to a maximum length - * of bytes. Returns 0 on success, or non-zero if the memory could not be - * accessed (ie. because it is an invalid address). - * - * writev-intensive code may want this to prefault several iovecs -- that - * would be possible (callers must not rely on the fact that _only_ the - * first iovec will be faulted with the current implementation). - */ -int iov_iter_fault_in_readable(struct iov_iter *i, size_t bytes) -{ - char __user *buf = i->iov->iov_base + i->iov_offset; - bytes = min(bytes, i->iov->iov_len - i->iov_offset); - return fault_in_pages_readable(buf, bytes); -} -EXPORT_SYMBOL(iov_iter_fault_in_readable); - -/* - * Return the count of just the current iov_iter segment. - */ -size_t iov_iter_single_seg_count(const struct iov_iter *i) -{ - const struct iovec *iov = i->iov; - if (i->nr_segs == 1) - return i->count; - else - return min(i->count, iov->iov_len - i->iov_offset); -} -EXPORT_SYMBOL(iov_iter_single_seg_count); - /* * Performs necessary checks before doing a write * diff --git a/mm/iov_iter.c b/mm/iov_iter.c new file mode 100644 index 000000000000..10e46cd721de --- /dev/null +++ b/mm/iov_iter.c @@ -0,0 +1,224 @@ +#include +#include +#include + +size_t copy_page_to_iter(struct page *page, size_t offset, size_t bytes, + struct iov_iter *i) +{ + size_t skip, copy, left, wanted; + const struct iovec *iov; + char __user *buf; + void *kaddr, *from; + + if (unlikely(bytes > i->count)) + bytes = i->count; + + if (unlikely(!bytes)) + return 0; + + wanted = bytes; + iov = i->iov; + skip = i->iov_offset; + buf = iov->iov_base + skip; + copy = min(bytes, iov->iov_len - skip); + + if (!fault_in_pages_writeable(buf, copy)) { + kaddr = kmap_atomic(page); + from = kaddr + offset; + + /* first chunk, usually the only one */ + left = __copy_to_user_inatomic(buf, from, copy); + copy -= left; + skip += copy; + from += copy; + bytes -= copy; + + while (unlikely(!left && bytes)) { + iov++; + buf = iov->iov_base; + copy = min(bytes, iov->iov_len); + left = __copy_to_user_inatomic(buf, from, copy); + copy -= left; + skip = copy; + from += copy; + bytes -= copy; + } + if (likely(!bytes)) { + kunmap_atomic(kaddr); + goto done; + } + offset = from - kaddr; + buf += copy; + kunmap_atomic(kaddr); + copy = min(bytes, iov->iov_len - skip); + } + /* Too bad - revert to non-atomic kmap */ + kaddr = kmap(page); + from = kaddr + offset; + left = __copy_to_user(buf, from, copy); + copy -= left; + skip += copy; + from += copy; + bytes -= copy; + while (unlikely(!left && bytes)) { + iov++; + buf = iov->iov_base; + copy = min(bytes, iov->iov_len); + left = __copy_to_user(buf, from, copy); + copy -= left; + skip = copy; + from += copy; + bytes -= copy; + } + kunmap(page); +done: + i->count -= wanted - bytes; + i->nr_segs -= iov - i->iov; + i->iov = iov; + i->iov_offset = skip; + return wanted - bytes; +} +EXPORT_SYMBOL(copy_page_to_iter); + +static size_t __iovec_copy_from_user_inatomic(char *vaddr, + const struct iovec *iov, size_t base, size_t bytes) +{ + size_t copied = 0, left = 0; + + while (bytes) { + char __user *buf = iov->iov_base + base; + int copy = min(bytes, iov->iov_len - base); + + base = 0; + left = __copy_from_user_inatomic(vaddr, buf, copy); + copied += copy; + bytes -= copy; + vaddr += copy; + iov++; + + if (unlikely(left)) + break; + } + return copied - left; +} + +/* + * Copy as much as we can into the page and return the number of bytes which + * were successfully copied. If a fault is encountered then return the number of + * bytes which were copied. + */ +size_t iov_iter_copy_from_user_atomic(struct page *page, + struct iov_iter *i, unsigned long offset, size_t bytes) +{ + char *kaddr; + size_t copied; + + kaddr = kmap_atomic(page); + if (likely(i->nr_segs == 1)) { + int left; + char __user *buf = i->iov->iov_base + i->iov_offset; + left = __copy_from_user_inatomic(kaddr + offset, buf, bytes); + copied = bytes - left; + } else { + copied = __iovec_copy_from_user_inatomic(kaddr + offset, + i->iov, i->iov_offset, bytes); + } + kunmap_atomic(kaddr); + + return copied; +} +EXPORT_SYMBOL(iov_iter_copy_from_user_atomic); + +/* + * This has the same sideeffects and return value as + * iov_iter_copy_from_user_atomic(). + * The difference is that it attempts to resolve faults. + * Page must not be locked. + */ +size_t iov_iter_copy_from_user(struct page *page, + struct iov_iter *i, unsigned long offset, size_t bytes) +{ + char *kaddr; + size_t copied; + + kaddr = kmap(page); + if (likely(i->nr_segs == 1)) { + int left; + char __user *buf = i->iov->iov_base + i->iov_offset; + left = __copy_from_user(kaddr + offset, buf, bytes); + copied = bytes - left; + } else { + copied = __iovec_copy_from_user_inatomic(kaddr + offset, + i->iov, i->iov_offset, bytes); + } + kunmap(page); + return copied; +} +EXPORT_SYMBOL(iov_iter_copy_from_user); + +void iov_iter_advance(struct iov_iter *i, size_t bytes) +{ + BUG_ON(i->count < bytes); + + if (likely(i->nr_segs == 1)) { + i->iov_offset += bytes; + i->count -= bytes; + } else { + const struct iovec *iov = i->iov; + size_t base = i->iov_offset; + unsigned long nr_segs = i->nr_segs; + + /* + * The !iov->iov_len check ensures we skip over unlikely + * zero-length segments (without overruning the iovec). + */ + while (bytes || unlikely(i->count && !iov->iov_len)) { + int copy; + + copy = min(bytes, iov->iov_len - base); + BUG_ON(!i->count || i->count < copy); + i->count -= copy; + bytes -= copy; + base += copy; + if (iov->iov_len == base) { + iov++; + nr_segs--; + base = 0; + } + } + i->iov = iov; + i->iov_offset = base; + i->nr_segs = nr_segs; + } +} +EXPORT_SYMBOL(iov_iter_advance); + +/* + * Fault in the first iovec of the given iov_iter, to a maximum length + * of bytes. Returns 0 on success, or non-zero if the memory could not be + * accessed (ie. because it is an invalid address). + * + * writev-intensive code may want this to prefault several iovecs -- that + * would be possible (callers must not rely on the fact that _only_ the + * first iovec will be faulted with the current implementation). + */ +int iov_iter_fault_in_readable(struct iov_iter *i, size_t bytes) +{ + char __user *buf = i->iov->iov_base + i->iov_offset; + bytes = min(bytes, i->iov->iov_len - i->iov_offset); + return fault_in_pages_readable(buf, bytes); +} +EXPORT_SYMBOL(iov_iter_fault_in_readable); + +/* + * Return the count of just the current iov_iter segment. + */ +size_t iov_iter_single_seg_count(const struct iov_iter *i) +{ + const struct iovec *iov = i->iov; + if (i->nr_segs == 1) + return i->count; + else + return min(i->count, iov->iov_len - i->iov_offset); +} +EXPORT_SYMBOL(iov_iter_single_seg_count); -- cgit v1.2.3 From 41fc56d573c35a212688b12b48af8c303f9bb6d2 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 9 Feb 2014 12:58:52 -0500 Subject: kill the 4th argument of __generic_file_aio_write() It's always equal to &iocb->ki_pos, where iocb is the value of the 1st argument. Signed-off-by: Al Viro --- fs/block_dev.c | 2 +- fs/ext4/file.c | 2 +- fs/udf/file.c | 2 +- include/linux/fs.h | 3 +-- mm/filemap.c | 13 ++++++------- 5 files changed, 10 insertions(+), 12 deletions(-) (limited to 'mm/filemap.c') diff --git a/fs/block_dev.c b/fs/block_dev.c index 1e86823a9cbd..764bd3b8d2fa 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -1518,7 +1518,7 @@ ssize_t