From e92adcba261fd391591bb63c1703185a04a41554 Mon Sep 17 00:00:00 2001 From: Jeff Moyer Date: Mon, 28 Apr 2008 02:12:04 -0700 Subject: aio: io_getevents() should return if io_destroy() is invoked This patch wakes up a thread waiting in io_getevents if another thread destroys the context. This was tested using a small program that spawns a thread to wait in io_getevents while the parent thread destroys the io context and then waits for the getevents thread to exit. Without this patch, the program hangs indefinitely. With the patch, the program exits as expected. Signed-off-by: Jeff Moyer Cc: Zach Brown Cc: Christopher Smith Cc: Benjamin LaHaise Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/aio.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/aio.c b/fs/aio.c index 228368610dfa..ae94e1dea266 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -1166,7 +1166,10 @@ retry: break; if (min_nr <= i) break; - ret = 0; + if (unlikely(ctx->dead)) { + ret = -EINVAL; + break; + } if (to.timed_out) /* Only check after read evt */ break; /* Try to only show up in io wait if there are ops @@ -1231,6 +1234,13 @@ static void io_destroy(struct kioctx *ioctx) aio_cancel_all(ioctx); wait_for_all_aios(ioctx); + + /* + * Wake up any waiters. The setting of ctx->dead must be seen + * by other CPUs at this point. Right now, we rely on the + * locking done by the above calls to ensure this consistency. + */ + wake_up(&ioctx->wait); put_ioctx(ioctx); /* once for the lookup */ } -- cgit From 488514d1798289f56f80ed018e246179fe500383 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Mon, 28 Apr 2008 02:12:05 -0700 Subject: Remove set_migrateflags() Migrate flags must be set on slab creation as agreed upon when the antifrag logic was reviewed. Otherwise some slabs of a slabcache will end up in the unmovable and others in the reclaimable section depending on which flag was active when a new slab page was allocated. This likely slid in somehow when antifrag was merged. Remove it. The buffer_heads are always allocated with __GFP_RECLAIMABLE because the SLAB_RECLAIM_ACCOUNT option is set. The set_migrateflags() never had any effect there. Radix tree allocations are not directly reclaimable but they are allocated with __GFP_RECLAIMABLE set on each allocation. We now set SLAB_RECLAIM_ACCOUNT on radix tree slab creation making sure that radix tree slabs are consistently placed in the reclaimable section. Radix tree slabs will also be accounted as such. There is then no user left of set_migratepages. So remove it. Signed-off-by: Christoph Lameter Cc: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 39ff14403d13..8b9807523efe 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -3180,8 +3180,7 @@ static void recalc_bh_state(void) struct buffer_head *alloc_buffer_head(gfp_t gfp_flags) { - struct buffer_head *ret = kmem_cache_alloc(bh_cachep, - set_migrateflags(gfp_flags, __GFP_RECLAIMABLE)); + struct buffer_head *ret = kmem_cache_alloc(bh_cachep, gfp_flags); if (ret) { INIT_LIST_HEAD(&ret->b_assoc_buffers); get_cpu_var(bh_accounting).nr++; -- cgit From 3c18ddd160d1fcd46d1131d9ad6c594dd8e9af99 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Mon, 28 Apr 2008 02:12:10 -0700 Subject: mm: remove nopage Nothing in the tree uses nopage any more. Remove support for it in the core mm code and documentation (and a few stray references to it in comments). Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/gfs2/ops_address.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index 90a04a6e3789..f55394e57cb2 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -438,7 +438,7 @@ static int stuffed_readpage(struct gfs2_inode *ip, struct page *page) int error; /* - * Due to the order of unstuffing files and ->nopage(), we can be + * Due to the order of unstuffing files and ->fault(), we can be * asked for a zero page in the case of a stuffed file being extended, * so we need to supply one here. It doesn't happen often. */ -- cgit From 9d02dbc8137759e4c2f91db0b7f9c8a1ec2a9276 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Mon, 28 Apr 2008 02:12:11 -0700 Subject: make swap_pte_to_pagemap_entry() static Make the needlessly global swap_pte_to_pagemap_entry() static. Signed-off-by: Adrian Bunk Acked-by: Matt Mackall Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/task_mmu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 9dfb5ff24209..f4ab76c7c662 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -579,7 +579,7 @@ static int pagemap_pte_hole(unsigned long start, unsigned long end, return err; } -u64 swap_pte_to_pagemap_entry(pte_t pte) +static u64 swap_pte_to_pagemap_entry(pte_t pte) { swp_entry_t e = pte_to_swp_entry(pte); return swp_type(e) | (swp_offset(e) << MAX_SWAPFILES_SHIFT); -- cgit From dac1d27bc8d5ca636d3014ecfdf94407031d1970 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Mon, 28 Apr 2008 02:12:12 -0700 Subject: mm: use zonelists instead of zones when direct reclaiming pages The following patches replace multiple zonelists per node with two zonelists that are filtered based on the GFP flags. The patches as a set fix a bug with regard to the use of MPOL_BIND and ZONE_MOVABLE. With this patchset, the MPOL_BIND will apply to the two highest zones when the highest zone is ZONE_MOVABLE. This should be considered as an alternative fix for the MPOL_BIND+ZONE_MOVABLE in 2.6.23 to the previously discussed hack that filters only custom zonelists. The first patch cleans up an inconsistency where direct reclaim uses zonelist->zones where other places use zonelist. The second patch introduces a helper function node_zonelist() for looking up the appropriate zonelist for a GFP mask which simplifies patches later in the set. The third patch defines/remembers the "preferred zone" for numa statistics, as it is no longer always the first zone in a zonelist. The forth patch replaces multiple zonelists with two zonelists that are filtered. The two zonelists are due to the fact that the memoryless patchset introduces a second set of zonelists for __GFP_THISNODE. The fifth patch introduces helper macros for retrieving the zone and node indices of entries in a zonelist. The final patch introduces filtering of the zonelists based on a nodemask. Two zonelists exist per node, one for normal allocations and one for __GFP_THISNODE. Performance results varied depending on the machine configuration. In real workloads the gain/loss will depend on how much the userspace portion of the benchmark benefits from having more cache available due to reduced referencing of zonelists. These are the range of performance losses/gains when running against 2.6.24-rc4-mm1. The set and these machines are a mix of i386, x86_64 and ppc64 both NUMA and non-NUMA. loss to gain Total CPU time on Kernbench: -0.86% to 1.13% Elapsed time on Kernbench: -0.79% to 0.76% page_test from aim9: -4.37% to 0.79% brk_test from aim9: -0.71% to 4.07% fork_test from aim9: -1.84% to 4.60% exec_test from aim9: -0.71% to 1.08% This patch: The allocator deals with zonelists which indicate the order in which zones should be targeted for an allocation. Similarly, direct reclaim of pages iterates over an array of zones. For consistency, this patch converts direct reclaim to use a zonelist. No functionality is changed by this patch. This simplifies zonelist iterators in the next patch. Signed-off-by: Mel Gorman Acked-by: Christoph Lameter Signed-off-by: Lee Schermerhorn Cc: KAMEZAWA Hiroyuki Cc: Mel Gorman Cc: Christoph Lameter Cc: Hugh Dickins Cc: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 8b9807523efe..1dae94acb3fe 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -360,16 +360,16 @@ void invalidate_bdev(struct block_device *bdev) */ static void free_more_memory(void) { - struct zone **zones; + struct zonelist *zonelist; pg_data_t *pgdat; wakeup_pdflush(1024); yield(); for_each_online_pgdat(pgdat) { - zones = pgdat->node_zonelists[gfp_zone(GFP_NOFS)].zones; - if (*zones) - try_to_free_pages(zones, 0, GFP_NOFS); + zonelist = &pgdat->node_zonelists[gfp_zone(GFP_NOFS)]; + if (zonelist->zones[0]) + try_to_free_pages(zonelist, 0, GFP_NOFS); } } -- cgit From 0e88460da6ab7bb6a7ef83675412ed5b6315d741 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Mon, 28 Apr 2008 02:12:14 -0700 Subject: mm: introduce node_zonelist() for accessing the zonelist for a GFP mask Introduce a node_zonelist() helper function. It is used to lookup the appropriate zonelist given a node and a GFP mask. The patch on its own is a cleanup but it helps clarify parts of the two-zonelist-per-node patchset. If necessary, it can be merged with the next patch in this set without problems. Reviewed-by: Christoph Lameter Signed-off-by: Mel Gorman Signed-off-by: Lee Schermerhorn Cc: KAMEZAWA Hiroyuki Cc: Mel Gorman Cc: Christoph Lameter Cc: Hugh Dickins Cc: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 1dae94acb3fe..71358499bc57 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -361,13 +361,13 @@ void invalidate_bdev(struct block_device *bdev) static void free_more_memory(void) { struct zonelist *zonelist; - pg_data_t *pgdat; + int nid; wakeup_pdflush(1024); yield(); - for_each_online_pgdat(pgdat) { - zonelist = &pgdat->node_zonelists[gfp_zone(GFP_NOFS)]; + for_each_online_node(nid) { + zonelist = node_zonelist(nid, GFP_NOFS); if (zonelist->zones[0]) try_to_free_pages(zonelist, 0, GFP_NOFS); } -- cgit From 54a6eb5c4765aa573a030ceeba2c14e3d2ea5706 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Mon, 28 Apr 2008 02:12:16 -0700 Subject: mm: use two zonelist that are filtered by GFP mask Currently a node has two sets of zonelists, one for each zone type in the system and a second set for GFP_THISNODE allocations. Based on the zones allowed by a gfp mask, one of these zonelists is selected. All of these zonelists consume memory and occupy cache lines. This patch replaces the multiple zonelists per-node with two zonelists. The first contains all populated zones in the system, ordered by distance, for fallback allocations when the target/preferred node has no free pages. The second contains all populated zones in the node suitable for GFP_THISNODE allocations. An iterator macro is introduced called for_each_zone_zonelist() that interates through each zone allowed by the GFP flags in the selected zonelist. Signed-off-by: Mel Gorman Acked-by: Christoph Lameter Signed-off-by: Lee Schermerhorn Cc: KAMEZAWA Hiroyuki Cc: Mel Gorman Cc: Christoph Lameter Cc: Hugh Dickins Cc: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 71358499bc57..9b5434a80479 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -360,16 +360,18 @@ void invalidate_bdev(struct block_device *bdev) */ static void free_more_memory(void) { - struct zonelist *zonelist; + struct zone **zones; int nid; wakeup_pdflush(1024); yield(); for_each_online_node(nid) { - zonelist = node_zonelist(nid, GFP_NOFS); - if (zonelist->zones[0]) - try_to_free_pages(zonelist, 0, GFP_NOFS); + zones = first_zones_zonelist(node_zonelist(nid, GFP_NOFS), + gfp_zone(GFP_NOFS)); + if (*zones) + try_to_free_pages(node_zonelist(nid, GFP_NOFS), 0, + GFP_NOFS); } } -- cgit From dd1a239f6f2d4d3eedd318583ec319aa145b324c Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Mon, 28 Apr 2008 02:12:17 -0700 Subject: mm: have zonelist contains structs with both a zone pointer and zone_idx Filtering zonelists requires very frequent use of zone_idx(). This is costly as it involves a lookup of another structure and a substraction operation. As the zone_idx is often required, it should be quickly accessible. The node idx could also be stored here if it was found that accessing zone->node is significant which may be the case on workloads where nodemasks are heavily used. This patch introduces a struct zoneref to store a zone pointer and a zone index. The zonelist then consists of an array of these struct zonerefs which are looked up as necessary. Helpers are given for accessing the zone index as well as the node index. [kamezawa.hiroyu@jp.fujitsu.com: Suggested struct zoneref instead of embedding information in pointers] [hugh@veritas.com: mm-have-zonelist: fix memcg ooms] [hugh@veritas.com: just return do_try_to_free_pages] [hugh@veritas.com: do_try_to_free_pages gfp_mask redundant] Signed-off-by: Mel Gorman Acked-by: Christoph Lameter Acked-by: David Rientjes Signed-off-by: Lee Schermerhorn Cc: KAMEZAWA Hiroyuki Cc: Mel Gorman Cc: Christoph Lameter Cc: Nick Piggin Signed-off-by: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 9b5434a80479..ac84cd13075d 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -360,16 +360,16 @@ void invalidate_bdev(struct block_device *bdev) */ static void free_more_memory(void) { - struct zone **zones; + struct zoneref *zrefs; int nid; wakeup_pdflush(1024); yield(); for_each_online_node(nid) { - zones = first_zones_zonelist(node_zonelist(nid, GFP_NOFS), + zrefs = first_zones_zonelist(node_zonelist(nid, GFP_NOFS), gfp_zone(GFP_NOFS)); - if (*zones) + if (zrefs->zone) try_to_free_pages(node_zonelist(nid, GFP_NOFS), 0, GFP_NOFS); } -- cgit From 19770b32609b6bf97a3dece2529089494cbfc549 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Mon, 28 Apr 2008 02:12:18 -0700 Subject: mm: filter based on a nodemask as well as a gfp_mask The MPOL_BIND policy creates a zonelist that is used for allocations controlled by that mempolicy. As the per-node zonelist is already being filtered based on a zone id, this patch adds a version of __alloc_pages() that takes a nodemask for further filtering. This eliminates the need for MPOL_BIND to create a custom zonelist. A positive benefit of this is that allocations using MPOL_BIND now use the local node's distance-ordered zonelist instead of a custom node-id-ordered zonelist. I.e., pages will be allocated from the closest allowed node with available memory. [Lee.Schermerhorn@hp.com: Mempolicy: update stale documentation and comments] [Lee.Schermerhorn@hp.com: Mempolicy: make dequeue_huge_page_vma() obey MPOL_BIND nodemask] [Lee.Schermerhorn@hp.com: Mempolicy: make dequeue_huge_page_vma() obey MPOL_BIND nodemask rework] Signed-off-by: Mel Gorman Acked-by: Christoph Lameter Signed-off-by: Lee Schermerhorn Cc: KAMEZAWA Hiroyuki Cc: Mel Gorman Cc: Hugh Dickins Cc: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index ac84cd13075d..7d51e649b19a 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -360,16 +360,17 @@ void invalidate_bdev(struct block_device *bdev) */ static void free_more_memory(void) { - struct zoneref *zrefs; + struct zone *zone; int nid; wakeup_pdflush(1024); yield(); for_each_online_node(nid) { - zrefs = first_zones_zonelist(node_zonelist(nid, GFP_NOFS), - gfp_zone(GFP_NOFS)); - if (zrefs->zone) + (void)first_zones_zonelist(node_zonelist(nid, GFP_NOFS), + gfp_zone(GFP_NOFS), NULL, + &zone); + if (zone) try_to_free_pages(node_zonelist(nid, GFP_NOFS), 0, GFP_NOFS); } -- cgit From 028fec414d803117eb4b2ed12acb4dd5da65b32d Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Mon, 28 Apr 2008 02:12:25 -0700 Subject: mempolicy: support optional mode flags With the evolution of mempolicies, it is necessary to support mempolicy mode flags that specify how the policy shall behave in certain circumstances. The most immediate need for mode flag support is to suppress remapping the nodemask of a policy at the time of rebind. Both the mempolicy mode and flags are passed by the user in the 'int policy' formal of either the set_mempolicy() or mbind() syscall. A new constant, MPOL_MODE_FLAGS, represents the union of legal optional flags that may be passed as part of this int. Mempolicies that include illegal flags as part of their policy are rejected as invalid. An additional member to struct mempolicy is added to support the mode flags: struct mempolicy { ... unsigned short policy; unsigned short flags; } The splitting of the 'int' actual passed by the user is done in sys_set_mempolicy() and sys_mbind() for their respective syscalls. This is done by intersecting the actual with MPOL_MODE_FLAGS, rejecting the syscall of there are additional flags, and storing it in the new 'flags' member of struct mempolicy. The intersection of the actual with ~MPOL_MODE_FLAGS is stored in the 'policy' member of the struct and all current users of pol->policy remain unchanged. The union of the policy mode and optional mode flags is passed back to the user in get_mempolicy(). This combination of mode and flags within the same actual does not break userspace code that relies on get_mempolicy(&policy, ...) and either switch (policy) { case MPOL_BIND: ... case MPOL_INTERLEAVE: ... }; statements or if (policy == MPOL_INTERLEAVE) { ... } statements. Such applications would need to use optional mode flags when calling set_mempolicy() or mbind() for these previously implemented statements to stop working. If an application does start using optional mode flags, it will need to mask the optional flags off the policy in switch and conditional statements that only test mode. An additional member is also added to struct shmem_sb_info to store the optional mode flags. [hugh@veritas.com: shmem mpol: fix build warning] Cc: Paul Jackson Cc: Christoph Lameter Cc: Lee Schermerhorn Cc: Andi Kleen Signed-off-by: David Rientjes Signed-off-by: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hugetlbfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 6846785fe904..2e9e5bdd5629 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -504,7 +504,7 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb, uid_t uid, inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; INIT_LIST_HEAD(&inode->i_mapping->private_list); info = HUGETLBFS_I(inode); - mpol_shared_policy_init(&info->policy, MPOL_DEFAULT, NULL); + mpol_shared_policy_init(&info->policy, MPOL_DEFAULT, 0, NULL); switch (mode & S_IFMT) { default: init_special_inode(inode, mode, dev); -- cgit From a10aa579878fc6f9cd17455067380bbdf1d53c91 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Mon, 28 Apr 2008 02:12:40 -0700 Subject: vmalloc: show vmalloced areas via /proc/vmallocinfo Implement a new proc file that allows the display of the currently allocated vmalloc memory. It allows to see the users of vmalloc. That is important if vmalloc space is scarce (i386 for example). And it's going to be important for the compound page fallback to vmalloc. Many of the current users can be switched to use compound pages with fallback. This means that the number of users of vmalloc is reduced and page tables no longer necessary to access the memory. /proc/vmallocinfo allows to review how that reduction occurs. If memory becomes fragmented and larger order allocations are no longer possible then /proc/vmallocinfo allows to see which compound page allocations fell back to virtual compound pages. That is important for new users of virtual compound pages. Such as order 1 stack allocation etc that may fallback to virtual compound pages in the future. /proc/vmallocinfo permissions are made readable-only-by-root to avoid possible information leakage. [akpm@linux-foundation.org: coding-style fixes] [akpm@linux-foundation.org: CONFIG_MMU=n build fix] Signed-off-by: Christoph Lameter Reviewed-by: KOSAKI Motohiro Cc: Hugh Dickins Cc: Nick Piggin Cc: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/proc_misc.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) (limited to 'fs') diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index 2d563979cb02..441a32f0e5f2 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -456,6 +456,20 @@ static const struct file_operations proc_slabstats_operations = { #endif #endif +#ifdef CONFIG_MMU +static int vmalloc_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &vmalloc_op); +} + +static const struct file_operations proc_vmalloc_operations = { + .open = vmalloc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; +#endif + static int show_stat(struct seq_file *p, void *v) { int i; @@ -868,6 +882,9 @@ void __init proc_misc_init(void) #ifdef CONFIG_DEBUG_SLAB_LEAK create_seq_entry("slab_allocators", 0 ,&proc_slabstats_operations); #endif +#endif +#ifdef CONFIG_MMU + proc_create("vmallocinfo", S_IRUSR, NULL, &proc_vmalloc_operations); #endif create_seq_entry("buddyinfo",S_IRUGO, &fragmentation_file_operations); create_seq_entry("pagetypeinfo", S_IRUGO, &pagetypeinfo_file_ops); -- cgit From 214e471ff99064726b2d8af3aa0e24a73c775531 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 28 Apr 2008 02:12:55 -0700 Subject: smaps: account swap entries Show the amount of swap for each vma. This can be used to see where all the swap goes. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Peter Zijlstra Acked-by: Matt Mackall Acked-by: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/task_mmu.