)]}' { "log": [ { "commit": "912985dce45ef18fcdd9f5439fef054e0e22302a", "tree": "394b3be51134bddef200f86dde48aa3e3c7ae94c", "parents": [ "40c42076ebd362dc69210cccea101ac80b6d4bd4" ], "author": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Tue Aug 12 17:52:52 2008 -0500" }, "committer": { "name": "Rusty Russell", "email": "rusty@rustcorp.com.au", "time": "Tue Aug 12 17:52:53 2008 +1000" }, "message": "mm: Make generic weak get_user_pages_fast and EXPORT_GPL it\n\nOut of line get_user_pages_fast fallback implementation, make it a weak\nsymbol, get rid of CONFIG_HAVE_GET_USER_PAGES_FAST.\n\nExport the symbol to modules so lguest can use it.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Rusty Russell \u003crusty@rustcorp.com.au\u003e\n" }, { "commit": "cddb8a5c14aa89810b40495d94d3d2a0faee6619", "tree": "d0b47b071f7d2dd1d6f9c36084aa8cfcef90d1da", "parents": [ "7906d00cd1f687268f0a3599442d113767795ae6" ], "author": { "name": "Andrea Arcangeli", "email": "andrea@qumranet.com", "time": "Mon Jul 28 15:46:29 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Jul 28 16:30:21 2008 -0700" }, "message": "mmu-notifiers: core\n\nWith KVM/GFP/XPMEM there isn\u0027t just the primary CPU MMU pointing to pages.\n There are secondary MMUs (with secondary sptes and secondary tlbs) too.\nsptes in the kvm case are shadow pagetables, but when I say spte in\nmmu-notifier context, I mean \"secondary pte\". In GRU case there\u0027s no\nactual secondary pte and there\u0027s only a secondary tlb because the GRU\nsecondary MMU has no knowledge about sptes and every secondary tlb miss\nevent in the MMU always generates a page fault that has to be resolved by\nthe CPU (this is not the case of KVM where the a secondary tlb miss will\nwalk sptes in hardware and it will refill the secondary tlb transparently\nto software if the corresponding spte is present). The same way\nzap_page_range has to invalidate the pte before freeing the page, the spte\n(and secondary tlb) must also be invalidated before any page is freed and\nreused.\n\nCurrently we take a page_count pin on every page mapped by sptes, but that\nmeans the pages can\u0027t be swapped whenever they\u0027re mapped by any spte\nbecause they\u0027re part of the guest working set. Furthermore a spte unmap\nevent can immediately lead to a page to be freed when the pin is released\n(so requiring the same complex and relatively slow tlb_gather smp safe\nlogic we have in zap_page_range and that can be avoided completely if the\nspte unmap event doesn\u0027t require an unpin of the page previously mapped in\nthe secondary MMU).\n\nThe mmu notifiers allow kvm/GRU/XPMEM to attach to the tsk-\u003emm and know\nwhen the VM is swapping or freeing or doing anything on the primary MMU so\nthat the secondary MMU code can drop sptes before the pages are freed,\navoiding all page pinning and allowing 100% reliable swapping of guest\nphysical address space. Furthermore it avoids the code that teardown the\nmappings of the secondary MMU, to implement a logic like tlb_gather in\nzap_page_range that would require many IPI to flush other cpu tlbs, for\neach fixed number of spte unmapped.\n\nTo make an example: if what happens on the primary MMU is a protection\ndowngrade (from writeable to wrprotect) the secondary MMU mappings will be\ninvalidated, and the next secondary-mmu-page-fault will call\nget_user_pages and trigger a do_wp_page through get_user_pages if it\ncalled get_user_pages with write\u003d1, and it\u0027ll re-establishing an updated\nspte or secondary-tlb-mapping on the copied page. Or it will setup a\nreadonly spte or readonly tlb mapping if it\u0027s a guest-read, if it calls\nget_user_pages with write\u003d0. This is just an example.