blkdev_aio_write(struct kiocb *iocb, const struct iovec *iov, BUG_ON(iocb->ki_pos != pos); blk_start_plug(&plug); - ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos); + ret = __generic_file_aio_write(iocb, iov, nr_segs); if (ret > 0) { ssize_t err; diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 1a5073959f32..d564bcfb23c5 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -146,7 +146,7 @@ ext4_file_dio_write(struct kiocb *iocb, const struct iovec *iov, overwrite = 1; } - ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos); + ret = __generic_file_aio_write(iocb, iov, nr_segs); mutex_unlock(&inode->i_mutex); if (ret > 0) { diff --git a/fs/udf/file.c b/fs/udf/file.c index 1037637957c7..d2c170f8b035 100644 --- a/fs/udf/file.c +++ b/fs/udf/file.c @@ -171,7 +171,7 @@ static ssize_t udf_file_aio_write(struct kiocb *iocb, const struct iovec *iov, } else up_write(&iinfo->i_data_sem); - retval = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos); + retval = __generic_file_aio_write(iocb, iov, nr_segs); mutex_unlock(&inode->i_mutex); if (retval > 0) { diff --git a/include/linux/fs.h b/include/linux/fs.h index 2a5b1744f80a..e677d1e1189f 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2392,8 +2392,7 @@ extern int generic_file_remap_pages(struct vm_area_struct *, unsigned long addr, unsigned long size, pgoff_t pgoff); int generic_write_checks(struct file *file, loff_t *pos, size_t *count, int isblk); extern ssize_t generic_file_aio_read(struct kiocb *, const struct iovec *, unsigned long, loff_t); -extern ssize_t __generic_file_aio_write(struct kiocb *, const struct iovec *, unsigned long, - loff_t *); +extern ssize_t __generic_file_aio_write(struct kiocb *, const struct iovec *, unsigned long); extern ssize_t generic_file_aio_write(struct kiocb *, const struct iovec *, unsigned long, loff_t); extern ssize_t generic_file_direct_write(struct kiocb *, const struct iovec *, unsigned long *, loff_t, loff_t *, size_t, size_t); diff --git a/mm/filemap.c b/mm/filemap.c index c4730efa5d9e..ce2246dd90de 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -2222,14 +2222,14 @@ EXPORT_SYMBOL(generic_file_buffered_write); * avoid syncing under i_mutex. */ ssize_t __generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t *ppos) + unsigned long nr_segs) { struct file *file = iocb->ki_filp; struct address_space * mapping = file->f_mapping; size_t ocount; /* original count */ size_t count; /* after file limit checks */ struct inode *inode = mapping->host; - loff_t pos; + loff_t pos = iocb->ki_pos; ssize_t written; ssize_t err; @@ -2239,7 +2239,6 @@ ssize_t __generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, return err; count = ocount; - pos = *ppos; /* We can write back this queue in page reclaim */ current->backing_dev_info = mapping->backing_dev_info; @@ -2266,7 +2265,7 @@ ssize_t __generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, ssize_t written_buffered; written = generic_file_direct_write(iocb, iov, &nr_segs, pos, - ppos, count, ocount); + &iocb->ki_pos, count, ocount); if (written < 0 || written == count) goto out; /* @@ -2276,7 +2275,7 @@ ssize_t __generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, pos += written; count -= written; written_buffered = generic_file_buffered_write(iocb, iov, - nr_segs, pos, ppos, count, + nr_segs, pos, &iocb->ki_pos, count, written); /* * If generic_file_buffered_write() retuned a synchronous error @@ -2310,7 +2309,7 @@ ssize_t __generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, } } else { written = generic_file_buffered_write(iocb, iov, nr_segs, - pos, ppos, count, written); + pos, &iocb->ki_pos, count, written); } out: current->backing_dev_info = NULL; @@ -2339,7 +2338,7 @@ ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, BUG_ON(iocb->ki_pos != pos); mutex_lock(&inode->i_mutex); - ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos); + ret = __generic_file_aio_write(iocb, iov, nr_segs); mutex_unlock(&inode->i_mutex); if (ret > 0) { -- cgit v1.2.3 From fcacafd269adc88f41b68cb77a3f957a66563428 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 9 Feb 2014 13:37:49 -0500 Subject: kill the 5th argument of generic_file_buffered_write() same story - it's &iocb->ki_pos in all cases Signed-off-by: Al Viro --- fs/ceph/file.c | 3 +-- fs/ocfs2/file.c | 2 +- fs/xfs/xfs_file.c | 2 +- include/linux/fs.h | 2 +- mm/filemap.c | 9 ++++----- 5 files changed, 8 insertions(+), 10 deletions(-) (limited to 'mm/filemap.c') diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 09c7afe32e49..a798db5e5e39 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -978,8 +978,7 @@ retry_snap: * can not run at the same time */ written = generic_file_buffered_write(iocb, iov, nr_segs, - pos, &iocb->ki_pos, - count, 0); + pos, count, 0); mutex_unlock(&inode->i_mutex); } diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 51632c40e896..89099cce14fe 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -2383,7 +2383,7 @@ relock: } else { current->backing_dev_info = file->f_mapping->backing_dev_info; written = generic_file_buffered_write(iocb, iov, nr_segs, *ppos, - ppos, count, 0); + count, 0); current->backing_dev_info = NULL; } diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 64b48eade91d..175ce58fbfa3 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -738,7 +738,7 @@ xfs_file_buffered_aio_write( write_retry: trace_xfs_file_buffered_write(ip, count, iocb->ki_pos, 0); ret = generic_file_buffered_write(iocb, iovp, nr_segs, - pos, &iocb->ki_pos, count, 0); + pos, count, 0); /* * If we just got an ENOSPC, try to write back all dirty inodes to diff --git a/include/linux/fs.h b/include/linux/fs.h index e677d1e1189f..830e37420f5e 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2397,7 +2397,7 @@ extern ssize_t generic_file_aio_write(struct kiocb *, const struct iovec *, unsi extern ssize_t generic_file_direct_write(struct kiocb *, const struct iovec *, unsigned long *, loff_t, loff_t *, size_t, size_t); extern ssize_t generic_file_buffered_write(struct kiocb *, const struct iovec *, - unsigned long, loff_t, loff_t *, size_t, ssize_t); + unsigned long, loff_t, size_t, ssize_t); extern ssize_t do_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos); extern ssize_t do_sync_write(struct file *filp, const char __user *buf, size_t len, loff_t *ppos); extern int generic_segment_checks(const struct iovec *iov, diff --git a/mm/filemap.c b/mm/filemap.c index ce2246dd90de..9d515a1a2372 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -2183,7 +2183,7 @@ again: ssize_t generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t pos, loff_t *ppos, + unsigned long nr_segs, loff_t pos, size_t count, ssize_t written) { struct file *file = iocb->ki_filp; @@ -2195,7 +2195,7 @@ generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, if (likely(status >= 0)) { written += status; - *ppos = pos + status; + iocb->ki_pos = pos + status; } return written ? written : status; @@ -2275,8 +2275,7 @@ ssize_t __generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, pos += written; count -= written; written_buffered = generic_file_buffered_write(iocb, iov, - nr_segs, pos, &iocb->ki_pos, count, - written); + nr_segs, pos, count, written); /* * If generic_file_buffered_write() retuned a synchronous error * then we want to return the number of