From cfe79c00a2f4f687eed8b7534d1d3d3d35540c29 Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Wed, 6 Jan 2010 17:23:23 +0000 Subject: NOMMU: Avoiding duplicate icache flushes of shared maps When working with FDPIC, there are many shared mappings of read-only code regions between applications (the C library, applet packages like busybox, etc.), but the current do_mmap_pgoff() function will issue an icache flush whenever a VMA is added to an MM instead of only doing it when the map is initially created. The flush can instead be done when a region is first mmapped PROT_EXEC. Note that we may not rely on the first mapping of a region being executable - it's possible for it to be PROT_READ only, so we have to remember whether we've flushed the region or not, and then flush the entire region when a bit of it is made executable. However, this also affects the brk area. That will no longer be executable. We can mprotect() it to PROT_EXEC on MPU-mode kernels, but for NOMMU mode kernels, when it increases the brk allocation, making sys_brk() flush the extra from the icache should suffice. The brk area probably isn't used by NOMMU programs since the brk area can only use up the leavings from the stack allocation, where the stack allocation is larger than requested. Signed-off-by: David Howells Signed-off-by: Mike Frysinger Signed-off-by: Linus Torvalds --- mm/nommu.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) (limited to 'mm') diff --git a/mm/nommu.c b/mm/nommu.c index 6f9248f89bd..a8d17521624 100644 --- a/mm/nommu.c +++ b/mm/nommu.c @@ -432,6 +432,7 @@ SYSCALL_DEFINE1(brk, unsigned long, brk) /* * Ok, looks good - let it rip. */ + flush_icache_range(mm->brk, brk); return mm->brk = brk; } @@ -1353,10 +1354,14 @@ unsigned long do_mmap_pgoff(struct file *file, share: add_vma_to_mm(current->mm, vma); - up_write(&nommu_region_sem); + /* we flush the region from the icache only when the first executable + * mapping of it is made */ + if (vma->vm_flags & VM_EXEC && !region->vm_icache_flushed) { + flush_icache_range(region->vm_start, region->vm_end); + region->vm_icache_flushed = true; + } - if (prot & PROT_EXEC) - flush_icache_range(result, result + len); + up_write(&nommu_region_sem); kleave(" = %lx", result); return result; -- cgit v1.2.3 From 7959722b951cffcd61a0a35229d007deeed8c2dd Mon Sep 17 00:00:00 2001 From: Jie Zhang Date: Wed, 6 Jan 2010 17:23:28 +0000 Subject: NOMMU: Use copy_*_user_page() in access_process_vm() The MMU code uses the copy_*_user_page() variants in access_process_vm() rather than copy_*_user() as the former includes an icache flush. This is important when doing things like setting software breakpoints with gdb. So switch the NOMMU code over to do the same. This patch makes the reasonable assumption that copy_from_user_page() won't fail - which is probably fine, as we've checked the VMA from which we're copying is usable, and the copy is not allowed to cross VMAs. The one case where it might go wrong is if the VMA is a device rather than RAM, and that device returns an error which - in which case rubbish will be returned rather than EIO. Signed-off-by: Jie Zhang Signed-off-by: Mike Frysinger Signed-off-by: David Howells Acked-by: David McCullough Acked-by: Paul Mundt Acked-by: Greg Ungerer Signed-off-by: Linus Torvalds --- mm/nommu.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'mm') diff --git a/mm/nommu.c b/mm/nommu.c index a8d17521624..17773862619 100644 --- a/mm/nommu.c +++ b/mm/nommu.c @@ -1921,9 +1921,11 @@ int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, in /* only read or write mappings where it is permitted */ if (write && vma->vm_flags & VM_MAYWRITE) - len -= copy_to_user((void *) addr, buf, len); + copy_to_user_page(vma, NULL, addr, + (void *) addr, buf, len); else if (!write && vma->vm_flags & VM_MAYREAD) - len -= copy_from_user(buf, (void *) addr, len); + copy_from_user_page(vma, NULL, addr, + buf, (void *) addr, len); else len = 0; } else { -- cgit v1.2.3 From 6144a85a0e018c19bc4b24f7eb6c1f3f7431813d Mon Sep 17 00:00:00 2001 From: Jason Wessel Date: Thu, 7 Jan 2010 11:58:36 -0600 Subject: maccess,probe_kernel: Allow arch specific override probe_kernel_(read|write) Some archs such as blackfin, would like to have an arch specific probe_kernel_read() and probe_kernel_write() implementation which can fall back to the generic implementation if no special operations are needed. CC: Thomas Gleixner CC: Ingo Molnar Signed-off-by: Jason Wessel Signed-off-by: Mike Frysinger --- mm/maccess.