aboutsummaryrefslogtreecommitdiff
path: root/mm/memory.c
AgeCommit message (Collapse)Author
2006-03-22[PATCH] hugepage: Fix hugepage logic in free_pgtables() harderDavid Gibson
Turns out the hugepage logic in free_pgtables() was doubly broken. The loop coalescing multiple normal page VMAs into one call to free_pgd_range() had an off by one error, which could mean it would coalesce one hugepage VMA into the same bundle (checking 'vma' not 'next' in the loop). I transferred this bug into the new is_vm_hugetlb_page() based version. Here's the fix. This one didn't bite on powerpc previously for the same reason the is_hugepage_only_range() problem didn't: powerpc's hugetlb_free_pgd_range() is identical to free_pgd_range(). It didn't bite on ia64 because the hugepage region is distant enough from any other region that the separated PMD_SIZE distance test would always prevent coalescing the two together. No libhugetlbfs testsuite regressions (ppc64, POWER5). Signed-off-by: David Gibson <dwg@au1.ibm.com> Cc: William Lee Irwin III <wli@holomorphy.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-22[PATCH] hugepage: Fix hugepage logic in free_pgtables()David Gibson
free_pgtables() has special logic to call hugetlb_free_pgd_range() instead of the normal free_pgd_range() on hugepage VMAs. However, the test it uses to do so is incorrect: it calls is_hugepage_only_range on a hugepage sized range at the start of the vma. is_hugepage_only_range() will return true if the given range has any intersection with a hugepage address region, and in this case the given region need not be hugepage aligned. So, for example, this test can return true if called on, say, a 4k VMA immediately preceding a (nicely aligned) hugepage VMA. At present we get away with this because the powerpc version of hugetlb_free_pgd_range() is just a call to free_pgd_range(). On ia64 (the only other arch with a non-trivial is_hugepage_only_range()) we get away with it for a different reason; the hugepage area is not contiguous with the rest of the user address space, and VMAs are not permitted in between, so the test can't return a false positive there. Nonetheless this should be fixed. We do that in the patch below by replacing the is_hugepage_only_range() test with an explicit test of the VMA using is_vm_hugetlb_page(). This in turn changes behaviour for platforms where is_hugepage_only_range() returns false always (everything except powerpc and ia64). We address this by ensuring that hugetlb_free_pgd_range() is defined to be identical to free_pgd_range() (instead of a no-op) on everything except ia64. Even so, it will prevent some otherwise possible coalescing of calls down to free_pgd_range(). Since this only happens for hugepage VMAs, removing this small optimization seems unlikely to cause any trouble. This patch causes no regressions on the libhugetlbfs testsuite - ppc64 POWER5 (8-way), ppc64 G5 (2-way) and i386 Pentium M (UP). Signed-off-by: David Gibson <dwg@au1.ibm.com> Cc: William Lee Irwin III <wli@holomorphy.com> Acked-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-22[PATCH] mm: more CONFIG_DEBUG_VMNick Piggin
Put a few more checks under CONFIG_DEBUG_VM Signed-off-by: Nick Piggin <npiggin@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-22[PATCH] mm: split highorder pagesNick Piggin
Have an explicit mm call to split higher order pages into individual pages. Should help to avoid bugs and be more explicit about the code's intention. Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: Russell King <rmk@arm.linux.org.uk> Cc: David Howells <dhowells@redhat.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mundt <lethal@linux-sh.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Chris Zankel <chris@zankel.net> Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-17[PATCH] fix free swap cache latencyHugh Dickins
Lee Revell reported 28ms latency when process with lots of swapped memory exits. 2.6.15 introduced a latency regression when unmapping: in accounting the zap_work latency breaker, pte_none counted 1, pte_present PAGE_SIZE, but a swap entry counted nothing at all. We think of pages present as the slow case, but Lee's trace shows that free_swap_and_cache's radix tree lookup can make a lot of work - and we could have been doing it many thousands of times without a latency break. Move the zap_work update up to account swap entries like pages present. This does account non-linear pte_file entries, and unmap_mapping_range skipping over swap entries, by the same amount even though they're quick: but neither of those cases deserves complicating the code (and they're treated no worse than they were in 2.6.14). Signed-off-by: Hugh Dickins <hugh@veritas.com> Acked-by: Nick Piggin <npiggin@suse.de> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-17[PATCH] x86_64: Add boot option to disable randomized mappings and cleanupAndi Kleen