bytes which were @@ -2309,7 +2308,7 @@ ssize_t __generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, } } else { written = generic_file_buffered_write(iocb, iov, nr_segs, - pos, &iocb->ki_pos, count, written); + pos, count, written); } out: current->backing_dev_info = NULL; -- cgit v1.2.3 From 5cb6c6c7eb1ed24744b41fad47d9a25b72207098 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 11 Feb 2014 20:58:20 -0500 Subject: generic_file_direct_write(): get rid of ppos argument always equal to &iocb->ki_pos. Signed-off-by: Al Viro --- fs/btrfs/file.c | 2 +- fs/fuse/file.c | 3 +-- fs/ocfs2/file.c | 2 +- fs/xfs/xfs_file.c | 2 +- include/linux/fs.h | 2 +- mm/filemap.c | 6 +++--- 6 files changed, 8 insertions(+), 9 deletions(-) (limited to 'mm/filemap.c') diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index f6032a2bfab9..8ed4b165abbd 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1640,7 +1640,7 @@ static ssize_t __btrfs_direct_write(struct kiocb *iocb, loff_t endbyte; int err; - written = generic_file_direct_write(iocb, iov, &nr_segs, pos, &iocb->ki_pos, + written = generic_file_direct_write(iocb, iov, &nr_segs, pos, count, ocount); if (written < 0 || written == count) diff --git a/fs/fuse/file.c b/fs/fuse/file.c index a91d3b4d32f3..fd06d1ebc2eb 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -1143,8 +1143,7 @@ static ssize_t fuse_file_aio_write(struct kiocb *iocb, const struct iovec *iov, goto out; if (file->f_flags & O_DIRECT) { - written = generic_file_direct_write(iocb, iov, &nr_segs, - pos, &iocb->ki_pos, + written = generic_file_direct_write(iocb, iov, &nr_segs, pos, count, ocount); if (written < 0 || written == count) goto out; diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 89099cce14fe..77b8a742866f 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -2375,7 +2375,7 @@ relock: if (direct_io) { written = generic_file_direct_write(iocb, iov, &nr_segs, *ppos, - ppos, count, ocount); + count, ocount); if (written < 0) { ret = written; goto out_dio; diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 175ce58fbfa3..e593554ce65e 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -699,7 +699,7 @@ xfs_file_dio_aio_write( trace_xfs_file_direct_write(ip, count, iocb->ki_pos, 0); ret = generic_file_direct_write(iocb, iovp, - &nr_segs, pos, &iocb->ki_pos, count, ocount); + &nr_segs, pos, count, ocount); out: xfs_rw_iunlock(ip, iolock); diff --git a/include/linux/fs.h b/include/linux/fs.h index 830e37420f5e..9dfd7c7ff8e3 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2395,7 +2395,7 @@ extern ssize_t generic_file_aio_read(struct kiocb *, const struct iovec *, unsig extern ssize_t __generic_file_aio_write(struct kiocb *, const struct iovec *, unsigned long); extern ssize_t generic_file_aio_write(struct kiocb *, const struct iovec *, unsigned long, loff_t); extern ssize_t generic_file_direct_write(struct kiocb *, const struct iovec *, - unsigned long *, loff_t, loff_t *, size_t, size_t); + unsigned long *, loff_t, size_t, size_t); extern ssize_t generic_file_buffered_write(struct kiocb *, const struct iovec *, unsigned long, loff_t, size_t, ssize_t); extern ssize_t do_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos); diff --git a/mm/filemap.c b/mm/filemap.c index 9d515a1a2372..93e9cf576452 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1985,7 +1985,7 @@ EXPORT_SYMBOL(pagecache_write_end); ssize_t generic_file_direct_write(struct kiocb *iocb, const struct iovec *iov, - unsigned long *nr_segs, loff_t pos, loff_t *ppos, + unsigned long *nr_segs, loff_t pos, size_t count, size_t ocount) { struct file *file = iocb->ki_filp; @@ -2046,7 +2046,7 @@ generic_file_direct_write(struct kiocb *iocb, const struct iovec *iov, i_size_write(inode, pos); mark_inode_dirty(inode); } - *ppos = pos; + iocb->ki_pos = pos; } out: return written; @@ -2265,7 +2265,7 @@ ssize_t __generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, ssize_t written_buffered; written = generic_file_direct_write(iocb, iov, &nr_segs, pos, - &iocb->ki_pos, count, ocount); + count, ocount); if (written < 0 || written == count) goto out; /* -- cgit v1.2.3 From 3b93f911d55cf8b3540f82fac4eeddfee7b5de0c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 11 Feb 2014 21:34:08 -0500 Subject: export generic_perform_write(), start getting rid of generic_file_buffer_write() Signed-off-by: Al Viro --- include/linux/fs.h | 2 ++ mm/filemap.c | 37 +++++++++++++++++++++---------------- 2 files changed, 23 insertions(+), 16 deletions(-) (limited to 'mm/filemap.c') diff --git a/include/linux/fs.h b/include/linux/fs.h index 9dfd7c7ff8e3..87be51aff9d7 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -48,6 +48,7 @@ struct cred; struct swap_info_struct; struct seq_file; struct workqueue_struct; +struct iov_iter; extern void __init inode_init(void); extern void __init inode_init_early(void); @@ -2396,6 +2397,7 @@ extern ssize_t __generic_file_aio_write(struct kiocb *, const struct iovec *, un extern ssize_t generic_file_aio_write(struct kiocb *, const struct iovec *, unsigned long, loff_t); extern ssize_t generic_file_direct_write(struct kiocb *, const struct iovec *, unsigned long *, loff_t, size_t, size_t); +extern ssize_t generic_perform_write(struct file *, struct iov_iter *, loff_t); extern ssize_t generic_file_buffered_write(struct kiocb *, const struct iovec *, unsigned long, loff_t, size_t, ssize_t); extern ssize_t do_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos); diff --git a/mm/filemap.c b/mm/filemap.c index 93e9cf576452..09bfc9b3bb51 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -2092,7 +2092,7 @@ found: } EXPORT_SYMBOL(grab_cache_page_write_begin); -static ssize_t generic_perform_write(struct file *file, +ssize_t generic_perform_write(struct file *file, struct iov_iter *i, loff_t pos) { struct address_space *mapping = file->f_mapping; @@ -2180,6 +2180,7 @@ again: return written ? written : status; } +EXPORT_SYMBOL(generic_perform_write); ssize_t generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, @@ -2230,8 +2231,10 @@ ssize_t __generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, size_t count; /* after file limit checks */ struct inode *inode = mapping->host; loff_t pos = iocb->ki_pos; - ssize_t written; + ssize_t written = 0; ssize_t err; + ssize_t status; + struct iov_iter from; ocount = 0; err = generic_segment_checks(iov, &nr_segs, &ocount, VERIFY_READ); @@ -2242,8 +2245,6 @@ ssize_t __generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, /* We can write back this queue in page reclaim */ current->backing_dev_info = mapping->backing_dev_info; - written = 0; - err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode)); if (err) goto out; @@ -2259,44 +2260,47 @@ ssize_t __generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, if (err) goto out; + iov_iter_init(&from, iov, nr_segs, count, 0); + /* coalesce the iovecs and go direct-to-BIO for O_DIRECT */ if (unlikely(file->f_flags & O_DIRECT)) { loff_t endbyte; - ssize_t written_buffered; - written = generic_file_direct_write(iocb, iov, &nr_segs, pos, + written = generic_file_direct_write(iocb, iov, &from.nr_segs, pos, count, ocount); if (written < 0 || written == count) goto out; + iov_iter_advance(&from, written); + /* * direct-io write to a hole: fall through to buffered I/O * for completing the