- 14 Apr, 2015 40 commits
-
-
Toshi Kani authored
ioremap() and its related interfaces are used to create I/O mappings to memory-mapped I/O devices. The mapping sizes of the traditional I/O devices are relatively small. Non-volatile memory (NVM), however, has many GB and is going to have TB soon. It is not very efficient to create large I/O mappings with 4KB. This patchset extends the ioremap() interfaces to transparently create I/O mappings with huge pages whenever possible. ioremap() continues to use 4KB mappings when a huge page does not fit into a requested range. There is no change necessary to the drivers using ioremap(). A requested physical address must be aligned by a huge page size (1GB or 2MB on x86) for using huge page mapping, though. The kernel huge I/O mapping will improve performance of NVM and other devices with large memory, and reduce the time to create their mappings as well. On x86, MTRRs can override PAT memory types with a 4KB granularity. When using a huge page, MTRRs can override the memory type of the huge page, which may lead a performance penalty. The processor can also behave in an undefined manner if a huge page is mapped to a memory range that MTRRs have mapped with multiple different memory types. Therefore, the mapping code falls back to use a smaller page size toward 4KB when a mapping range is covered by non-WB type of MTRRs. The WB type of MTRRs has no affect on the PAT memory types. The patchset introduces HAVE_ARCH_HUGE_VMAP, which indicates that the arch supports huge KVA mappings for ioremap(). User may specify a new kernel option "nohugeiomap" to disable the huge I/O mapping capability of ioremap() when necessary. Patch 1-4 change common files to support huge I/O mappings. There is no change in the functinalities unless HAVE_ARCH_HUGE_VMAP is defined on the architecture of the system. Patch 5-6 implement the HAVE_ARCH_HUGE_VMAP funcs on x86, and set HAVE_ARCH_HUGE_VMAP on x86. This patch (of 6): __get_vm_area_node() takes unsigned long size, which is a 64-bit value on a 64-bit kernel. However, fls(size) simply ignores the upper 32-bit. Change to use fls_long() to handle the size properly. Signed-off-by:
Toshi Kani <toshi.kani@hp.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Robert Elliott <Elliott@hp.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Yaowei Bai authored
Signed-off-by:
Yaowei Bai <bywxiaobai@163.com> Signed-off-by:
-
Michal Hocko authored
Commit 920c3ed7 ("[SPARC64]: Add basic infrastructure for MD add/remove notification") has added __GFP_NOFAIL for the allocation request but it hasn't mentioned why is this strict requirement really needed. The code was handling an allocation failure and propagated it properly up the callchain so it is not clear why it is needed. Dave has clarified the intention when I tried to remove the flag as not being necessary: : It is a serious failure. : : If we miss an MDESC update due to this allocation failure, the update : is not an event which gets retransmitted so we will lose the updated : machine description forever. : : We really need this allocation to succeed. So add a comment to clarify the nofail flag and get rid of the failure check because __GFP_NOFAIL allocation doesn't fail. Signed-off-by:
Michal Hocko <mhocko@suse.cz> Cc: David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Dave Chinner <david@fromorbit.com> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: "David S. Miller" <davem@davemloft.net> Cc: Vipul Pandya <vipul@chelsio.com> Cc: Jan Kara <jack@suse.cz> Signed-off-by:
-
Michal Hocko authored
__GFP_NOFAIL is documented as a deprecated flag since commit 478352e7 ("mm: add comment about deprecation of __GFP_NOFAIL"). This has discouraged people from using it but in some cases an opencoded endless loop around allocator has been used instead. So the allocator is not aware of the de facto __GFP_NOFAIL allocation because this information was not communicated properly. Let's make clear that if the allocation context really cannot afford failure because there is no good failure policy then using __GFP_NOFAIL is preferable to opencoding the loop outside of the allocator. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by:
David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Dave Chinner <david@fromorbit.com> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Mel Gorman <mgorman@suse.de> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Cc: "David S. Miller" <davem@davemloft.net> Cc: Vipul Pandya <vipul@chelsio.com> Cc: Jan Kara <jack@suse.cz> Signed-off-by:
-
Sasha Levin authored