AMD SimNow!'s JIT doesn't like them at all in the guest. For distribution installation it's easiest if it's a boot time option. Also I moved the variable to a more appropiate place and make it independent from sysctl And marked __read_mostly which it is. Signed-off-by: Andi Kleen <ak@suse.de> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-01[PATCH] Direct Migration V9: PageSwapCache checksChristoph Lameter
Check for PageSwapCache after looking up and locking a swap page. The page migration code may change a swap pte to point to a different page under lock_page(). If that happens then the vm must retry the lookup operation in the swap space to find the correct page number. There are a couple of locations in the VM where a lock_page() is done on a swap page. In these locations we need to check afterwards if the page was migrated. If the page was migrated then the old page that was looked up before was freed and no longer has the PageSwapCache bit set. Signed-off-by: Hirokazu Takahashi <taka@valinux.co.jp> Signed-off-by: Dave Hansen <haveblue@us.ibm.com> Signed-off-by: Christoph Lameter <clameter@@sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-09[PATCH] mutex subsystem, semaphore to mutex: VFS, ->i_semJes Sorensen
This patch converts the inode semaphore to a mutex. I have tested it on XFS and compiled as much as one can consider on an ia64. Anyway your luck with it might be different. Modified-by: Ingo Molnar <mingo@elte.hu> (finished the conversion) Signed-off-by: Jes Sorensen <jes@sgi.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2006-01-09[PATCH] spufs: The SPU file system, baseArnd Bergmann
This is the current version of the spu file system, used for driving SPEs on the Cell Broadband Engine. This release is almost identical to the version for the 2.6.14 kernel posted earlier, which is available as part of the Cell BE Linux distribution from http://www.bsc.es/projects/deepcomputing/linuxoncell/. The first patch provides all the interfaces for running spu application, but does not have any support for debugging SPU tasks or for scheduling. Both these functionalities are added in the subsequent patches. See Documentation/filesystems/spufs.txt on how to use spufs. Signed-off-by: Arnd Bergmann <arndb@de.ibm.com> Signed-off-by: Paul Mackerras <paulus@samba.org>
2006-01-06[PATCH] mm: pfault optimisationNick Piggin
This atomic operation is superfluous: the pte will be added with the referenced bit set, and the page will be referenced through this mapping after the page fault handler returns anyway. Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-06[PATCH] mm: rmap optimisationNick Piggin
Optimise rmap functions by minimising atomic operations when we know there will be no concurrent modifications. Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-06[PATCH] madvise(MADV_REMOVE): remove pages from tmpfs shm backing storeBadari Pulavarty
Here is the patch to implement madvise(MADV_REMOVE) - which frees up a given range of pages & its associated backing store. Current implementation supports only shmfs/tmpfs and other filesystems return -ENOSYS. "Some app allocates large tmpfs files, then when some task quits and some client disconnect, some memory can be released. However the only way to release tmpfs-swap is to MADV_REMOVE". - Andrea Arcangeli Databases want to use this feature to drop a section of their bufferpool (shared memory segments) - without writing back to disk/swap space. This feature is also useful for supporting hot-plug memory on UML. Concerns raised by Andrew Morton: - "We have no plan for holepunching! If we _do_ have such a plan (or might in the future) then what would the API look like? I think sys_holepunch(fd, start, len), so we should start out with that." - Using madvise is very weird, because people will ask "why do I need to mmap my file before I can stick a hole in it?" - None of the other madvise operations call into the filesystem in this manner. A broad question is: is this capability an MM operation or a filesytem operation? truncate, for example, is a filesystem operation which sometimes has MM side-effects. madvise is an mm operation and with this patch, it gains FS side-effects, only they're really, really significant ones." Comments: - Andrea suggested the fs operation too but then it's more efficient to have it as a mm operation with fs side effects, because they don't immediatly know fd and physical offset of the range. It's possible to fixup in userland and to use the fs operation but it's more expensive, the vmas are already in the kernel and we can use them. Short term plan & Future Direction: - We seem to need this interface only for shmfs/tmpfs files in the short term. We have to add hooks into the filesystem for correctness and completeness. This is what this patch does. - In the future, plan is to support both fs and mmap apis also. This also involves (other) filesystem specific functions to be implemented. - Current patch doesn't support VM_NONLINEAR - which can be addressed in the future. Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com> Cc: Hugh Dickins <hugh@veritas.com> Cc: Andrea Arcangeli <andrea@suse.de> Cc: Michael Kerrisk <mtk-manpages@gmx.net> Cc: Ulrich Drepper <drepper@redhat.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-12-16Make sure we copy pages inserted with "vm_insert_page()" on forkLinus Torvalds
The logic that decides that a fork() might be able to avoid copying a VM area when it can be re-created by page faults didn't know about the new vm_insert_page() case. Also make some things a bit more anal wrt VM_PFNMAP. Pointed out by Hugh Dickins <hugh@veritas.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-12-12get_user_pages: don't try to follow PFNMAP pagesLinus Torvalds
Nick Piggin points out that a few drivers play games with VM_IO (why? who knows..) and thus a pfn-remapped area may not have that bit set even if remap_pfn_range() set it originally. So make it explicit in get_user_pages() that we don't follow VM_PFNMAP pages, since pretty much by definition they do not have a "struct page" associated with them. Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-12-11Allow arbitrary read-only shared pfn-remapping tooLinus Torvalds
The VM layer (for historical reasons) turns a read-only shared mmap into a private-like mapping with the VM_MAYWRITE bit clear. Thus checking just VM_SHARED isn't actually sufficient. So use a trivial helper function for the cases where we wanted to inquire if a mapping was COW-like or not. Moo! Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-12-11Remove (at least temporarily) the "incomplete PFN mapping" supportLinus Torvalds
With the previous commit, we can handle arbitrary shared re-mappings even without this complexity, and since the only known private mappings are for strange users of /dev/mem (which never create an incomplete one), there seems to be no reason to support it. Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-12-11Allow arbitrary shared PFNMAP'sLinus Torvalds
A shared mapping doesn't cause COW-pages, so we don't need to worry about the whole vm_pgoff logic to decide if a PFN-remapped page has gone through COW or not. This makes it possible to entirely avoid the special "partial remapping" logic for the common case. Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-12-03Make vm_insert_page() available to NVidia moduleLinus Torvalds
It used to use remap_pfn_range(), which wasn't GPL-only either, and the new interface is actually simpler and does more checking, so we shouldn't unnecessarily discourage people from switching over. Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-30VM: add "vm_insert_page()" functionLinus Torvalds
This is what a lot of drivers will actually want to use to insert individual pages into a user VMA. It doesn't have the old PageReserved restrictions of remap_pfn_range(), and it doesn't complain about partial remappings. The page you insert needs to be a nice clean kernel allocation, so you can't insert arbitrary page mappings with this, but that's not what people want. Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-29[PATCH] VM: Fix typos in get_locked_pteTrond Myklebust
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-29[PATCH] pfnmap: do_no_page BUG_ON againHugh Dickins
Use copy_user_highpage directly instead of cow_user_page in do_no_page: in the immediately following page_cache_release, and elsewhere, it is assuming that new_page is normal. If any VM_PFNMAP driver can get to do_no_page, it's just a BUG (but not in the case of do_anonymous_page). Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-29[PATCH] pfnmap: remove src_page from do_wp_pageHugh Dickins
Clean away do_wp_page's "src_page": cow_user_page makes it unnecessary. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-29cow_user_page: fix page alignmentLinus Torvalds
High Dickins points out that the user virtual address passed to the page fault handler isn't necessarily page-aligned. Also, add a comment on why the copy could fail for the user address case. Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-29VM: add common helper function to create the page tablesLinus Torvalds
This logic was duplicated four times, for no good reason. Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-29Support strange discontiguous PFN remappingsLinus Torvalds