\n\nThis allows to map any page pointed by any pte (and in turn visible in the\nprimary CPU MMU), into a secondary MMU (be it a pure tlb like GRU, or an\nfull MMU with both sptes and secondary-tlb like the shadow-pagetable layer\nwith kvm), or a remote DMA in software like XPMEM (hence needing of\nschedule in XPMEM code to send the invalidate to the remote node, while no\nneed to schedule in kvm/gru as it\u0027s an immediate event like invalidating\nprimary-mmu pte).\n\nAt least for KVM without this patch it\u0027s impossible to swap guests\nreliably. And having this feature and removing the page pin allows\nseveral other optimizations that simplify life considerably.\n\nDependencies:\n\n1) mm_take_all_locks() to register the mmu notifier when the whole VM\n isn\u0027t doing anything with \"mm\". This allows mmu notifier users to keep\n track if the VM is in the middle of the invalidate_range_begin/end\n critical section with an atomic counter incraese in range_begin and\n decreased in range_end. No secondary MMU page fault is allowed to map\n any spte or secondary tlb reference, while the VM is in the middle of\n range_begin/end as any page returned by get_user_pages in that critical\n section could later immediately be freed without any further\n -\u003einvalidate_page notification (invalidate_range_begin/end works on\n ranges and -\u003einvalidate_page isn\u0027t called immediately before freeing\n the page). To stop all page freeing and pagetable overwrites the\n mmap_sem must be taken in write mode and all other anon_vma/i_mmap\n locks must be taken too.\n\n2) It\u0027d be a waste to add branches in the VM if nobody could possibly\n run KVM/GRU/XPMEM on the kernel, so mmu notifiers will only enabled if\n CONFIG_KVM\u003dm/y. In the current kernel kvm won\u0027t yet take advantage of\n mmu notifiers, but this already allows to compile a KVM external module\n against a kernel with mmu notifiers enabled and from the next pull from\n kvm.git we\u0027ll start using them. And GRU/XPMEM will also be able to\n continue the development by enabling KVM\u003dm in their config, until they\n submit all GRU/XPMEM GPLv2 code to the mainline kernel. Then they can\n also enable MMU_NOTIFIERS in the same way KVM does it (even if KVM\u003dn).\n This guarantees nobody selects MMU_NOTIFIER\u003dy if KVM and GRU and XPMEM\n are all \u003dn.\n\nThe mmu_notifier_register call can fail because mm_take_all_locks may be\ninterrupted by a signal and return -EINTR. Because mmu_notifier_reigster\nis used when a driver startup, a failure can be gracefully handled. Here\nan example of the change applied to kvm to register the mmu notifiers.\nUsually when a driver startups other allocations are required anyway and\n-ENOMEM failure paths exists already.\n\n struct kvm *kvm_arch_create_vm(void)\n {\n struct kvm *kvm \u003d kzalloc(sizeof(struct kvm), GFP_KERNEL);\n+ int err;\n\n if (!kvm)\n return ERR_PTR(-ENOMEM);\n\n INIT_LIST_HEAD(\u0026kvm-\u003earch.active_mmu_pages);\n\n+ kvm-\u003earch.mmu_notifier.ops \u003d \u0026kvm_mmu_notifier_ops;\n+ err \u003d mmu_notifier_register(\u0026kvm-\u003earch.mmu_notifier, current-\u003emm);\n+ if (err) {\n+ kfree(kvm);\n+ return ERR_PTR(err);\n+ }\n+\n return kvm;\n }\n\nmmu_notifier_unregister returns void and it\u0027s reliable.\n\nThe patch also adds a few needed but missing includes that would prevent\nkernel to compile after these changes on non-x86 archs (x86 didn\u0027t need\nthem by luck).