Constify function parameters and use correct signness where needed. Signed-off-by:
Sasha Levin <sasha.levin@oracle.com> Cc: Michal Nazarewicz <mina86@mina86.com> Cc: Marek Szyprowski <m.szyprowski@samsung.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com> Acked-by:
Gregory Fong <gregory.0xf0@gmail.com> Cc: Pintu Kumar <pintu.k@samsung.com> Signed-off-by:
-
David Rientjes authored
Nothing calls __cpuset_node_allowed() with __GFP_THISNODE set anymore, so remove the obscure comment about it and its special-case exception. Signed-off-by:
Vlastimil Babka <vbabka@suse.cz> Cc: Christoph Lameter <cl@linux.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Pravin Shelar <pshelar@nicira.com> Cc: Jarno Rajahalme <jrajahalme@nicira.com> Cc: Li Zefan <lizefan@huawei.com> Cc: Greg Thelen <gthelen@google.com> Cc: Tejun Heo <tj@kernel.org> Signed-off-by:
-
David Rientjes authored
Commit 077fcf11 ("mm/thp: allocate transparent hugepages on local node") restructured alloc_hugepage_vma() with the intent of only allocating transparent hugepages locally when there was not an effective interleave mempolicy. alloc_pages_exact_node() does not limit the allocation to the single node, however, but rather prefers it. This is because __GFP_THISNODE is not set which would cause the node-local nodemask to be passed. Without it, only a nodemask that prefers the local node is passed. Fix this by passing __GFP_THISNODE and falling back to small pages when the allocation fails. Commit 9f1b868a ("mm: thp: khugepaged: add policy for finding target node") suffers from a similar problem for khugepaged, which is also fixed. Fixes: 077fcf11 ("mm/thp: allocate transparent hugepages on local node") Fixes: 9f1b868a ("mm: thp: khugepaged: add policy for finding target node") Signed-off-by:
-
David Rientjes authored
NOTE: this is not about __GFP_THISNODE, this is only about GFP_THISNODE. GFP_THISNODE is a secret combination of gfp bits that have different behavior than expected. It is a combination of __GFP_THISNODE, __GFP_NORETRY, and __GFP_NOWARN and is special-cased in the page allocator slowpath to fail without trying reclaim even though it may be used in combination with __GFP_WAIT. An example of the problem this creates: commit e97ca8e5 ("mm: fix GFP_THISNODE callers and clarify") fixed up many users of GFP_THISNODE that really just wanted __GFP_THISNODE. The problem doesn't end there, however, because even it was a no-op for alloc_misplaced_dst_page(), which also sets __GFP_NORETRY and __GFP_NOWARN, and migrate_misplaced_transhuge_page(), where __GFP_NORETRY and __GFP_NOWAIT is set in GFP_TRANSHUGE. Converting GFP_THISNODE to __GFP_THISNODE is a no-op in these cases since the page allocator special-cases __GFP_THISNODE && __GFP_NORETRY && __GFP_NOWARN. It's time to just remove GFP_THISNODE entirely. We leave __GFP_THISNODE to restrict an allocation to a local node, but remove GFP_THISNODE and its obscurity. Instead, we require that a caller clear __GFP_WAIT if it wants to avoid reclaim. This allows the aforementioned functions to actually reclaim as they should. It also enables any future callers that want to do __GFP_THISNODE but also __GFP_NORETRY && __GFP_NOWARN to reclaim. The rule is simple: if you don't want to reclaim, then don't set __GFP_WAIT. Aside: ovs_flow_stats_update() really wants to avoid reclaim as well, so it is unchanged. Signed-off-by:
Pekka Enberg <penberg@kernel.org> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Acked-by:
Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Pravin Shelar <pshelar@nicira.com> Cc: Jarno Rajahalme <jrajahalme@nicira.com> Cc: Li Zefan <lizefan@huawei.com> Cc: Greg Thelen <gthelen@google.com> Cc: Tejun Heo <tj@kernel.org> Signed-off-by:
-
David Rientjes authored
migrate_to_node() is intended to migrate a page from one source node to a target node. Today, migrate_to_node() could end up migrating to any node, not only the target node. This is because the page migration allocator, new_node_page() does not pass __GFP_THISNODE to alloc_pages_exact_node(). This causes the target node to be preferred but allows fallback to any other node in order of affinity. Prevent this by allocating with __GFP_THISNODE. If memory is not available, -ENOMEM will be returned as appropriate. Signed-off-by:
Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Cc: Mel Gorman <mel@csn.ul.ie> Signed-off-by:
-
Vladimir Davydov authored