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index f4ab76c7c662..7415eeb7cc3a 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -338,8 +338,7 @@ const struct file_operations proc_maps_operations = { #define PSS_SHIFT 12 #ifdef CONFIG_PROC_PAGE_MONITOR -struct mem_size_stats -{ +struct mem_size_stats { struct vm_area_struct *vma; unsigned long resident; unsigned long shared_clean; @@ -347,6 +346,7 @@ struct mem_size_stats unsigned long private_clean; unsigned long private_dirty; unsigned long referenced; + unsigned long swap; u64 pss; }; @@ -363,6 +363,12 @@ static int smaps_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end, pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl); for (; addr != end; pte++, addr += PAGE_SIZE) { ptent = *pte; + + if (is_swap_pte(ptent)) { + mss->swap += PAGE_SIZE; + continue; + } + if (!pte_present(ptent)) continue; @@ -421,7 +427,8 @@ static int show_smap(struct seq_file *m, void *v) "Shared_Dirty: %8lu kB\n" "Private_Clean: %8lu kB\n" "Private_Dirty: %8lu kB\n" - "Referenced: %8lu kB\n", + "Referenced: %8lu kB\n" + "Swap: %8lu kB\n", (vma->vm_end - vma->vm_start) >> 10, mss.resident >> 10, (unsigned long)(mss.pss >> (10 + PSS_SHIFT)), @@ -429,7 +436,8 @@ static int show_smap(struct seq_file *m, void *v) mss.shared_dirty >> 10, mss.private_clean >> 10, mss.private_dirty >> 10, - mss.referenced >> 10); + mss.referenced >> 10, + mss.swap >> 10); return ret; } -- cgit From 30afcb4bd2762fa4b87b17ada9500aa46dc10b1b Mon Sep 17 00:00:00 2001 From: Jared Hulbert Date: Mon, 28 Apr 2008 02:13:02 -0700 Subject: return pfn from direct_access, for XIP Alter the block device ->direct_access() API to work with the new get_xip_mem() API (that requires both kaddr and pfn are returned). Some architectures will not do the right thing in their virt_to_page() for use by XIP (to translate from the kernel virtual address returned by direct_access(), to a user mappable pfn in XIP's page fault handler. However, we can't switch it to just return the pfn and not the kaddr, because we have no good way to get a kva from a pfn, and XIP requires the kva for its read(2) and write(2) handlers. So we have to return both. Signed-off-by: Jared Hulbert Signed-off-by: Nick Piggin Cc: Carsten Otte Cc: Heiko Carstens Cc: linux-mm@kvack.org Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext2/xip.c | 24 ++++++++++++++---------- 1 file changed, 14 insertions(+), 10 deletions(-) (limited to 'fs') diff --git a/fs/ext2/xip.c b/fs/ext2/xip.c index ca7f00312388..430b4c8ee971 100644 --- a/fs/ext2/xip.c +++ b/fs/ext2/xip.c @@ -16,11 +16,13 @@ static inline int __inode_direct_access(struct inode *inode, sector_t sector, - unsigned long *data) + void **kaddr, unsigned long *pfn) { - BUG_ON(!inode->i_sb->s_bdev->bd_disk->fops->direct_access); - return inode->i_sb->s_bdev->bd_disk->fops - ->direct_access(inode->i_sb->s_bdev,sector,data); + struct block_device *bdev = inode->i_sb->s_bdev; + struct block_device_operations *ops = bdev->bd_disk->fops; + + BUG_ON(!ops->direct_access); + return ops->direct_access(bdev, sector, kaddr, pfn); } static inline int @@ -48,12 +50,13 @@ int ext2_clear_xip_target(struct inode *inode, int block) { sector_t sector = block * (PAGE_SIZE/512); - unsigned long data; + void *kaddr; + unsigned long pfn; int rc; - rc = __inode_direct_access(inode, sector, &data); + rc = __inode_direct_access(inode, sector, &kaddr, &pfn); if (!rc) - clear_page((void*)data); + clear_page(kaddr); return rc; } @@ -74,7 +77,8 @@ ext2_get_xip_page(struct address_space *mapping, sector_t offset, int create) { int rc; - unsigned long data; + void *kaddr; + unsigned long pfn; sector_t sector; /* first, retrieve the sector number */ @@ -84,9 +88,9 @@ ext2_get_xip_page(struct address_space *mapping, sector_t offset, /* retrieve address of the target data */ rc = __inode_direct_access - (mapping->host, sector * (PAGE_SIZE/512), &data); + (mapping->host, sector * (PAGE_SIZE/512), &kaddr, &pfn); if (!rc) - return virt_to_page(data); + return pfn_to_page(pfn); error: return ERR_PTR(rc); -- cgit From 70688e4dd1647f0ceb502bbd5964fa344c5eb411 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Mon, 28 Apr 2008 02:13:02 -0700 Subject: xip: support non-struct page backed memory Convert XIP to support non-struct page backed memory, using VM_MIXEDMAP for the user mappings. This requires the get_xip_page API to be changed to an address based one. Improve the API layering a little bit too, while we're here. This is required in order to support XIP filesystems on memory that isn't backed with struct page (but memory with struct page is still supported too). Signed-off-by: Nick Piggin Acked-by: Carsten Otte Cc: Jared Hulbert Cc: Martin Schwidefsky Cc: Heiko Carstens Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext2/inode.c | 2 +- fs/ext2/xip.c | 37 +++++++++++++++---------------------- fs/ext2/xip.h | 9 +++++---- fs/open.c | 2 +- 4 files changed, 22 insertions(+), 28 deletions(-) (limited to 'fs') diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index b8a2990bab83..687023bdfd1e 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -796,7 +796,7 @@ const struct address_space_operations ext2_aops = { const struct address_space_operations ext2_aops_xip = { .bmap = ext2_bmap, - .get_xip_page = ext2_get_xip_page, + .get_xip_mem = ext2_get_xip_mem, }; const struct address_space_operations ext2_nobh_aops = { diff --git a/fs/ext2/xip.c b/fs/ext2/xip.c index 430b4c8ee971..233f7fdbe31d 100644 --- a/fs/ext2/xip.c +++ b/fs/ext2/xip.c @@ -15,26 +15,28 @@ #include "xip.h" static inline int -__inode_direct_access(struct inode *inode, sector_t sector, +__inode_direct_access(struct inode *inode, sector_t block, void **kaddr, unsigned long *pfn) { struct block_device *bdev = inode->i_sb->s_bdev; struct block_device_operations *ops = bdev->bd_disk->fops; + sector_t sector; + + sector = block * (PAGE_SIZE / 512); /* ext2 block to bdev sector */ BUG_ON(!ops->direct_access); return ops->direct_access(bdev, sector, kaddr, pfn); } static inline int -__ext2_get_sector(struct inode *inode, sector_t offset, int create, +__ext2_get_block(struct inode *inode, pgoff_t pgoff, int create, sector_t *result) { struct buffer_head tmp; int rc; memset(&tmp, 0, sizeof(struct buffer_head)); - rc = ext2_get_block(inode, offset/ (PAGE_SIZE/512), &tmp, - create); + rc = ext2_get_block(inode, pgoff, &tmp, create); *result = tmp.b_blocknr; /* did we get a sparse block (hole in the file)? */ @@ -47,14 +49,13 @@ __ext2_get_sector(struct inode *inode, sector_t offset, int create, } int -ext2_clear_xip_target(struct inode *inode, int block) +ext2_clear_xip_target(struct inode *inode, sector_t block) { - sector_t sector = block * (PAGE_SIZE/512); void *kaddr; unsigned long pfn; int rc; - rc = __inode_direct_access(inode, sector, &kaddr, &pfn); + rc = __inode_direct_access(inode, block, &kaddr, &pfn); if (!rc) clear_page(kaddr); return rc; @@ -72,26 +73,18 @@ void ext2_xip_verify_sb(struct super_block *sb) } } -struct page * -ext2_get_xip_page(struct address_space *mapping, sector_t offset, - int create) +int ext2_get_xip_mem(struct address_space *mapping, pgoff_t pgoff, int create, + void **kmem, unsigned long *pfn) { int rc; - void *kaddr; - unsigned long pfn; - sector_t sector; + sector_t block; /* first, retrieve the sector number */ - rc = __ext2_get_sector(mapping->host, offset, create, §or); + rc = __ext2_get_block(mapping->host, pgoff, create, &block); if (rc) - goto error; + return rc; /* retrieve address of the target data */ - rc = __inode_direct_access - (mapping->host, sector * (PAGE_SIZE/512), &kaddr, &pfn); - if (!rc) - return pfn_to_page(pfn); - - error: - return ERR_PTR(rc); + rc = __inode_direct_access(mapping->host, block, kmem, pfn); + return rc; } diff --git a/fs/ext2/xip.h b/fs/ext2/xip.h index aa85331d6c56..18b34d2f31b3 100644 --- a/fs/ext2/xip.h +++ b/fs/ext2/xip.h @@ -7,19 +7,20 @@ #ifdef CONFIG_EXT2_FS_XIP extern void ext2_xip_verify_sb (struct super_block *); -extern int ext2_clear_xip_target (struct inode *, int); +extern int ext2_clear_xip_target (struct inode *, sector_t); static inline int ext2_use_xip (struct super_block *sb) { struct ext2_sb_info *sbi = EXT2_SB(sb); return (sbi->s_mount_opt & EXT2_MOUNT_XIP); } -struct page* ext2_get_xip_page (struct address_space *, sector_t, int); -#define mapping_is_xip(map) unlikely(map->a_ops->get_xip_page) +int ext2_get_xip_mem(struct address_space *, pgoff_t, int, + void **, unsigned long *); +#define mapping_is_xip(map) unlikely(map->a_ops->get_xip_mem) #else #define mapping_is_xip(map) 0 #define ext2_xip_verify_sb(sb) do { } while (0) #define ext2_use_xip(sb) 0 #define ext2_clear_xip_target(inode, chain) 0 -#define ext2_get_xip_page NULL +#define ext2_get_xip_mem NULL #endif diff --git a/fs/open.c b/fs/open.c index b70e7666bb2c..7af1f05d5978 100644 --- a/fs/open.c +++ b/fs/open.c @@ -837,7 +837,7 @@ static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt, if (f->f_flags & O_DIRECT) { if (!f->f_mapping->a_ops || ((!f->f_mapping->a_ops->direct_IO) && - (!f->f_mapping->a_ops->get_xip_page))) { + (!f->f_mapping->a_ops->get_xip_mem))) { fput(f); f = ERR_PTR(-EINVAL); } -- cgit From 71fe804b6d56d6a7aed680e096901434cef6a2c3 Mon Sep 17 00:00:00 2001 From: Lee Schermerhorn Date: Mon, 28 Apr 2008 02:13:26 -0700 Subject: mempolicy: use struct mempolicy pointer in shmem_sb_info This patch replaces the mempolicy mode, mode_flags, and nodemask in the shmem_sb_info struct with a struct mempolicy pointer, initialized to NULL. This removes dependency on the details of mempolicy from shmem.c and hugetlbfs inode.c and simplifies the interfaces. mpol_parse_str() in mempolicy.c is changed to return, via a pointer to a pointer arg, a struct mempolicy pointer on success. For MPOL_DEFAULT, the returned pointer is NULL. Further, mpol_parse_str() now takes a 'no_context' argument that causes the input nodemask to be stored in the w.user_nodemask of the created mempolicy for use when the mempolicy is installed in a tmpfs inode shared policy tree. At that time, any cpuset contextualization is applied to the original input nodemask. This preserves the previous behavior where the input nodemask was stored in the superblock. We can think of the returned mempolicy as "context free". Because mpol_parse_str() is now calling mpol_new(), we can remove from mpol_to_str() the semantic checks that mpol_new() already performs. Add 'no_context' parameter to mpol_to_str() to specify that it should format the nodemask in w.user_nodemask for 'bind' and 'interleave' policies. Change mpol_shared_policy_init() to take a pointer to a "context free" struct mempolicy and to create a new, "contextualized" mempolicy using the mode, mode_flags and user_nodemask from the input mempolicy. Note: we know that the mempolicy passed to mpol_to_str() or mpol_shared_policy_init() from a tmpfs superblock is "context free". This is currently the only instance thereof. However, if we found more uses for this concept, and introduced any ambiguity as to whether a mempolicy was context free or not, we could add another internal mode flag to identify context free mempolicies. Then, we could remove the 'no_context' argument from mpol_to_str(). Added shmem_get_sbmpol() to return a reference counted superblock mempolicy, if one exists, to pass to mpol_shared_policy_init(). We must add the reference under the sb stat_lock to prevent races with replacement of the mpol by remount. This reference is removed in mpol_shared_policy_init(). [akpm@linux-foundation.org: build fix] [akpm@linux-foundation.org: another build fix] [akpm@linux-foundation.org: yet another build fix] Signed-off-by: Lee Schermerhorn Cc: Christoph Lameter Cc: David Rientjes Cc: Mel Gorman Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hugetlbfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 2e9e5bdd5629..9783723e8ffe 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -504,7 +504,7 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb, uid_t uid, inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; INIT_LIST_HEAD(&inode->i_mapping->private_list); info = HUGETLBFS_I(inode); - mpol_shared_policy_init(&info->policy, MPOL_DEFAULT, 0, NULL); + mpol_shared_policy_init(&info->policy, NULL); switch (mode & S_IFMT) { default: init_special_inode(inode, mode, dev); -- cgit From cdf8803768db6f652d498628fe1421a23c025253 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Mon, 28 Apr 2008 02:13:59 -0700 Subject: ncpfs: add prototypes to ncp_fs.h Removes some externs from C files, noticed from the sparse warnings: fs/ncpfs/dir.c:90:26: warning: symbol 'ncp_root_dentry_operations' was not declared. Should it be static? fs/ncpfs/symlink.c:107:5: warning: symbol 'ncp_symlink' was not declared. Should it be static? fs/ncpfs/symlink.c:101:39: warning: symbol 'ncp_symlink_aops' was not declared. Should it be static? Signed-off-by: Harvey Harrison Acked-by: Petr Vandrovec Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ncpfs/inode.c | 6 ------ 1 file changed, 6 deletions(-) (limited to 'fs') diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index fbbb9f7afa1a..2e5ab1204dec 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -107,12 +107,6 @@ static const struct super_operations ncp_sops = .show_options = ncp_show_options, }; -extern struct dentry_operations ncp_root_dentry_operations; -#if defined(CONFIG_NCPFS_EXTRAS) || defined(CONFIG_NCPFS_NFS_NS) -extern const struct address_space_operations ncp_symlink_aops; -extern int ncp_symlink(struct inode*, struct dentry*, const char*); -#endif - /* * Fill in the ncpfs-specific information in the inode. */ -- cgit From 305787e44ebc21d87ab4d4949da5b97d4252aa9b Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Mon, 28 Apr 2008 02:14:00 -0700 Subject: ncpfs: fix sparse warnings in ioctl.c In both cases, these inode variables arebeing used to test the server's root inode against NULL. Change them to s_inode. fs/ncpfs/ioctl.c:391:18: warning: symbol 'inode' shadows an earlier one fs/ncpfs/ioctl.c:264:28: originally declared here fs/ncpfs/ioctl.c:441:17: warning: symbol 'inode' shadows an earlier one fs/ncpfs/ioctl.c:264:28: originally declared here In this case, we are about to return anyway, just reuse result. fs/ncpfs/ioctl.c:521:8: warning: symbol 'result' shadows an earlier one fs/ncpfs/ioctl.c:268:6: originally declared here Signed-off-by: Harvey Harrison Acked-by: Petr Vandrovec Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ncpfs/ioctl.c | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/ncpfs/ioctl.c b/fs/ncpfs/ioctl.c index ad8f167e54bc..3a97c95e1ca2 100644 --- a/fs/ncpfs/ioctl.c +++ b/fs/ncpfs/ioctl.c @@ -389,11 +389,11 @@ static int __ncp_ioctl(struct inode *inode, struct file *filp, struct dentry* dentry = inode->i_sb->s_root; if (dentry) { - struct inode* inode = dentry->d_inode; + struct inode* s_inode = dentry->d_inode; - if (inode) { - sr.volNumber = NCP_FINFO(inode)->volNumber; - sr.dirEntNum = NCP_FINFO(inode)->dirEntNum; + if (s_inode) { + sr.volNumber = NCP_FINFO(s_inode)->volNumber; + sr.dirEntNum = NCP_FINFO(s_inode)->dirEntNum; sr.namespace = server->name_space[sr.volNumber]; } else DPRINTK("ncpfs: s_root->d_inode==NULL\n"); @@ -439,12 +439,12 @@ static int __ncp_ioctl(struct inode *inode, struct file *filp, dentry = inode->i_sb->s_root; server->root_setuped = 1; if (dentry) { - struct inode* inode = dentry->d_inode; + struct inode* s_inode = dentry->d_inode; if (inode) { - NCP_FINFO(inode)->volNumber = vnum; - NCP_FINFO(inode)->dirEntNum = de; - NCP_FINFO(inode)->DosDirNum = dosde; + NCP_FINFO(s_inode)->volNumber = vnum; + NCP_FINFO(s_inode)->dirEntNum = de; + NCP_FINFO(s_inode)->DosDirNum = dosde; } else DPRINTK("ncpfs: s_root->d_inode==NULL\n"); } else @@ -519,7 +519,6 @@ static int __ncp_ioctl(struct inode *inode, struct file *filp, } { struct ncp_lock_ioctl rqdata; - int result; if (copy_from_user(&rqdata, argp, sizeof(rqdata))) return -EFAULT; -- cgit From eee3754f5e45bd27e001ea41823bdbcdd0d192d4 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Mon, 28 Apr 2008 02:14:01 -0700 Subject: ncpfs: fix sparse warning in ncpsign_kernel.c We're casting anyway, might as well cast to the correct sign. Specific to i386 (ifdef __i386__) fs/ncpfs/ncpsign_kernel.c:58:23: warning: incorrect type in initializer (different signedness) fs/ncpfs/ncpsign_kernel.c:58:23: expected unsigned int *data2 fs/ncpfs/ncpsign_kernel.c:58:23: got int * Signed-off-by: Harvey Harrison Acked-by: Petr Vandrovec Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ncpfs/ncpsign_kernel.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ncpfs/ncpsign_kernel.c b/fs/ncpfs/ncpsign_kernel.c index 749a18d33599..7c0b5c21e6cf 100644 --- a/fs/ncpfs/ncpsign_kernel.c +++ b/fs/ncpfs/ncpsign_kernel.c @@ -55,7 +55,7 @@ static void nwsign(char *r_data1, char *r_data2, char *outdata) { unsigned int w0,w1,w2,w3; static int rbit[4]={0, 2, 1, 3}; #ifdef __i386__ - unsigned int *data2=(int *)r_data2; + unsigned int *data2=(unsigned int *)r_data2; #else unsigned int data2[16]; for (i=0;i<16;i++) -- cgit From 338bf9afda91ec005a1e9a0de4af0271cc167d56 Mon Sep 17 00:00:00 2001 From: Andrew Perepechko Date: Mon, 28 Apr 2008 02:14:31 -0700 Subject: quota: do not allow setting of quota limits to too high values We should check whether quota limits set via Q_SETQUOTA are not exceeding limits which quota format is able to handle. Signed-off-by: Andrew Perepechko Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/dquot.c | 22 +++++++++++++++++----- fs/quota_v1.c | 3 +++ fs/quota_v2.c | 3 +++ 3 files changed, 23 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/dquot.c b/fs/dquot.c index 41b9dbd68b0e..24eef582d2a0 100644 --- a/fs/dquot.c +++ b/fs/dquot.c @@ -1709,10 +1709,19 @@ int vfs_get_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *d } /* Generic routine for setting common part of quota structure */ -static void do_set_dqblk(struct dquot *dquot, struct if_dqblk *di) +static int do_set_dqblk(struct dquot *dquot, struct if_dqblk *di) { struct mem_dqblk *dm = &dquot->dq_dqb; int check_blim = 0, check_ilim = 0; + struct mem_dqinfo *dqi = &sb_dqopt(dquot->dq_sb)->info[dquot->dq_type]; + + if ((di->dqb_valid & QIF_BLIMITS && + (di->dqb_bhardlimit > dqi->dqi_maxblimit || + di->dqb_bsoftlimit > dqi->dqi_maxblimit)) || + (di->dqb_valid & QIF_ILIMITS && + (di->dqb_ihardlimit > dqi->dqi_maxilimit || + di->dqb_isoftlimit > dqi->dqi_maxilimit))) + return -ERANGE; spin_lock(&dq_data_lock); if (di->dqb_valid & QIF_SPACE) { @@ -1744,7 +1753,7 @@ static void do_set_dqblk(struct dquot *dquot, struct if_dqblk *di) clear_bit(DQ_BLKS_B, &dquot->dq_flags); } else if (!(di->dqb_valid & QIF_BTIME)) /* Set grace only if user hasn't provided his own... */ - dm->dqb_btime = get_seconds() + sb_dqopt(dquot->dq_sb)->info[dquot->dq_type].dqi_bgrace; + dm->dqb_btime = get_seconds() + dqi->dqi_bgrace; } if (check_ilim) { if (!dm->dqb_isoftlimit || dm->dqb_curinodes < dm->dqb_isoftlimit) { @@ -1752,7 +1761,7 @@ static void do_set_dqblk(struct dquot *dquot, struct if_dqblk *di) clear_bit(DQ_INODES_B, &dquot->dq_flags); } else if (!(di->dqb_valid & QIF_ITIME)) /* Set grace only if user hasn't provided his own... */ - dm->dqb_itime = get_seconds() + sb_dqopt(dquot->dq_sb)->info[dquot->dq_type].dqi_igrace; + dm->dqb_itime = get_seconds() + dqi->dqi_igrace; } if (dm->dqb_bhardlimit || dm->dqb_bsoftlimit || dm->dqb_ihardlimit || dm->dqb_isoftlimit) clear_bit(DQ_FAKE_B, &dquot->dq_flags); @@ -1760,21 +1769,24 @@ static void do_set_dqblk(struct dquot *dquot, struct if_dqblk *di) set_bit(DQ_FAKE_B, &dquot->dq_flags); spin_unlock(&dq_data_lock); mark_dquot_dirty(dquot); + + return 0; } int vfs_set_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *di) { struct dquot *dquot; + int rc; mutex_lock(&sb_dqopt(sb)->dqonoff_mutex); if (!(dquot = dqget(sb, id, type))) { mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex); return -ESRCH; } - do_set_dqblk(dquot, di); + rc = do_set_dqblk(dquot, di); dqput(dquot); mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex); - return 0; + return rc; } /* Generic routine for getting common part of quota file information */ diff --git a/fs/quota_v1.c b/fs/quota_v1.c index f3841f233069..a6cf9269105c 100644 --- a/fs/quota_v1.c +++ b/fs/quota_v1.c @@ -139,6 +139,9 @@ static int v1_read_file_info(struct super_block *sb, int type) goto out; } ret = 0; + /* limits are stored as unsigned 32-bit data */ + dqopt->info[type].dqi_maxblimit = 0xffffffff; + dqopt->info[type].dqi_maxilimit = 0xffffffff; dqopt->info[type].dqi_igrace = dqblk.dqb_itime ? dqblk.dqb_itime : MAX_IQ_TIME; dqopt->info[type].dqi_bgrace = dqblk.dqb_btime ? dqblk.dqb_btime : MAX_DQ_TIME; out: diff --git a/fs/quota_v2.c b/fs/quota_v2.c index c519a583e681..23b647f25d08 100644 --- a/fs/quota_v2.c +++ b/fs/quota_v2.c @@ -59,6 +59,9 @@ static int v2_read_file_info(struct super_block *sb, int type) sb->s_id); return -1; } + /* limits are stored as unsigned 32-bit data */ + info->dqi_maxblimit = 0xffffffff; + info->dqi_maxilimit = 0xffffffff; info->dqi_bgrace = le32_to_cpu(dinfo.dqi_bgrace); info->dqi_igrace = le32_to_cpu(dinfo.dqi_igrace); info->dqi_flags = le32_to_cpu(dinfo.dqi_flags); -- cgit From 8794b5b246cf6f67baf57bd9db386e79ca5cac33 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 28 Apr 2008 02:14:32 -0700 Subject: quota: remove superfluous DQUOT_OFF() in fs/namespace.c We don't need to turn quotas off before remounting root ro, because do_remount_sb() already handles this. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/namespace.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs') diff --git a/fs/namespace.c b/fs/namespace.c index f48f98110c30..fe376805cf5f 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -14,7 +14,6 @@ #include #include #include -#include #include #include #include @@ -1084,7 +1083,6 @@ static int do_umount(struct vfsmount *mnt, int flags) down_write(&sb->s_umount); if (!(sb->s_flags & MS_RDONLY)) { lock_kernel(); - DQUOT_OFF(sb); retval = do_remount_sb(sb, MS_RDONLY, NULL, 0); unlock_kernel(); } -- cgit From 03f6e92bdd467aed9d7571a571868563ae6ad288 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 28 Apr 2008 02:14:32 -0700 Subject: quota: various style cleanups Cleanups in quota code: Change __inline__ to inline. Change some macros to inline functions. Remove vfs_quota_off_mount() macro. DQUOT_OFF() should be (0) is CONFIG_QUOTA is disabled. Move declaration of mark_dquot_dirty and dirty_dquot from quota.h to dquot.c [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/dquot.c | 10 +++++++++- fs/reiserfs/super.c | 2 +- 2 files changed, 10 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/dquot.c b/fs/dquot.c index 24eef582d2a0..fc26d1097d3c 100644 --- a/fs/dquot.c +++ b/fs/dquot.c @@ -289,7 +289,15 @@ static void wait_on_dquot(struct dquot *dquot) mutex_unlock(&dquot->dq_lock); } -#define mark_dquot_dirty(dquot) ((dquot)->dq_sb->dq_op->mark_dirty(dquot)) +static inline int dquot_dirty(struct dquot *dquot) +{ + return test_bit(DQ_MOD_B, &dquot->dq_flags); +} + +static inline int mark_dquot_dirty(struct dquot *dquot) +{ + return dquot->dq_sb->dq_op->mark_dirty(dquot); +} int dquot_mark_dquot_dirty(struct dquot *dquot) { diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 393cc22c1717..3e1972d31e44 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -304,7 +304,7 @@ static int finish_unfinished(struct super_block *s) /* Turn quotas off */ for (i = 0; i < MAXQUOTAS; i++) { if (sb_dqopt(s)->files[i]) - vfs_quota_off_mount(s, i); + vfs_quota_off(s, i); } if (ms_active_set) /* Restore the flag back */ -- cgit From 0ff5af8340aa6be44220d7237ef4a654314cf795 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 28 Apr 2008 02:14:33 -0700 Subject: quota: quota core changes for quotaon on remount Currently, we just turn quotas off on remount of filesystem to read-only state. The patch below adds necessary framework so that we can turn quotas off on remount RO but we are able to automatically reenable them again when filesystem is remounted to RW state. All we need to do is to keep references to inodes of quota files when remounting RO and using these references to reenable quotas when remounting RW. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/dquot.c | 77 ++++++++++++++++++++++++++++++++++++++++++++++++++++++-------- fs/quota.c | 5 ++-- fs/super.c | 10 ++++++-- 3 files changed, 78 insertions(+), 14 deletions(-) (limited to 'fs') diff --git a/fs/dquot.c b/fs/dquot.c index fc26d1097d3c..dfba1623cccb 100644 --- a/fs/dquot.c +++ b/fs/dquot.c @@ -1449,31 +1449,43 @@ static inline void set_enable_flags(struct quota_info *dqopt, int type) switch (type) { case USRQUOTA: dqopt->flags |= DQUOT_USR_ENABLED; + dqopt->flags &= ~DQUOT_USR_SUSPENDED; break; case GRPQUOTA: dqopt->flags |= DQUOT_GRP_ENABLED; + dqopt->flags &= ~DQUOT_GRP_SUSPENDED; break; } } -static inline void reset_enable_flags(struct quota_info *dqopt, int type) +static inline void reset_enable_flags(struct quota_info *dqopt, int type, + int remount) { switch (type) { case USRQUOTA: dqopt->flags &= ~DQUOT_USR_ENABLED; + if (remount) + dqopt->flags |= DQUOT_USR_SUSPENDED; + else + dqopt->flags &= ~DQUOT_USR_SUSPENDED; break; case GRPQUOTA: dqopt->flags &= ~DQUOT_GRP_ENABLED; + if (remount) + dqopt->flags |= DQUOT_GRP_SUSPENDED; + else + dqopt->flags &= ~DQUOT_GRP_SUSPENDED; break; } } + /* * Turn quota off on a device. type == -1 ==> quotaoff for all types (umount) */ -int vfs_quota_off(struct super_block *sb, int type) +int vfs_quota_off(struct super_block *sb, int type, int remount) { - int cnt; + int cnt, ret = 0; struct quota_info *dqopt = sb_dqopt(sb); struct inode *toputinode[MAXQUOTAS]; @@ -1483,9 +1495,17 @@ int vfs_quota_off(struct super_block *sb, int type) toputinode[cnt] = NULL; if (type != -1 && cnt != type) continue; + /* If we keep inodes of quota files after remount and quotaoff + * is called, drop kept inodes. */ + if (!remount && sb_has_quota_suspended(sb, cnt)) { + iput(dqopt->files[cnt]); + dqopt->files[cnt] = NULL; + reset_enable_flags(dqopt, cnt, 0); + continue; + } if (!sb_has_quota_enabled(sb, cnt)) continue; - reset_enable_flags(dqopt, cnt); + reset_enable_flags(dqopt, cnt, remount); /* Note: these are blocking operations */ drop_dquot_ref(sb, cnt); @@ -1501,7 +1521,8 @@ int vfs_quota_off(struct super_block *sb, int type) put_quota_format(dqopt->info[cnt].dqi_format); toputinode[cnt] = dqopt->files[cnt]; - dqopt->files[cnt] = NULL; + if (!remount) + dqopt->files[cnt] = NULL; dqopt->info[cnt].dqi_flags = 0; dqopt->info[cnt].dqi_igrace = 0; dqopt->info[cnt].dqi_bgrace = 0; @@ -1531,12 +1552,19 @@ int vfs_quota_off(struct super_block *sb, int type) mutex_unlock(&toputinode[cnt]->i_mutex); mark_inode_dirty(toputinode[cnt]); } - iput(toputinode[cnt]); mutex_unlock(&dqopt->dqonoff_mutex); + /* On remount RO, we keep the inode pointer so that we + * can reenable quota on the subsequent remount RW. + * But we have better not keep inode pointer when there + * is pending delete on the quota file... */ + if (!remount) + iput(toputinode[cnt]); + else if (!toputinode[cnt]->i_nlink) + ret = -EBUSY; } if (sb->s_bdev) invalidate_bdev(sb->s_bdev); - return 0; + return ret; } /* @@ -1574,7 +1602,8 @@ static int vfs_quota_on_inode(struct inode *inode, int type, int format_id) invalidate_bdev(sb->s_bdev); mutex_lock(&inode->i_mutex); mutex_lock(&dqopt->dqonoff_mutex); - if (sb_has_quota_enabled(sb, type)) { + if (sb_has_quota_enabled(sb, type) || + sb_has_quota_suspended(sb, type)) { error = -EBUSY; goto out_lock; } @@ -1597,6 +1626,7 @@ static int vfs_quota_on_inode(struct inode *inode, int type, int format_id) dqopt->ops[type] = fmt->qf_ops; dqopt->info[type].dqi_format = fmt; + dqopt->info[type].dqi_fmt_id = format_id; INIT_LIST_HEAD(&dqopt->info[type].dqi_dirty_list); mutex_lock(&dqopt->dqio_mutex); if ((error = dqopt->ops[type]->read_file_info(sb, type)) < 0) { @@ -1632,12 +1662,41 @@ out_fmt: return error; } +/* Reenable quotas on remount RW */ +static int vfs_quota_on_remount(struct super_block *sb, int type) +{ + struct quota_info *dqopt = sb_dqopt(sb); + struct inode *inode; + int ret; + + mutex_lock(&dqopt->dqonoff_mutex); + if (!sb_has_quota_suspended(sb, type)) { + mutex_unlock(&dqopt->dqonoff_mutex); + return 0; + } + BUG_ON(sb_has_quota_enabled(sb, type)); + + inode = dqopt->files[type]; + dqopt->files[type] = NULL; + reset_enable_flags(dqopt, type, 0); + mutex_unlock(&dqopt->dqonoff_mutex); + + ret = vfs_quota_on_inode(inode, type, dqopt->info[type].dqi_fmt_id); + iput(inode); + + return ret; +} + /* Actual function called from quotactl() */ -int vfs_quota_on(struct super_block *sb, int type, int format_id, char *path) +int vfs_quota_on(struct super_block *sb, int type, int format_id, char *path, + int remount) { struct nameidata nd; int error; + if (remount) + return vfs_quota_on_remount(sb, type); + error = path_lookup(path, LOOKUP_FOLLOW, &nd); if (error < 0) return error; diff --git a/fs/quota.c b/fs/quota.c index 84f28dd72116..db1cc9f3c7aa 100644 --- a/fs/quota.c +++ b/fs/quota.c @@ -69,7 +69,6 @@ static int generic_quotactl_valid(struct super_block *sb, int type, int cmd, qid switch (cmd) { case Q_GETFMT: case Q_GETINFO: - case Q_QUOTAOFF: case Q_SETINFO: case Q_SETQUOTA: case Q_GETQUOTA: @@ -229,12 +228,12 @@ static int do_quotactl(struct super_block *sb, int type, int cmd, qid_t id, void if (IS_ERR(pathname = getname(addr))) return PTR_ERR(pathname); - ret = sb->s_qcop->quota_on(sb, type, id, pathname); + ret = sb->s_qcop->quota_on(sb, type, id, pathname, 0); putname(pathname); return ret; } case Q_QUOTAOFF: - return sb->s_qcop->quota_off(sb, type); + return sb->s_qcop->quota_off(sb, type, 0); case Q_GETFMT: { __u32 fmt; diff --git a/fs/super.c b/fs/super.c index 4798350b2bc9..a5a4aca7e22f 100644 --- a/fs/super.c +++ b/fs/super.c @@ -179,7 +179,7 @@ void deactivate_super(struct super_block *s) if (atomic_dec_and_lock(&s->s_active, &sb_lock)) { s->s_count -= S_BIAS-1; spin_unlock(&sb_lock); - DQUOT_OFF(s); + DQUOT_OFF(s, 0); down_write(&s->s_umount); fs->kill_sb(s); put_filesystem(fs); @@ -608,6 +608,7 @@ retry: int do_remount_sb(struct super_block *sb, int flags, void *data, int force) { int retval; + int remount_rw; #ifdef CONFIG_BLOCK if (!(flags & MS_RDONLY) && bdev_read_only(sb->s_bdev)) @@ -625,8 +626,11 @@ int do_remount_sb(struct super_block *sb, int flags, void *data, int force) mark_files_ro(sb); else if (!fs_may_remount_ro(sb)) return -EBUSY; - DQUOT_OFF(sb); + retval = DQUOT_OFF(sb, 1); + if (retval < 0 && retval != -ENOSYS) + return -EBUSY; } + remount_rw = !(flags & MS_RDONLY) && (sb->s_flags & MS_RDONLY); if (sb->s_op->remount_fs) { lock_super(sb); @@ -636,6 +640,8 @@ int do_remount_sb(struct super_block *sb, int flags, void *data, int force) return retval; } sb->s_flags = (sb->s_flags & ~MS_RMT_MASK) | (flags & MS_RMT_MASK); + if (remount_rw) + DQUOT_ON_REMOUNT(sb); return 0; } -- cgit From 2fd83a4f3cd5a725168e3a269746dfce2adfa56a Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 28 Apr 2008 02:14:34 -0700 Subject: quota: ext3: make ext3 handle quotaon on remount Update ext3 handle quotaon on remount RW. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/super.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/ext3/super.c b/fs/ext3/super.c index ad5360664082..883ff965d984 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -685,7 +685,8 @@ static int ext3_acquire_dquot(struct dquot *dquot); static int ext3_release_dquot(struct dquot *dquot); static int ext3_mark_dquot_dirty(struct dquot *dquot); static int ext3_write_info(struct super_block *sb, int type); -static int ext3_quota_on(struct super_block *sb, int type, int format_id, char *path); +static int ext3_quota_on(struct super_block *sb, int type, int format_id, + char *path, int remount); static int ext3_quota_on_mount(struct super_block *sb, int type); static ssize_t ext3_quota_read(struct super_block *sb, int type, char *data, size_t len, loff_t off); @@ -1415,7 +1416,7 @@ static void ext3_orphan_cleanup (struct super_block * sb, /* Turn quotas off */ for (i = 0; i < MAXQUOTAS; i++) { if (sb_dqopt(sb)->files[i]) - vfs_quota_off(sb, i); + vfs_quota_off(sb, i, 0); } #endif sb->s_flags = s_flags; /* Restore MS_RDONLY status */ @@ -2743,17 +2744,17 @@ static int ext3_quota_on_mount(struct super_block *sb, int type) * Standard function to be called on quota_on */ static int ext3_quota_on(struct super_block *sb, int type, int format_id, - char *path) + char *path, int remount) { int err; struct nameidata nd; if (!test_opt(sb, QUOTA)) return -EINVAL; - /* Not journalling quota? */ - if (!EXT3_SB(sb)->s_qf_names[USRQUOTA] && - !EXT3_SB(sb)->s_qf_names[GRPQUOTA]) - return vfs_quota_on(sb, type, format_id, path); + /* Not journalling quota or remount? */ + if ((!EXT3_SB(sb)->s_qf_names[USRQUOTA] && + !EXT3_SB(sb)->s_qf_names[GRPQUOTA]) || remount) + return vfs_quota_on(sb, type, format_id, path, remount); err = path_lookup(path, LOOKUP_FOLLOW, &nd); if (err) return err; @@ -2762,13 +2763,13 @@ static int ext3_quota_on(struct super_block *sb, int type, int format_id, path_put(&nd.path); return -EXDEV; } - /* Quotafile not of fs root? */ + /* Quotafile not in fs root? */ if (nd.path.dentry->d_parent->d_inode != sb->s_root->d_inode) printk(KERN_WARNING "EXT3-fs: Quota file not on filesystem root. " "Journalled quota will not work.\n"); path_put(&nd.path); - return vfs_quota_on(sb, type, format_id, path); + return vfs_quota_on(sb, type, format_id, path, remount); } /* Read data from quotafile - avoid pagecache and such because we cannot afford -- cgit From 6f28e08794749f3431e89302728e612343d9d9e4 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 28 Apr 2008 02:14:34 -0700 Subject: quota: ext4: make ext4 handle quotaon on remount Update ext4 to handle quotaon on remount RW. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext4/super.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 13383ba18f1d..c81a8e759bad 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -813,7 +813,8 @@ static int ext4_acquire_dquot(struct dquot *dquot); static int ext4_release_dquot(struct dquot *dquot); static int ext4_mark_dquot_dirty(struct dquot *dquot); static int ext4_write_info(struct super_block *sb, int type); -static int ext4_quota_on(struct super_block *sb, int type, int format_id, char *path); +static int ext4_quota_on(struct super_block *sb, int type, int format_id, + char *path, int remount); static int ext4_quota_on_mount(struct super_block *sb, int type); static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data, size_t len, loff_t off); @@ -1632,7 +1633,7 @@ static void ext4_orphan_cleanup (struct super_block * sb, /* Turn quotas off */ for (i = 0; i < MAXQUOTAS; i++) { if (sb_dqopt(sb)->files[i]) - vfs_quota_off(sb, i); + vfs_quota_off(sb, i, 0); } #endif sb->s_flags = s_flags; /* Restore MS_RDONLY status */ @@ -3143,7 +3144,7 @@ static int ext4_quota_on_mount(struct super_block *sb, int type) * Standard function to be called on quota_on */ static int ext4_quota_on(struct super_block *sb, int type, int format_id, - char *path) + char *path, int remount) { int err; struct nameidata nd; @@ -3151,9 +3152,9 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id, if (!test_opt(sb, QUOTA)) return -EINVAL; /* Not journalling quota? */ - if (!EXT4_SB(sb)->s_qf_names[USRQUOTA] && - !EXT4_SB(sb)->s_qf_names[GRPQUOTA]) - return vfs_quota_on(sb, type, format_id, path); + if ((!EXT4_SB(sb)->s_qf_names[USRQUOTA] && + !EXT4_SB(sb)->s_qf_names[GRPQUOTA]) || remount) + return vfs_quota_on(sb, type, format_id, path, remount); err = path_lookup(path, LOOKUP_FOLLOW, &nd); if (err) return err; @@ -3168,7 +3169,7 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id, "EXT4-fs: Quota file not on filesystem root. " "Journalled quota will not work.\n"); path_put(&nd.path); - return vfs_quota_on(sb, type, format_id, path); + return vfs_quota_on(sb, type, format_id, path, remount); } /* Read data from quotafile - avoid pagecache and such because we cannot afford -- cgit From 1b445a9c21f593be9d3c4ab912359d2c51c371dd Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 28 Apr 2008 02:14:35 -0700 Subject: quota: reiserfs: make reiserfs handle quotaon on remount Update reiserfs to handle quotaon on remount RW. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/super.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 3e1972d31e44..f8f9a473e670 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -304,7 +304,7 @@ static int finish_unfinished(struct super_block *s) /* Turn quotas off */ for (i = 0; i < MAXQUOTAS; i++) { if (sb_dqopt(s)->files[i]) - vfs_quota_off(s, i); + vfs_quota_off(s, i, 0); } if (ms_active_set) /* Restore the flag back */ @@ -634,7 +634,7 @@ static int reiserfs_acquire_dquot(struct dquot *); static int reiserfs_release_dquot(struct dquot *); static int reiserfs_mark_dquot_dirty(struct dquot *); static int reiserfs_write_info(struct super_block *, int); -static int reiserfs_quota_on(struct super_block *, int, int, char *); +static int reiserfs_quota_on(struct super_block *, int, int, char *, int); static struct dquot_operations reiserfs_quota_operations = { .initialize = reiserfs_dquot_initialize, @@ -2015,13 +2015,16 @@ static int reiserfs_quota_on_mount(struct super_block *sb, int type) * Standard function to be called on quota_on */ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id, - char *path) + char *path, int remount) { int err; struct nameidata nd; if (!(REISERFS_SB(sb)->s_mount_opt & (1 << REISERFS_QUOTA))) return -EINVAL; + /* No more checks needed? Path and format_id are bogus anyway... */ + if (remount) + return vfs_quota_on(sb, type, format_id, path, 1); err = path_lookup(path, LOOKUP_FOLLOW, &nd); if (err) return err; @@ -2041,7 +2044,7 @@ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id, if (!REISERFS_SB(sb)->s_qf_names[USRQUOTA] && !REISERFS_SB(sb)->s_qf_names[GRPQUOTA]) { path_put(&nd.path); - return vfs_quota_on(sb, type, format_id, path); + return vfs_quota_on(sb, type, format_id, path, 0); } /* Quotafile not of fs root? */ if (nd.path.dentry->d_parent->d_inode != sb->s_root->d_inode) @@ -2049,7 +2052,7 @@ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id, "reiserfs: Quota file not on filesystem root. " "Journalled quota will not work."); path_put(&nd.path); - return vfs_quota_on(sb, type, format_id, path); + return vfs_quota_on(sb, type, format_id, path, 0); } /* Read data from quotafile - avoid pagecache and such because we cannot afford -- cgit From fba4d3997f4c3dc6de927f7f516168580033ae1a Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Mon, 28 Apr 2008 02:15:59 -0700 Subject: ext2: le*_add_cpu conversion replace all: little_endian_variable = cpu_to_leX(leX_to_cpu(little_endian_variable) + expression_in_cpu_byteorder); with: leX_add_cpu(&little_endian_variable, expression_in_cpu_byteorder); generated with semantic patch Signed-off-by: Marcin Slusarz Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext2/ialloc.c | 12 ++++-------- fs/ext2/super.c | 2 +- fs/ext2/xattr.c | 9 +++------ 3 files changed, 8 insertions(+), 15 deletions(-) (limited to 'fs') diff --git a/fs/ext2/ialloc.c b/fs/ext2/ialloc.c index 08f647d8188d..f59741346760 100644 --- a/fs/ext2/ialloc.c +++ b/fs/ext2/ialloc.c @@ -75,11 +75,9 @@ static void ext2_release_inode(struct super_block *sb, int group, int dir) } spin_lock(sb_bgl_lock(EXT2_SB(sb), group)); - desc->bg_free_inodes_count = - cpu_to_le16(le16_to_cpu(desc->bg_free_inodes_count) + 1); + le16_add_cpu(&desc->bg_free_inodes_count, 1); if (dir) - desc->bg_used_dirs_count = - cpu_to_le16(le16_to_cpu(desc->bg_used_dirs_count) - 1); + le16_add_cpu(&desc->bg_used_dirs_count, -1); spin_unlock(sb_bgl_lock(EXT2_SB(sb), group)); if (dir) percpu_counter_dec(&EXT2_SB(sb)->s_dirs_counter); @@ -539,13 +537,11 @@ got: percpu_counter_inc(&sbi->s_dirs_counter); spin_lock(sb_bgl_lock(sbi, group)); - gdp->bg_free_inodes_count = - cpu_to_le16(le16_to_cpu(gdp->bg_free_inodes_count) - 1); + le16_add_cpu(&gdp->bg_free_inodes_count, -1); if (S_ISDIR(mode)) { if (sbi->s_debts[group] < 255) sbi->s_debts[group]++; - gdp->bg_used_dirs_count = - cpu_to_le16(le16_to_cpu(gdp->bg_used_dirs_count) + 1); + le16_add_cpu(&gdp->bg_used_dirs_count, 1); } else { if (sbi->s_debts[group]) sbi->s_debts[group]--; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 088b011bb97e..40d934c3e403 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -604,7 +604,7 @@ static int ext2_setup_super (struct super_block * sb, "running e2fsck is recommended\n"); if (!le16_to_cpu(es->s_max_mnt_count)) es->s_max_mnt_count = cpu_to_le16(EXT2_DFL_MAX_MNT_COUNT); - es->s_mnt_count=cpu_to_le16(le16_to_cpu(es->s_mnt_count) + 1); + le16_add_cpu(&es->s_mnt_count, 1); ext2_write_super(sb); if (test_opt (sb, DEBUG)) printk ("[EXT II FS %s, %s, bs=%lu, fs=%lu, gc=%lu, " diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c index a99d46f3b26e..b86f02747eaf 100644 --- a/fs/ext2/xattr.c +++ b/fs/ext2/xattr.c @@ -646,8 +646,7 @@ ext2_xattr_set2(struct inode *inode, struct buffer_head *old_bh, unlock_buffer(new_bh); goto cleanup; } - HDR(new_bh)->h_refcount = cpu_to_le32(1 + - le32_to_cpu(HDR(new_bh)->h_refcount)); + le32_add_cpu(&HDR(new_bh)->h_refcount, 1); ea_bdebug(new_bh, "refcount now=%d", le32_to_cpu(HDR(new_bh)->h_refcount)); } @@ -731,8 +730,7 @@ ext2_xattr_set2(struct inode *inode, struct buffer_head *old_bh, bforget(old_bh); } else { /* Decrement the refcount only. */ - HDR(old_bh)->h_refcount = cpu_to_le32( - le32_to_cpu(HDR(old_bh)->h_refcount) - 1); + le32_add_cpu(&HDR(old_bh)->h_refcount, -1); if (ce) mb_cache_entry_release(ce); DQUOT_FREE_BLOCK(inode, 1); @@ -789,8 +787,7 @@ ext2_xattr_delete_inode(struct inode *inode) bforget(bh); unlock_buffer(bh); } else { - HDR(bh)->h_refcount = cpu_to_le32( - le32_to_cpu(HDR(bh)->h_refcount) - 1); + le32_add_cpu(&HDR(bh)->h_refcount, -1); if (ce) mb_cache_entry_release(ce); ea_bdebug(bh, "refcount now=%d", -- cgit From 31f68e130149809c5d49583e1ff225dd5d404d83 Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Mon, 28 Apr 2008 02:16:00 -0700 Subject: ext2: convert byte order of constant instead of variable Convert byte order of constant instead of variable which can be done at compile time (vs run time). Signed-off-by: Marcin Slusarz Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext2/super.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 40d934c3e403..7e6867329220 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -51,8 +51,7 @@ void ext2_error (struct super_block * sb, const char * function, if (!(sb->s_flags & MS_RDONLY)) { sbi->s_mount_state |= EXT2_ERROR_FS; - es->s_state = - cpu_to_le16(le16_to_cpu(es->s_state) | EXT2_ERROR_FS); + es->s_state |= cpu_to_le16(EXT2_ERROR_FS); ext2_sync_super(sb, es); } @@ -1126,10 +1125,9 @@ void ext2_write_super (struct super_block * sb) if (!(sb->s_flags & MS_RDONLY)) { es = EXT2_SB(sb)->s_es; - if (le16_to_cpu(es->s_state) & EXT2_VALID_FS) { + if (es->s_state & cpu_to_le16(EXT2_VALID_FS)) { ext2_debug ("setting valid to 0\n"); - es->s_state = cpu_to_le16(le16_to_cpu(es->s_state) & - ~EXT2_VALID_FS); + es->s_state &= cpu_to_le16(~EXT2_VALID_FS); es->s_free_blocks_count = cpu_to_le32(ext2_count_free_blocks(sb)); es->s_free_inodes_count = cpu_to_le32(ext2_count_free_inodes(sb)); es->s_mtime = cpu_to_le32(get_seconds()); -- cgit From bbff28602436cc7ca660757bba81f03b99e8586d Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 28 Apr 2008 02:16:00 -0700 Subject: ext2: improve ext2_readdir() return value Improve ext2_readdir() return value for ext2_get_page() failure by using the actual result of ext2_get_page(). Signed-off-by: Akinobu Mita Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext2/dir.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext2/dir.c b/fs/ext2/dir.c index 8dededd80fe2..a0995eb2dc23 100644 --- a/fs/ext2/dir.c +++ b/fs/ext2/dir.c @@ -299,7 +299,7 @@ ext2_readdir (struct file * filp, void * dirent, filldir_t filldir) "bad page in #%lu", inode->i_ino); filp->f_pos += PAGE_CACHE_SIZE - offset; - return -EIO; + return PTR_ERR(page); } kaddr = page_address(page); if (unlikely(need_revalidate)) { -- cgit From 24097d12efbb97bff14fb6e350508853db0f9595 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 28 Apr 2008 02:16:01 -0700 Subject: ext2: use ext2_group_first_block_no() Use ext2_group_first_block_no() and assign the return values to ext2_fsblk_t variables. Signed-off-by: Akinobu Mita Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext2/inode.c | 3 +-- fs/ext2/super.c | 11 +++++------ fs/ext2/xattr.c | 6 ++---- 3 files changed, 8 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index 687023bdfd1e..be0d02ac7ac4 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -275,8 +275,7 @@ static unsigned long ext2_find_near(struct inode *inode, Indirect *ind) * It is going to be refered from inode itself? OK, just put it into * the same cylinder group then. */ - bg_start = (ei->i_block_group * EXT2_BLOCKS_PER_GROUP(inode->i_sb)) + - le32_to_cpu(EXT2_SB(inode->i_sb)->s_es->s_first_data_block); + bg_start = ext2_group_first_block_no(inode->i_sb, ei->i_block_group); colour = (current->pid % 16) * (EXT2_BLOCKS_PER_GROUP(inode->i_sb) / 16); return bg_start + colour; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 7e6867329220..3f745fb5fff3 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -621,13 +621,13 @@ static int ext2_check_descriptors(struct super_block *sb) { int i; struct ext2_sb_info *sbi = EXT2_SB(sb); - unsigned long first_block = le32_to_cpu(sbi->s_es->s_first_data_block); - unsigned long last_block; ext2_debug ("Checking group descriptors"); for (i = 0; i < sbi->s_groups_count; i++) { struct ext2_group_desc *gdp = ext2_get_group_desc(sb, i, NULL); + ext2_fsblk_t first_block = ext2_group_first_block_no(sb, i); + ext2_fsblk_t last_block; if (i == sbi->s_groups_count - 1) last_block = le32_to_cpu(sbi->s_es->s_blocks_count) - 1; @@ -663,7 +663,6 @@ static int ext2_check_descriptors(struct super_block *sb) i, (unsigned long) le32_to_cpu(gdp->bg_inode_table)); return 0; } - first_block += EXT2_BLOCKS_PER_GROUP(sb); } return 1; } @@ -720,10 +719,9 @@ static unsigned long descriptor_loc(struct super_block *sb, int nr) { struct ext2_sb_info *sbi = EXT2_SB(sb); - unsigned long bg, first_data_block, first_meta_bg; + unsigned long bg, first_meta_bg; int has_super = 0; - first_data_block = le32_to_cpu(sbi->s_es->s_first_data_block); first_meta_bg = le32_to_cpu(sbi->s_es->s_first_meta_bg); if (!EXT2_HAS_INCOMPAT_FEATURE(sb, EXT2_FEATURE_INCOMPAT_META_BG) || @@ -732,7 +730,8 @@ static unsigned long descriptor_loc(struct super_block *sb, bg = sbi->s_desc_per_block * nr; if (ext2_bg_has_super(sb, bg)) has_super = 1; - return (first_data_block + has_super + (bg * sbi->s_blocks_per_group)); + + return ext2_group_first_block_no(sb, bg) + has_super; } static int ext2_fill_super(struct super_block *sb, void *data, int silent) diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c index b86f02747eaf..987a5261cc2e 100644 --- a/fs/ext2/xattr.c +++ b/fs/ext2/xattr.c @@ -659,10 +659,8 @@ ext2_xattr_set2(struct inode *inode, struct buffer_head *old_bh, ext2_xattr_cache_insert(new_bh); } else { /* We need to allocate a new block */ - int goal = le32_to_cpu(EXT2_SB(sb)->s_es-> - s_first_data_block) + - EXT2_I(inode)->i_block_group * - EXT2_BLOCKS_PER_GROUP(sb); + ext2_fsblk_t goal = ext2_group_first_block_no(sb, + EXT2_I(inode)->i_block_group); int block = ext2_new_block(inode, goal, &error); if (error) goto cleanup; -- cgit From 4c8b3125f8dcefd1ee2990113fe69cfa9e9fa609 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 28 Apr 2008 02:16:02 -0700 Subject: ext2: use ext2_fsblk_t type Use ext2_fsblk_t type for filesystem-wide blocks number Signed-off-by: Akinobu Mita Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext2/inode.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index be0d02ac7ac4..384fc0d1dd74 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -254,13 +254,13 @@ no_block: * Caller must make sure that @ind is valid and will stay that way. */ -static unsigned long ext2_find_near(struct inode *inode, Indirect *ind) +static ext2_fsblk_t ext2_find_near(struct inode *inode, Indirect *ind) { struct ext2_inode_info *ei = EXT2_I(inode); __le32 *start = ind->bh ? (__le32 *) ind->bh->b_data : ei->i_data; __le32 *p; - unsigned long bg_start; - unsigned long colour; + ext2_fsblk_t bg_start; + ext2_fsblk_t colour; /* Try to find previous block */ for (p = ind->p - 1; p >= start; p--) @@ -290,8 +290,8 @@ static unsigned long ext2_find_near(struct inode *inode, Indirect *ind) * Returns preferred place for a block (the goal). */ -static inline int ext2_find_goal(struct inode *inode, long block, - Indirect *partial) +static inline ext2_fsblk_t ext2_find_goal(struct inode *inode, long block, + Indirect *partial) { struct ext2_block_alloc_info *block_i; -- cgit From 2c11619a590e13fecc801c9c50859cba29740f38 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Mon, 28 Apr 2008 02:16:02 -0700 Subject: fs/ext2: use BUG_ON if (...) BUG(); should be replaced with BUG_ON(...) when the test has no side-effects to allow a definition of BUG_ON that drops the code completely. The semantic patch that makes this change is as follows: (http://www.emn.fr/x-info/coccinelle/) // @ disable unlikely @ expression E,f; @@ ( if (<... f(...) ...