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'mm') diff --git a/mm/maccess.c b/mm/maccess.c index 9073695ff25..4e348dbaecd 100644 --- a/mm/maccess.c +++ b/mm/maccess.c @@ -14,7 +14,11 @@ * Safely read from address @src to the buffer at @dst. If a kernel fault * happens, handle that and return -EFAULT. */ -long probe_kernel_read(void *dst, void *src, size_t size) + +long __weak probe_kernel_read(void *dst, void *src, size_t size) + __attribute__((alias("__probe_kernel_read"))); + +long __probe_kernel_read(void *dst, void *src, size_t size) { long ret; mm_segment_t old_fs = get_fs(); @@ -39,7 +43,10 @@ EXPORT_SYMBOL_GPL(probe_kernel_read); * Safely write to address @dst from the buffer at @src. If a kernel fault * happens, handle that and return -EFAULT. */ -long notrace __weak probe_kernel_write(void *dst, void *src, size_t size) +long __weak probe_kernel_write(void *dst, void *src, size_t size) + __attribute__((alias("__probe_kernel_write"))); + +long __probe_kernel_write(void *dst, void *src, size_t size) { long ret; mm_segment_t old_fs = get_fs(); -- cgit v1.2.3 From 129182e5626972ac0df85d43a36dd46ad61c64e1 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Fri, 8 Jan 2010 14:42:39 -0800 Subject: percpu: avoid calling __pcpu_ptr_to_addr(NULL) __pcpu_ptr_to_addr() can be overridden by the architecture and might not behave well if passed a NULL pointer. So avoid calling it until we have verified that its arg is not NULL. Cc: Rusty Russell Cc: Kamalesh Babulal Acked-by: Tejun Heo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/percpu.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'mm') diff --git a/mm/percpu.c b/mm/percpu.c index 442010cc91c..083e7c91e5f 100644 --- a/mm/percpu.c +++ b/mm/percpu.c @@ -1271,7 +1271,7 @@ static void pcpu_reclaim(struct work_struct *work) */ void free_percpu(void *ptr) { - void *addr = __pcpu_ptr_to_addr(ptr); + void *addr; struct pcpu_chunk *chunk; unsigned long flags; int off; @@ -1279,6 +1279,8 @@ void free_percpu(void *ptr) if (!ptr) return; + addr = __pcpu_ptr_to_addr(ptr); + spin_lock_irqsave(&pcpu_lock, flags); chunk = pcpu_chunk_addr_search(addr); -- cgit v1.2.3 From 74dbdd239bb1348ad86d28b18574d9c1f28b62ca Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Fri, 8 Jan 2010 14:43:05 -0800 Subject: mm: hugetlb: fix clear_huge_page() sz is in bytes, MAX_ORDER_NR_PAGES is in pages. Signed-off-by: Andrea Arcangeli Acked-by: David Gibson Cc: Mel Gorman Cc: David Rientjes Cc: Lee Schermerhorn Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm') diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 65f38c21820..e91b81b6367 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -402,7 +402,7 @@ static void clear_huge_page(struct page *page, { int i; - if (unlikely(sz > MAX_ORDER_NR_PAGES)) { + if (unlikely(sz/PAGE_SIZE > MAX_ORDER_NR_PAGES)) { clear_gigantic_page(page, addr, sz); return; } -- cgit v1.2.3 From cedabed49b39b4319bccc059a63344b6232b619c Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Wed, 13 Jan 2010 21:14:09 +0900 Subject: vfs: Fix vmtruncate() regression If __block_prepare_write() was failed in block_write_begin(), the allocated blocks can be outside of ->i_size. But new truncate_pagecache() in vmtuncate() does nothing if new < old. It means the above usage is not working anymore. So, this patch fixes it by removing "new < old" check. It would need more cleanup/change. But, now -rc and truncate working is in progress, so, this tried to fix it minimum change. Acked-by: Nick Piggin Signed-off-by: OGAWA Hirofumi Signed-off-by: Linus Torvalds --- mm/truncate.c | 30 ++++++++++++++---------------- 1 file changed, 14 insertions(+), 16 deletions(-) (limited to 'mm') diff --git a/mm/truncate.c b/mm/truncate.c index 342deee2268..e87e3724482 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -522,22 +522,20 @@ EXPORT_SYMBOL_GPL(invalidate_inode_pages2); */ void truncate_pagecache(struct inode *inode, loff_t old, loff_t new) { - if (new < old) { - struct address_space *mapping = inode->i_mapping; - - /* - * unmap_mapping_range is called twice, first simply for - * efficiency so that truncate_inode_pages does fewer - * single-page unmaps. However after this first call, and - * before truncate_inode_pages finishes, it is possible for - * private pages to be COWed, which remain after - * truncate_inode_pages finishes, hence the second - * unmap_mapping_range call must be made for correctness. - */ - unmap_mapping_range(mapping, new + PAGE_SIZE - 1, 0, 1); - truncate_inode_pages(mapping, new); - unmap_mapping_range(mapping, new + PAGE_SIZE - 1, 0, 1); - } + struct address_space *mapping = inode->i_mapping; + + /* + * unmap_mapping_range is called twice, first simply for + * efficiency so that truncate_inode_pages does fewer + * single-page unmaps. However after this first call, and + * before truncate_inode_pages finishes, it is possible for + * private pages to be COWed, which remain after + * truncate_inode_pages finishes, hence the second + * unmap_mapping_range call must be made for correctness. + */ + unmap_mapping_range(mapping, new + PAGE_SIZE - 1, 0, 1); + truncate_inode_pages(mapping, new); + unmap_mapping_range(mapping, new + PAGE_SIZE - 1, 0, 1); } EXPORT_SYMBOL(truncate_pagecache); -- cgit v1.2.3 From d2dbe08ddceb4ba2b274abb84326d7e69d454e5c Mon Sep 17 00:00:00 2001 From: Kazuhisa Ichikawa Date: Fri, 15 Jan 2010 17:01:20 -0800 Subject: mm/page_alloc: fix the range