\n\n[akpm@linux-foundation.org: coding-style fixes]\n[akpm@linux-foundation.org: fix mm/filemap_xip.c build]\n[akpm@linux-foundation.org: fix mm/mmu_notifier.c build]\nSigned-off-by: Andrea Arcangeli \u003candrea@qumranet.com\u003e\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Christoph Lameter \u003ccl@linux-foundation.org\u003e\nCc: Jack Steiner \u003csteiner@sgi.com\u003e\nCc: Robin Holt \u003cholt@sgi.com\u003e\nCc: Nick Piggin \u003cnpiggin@suse.de\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Kanoj Sarcar \u003ckanojsarcar@yahoo.com\u003e\nCc: Roland Dreier \u003crdreier@cisco.com\u003e\nCc: Steve Wise \u003cswise@opengridcomputing.com\u003e\nCc: Avi Kivity \u003cavi@qumranet.com\u003e\nCc: Hugh Dickins \u003chugh@veritas.com\u003e\nCc: Rusty Russell \u003crusty@rustcorp.com.au\u003e\nCc: Anthony Liguori \u003caliguori@us.ibm.com\u003e\nCc: Chris Wright \u003cchrisw@redhat.com\u003e\nCc: Marcelo Tosatti \u003cmarcelo@kvack.org\u003e\nCc: Eric Dumazet \u003cdada1@cosmosbay.com\u003e\nCc: \"Paul E. McKenney\" \u003cpaulmck@us.ibm.com\u003e\nCc: Izik Eidus \u003cizike@qumranet.com\u003e\nCc: Anthony Liguori \u003caliguori@us.ibm.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "8174c430e445a93016ef18f717fe570214fa38bf", "tree": "f1b4426eae7401425e9102c7b3e141be86f0930c", "parents": [ "21cc199baa815d7b3f1ace4be20b9558cbddc00f" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Fri Jul 25 19:45:24 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Jul 26 12:00:06 2008 -0700" }, "message": "x86: lockless get_user_pages_fast()\n\nImplement get_user_pages_fast without locking in the fastpath on x86.\n\nDo an optimistic lockless pagetable walk, without taking mmap_sem or any\npage table locks or even mmap_sem. Page table existence is guaranteed by\nturning interrupts off (combined with the fact that we\u0027re always looking\nup the current mm, means we can do the lockless page table walk within the\nconstraints of the TLB shootdown design). Basically we can do this\nlockless pagetable walk in a similar manner to the way the CPU\u0027s pagetable\nwalker does not have to take any locks to find present ptes.\n\nThis patch (combined with the subsequent ones to convert direct IO to use\nit) was found to give about 10% performance improvement on a 2 socket 8\ncore Intel Xeon system running an OLTP workload on DB2 v9.5\n\n \"To test the effects of the patch, an OLTP workload was run on an IBM\n x3850 M2 server with 2 processors (quad-core Intel Xeon processors at\n 2.93 GHz) using IBM DB2 v9.5 running Linux 2.6.24rc7 kernel. Comparing\n runs with and without the patch resulted in an overall performance\n benefit of ~9.8%. Correspondingly, oprofiles showed that samples from\n __up_read and __down_read routines that is seen during thread contention\n for system resources was reduced from 2.8% down to .05%. Monitoring the\n /proc/vmstat output from the patched run showed that the counter for\n fast_gup contained a very high number while the fast_gup_slow value was\n zero.\"\n\n(fast_gup is the old name for get_user_pages_fast, fast_gup_slow is a\ncounter we had for the number of times the slowpath was invoked).\n\nThe main reason for the improvement is that DB2 has multiple threads each\nissuing direct-IO. Direct-IO uses get_user_pages, and thus the threads\ncontend the mmap_sem cacheline, and can also contend on page table locks.\n\nI would anticipate larger performance gains on larger systems, however I\nthink DB2 uses an adaptive mix of threads and processes, so it could be\nthat thread contention remains pretty constant as machine size increases.\nIn which case, we stuck with \"only\" a 10% gain.\n\nThe downside of using get_user_pages_fast is that if there is not a pte\nwith the correct permissions for the access, we end up falling back to\nget_user_pages and so the get_user_pages_fast is a bit of extra work.\nHowever this should not be the common case in most performance critical\ncode.