These get created by some drivers that don't generally even want a pfn remapping at all, but would really mostly prefer to just map pages they've allocated individually instead. For now, create a helper function that turns such an incomplete PFN remapping call into a loop that does that explicit mapping. In the long run we almost certainly want to export a totally different interface for that, though. Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-29[PATCH] Fix missing pfn variables caused by vm changesBen Collins
I image this showed up because of "unused var..." when the changes occured, because flush_cache_page() is a noop in most places. This showed up for me on parisc however, where flush_cache_page() is a real function. Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-29[PATCH] Fix vma argument in get_usr_pages() for gate areasNick Piggin
The system call gate area handling called vm_normal_page() with the wrong vma (which was always NULL, and caused an oops). Signed-off-by: Nick Piggin <npiggin@suse.de> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-28[PATCH] Workaround for gcc 2.96 (undefined references)Alan Stern
LD .tmp_vmlinux1 mm/built-in.o(.text+0x100d6): In function `copy_page_range': : undefined reference to `__pud_alloc' mm/built-in.o(.text+0x1010b): In function `copy_page_range': : undefined reference to `__pmd_alloc' mm/built-in.o(.text+0x11ef4): In function `__handle_mm_fault': : undefined reference to `__pud_alloc' fs/built-in.o(.text+0xc930): In function `install_arg_page': : undefined reference to `__pud_alloc' make: *** [.tmp_vmlinux1] Error 1 Those missing references in mm/memory.c arise from this code in include/linux/mm.h, combined with the fact that __PGTABLE_PMD_FOLDED and __PGTABLE_PUD_FOLDED are both set and __ARCH_HAS_4LEVEL_HACK is not: /* * The following ifdef needed to get the 4level-fixup.h header to work. * Remove it when 4level-fixup.h has been removed. */ #if defined(CONFIG_MMU) && !defined(__ARCH_HAS_4LEVEL_HACK) static inline pud_t *pud_alloc(struct mm_struct *mm, pgd_t *pgd, unsigned long address) { return (unlikely(pgd_none(*pgd)) && __pud_alloc(mm, pgd, address))? NULL: pud_offset(pgd, address); } static inline pmd_t *pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address) { return (unlikely(pud_none(*pud)) && __pmd_alloc(mm, pud, address))? NULL: pmd_offset(pud, address); } #endif /* CONFIG_MMU && !__ARCH_HAS_4LEVEL_HACK */ With my configuration the pgd_none and pud_none routines are inlines returning a constant 0. Apparently the old compiler avoids generating calls to __pud_alloc and __pmd_alloc but still lists them as undefined references in the module's symbol table. I don't know which change caused this problem. I think it was added somewhere between 2.6.14 and 2.6.15-rc1, because I remember building several 2.6.14-rc kernels without difficulty. However I can't point to an individual culprit. Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-28mm: re-architect the VM_UNPAGED logicLinus Torvalds
This replaces the (in my opinion horrible) VM_UNMAPPED logic with very explicit support for a "remapped page range" aka VM_PFNMAP. It allows a VM area to contain an arbitrary range of page table entries that the VM never touches, and never considers to be normal pages. Any user of "remap_pfn_range()" automatically gets this new functionality, and doesn't even have to mark the pages reserved or indeed mark them any other way. It just works. As a side effect, doing mmap() on /dev/mem works for arbitrary ranges. Sparc update from David in the next commit. Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-22[PATCH] unpaged: ZERO_PAGE in VM_UNPAGEDHugh Dickins
It's strange enough to be looking out for anonymous pages in VM_UNPAGED areas, let's not insert the ZERO_PAGE there - though whether it would matter will depend on what we decide about ZERO_PAGE refcounting. But whereas do_anonymous_page may (exceptionally) be called on a VM_UNPAGED area, do_no_page should never be: just BUG_ON. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-22[PATCH] unpaged: anon in VM_UNPAGEDHugh Dickins
copy_one_pte needs to copy the anonymous COWed pages in a VM_UNPAGED area, zap_pte_range needs to free them, do_wp_page needs to COW them: just like ordinary pages, not like the unpaged. But recognizing them is a little subtle: because PageReserved is no longer a condition for remap_pfn_range, we can now mmap all of /dev/mem (whether the distro permits, and whether it's advisable on this or that architecture, is another matter). So if we can see a PageAnon, it may not be ours to mess with (or may be ours from elsewhere in the address space). I suspect there's an entertaining insoluble self-referential problem here, but the page_is_anon function does a good practical job, and MAP_PRIVATE PROT_WRITE VM_UNPAGED will always be an odd choice. In updating the comment on page_address_in_vma, noticed a potential NULL dereference, in a path we don't actually take, but fixed it. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-22[PATCH] unpaged: COW on VM_UNPAGEDHugh Dickins