check for backward merging The current check for 'backward merging' within add_active_range() does not seem correct. start_pfn must be compared against early_node_map[i].start_pfn (and NOT against .end_pfn) to find out whether the new region is backward-mergeable with the existing range. Signed-off-by: Kazuhisa Ichikawa Acked-by: David Rientjes Cc: KOSAKI Motohiro Cc: Mel Gorman Cc: Christoph Lameter Cc: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 4e9f5cc5fb5..6ea4966a633 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3998,7 +3998,7 @@ void __init add_active_range(unsigned int nid, unsigned long start_pfn, } /* Merge backward if suitable */ - if (start_pfn < early_node_map[i].end_pfn && + if (start_pfn < early_node_map[i].start_pfn && end_pfn >= early_node_map[i].start_pfn) { early_node_map[i].start_pfn = start_pfn; return; -- cgit v1.2.3 From de3fab39348dff18c69a0cd04efee9c276a02f51 Mon Sep 17 00:00:00 2001 From: KOSAKI Motohiro Date: Fri, 15 Jan 2010 17:01:25 -0800 Subject: vmscan: kswapd: don't retry balance_pgdat() if all zones are unreclaimable Commit f50de2d3 (vmscan: have kswapd sleep for a short interval and double check it should be asleep) can cause kswapd to enter an infinite loop if running on a single-CPU system. If all zones are unreclaimble, sleeping_prematurely return 1 and kswapd will call balance_pgdat() again. but it's totally meaningless, balance_pgdat() doesn't anything against unreclaimable zone! Signed-off-by: KOSAKI Motohiro Cc: Mel Gorman Reported-by: Will Newton Reviewed-by: Minchan Kim Reviewed-by: Rik van Riel Tested-by: Will Newton Reviewed-by: Wu Fengguang Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmscan.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'mm') diff --git a/mm/vmscan.c b/mm/vmscan.c index 885207a6b6b..c26986c85ce 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1922,6 +1922,9 @@ static int sleeping_prematurely(pg_data_t *pgdat, int order, long remaining) if (!populated_zone(zone)) continue; + if (zone_is_all_unreclaimable(zone)) + continue; + if (!zone_watermark_ok(zone, order, high_wmark_pages(zone), 0, 0)) return 1; -- cgit v1.2.3 From fce66477578d081f19aef5ea218664ff7758c33a Mon Sep 17 00:00:00 2001 From: Daisuke Nishimura Date: Fri, 15 Jan 2010 17:01:30 -0800 Subject: memcg: ensure list is empty at rmdir Current mem_cgroup_force_empty() only ensures mem->res.usage == 0 on success. But this doesn't guarantee memcg's LRU is really empty, because there are some cases in which !PageCgrupUsed pages exist on memcg's LRU. For example: - Pages can be uncharged by its owner process while they are on LRU. - race between mem_cgroup_add_lru_list() and __mem_cgroup_uncharge_common(). So there can be a case in which the usage is zero but some of the LRUs are not empty. OTOH, mem_cgroup_del_lru_list(), which can be called asynchronously with rmdir, accesses the mem_cgroup, so this access can cause a problem if it races with rmdir because the mem_cgroup might have been freed by rmdir. Actually, I saw a bug which seems to be caused by this race. [1530745.949906] BUG: unable to handle kernel NULL pointer dereference at 0000000000000230 [1530745.950651] IP: [] mem_cgroup_del_lru_list+0x30/0x80 [1530745.950651] PGD 3863de067 PUD 3862c7067 PMD 0 [1530745.950651] Oops: 0002 [#1] SMP [1530745.950651] last sysfs file: /sys/devices/system/cpu/cpu7/cache/index1/shared_cpu_map [1530745.950651] CPU 3 [1530745.950651] Modules linked in: configs ipt_REJECT xt_tcpudp iptable_filter ip_tables x_tables bridge stp nfsd nfs_acl auth_rpcgss exportfs autofs4 hidp rfcomm l2cap crc16 bluetooth lockd sunrpc ib_iser rdma_cm ib_cm iw_cm ib_sa ib_mad ib_core ib_addr iscsi_tcp bnx2i cnic uio ipv6 cxgb3i cxgb3 mdio libiscsi_tcp libiscsi scsi_transport_iscsi dm_mirror dm_multipath scsi_dh video output sbs sbshc battery ac lp kvm_intel kvm sg ide_cd_mod cdrom serio_raw tpm_tis tpm tpm_bios acpi_memhotplug button parport_pc parport rtc_cmos rtc_core rtc_lib e1000 i2c_i801 i2c_core pcspkr dm_region_hash dm_log dm_mod ata_piix libata shpchp megaraid_mbox sd_mod scsi_mod megaraid_mm ext3 jbd uhci_hcd ohci_hcd ehci_hcd [last unloaded: freq_table] [1530745.950651] Pid: 19653, comm: shmem_test_02 Tainted: G M 2.6.32-mm1-00701-g2b04386 #3 Express5800/140Rd-4 [N8100-1065] [1530745.950651] RIP: 0010:[] [] mem_cgroup_del_lru_list+0x30/0x80 [1530745.950651] RSP: 0018:ffff8803863ddcb8 EFLAGS: 00010002 [1530745.950651] RAX: 00000000000001e0 RBX: ffff8803abc02238 RCX: 00000000000001e0 [1530745.950651] RDX: 0000000000000000 RSI: ffff88038611a000 RDI: ffff8803abc02238 [1530745.950651] RBP: ffff8803863ddcc8 R08: 0000000000000002 R09: ffff8803a04c8643 [1530745.950651] R10: 0000000000000000 R11: ffffffff810c7333 R12: 0000000000000000 [1530745.950651] R13: ffff880000017f00 R14: 0000000000000092 R15: ffff8800179d0310 [1530745.950651] FS: 0000000000000000(0000) GS:ffff880017800000(0000) knlGS:0000000000000000 [1530745.950651] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b [1530745.950651] CR2: 0000000000000230 CR3: 0000000379d87000 CR4: 00000000000006e0 [1530745.950651] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [1530745.950651] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 [1530745.950651] Process shmem_test_02 (pid: 19653, threadinfo ffff8803863dc000, task ffff88038612a8a0) [1530745.950651] Stack: [1530745.950651] ffffea00040c2fe8 0000000000000000 ffff8803863ddd98 ffffffff810c739a [1530745.950651] <0> 00000000863ddd18 000000000000000c 0000000000000000 0000000000000000 [1530745.950651] <0> 0000000000000002 0000000000000000 ffff8803863ddd68 0000000000000046 [1530745.950651] Call Trace: [1530745.950651] [] release_pages+0x142/0x1e7 [1530745.950651] [] ? pagevec_move_tail+0x6e/0x112 [1530745.950651] [] pagevec_move_tail+0xfd/0x112 [1530745.950651] [] lru_add_drain+0x76/0x94 [1530745.950651] [] exit_mmap+0x6e/0x145 [1530745.950651] [] mmput+0x5e/0xcf [1530745.950651] [] exit_mm+0x11c/0x129 [1530745.950651] [] ? audit_free+0x196/0x1c9 [1530745.950651] [] do_exit+0x1f5/0x6b7 [1530745.950651] [] ? up_read+0x2b/0x2f [1530745.950651] [] ? lockdep_sys_exit_thunk+0x35/0x67 [1530745.950651] [] do_group_exit+0x83/0xb0 [1530745.950651] [] sys_exit_group+0x17/0x1b [1530745.950651] [] system_call_fastpath+0x16/0x1b [1530745.950651] Code: 54 53 0f 1f 44 00 00 83 3d cc 29 7c 00 00 41 89 f4 75 63 eb 4e 48 83 7b 08 00 75 04 0f 0b eb fe 48 89 df e8 18 f3 ff ff 44 89 e2 <48> ff 4c d0 50 48 8b 05 2b 2d 7c 00 48 39 43 08 74 39 48 8b 4b [1530745.950651] RIP [] mem_cgroup_del_lru_list+0x30/0x80 [1530745.950651] RSP [1530745.950651] CR2: 0000000000000230 [1530745.950651] ---[ end trace c3419c1bb8acc34f ]--- [1530745.950651] Fixing recursive fault but reboot is needed! The problem here is pages on LRU may contain pointer to stale memcg. To make res->usage to be 0, all pages on memcg must be uncharged or moved to another(parent) memcg. Moved page_cgroup have already removed from original LRU, but uncharged page_cgroup contains pointer to memcg withou PCG_USED bit. (This asynchronous LRU work is for improving performance.) If PCG_USED bit is not set, page_cgroup will never be added to memcg's LRU. So, about pages not on LRU, they never access stale pointer. Then, what we have to take care of is page_cgroup _on_ LRU list. This patch fixes this problem by making mem_cgroup_force_empty() visit all LRUs before exiting its loop and guarantee there are no pages on its LRU. Signed-off-by: Daisuke Nishimura Acked-by: KAMEZAWA Hiroyuki Cc: Balbir Singh Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memcontrol.c | 11 ++++------- 1 file changed, 4 insertions(+), 7 deletions(-) (limited to 'mm') diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 488b644e0e8..954032b80be 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -2586,7 +2586,7 @@ static int mem_cgroup_force_empty(struct mem_cgroup *mem, bool free_all) if (free_all) goto try_to_free; move_account: - while (mem->res.usage > 0) { + do { ret = -EBUSY; if (cgroup_task_count(cgrp) || !list_empty(&cgrp->children)) goto out; @@ -2614,8 +2614,8 @@ move_account: if (ret == -ENOMEM) goto try_to_free; cond_resched(); - } - ret = 0; + /* "ret" should also be checked to ensure all lists are empty. */ + } while (mem->res.usage > 0 || ret); out: css_put(&mem->css); return ret; @@ -2648,10 +2648,7 @@ try_to_free: } lru_add_drain(); /* try move_account...there may be some *locked* pages. */ - if (mem->res.usage) - goto move_account; - ret = 0; - goto out; + goto move_account; } int mem_cgroup_force_empty_write(struct cgroup *cont, unsigned int event) -- cgit v1.2.3 From 1e2ae599d37e60958c03ca5e46b1f657619a30cd Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 15 Jan 2010 17:01:33 -0800 Subject: nommu: struct vm_region's vm_usage count need not be atomic The vm_usage count field in struct vm_region does not need to be atomic as it's only even modified whilst nommu_region_sem is write locked. Signed-off-by: David Howells Acked-by: Al Viro Cc: Greg Ungerer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/nommu.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'mm') diff --git a/mm/nommu.c b/mm/nommu.c index 17773862619..5e39294f8ea 100644 --- a/mm/nommu.c +++ b/mm/nommu.c @@ -552,11 +552,11 @@ static void free_page_series(unsigned long from, unsigned long to) static void __put_nommu_region(struct vm_region *region) __releases(nommu_region_sem) { - kenter("%p{%d}", region, atomic_read(®ion->vm_usage)); + kenter("%p{%d}", region, region->vm_usage); BUG_ON(!nommu_region_tree.rb_node); - if (atomic_dec_and_test(®ion->vm_usage)) { + if (--region->vm_usage == 0) { if (region->vm_top > region->vm_start) delete_nommu_region(region); up_write(&nommu_region_sem); @@ -1205,7 +1205,7 @@ unsigned long do_mmap_pgoff(struct file *file, if (!vma) goto error_getting_vma; - atomic_set(®ion->vm_usage, 1); + region->vm_usage = 1; region->vm_flags = vm_flags; region->vm_pgoff = pgoff; @@ -1272,7 +1272,7 @@ unsigned long do_mmap_pgoff(struct