\n\n[akpm@linux-foundation.org: coding-style fixes]\n[akpm@linux-foundation.org: build fix]\n[akpm@linux-foundation.org: Kconfig fix]\n[akpm@linux-foundation.org: Makefile fix/cleanup]\n[akpm@linux-foundation.org: warning fix]\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nCc: Dave Kleikamp \u003cshaggy@austin.ibm.com\u003e\nCc: Andy Whitcroft \u003capw@shadowen.org\u003e\nCc: Ingo Molnar \u003cmingo@elte.hu\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: Andi Kleen \u003candi@firstfloor.org\u003e\nCc: Dave Kleikamp \u003cshaggy@austin.ibm.com\u003e\nCc: Badari Pulavarty \u003cpbadari@us.ibm.com\u003e\nCc: Zach Brown \u003czach.brown@oracle.com\u003e\nCc: Jens Axboe \u003cjens.axboe@oracle.com\u003e\nReviewed-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "83d1674a946141c3c59d430e96c224f7937e6158", "tree": "03420c9fdf56ad35de685b7c0b48899d886bd7ff", "parents": [ "9ca908f47bc784c90e17a553ce33e756c73feac4" ], "author": { "name": "Gerald Schaefer", "email": "gerald.schaefer@de.ibm.com", "time": "Wed Jul 23 21:28:22 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Jul 24 10:47:21 2008 -0700" }, "message": "mm: make CONFIG_MIGRATION available w/o CONFIG_NUMA\n\nWe\u0027d like to support CONFIG_MEMORY_HOTREMOVE on s390, which depends on\nCONFIG_MIGRATION. So far, CONFIG_MIGRATION is only available with NUMA\nsupport.\n\nThis patch makes CONFIG_MIGRATION selectable for architectures that define\nARCH_ENABLE_MEMORY_HOTREMOVE. When MIGRATION is enabled w/o NUMA, the\nkernel won\u0027t compile because migrate_vmas() does not know about\nvm_ops-\u003emigrate() and vma_migratable() does not know about policy_zone.\nTo fix this, those two functions can be restricted to \u0027#ifdef CONFIG_NUMA\u0027\nbecause they are not being used w/o NUMA. vma_migratable() is moved over\nfrom migrate.h to mempolicy.h.\n\n[kosaki.motohiro@jp.fujitsu.com: build fix]\nAcked-by: Christoph Lameter \u003ccl@linux-foundation.org\u003e\nSigned-off-by: Gerald Schaefer \u003cgerald.schaefer@de.ibm.com\u003e\nCc: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nCc: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nSigned-off-by: KOSAKI Motorhiro \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "6c118e43dc513a7118b49b9ff953fe61e14515dc", "tree": "ef654ba1982c17a516697d25b3d0405acb6aeac1", "parents": [ "847106ff628805e1a0aa91e7f53381f3fdfcd839", "72f6befeea7dc634a83219287d5b874734b85637" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Jul 14 13:37:29 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Jul 14 13:37:29 2008 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/hskinnemoen/avr32-2.6\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/hskinnemoen/avr32-2.6: (31 commits)\n avr32: Fix typo of IFSR in a comment in the PIO header file\n avr32: Power Management support (\"standby\" and \"mem\" modes)\n avr32: Add system device for the internal interrupt controller (intc)\n avr32: Add simple SRAM allocator\n avr32: Enable SDRAMC clock at startup\n rtc-at32ap700x: Enable wakeup\n macb: Basic suspend/resume support\n atmel_serial: Drain console TX shifter before suspending\n atmel_serial: Fix build on avr32 with CONFIG_PM enabled\n avr32: Use a quicklist for PTE allocation as well\n avr32: Use a quicklist for PGD allocation\n avr32: Cover the kernel page tables in the user PGDs\n avr32: Store virtual addresses in the PGD\n avr32: Remove useless zeroing of swapper_pg_dir at startup\n avr32: Clean up and optimize the TLB operations\n avr32: Rename at32ap.c -\u003e pdc.c\n avr32: Move setup_platform() into chip-specific file\n avr32: Kill special exception handler sections\n avr32: Kill unneeded #include \u003casm/pgalloc.h\u003e from asm/mmu_context.h\n avr32: Clean up time.c #includes\n ...\n" }, { "commit": "421c175c4d609864350df495b34d3e99f9fb1bdd", "tree": "ea3ade04452a6a6c578ae7895b5451c1bf231bc9", "parents": [ "0788fea4d583a3b7d199696819940ff3387d79a3" ], "author": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Mon Jul 14 09:59:18 2008 +0200" }, "committer": { "name": "Heiko Carstens", "email": "heiko.carstens@de.ibm.com", "time": "Mon Jul 14 10:02:16 2008 +0200" }, "message": "[S390] Add support for memory hot-add.