Remove the BUG_ON(vma->vm_flags & VM_UNPAGED) from do_wp_page, and let it do Copy-On-Write without touching the VM_UNPAGED's page counts - but this is incomplete, because the anonymous page it inserts will itself need to be handled, here and in other functions - next patch. We still don't copy the page if the pfn is invalid, because the copy_user_highpage interface does not allow it. But that's not been a problem in the past: can be added in later if the need arises. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-22[PATCH] unpaged: VM_UNPAGEDHugh Dickins
Although we tend to associate VM_RESERVED with remap_pfn_range, quite a few drivers set VM_RESERVED on areas which are then populated by nopage. The PageReserved removal in 2.6.15-rc1 changed VM_RESERVED not to free pages in zap_pte_range, without changing those drivers not to set it: so their pages just leak away. Let's not change miscellaneous drivers now: introduce VM_UNPAGED at the core, to flag the special areas where the ptes may have no struct page, or if they have then it's not to be touched. Replace most instances of VM_RESERVED in core mm by VM_UNPAGED. Force it on in remap_pfn_range, and the sparc and sparc64 io_remap_pfn_range. Revert addition of VM_RESERVED to powerpc vdso, it's not needed there. Is it needed anywhere? It still governs the mm->reserved_vm statistic, and special vmas not to be merged, and areas not to be core dumped; but could probably be eliminated later (the drivers are probably specifying it because in 2.4 it kept swapout off the vma, but in 2.6 we work from the LRU, which these pages don't get on). Use the VM_SHM slot for VM_UNPAGED, and define VM_SHM to 0: it serves no purpose whatsoever, and should be removed from drivers when we clean up. Signed-off-by: Hugh Dickins <hugh@veritas.com> Acked-by: William Irwin <wli@holomorphy.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-22[PATCH] unpaged: get_user_pages VM_RESERVEDHugh Dickins
The PageReserved removal in 2.6.15-rc1 prohibited get_user_pages on the areas flagged VM_RESERVED in place of PageReserved. That is correct in theory - we ought not to interfere with struct pages in such a reserved area; but in practice it broke BTTV for one. So revert to prohibiting only on VM_IO: if someone gets into trouble with get_user_pages on VM_RESERVED, it'll just be a "don't do that". You can argue that videobuf_mmap_mapper shouldn't set VM_RESERVED in the first place, but now's not the time for breaking drivers without notice. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-13[PATCH] mm: ZAP_BLOCK causes redundant workRobin Holt
The address based work estimate for unmapping (for lockbreak) is and always was horribly inefficient for sparse mappings. The problem is most simply explained with an example: If we find a pgd is clear, we still have to call into unmap_page_range PGDIR_SIZE / ZAP_BLOCK_SIZE times, each time checking the clear pgd, in order to progress the working address to the next pgd. The fundamental way to solve the problem is to keep track of the end address we've processed and pass it back to the higher layers. From: Nick Piggin <npiggin@suse.de> Modification to completely get away from address based work estimate and instead use an abstract count, with a very small cost for empty entries as opposed to present pages. On 2.6.14-git2, ppc64, and CONFIG_PREEMPT=y, mapping and unmapping 1TB of virtual address space takes 1.69s; with the following patch applied, this operation can be done 1000 times in less than 0.01s From: Andrew Morton <akpm@osdl.org> With CONFIG_HUTETLB_PAGE=n: mm/memory.c: In function `unmap_vmas': mm/memory.c:779: warning: division by zero Due to zap_work -= (end - start) / (HPAGE_SIZE / PAGE_SIZE); So make the dummy HPAGE_SIZE non-zero Signed-off-by: Robin Holt <holt@sgi.com> Signed-off-by: Nick Piggin <npiggin@suse.de> Cc: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29[PATCH] .text page fault SMP scalability optimizationAndrea Arcangeli
We had a problem on ppc64 where with more than 4 threads a large system wouldn't scale well while faulting in the .text (most of the time was spent in the kernel despite it was an userland compute intensive app). The reason is the useless overwrite of the same pte from all cpu. I fixed it this way (verified on an older kernel but the forward port is almost identical). This will benefit all archs not just ppc64. Signed-off-by: Andrea Arcangeli <andrea@suse.de> Cc: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29[PATCH] mm: fix rss and mmlist lockingHugh Dickins