file *file, } /* we've found a region we can share */ - atomic_inc(&pregion->vm_usage); + pregion->vm_usage++; vma->vm_region = pregion; start = pregion->vm_start; start += (pgoff - pregion->vm_pgoff) << PAGE_SHIFT; @@ -1289,7 +1289,7 @@ unsigned long do_mmap_pgoff(struct file *file, vma->vm_region = NULL; vma->vm_start = 0; vma->vm_end = 0; - atomic_dec(&pregion->vm_usage); + pregion->vm_usage--; pregion = NULL; goto error_just_free; } @@ -1444,7 +1444,7 @@ int split_vma(struct mm_struct *mm, struct vm_area_struct *vma, /* we're only permitted to split anonymous regions that have a single * owner */ if (vma->vm_file || - atomic_read(&vma->vm_region->vm_usage) != 1) + vma->vm_region->vm_usage != 1) return -ENOMEM; if (mm->map_count >= sysctl_max_map_count) @@ -1518,7 +1518,7 @@ static int shrink_vma(struct mm_struct *mm, /* cut the backing region down to size */ region = vma->vm_region; - BUG_ON(atomic_read(®ion->vm_usage) != 1); + BUG_ON(region->vm_usage != 1); down_write(&nommu_region_sem); delete_nommu_region(region); -- cgit v1.2.3 From 779c10232ceb11c1b259232c4845cfb2850287b7 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 15 Jan 2010 17:01:34 -0800 Subject: nommu: remove a superfluous check of vm_region::vm_usage In split_vma(), there's no need to check if the VMA being split has a region that's in use by more than one VMA because: (1) The preceding test prohibits splitting of non-anonymous VMAs and regions (eg: file or chardev backed VMAs). (2) Anonymous regions can't be mapped multiple times because there's no handle by which to refer to the already existing region. (3) If a VMA has previously been split, then the region backing it has also been split into two regions, each of usage 1. Signed-off-by: David Howells Acked-by: Al Viro Cc: Greg Ungerer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/nommu.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'mm') diff --git a/mm/nommu.c b/mm/nommu.c index 5e39294f8ea..d6dd656264a 100644 --- a/mm/nommu.c +++ b/mm/nommu.c @@ -1441,10 +1441,9 @@ int split_vma(struct mm_struct *mm, struct vm_area_struct *vma, kenter(""); - /* we're only permitted to split anonymous regions that have a single - * owner */ - if (vma->vm_file || - vma->vm_region->vm_usage != 1) + /* we're only permitted to split anonymous regions (these should have + * only a single usage on the region) */ + if (vma->vm_file) return -ENOMEM; if (mm->map_count >= sysctl_max_map_count) -- cgit v1.2.3 From efc1a3b16930c41d64ffefde16b87d82f603a8a0 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 15 Jan 2010 17:01:35 -0800 Subject: nommu: don't need get_unmapped_area() for NOMMU get_unmapped_area() is unnecessary for NOMMU as no-one calls it. Signed-off-by: David Howells Acked-by: Al Viro Cc: Greg Ungerer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/nommu.c | 21 --------------------- mm/util.c | 2 +- 2 files changed, 1 insertion(+), 22 deletions(-) (limited to 'mm') diff --git a/mm/nommu.c b/mm/nommu.c index d6dd656264a..32be0cf51ba 100644 --- a/mm/nommu.c +++ b/mm/nommu.c @@ -1760,27 +1760,6 @@ void unmap_mapping_range(struct address_space *mapping, } EXPORT_SYMBOL(unmap_mapping_range); -/* - * ask for an unmapped area at which to create a mapping on a file - */ -unsigned long get_unmapped_area(struct file *file, unsigned long addr, - unsigned long len, unsigned long pgoff, - unsigned long flags) -{ - unsigned long (*get_area)(struct file *, unsigned long, unsigned long, - unsigned long, unsigned long); - - get_area = current->mm->get_unmapped_area; - if (file && file->f_op && file->f_op->get_unmapped_area) - get_area = file->f_op->get_unmapped_area; - - if (!get_area) - return -ENOSYS; - - return get_area(file, addr, len, pgoff, flags); -} -EXPORT_SYMBOL(get_unmapped_area); - /* * Check that a process has enough memory to allocate a new virtual * mapping. 0 means there is enough memory for the allocation to diff --git a/mm/util.c b/mm/util.c index 7c35ad95f92..834db7be240 100644 --- a/mm/util.c +++ b/mm/util.c @@ -220,7 +220,7 @@ char *strndup_user(const char __user *s, long n) } EXPORT_SYMBOL(strndup_user); -#ifndef HAVE_ARCH_PICK_MMAP_LAYOUT +#if defined(CONFIG_MMU) && !defined(HAVE_ARCH_PICK_MMAP_LAYOUT) void arch_pick_mmap_layout(struct mm_struct *mm) { mm->mmap_base = TASK_UNMAPPED_BASE; -- cgit v1.2.3 From 7e6608724c640924aad1d556d17df33ebaa6124d Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 15 Jan 2010 17:01:39 -0800 Subject: nommu: fix shared mmap after truncate shrinkage problems Fix a problem in NOMMU mmap with ramfs whereby a shared mmap can happen over the end of a truncation. The problem is that ramfs_nommu_check_mappings() checks that the reduced file size against the VMA tree, but not the vm_region tree. The following sequence of events can cause the problem: fd = open("/tmp/x", O_RDWR|O_TRUNC|O_CREAT, 0600); ftruncate(fd, 32 * 1024); a = mmap(NULL, 32 * 1024, PROT_READ|PROT_WRITE, MAP_SHARED, fd, 0); b = mmap(NULL, 16 * 