>) { BUG(); } | - if (unlikely(E)) { BUG(); } + BUG_ON(E); ) @@ expression E,f; @@ ( if (<... f(...) ...>) { BUG(); } | - if (E) { BUG(); } + BUG_ON(E); ) // Signed-off-by: Julia Lawall Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext2/balloc.c | 3 +-- fs/ext2/dir.c | 4 ++-- 2 files changed, 3 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/ext2/balloc.c b/fs/ext2/balloc.c index e7b2bafa1dd9..c834f512dce5 100644 --- a/fs/ext2/balloc.c +++ b/fs/ext2/balloc.c @@ -245,8 +245,7 @@ restart: prev = rsv; } printk("Window map complete.\n"); - if (bad) - BUG(); + BUG_ON(bad); } #define rsv_window_dump(root, verbose) \ __rsv_window_dump((root), (verbose), __FUNCTION__) diff --git a/fs/ext2/dir.c b/fs/ext2/dir.c index a0995eb2dc23..dd9dbb25eab2 100644 --- a/fs/ext2/dir.c +++ b/fs/ext2/dir.c @@ -41,8 +41,8 @@ static inline __le16 ext2_rec_len_to_disk(unsigned len) { if (len == (1 << 16)) return cpu_to_le16(EXT2_MAX_REC_LEN); - else if (len > (1 << 16)) - BUG(); + else + BUG_ON(len > (1 << 16)); return cpu_to_le16(len); } -- cgit From 605afd60ef6dde254fc9b8667c5bccebade251c3 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Mon, 28 Apr 2008 02:16:03 -0700 Subject: ext2: replace remaining __FUNCTION__ occurrences __FUNCTION__ is gcc-specific, use __func__ Signed-off-by: Harvey Harrison Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext2/balloc.c | 10 +++++----- fs/ext2/dir.c | 14 +++++++------- fs/ext2/super.c | 6 +++--- fs/ext2/xip.c | 2 +- 4 files changed, 16 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/ext2/balloc.c b/fs/ext2/balloc.c index c834f512dce5..25dec8633c94 100644 --- a/fs/ext2/balloc.c +++ b/fs/ext2/balloc.c @@ -106,7 +106,7 @@ static int ext2_valid_block_bitmap(struct super_block *sb, return 1; err_out: - ext2_error(sb, __FUNCTION__, + ext2_error(sb, __func__, "Invalid block bitmap - " "block_group = %d, block = %lu", block_group, bitmap_blk); @@ -132,7 +132,7 @@ read_block_bitmap(struct super_block *sb, unsigned int block_group) bitmap_blk = le32_to_cpu(desc->bg_block_bitmap); bh = sb_getblk(sb, bitmap_blk); if (unlikely(!bh)) { - ext2_error(sb, __FUNCTION__, + ext2_error(sb, __func__, "Cannot read block bitmap - " "block_group = %d, block_bitmap = %u", block_group, le32_to_cpu(desc->bg_block_bitmap)); @@ -143,7 +143,7 @@ read_block_bitmap(struct super_block *sb, unsigned int block_group) if (bh_submit_read(bh) < 0) { brelse(bh); - ext2_error(sb, __FUNCTION__, + ext2_error(sb, __func__, "Cannot read block bitmap - " "block_group = %d, block_bitmap = %u", block_group, le32_to_cpu(desc->bg_block_bitmap)); @@ -248,7 +248,7 @@ restart: BUG_ON(bad); } #define rsv_window_dump(root, verbose) \ - __rsv_window_dump((root), (verbose), __FUNCTION__) + __rsv_window_dump((root), (verbose), __func__) #else #define rsv_window_dump(root, verbose) do {} while (0) #endif @@ -547,7 +547,7 @@ do_more: for (i = 0, group_freed = 0; i < count; i++) { if (!ext2_clear_bit_atomic(sb_bgl_lock(sbi, block_group), bit + i, bitmap_bh->b_data)) { - ext2_error(sb, __FUNCTION__, + ext2_error(sb, __func__, "bit already cleared for block %lu", block + i); } else { group_freed++; diff --git a/fs/ext2/dir.c b/fs/ext2/dir.c index dd9dbb25eab2..a78c6b4af060 100644 --- a/fs/ext2/dir.c +++ b/fs/ext2/dir.c @@ -295,7 +295,7 @@ ext2_readdir (struct file * filp, void * dirent, filldir_t filldir) struct page *page = ext2_get_page(inode, n); if (IS_ERR(page)) { - ext2_error(sb, __FUNCTION__, + ext2_error(sb, __func__, "bad page in #%lu", inode->i_ino); filp->f_pos += PAGE_CACHE_SIZE - offset; @@ -314,7 +314,7 @@ ext2_readdir (struct file * filp, void * dirent, filldir_t filldir) limit = kaddr + ext2_last_byte(inode, n) - EXT2_DIR_REC_LEN(1); for ( ;(char*)de <= limit; de = ext2_next_entry(de)) { if (de->rec_len == 0) { - ext2_error(sb, __FUNCTION__, + ext2_error(sb, __func__, "zero-length directory entry"); ext2_put_page(page); return -EIO; @@ -381,7 +381,7 @@ struct ext2_dir_entry_2 * ext2_find_entry (struct inode * dir, kaddr += ext2_last_byte(dir, n) - reclen; while ((char *) de <= kaddr) { if (de->rec_len == 0) { - ext2_error(dir->i_sb, __FUNCTION__, + ext2_error(dir->i_sb, __func__, "zero-length directory entry"); ext2_put_page(page); goto out; @@ -396,7 +396,7 @@ struct ext2_dir_entry_2 * ext2_find_entry (struct inode * dir, n = 0; /* next page is past the blocks we've got */ if (unlikely(n > (dir->i_blocks >> (PAGE_CACHE_SHIFT - 9)))) { - ext2_error(dir->i_sb, __FUNCTION__, + ext2_error(dir->i_sb, __func__, "dir %lu size %lld exceeds block count %llu", dir->i_ino, dir->i_size, (unsigned long long)dir->i_blocks); @@ -506,7 +506,7 @@ int ext2_add_link (struct dentry *dentry, struct inode *inode) goto got_it; } if (de->rec_len == 0) { - ext2_error(dir->i_sb, __FUNCTION__, + ext2_error(dir->i_sb, __func__, "zero-length directory entry"); err = -EIO; goto out_unlock; @@ -578,7 +578,7 @@ int ext2_delete_entry (struct ext2_dir_entry_2 * dir, struct page * page ) while ((char*)de < (char*)dir) { if (de->rec_len == 0) { - ext2_error(inode->i_sb, __FUNCTION__, + ext2_error(inode->i_sb, __func__, "zero-length directory entry"); err = -EIO; goto out; @@ -670,7 +670,7 @@ int ext2_empty_dir (struct inode * inode) while ((char *)de <= kaddr) { if (de->rec_len == 0) { - ext2_error(inode->i_sb, __FUNCTION__, + ext2_error(inode->i_sb, __func__, "zero-length directory entry"); printk("kaddr=%p, de=%p\n", kaddr, de); goto not_empty; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 3f745fb5fff3..ef50cbc792db 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -89,7 +89,7 @@ void ext2_update_dynamic_rev(struct super_block *sb) if (le32_to_cpu(es->s_rev_level) > EXT2_GOOD_OLD_REV) return; - ext2_warning(sb, __FUNCTION__, + ext2_warning(sb, __func__, "updating to rev %d because of new feature flag, " "running e2fsck is recommended", EXT2_DYNAMIC_REV); @@ -1060,7 +1060,7 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) goto failed_mount3; } if (EXT2_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_HAS_JOURNAL)) - ext2_warning(sb, __FUNCTION__, + ext2_warning(sb, __func__, "mounting ext3 filesystem as ext2"); ext2_setup_super (sb, es, sb->s_flags & MS_RDONLY); return 0; @@ -1177,7 +1177,7 @@ static int ext2_remount (struct super_block * sb, int * flags, char * data) if (((sbi->s_mount_opt & EXT2_MOUNT_XIP) != (old_mount_opt & EXT2_MOUNT_XIP)) && invalidate_inodes(sb)) - ext2_warning(sb, __FUNCTION__, "busy inodes while remounting "\ + ext2_warning(sb, __func__, "busy inodes while remounting "\ "xip remain in cache (no functional problem)"); if ((*flags & MS_RDONLY) == (sb->s_flags & MS_RDONLY)) return 0; diff --git a/fs/ext2/xip.c b/fs/ext2/xip.c index 233f7fdbe31d..4fb94c20041b 100644 --- a/fs/ext2/xip.c +++ b/fs/ext2/xip.c @@ -68,7 +68,7 @@ void ext2_xip_verify_sb(struct super_block *sb) if ((sbi->s_mount_opt & EXT2_MOUNT_XIP) && !sb->s_bdev->bd_disk->fops->direct_access) { sbi->s_mount_opt &= (~EXT2_MOUNT_XIP); - ext2_warning(sb, __FUNCTION__, + ext2_warning(sb, __func__, "ignoring xip option - not supported by bdev"); } } -- cgit From 8b91582500ae750db22bd515379616e5e9ad06ab Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Mon, 28 Apr 2008 02:16:04 -0700 Subject: ext2: retry block allocation if new blocks are allocated from system zone If the block allocator gets blocks out of system zone ext2 calls ext2_error. But if the file system is mounted with errors=continue retry block allocation. We need to mark the system zone blocks as in use to make sure retry don't pick them again System zone is the block range mapping block bitmap, inode bitmap and inode table. [akpm@linux-foundation.org: fix typo in comment] Signed-off-by: Aneesh Kumar K.V Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext2/balloc.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/ext2/balloc.c b/fs/ext2/balloc.c index 25dec8633c94..10bb02c3f25c 100644 --- a/fs/ext2/balloc.c +++ b/fs/ext2/balloc.c @@ -149,11 +149,12 @@ read_block_bitmap(struct super_block *sb, unsigned int block_group) block_group, le32_to_cpu(desc->bg_block_bitmap)); return NULL; } - if (!ext2_valid_block_bitmap(sb, desc, block_group, bh)) { - brelse(bh); - return NULL; - } + ext2_valid_block_bitmap(sb, desc, block_group, bh); + /* + * file system mounted not to panic on error, continue with corrupt + * bitmap + */ return bh; } @@ -1380,7 +1381,12 @@ allocated: "Allocating block in system zone - " "blocks from "E2FSBLK", length %lu", ret_block, num); - goto out; + /* + * ext2_try_to_allocate marked the blocks we allocated as in + * use. So we may want to selectively mark some of the blocks + * as free + */ + goto retry_alloc; } performed_allocation = 1; -- cgit From 3d61f75eefedf75914ab4453c67aaa2ee64bcf93 Mon Sep 17 00:00:00 2001 From: Hisashi Hifumi Date: Mon, 28 Apr 2008 02:16:05 -0700 Subject: ext3: fdatasync should skip metadata writeout when overwriting Currently fdatasync is identical to fsync in ext3. I think fdatasync should skip journal flush in data=ordered and data=writeback mode when it overwrites to already-instantiated blocks on HDD. When I_DIRTY_DATASYNC flag is not set, fdatasync should skip journal writeout because this indicates only atime or/and mtime updates. Following patch is the same approach of ext2's fsync code(ext2_sync_file). I did a performance test using the sysbench. #sysbench --num-threads=128 --max-requests=50000 --test=fileio --file-total-size=128G --file-test-mode=rndwr --file-fsync-mode=fdatasync run The result on ext3 was: -2.6.24 Operations performed: 0 Read, 50080 Write, 59600 Other = 109680 Total Read 0b Written 782.5Mb Total transferred 782.5Mb (12.116Mb/sec) 775.45 Requests/sec executed Test execution summary: total time: 64.5814s total number of events: 50080 total time taken by event execution: 3713.9836 per-request statistics: min: 0.0000s avg: 0.0742s max: 0.9375s approx. 95 percentile: 0.2901s Threads fairness: events (avg/stddev): 391.2500/23.26 execution time (avg/stddev): 29.0155/1.99 -2.6.24-patched Operations performed: 0 Read, 50009 Write, 61596 Other = 111605 Total Read 0b Written 781.39Mb Total transferred 781.39Mb (16.419Mb/sec) 1050.83 Requests/sec executed Test execution summary: total time: 47.5900s total number of events: 50009 total time taken by event execution: 2934.5768 per-request statistics: min: 0.0000s avg: 0.0587s max: 0.8938s approx. 95 percentile: 0.1993s Threads fairness: events (avg/stddev): 390.6953/22.64 execution time (avg/stddev): 22.9264/1.17 Filesystem I/O throughput was improved. Signed-off-by :Hisashi Hifumi Acked-by: Jan Kara Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/fsync.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs') diff --git a/fs/ext3/fsync.c b/fs/ext3/fsync.c index a588e23841d4..d33634119e17 100644 --- a/fs/ext3/fsync.c +++ b/fs/ext3/fsync.c @@ -72,6 +72,9 @@ int ext3_sync_file(struct file * file, struct dentry *dentry, int datasync) goto out; } + if (datasync && !(inode->i_state & I_DIRTY_DATASYNC)) + goto out; + /* * The VFS has written the file data. If the inode is unaltered * then we need not start a commit. -- cgit From e7f23ebdef879226817ce94ae6e298afc8cd093d Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Mon, 28 Apr 2008 02:16:06 -0700 Subject: ext3: convert byte order of constant instead of variable Convert byte order of constant instead of variable which can be done at compile time (vs run time). Signed-off-by: Marcin Slusarz Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 883ff965d984..d7f242a93b7c 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -1219,7 +1219,7 @@ static int ext3_setup_super(struct super_block *sb, struct ext3_super_block *es, inconsistencies, to force a fsck at reboot. But for a plain journaled filesystem we can keep it set as valid forever! :) */ - es->s_state = cpu_to_le16(le16_to_cpu(es->s_state) & ~EXT3_VALID_FS); + es->s_state &= cpu_to_le16(~EXT3_VALID_FS); #endif if (!(__s16) le16_to_cpu(es->s_max_mnt_count)) es->s_max_mnt_count = cpu_to_le16(EXT3_DFL_MAX_MNT_COUNT); -- cgit From 15633005e07883b57c6c7ca539c32148c3a7f588 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Mon, 28 Apr 2008 02:16:06 -0700 Subject: make ext3_xattr_list() static Make the needlessly global ext3_xattr_list() static. Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/xattr.c | 4 +++- fs/ext3/xattr.h | 7 ------- 2 files changed, 3 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/ext3/xattr.c b/fs/ext3/xattr.c index 42856541e9a5..8516831709b8 100644 --- a/fs/ext3/xattr.c +++ b/fs/ext3/xattr.c @@ -99,6 +99,8 @@ static struct buffer_head *ext3_xattr_cache_find(struct inode *, struct mb_cache_entry **); static void ext3_xattr_rehash(struct ext3_xattr_header *, struct ext3_xattr_entry *); +static int ext3_xattr_list(struct inode *inode, char *buffer, + size_t buffer_size); static struct mb_cache *ext3_xattr_cache; @@ -427,7 +429,7 @@ cleanup: * Returns a negative error number on failure, or the number of bytes * used / required on success. */ -int +static int ext3_xattr_list(struct inode *inode, char *buffer, size_t buffer_size) { int i_error, b_error; diff --git a/fs/ext3/xattr.h b/fs/ext3/xattr.h index 6b1ae1c6182c..148a4dfc82ab 100644 --- a/fs/ext3/xattr.h +++ b/fs/ext3/xattr.h @@ -67,7 +67,6 @@ extern struct xattr_handler ext3_xattr_security_handler; extern ssize_t ext3_listxattr(struct dentry *, char *, size_t); extern int ext3_xattr_get(struct inode *, int, const char *, void *, size_t); -extern int ext3_xattr_list(struct inode *, char *, size_t); extern int ext3_xattr_set(struct inode *, int, const char *, const void *, size_t, int); extern int ext3_xattr_set_handle(handle_t *, struct inode *, int, const char *, const void *, size_t, int); @@ -88,12 +87,6 @@ ext3_xattr_get(struct inode *inode, int name_index, const char *name, return -EOPNOTSUPP; } -static inline int -ext3_xattr_list(struct inode *inode, void *buffer, size_t size) -{ - return -EOPNOTSUPP; -} - static inline int ext3_xattr_set(struct inode *inode, int name_index, const char *name, const void *value, size_t size, int flags) -- cgit From 1eaafeae4b6f87eabdbabe3277826696f4ca196f Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 28 Apr 2008 02:16:07 -0700 Subject: ext3: use ext3_group_first_block_no() Use ext3_group_first_block_no() Signed-off-by: Akinobu Mita Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/super.c | 5 ++--- fs/ext3/xattr.c | 6 ++---- 2 files changed, 4 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/ext3/super.c b/fs/ext3/super.c index d7f242a93b7c..770bbb4fa58c 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -1254,14 +1254,14 @@ static int ext3_setup_super(struct super_block *sb, struct ext3_super_block *es, static int ext3_check_descriptors(struct super_block *sb) { struct ext3_sb_info *sbi = EXT3_SB(sb); - ext3_fsblk_t first_block = le32_to_cpu(sbi->s_es->s_first_data_block); - ext3_fsblk_t last_block; int i; ext3_debug ("Checking group descriptors"); for (i = 0; i < sbi->s_groups_count; i++) { struct ext3_group_desc *gdp = ext3_get_group_desc(sb, i, NULL); + ext3_fsblk_t first_block = ext3_group_first_block_no(sb, i); + ext3_fsblk_t last_block; if (i == sbi->s_groups_count - 1) last_block = le32_to_cpu(sbi->s_es->s_blocks_count) - 1; @@ -1300,7 +1300,6 @@ static int ext3_check_descriptors(struct super_block *sb) le32_to_cpu(gdp->bg_inode_table)); return 0; } - first_block += EXT3_BLOCKS_PER_GROUP(sb); } sbi->s_es->s_free_blocks_count=cpu_to_le32(ext3_count_free_blocks(sb)); diff --git a/fs/ext3/xattr.c b/fs/ext3/xattr.c index 8516831709b8..ca591ad1540d 100644 --- a/fs/ext3/xattr.c +++ b/fs/ext3/xattr.c @@ -799,10 +799,8 @@ inserted: get_bh(new_bh); } else { /* We need to allocate a new block */ - ext3_fsblk_t goal = le32_to_cpu( - EXT3_SB(sb)->s_es->s_first_data_block) + - (ext3_fsblk_t)EXT3_I(inode)->i_block_group * - EXT3_BLOCKS_PER_GROUP(sb); + ext3_fsblk_t goal = ext3_group_first_block_no(sb, + EXT3_I(inode)->i_block_group); ext3_fsblk_t block = ext3_new_block(handle, inode, goal, &error); if (error) -- cgit From 22a5daf5375a900e1a4efe8ffe2daef9be01e873 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 28 Apr 2008 02:16:07 -0700 Subject: ext3: add missing ext3_journal_stop() Add missing ext3_journal_stop() in error handling. Signed-off-by: Akinobu Mita Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/resize.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs') diff --git a/fs/ext3/resize.c b/fs/ext3/resize.c index 0e97b6e07cb0..7a9dac1d3653 100644 --- a/fs/ext3/resize.c +++ b/fs/ext3/resize.c @@ -1049,6 +1049,7 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, ext3_warning(sb, __FUNCTION__, "multiple resizers run on filesystem!"); unlock_super(sb); + ext3_journal_stop(handle); err = -EBUSY; goto exit_put; } -- cgit From e0e369a7dd39894465b6501a9492173e8104a19b Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 28 Apr 2008 02:16:08 -0700 Subject: ext3: use ext3_get_group_desc() Use ext3_get_group_desc() Signed-off-by: Akinobu Mita Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/inode.c | 23 +++++------------------ 1 file changed, 5 insertions(+), 18 deletions(-) (limited to 'fs') diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index c683609b0e3a..e2acbc7a3c3f 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -2454,11 +2454,10 @@ out_stop: static ext3_fsblk_t ext3_get_inode_block(struct super_block *sb, unsigned long ino, struct ext3_iloc *iloc) { - unsigned long desc, group_desc, block_group; + unsigned long block_group; unsigned long offset; ext3_fsblk_t block; - struct buffer_head *bh; - struct ext3_group_desc * gdp; + struct ext3_group_desc *gdp; if (!ext3_valid_inum(sb, ino)) { /* @@ -2470,27 +2469,15 @@ static ext3_fsblk_t ext3_get_inode_block(struct super_block *sb, } block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb); - if (block_group >= EXT3_SB(sb)->s_groups_count) { - ext3_error(sb,"ext3_get_inode_block","group >= groups count"); - return 0; - } - smp_rmb(); - group_desc = block_group >> EXT3_DESC_PER_BLOCK_BITS(sb); - desc = block_group & (EXT3_DESC_PER_BLOCK(sb) - 1); - bh = EXT3_SB(sb)->s_group_desc[group_desc]; - if (!bh) { - ext3_error (sb, "ext3_get_inode_block", - "Descriptor not loaded"); + gdp = ext3_get_group_desc(sb, block_group, NULL); + if (!gdp) return 0; - } - - gdp = (struct ext3_group_desc *)bh->b_data; /* * Figure out the offset within the block group inode table */ offset = ((ino - 1) % EXT3_INODES_PER_GROUP(sb)) * EXT3_INODE_SIZE(sb); - block = le32_to_cpu(gdp[desc].bg_inode_table) + + block = le32_to_cpu(gdp->bg_inode_table) + (offset >> EXT3_BLOCK_SIZE_BITS(sb)); iloc->block_group = block_group; -- cgit From 33575f8ffe99bf9e381161ccd76b39079c0aa92f Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 28 Apr 2008 02:16:08 -0700 Subject: ext3: check ext3_journal_get_write_access() errors Check ext3_journal_get_write_access() errors. Signed-off-by: Akinobu Mita Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/namei.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index dec3e0d88ab1..ba55865caaaf 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -57,10 +57,15 @@ static struct buffer_head *ext3_append(handle_t *handle, *block = inode->i_size >> inode->i_sb->s_blocksize_bits; - if ((bh = ext3_bread(handle, inode, *block, 1, err))) { + bh = ext3_bread(handle, inode, *block, 1, err); + if (bh) { inode->i_size += inode->i_sb->s_blocksize; EXT3_I(inode)->i_disksize = inode->i_size; - ext3_journal_get_write_access(handle,bh); + *err = ext3_journal_get_write_access(handle, bh); + if (*err) { + brelse(bh); + bh = NULL; + } } return bh; } -- cgit From 269b26191650be46ce6c91dec24cf20f59650529 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Mon, 28 Apr 2008 02:16:09 -0700 Subject: fs/ext3: use BUG_ON if (...) BUG(); should be replaced with BUG_ON(...) when the test has no side-effects to allow a definition of BUG_ON that drops the code completely. The semantic patch that makes this change is as follows: (http://www.emn.fr/x-info/coccinelle/) // @ disable unlikely @ expression E,f; @@ ( if (<... f(...) ...>) { BUG(); } | - if (unlikely(E)) { BUG(); } + BUG_ON(E); ) @@ expression E,f; @@ ( if (<... f(...) ...>) { BUG(); } | - if (E) { BUG(); } + BUG_ON(E); ) // Signed-off-by: Julia Lawall Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/balloc.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs') diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c index da0cb2c0e437..dd4237e5e9ec 100644 --- a/fs/ext3/balloc.c +++ b/fs/ext3/balloc.c @@ -232,8 +232,7 @@ restart: prev = rsv; } printk("Window map complete.