The limit equals 32 and is imposed by the number of entries in the fs_poolid_map and shared_fs_poolid_map. Nowadays it is insufficient, because with containers on board a Linux host can have hundreds of active fs mounts. These maps were introduced by commit 49a9ab81 ("mm: cleancache: lazy initialization to allow tmem backends to build/run as modules") in order to allow compiling cleancache drivers as modules. Real pool ids are stored in these maps while super_block->cleancache_poolid points to an entry in the map, so that on cleancache registration we can walk over all (if there are <= 32 of them, of course) cleancache-enabled super blocks and assign real pool ids. Actually, there is absolutely no need in these maps, because we can iterate over all super blocks immediately using iterate_supers. This is not racy, because cleancache_init_ops is called from mount_fs with super_block->s_umount held for writing, while iterate_supers takes this semaphore for reading, so if we call iterate_supers after setting cleancache_ops, all super blocks that had been created before cleancache_register_ops was called will be assigned pool ids by the action function of iterate_supers while all newer super blocks will receive it in cleancache_init_fs. This patch therefore removes the maps and hence the artificial limit on the number of cleancache enabled filesystems. Signed-off-by:
Vladimir Davydov <vdavydov@parallels.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com> Cc: David Vrabel <david.vrabel@citrix.com> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Stefan Hengelein <ilendir@googlemail.com> Cc: Florian Schmaus <fschmaus@gmail.com> Cc: Andor Daam <andor.daam@googlemail.com> Cc: Dan Magenheimer <dan.magenheimer@oracle.com> Cc: Bob Liu <lliubbo@gmail.com> Signed-off-by:
-
Vladimir Davydov authored
Currently, cleancache_register_ops returns the previous value of cleancache_ops to allow chaining. However, chaining, as it is implemented now, is extremely dangerous due to possible pool id collisions. Suppose, a new cleancache driver is registered after the previous one assigned an id to a super block. If the new driver assigns the same id to another super block, which is perfectly possible, we will have two different filesystems using the same id. No matter if the new driver implements chaining or not, we are likely to get data corruption with such a configuration eventually. This patch therefore disables the ability to override cleancache_ops altogether as potentially dangerous. If there is already cleancache driver registered, all further calls to cleancache_register_ops will return EBUSY. Since no user of cleancache implements chaining, we only need to make minor changes to the code outside the cleancache core. Signed-off-by:
-
Vladimir Davydov authored
Use super_block->s_uuid instead. Every shared filesystem using cleancache must now initialize super_block->s_uuid before calling cleancache_init_shared_fs. The only one on the tree, ocfs2, already meets this requirement. Signed-off-by:
-
Vladimir Davydov authored
Currently, maximal number of cleancache enabled filesystems equals 32, which is insufficient nowadays, because a Linux host can have hundreds of containers on board, each of which might want its own filesystem. This patch set targets at removing this limitation - see patch 4 for more details. Patches 1-3 prepare the code for this change. This patch (of 4): This will allow us to remove the uuid argument from cleancache_init_shared_fs. Signed-off-by:
-
Shachar Raindel authored
The do_wp_page function is extremely long. Extract the logic for handling a page belonging to a shared vma into a function of its own. This helps the readability of the code, without doing any functional change in it. Signed-off-by:
Shachar Raindel <raindel@mellanox.com> Acked-by:
Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by:
Rik van Riel <riel@redhat.com> Acked-by:
Andi Kleen <ak@linux.intel.com> Acked-by:
Haggai Eran <haggaie@mellanox.com> Acked-by:
-
Shachar Raindel authored
In some cases, do_wp_page had to copy the page suffering a write fault to a new location. If the function logic decided that to do this, it was done by jumping with a "goto" operation to the relevant code block. This made the code really hard to understand. It is also against the kernel coding style guidelines. This patch extracts the page copy and page table update logic to a separate function. It also clean up the naming, from "gotten" to "wp_page_copy", and adds few comments. Signed-off-by:
-
Shachar Raindel authored
When do_wp_page is ending, in several cases it needs to unlock the pages and ptls it was accessing. Currently, this logic was "called" by using a goto jump. This makes following the control flow of the function harder. Readability was further hampered by the unlock case containing large amount of logic needed only in one of the 3 cases. Using goto for cleanup is generally allowed. However, moving the trivial unlocking flows to the relevant call sites allow deeper refactoring in the next patch. Signed-off-by:
-
Shachar Raindel authored
Currently do_wp_page contains 265 code lines. It also contains 9 goto statements, of which 5 are targeting labels which are not cleanup related. This makes the function extremely difficult to understand. The following patches are an attempt at breaking the function to its basic components, and making it easier to understand. The patches are straight forward function extractions from do_wp_page. As we extract functions, we remove unneeded parameters and simplify the code as much as possible. However, the functionality is supposed to remain completely unchanged. The patches also attempt to document the functionality of each extracted function. In patch 2, we split the unlock logic to the contain logic relevant to specific needs of each use case, instead of having huge number of conditional decisions in a single unlock flow. This patch (of 4): When do_wp_page is ending, in several cases it needs to reuse the existing page. This is achieved by making the page table writable, and possibly updating the page-cache state. Currently, this logic was "called" by using a goto jump. This makes following the control flow of the function harder. It is also against the coding style guidelines for using goto. As the code can easily be refactored into a specialized function, refactor it out and simplify the code flow in do_wp_page. Acked-by:
-
Kirill A. Shutemov authored
CONFIG_PGTABLE_LEVELS is now available on every architecture and we can use it to check if we need to add nr_pmds into mm_struct. Signed-off-by:
Guenter Roeck <linux@roeck-us.net> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Matt Turner <mattst88@gmail.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Chris Metcalf <cmetcalf@ezchip.com> Cc: David Howells <dhowells@redhat.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Helge Deller <deller@gmx.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jeff Dike <jdike@addtoit.com> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Koichi Yasutake <yasutake.koichi@jp.panasonic.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Paul Mackerras <paulus@samba.org> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Richard Weinberger <richard@nod.at> Cc: Russell King <linux@arm.linux.org.uk> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Tony Luck <tony.luck@intel.com> Cc: Will Deacon <will.deacon@arm.com> Signed-off-by:
-
Kirill A. Shutemov authored
By this time all architectures which support more than two page table levels should be covered. This patch add default definiton of PGTABLE_LEVELS equal 2. We also add assert to detect inconsistence between CONFIG_PGTABLE_LEVELS and __PAGETABLE_PMD_FOLDED/__PAGETABLE_PUD_FOLDED. Signed-off-by:
-
Kirill A. Shutemov authored
We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. Signed-off-by:
-
Kirill A. Shutemov authored
We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. Signed-off-by:
Richard Weinberger <richard@nod.at> Cc: Jeff Dike <jdike@addtoit.com> Tested-by:
-
Kirill A. Shutemov authored
We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. Signed-off-by:
Chris Metcalf <cmetcalf@ezchip.com> Tested-by:
-
Kirill A. Shutemov authored
We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. Signed-off-by:
-
Kirill A. Shutemov authored
We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. Signed-off-by:
-
Kirill A. Shutemov authored
We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. Core mm expects __PAGETABLE_{PUD,PMD}_FOLDED to be defined if these page table levels folded. Signed-off-by: -
Kirill A. Shutemov authored
We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. Signed-off-by:
-
Kirill A. Shutemov authored
We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. Signed-off-by:
-
Kirill A. Shutemov authored
We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. Signed-off-by:
-
Kirill A. Shutemov authored
We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. Core mm expects __PAGETABLE_{PUD,PMD}_FOLDED to be defined if these page table levels folded. Signed-off-by: -
Kirill A. Shutemov authored
We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. We need to define PGTABLE_LEVELS before sourcing init/Kconfig: arch/Kconfig will define default value and it's sourced from init/Kconfig. Signed-off-by:
-
Kirill A. Shutemov authored
We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. Signed-off-by:
-