1024, PROT_READ|PROT_WRITE, MAP_SHARED, fd, 0); munmap(a, 32 * 1024); ftruncate(fd, 16 * 1024); c = mmap(NULL, 32 * 1024, PROT_READ|PROT_WRITE, MAP_SHARED, fd, 0); Mapping 'a' creates a vm_region covering 32KB of the file. Mapping 'b' sees that the vm_region from 'a' is covering the region it wants and so shares it, pinning it in memory. Mapping 'a' then goes away and the file is truncated to the end of VMA 'b'. However, the region allocated by 'a' is still in effect, and has _not_ been reduced. Mapping 'c' is then created, and because there's a vm_region covering the desired region, get_unmapped_area() is _not_ called to repeat the check, and the mapping is granted, even though the pages from the latter half of the mapping have been discarded. However: d = mmap(NULL, 16 * 1024, PROT_READ|PROT_WRITE, MAP_SHARED, fd, 0); Mapping 'd' should work, and should end up sharing the region allocated by 'a'. To deal with this, we shrink the vm_region struct during the truncation, lest do_mmap_pgoff() take it as licence to share the full region automatically without calling the get_unmapped_area() file op again. Signed-off-by: David Howells Acked-by: Al Viro Cc: Greg Ungerer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/nommu.c | 62 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 62 insertions(+) (limited to 'mm') diff --git a/mm/nommu.c b/mm/nommu.c index 32be0cf51ba..48a2ecfaf05 100644 --- a/mm/nommu.c +++ b/mm/nommu.c @@ -1914,3 +1914,65 @@ int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, in mmput(mm); return len; } + +/** + * nommu_shrink_inode_mappings - Shrink the shared mappings on an inode + * @inode: The inode to check + * @size: The current filesize of the inode + * @newsize: The proposed filesize of the inode + * + * Check the shared mappings on an inode on behalf of a shrinking truncate to + * make sure that that any outstanding VMAs aren't broken and then shrink the + * vm_regions that extend that beyond so that do_mmap_pgoff() doesn't + * automatically grant mappings that are too large. + */ +int nommu_shrink_inode_mappings(struct inode *inode, size_t size, + size_t newsize) +{ + struct vm_area_struct *vma; + struct prio_tree_iter iter; + struct vm_region *region; + pgoff_t low, high; + size_t r_size, r_top; + + low = newsize >> PAGE_SHIFT; + high = (size + PAGE_SIZE - 1) >> PAGE_SHIFT; + + down_write(&nommu_region_sem); + + /* search for VMAs that fall within the dead zone */ + vma_prio_tree_foreach(vma, &iter, &inode->i_mapping->i_mmap, + low, high) { + /* found one - only interested if it's shared out of the page + * cache */ + if (vma->vm_flags & VM_SHARED) { + up_write(&nommu_region_sem); + return -ETXTBSY; /* not quite true, but near enough */ + } + } + + /* reduce any regions that overlap the dead zone - if in existence, + * these will be pointed to by VMAs that don't overlap the dead zone + * + * we don't check for any regions that start beyond the EOF as there + * shouldn't be any + */ + vma_prio_tree_foreach(vma, &iter, &inode->i_mapping->i_mmap, + 0, ULONG_MAX) { + if (!(vma->vm_flags & VM_SHARED)) + continue; + + region = vma->vm_region; + r_size = region->vm_top - region->vm_start; + r_top = (region->vm_pgoff << PAGE_SHIFT) + r_size; + + if (r_top > newsize) { + region->vm_top -= r_top - newsize; + if (region->vm_end > region->vm_top) + region->vm_end = region->vm_top; + } + } + + up_write(&nommu_region_sem); + return 0; +} -- cgit v1.2.3 From 6ccf80eb15ccaca4d3f1ab5162b9ded5eecd9971 Mon Sep 17 00:00:00 2001 From: KOSAKI Motohiro Date: Fri, 15 Jan 2010 17:01:18 -0800 Subject: page allocator: update NR_FREE_PAGES only when necessary commit f2260e6b (page allocator: update NR_FREE_PAGES only as necessary) made one minor regression. if __rmqueue() was failed, NR_FREE_PAGES stat go wrong. this patch fixes it. Signed-off-by: KOSAKI Motohiro Cc: Mel Gorman Reviewed-by: Minchan Kim Reported-by: Huang Shijie Reviewed-by: Christoph Lameter Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 6ea4966a633..d2a8889b4c5 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1222,10 +1222,10 @@ again: } spin_lock_irqsave(&zone->lock, flags); page = __rmqueue(zone, order, migratetype); - __mod_zone_page_state(zone, NR_FREE_PAGES, -(1 << order)); spin_unlock(&zone->lock); if (!page) goto failed; + __mod_zone_page_state(zone, NR_FREE_PAGES, -(1 << order)); } __count_zone_vm_events(PGALLOC, zone, 1 << order); -- cgit v1.2.3 From 88f5004430babb836cfce886d5d54c82166f8ba4 Mon Sep 17 00:00:00 2001 From: Yongseok Koh Date: Tue, 19 Jan 2010 17:33:49 +0900 Subject: vmalloc: remove BUG_ON due to racy counting of VM_LAZY_FREE In free_unmap_area_noflush(), va->flags is marked as VM_LAZY_FREE first, and then vmap_lazy_nr is increased atomically. But, in __purge_vmap_area_lazy(), while traversing of vmap_are_list, nr is counted by checking VM_LAZY_FREE is set to va->flags. After counting the variable nr, kernel reads vmap_lazy_nr atomically and checks a BUG_ON condition whether