\n"); - if (bad) - BUG(); + BUG_ON(bad); } #define rsv_window_dump(root, verbose) \ __rsv_window_dump((root), (verbose), __FUNCTION__) -- cgit From 5bc833feaa8b2236265764e7e81f44937be46eda Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Mon, 28 Apr 2008 02:16:10 -0700 Subject: jbd: fix the way the b_modified flag is cleared Currently at the start of a journal commit we loop through all of the buffers on the committing transaction and clear the b_modified flag (the flag that is set when a transaction modifies the buffer) under the j_list_lock. The problem is that everywhere else this flag is modified only under the jbd lock buffer flag, so it will race with a running transaction who could potentially set it, and have it unset by the committing transaction. This is also a big waste, you can have several thousands of buffers that you are clearing the modified flag on when you may not need to. This patch removes this code and instead clears the b_modified flag upon entering do_get_write_access/journal_get_create_access, so if that transaction does indeed use the buffer then it will be accounted for properly, and if it does not then we know we didn't use it. That will be important for the next patch in this series. Tested thoroughly by myself using postmark/iozone/bonnie++. Signed-off-by: Josef Bacik Cc: Acked-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/jbd/commit.c | 16 ---------------- fs/jbd/transaction.c | 13 +++++++++++++ 2 files changed, 13 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c index a38c7186c570..d2f652e6e22e 100644 --- a/fs/jbd/commit.c +++ b/fs/jbd/commit.c @@ -406,22 +406,6 @@ void journal_commit_transaction(journal_t *journal) jbd_debug (3, "JBD: commit phase 2\n"); - /* - * First, drop modified flag: all accesses to the buffers - * will be tracked for a new trasaction only -bzzz - */ - spin_lock(&journal->j_list_lock); - if (commit_transaction->t_buffers) { - new_jh = jh = commit_transaction->t_buffers->b_tnext; - do { - J_ASSERT_JH(new_jh, new_jh->b_modified == 1 || - new_jh->b_modified == 0); - new_jh->b_modified = 0; - new_jh = new_jh->b_tnext; - } while (new_jh != jh); - } - spin_unlock(&journal->j_list_lock); - /* * Now start flushing things to disk, in the order they appear * on the transaction lists. Data blocks go first. diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index 2c9e8f5d13aa..2e9670e2237e 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -608,6 +608,12 @@ repeat: jh->b_next_transaction == transaction) goto done; + /* + * this is the first time this transaction is touching this buffer, + * reset the modified flag + */ + jh->b_modified = 0; + /* * If there is already a copy-out version of this buffer, then we don't * need to make another one @@ -820,9 +826,16 @@ int journal_get_create_access(handle_t *handle, struct buffer_head *bh) if (jh->b_transaction == NULL) { jh->b_transaction = transaction; + + /* first access by this transaction */ + jh->b_modified = 0; + JBUFFER_TRACE(jh, "file as BJ_Reserved"); __journal_file_buffer(jh, transaction, BJ_Reserved); } else if (jh->b_transaction == journal->j_committing_transaction) { + /* first access by this transaction */ + jh->b_modified = 0; + JBUFFER_TRACE(jh, "set next transaction"); jh->b_next_transaction = transaction; } -- cgit From 5b9a499d77e9dd39c9e6611ea10c56a31604f274 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Mon, 28 Apr 2008 02:16:12 -0700 Subject: jbd: fix possible journal overflow issues There are several cases where the running transaction can get buffers added to its BJ_Metadata list which it never dirtied, which makes its t_nr_buffers counter end up larger than its t_outstanding_credits counter. This will cause issues when starting new transactions as while we are logging buffers we decrement t_outstanding_buffers, so when t_outstanding_buffers goes negative, we will report that we need less space in the journal than we actually need, so transactions will be started even though there may not be enough room for them. In the worst case scenario (which admittedly is almost impossible to reproduce) this will result in the journal running out of space. The fix is to only refile buffers from the committing transaction to the running transactions BJ_Modified list when b_modified is set on that journal, which is the only way to be sure if the running transaction has modified that buffer. This patch also fixes an accounting error in journal_forget, it is possible that we can call journal_forget on a buffer without having modified it, only gotten write access to it, so instead of freeing a credit, we only do so if the buffer was modified. The assert will help catch if this problem occurs. Without these two patches I could hit this assert within minutes of running postmark, with them this issue no longer arises. Thank you, Signed-off-by: Josef Bacik Cc: Acked-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/jbd/commit.c | 3 +++ fs/jbd/transaction.c | 21 ++++++++++++++++++--- 2 files changed, 21 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c index d2f652e6e22e..cd931ef1f000 100644 --- a/fs/jbd/commit.c +++ b/fs/jbd/commit.c @@ -472,6 +472,9 @@ void journal_commit_transaction(journal_t *journal) */ commit_transaction->t_state = T_COMMIT; + J_ASSERT(commit_transaction->t_nr_buffers <= + commit_transaction->t_outstanding_credits); + descriptor = NULL; bufs = 0; while (commit_transaction->t_buffers) { diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index 2e9670e2237e..d8dd3d915fd9 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -1235,6 +1235,7 @@ int journal_forget (handle_t *handle, struct buffer_head *bh) struct journal_head *jh; int drop_reserve = 0; int err = 0; + int was_modified = 0; BUFFER_TRACE(bh, "entry"); @@ -1253,6 +1254,9 @@ int journal_forget (handle_t *handle, struct buffer_head *bh) goto not_jbd; } + /* keep track of wether or not this transaction modified us */ + was_modified = jh->b_modified; + /* * The buffer's going from the transaction, we must drop * all references -bzzz @@ -1270,7 +1274,12 @@ int journal_forget (handle_t *handle, struct buffer_head *bh) JBUFFER_TRACE(jh, "belongs to current transaction: unfile"); - drop_reserve = 1; + /* + * we only want to drop a reference if this transaction + * modified the buffer + */ + if (was_modified) + drop_reserve = 1; /* * We are no longer going to journal this buffer. @@ -1310,7 +1319,13 @@ int journal_forget (handle_t *handle, struct buffer_head *bh) if (jh->b_next_transaction) { J_ASSERT(jh->b_next_transaction == transaction); jh->b_next_transaction = NULL; - drop_reserve = 1; + + /* + * only drop a reference if this transaction modified + * the buffer + */ + if (was_modified) + drop_reserve = 1; } } @@ -2082,7 +2097,7 @@ void __journal_refile_buffer(struct journal_head *jh) jh->b_transaction = jh->b_next_transaction; jh->b_next_transaction = NULL; __journal_file_buffer(jh, jh->b_transaction, - was_dirty ? BJ_Metadata : BJ_Reserved); + jh->b_modified ? BJ_Metadata : BJ_Reserved); J_ASSERT_JH(jh, jh->b_transaction->t_state == T_RUNNING); if (was_dirty) -- cgit From 0b23076988b44b2c165e060248345de6f2337387 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 28 Apr 2008 02:16:12 -0700 Subject: ext3: fix update of mtime and ctime on rename Make ext3 update mtime and ctime of the directory into which we move file even if the directory entry already exists. Signed-off-by: Jan Kara Cc: Al Viro Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/namei.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index ba55865caaaf..1a106572d63e 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -2323,6 +2323,8 @@ static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry, EXT3_FEATURE_INCOMPAT_FILETYPE)) new_de->file_type = old_de->file_type; new_dir->i_version++; + new_dir->i_ctime = new_dir->i_mtime = CURRENT_TIME_SEC; + ext3_mark_inode_dirty(handle, new_dir); BUFFER_TRACE(new_bh, "call ext3_journal_dirty_metadata"); ext3_journal_dirty_metadata(handle, new_bh); brelse(new_bh); -- cgit From 07c9938a4e2c92b796b163dc70e99d3d1870aaee Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 28 Apr 2008 02:16:13 -0700 Subject: ext3: fix hang on umount with quotas when journal is aborted Call dquot_drop() from ext3_dquot_drop() even if we fail to start a transaction. Otherwise we never get to dropping references to quota structures from the inode and umount will hang indefinitely. Thanks to Payphone LIOU for spotting the problem. Signed-off-by: Jan Kara Cc: Payphone LIOU Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/super.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 770bbb4fa58c..03d5c626c1cd 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -2639,8 +2639,14 @@ static int ext3_dquot_drop(struct inode *inode) /* We may delete quota structure so we need to reserve enough blocks */ handle = ext3_journal_start(inode, 2*EXT3_QUOTA_DEL_BLOCKS(inode->i_sb)); - if (IS_ERR(handle)) + if (IS_ERR(handle)) { + /* + * We call dquot_drop() anyway to at least release references + * to quota structures so that umount does not hang. + */ + dquot_drop(inode); return PTR_ERR(handle); + } ret = dquot_drop(inode); err = ext3_journal_stop(handle); if (!ret) -- cgit From 2588ef83f7933d8ae42868d7bf68fc8a3001186b Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Mon, 28 Apr 2008 02:16:14 -0700 Subject: ext3: retry block allocation if new blocks are allocated from system zone If the block allocator gets blocks out of system zone ext3 calls ext3_error. But if the file system is mounted with errors=continue retry block allocation. We need to mark the system zone blocks as in use to make sure retry don't pick them again System zone is the block range mapping block bitmap, inode bitmap and inode table. [akpm@linux-foundation.org: fix typo in comment] Signed-off-by: Aneesh Kumar K.V Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/balloc.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c index dd4237e5e9ec..731fbc0f1291 100644 --- a/fs/ext3/balloc.c +++ b/fs/ext3/balloc.c @@ -164,10 +164,11 @@ read_block_bitmap(struct super_block *sb, unsigned int block_group) block_group, le32_to_cpu(desc->bg_block_bitmap)); return NULL; } - if (!ext3_valid_block_bitmap(sb, desc, block_group, bh)) { - brelse(bh); - return NULL; - } + ext3_valid_block_bitmap(sb, desc, block_group, bh); + /* + * file system mounted not to panic on error, continue with corrupt + * bitmap + */ return bh; } /* @@ -1641,7 +1642,11 @@ allocated: "Allocating block in system zone - " "blocks from "E3FSBLK", length %lu", ret_block, num); - goto out; + /* + * claim_block() marked the blocks we allocated as in use. So we + * may want to selectively mark some of the blocks as free. + */ + goto retry_alloc; } performed_allocation = 1; -- cgit From fa1ff1e02fee908dfdc3f92902d39acc38041e4c Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 28 Apr 2008 02:16:14 -0700 Subject: ext3: fix mount messages when quota disabled When quota is disabled, we should not print 'journaled quota not supported' when user tried to mount non-journaled quota. Also fix typo in the message. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/super.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 03d5c626c1cd..6303adfbdae4 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -1097,6 +1097,9 @@ clear_qf_name: case Opt_quota: case Opt_usrquota: case Opt_grpquota: + printk(KERN_ERR + "EXT3-fs: quota options not supported.\n"); + break; case Opt_usrjquota: case Opt_grpjquota: case Opt_offusrjquota: @@ -1104,7 +1107,7 @@ clear_qf_name: case Opt_jqfmt_vfsold: case Opt_jqfmt_vfsv0: printk(KERN_ERR - "EXT3-fs: journalled quota options not " + "EXT3-fs: journaled quota options not " "supported.\n"); break; case Opt_noquota: -- cgit From e05b6b524bd5c5c2bae1b64a7cbe08d46d57a6fe Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Mon, 28 Apr 2008 02:16:15 -0700 Subject: ext3: replace remaining __FUNCTION__ occurrences __FUNCTION__ is gcc-specific, use __func__ Signed-off-by: Harvey Harrison Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/balloc.c | 12 +++++----- fs/ext3/ext3_jbd.c | 12 +++++----- fs/ext3/ialloc.c | 6 ++--- fs/ext3/inode.c | 6 ++--- fs/ext3/namei.c | 26 ++++++++++---------- fs/ext3/resize.c | 70 +++++++++++++++++++++++++++--------------------------- fs/ext3/super.c | 14 +++++------ fs/ext3/xattr.c | 14 +++++------ 8 files changed, 80 insertions(+), 80 deletions(-) (limited to 'fs') diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c index 731fbc0f1291..92fd0338a6eb 100644 --- a/fs/ext3/balloc.c +++ b/fs/ext3/balloc.c @@ -117,7 +117,7 @@ static int ext3_valid_block_bitmap(struct super_block *sb, return 1; err_out: - ext3_error(sb, __FUNCTION__, + ext3_error(sb, __func__, "Invalid block bitmap - " "block_group = %d, block = %lu", block_group, bitmap_blk); @@ -147,7 +147,7 @@ read_block_bitmap(struct super_block *sb, unsigned int block_group) bitmap_blk = le32_to_cpu(desc->bg_block_bitmap); bh = sb_getblk(sb, bitmap_blk); if (unlikely(!bh)) { - ext3_error(sb, __FUNCTION__, + ext3_error(sb, __func__, "Cannot read block bitmap - " "block_group = %d, block_bitmap = %u", block_group, le32_to_cpu(desc->bg_block_bitmap)); @@ -158,7 +158,7 @@ read_block_bitmap(struct super_block *sb, unsigned int block_group) if (bh_submit_read(bh) < 0) { brelse(bh); - ext3_error(sb, __FUNCTION__, + ext3_error(sb, __func__, "Cannot read block bitmap - " "block_group = %d, block_bitmap = %u", block_group, le32_to_cpu(desc->bg_block_bitmap)); @@ -236,7 +236,7 @@ restart: BUG_ON(bad); } #define rsv_window_dump(root, verbose) \ - __rsv_window_dump((root), (verbose), __FUNCTION__) + __rsv_window_dump((root), (verbose), __func__) #else #define rsv_window_dump(root, verbose) do {} while (0) #endif @@ -618,7 +618,7 @@ do_more: if (!ext3_clear_bit_atomic(sb_bgl_lock(sbi, block_group), bit + i, bitmap_bh->b_data)) { jbd_unlock_bh_state(bitmap_bh); - ext3_error(sb, __FUNCTION__, + ext3_error(sb, __func__, "bit already cleared for block "E3FSBLK, block + i); jbd_lock_bh_state(bitmap_bh); @@ -1672,7 +1672,7 @@ allocated: if (ext3_test_bit(grp_alloc_blk+i, bh2jh(bitmap_bh)->b_committed_data)) { printk("%s: block was unexpectedly set in " - "b_committed_data\n", __FUNCTION__); + "b_committed_data\n", __func__); } } } diff --git a/fs/ext3/ext3_jbd.c b/fs/ext3/ext3_jbd.c index e1f91fd26a93..d401f148d74d 100644 --- a/fs/ext3/ext3_jbd.c +++ b/fs/ext3/ext3_jbd.c @@ -9,7 +9,7 @@ int __ext3_journal_get_undo_access(const char *where, handle_t *handle, { int err = journal_get_undo_access(handle, bh); if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext3_journal_abort_handle(where, __func__, bh, handle,err); return err; } @@ -18,7 +18,7 @@ int __ext3_journal_get_write_access(const char *where, handle_t *handle, { int err = journal_get_write_access(handle, bh); if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext3_journal_abort_handle(where, __func__, bh, handle,err); return err; } @@ -27,7 +27,7 @@ int __ext3_journal_forget(const char *where, handle_t *handle, { int err = journal_forget(handle, bh); if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext3_journal_abort_handle(where, __func__, bh, handle,err); return err; } @@ -36,7 +36,7 @@ int __ext3_journal_revoke(const char *where, handle_t *handle, { int err = journal_revoke(handle, blocknr, bh); if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext3_journal_abort_handle(where, __func__, bh, handle,err); return err; } @@ -45,7 +45,7 @@ int __ext3_journal_get_create_access(const char *where, { int err = journal_get_create_access(handle, bh); if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext3_journal_abort_handle(where, __func__, bh, handle,err); return err; } @@ -54,6 +54,6 @@ int __ext3_journal_dirty_metadata(const char *where, { int err = journal_dirty_metadata(handle, bh); if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + ext3_journal_abort_handle(where, __func__, bh, handle,err); return err; } diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c index 96dd5573e49b..77126821b2e9 100644 --- a/fs/ext3/ialloc.c +++ b/fs/ext3/ialloc.c @@ -644,7 +644,7 @@ struct inode *ext3_orphan_get(struct super_block *sb, unsigned long ino) /* Error cases - e2fsck has already cleaned up for us */ if (ino > max_ino) { - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "bad orphan ino %lu! e2fsck was run?", ino); goto error; } @@ -653,7 +653,7 @@ struct inode *ext3_orphan_get(struct super_block *sb, unsigned long ino) bit = (ino - 1) % EXT3_INODES_PER_GROUP(sb); bitmap_bh = read_inode_bitmap(sb, block_group); if (!bitmap_bh) { - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "inode bitmap error for orphan %lu", ino); goto error; } @@ -678,7 +678,7 @@ iget_failed: err = PTR_ERR(inode); inode = NULL; bad_orphan: - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "bad orphan inode %lu! e2fsck was run?", ino); printk(KERN_NOTICE "ext3_test_bit(bit=%d, block=%llu) = %d\n", bit, (unsigned long long)bitmap_bh->b_blocknr, diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index e2acbc7a3c3f..cc47b76091bf 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -95,7 +95,7 @@ int ext3_forget(handle_t *handle, int is_metadata, struct inode *inode, BUFFER_TRACE(bh, "call ext3_journal_revoke"); err = ext3_journal_revoke(handle, blocknr, bh); if (err) - ext3_abort(inode->i_sb, __FUNCTION__, + ext3_abort(inode->i_sb, __func__, "error %d when attempting revoke", err); BUFFER_TRACE(bh, "exit"); return err; @@ -1190,7 +1190,7 @@ int ext3_journal_dirty_data(handle_t *handle, struct buffer_head *bh) { int err = journal_dirty_data(handle, bh); if (err) - ext3_journal_abort_handle(__FUNCTION__, __FUNCTION__, + ext3_journal_abort_handle(__func__, __func__, bh, handle, err); return err; } @@ -3201,7 +3201,7 @@ void ext3_dirty_inode(struct inode *inode) current_handle->h_transaction != handle->h_transaction) { /* This task has a transaction open against a different fs */ printk(KERN_EMERG "%s: transactions do not match!\n", - __FUNCTION__); + __func__); } else { jbd_debug(5, "marking dirty. outer handle=%p\n", current_handle); diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index 1a106572d63e..0b8cf80154f1 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -361,7 +361,7 @@ dx_probe(struct dentry *dentry, struct inode *dir, if (root->info.hash_version != DX_HASH_TEA && root->info.hash_version != DX_HASH_HALF_MD4 && root->info.hash_version != DX_HASH_LEGACY) { - ext3_warning(dir->i_sb, __FUNCTION__, + ext3_warning(dir->i_sb, __func__, "Unrecognised inode hash code %d", root->info.hash_version); brelse(bh); @@ -375,7 +375,7 @@ dx_probe(struct dentry *dentry, struct inode *dir, hash = hinfo->hash; if (root->info.unused_flags & 1) { - ext3_warning(dir->i_sb, __FUNCTION__, + ext3_warning(dir->i_sb, __func__, "Unimplemented inode hash flags: %#06x", root->info.unused_flags); brelse(bh); @@ -384,7 +384,7 @@ dx_probe(struct dentry *dentry, struct inode *dir, } if ((indirect = root->info.indirect_levels) > 1) { - ext3_warning(dir->i_sb, __FUNCTION__, + ext3_warning(dir->i_sb, __func__, "Unimplemented inode hash depth: %#06x", root->info.indirect_levels); brelse(bh); @@ -397,7 +397,7 @@ dx_probe(struct dentry *dentry, struct inode *dir, if (dx_get_limit(entries) != dx_root_limit(dir, root->info.info_length)) { - ext3_warning(dir->i_sb, __FUNCTION__, + ext3_warning(dir->i_sb, __func__, "dx entry: limit != root limit"); brelse(bh); *err = ERR_BAD_DX_DIR; @@ -409,7 +409,7 @@ dx_probe(struct dentry *dentry, struct inode *dir, { count = dx_get_count(entries); if (!count || count > dx_get_limit(entries)) { - ext3_warning(dir->i_sb, __FUNCTION__, + ext3_warning(dir->i_sb, __func__, "dx entry: no count or count > limit"); brelse(bh); *err = ERR_BAD_DX_DIR; @@ -454,7 +454,7 @@ dx_probe(struct dentry *dentry, struct inode *dir, goto fail2; at = entries = ((struct dx_node *) bh->b_data)->entries; if (dx_get_limit(entries) != dx_node_limit (dir)) { - ext3_warning(dir->i_sb, __FUNCTION__, + ext3_warning(dir->i_sb, __func__, "dx entry: limit != node limit"); brelse(bh); *err = ERR_BAD_DX_DIR; @@ -470,7 +470,7 @@ fail2: } fail: if (*err == ERR_BAD_DX_DIR) - ext3_warning(dir->i_sb, __FUNCTION__, + ext3_warning(dir->i_sb, __func__, "Corrupt dir inode %ld, running e2fsck is " "recommended.", dir->i_ino); return NULL; @@ -918,7 +918,7 @@ restart: wait_on_buffer(bh); if (!buffer_uptodate(bh)) { /* read error, skip block & hope for the best */ - ext3_error(sb, __FUNCTION__, "reading directory #%lu " + ext3_error(sb, __func__, "reading directory #%lu " "offset %lu", dir->i_ino, block); brelse(bh); goto next; @@ -1010,7 +1010,7 @@ static struct buffer_head * ext3_dx_find_entry(struct dentry *dentry, retval = ext3_htree_next_block(dir, hash, frame, frames, NULL); if (retval < 0) { - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "error reading index page in directory #%lu", dir->i_ino); *err = retval; @@ -1535,7 +1535,7 @@ static int ext3_dx_add_entry(handle_t *handle, struct dentry *dentry, if (levels && (dx_get_count(frames->entries) == dx_get_limit(frames->entries))) { - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "Directory index full!"); err = -ENOSPC; goto cleanup; @@ -1837,11 +1837,11 @@ static int empty_dir (struct inode * inode) if (inode->i_size < EXT3_DIR_REC_LEN(1) + EXT3_DIR_REC_LEN(2) || !(bh = ext3_bread (NULL, inode, 0, 0, &err))) { if (err) - ext3_error(inode->i_sb, __FUNCTION__, + ext3_error(inode->i_sb, __func__, "error %d reading directory #%lu offset 0", err, inode->i_ino); else - ext3_warning(inode->i_sb, __FUNCTION__, + ext3_warning(inode->i_sb, __func__, "bad directory (dir #%lu) - no data block", inode->i_ino); return 1; @@ -1870,7 +1870,7 @@ static int empty_dir (struct inode * inode) offset >> EXT3_BLOCK_SIZE_BITS(sb), 0, &err); if (!bh) { if (err) - ext3_error(sb, __FUNCTION__, + ext3_error(sb, __func__, "error %d reading directory" " #%lu offset %lu", err, inode->i_ino, offset); diff --git a/fs/ext3/resize.c b/fs/ext3/resize.c index 7a9dac1d3653..28cfd0b40527 100644 --- a/fs/ext3/resize.c +++ b/fs/ext3/resize.c @@ -48,60 +48,60 @@ static int verify_group_input(struct super_block *sb, free_blocks_count, input->reserved_blocks); if (group != sbi->s_groups_count) - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "Cannot add at group %u (only %lu groups)", input->group, sbi->s_groups_count); else if ((start - le32_to_cpu(es->s_first_data_block)) % EXT3_BLOCKS_PER_GROUP(sb)) - ext3_warning(sb, __FUNCTION__, "Last group not full"); + ext3_warning(sb, __func__, "Last group not full"); else if (input->reserved_blocks > input->blocks_count / 5) - ext3_warning(sb, __FUNCTION__, "Reserved blocks too high (%u)", + ext3_warning(sb, __func__, "Reserved blocks too high (%u)", input->reserved_blocks); else if (free_blocks_count < 0) - ext3_warning(sb, __FUNCTION__, "Bad blocks count %u", + ext3_warning(sb, __func__, "Bad blocks count %u", input->blocks_count); else if (!