A couple of oddities were guarded by page_table_lock, no longer properly guarded when that is split. The mm_counters of file_rss and anon_rss: make those an atomic_t, or an atomic64_t if the architecture supports it, in such a case. Definitions by courtesy of Christoph Lameter: who spent considerable effort on more scalable ways of counting, but found insufficient benefit in practice. And adding an mm with swap to the mmlist for swapoff: the list is well- guarded by its own lock, but the list_empty check now has to be repeated inside it. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29[PATCH] mm: split page table lockHugh Dickins
Christoph Lameter demonstrated very poor scalability on the SGI 512-way, with a many-threaded application which concurrently initializes different parts of a large anonymous area. This patch corrects that, by using a separate spinlock per page table page, to guard the page table entries in that page, instead of using the mm's single page_table_lock. (But even then, page_table_lock is still used to guard page table allocation, and anon_vma allocation.) In this implementation, the spinlock is tucked inside the struct page of the page table page: with a BUILD_BUG_ON in case it overflows - which it would in the case of 32-bit PA-RISC with spinlock debugging enabled. Splitting the lock is not quite for free: another cacheline access. Ideally, I suppose we would use split ptlock only for multi-threaded processes on multi-cpu machines; but deciding that dynamically would have its own costs. So for now enable it by config, at some number of cpus - since the Kconfig language doesn't support inequalities, let preprocessor compare that with NR_CPUS. But I don't think it's worth being user-configurable: for good testing of both split and unsplit configs, split now at 4 cpus, and perhaps change that to 8 later. There is a benefit even for singly threaded processes: kswapd can be attacking one part of the mm while another part is busy faulting. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29[PATCH] mm: follow_page with inner ptlockHugh Dickins
Final step in pushing down common core's page_table_lock. follow_page no longer wants caller to hold page_table_lock, uses pte_offset_map_lock itself; and so no page_table_lock is taken in get_user_pages itself. But get_user_pages (and get_futex_key) do then need follow_page to pin the page for them: take Daniel's suggestion of bitflags to follow_page. Need one for WRITE, another for TOUCH (it was the accessed flag before: vanished along with check_user_page_readable, but surely get_numa_maps is wrong to mark every page it finds as accessed), another for GET. And another, ANON to dispose of untouched_anonymous_page: it seems silly for that to descend a second time, let follow_page observe if there was no page table and return ZERO_PAGE if so. Fix minor bug in that: check VM_LOCKED - make_pages_present ought to make readonly anonymous present. Give get_numa_maps a cond_resched while we're there. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29[PATCH] mm: kill check_user_page_readableHugh Dickins
check_user_page_readable is a problematic variant of follow_page. It's used only by oprofile's i386 and arm backtrace code, at interrupt time, to establish whether a userspace stackframe is currently readable. This is problematic, because we want to push the page_table_lock down inside follow_page, and later split it; whereas oprofile is doing a spin_trylock on it (in the i386 case, forgotten in the arm case), and needs that to pin perhaps two pages spanned by the stackframe (which might be covered by different locks when we split). I think oprofile is going about this in the wrong way: it doesn't need to know the area is readable (neither i386 nor arm uses read protection of user pages), it doesn't need to pin the memory, it should simply __copy_from_user_inatomic, and see if that succeeds or not. Sorry, but I've not got around to devising the sparse __user annotations for this. Then we can eliminate check_user_page_readable, and return to a single follow_page without the __follow_page variants. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29[PATCH] mm: unmap_vmas with inner ptlockHugh Dickins
Remove the page_table_lock from around the calls to unmap_vmas, and replace the pte_offset_map in zap_pte_range by pte_offset_map_lock: all callers are now safe to descend without page_table_lock. Don't attempt fancy locking for hugepages, just take page_table_lock in unmap_hugepage_range. Which makes zap_hugepage_range, and the hugetlb test in zap_page_range, redundant: unmap_vmas calls unmap_hugepage_range anyway. Nor does unmap_vmas have much use for its mm arg now. The tlb_start_vma and tlb_end_vma in unmap_page_range are now called without page_table_lock: if they're implemented at all, they typically come down to flush_cache_range (usually done outside page_table_lock) and flush_tlb_range (which we already audited for the mprotect case). Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29[PATCH] mm: unlink vma before pagetablesHugh Dickins