nr is greater than vmap_lazy_nr to prevent vmap_lazy_nr from being negative. The problem is that, if interrupted right after marking VM_LAZY_FREE, increment of vmap_lazy_nr can be delayed. Consequently, BUG_ON condition can be met because nr is counted more than vmap_lazy_nr. It is highly probable when vmalloc/vfree are called frequently. This scenario have been verified by adding delay between marking VM_LAZY_FREE and increasing vmap_lazy_nr in free_unmap_area_noflush(). Even the vmap_lazy_nr is for checking high watermark, it never be the strict watermark. Although the BUG_ON condition is to prevent vmap_lazy_nr from being negative, vmap_lazy_nr is signed variable. So, it could go down to negative value temporarily. Consequently, removing the BUG_ON condition is proper. A possible BUG_ON message is like the below. kernel BUG at mm/vmalloc.c:517! invalid opcode: 0000 [#1] SMP EIP: 0060:[] EFLAGS: 00010297 CPU: 3 EIP is at __purge_vmap_area_lazy+0x144/0x150 EAX: ee8a8818 EBX: c08e77d4 ECX: e7c7ae40 EDX: c08e77ec ESI: 000081fe EDI: e7c7ae60 EBP: e7c7ae64 ESP: e7c7ae3c DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068 Call Trace: [] free_unmap_vmap_area_noflush+0x69/0x70 [] remove_vm_area+0x22/0x70 [] __vunmap+0x45/0xe0 [] vmalloc+0x2c/0x30 Code: 8d 59 e0 eb 04 66 90 89 cb 89 d0 e8 87 fe ff ff 8b 43 20 89 da 8d 48 e0 8d 43 20 3b 04 24 75 e7 fe 05 a8 a5 a3 c0 e9 78 ff ff ff <0f> 0b eb fe 90 8d b4 26 00 00 00 00 56 89 c6 b8 ac a5 a3 c0 31 EIP: [] __purge_vmap_area_lazy+0x144/0x150 SS:ESP 0068:e7c7ae3c [ See also http://marc.info/?l=linux-kernel&m=126335856228090&w=2 ] Signed-off-by: Yongseok Koh Reviewed-by: Minchan Kim Cc: Nick Piggin Cc: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmalloc.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'mm') diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 37e69295f25..d55d905463e 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -555,10 +555,8 @@ static void __purge_vmap_area_lazy(unsigned long *start, unsigned long *end, } rcu_read_unlock(); - if (nr) { - BUG_ON(nr > atomic_read(&vmap_lazy_nr)); + if (nr) atomic_sub(nr, &vmap_lazy_nr); - } if (nr || force_flush) flush_tlb_kernel_range(*start, *end); -- cgit v1.2.3 From 0531b2aac59c2296570ac52bfc032ef2ace7d5e1 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 27 Jan 2010 09:20:03 -0800 Subject: mm: add new 'read_cache_page_gfp()' helper function It's a simplified 'read_cache_page()' which takes a page allocation flag, so that different paths can control how aggressive the memory allocations are that populate a address space. In particular, the intel GPU object mapping code wants to be able to do a certain amount of own internal memory management by automatically shrinking the address space when memory starts getting tight. This allows it to dynamically use different memory allocation policies on a per-allocation basis, rather than depend on the (static) address space gfp policy. The actual new function is a one-liner, but re-organizing the helper functions to the point where you can do this with a single line of code is what most of the patch is all about. Tested-by: Chris Wilson Signed-off-by: Linus Torvalds --- mm/filemap.c | 100 ++++++++++++++++++++++++++++++++++++++++------------------- 1 file changed, 68 insertions(+), 32 deletions(-) (limited to 'mm') diff --git a/mm/filemap.c b/mm/filemap.c index 96ac6b0eb6c..e3736923220 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1634,14 +1634,15 @@ EXPORT_SYMBOL(generic_file_readonly_mmap); static struct page *__read_cache_page(struct address_space *mapping, pgoff_t index, int (*filler)(void *,struct page*), - void *data) + void *data, + gfp_t gfp) { struct page *page; int err; repeat: page = find_get_page(mapping, index); if (!page) { - page = page_cache_alloc_cold(mapping); + page = __page_cache_alloc(gfp | __GFP_COLD); if (!page) return ERR_PTR(-ENOMEM); err = add_to_page_cache_lru(page, mapping, index, GFP_KERNEL); @@ -1661,31 +1662,18 @@ repeat: return page; } -/** - * read_cache_page_async - read into page cache, fill it if needed - * @mapping: the page's address_space - * @index: the page index - * @filler: function to perform the read - * @data: destination for read data - * - * Same as read_cache_page, but don't wait for page to become unlocked - * after submitting it to the filler. - * - * Read into the page cache. If a page already exists, and PageUptodate() is - * not set, try to fill the page but don't wait for it to become unlocked. - * - * If the page does not get brought uptodate, return -EIO. - */ -struct page *read_cache_page_async(struct address_space *mapping, +static struct page *do_read_cache_page(struct address_space *mapping, pgoff_t index, int (*filler)(void *,struct page*), - void *data) + void *data, + gfp_t gfp) + { struct page *page; int err; retry: - page = __read_cache_page(mapping, index, filler, data); + page = __read_cache_page(mapping, index, filler, data, gfp); if (IS_ERR(page)) return