(bh = sb_bread(sb, end - 1))) - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "Cannot read last block ("E3FSBLK")", end - 1); else if (outside(input->block_bitmap, start, end)) - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "Block bitmap not in group (block %u)", input->block_bitmap); else if (outside(input->inode_bitmap, start, end)) - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "Inode bitmap not in group (block %u)", input->inode_bitmap); else if (outside(input->inode_table, start, end) || outside(itend - 1, start, end)) - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "Inode table not in group (blocks %u-"E3FSBLK")", input->inode_table, itend - 1); else if (input->inode_bitmap == input->block_bitmap) - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "Block bitmap same as inode bitmap (%u)", input->block_bitmap); else if (inside(input->block_bitmap, input->inode_table, itend)) - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "Block bitmap (%u) in inode table (%u-"E3FSBLK")", input->block_bitmap, input->inode_table, itend-1); else if (inside(input->inode_bitmap, input->inode_table, itend)) - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "Inode bitmap (%u) in inode table (%u-"E3FSBLK")", input->inode_bitmap, input->inode_table, itend-1); else if (inside(input->block_bitmap, start, metaend)) - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "Block bitmap (%u) in GDT table" " ("E3FSBLK"-"E3FSBLK")", input->block_bitmap, start, metaend - 1); else if (inside(input->inode_bitmap, start, metaend)) - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "Inode bitmap (%u) in GDT table" " ("E3FSBLK"-"E3FSBLK")", input->inode_bitmap, start, metaend - 1); else if (inside(input->inode_table, start, metaend) || inside(itend - 1, start, metaend)) - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "Inode table (%u-"E3FSBLK") overlaps" "GDT table ("E3FSBLK"-"E3FSBLK")", input->inode_table, itend - 1, start, metaend - 1); @@ -386,7 +386,7 @@ static int verify_reserved_gdb(struct super_block *sb, while ((grp = ext3_list_backups(sb, &three, &five, &seven)) < end) { if (le32_to_cpu(*p++) != grp * EXT3_BLOCKS_PER_GROUP(sb) + blk){ - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "reserved GDT "E3FSBLK " missing grp %d ("E3FSBLK")", blk, grp, @@ -440,7 +440,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, */ if (EXT3_SB(sb)->s_sbh->b_blocknr != le32_to_cpu(EXT3_SB(sb)->s_es->s_first_data_block)) { - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "won't resize using backup superblock at %llu", (unsigned long long)EXT3_SB(sb)->s_sbh->b_blocknr); return -EPERM; @@ -464,7 +464,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, data = (__le32 *)dind->b_data; if (le32_to_cpu(data[gdb_num % EXT3_ADDR_PER_BLOCK(sb)]) != gdblock) { - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "new group %u GDT block "E3FSBLK" not reserved", input->group, gdblock); err = -EINVAL; @@ -488,7 +488,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, GFP_NOFS); if (!n_group_desc) { err = -ENOMEM; - ext3_warning (sb, __FUNCTION__, + ext3_warning (sb, __func__, "not enough memory for %lu groups", gdb_num + 1); goto exit_inode; } @@ -586,7 +586,7 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode, /* Get each reserved primary GDT block and verify it holds backups */ for (res = 0; res < reserved_gdb; res++, blk++) { if (le32_to_cpu(*data) != blk) { - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "reserved block "E3FSBLK " not at offset %ld", blk, @@ -730,7 +730,7 @@ static void update_backups(struct super_block *sb, */ exit_err: if (err) { - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "can't update backup for group %d (err %d), " "forcing fsck on next reboot", group, err); sbi->s_mount_state &= ~EXT3_VALID_FS; @@ -770,33 +770,33 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) if (gdb_off == 0 && !EXT3_HAS_RO_COMPAT_FEATURE(sb, EXT3_FEATURE_RO_COMPAT_SPARSE_SUPER)) { - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "Can't resize non-sparse filesystem further"); return -EPERM; } if (le32_to_cpu(es->s_blocks_count) + input->blocks_count < le32_to_cpu(es->s_blocks_count)) { - ext3_warning(sb, __FUNCTION__, "blocks_count overflow\n"); + ext3_warning(sb, __func__, "blocks_count overflow\n"); return -EINVAL; } if (le32_to_cpu(es->s_inodes_count) + EXT3_INODES_PER_GROUP(sb) < le32_to_cpu(es->s_inodes_count)) { - ext3_warning(sb, __FUNCTION__, "inodes_count overflow\n"); + ext3_warning(sb, __func__, "inodes_count overflow\n"); return -EINVAL; } if (reserved_gdb || gdb_off == 0) { if (!EXT3_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_RESIZE_INODE)){ - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "No reserved GDT blocks, can't resize"); return -EPERM; } inode = ext3_iget(sb, EXT3_RESIZE_INO); if (IS_ERR(inode)) { - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "Error opening resize inode"); return PTR_ERR(inode); } @@ -825,7 +825,7 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) lock_super(sb); if (input->group != sbi->s_groups_count) { - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "multiple resizers run on filesystem!"); err = -EBUSY; goto exit_journal; @@ -988,13 +988,13 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, " too large to resize to %lu blocks safely\n", sb->s_id, n_blocks_count); if (sizeof(sector_t) < 8) - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "CONFIG_LBD not enabled\n"); return -EINVAL; } if (n_blocks_count < o_blocks_count) { - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "can't shrink FS - resize aborted"); return -EBUSY; } @@ -1004,7 +1004,7 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, EXT3_BLOCKS_PER_GROUP(sb); if (last == 0) { - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "need to use ext2online to resize further"); return -EPERM; } @@ -1012,7 +1012,7 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, add = EXT3_BLOCKS_PER_GROUP(sb) - last; if (o_blocks_count + add < o_blocks_count) { - ext3_warning(sb, __FUNCTION__, "blocks_count overflow"); + ext3_warning(sb, __func__, "blocks_count overflow"); return -EINVAL; } @@ -1020,7 +1020,7 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, add = n_blocks_count - o_blocks_count; if (o_blocks_count + add < n_blocks_count) - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "will only finish group ("E3FSBLK " blocks, %u new)", o_blocks_count + add, add); @@ -1028,7 +1028,7 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, /* See if the device is actually as big as what was requested */ bh = sb_bread(sb, o_blocks_count + add -1); if (!bh) { - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "can't read last block, resize aborted"); return -ENOSPC; } @@ -1040,13 +1040,13 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, handle = ext3_journal_start_sb(sb, 3); if (IS_ERR(handle)) { err = PTR_ERR(handle); - ext3_warning(sb, __FUNCTION__, "error %d on journal start",err); + ext3_warning(sb, __func__, "error %d on journal start",err); goto exit_put; } lock_super(sb); if (o_blocks_count != le32_to_cpu(es->s_blocks_count)) { - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "multiple resizers run on filesystem!"); unlock_super(sb); ext3_journal_stop(handle); @@ -1056,7 +1056,7 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, if ((err = ext3_journal_get_write_access(handle, EXT3_SB(sb)->s_sbh))) { - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "error %d on journal write access", err); unlock_super(sb); ext3_journal_stop(handle); diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 6303adfbdae4..fe3119a71ada 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -84,7 +84,7 @@ handle_t *ext3_journal_start_sb(struct super_block *sb, int nblocks) * take the FS itself readonly cleanly. */ journal = EXT3_SB(sb)->s_journal; if (is_journal_aborted(journal)) { - ext3_abort(sb, __FUNCTION__, + ext3_abort(sb, __func__, "Detected aborted journal"); return ERR_PTR(-EROFS); } @@ -304,7 +304,7 @@ void ext3_update_dynamic_rev(struct super_block *sb) if (le32_to_cpu(es->s_rev_level) > EXT3_GOOD_OLD_REV) return; - ext3_warning(sb, __FUNCTION__, + ext3_warning(sb, __func__, "updating to rev %d because of new feature flag, " "running e2fsck is recommended", EXT3_DYNAMIC_REV); @@ -1390,7 +1390,7 @@ static void ext3_orphan_cleanup (struct super_block * sb, if (inode->i_nlink) { printk(KERN_DEBUG "%s: truncating inode %lu to %Ld bytes\n", - __FUNCTION__, inode->i_ino, inode->i_size); + __func__, inode->i_ino, inode->i_size); jbd_debug(2, "truncating inode %lu to %Ld bytes\n", inode->i_ino, inode->i_size); ext3_truncate(inode); @@ -1398,7 +1398,7 @@ static void ext3_orphan_cleanup (struct super_block * sb, } else { printk(KERN_DEBUG "%s: deleting unreferenced inode %lu\n", - __FUNCTION__, inode->i_ino); + __func__, inode->i_ino); jbd_debug(2, "deleting unreferenced inode %lu\n", inode->i_ino); nr_orphans++; @@ -2301,9 +2301,9 @@ static void ext3_clear_journal_err(struct super_block * sb, char nbuf[16]; errstr = ext3_decode_error(sb, j_errno, nbuf); - ext3_warning(sb, __FUNCTION__, "Filesystem error recorded " + ext3_warning(sb, __func__, "Filesystem error recorded " "from previous mount: %s", errstr); - ext3_warning(sb, __FUNCTION__, "Marking fs in need of " + ext3_warning(sb, __func__, "Marking fs in need of " "filesystem check."); EXT3_SB(sb)->s_mount_state |= EXT3_ERROR_FS; @@ -2430,7 +2430,7 @@ static int ext3_remount (struct super_block * sb, int * flags, char * data) } if (sbi->s_mount_opt & EXT3_MOUNT_ABORT) - ext3_abort(sb, __FUNCTION__, "Abort forced by user"); + ext3_abort(sb, __func__, "Abort forced by user"); sb->s_flags = (sb->s_flags & ~MS_POSIXACL) | ((sbi->s_mount_opt & EXT3_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0); diff --git a/fs/ext3/xattr.c b/fs/ext3/xattr.c index ca591ad1540d..d4a4f0e9ff69 100644 --- a/fs/ext3/xattr.c +++ b/fs/ext3/xattr.c @@ -234,7 +234,7 @@ ext3_xattr_block_get(struct inode *inode, int name_index, const char *name, ea_bdebug(bh, "b_count=%d, refcount=%d", atomic_read(&(bh->b_count)), le32_to_cpu(BHDR(bh)->h_refcount)); if (ext3_xattr_check_block(bh)) { -bad_block: ext3_error(inode->i_sb, __FUNCTION__, +bad_block: ext3_error(inode->i_sb, __func__, "inode %lu: bad block "E3FSBLK, inode->i_ino, EXT3_I(inode)->i_file_acl); error = -EIO; @@ -376,7 +376,7 @@ ext3_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size) ea_bdebug(bh, "b_count=%d, refcount=%d", atomic_read(&(bh->b_count)), le32_to_cpu(BHDR(bh)->h_refcount)); if (ext3_xattr_check_block(bh)) { - ext3_error(inode->i_sb, __FUNCTION__, + ext3_error(inode->i_sb, __func__, "inode %lu: bad block "E3FSBLK, inode->i_ino, EXT3_I(inode)->i_file_acl); error = -EIO; @@ -651,7 +651,7 @@ ext3_xattr_block_find(struct inode *inode, struct ext3_xattr_info *i, atomic_read(&(bs->bh->b_count)), le32_to_cpu(BHDR(bs->bh)->h_refcount)); if (ext3_xattr_check_block(bs->bh)) { - ext3_error(sb, __FUNCTION__, + ext3_error(sb, __func__, "inode %lu: bad block "E3FSBLK, inode->i_ino, EXT3_I(inode)->i_file_acl); error = -EIO; @@ -852,7 +852,7 @@ cleanup_dquot: goto cleanup; bad_block: - ext3_error(inode->i_sb, __FUNCTION__, + ext3_error(inode->i_sb, __func__, "inode %lu: bad block "E3FSBLK, inode->i_ino, EXT3_I(inode)->i_file_acl); goto cleanup; @@ -1081,14 +1081,14 @@ ext3_xattr_delete_inode(handle_t *handle, struct inode *inode) goto cleanup; bh = sb_bread(inode->i_sb, EXT3_I(inode)->i_file_acl); if (!bh) { - ext3_error(inode->i_sb, __FUNCTION__, + ext3_error(inode->i_sb, __func__, "inode %lu: block "E3FSBLK" read error", inode->i_ino, EXT3_I(inode)->i_file_acl); goto cleanup; } if (BHDR(bh)->h_magic != cpu_to_le32(EXT3_XATTR_MAGIC) || BHDR(bh)->h_blocks != cpu_to_le32(1)) { - ext3_error(inode->i_sb, __FUNCTION__, + ext3_error(inode->i_sb, __func__, "inode %lu: bad block "E3FSBLK, inode->i_ino, EXT3_I(inode)->i_file_acl); goto cleanup; @@ -1215,7 +1215,7 @@ again: } bh = sb_bread(inode->i_sb, ce->e_block); if (!bh) { - ext3_error(inode->i_sb, __FUNCTION__, + ext3_error(inode->i_sb, __func__, "inode %lu: block %lu read error", inode->i_ino, (unsigned long) ce->e_block); } else if (le32_to_cpu(BHDR(bh)->h_refcount) >= -- cgit From 08fc99bfc39d5b394aa27498ac5f14df0ef8e52a Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Mon, 28 Apr 2008 02:16:16 -0700 Subject: jbd: replace remaining __FUNCTION__ occurrences __FUNCTION__ is gcc-specific, use __func__ Signed-off-by: Harvey Harrison Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/jbd/journal.c | 18 +++++++++--------- fs/jbd/revoke.c | 2 +- fs/jbd/transaction.c | 4 ++-- 3 files changed, 12 insertions(+), 12 deletions(-) (limited to 'fs') diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index 0e081d5f32e8..b99c3b3654c4 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -534,7 +534,7 @@ int log_wait_commit(journal_t *journal, tid_t tid) if (!tid_geq(journal->j_commit_request, tid)) { printk(KERN_EMERG "%s: error: j_commit_request=%d, tid=%d\n", - __FUNCTION__, journal->j_commit_request, tid); + __func__, journal->j_commit_request, tid); } spin_unlock(&journal->j_state_lock); #endif @@ -599,7 +599,7 @@ int journal_bmap(journal_t *journal, unsigned long blocknr, printk(KERN_ALERT "%s: journal block not found " "at offset %lu on %s\n", - __FUNCTION__, + __func__, blocknr, bdevname(journal->j_dev, b)); err = -EIO; @@ -728,7 +728,7 @@ journal_t * journal_init_dev(struct block_device *bdev, journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL); if (!journal->j_wbuf) { printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n", - __FUNCTION__); + __func__); kfree(journal); journal = NULL; goto out; @@ -782,7 +782,7 @@ journal_t * journal_init_inode (struct inode *inode) journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL); if (!journal->j_wbuf) { printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n", - __FUNCTION__); + __func__); kfree(journal); return NULL; } @@ -791,7 +791,7 @@ journal_t * journal_init_inode (struct inode *inode) /* If that failed, give up */ if (err) { printk(KERN_ERR "%s: Cannnot locate journal superblock\n", - __FUNCTION__); + __func__); kfree(journal); return NULL; } @@ -877,7 +877,7 @@ int journal_create(journal_t *journal) */ printk(KERN_EMERG "%s: creation of journal on external device!\n", - __FUNCTION__); + __func__); BUG(); } @@ -1657,7 +1657,7 @@ static struct journal_head *journal_alloc_journal_head(void) jbd_debug(1, "out of memory for journal_head\n"); if (time_after(jiffies, last_warning + 5*HZ)) { printk(KERN_NOTICE "ENOMEM in %s, retrying.\n", - __FUNCTION__); + __func__); last_warning = jiffies; } while (ret == NULL) { @@ -1794,13 +1794,13 @@ static void __journal_remove_journal_head(struct buffer_head *bh) if (jh->b_frozen_data) { printk(KERN_WARNING "%s: freeing " "b_frozen_data\n", - __FUNCTION__); + __func__); jbd_free(jh->b_frozen_data, bh->b_size); } if (jh->b_committed_data) { printk(KERN_WARNING "%s: freeing " "b_committed_data\n", - __FUNCTION__); + __func__); jbd_free(jh->b_committed_data, bh->b_size); } bh->b_private = NULL; diff --git a/fs/jbd/revoke.c b/fs/jbd/revoke.c index d5f8eee7c88c..1bb43e987f4b 100644 --- a/fs/jbd/revoke.c +++ b/fs/jbd/revoke.c @@ -138,7 +138,7 @@ repeat: oom: if (!journal_oom_retry) return -ENOMEM; - jbd_debug(1, "ENOMEM in %s, retrying\n", __FUNCTION__); + jbd_debug(1, "ENOMEM in %s, retrying\n", __func__); yield(); goto repeat; } diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index d8dd3d915fd9..67ff2024c23c 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -687,7 +687,7 @@ repeat: if (!frozen_buffer) { printk(KERN_EMERG "%s: OOM for frozen_buffer\n", - __FUNCTION__); + __func__); JBUFFER_TRACE(jh, "oom!"); error = -ENOMEM; jbd_lock_bh_state(bh); @@ -904,7 +904,7 @@ repeat: committed_data = jbd_alloc(jh2bh(jh)->b_size, GFP_NOFS); if (!committed_data) { printk(KERN_EMERG "%s: No memory for committed data\n", - __FUNCTION__); + __func__); err = -ENOMEM; goto out; } -- cgit From 3c5afae2bac974fe2c15cd9439f921526b3ea2d7 Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Mon, 28 Apr 2008 02:16:16 -0700 Subject: ufs: [bl]e*_add_cpu conversion replace all: big/little_endian_variable = cpu_to_[bl]eX([bl]eX_to_cpu(big/little_endian_variable) + expression_in_cpu_byteorder); with: [bl]eX_add_cpu(&big/little_endian_variable, expression_in_cpu_byteorder); generated with semantic patch Signed-off-by: Marcin Slusarz Cc: Evgeniy Dushistov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ufs/swab.h | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/ufs/swab.h b/fs/ufs/swab.h index 1683d2bee614..a4340d0267bb 100644 --- a/fs/ufs/swab.h +++ b/fs/ufs/swab.h @@ -80,18 +80,18 @@ static inline void fs32_add(struct super_block *sbp, __fs32 *n, int d) { if (UFS_SB(sbp)->s_bytesex == BYTESEX_LE) - *(__le32 *)n = cpu_to_le32(le32_to_cpu(*(__le32 *)n)+d); + le32_add_cpu((__le32 *)n, d); else - *(__be32 *)n = cpu_to_be32(be32_to_cpu(*(__be32 *)n)+d); + be32_add_cpu((__be32 *)n, d); } static inline void fs32_sub(struct super_block *sbp, __fs32 *n, int d) { if (UFS_SB(sbp)->s_bytesex == BYTESEX_LE) - *(__le32 *)n = cpu_to_le32(le32_to_cpu(*(__le32 *)n)-d); + le32_add_cpu((__le32 *)n, -d); else - *(__be32 *)n = cpu_to_be32(be32_to_cpu(*(__be32 *)n)-d); + be32_add_cpu((__be32 *)n, -d); } static inline u16 @@ -116,18 +116,18 @@ static inline void fs16_add(struct super_block *sbp, __fs16 *n, int d) { if (UFS_SB(sbp)->s_bytesex == BYTESEX_LE) - *(__le16 *)n = cpu_to_le16(le16_to_cpu(*(__le16 *)n)+d); + le16_add_cpu((__le16 *)n, d); else - *(__be16 *)n = cpu_to_be16(be16_to_cpu(*(__be16 *)n)+d); + be16_add_cpu((__be16 *)n, d); } static inline void fs16_sub(struct super_block *sbp, __fs16 *n, int d) { if (UFS_SB(sbp)->s_bytesex == BYTESEX_LE) - *(__le16 *)n = cpu_to_le16(le16_to_cpu(*(__le16 *)n)-d); + le16_add_cpu((__le16 *)n, -d); else - *(__be16 *)n = cpu_to_be16(be16_to_cpu(*(__be16 *)n)-d); + be16_add_cpu((__be16 *)n, -d); } #endif /* _UFS_SWAB_H */ -- cgit From 9746077a71d01b8d86f5e471a3a2c8f1fa91ef94 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Mon, 28 Apr 2008 02:16:17 -0700 Subject: ufs: replace remaining __FUNCTION__ occurrences __FUNCTION__ is gcc-specific, use __func__ Signed-off-by: Harvey Harrison Cc: Jan Engelhardt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ufs/balloc.c | 4 ++-- fs/ufs/dir.c | 14 +++++++------- fs/ufs/inode.c | 2 +- fs/ufs/ufs.h | 2 +- 4 files changed, 11 insertions(+), 11 deletions(-) (limited to 'fs') diff --git a/fs/ufs/balloc.c b/fs/ufs/balloc.c index 1e7598fb9787..0d9ada173739 100644 --- a/fs/ufs/balloc.c +++ b/fs/ufs/balloc.c @@ -277,7 +277,7 @@ static void ufs_change_blocknr(struct inode *inode, sector_t beg, if (!page)/* it was truncated */ continue; if (IS_ERR(page)) {/* or EIO */ - ufs_error(inode->i_sb, __FUNCTION__, + ufs_error(inode->i_sb, __func__, "read of page %llu failed\n", (unsigned long long)index); continue; @@ -308,7 +308,7 @@ static void ufs_change_blocknr(struct inode *inode, sector_t beg, ll_rw_block(READ, 1, &bh); wait_on_buffer(bh); if (!buffer_uptodate(bh)) { - ufs_error(inode->i_sb, __FUNCTION__, + ufs_error(inode->i_sb, __func__, "read of block failed\n"); break; } diff --git a/fs/ufs/dir.c b/fs/ufs/dir.c index ef563fc8d72c..df0bef18742d 100644 --- a/fs/ufs/dir.c +++ b/fs/ufs/dir.c @@ -179,7 +179,7 @@ bad_entry: goto fail; Eend: p = (struct ufs_dir_entry *)(kaddr + offs); - ufs_error(sb, __FUNCTION__, + ufs_error(sb, __func__, "entry in directory #%lu spans the page boundary" "offset=%lu", dir->i_ino, (page->index<d_reclen == 0) { - ufs_error(dir->i_sb, __FUNCTION__, + ufs_error(dir->i_sb, __func__, "zero-length directory entry"); ufs_put_page(page); goto out; @@ -356,7 +356,7 @@ int ufs_add_link(struct dentry *dentry, struct inode *inode) goto got_it; } if (de->d_reclen == 0) { - ufs_error(dir->i_sb, __FUNCTION__, + ufs_error(dir->i_sb, __func__, "zero-length directory entry"); err = -EIO; goto out_unlock; @@ -456,7 +456,7 @@ ufs_readdir(struct file *filp, void *dirent, filldir_t filldir) struct page *page = ufs_get_page(inode, n); if (IS_ERR(page)) { - ufs_error(sb, __FUNCTION__, + ufs_error(sb, __func__, "bad page in #%lu", inode->i_ino); filp->f_pos += PAGE_CACHE_SIZE - offset; @@ -475,7 +475,7 @@ ufs_readdir(struct file *filp, void *dirent, filldir_t filldir) limit = kaddr + ufs_last_byte(inode, n) - UFS_DIR_REC_LEN(1); for ( ;(char*)de <= limit; de = ufs_next_entry(sb, de)) { if (de->d_reclen == 0) { - ufs_error(sb, __FUNCTION__, + ufs_error(sb, __func__, "zero-length directory entry"); ufs_put_page(page); return -EIO; @@ -536,7 +536,7 @@ int ufs_delete_entry(struct inode *inode, struct ufs_dir_entry *dir, while ((char*)de < (char*)dir) { if (de->d_reclen == 0) { - ufs_error(inode->i_sb, __FUNCTION__, + ufs_error(inode->i_sb, __func__, "zero-length directory entry"); err = -EIO; goto out; @@ -633,7 +633,7 @@ int ufs_empty_dir(struct inode * inode) while ((char *)de <= kaddr) { if (de->d_reclen == 0) { - ufs_error(inode->i_sb, __FUNCTION__, + ufs_error(inode->i_sb, __func__, "zero-length directory entry: " "kaddr=%p, de=%p\n", kaddr, de); goto not_empty; diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c index 5446b888fc8e..39f877898565 100644 --- a/fs/ufs/inode.c +++ b/fs/ufs/inode.c @@ -929,7 +929,7 @@ void ufs_delete_inode (struct inode * inode) old_i_size = inode->i_size; inode->i_size = 0; if (inode->i_blocks && ufs_truncate(inode, old_i_size)) - ufs_warning(inode->i_sb, __FUNCTION__, "ufs_truncate failed\n"); + ufs_warning(inode->i_sb, __func__, "ufs_truncate failed\n"); ufs_free_inode (inode); unlock_kernel(); return; diff --git a/fs/ufs/ufs.h b/fs/ufs/ufs.h index fcb9231bb9ed..244a1aaa940e 100644 --- a/fs/ufs/ufs.h +++ b/fs/ufs/ufs.h @@ -66,7 +66,7 @@ struct ufs_inode_info { #ifdef CONFIG_UFS_DEBUG # define UFSD(f, a...) { \ printk ("UFSD (%s, %d): %s:", \ - __FILE__, __LINE__, __FUNCTION__); \ + __FILE__, __LINE__, __func__); \ printk (f, ## a); \ } #else -- cgit From 0045edaaf980e577272eeb11832daebaeb5d63c4 Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Mon, 28 Apr 2008 02:16:18 -0700 Subject: ufs: remove unused fs64_add and fs64_sub remove fs64_add and fs64_sub - they probably weren't ever used because their prototypes used u32 instead of __fs64 Signed-off-by: Marcin Slusarz Cc: Evgeniy Dushistov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ufs/swab.h | 18 ------------------ 1 file changed, 18 deletions(-) (limited to 'fs') diff --git a/fs/ufs/swab.h b/fs/ufs/swab.h index a4340d0267bb..c89e6ad891fb 100644 --- a/fs/ufs/swab.h +++ b/fs/ufs/swab.h @@ -40,24 +40,6 @@ cpu_to_fs64(struct super_block *sbp, u64 n) return (__force __fs64)cpu_to_be64(n); } -static __inline u32 -fs64_add(struct super_block *sbp, u32 *n, int d) -{ - if (UFS_SB(sbp)->s_bytesex == BYTESEX_LE) - return *n = cpu_to_le64(le64_to_cpu(*n)+d); - else - return *n = cpu_to_be64(be64_to_cpu(*n)+d); -} - -static __inline u32 -fs64_sub(struct super_block *sbp, u32 *n, int d) -{ - if (UFS_SB(sbp)->s_bytesex == BYTESEX_LE) - return *n = cpu_to_le64(le64_to_cpu(*n)-d); - else - return *n = cpu_to_be64(be64_to_cpu(*n)-d); -} - static __inline u32 fs32_to_cpu(struct super_block *sbp, __fs32 n) { -- cgit From 36a53ddf859b0a9aa56e2e72978997bf68eec37a Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Mon, 28 Apr 2008 02:16:18 -0700 Subject: ufs: replace __inline with inline Signed-off-by: Harvey Harrison Cc: Evgeniy Dushistov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ufs/swab.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/ufs/swab.h b/fs/ufs/swab.h index c89e6ad891fb..8d974c4fd18b 100644 --- a/fs/ufs/swab.h +++ b/fs/ufs/swab.h @@ -40,7 +40,7 @@ cpu_to_fs64(struct super_block *sbp, u64 n) return (__force __fs64)cpu_to_be64(n); } -static __inline u32 +static inline u32 fs32_to_cpu(struct super_block *sbp, __fs32 n) { if (UFS_SB(sbp)->s_bytesex == BYTESEX_LE) -- cgit From 78e917d59c30c3d5a4cda7f47e0f40f1c98f9b02 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Mon, 28 Apr 2008 02:16:19 -0700 Subject: udf: fix sparse warning in namei.c Let's use bsize instead. fs/udf/namei.c:960:12: warning: symbol 'elen' shadows an earlier one fs/udf/namei.c:937:15: originally declared here Signed-off-by: Harvey Harrison Cc: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/udf/namei.