In most places the descent from pgd to pud to pmd to pte holds mmap_sem (exclusively or not), which ensures that free_pgtables cannot be freeing page tables from any level at the same time. But truncation and reverse mapping descend without mmap_sem. No problem: just make sure that a vma is unlinked from its prio_tree (or nonlinear list) and from its anon_vma list, after zapping the vma, but before freeing its page tables. Then neither vmtruncate nor rmap can reach that vma whose page tables are now volatile (nor do they need to reach it, since all its page entries have been zapped by this stage). The i_mmap_lock and anon_vma->lock already serialize this correctly; but the locking hierarchy is such that we cannot take them while holding page_table_lock. Well, we're trying to push that down anyway. So in this patch, move anon_vma_unlink and unlink_file_vma into free_pgtables, at the same time as moving page_table_lock around calls to unmap_vmas. tlb_gather_mmu and tlb_finish_mmu then fall outside the page_table_lock, but we made them preempt_disable and preempt_enable earlier; and a long source audit of all the architectures has shown no problem with removing page_table_lock from them. free_pgtables doesn't need page_table_lock for itself, nor for what it calls; tlb->mm->nr_ptes is usually protected by page_table_lock, but partly by non-exclusive mmap_sem - here it's decremented with exclusive mmap_sem, or mm_users 0. update_hiwater_rss and vm_unacct_memory don't need page_table_lock either. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29[PATCH] mm: page fault handler lockingHugh Dickins
On the page fault path, the patch before last pushed acquiring the page_table_lock down to the head of handle_pte_fault (though it's also taken and dropped earlier when a new page table has to be allocated). Now delete that line, read "entry = *pte" without it, and go off to this or that page fault handler on the basis of this unlocked peek. Usually the handler can proceed without the lock, relying on the subsequent locked pte_same or pte_none test to back out when necessary; though do_wp_page needs the lock immediately, and do_file_page doesn't check (if there's a race, install_page just zaps the entry and reinstalls it). But on those architectures (notably i386 with PAE) whose pte is too big to be read atomically, if SMP or preemption is enabled, do_swap_page and do_file_page might cause irretrievable damage if passed a Frankenstein entry stitched together from unrelated parts. In those configs, "pte_unmap_same" has to take page_table_lock, validate orig_pte still the same, and drop page_table_lock before unmapping, before proceeding. Use pte_offset_map_lock and pte_unmap_unlock throughout the handlers; but lock avoidance leaves more lone maps and unmaps than elsewhere. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29[PATCH] mm: ptd_alloc take ptlockHugh Dickins
Second step in pushing down the page_table_lock. Remove the temporary bridging hack from __pud_alloc, __pmd_alloc, __pte_alloc: expect callers not to hold page_table_lock, whether it's on init_mm or a user mm; take page_table_lock internally to check if a racing task already allocated. Convert their callers from common code. But avoid coming back to change them again later: instead of moving the spin_lock(&mm->page_table_lock) down, switch over to new macros pte_alloc_map_lock and pte_unmap_unlock, which encapsulate the mapping+locking and unlocking+unmapping together, and in the end may use alternatives to the mm page_table_lock itself. These callers all hold mmap_sem (some exclusively, some not), so at no level can a page table be whipped away from beneath them; and pte_alloc uses the "atomic" pmd_present to test whether it needs to allocate. It appears that on all arches we can safely descend without page_table_lock. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29[PATCH] mm: ptd_alloc inline and outHugh Dickins
It seems odd to me that, whereas pud_alloc and pmd_alloc test inline, only calling out-of-line __pud_alloc __pmd_alloc if allocation needed, pte_alloc_map and pte_alloc_kernel are entirely out-of-line. Though it does add a little to kernel size, change them to macros testing inline, calling __pte_alloc or __pte_alloc_kernel to allocate out-of-line. Mark none of them as fastcalls, leave that to CONFIG_REGPARM or not. It also seems more natural for the out-of-line functions to leave the offset calculation and map to the inline, which has to do it anyway for the common case. At least mremap move wants __pte_alloc without _map. Macros rather than inline functions, certainly to avoid the header file issues which arise from CONFIG_HIGHPTE needing kmap_types.h, but also in case any architectures I haven't built would have other such problems. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29[PATCH] mm: init_mm without ptlockHugh Dickins