page; if (PageUptodate(page)) @@ -1710,8 +1698,67 @@ out: mark_page_accessed(page); return page; } + +/** + * read_cache_page_async - read into page cache, fill it if needed + * @mapping: the page's address_space + * @index: the page index + * @filler: function to perform the read + * @data: destination for read data + * + * Same as read_cache_page, but don't wait for page to become unlocked + * after submitting it to the filler. + * + * Read into the page cache. If a page already exists, and PageUptodate() is + * not set, try to fill the page but don't wait for it to become unlocked. + * + * If the page does not get brought uptodate, return -EIO. + */ +struct page *read_cache_page_async(struct address_space *mapping, + pgoff_t index, + int (*filler)(void *,struct page*), + void *data) +{ + return do_read_cache_page(mapping, index, filler, data, mapping_gfp_mask(mapping)); +} EXPORT_SYMBOL(read_cache_page_async); +static struct page *wait_on_page_read(struct page *page) +{ + if (!IS_ERR(page)) { + wait_on_page_locked(page); + if (!PageUptodate(page)) { + page_cache_release(page); + page = ERR_PTR(-EIO); + } + } + return page; +} + +/** + * read_cache_page_gfp - read into page cache, using specified page allocation flags. + * @mapping: the page's address_space + * @index: the page index + * @gfp: the page allocator flags to use if allocating + * + * This is the same as "read_mapping_page(mapping, index, NULL)", but with + * any new page allocations done using the specified allocation flags. Note + * that the Radix tree operations will still use GFP_KERNEL, so you can't + * expect to do this atomically or anything like that - but you can pass in + * other page requirements. + * + * If the page does not get brought uptodate, return -EIO. + */ +struct page *read_cache_page_gfp(struct address_space *mapping, + pgoff_t index, + gfp_t gfp) +{ + filler_t *filler = (filler_t *)mapping->a_ops->readpage; + + return wait_on_page_read(do_read_cache_page(mapping, index, filler, NULL, gfp)); +} +EXPORT_SYMBOL(read_cache_page_gfp); + /** * read_cache_page - read into page cache, fill it if needed * @mapping: the page's address_space @@ -1729,18 +1776,7 @@ struct page *read_cache_page(struct address_space *mapping, int (*filler)(void *,struct page*), void *data) { - struct page *page; - - page = read_cache_page_async(mapping, index, filler, data); - if (IS_ERR(page)) - goto out; - wait_on_page_locked(page); - if (!PageUptodate(page)) { - page_cache_release(page); - page = ERR_PTR(-EIO); - } - out: - return page; + return wait_on_page_read(read_cache_page_async(mapping, index, filler, data)); } EXPORT_SYMBOL(read_cache_page); -- cgit v1.2.3 From a7016235a61d520e6806f38129001d935c4b6661 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Fri, 29 Jan 2010 17:46:34 +0000 Subject: mm: fix migratetype bug which slowed swapping After memory pressure has forced it to dip into the reserves, 2.6.32's 5f8dcc21211a3d4e3a7a5ca366b469fb88117f61 "page-allocator: split per-cpu list into one-list-per-migrate-type" has been returning MIGRATE_RESERVE pages to the MIGRATE_MOVABLE free_list: in some sense depleting reserves. Fix that in the most straightforward way (which, considering the overheads of alternative approaches, is Mel's preference): the right migratetype is already in page_private(page), but free_pcppages_bulk() wasn't using it. How did this bug show up? As a 20% slowdown in my tmpfs loop kbuild swapping tests, on PowerMac G5 with SLUB allocator. Bisecting to that commit was easy, but explaining the magnitude of the slowdown not easy. The same effect appears, but much less markedly, with SLAB, and even less markedly on other machines (the PowerMac divides into fewer zones than x86, I think that may be a factor). We guess that lumpy reclaim of short-lived high-order pages is implicated in some way, and probably this bug has been tickling a poor decision somewhere in page reclaim. But instrumentation hasn't told me much, I've run out of time and imagination to determine exactly what's going on, and shouldn't hold up the fix any longer: it's valid, and might even fix other misbehaviours. Signed-off-by: Hugh Dickins Acked-by: Mel Gorman Cc: stable@kernel.org Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'mm') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index d2a8889b4c5..8deb9d0fd5b 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -556,8 +556,9 @@ static void free_pcppages_bulk(struct zone *zone, int count, page = list_entry(list->prev, struct page, lru); /* must delete as __free_one_page list manipulates */ list_del(&page->lru); - __free_one_page(page, zone, 0, migratetype); - trace_mm_page_pcpu_drain(page, 0, migratetype); + /* MIGRATE_MOVABLE list may include MIGRATE_RESERVEs */ + __free_one_page(page, zone, 0, page_private(page)); + trace_mm_page_pcpu_drain(page, 0, page_private(page)); } while (--count && --batch_free && !list_empty(list)); } spin_unlock(&zone->lock); -- cgit v1.2.3