\n\nCc: Gerald Schaefer \u003cgerald.schaefer@de.ibm.com\u003e\nSigned-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nSigned-off-by: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\n" }, { "commit": "38510754a50192a072210e24fdc4ae65592182f0", "tree": "f14de78921e4e6a5f56aee93d5ba921c57c2a82f", "parents": [ "5a4d5292779b6163aa41e594a56307e442fbe73c" ], "author": { "name": "Haavard Skinnemoen", "email": "haavard.skinnemoen@atmel.com", "time": "Mon Jan 14 23:35:32 2008 +0100" }, "committer": { "name": "Haavard Skinnemoen", "email": "haavard.skinnemoen@atmel.com", "time": "Wed Jul 02 11:01:29 2008 +0200" }, "message": "avr32: Use a quicklist for PTE allocation as well\n\nUsing a quicklist to allocate PTEs might be slightly faster than using\nthe page allocator directly since we might avoid zeroing the page\nafter each allocation.\n\nSigned-off-by: Haavard Skinnemoen \u003chaavard.skinnemoen@atmel.com\u003e\n" }, { "commit": "e20b8cca760ed2a6abcfe37ef56f2306790db648", "tree": "85e5610364f73da193ee781be61710fddfbe045f", "parents": [ "97965478a66fbdf0f4ad5e4ecc4828f0cb548a45" ], "author": { "name": "Christoph Lameter", "email": "clameter@sgi.com", "time": "Mon Apr 28 02:12:55 2008 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Apr 28 08:58:22 2008 -0700" }, "message": "PAGEFLAGS_EXTENDED and separate page flags for Head and Tail\n\nHaving separate page flags for the head and the tail of a compound page allows\nthe compiler to use bitops instead of operations on a word to check for a tail\npage. That is f.e. important for virt_to_head_page() which is used in\nvarious critical code paths (kfree for example):\n\nCode for PageTail(page)\n\nBefore:\n\n mov (%rdi),%rdx\t\tpage-\u003eflags\n mov %rdx,%rax\t\t3 bytes\n and $0x12000,%eax\t\t5 bytes\n cmp $0x12000,%rax\t\t6 bytes\n je 897 \u003ckfree+0xa7\u003e\n\nAfter:\n\n mov (%rdi),%rax\n test $0x40,%ah\t\t\t(3 bytes)\n jne 887 \u003ckfree+0x97\u003e\n\nSo we go from 14 bytes to 3 bytes and from 3 instructions to one. From the\nuse of 2 registers we go to none.\n\nWe can only use page flags for this if we have page flags available. This\npatch introduces CONFIG_PAGEFLAGS_EXTENDED that is set if pageflags are not\nscarce due to SPARSEMEM using page flags for its sectionid on 32 bit NUMA\nplatforms.\n\nAdditional page flag definitions can be added to the CONFIG_PAGEFLAGS_EXTENDED\nsection in page-flags.h if the functionality depends on PAGEFLAGS_EXTENDED or\nif more page flag overlapping tricks are used for the !PAGEFLAGS_EXTENDED\nfallback (the upcoming virtual compound patch may hook in here and Rik\u0027s/Lee\u0027s\nadditional page flags to solve the reclaim issues could also be added there\n[hint... hint... where are these patchsets?]).\n\nAvoiding the overlaying of Pg_reclaim also clears the way for possible use of\ncompound pages for the pagecache or on the LRU.\n\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nCc: Nick Piggin \u003cnickpiggin@yahoo.com.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "d5f68c6dbda8e63df258a0c639f03d7565cf1d50", "tree": "490df8e0dcdfd38e83754711d6db6ce0d055b121", "parents": [ "29e0d209b38479e6817318a971082421a7d630b7" ], "author": { "name": "Paul Mundt", "email": "lethal@linux-sh.org", "time": "Thu Nov 22 16:28:26 2007 +0900" }, "committer": { "name": "Paul Mundt", "email": "lethal@linux-sh.org", "time": "Mon Jan 28 13:18:55 2008 +0900" }, "message": "sh: Bump number of quicklists for SH-5.\n\nSync up with the SH definitions.