rest of the request. */ pos += written; count -= written; - written_buffered = generic_file_buffered_write(iocb, iov, - nr_segs, pos, count, written); + + status = generic_perform_write(file, &from, pos); /* - * If generic_file_buffered_write() retuned a synchronous error + * If generic_perform_write() returned a synchronous error * then we want to return the number of bytes which were * direct-written, or the error code if that was zero. Note * that this differs from normal direct-io semantics, which * will return -EFOO even if some bytes were written. */ - if (written_buffered < 0) { - err = written_buffered; + if (unlikely(status < 0) && !written) { + err = status; goto out; } - + iocb->ki_pos = pos + status; /* * We need to ensure that the page cache pages are written to * disk and invalidated to preserve the expected O_DIRECT * semantics. */ - endbyte = pos + written_buffered - written - 1; + endbyte = pos + status - 1; err = filemap_write_and_wait_range(file->f_mapping, pos, endbyte); if (err == 0) { - written = written_buffered; + written += status; invalidate_mapping_pages(mapping, pos >> PAGE_CACHE_SHIFT, endbyte >> PAGE_CACHE_SHIFT); @@ -2307,8 +2311,9 @@ ssize_t __generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, */ } } else { - written = generic_file_buffered_write(iocb, iov, nr_segs, - pos, count, written); + written = generic_perform_write(file, &from, pos); + if (likely(written >= 0)) + iocb->ki_pos = pos + written; } out: current->backing_dev_info = NULL; -- cgit v1.2.3 From ccad2365668f12336dd497d9039e8584af836070 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 11 Feb 2014 22:36:48 -0500 Subject: kill generic_file_buffered_write() Signed-off-by: Al Viro --- include/linux/fs.h | 2 -- mm/filemap.c | 23 +---------------------- 2 files changed, 1 insertion(+), 24 deletions(-) (limited to 'mm/filemap.c') diff --git a/include/linux/fs.h b/include/linux/fs.h index 87be51aff9d7..c309b7a0da2d 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2398,8 +2398,6 @@ extern ssize_t generic_file_aio_write(struct kiocb *, const struct iovec *, unsi extern ssize_t generic_file_direct_write(struct kiocb *, const struct iovec *, unsigned long *, loff_t, size_t, size_t); extern ssize_t generic_perform_write(struct file *, struct iov_iter *, loff_t); -extern ssize_t generic_file_buffered_write(struct kiocb *, const struct iovec *, - unsigned long, loff_t, size_t, ssize_t); extern ssize_t do_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos); extern ssize_t do_sync_write(struct file *filp, const char __user *buf, size_t len, loff_t *ppos); extern int generic_segment_checks(const struct iovec *iov, diff --git a/mm/filemap.c b/mm/filemap.c index 09bfc9b3bb51..1a455142784d 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -76,7 +76,7 @@ * ->mmap_sem * ->lock_page (access_process_vm) * - * ->i_mutex (generic_file_buffered_write) + * ->i_mutex (generic_perform_write) * ->mmap_sem (fault_in_pages_readable->do_page_fault) * * bdi->wb.list_lock @@ -2182,27 +2182,6 @@ again: } EXPORT_SYMBOL(generic_perform_write); -ssize_t -generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t pos, - size_t count, ssize_t written) -{ - struct file *file = iocb->ki_filp; - ssize_t status; - struct iov_iter i; - - iov_iter_init(&i, iov, nr_segs, count, written); - status = generic_perform_write(file, &i, pos); - - if (likely(status >= 0)) { - written += status; - iocb->ki_pos = pos + status; - } - - return written ? written : status; -} -EXPORT_SYMBOL(generic_file_buffered_write); - /** * __generic_file_aio_write - write data to a file * @iocb: IO state structure (file, offset, etc.) -- cgit v1.2.3