Kirill A. Shutemov authored
We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. ARM64_PGTABLE_LEVELS is renamed to PGTABLE_LEVELS and defined before sourcing init/Kconfig: arch/Kconfig will define default value and it's sourced from init/Kconfig. Signed-off-by:
Catalin Marinas <catalin.marinas@arm.com> Cc: Will Deacon <will.deacon@arm.com> Tested-by:
-
Kirill A. Shutemov authored
I've implemented accounting for pmd page tables as we have for pte (see mm->nr_ptes). It's requires a new counter in mm_struct: mm->nr_pmds. But the feature doesn't make any sense if an architecture has PMD level folded and it would be nice get rid of the counter in this case. The problem is that we cannot use __PAGETABLE_PMD_FOLDED in <linux/mm_types.h> due to circular dependencies: <linux/mm_types> -> <asm/pgtable.h> -> <linux/mm_types.h> In most cases <asm/pgtable.h> wants <linux/mm_types.h> to get definition of struct page and struct vm_area_struct. I've tried to split mm_struct into separate header file to be able to user <asm/pgtable.h> there. But it doesn't fly on some architectures, like ARM: it wants mm_struct <asm/pgtable.h> to implement tlb flushing. I don't see how to fix it without massive de-inlining or coverting a lot for inline functions to macros. This is other approach: expose number of page tables in use via Kconfig and use it in <linux/mm_types.h> instead of __PAGETABLE_PMD_FOLDED from <asm/pgtable.h>. This patch (of 19): We would want to use number of page table level to define mm_struct. Let's expose it as CONFIG_PGTABLE_LEVELS. Signed-off-by:
Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Matt Turner <mattst88@gmail.com> Tested-by:
-
Konstantin Khlebnikov authored
It seems nobody needs this. Signed-off-by:
Konstantin Khlebnikov <koct9i@gmail.com> Cc: Michal Hocko <mhocko@suse.cz> Cc: David Rientjes <rientjes@google.com> Signed-off-by:
-
Konstantin Khlebnikov authored
This makes show_mem() much less verbose on huge machines. Instead of huge and almost useless dump of counters for each per-zone per-cpu lists this patch prints the sum of these counters for each zone (free_pcp) and size of per-cpu list for current cpu (local_pcp). The filter flag SHOW_MEM_PERCPU_LISTS reverts to the old verbose mode. [akpm@linux-foundation.org: update show_free_areas comment] Signed-off-by:
Konstantin Khlebnikov <khlebnikov@yandex-team.ru> Acked-by:
-
Konstantin Khlebnikov authored
This patch replaces cancel_dirty_page() with a helper function account_page_cleaned() which only updates counters. It's called from truncate_complete_page() and from try_to_free_buffers() (hack for ext3). Page is locked in both cases, page-lock protects against concurrent dirtiers: see commit 2d6d7f98 ("mm: protect set_page_dirty() from ongoing truncation"). Delete_from_page_cache() shouldn't be called for dirty pages, they must be handled by caller (either written or truncated). This patch treats final dirty accounting fixup at the end of __delete_from_page_cache() as a debug check and adds WARN_ON_ONCE() around it. If something removes dirty pages without proper handling that might be a bug and unwritten data might be lost. Hugetlbfs has no dirty pages accounting, ClearPageDirty() is enough here. cancel_dirty_page() in nfs_wb_page_cancel() is redundant. This is helper for nfs_invalidate_page() and it's called only in case complete invalidation. The mess was started in v2.6.20 after commits 46d2277c ("Clean up and make try_to_free_buffers() not race with dirty pages") and 3e67c098 ("truncate: clear page dirtiness before running try_to_free_buffers()") first was reverted right in v2.6.20 in commit ecdfc978 ("Resurrect 'try_to_free_buffers()' VM hackery"), second in v2.6.25 commit a2b34564 ("Fix dirty page accounting leak with ext3 data=journal"). Custom fixes were introduced between these points. NFS in v2.6.23, commit 1b3b4a1a ("NFS: Fix a write request leak in nfs_invalidate_page()"). Kludge in __delete_from_page_cache() in v2.6.24, commit 3a692790 ("Do dirty page accounting when removing a page from the page cache"). Since v2.6.25 all of them are redundant. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by:
-
Ebru Akagunduz authored
This patch improves THP collapse rates, by allowing zero pages. Currently THP can collapse 4kB pages into a THP when there are up to khugepaged_max_ptes_none pte_none ptes in a 2MB range. This patch counts pte none and mapped zero pages with the same variable. The patch was tested with a program that allocates 800MB of memory, and performs interleaved reads and writes, in a pattern that causes some 2MB areas to first see read accesses, resulting in the zero pfn being mapped there. To simulate memory fragmentation at allocation time, I modified do_huge_pmd_anonymous_page to return VM_FAULT_FALLBACK for read faults. Without the patch, only %50 of the program was collapsed into THP and the percentage did not increase over time. With this patch after 10 minutes of waiting khugepaged had collapsed %99 of the program's memory. [aarcange@redhat.com: fix bogus BUG()] Signed-off-by:
Ebru Akagunduz <ebru.akagunduz@gmail.com> Reviewed-by:
Andrea Arcangeli <aarcange@redhat.com> Acked-by:
-
Joonsoo Kim authored
Compaction has anti fragmentation algorithm. It is that freepage should be more than pageblock order to finish the compaction if we don't find any freepage in requested migratetype buddy list. This is for mitigating fragmentation, but, there is a lack of migratetype consideration and it is too excessive compared to page allocator's anti fragmentation algorithm. Not considering migratetype would cause premature finish of compaction. For example, if allocation request is for unmovable migratetype, freepage with CMA migratetype doesn't help that allocation and compaction should not be stopped. But, current logic regards this situation as compaction is no longer needed, so finish the compaction. Secondly, condition is too excessive compared to page allocator's logic. We can steal freepage from other migratetype and change pageblock migratetype on more relaxed conditions in page allocator. This is designed to prevent fragmentation and we can use it here. Imposing hard constraint only to the compaction doesn't help much in this case since page allocator would cause fragmentation again. To solve these problems, this patch borrows anti fragmentation logic from page allocator. It will reduce premature compaction finish in some cases and reduce excessive compaction work. stress-highalloc test in mmtests with non movable order 7 allocation shows considerable increase of compaction success rate. Compaction success rate (Compaction success * 100 / Compaction stalls, %) 31.82 : 42.20 I tested it on non-reboot 5 runs stress-highalloc benchmark and found that there is no more degradation on allocation success rate than before. That roughly means that this patch doesn't result in more fragmentations. Vlastimil suggests additional idea that we only test for fallbacks when migration scanner has scanned a whole pageblock. It looked good for fragmentation because chance of stealing increase due to making more free pages in certain pageblock. So, I tested it, but, it results in decreased compaction success rate, roughly 38.00. I guess the reason that if system is low memory condition, watermark check could be failed due to not enough order 0 free page and so, sometimes, we can't reach a fallback check although migrate_pfn is aligned to pageblock_nr_pages. I can insert code to cope with this situation but it makes code more complicated so I don't include his idea at this patch. [akpm@linux-foundation.org: fix CONFIG_CMA=n build] Signed-off-by:
Joonsoo Kim <iamjoonsoo.kim@lge.com> Acked-by:
-
Joonsoo Kim authored
This is preparation step to use page allocator's anti fragmentation logic in compaction. This patch just separates fallback freepage checking part from fallback freepage management part. Therefore, there is no functional change. Signed-off-by:
-
Joonsoo Kim authored
Freepage with MIGRATE_CMA can be used only for MIGRATE_MOVABLE and they should not be expanded to other migratetype buddy list to protect them from unmovable/reclaimable allocation. Implementing these requirements in __rmqueue_fallback(), that is, finding largest possible block of freepage has bad effect that high order freepage with MIGRATE_CMA are broken continually although there are suitable order CMA freepage. Reason is that they are not be expanded to other migratetype buddy list and next __rmqueue_fallback() invocation try to finds another largest block of freepage and break it again. So, MIGRATE_CMA fallback should be handled separately. This patch introduces __rmqueue_cma_fallback(), that just wrapper of __rmqueue_smallest() and call it before __rmqueue_fallback() if migratetype == MIGRATE_MOVABLE. This results in unintended behaviour change that MIGRATE_CMA freepage is always used first rather than other migratetype as movable allocation's fallback. But, as already mentioned above, MIGRATE_CMA can be used only for MIGRATE_MOVABLE, so it is better to use MIGRATE_CMA freepage first as much as possible. Otherwise, we needlessly take up precious freepages with other migratetype and increase chance of fragmentation. Signed-off-by:
-