\n\nSigned-off-by: Paul Mundt \u003clethal@linux-sh.org\u003e\n" }, { "commit": "a5ee6daa525c04079baee6f393c0b2dab3f61253", "tree": "5024bdedd14111e3cf6723a41390b2f80dec8bfd", "parents": [ "2e12a7fb0d79d011ff9e0b09b53ca4438e5604de" ], "author": { "name": "Geoff Levand", "email": "geoffrey.levand@am.sony.com", "time": "Mon Dec 17 16:19:53 2007 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Mon Dec 17 19:28:16 2007 -0800" }, "message": "sparsemem: make SPARSEMEM_VMEMMAP selectable\n\nSPARSEMEM_VMEMMAP needs to be a selectable config option to support\nbuilding the kernel both with and without sparsemem vmemmap support. This\nselection is desirable for platforms which could be configured one way for\nplatform specific builds and the other for multi-platform builds.\n\nSigned-off-by: Miguel Botón \u003cmboton@gmail.com\u003e\nSigned-off-by: Geoff Levand \u003cgeoffrey.levand@am.sony.com\u003e\nAcked-by: Yasunori Goto \u003cy-goto@jp.fujitsu.com\u003e\nCc: Christoph Lameter \u003cclameter@sgi.com\u003e\nCc: Andy Whitcroft \u003capw@shadowen.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "ad3d0a3827a3ce45ee4141de81be7375157b42de", "tree": "47c96799b8108e57dd24036beae314441eb8822f", "parents": [ "2de206d8f09c8b01d3721f860c28258838953a18" ], "author": { "name": "Philipp Marek", "email": "philipp.marek@bmlv.gv.at", "time": "Sat Oct 20 02:46:58 2007 +0200" }, "committer": { "name": "Adrian Bunk", "email": "bunk@kernel.org", "time": "Sat Oct 20 02:46:58 2007 +0200" }, "message": "small documentation fixes\n\nSigned-off-by: Adrian Bunk \u003cbunk@kernel.org\u003e\n" }, { "commit": "fb9fc395174138983a49f2da982ed14caabbe741", "tree": "5d5d3643ee6853a899205613da272cc343fdc1a4", "parents": [ "0eafaae84e21ac033815cc9f33c3ae889cd7ccfe", "ace2e92e193126711cb3a83a3752b2c5b8396950" ], "author": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Wed Oct 17 11:10:11 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Wed Oct 17 11:10:11 2007 -0700" }, "message": "Merge branch \u0027xen-upstream\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jeremy/xen\n\n* \u0027xen-upstream\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jeremy/xen:\n xfs: eagerly remove vmap mappings to avoid upsetting Xen\n xen: add some debug output for failed multicalls\n xen: fix incorrect vcpu_register_vcpu_info hypercall argument\n xen: ask the hypervisor how much space it needs reserved\n xen: lock pte pages while pinning/unpinning\n xen: deal with stale cr3 values when unpinning pagetables\n xen: add batch completion callbacks\n xen: yield to IPI target if necessary\n Clean up duplicate includes in arch/i386/xen/\n remove dead code in pgtable_cache_init\n paravirt: clean up lazy mode handling\n paravirt: refactor struct paravirt_ops into smaller pv_*_ops\n" }, { "commit": "74260714c56de4f967fcb2f17a8656bc574b75be", "tree": "f02bcd991285a20a543fae69f916577c8447b8f4", "parents": [ "9f79991d4186089e228274196413572cc000143b" ], "author": { "name": "Jeremy Fitzhardinge", "email": "jeremy@xensource.com", "time": "Tue Oct 16 11:51:30 2007 -0700" }, "committer": { "name": "Jeremy Fitzhardinge", "email": "jeremy@goop.org", "time": "Tue Oct 16 11:51:30 2007 -0700" }, "message": "xen: lock pte pages while pinning/unpinning\n\nWhen a pagetable is created, it is made globally visible in the rmap\nprio tree before it is pinned via arch_dup_mmap(), and remains in the\nrmap tree while it is unpinned with arch_exit_mmap().\n\nThis means that other CPUs may race with the pinning/unpinning\nprocess, and see a pte between when it gets marked RO and actually\npinned, causing any pte updates to fail with write-protect faults.\n\nAs a result, all pte pages must be properly locked, and only unlocked\nonce the pinning/unpinning process has finished.\n\nIn order to avoid taking spinlocks for the whole pagetable - which may\noverflow the PREEMPT_BITS portion of preempt counter - it locks and pins\neach pte page individually, and then finally pins the whole pagetable.