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs') diff --git a/fs/udf/namei.c b/fs/udf/namei.c index ba5537d4bc15..2b34c8ca6c83 100644 --- a/fs/udf/namei.c +++ b/fs/udf/namei.c @@ -890,7 +890,7 @@ static int udf_symlink(struct inode *dir, struct dentry *dentry, if (iinfo->i_alloc_type != ICBTAG_FLAG_AD_IN_ICB) { kernel_lb_addr eloc; - uint32_t elen; + uint32_t bsize; block = udf_new_block(inode->i_sb, inode, iinfo->i_location.partitionReferenceNum, @@ -903,9 +903,9 @@ static int udf_symlink(struct inode *dir, struct dentry *dentry, eloc.logicalBlockNum = block; eloc.partitionReferenceNum = iinfo->i_location.partitionReferenceNum; - elen = inode->i_sb->s_blocksize; - iinfo->i_lenExtents = elen; - udf_add_aext(inode, &epos, eloc, elen, 0); + bsize = inode->i_sb->s_blocksize; + iinfo->i_lenExtents = bsize; + udf_add_aext(inode, &epos, eloc, bsize, 0); brelse(epos.bh); block = udf_get_pblock(inode->i_sb, block, -- cgit From 9e902df6be2cb7444e5a0f7e2e72bcbf3b978f3e Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Mon, 28 Apr 2008 02:16:20 -0700 Subject: reiserfs: le*_add_cpu conversion replace all: little_endian_variable = cpu_to_leX(leX_to_cpu(little_endian_variable) + expression_in_cpu_byteorder); with: leX_add_cpu(&little_endian_variable, expression_in_cpu_byteorder); generated with semantic patch Signed-off-by: Marcin Slusarz Cc: Jeff Mahoney Cc: Chris Mason Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/objectid.c | 5 ++--- fs/reiserfs/stree.c | 3 +-- 2 files changed, 3 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/objectid.c b/fs/reiserfs/objectid.c index 65feba4deb69..f0c154372f50 100644 --- a/fs/reiserfs/objectid.c +++ b/fs/reiserfs/objectid.c @@ -114,7 +114,7 @@ void reiserfs_release_objectid(struct reiserfs_transaction_handle *th, if (objectid_to_release == le32_to_cpu(map[i])) { /* This incrementation unallocates the objectid. */ //map[i]++; - map[i] = cpu_to_le32(le32_to_cpu(map[i]) + 1); + le32_add_cpu(&map[i], 1); /* Did we unallocate the last member of an odd sequence, and can shrink oids? */ if (map[i] == map[i + 1]) { @@ -138,8 +138,7 @@ void reiserfs_release_objectid(struct reiserfs_transaction_handle *th, /* size of objectid map is not changed */ if (objectid_to_release + 1 == le32_to_cpu(map[i + 1])) { //objectid_map[i+1]--; - map[i + 1] = - cpu_to_le32(le32_to_cpu(map[i + 1]) - 1); + le32_add_cpu(&map[i + 1], -1); return; } diff --git a/fs/reiserfs/stree.c b/fs/reiserfs/stree.c index d2db2417b2bd..abbc64dcc8d4 100644 --- a/fs/reiserfs/stree.c +++ b/fs/reiserfs/stree.c @@ -1419,8 +1419,7 @@ int reiserfs_delete_object(struct reiserfs_transaction_handle *th, inode_generation = &REISERFS_SB(th->t_super)->s_rs->s_inode_generation; - *inode_generation = - cpu_to_le32(le32_to_cpu(*inode_generation) + 1); + le32_add_cpu(inode_generation, 1); } /* USE_INODE_GENERATION_COUNTER */ #endif -- cgit From e13601bc6aa427a6483e7f66fde55ac3b5fed335 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Mon, 28 Apr 2008 02:16:21 -0700 Subject: reiserfs: fix sparse warning in journal.c fs/reiserfs/journal.c:4319:2: warning: returning void-valued expression Signed-off-by: Harvey Harrison Cc: Chris Mason Cc: Jeff Mahoney Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/journal.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 060eb3f598e7..2e87b6aaec72 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -4316,5 +4316,5 @@ static void __reiserfs_journal_abort_soft(struct super_block *sb, int errno) void reiserfs_journal_abort(struct super_block *sb, int errno) { - return __reiserfs_journal_abort_soft(sb, errno); + __reiserfs_journal_abort_soft(sb, errno); } -- cgit From 8acc570fabc022d41baedc0c9bf976a4b9d16d6f Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Mon, 28 Apr 2008 02:16:21 -0700 Subject: reiserfs: fix more sparse warnings in do_balan.c fs/reiserfs/do_balan.c:1467:10: warning: symbol 'ret_val' shadows an earlier one fs/reiserfs/do_balan.c:275:6: originally declared here fs/reiserfs/do_balan.c:1471:23: warning: symbol 'ih' shadows an earlier one fs/reiserfs/do_balan.c:249:67: originally declared here Signed-off-by: Harvey Harrison Cc: Chris Mason Cc: Jeff Mahoney Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/do_balan.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/do_balan.c b/fs/reiserfs/do_balan.c index 7ee4208793b6..2f87f5b14630 100644 --- a/fs/reiserfs/do_balan.c +++ b/fs/reiserfs/do_balan.c @@ -1464,29 +1464,29 @@ static int balance_leaf(struct tree_balance *tb, struct item_head *ih, /* item h } else /* item falls wholly into S_new[i] */ { - int ret_val; + int leaf_mi; struct item_head *pasted; #ifdef CONFIG_REISERFS_CHECK - struct item_head *ih = + struct item_head *ih_check = B_N_PITEM_HEAD(tbS0, item_pos); - if (!is_direntry_le_ih(ih) - && (pos_in_item != ih_item_len(ih) + if (!is_direntry_le_ih(ih_check) + && (pos_in_item != ih_item_len(ih_check) || tb->insert_size[0] <= 0)) reiserfs_panic(tb->tb_sb, "PAP-12235: balance_leaf: pos_in_item must be equal to ih_item_len"); #endif /* CONFIG_REISERFS_CHECK */ - ret_val = + leaf_mi = leaf_move_items(LEAF_FROM_S_TO_SNEW, tb, snum[i], sbytes[i], S_new[i]); - RFALSE(ret_val, + RFALSE(leaf_mi, "PAP-12240: unexpected value returned by leaf_move_items (%d)", - ret_val); + leaf_mi); /* paste into item */ bi.tb = tb; -- cgit From fbe5498b3d5af44a8671800e85b2858c2ea236fa Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Mon, 28 Apr 2008 02:16:22 -0700 Subject: reiserfs: replace remaining __FUNCTION__ occurrences __FUNCTION__ is gcc-specific, use __func__ Signed-off-by: Harvey Harrison Cc: Chris Mason Cc: Jeff Mahoney Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/bitmap.c | 8 ++++---- fs/reiserfs/journal.c | 8 ++++---- fs/reiserfs/namei.c | 8 ++++---- fs/reiserfs/objectid.c | 2 +- 4 files changed, 13 insertions(+), 13 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/bitmap.c b/fs/reiserfs/bitmap.c index f491ceb5af02..4646caa60455 100644 --- a/fs/reiserfs/bitmap.c +++ b/fs/reiserfs/bitmap.c @@ -479,7 +479,7 @@ static void __discard_prealloc(struct reiserfs_transaction_handle *th, if (ei->i_prealloc_count < 0) reiserfs_warning(th->t_super, "zam-4001:%s: inode has negative prealloc blocks count.", - __FUNCTION__); + __func__); #endif while (ei->i_prealloc_count > 0) { reiserfs_free_prealloc_block(th, inode, ei->i_prealloc_block); @@ -517,7 +517,7 @@ void reiserfs_discard_all_prealloc(struct reiserfs_transaction_handle *th) if (!ei->i_prealloc_count) { reiserfs_warning(th->t_super, "zam-4001:%s: inode is in prealloc list but has no preallocated blocks.", - __FUNCTION__); + __func__); } #endif __discard_prealloc(th, ei); @@ -632,7 +632,7 @@ int reiserfs_parse_alloc_options(struct super_block *s, char *options) } reiserfs_warning(s, "zam-4001: %s : unknown option - %s", - __FUNCTION__, this_char); + __func__, this_char); return 1; } @@ -1254,7 +1254,7 @@ struct buffer_head *reiserfs_read_bitmap_block(struct super_block *sb, bh = sb_bread(sb, block); if (bh == NULL) reiserfs_warning(sb, "sh-2029: %s: bitmap block (#%u) " - "reading failed", __FUNCTION__, block); + "reading failed", __func__, block); else { if (buffer_locked(bh)) { PROC_INFO_INC(sb, scan_bitmap.wait); diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 2e87b6aaec72..da86042b3e03 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -1187,7 +1187,7 @@ static int flush_commit_list(struct super_block *s, if (retval) reiserfs_abort(s, retval, "Journal write error in %s", - __FUNCTION__); + __func__); put_fs_excl(); return retval; } @@ -1534,7 +1534,7 @@ static int flush_journal_list(struct super_block *s, reiserfs_warning(s, "clm-2082: Unable to flush buffer %llu in %s", (unsigned long long)saved_bh-> - b_blocknr, __FUNCTION__); + b_blocknr, __func__); } free_cnode: last = cn; @@ -1586,7 +1586,7 @@ static int flush_journal_list(struct super_block *s, if (err) reiserfs_abort(s, -EIO, "Write error while pushing transaction to disk in %s", - __FUNCTION__); + __func__); flush_older_and_return: /* before we can update the journal header block, we _must_ flush all @@ -1616,7 +1616,7 @@ static int flush_journal_list(struct super_block *s, if (err) reiserfs_abort(s, -EIO, "Write error while updating journal header in %s", - __FUNCTION__); + __func__); } remove_all_from_journal_list(s, jl, 0); list_del_init(&jl->j_list); diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c index 8867533cb727..c1add28dd45e 100644 --- a/fs/reiserfs/namei.c +++ b/fs/reiserfs/namei.c @@ -301,7 +301,7 @@ static int reiserfs_find_entry(struct inode *dir, const char *name, int namelen, path_to_entry, de); if (retval == IO_ERROR) { reiserfs_warning(dir->i_sb, "zam-7001: io error in %s", - __FUNCTION__); + __func__); return IO_ERROR; } @@ -496,7 +496,7 @@ static int reiserfs_add_entry(struct reiserfs_transaction_handle *th, reiserfs_warning(dir->i_sb, "zam-7002:%s: \"reiserfs_find_entry\" " "has returned unexpected value (%d)", - __FUNCTION__, retval); + __func__, retval); } return -EEXIST; @@ -907,7 +907,7 @@ static int reiserfs_rmdir(struct inode *dir, struct dentry *dentry) if (inode->i_nlink != 2 && inode->i_nlink != 1) reiserfs_warning(inode->i_sb, "%s: empty directory has nlink " - "!= 2 (%d)", __FUNCTION__, inode->i_nlink); + "!= 2 (%d)", __func__, inode->i_nlink); clear_nlink(inode); inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC; @@ -984,7 +984,7 @@ static int reiserfs_unlink(struct inode *dir, struct dentry *dentry) if (!inode->i_nlink) { reiserfs_warning(inode->i_sb, "%s: deleting nonexistent file " - "(%s:%lu), %d", __FUNCTION__, + "(%s:%lu), %d", __func__, reiserfs_bdevname(inode->i_sb), inode->i_ino, inode->i_nlink); inode->i_nlink = 1; diff --git a/fs/reiserfs/objectid.c b/fs/reiserfs/objectid.c index f0c154372f50..ea0cf8c28a99 100644 --- a/fs/reiserfs/objectid.c +++ b/fs/reiserfs/objectid.c @@ -61,7 +61,7 @@ __u32 reiserfs_get_unused_objectid(struct reiserfs_transaction_handle *th) /* comment needed -Hans */ unused_objectid = le32_to_cpu(map[1]); if (unused_objectid == U32_MAX) { - reiserfs_warning(s, "%s: no more object ids", __FUNCTION__); + reiserfs_warning(s, "%s: no more object ids", __func__); reiserfs_restore_prepared_buffer(s, SB_BUFFER_WITH_SB(s)); return 0; } -- cgit From a2fe594fa3732298947377409b5d8111b14f56ec Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 28 Apr 2008 02:16:23 -0700 Subject: reiserfs: fix hang on umount with quotas when journal is aborted Call dquot_drop() from reiserfs_dquot_drop() even if we fail to start a transaction. Otherwise we never get to dropping references to quota structures from the inode and umount will hang indefinitely. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/super.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index f8f9a473e670..a4c9770a6012 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -1890,8 +1890,14 @@ static int reiserfs_dquot_drop(struct inode *inode) ret = journal_begin(&th, inode->i_sb, 2 * REISERFS_QUOTA_DEL_BLOCKS(inode->i_sb)); - if (ret) + if (ret) { + /* + * We call dquot_drop() anyway to at least release references + * to quota structures so that umount does not hang. + */ + dquot_drop(inode); goto out; + } ret = dquot_drop(inode); err = journal_end(&th, inode->i_sb, -- cgit From d5dee5c395062a55236318ac4eec1f4ebb9de6db Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 28 Apr 2008 02:16:23 -0700 Subject: reiserfs: unpack tails on quota files Quota files cannot have tails because quota_write and quota_read functions do not support them. So far when quota files did have tail, we just refused to turn quotas on it. Sadly this check has been wrong and so there are now plenty installations where quota files don't have NOTAIL flag set and so now after fixing the check, they suddently fail to turn quotas on. Since it's easy to unpack the tail from kernel, do this from reiserfs_quota_on() which solves the problem and is generally nicer to users anyway. Signed-off-by: Jan Kara Reported-by: Cc: Jeff Mahoney Cc: Chris Mason Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/ioctl.c | 4 +--- fs/reiserfs/super.c | 17 ++++++++++++----- 2 files changed, 13 insertions(+), 8 deletions(-) (limited to 'fs') diff --git a/fs/reiserfs/ioctl.c b/fs/reiserfs/ioctl.c index 74363a7aacbc..830332021ed4 100644 --- a/fs/reiserfs/ioctl.c +++ b/fs/reiserfs/ioctl.c @@ -12,8 +12,6 @@ #include #include -static int reiserfs_unpack(struct inode *inode, struct file *filp); - /* ** reiserfs_ioctl - handler for ioctl for inode ** supported commands: @@ -159,7 +157,7 @@ int reiserfs_prepare_write(struct file *f, struct page *page, ** Function try to convert tail from direct item into indirect. ** It set up nopack attribute in the REISERFS_I(inode)->nopack */ -static int reiserfs_unpack(struct inode *inode, struct file *filp) +int reiserfs_unpack(struct inode *inode, struct file *filp) { int retval = 0; int index; diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index a4c9770a6012..ed424d708e69 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -2025,6 +2025,7 @@ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id, { int err; struct nameidata nd; + struct inode *inode; if (!(REISERFS_SB(sb)->s_mount_opt & (1 << REISERFS_QUOTA))) return -EINVAL; @@ -2039,12 +2040,18 @@ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id, path_put(&nd.path); return -EXDEV; } + inode = nd.path.dentry->d_inode; /* We must not pack tails for quota files on reiserfs for quota IO to work */ - if (!(REISERFS_I(nd.path.dentry->d_inode)->i_flags & i_nopack_mask)) { - reiserfs_warning(sb, - "reiserfs: Quota file must have tail packing disabled."); - path_put(&nd.path); - return -EINVAL; + if (!(REISERFS_I(inode)->i_flags & i_nopack_mask)) { + err = reiserfs_unpack(inode, NULL); + if (err) { + reiserfs_warning(sb, + "reiserfs: Unpacking tail of quota file failed" + " (%d). Cannot turn on quotas.", err); + path_put(&nd.path); + return -EINVAL; + } + mark_inode_dirty(inode); } /* Not journalling quota? No more tests needed... */ if (!REISERFS_SB(sb)->s_qf_names[USRQUOTA] && -- cgit From 3754a544474b9fd4fdd32a96d98bc4ab558b2bb6 Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Mon, 28 Apr 2008 02:16:24 -0700 Subject: fat: kill is_bad_inode() check FAT doesn't need to check bad inode anymore. Signed-off-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fat/inode.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) (limited to 'fs') diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 53f3cf62b7c1..f22f287e507a 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -433,11 +433,8 @@ EXPORT_SYMBOL_GPL(fat_build_inode); static void fat_delete_inode(struct inode *inode) { truncate_inode_pages(&inode->i_data, 0); - - if (!is_bad_inode(inode)) { - inode->i_size = 0; - fat_truncate(inode); - } + inode->i_size = 0; + fat_truncate(inode); clear_inode(inode); } @@ -445,8 +442,6 @@ static void fat_clear_inode(struct inode *inode) { struct msdos_sb_info *sbi = MSDOS_SB(inode->i_sb); - if (is_bad_inode(inode)) - return; lock_kernel(); spin_lock(&sbi->inode_hash_lock); fat_cache_inval_inode(inode); -- cgit From 1278fdd34b12214b5c2e91e64848a5e54e57ed96 Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Mon, 28 Apr 2008 02:16:25 -0700 Subject: fat: fat_notify_change() and check_mode() cleanup - Rename fat_notify_change() to fat_setattr() - check_mode() cleanup - Change layout of code Signed-off-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fat/file.c | 183 +++++++++++++++++++++++++------------------------------ fs/msdos/namei.c | 2 +- fs/vfat/namei.c | 2 +- 3 files changed, 86 insertions(+), 101 deletions(-) (limited to 'fs') diff --git a/fs/fat/file.c b/fs/fat/file.c index 2a3bed967041..b61a98f5398a 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -157,104 +157,6 @@ out: return err; } -static int check_mode(const struct msdos_sb_info *sbi, mode_t mode) -{ - mode_t req = mode & ~S_IFMT; - - /* - * Of the r and x bits, all (subject to umask) must be present. Of the - * w bits, either all (subject to umask) or none must be present. - */ - - if (S_ISREG(mode)) { - req &= ~sbi->options.fs_fmask; - - if ((req & (S_IRUGO | S_IXUGO)) != - ((S_IRUGO | S_IXUGO) & ~sbi->options.fs_fmask)) - return -EPERM; - - if ((req & S_IWUGO) != 0 && - (req & S_IWUGO) != (S_IWUGO & ~sbi->options.fs_fmask)) - return -EPERM; - } else if (S_ISDIR(mode)) { - req &= ~sbi->options.fs_dmask; - - if ((req & (S_IRUGO | S_IXUGO)) != - ((S_IRUGO | S_IXUGO) & ~sbi->options.fs_dmask)) - return -EPERM; - - if ((req & S_IWUGO) != 0 && - (req & S_IWUGO) != (S_IWUGO & ~sbi->options.fs_dmask)) - return -EPERM; - } else { - return -EPERM; - } - - return 0; -} - -int fat_notify_change(struct dentry *dentry, struct iattr *attr) -{ - struct msdos_sb_info *sbi = MSDOS_SB(dentry->d_sb); - struct inode *inode = dentry->d_inode; - int mask, error = 0; - - lock_kernel(); - - /* - * Expand the file. Since inode_setattr() updates ->i_size - * before calling the ->truncate(), but FAT needs to fill the - * hole before it. - */ - if (attr->ia_valid & ATTR_SIZE) { - if (attr->ia_size > inode->i_size) { - error = fat_cont_expand(inode, attr->ia_size); - if (error || attr->ia_valid == ATTR_SIZE) - goto out; - attr->ia_valid &= ~ATTR_SIZE; - } - } - - error = inode_change_ok(inode, attr); - if (error) { - if (sbi->options.quiet) - error = 0; - goto out; - } - if (((attr->ia_valid & ATTR_UID) && - (attr->ia_uid != sbi->options.fs_uid)) || - ((attr->ia_valid & ATTR_GID) && - (attr->ia_gid != sbi->options.fs_gid))) - error = -EPERM; - - if (error) { - if (sbi->options.quiet) - error = 0; - goto out; - } - - if (attr->ia_valid & ATTR_MODE) { - error = check_mode(sbi, attr->ia_mode); - if (error != 0 && !sbi->options.quiet) - goto out; - } - - error = inode_setattr(inode, attr); - if (error) - goto out; - - if (S_ISDIR(inode->i_mode)) - mask = sbi->options.fs_dmask; - else - mask = sbi->options.fs_fmask; - inode->i_mode &= S_IFMT | (S_IRWXUGO & ~mask); -out: - unlock_kernel(); - return error; -} - -EXPORT_SYMBOL_GPL(fat_notify_change); - /* Free all clusters after the skip'th cluster. */ static int fat_free(struct inode *inode, int skip) { @@ -355,8 +257,91 @@ int fat_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) } EXPORT_SYMBOL_GPL(fat_getattr); +static int fat_check_mode(const struct msdos_sb_info *sbi, mode_t mode) +{ + mode_t mask, req = mode & ~S_IFMT; + + if (S_ISREG(mode)) + mask = sbi->options.fs_fmask; + else + mask = sbi->options.fs_dmask; + + /* + * Of the r and x bits, all (subject to umask) must be present. Of the + * w bits, either all (subject to umask) or none must be present. + */ + req &= ~mask; + if ((req & (S_IRUGO | S_IXUGO)) != ((S_IRUGO | S_IXUGO) & ~mask)) + return -EPERM; + if ((req & S_IWUGO) && ((req & S_IWUGO) != (S_IWUGO & ~mask))) + return -EPERM; + + return 0; +} + +int fat_setattr(struct dentry *dentry, struct iattr *attr) +{ + struct msdos_sb_info *sbi = MSDOS_SB(dentry->d_sb); + struct inode *inode = dentry->d_inode; + int mask, error = 0; + + lock_kernel(); + + /* + * Expand the file. Since inode_setattr() updates ->i_size + * before calling the ->truncate(), but FAT needs to fill the + * hole before it. + */ + if (attr->ia_valid & ATTR_SIZE) { + if (attr->ia_size > inode->i_size) { + error = fat_cont_expand(inode, attr->ia_size); + if (error || attr->ia_valid == ATTR_SIZE) + goto out; + attr->ia_valid &= ~ATTR_SIZE; + } + } + + error = inode_change_ok(inode, attr); + if (error) { + if (sbi->options.quiet) + error = 0; + goto out; + } + if (((attr->ia_valid & ATTR_UID) && + (attr->ia_uid != sbi->options.fs_uid)) || + ((attr->ia_valid & ATTR_GID) && + (attr->ia_gid != sbi->options.fs_gid))) + error = -EPERM; + + if (error) { + if (sbi->options.quiet) + error = 0; + goto out; + } + + if (attr->ia_valid & ATTR_MODE) { + error = fat_check_mode(sbi, attr->ia_mode); + if (error != 0 && !sbi->options.quiet) + goto out; + } + + error = inode_setattr(inode, attr); + if (error) + goto out; + + if (S_ISDIR(inode->i_mode)) + mask = sbi->options.fs_dmask; + else + mask = sbi->options.fs_fmask; + inode->i_mode &= S_IFMT | (S_IRWXUGO & ~mask); +out: + unlock_kernel(); + return error; +} +EXPORT_SYMBOL_GPL(fat_setattr); + const struct inode_operations fat_file_inode_operations = { .truncate = fat_truncate, - .setattr = fat_notify_change, + .setattr = fat_setattr, .getattr = fat_getattr, }; diff --git a/fs/msdos/namei.c b/fs/msdos/namei.c index 30f7d0ae2215..2d4358c59f68 100644 --- a/fs/msdos/namei.c +++ b/fs/msdos/namei.c @@ -653,7 +653,7 @@ static const struct inode_operations msdos_dir_inode_operations = { .mkdir = msdos_mkdir, .rmdir = msdos_rmdir, .rename = msdos_rename, - .setattr = fat_notify_change, + .setattr = fat_setattr, .getattr = fat_getattr, }; diff --git a/fs/vfat/namei.c b/fs/vfat/namei.c index cd450bea9f1a..efc70576e4b4 100644 --- a/fs/vfat/namei.c +++ b/fs/vfat/namei.c @@ -1003,7 +1003,7 @@ static const struct inode_operations vfat_dir_inode_operations = { .mkdir = vfat_mkdir, .rmdir = vfat_rmdir, .rename = vfat_rename, - .setattr = fat_notify_change, + .setattr = fat_setattr, .getattr = fat_getattr, }; -- cgit From e97e8de388723f9491514fa0434ddf1fd713a188 Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Mon, 28 Apr 2008 02:16:26 -0700 Subject: fat: fat_setattr() fix Fix fat_setattr() on the case of showexec option. If user specified showexec option, inode->i_mode may not have S_IXUGO. This just use inode->i_mode to fix it. And with this patch, we don't allow chmod() on memory inode, it's just bad behaviour. IOW, we allow changing S_IWUGO only which can be stored to disk. Signed-off-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fat/file.c | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-) (limited to 'fs') diff --git a/fs/fat/file.c b/fs/fat/file.c index b61a98f5398a..e73f13a13792 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -257,7 +257,8 @@ int fat_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) } EXPORT_SYMBOL_GPL(fat_getattr); -static int fat_check_mode(const struct msdos_sb_info *sbi, mode_t mode) +static int fat_check_mode(const struct msdos_sb_info *sbi, struct inode *inode, + mode_t mode) { mode_t mask, req = mode & ~S_IFMT; @@ -271,7 +272,7 @@ static int fat_check_mode(const struct msdos_sb_info *sbi, mode_t mode) * w bits, either all (subject to umask) or none must be present. */ req &= ~mask; - if ((req & (S_IRUGO | S_IXUGO)) != ((S_IRUGO | S_IXUGO) & ~mask)) + if ((req & (S_IRUGO | S_IXUGO)) != (inode->i_mode & (S_IRUGO|S_IXUGO))) return -EPERM; if ((req & S_IWUGO) && ((req & S_IWUGO) != (S_IWUGO & ~mask))) return -EPERM; @@ -310,7 +311,9 @@ int fat_setattr(struct dentry *dentry, struct iattr *attr) if (((attr->ia_valid & ATTR_UID) && (attr->ia_uid != sbi->options.fs_uid)) || ((attr->ia_valid & ATTR_GID) && - (attr->ia_gid != sbi->options.fs_gid))) + (attr->ia_gid != sbi->options.fs_gid)) || + ((attr->ia_valid & ATTR_MODE) && + fat_check_mode(sbi, inode, attr->ia_mode) < 0)) error = -EPERM; if (error) { @@ -319,12 +322,6 @@ int fat_setattr(struct dentry *dentry, struct iattr *attr) goto out; } - if (attr->ia_valid & ATTR_MODE) { - error = fat_check_mode(sbi, attr->ia_mode); - if (error != 0 && !sbi->options.quiet) - goto out; - } - error = inode_setattr(inode, attr); if (error) goto out; -- cgit From 1ae43f826b6cb951fc5b0f9c92372a8d5b63c7f9 Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Mon, 28 Apr 2008 02:16:26 -0700 Subject: fat: Add allow_utime option Normally utime(2) checks current process is owner of the file, or it has CAP_FOWNER capability. But FAT filesystem doesn't have uid/gid as on disk info, so normal check is too unflexible. With this option you can relax it. Signed-off-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fat/file.c | 24 ++++++++++++++++++++++++ fs/fat/inode.c | 18 +++++++++++++++--- 2 files changed, 39 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/fat/file.c b/fs/fat/file.c index e73f13a13792..d604bb132422 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -280,11 +280,27 @@ static int fat_check_mode(const struct msdos_sb_info *sbi, struct inode *inode, return 0; } +static int fat_allow_set_time(struct msdos_sb_info *sbi, struct inode *inode) +{ + mode_t allow_utime = sbi->options.allow_utime; + + if (current->fsuid != inode->i_uid) { + if (in_group_p(inode->i_gid)) + allow_utime >>= 3; + if (allow_utime & MAY_WRITE) + return 1; + } + + /* use a default check */ + return 0; +} + int fat_setattr(struct dentry *dentry, struct iattr *attr) { struct msdos_sb_info *sbi = MSDOS_SB(dentry->d_sb); struct inode *inode = dentry->d_inode; int mask, error = 0; + unsigned int ia_valid; lock_kernel(); @@ -302,7 +318,15 @@ int fat_setattr(struct dentry *dentry, struct iattr *attr) } } + /* Check for setting the inode time. */ + ia_valid = attr->ia_valid; + if (ia_valid & (ATTR_MTIME_SET | ATTR_ATIME_SET)) { + if (fat_allow_set_time(sbi, inode)) + attr->ia_valid &= ~(ATTR_MTIME_SET | ATTR_ATIME_SET); + } + error = inode_change_ok(inode, attr); + attr->ia_valid = ia_valid; if (error) { if (sbi->options.quiet) error = 0; diff --git a/fs/fat/inode.c b/fs/fat/inode.c index f22f287e507a..886f6095a378 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -785,6 +785,8 @@ static int fat_show_options(struct seq_file *m, struct vfsmount *mnt) seq_printf(m, ",gid=%u", opts->fs_gid); seq_printf(m, ",fmask=%04o", opts->fs_fmask); seq_printf(m, ",dmask=%04o", opts->fs_dmask); + if (opts->allow_utime) + seq_printf(m, ",allow_utime=%04o", opts->allow_utime); if (sbi->nls_disk) seq_printf(m, ",codepage=%s", sbi->nls_disk->charset); if (isvfat) { @@ -840,9 +842,9 @@ static int fat_show_options(struct seq_file *m, struct vfsmount *mnt) enum { Opt_check_n, Opt_check_r, Opt_check_s, Opt_uid, Opt_gid, - Opt_umask, Opt_dmask, Opt_fmask, Opt_codepage, Opt_usefree, Opt_nocase, - Opt_quiet, Opt_showexec, Opt_debug, Opt_immutable, - Opt_dots, Opt_nodots, + Opt_umask, Opt_dmask, Opt_fmask, Opt_allow_utime, Opt_codepage, + Opt_usefree, Opt_nocase, Opt_quiet, Opt_showexec, Opt_debug, + Opt_immutable, Opt_dots, Opt_nodots, Opt_charset, Opt_shortname_lower, Opt_shortname_win95, Opt_shortname_winnt, Opt_shortname_mixed, Opt_utf8_no, Opt_utf8_yes, Opt_uni_xl_no, Opt_uni_xl_yes, Opt_nonumtail_no, Opt_nonumtail_yes, @@ -861,6 +863,7 @@ static match_table_t fat_tokens = { {Opt_umask, "umask=%o"}, {Opt_dmask, "dmask=%o"}, {Opt_fmask, "fmask=%o"}, + {Opt_allow_utime, "allow_utime=%o"}, {Opt_codepage, "codepage=%u"}, {Opt_usefree, "usefree"}, {Opt_nocase, "nocase"}, @@ -932,6 +935,7 @@ static int parse_options(char *options, int is_vfat, int silent, int *debug, opts->fs_uid = current->uid; opts->fs_gid = current->gid; opts->fs_fmask = opts->fs_dmask = current->fs->umask; + opts->allow_utime = -1; opts->codepage = fat_default_codepage; opts->iocharset = fat_default_iocharset; if (is_vfat) @@ -1019,6 +1023,11 @@ static int parse_options(char *options, int is_vfat, int silent, int *debug, return 0; opts->fs_fmask = option; break; + case Opt_allow_utime: + if (match_octal(&args[0], &option)) + return 0; + opts->allow_utime = option & (S_IWGRP | S_IWOTH); + break; case Opt_codepage: if (match_int(&args[0], &option)) return 0; @@ -1101,6 +1110,9 @@ static int parse_options(char *options, int is_vfat, int silent, int *debug, " for FAT filesystems, filesystem will be case sensitive!