First step in pushing down the page_table_lock. init_mm.page_table_lock has been used throughout the architectures (usually for ioremap): not to serialize kernel address space allocation (that's usually vmlist_lock), but because pud_alloc,pmd_alloc,pte_alloc_kernel expect caller holds it. Reverse that: don't lock or unlock init_mm.page_table_lock in any of the architectures; instead rely on pud_alloc,pmd_alloc,pte_alloc_kernel to take and drop it when allocating a new one, to check lest a racing task already did. Similarly no page_table_lock in vmalloc's map_vm_area. Some temporary ugliness in __pud_alloc and __pmd_alloc: since they also handle user mms, which are converted only by a later patch, for now they have to lock differently according to whether or not it's init_mm. If sources get muddled, there's a danger that an arch source taking init_mm.page_table_lock will be mixed with common source also taking it (or neither take it). So break the rules and make another change, which should break the build for such a mismatch: remove the redundant mm arg from pte_alloc_kernel (ppc64 scrapped its distinct ioremap_mm in 2.6.13). Exceptions: arm26 used pte_alloc_kernel on user mm, now pte_alloc_map; ia64 used pte_alloc_map on init_mm, now pte_alloc_kernel; parisc had bad args to pmd_alloc and pte_alloc_kernel in unused USE_HPPA_IOREMAP code; ppc64 map_io_page forgot to unlock on failure; ppc mmu_mapin_ram and ppc64 im_free took page_table_lock for no good reason. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29[PATCH] mm: update_hiwaters just in timeHugh Dickins
update_mem_hiwater has attracted various criticisms, in particular from those concerned with mm scalability. Originally it was called whenever rss or total_vm got raised. Then many of those callsites were replaced by a timer tick call from account_system_time. Now Frank van Maarseveen reports that to be found inadequate. How about this? Works for Frank. Replace update_mem_hiwater, a poor combination of two unrelated ops, by macros update_hiwater_rss and update_hiwater_vm. Don't attempt to keep mm->hiwater_rss up to date at timer tick, nor every time we raise rss (usually by 1): those are hot paths. Do the opposite, update only when about to lower rss (usually by many), or just before final accounting in do_exit. Handle mm->hiwater_vm in the same way, though it's much less of an issue. Demand that whoever collects these hiwater statistics do the work of taking the maximum with rss or total_vm. And there has been no collector of these hiwater statistics in the tree. The new convention needs an example, so match Frank's usage by adding a VmPeak line above VmSize to /proc/<pid>/status, and also a VmHWM line above VmRSS (High-Water-Mark or High-Water-Memory). There was a particular anomaly during mremap move, that hiwater_vm might be captured too high. A fleeting such anomaly remains, but it's quickly corrected now, whereas before it would stick. What locking? None: if the app is racy then these statistics will be racy, it's not worth any overhead to make them exact. But whenever it suits, hiwater_vm is updated under exclusive mmap_sem, and hiwater_rss under page_table_lock (for now) or with preemption disabled (later on): without going to any trouble, minimize the time between reading current values and updating, to minimize those occasions when a racing thread bumps a count up and back down in between. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29[PATCH] mm: do_swap_page race majorHugh Dickins
Small adjustment: do_swap_page should report its !pte_same race as a major fault if it had to read into swap cache, because whatever raced with it will have found page already in cache and reported minor fault. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29[PATCH] mm: zap_pte_range dec rssHugh Dickins
Small adjustment: zap_pte_range decrement its rss counts from 0 then finally add, avoiding negations - we don't have or need a sub_mm_rss. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29[PATCH] mm: copy_one_pte inc rssHugh Dickins
Small adjustment, following Nick's suggestion: it's more straightforward for copy_pte_range to let copy_one_pte do the rss incrementation, than use an index it passed back. Saves a #define, and 16 bytes of .text. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>