\n"); } + /* If user doesn't specify allow_utime, it's initialized from dmask. */ + if (opts->allow_utime == (unsigned short)-1) + opts->allow_utime = ~opts->fs_dmask & (S_IWGRP | S_IWOTH); if (opts->unicode_xlate) opts->utf8 = 0; -- cgit From 606e423e43bac0c2f7b85b682eb1ddd2a634586e Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Mon, 28 Apr 2008 02:16:27 -0700 Subject: fat: Update free_clusters even if it is untrusted Currently, free_clusters is not updated until it is trusted, because Windows doesn't update it correctly. But if user is using FAT driver of Linux, it updates free_clusters correctly. Instead, this updates it even if it's untrusted, so if free_clustes is correct, now keep correct value. Signed-off-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fat/fatent.c | 7 +++++-- fs/fat/inode.c | 7 ++++--- 2 files changed, 9 insertions(+), 5 deletions(-) (limited to 'fs') diff --git a/fs/fat/fatent.c b/fs/fat/fatent.c index 5fb366992b73..c7c8ab2209e1 100644 --- a/fs/fat/fatent.c +++ b/fs/fat/fatent.c @@ -450,7 +450,8 @@ int fat_alloc_clusters(struct inode *inode, int *cluster, int nr_cluster) BUG_ON(nr_cluster > (MAX_BUF_PER_PAGE / 2)); /* fixed limit */ lock_fat(sbi); - if (sbi->free_clusters != -1 && sbi->free_clusters < nr_cluster) { + if (sbi->free_clusters != -1 && sbi->free_clus_valid && + sbi->free_clusters < nr_cluster) { unlock_fat(sbi); return -ENOSPC; } @@ -504,6 +505,7 @@ int fat_alloc_clusters(struct inode *inode, int *cluster, int nr_cluster) /* Couldn't allocate the free entries */ sbi->free_clusters = 0; + sbi->free_clus_valid = 1; sb->s_dirt = 1; err = -ENOSPC; @@ -615,7 +617,7 @@ int fat_count_free_clusters(struct super_block *sb) int err = 0, free; lock_fat(sbi); - if (sbi->free_clusters != -1) + if (sbi->free_clusters != -1 && sbi->free_clus_valid) goto out; reada_blocks = FAT_READA_SIZE >> sb->s_blocksize_bits; @@ -643,6 +645,7 @@ int fat_count_free_clusters(struct super_block *sb) } while (fat_ent_next(sbi, &fatent)); } sbi->free_clusters = free; + sbi->free_clus_valid = 1; sb->s_dirt = 1; fatent_brelse(&fatent); out: diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 886f6095a378..1604a8a5c01f 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -537,7 +537,7 @@ static int fat_statfs(struct dentry *dentry, struct kstatfs *buf) struct msdos_sb_info *sbi = MSDOS_SB(dentry->d_sb); /* If the count of free cluster is still unknown, counts it here. */ - if (sbi->free_clusters == -1) { + if (sbi->free_clusters == -1 || !sbi->free_clus_valid) { int err = fat_count_free_clusters(dentry->d_sb); if (err) return err; @@ -1274,6 +1274,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, sbi->fat_length = le16_to_cpu(b->fat_length); sbi->root_cluster = 0; sbi->free_clusters = -1; /* Don't know yet */ + sbi->free_clus_valid = 0; sbi->prev_free = FAT_START_ENT; if (!sbi->fat_length && b->fat32_length) { @@ -1309,8 +1310,8 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, sbi->fsinfo_sector); } else { if (sbi->options.usefree) - sbi->free_clusters = - le32_to_cpu(fsinfo->free_clusters); + sbi->free_clus_valid = 1; + sbi->free_clusters = le32_to_cpu(fsinfo->free_clusters); sbi->prev_free = le32_to_cpu(fsinfo->next_cluster); } -- cgit From e69be4c9c4ac1b6bb6380512557da0afa6262e98 Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Mon, 28 Apr 2008 02:16:27 -0700 Subject: fat: Remove fat_clusters_flush() This removes unneeded fat_clusters_flush(). Signed-off-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fat/fatent.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'fs') diff --git a/fs/fat/fatent.c b/fs/fat/fatent.c index c7c8ab2209e1..13ab763cc510 100644 --- a/fs/fat/fatent.c +++ b/fs/fat/fatent.c @@ -585,8 +585,6 @@ error: brelse(bhs[i]); unlock_fat(sbi); - fat_clusters_flush(sb); - return err; } -- cgit From 061e97469f46f924cf14bbf1dd4805b46986691a Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Mon, 28 Apr 2008 02:16:28 -0700 Subject: Add balance_dirty_pages_ratelimited() to cont_expand_zero() On the systems, ftruncate() which expand size for FAT became the cause of OOM. The cont_expand_zero() filled all memory with dirty pages, and since disk is very slow, limit of page scanning was exceeded, then it triggered OOM. This adds balance_dirty_pages_ratelimited() to avoid filling memory with dirty pages. Signed-off-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs') diff --git a/fs/buffer.c b/fs/buffer.c index 7d51e649b19a..3db4a26adc44 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2246,6 +2246,8 @@ int cont_expand_zero(struct file *file, struct address_space *mapping, goto out; BUG_ON(err != len); err = 0; + + balance_dirty_pages_ratelimited(mapping); } /* page covers the boundary, find the boundary offset */ -- cgit From f22032ba8de2960a64a3dd9719fb5c99b1f1ae6e Mon Sep 17 00:00:00 2001 From: Keith Mok Date: Mon, 28 Apr 2008 02:16:29 -0700 Subject: vfat: bug fix for vfat cannot handle filename with 255 This patch fix the problem that the buffer allocated for convert of unicode to utf8 in fat/dir.c is too small. And cannot handle filename with 255 asian characters when mounted with utf8 options. Also it fix the filename length limitation checking in vfat/namei.c that the filename length should be checked against the number of converted unicode characters. Not the length before NLS/UTF8 converted. Signed-off-by: Keith Mok Signed-off-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fat/dir.c | 44 ++++++++++++++++++++++++++++---------------- fs/vfat/namei.c | 25 +++++++++++-------------- 2 files changed, 39 insertions(+), 30 deletions(-) (limited to 'fs') diff --git a/fs/fat/dir.c b/fs/fat/dir.c index 72cbcd61bd95..7b62ffb663f6 100644 --- a/fs/fat/dir.c +++ b/fs/fat/dir.c @@ -124,8 +124,8 @@ static inline int fat_get_entry(struct inode *dir, loff_t *pos, * but ignore that right now. * Ahem... Stack smashing in ring 0 isn't fun. Fixed. */ -static int uni16_to_x8(unsigned char *ascii, wchar_t *uni, int uni_xlate, - struct nls_table *nls) +static int uni16_to_x8(unsigned char *ascii, wchar_t *uni, int len, + int uni_xlate, struct nls_table *nls) { wchar_t *ip, ec; unsigned char *op, nc; @@ -135,10 +135,11 @@ static int uni16_to_x8(unsigned char *ascii, wchar_t *uni, int uni_xlate, ip = uni; op = ascii; - while (*ip) { + while (*ip && ((len - NLS_MAX_CHARSET_SIZE) > 0)) { ec = *ip++; if ( (charlen = nls->uni2char(ec, op, NLS_MAX_CHARSET_SIZE)) > 0) { op += charlen; + len -= charlen; } else { if (uni_xlate == 1) { *op = ':'; @@ -149,16 +150,19 @@ static int uni16_to_x8(unsigned char *ascii, wchar_t *uni, int uni_xlate, ec >>= 4; } op += 5; + len -= 5; } else { *op++ = '?'; + len--; } } - /* We have some slack there, so it's OK */ - if (op>ascii+256) { - op = ascii + 256; - break; - } } + + if (unlikely(*ip)) { + printk(KERN_WARNING "FAT: filename was truncated while " + "converting."); + } + *op = 0; return (op - ascii); } @@ -311,9 +315,11 @@ int fat_search_long(struct inode *inode, const unsigned char *name, struct nls_table *nls_io = sbi->nls_io; struct nls_table *nls_disk = sbi->nls_disk; wchar_t bufuname[14]; - unsigned char xlate_len, nr_slots; + unsigned char nr_slots; + int xlate_len; wchar_t *unicode = NULL; - unsigned char work[MSDOS_NAME], bufname[260]; /* 256 + 4 */ + unsigned char work[MSDOS_NAME]; + unsigned char *bufname = NULL; int uni_xlate = sbi->options.unicode_xlate; int utf8 = sbi->options.utf8; int anycase = (sbi->options.name_check != 's'); @@ -321,6 +327,10 @@ int fat_search_long(struct inode *inode, const unsigned char *name, loff_t cpos = 0; int chl, i, j, last_u, err; + bufname = (unsigned char*)__get_free_page(GFP_KERNEL); + if (!bufname) + return -ENOMEM; + err = -ENOENT; while(1) { if (fat_get_entry(inode, &cpos, &bh, &de) == -1) @@ -386,8 +396,8 @@ parse_record: bufuname[last_u] = 0x0000; xlate_len = utf8 - ?utf8_wcstombs(bufname, bufuname, sizeof(bufname)) - :uni16_to_x8(bufname, bufuname, uni_xlate, nls_io); + ?utf8_wcstombs(bufname, bufuname, PAGE_SIZE) + :uni16_to_x8(bufname, bufuname, PAGE_SIZE, uni_xlate, nls_io); if (xlate_len == name_len) if ((!anycase && !memcmp(name, bufname, xlate_len)) || (anycase && !nls_strnicmp(nls_io, name, bufname, @@ -396,8 +406,8 @@ parse_record: if (nr_slots) { xlate_len = utf8 - ?utf8_wcstombs(bufname, unicode, sizeof(bufname)) - :uni16_to_x8(bufname, unicode, uni_xlate, nls_io); + ?utf8_wcstombs(bufname, unicode, PAGE_SIZE) + :uni16_to_x8(bufname, unicode, PAGE_SIZE, uni_xlate, nls_io); if (xlate_len != name_len) continue; if ((!anycase && !memcmp(name, bufname, xlate_len)) || @@ -416,6 +426,8 @@ Found: sinfo->i_pos = fat_make_i_pos(sb, sinfo->bh, sinfo->de); err = 0; EODir: + if (bufname) + free_page((unsigned long)bufname); if (unicode) free_page((unsigned long)unicode); @@ -598,7 +610,7 @@ parse_record: if (isvfat) { bufuname[j] = 0x0000; i = utf8 ? utf8_wcstombs(bufname, bufuname, sizeof(bufname)) - : uni16_to_x8(bufname, bufuname, uni_xlate, nls_io); + : uni16_to_x8(bufname, bufuname, sizeof(bufname), uni_xlate, nls_io); } fill_name = bufname; @@ -610,7 +622,7 @@ parse_record: int buf_size = PAGE_SIZE - (261 * sizeof(unicode[0])); int long_len = utf8 ? utf8_wcstombs(longname, unicode, buf_size) - : uni16_to_x8(longname, unicode, uni_xlate, nls_io); + : uni16_to_x8(longname, unicode, buf_size, uni_xlate, nls_io); if (!both) { fill_name = longname; diff --git a/fs/vfat/namei.c b/fs/vfat/namei.c index efc70576e4b4..ab4f3da770f0 100644 --- a/fs/vfat/namei.c +++ b/fs/vfat/namei.c @@ -176,15 +176,10 @@ static inline int vfat_is_used_badchars(const wchar_t *s, int len) for (i = 0; i < len; i++) if (vfat_bad_char(s[i])) return -EINVAL; - return 0; -} -static int vfat_valid_longname(const unsigned char *name, unsigned int len) -{ - if (name[len - 1] == ' ') + if (s[i - 1] == ' ') /* last character cannot be space */ return -EINVAL; - if (len >= 256) - return -ENAMETOOLONG; + return 0; } @@ -485,11 +480,14 @@ xlate_to_uni(const unsigned char *name, int len, unsigned char *outname, */ *outlen -= (name_len - len); + if (*outlen > 255) + return -ENAMETOOLONG; + op = &outname[*outlen * sizeof(wchar_t)]; } else { if (nls) { for (i = 0, ip = name, op = outname, *outlen = 0; - i < len && *outlen <= 260; + i < len && *outlen <= 255; *outlen += 1) { if (escape && (*ip == ':')) { @@ -525,18 +523,20 @@ xlate_to_uni(const unsigned char *name, int len, unsigned char *outname, op += 2; } } + if (i < len) + return -ENAMETOOLONG; } else { for (i = 0, ip = name, op = outname, *outlen = 0; - i < len && *outlen <= 260; + i < len && *outlen <= 255; i++, *outlen += 1) { *op++ = *ip++; *op++ = 0; } + if (i < len) + return -ENAMETOOLONG; } } - if (*outlen > 260) - return -ENAMETOOLONG; *longlen = *outlen; if (*outlen % 13) { @@ -574,9 +574,6 @@ static int vfat_build_slots(struct inode *dir, const unsigned char *name, loff_t offset; *nr_slots = 0; - err = vfat_valid_longname(name, len); - if (err) - return err; page = __get_free_page(GFP_KERNEL); if (!page) -- cgit From c7a6c4edc74ee5c3d990325fe93b6916f43499f4 Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Mon, 28 Apr 2008 02:16:29 -0700 Subject: fat: use __getname() __getname() is faster than __get_free_page(). Use it. Signed-off-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fat/dir.c | 20 ++++++++++---------- fs/vfat/namei.c | 10 ++++------ 2 files changed, 14 insertions(+), 16 deletions(-) (limited to 'fs') diff --git a/fs/fat/dir.c b/fs/fat/dir.c index 7b62ffb663f6..486725ee99ae 100644 --- a/fs/fat/dir.c +++ b/fs/fat/dir.c @@ -247,7 +247,7 @@ static int fat_parse_long(struct inode *dir, loff_t *pos, unsigned char id, slot, slots, alias_checksum; if (!*unicode) { - *unicode = (wchar_t *)__get_free_page(GFP_KERNEL); + *unicode = __getname(); if (!*unicode) { brelse(*bh); return -ENOMEM; @@ -327,7 +327,7 @@ int fat_search_long(struct inode *inode, const unsigned char *name, loff_t cpos = 0; int chl, i, j, last_u, err; - bufname = (unsigned char*)__get_free_page(GFP_KERNEL); + bufname = __getname(); if (!bufname) return -ENOMEM; @@ -396,8 +396,8 @@ parse_record: bufuname[last_u] = 0x0000; xlate_len = utf8 - ?utf8_wcstombs(bufname, bufuname, PAGE_SIZE) - :uni16_to_x8(bufname, bufuname, PAGE_SIZE, uni_xlate, nls_io); + ?utf8_wcstombs(bufname, bufuname, PATH_MAX) + :uni16_to_x8(bufname, bufuname, PATH_MAX, uni_xlate, nls_io); if (xlate_len == name_len) if ((!anycase && !memcmp(name, bufname, xlate_len)) || (anycase && !nls_strnicmp(nls_io, name, bufname, @@ -406,8 +406,8 @@ parse_record: if (nr_slots) { xlate_len = utf8 - ?utf8_wcstombs(bufname, unicode, PAGE_SIZE) - :uni16_to_x8(bufname, unicode, PAGE_SIZE, uni_xlate, nls_io); + ?utf8_wcstombs(bufname, unicode, PATH_MAX) + :uni16_to_x8(bufname, unicode, PATH_MAX, uni_xlate, nls_io); if (xlate_len != name_len) continue; if ((!anycase && !memcmp(name, bufname, xlate_len)) || @@ -427,9 +427,9 @@ Found: err = 0; EODir: if (bufname) - free_page((unsigned long)bufname); + __putname(bufname); if (unicode) - free_page((unsigned long)unicode); + __putname(unicode); return err; } @@ -619,7 +619,7 @@ parse_record: /* convert the unicode long name. 261 is maximum size * of unicode buffer. (13 * slots + nul) */ void *longname = unicode + 261; - int buf_size = PAGE_SIZE - (261 * sizeof(unicode[0])); + int buf_size = PATH_MAX - (261 * sizeof(unicode[0])); int long_len = utf8 ? utf8_wcstombs(longname, unicode, buf_size) : uni16_to_x8(longname, unicode, buf_size, uni_xlate, nls_io); @@ -652,7 +652,7 @@ EODir: FillFailed: brelse(bh); if (unicode) - free_page((unsigned long)unicode); + __putname(unicode); out: unlock_kernel(); return ret; diff --git a/fs/vfat/namei.c b/fs/vfat/namei.c index ab4f3da770f0..5b66162d0747 100644 --- a/fs/vfat/namei.c +++ b/fs/vfat/namei.c @@ -472,7 +472,7 @@ xlate_to_uni(const unsigned char *name, int len, unsigned char *outname, if (utf8) { int name_len = strlen(name); - *outlen = utf8_mbstowcs((wchar_t *)outname, name, PAGE_SIZE); + *outlen = utf8_mbstowcs((wchar_t *)outname, name, PATH_MAX); /* * We stripped '.'s before and set len appropriately, @@ -565,7 +565,6 @@ static int vfat_build_slots(struct inode *dir, const unsigned char *name, struct fat_mount_options *opts = &sbi->options; struct msdos_dir_slot *ps; struct msdos_dir_entry *de; - unsigned long page; unsigned char cksum, lcase; unsigned char msdos_name[MSDOS_NAME]; wchar_t *uname; @@ -575,11 +574,10 @@ static int vfat_build_slots(struct inode *dir, const unsigned char *name, *nr_slots = 0; - page = __get_free_page(GFP_KERNEL); - if (!page) + uname = __getname(); + if (!uname) return -ENOMEM; - uname = (wchar_t *)page; err = xlate_to_uni(name, len, (unsigned char *)uname, &ulen, &usize, opts->unicode_xlate, opts->utf8, sbi->nls_io); if (err) @@ -631,7 +629,7 @@ shortname: de->starthi = cpu_to_le16(cluster >> 16); de->size = 0; out_free: - free_page(page); + __putname(uname); return err; } -- cgit From 73f20e58b1d586e9f6d3ddc3aad872829aca7743 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 28 Apr 2008 02:16:30 -0700 Subject: FAT_VALID_MEDIA(): remove pointless test The on-disk media specification field in FAT is only 8-bits, so testing for <=0xff is pointless, and can generate a "comparison is always true due to limited range of data type" warning. While we're there, convert FAT_VALID_MEDIA() into a C function - the present implementation is buggy: it generates either one or two references to its argument. Cc: Frank Seidel Acked-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fat/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 1604a8a5c01f..bf918929e793 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -1215,7 +1215,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, */ media = b->media; - if (!FAT_VALID_MEDIA(media)) { + if (!fat_valid_media(media)) { if (!silent) printk(KERN_ERR "FAT: invalid media value (0x%02x)\n", media); -- cgit From 0607fd02587a6b4b086dc746d63123c1f284db68 Mon Sep 17 00:00:00 2001 From: Frank Seidel Date: Mon, 28 Apr 2008 02:16:31 -0700 Subject: fat: detect media without partition table correctly I received a complaint that some FAT formated medias (e.g. sd memory cards) trigger a "unknown partition table" message even though there is no partition table and they work correctly, while in general (when e.g. formated with mkdosfs or even Windows Vista) this message is not shown. Currently this seems only to happen when the medias get formatted with Windows XP (and possibly Win 2000). Then the boot indicator byte contains garbage (part of text message) and so do the other parts checked by msdos_paritition which then later triggers this message. References: novell bug #364365 Most fat formatted media without partition table contains zeros in the boot indication and the other tested bytes and so falls through the checks in msdos_partition, leading it to return with 1 (all is fine). But some (e.g. WinXP formatted) fat fomated medias don't use boot_ind and so the check fails and causes a "unkown partition table" warning eventhough there is none and everything would be fine. This additional check directly verifies if there is a fat formatted medium without a partition table. Signed-off-by: Frank Seidel Cc: Andreas Dilger Acked-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/partitions/msdos.c | 20 +++++++++++++++++--- 1 file changed, 17 insertions(+), 3 deletions(-) (limited to 'fs') diff --git a/fs/partitions/msdos.c b/fs/partitions/msdos.c index 5567ec0d03a3..796511886f28 100644 --- a/fs/partitions/msdos.c +++ b/fs/partitions/msdos.c @@ -18,7 +18,7 @@ * * Re-organised Feb 1998 Russell King */ - +#include #include "check.h" #include "msdos.h" @@ -419,6 +419,7 @@ int msdos_partition(struct parsed_partitions *state, struct block_device *bdev) Sector sect; unsigned char *data; struct partition *p; + struct fat_boot_sector *fb; int slot; data = read_dev_sector(bdev, 0, §); @@ -444,8 +445,21 @@ int msdos_partition(struct parsed_partitions *state, struct block_device *bdev) p = (struct partition *) (data + 0x1be); for (slot = 1; slot <= 4; slot++, p++) { if (p->boot_ind != 0 && p->boot_ind != 0x80) { - put_dev_sector(sect); - return 0; + /* + * Even without a valid boot inidicator value + * its still possible this is valid FAT filesystem + * without a partition table. + */ + fb = (struct fat_boot_sector *) data; + if (slot == 1 && fb->reserved && fb->fats + && fat_valid_media(fb->media)) { + printk("\n"); + put_dev_sector(sect); + return 1; + } else { + put_dev_sector(sect); + return 0; + } } } -- cgit From 9f354858b8ea29e969b82dd96caea376157d76ca Mon Sep 17 00:00:00 2001 From: Olof Johansson Date: Mon, 28 Apr 2008 02:16:32 -0700 Subject: fatfs: fix build warning with 64k PAGE_SIZE Annoying gcc warning: fs/fat/inode.c: In function 'fat_fill_super': fs/fat/inode.c:1222: warning: comparison is always false due to limited range of data type Change it to compare with 4K instead of PAGE_CACHE_SIZE, as suggested by OGAWA-san. [FAT spec says: logical_sector_size should be 512, 1024, 2048 4096] So, at least for now, we limit it to 4096. Signed-off-by: Olof Johansson Signed-off-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fat/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs') diff --git a/fs/fat/inode.c b/fs/fat/inode.c index bf918929e793..5f522a55b596 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -1226,7 +1226,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, le16_to_cpu(get_unaligned((__le16 *)&b->sector_size)); if (!is_power_of_2(logical_sector_size) || (logical_sector_size < 512) - || (PAGE_CACHE_SIZE < logical_sector_size)) { + || (logical_sector_size > 4096)) { if (!silent) printk(KERN_ERR "FAT: bogus logical sector size %u\n", logical_sector_size); -- cgit