\n\nSigned-off-by: Jeremy Fitzhardinge \u003cjeremy@xensource.com\u003e\nCc: Rik van Riel \u003criel@redhat.com\u003e\nCc: Hugh Dickens \u003chugh@veritas.com\u003e\nCc: David Rientjes \u003crientjes@google.com\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Andi Kleen \u003cak@suse.de\u003e\nCc: Keir Fraser \u003ckeir@xensource.com\u003e\nCc: Jan Beulich \u003cjbeulich@novell.com\u003e\n\n" }, { "commit": "0c0e6195896535481173df98935ad8db174f4d45", "tree": "2b35d3b81ba54b5d38e691d2a2019f4bcdfd1dce", "parents": [ "a5d76b54a3f3a40385d7f76069a2feac9f1bad63" ], "author": { "name": "KAMEZAWA Hiroyuki", "email": "kamezawa.hiroyu@jp.fujitsu.com", "time": "Tue Oct 16 01:26:12 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Oct 16 09:43:02 2007 -0700" }, "message": "memory unplug: page offline\n\nLogic.\n - set all pages in [start,end) as isolated migration-type.\n by this, all free pages in the range will be not-for-use.\n - Migrate all LRU pages in the range.\n - Test all pages in the range\u0027s refcnt is zero or not.\n\nTodo:\n - allocate migration destination page from better area.\n - confirm page_count(page)\u003d\u003d 0 \u0026\u0026 PageReserved(page) page is safe to be freed..\n (I don\u0027t like this kind of page but..\n - Find out pages which cannot be migrated.\n - more running tests.\n - Use reclaim for unplugging other memory type area.\n\nSigned-off-by: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nSigned-off-by: Yasunori Goto \u003cy-goto@jp.fujitsu.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "29c71111d0557385328211b130246a90f9223b46", "tree": "5588a49ee548d38e15bd7541cec29e069b9e457c", "parents": [ "8f6aac419bd590f535fb110875a51f7db2b62b5b" ], "author": { "name": "Andy Whitcroft", "email": "apw@shadowen.org", "time": "Tue Oct 16 01:24:14 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Oct 16 09:42:51 2007 -0700" }, "message": "vmemmap: generify initialisation via helpers\n\nConvert the common vmemmap population into initialisation helpers for use by\narchitecture vmemmap populators. All architecture implementing the\nSPARSEMEM_VMEMMAP variant supply an architecture specific vmemmap_populate()\ninitialiser, which may make use of the helpers.\n\nThis allows us to clean up and remove the initialisation Kconfig entries.\nWith this patch there is a single SPARSEMEM_VMEMMAP_ENABLE Kconfig option to\nindicate use of that variant.\n\nSigned-off-by: Andy Whitcroft \u003capw@shadowen.org\u003e\nAcked-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "67dd5a25f4efbfccf973159429cb20acdc5b0e0e", "tree": "9965b12c0d614b7cf82dd8483cfc1671f925af4e", "parents": [ "9f34073b4e54ad58541e0e2b4a87f4f6c1460e21" ], "author": { "name": "Jeremy Fitzhardinge", "email": "jeremy@goop.org", "time": "Fri Oct 05 17:19:35 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Sat Oct 06 09:31:30 2007 -0700" }, "message": "xen: disable split pte locks for now\n\nWhen pinning and unpinning pagetables, we must protect them against\nbeing used by other CPUs, lest they see the pagetable in an\nintermediate read-only-but-not-pinned state.\n\nWhen using split pte locks, doing this properly would require taking\nall the pte locks for the pagetable while pinning, but this may overflow\nthe PREEMPT_BITS part of the preempt counter if the process has mapped\nmore than about 512M of memory.\n\nHowever, failing to take the pte locks causes write-protect faults when\nthe pageout code is trying to clear the Access bit on a pte which is part\nof a freshy created and still being pinned process after fork.\n\nThis is a short-term fix until the problem is solved properly.\n\nSigned-off-by: Jeremy Fitzhardinge \u003cjeremy@xensource.com\u003e\nAcked-by: Rik van Riel \u003criel@redhat.com\u003e\nAcked-by: Hugh Dickins \u003chugh@veritas.com\u003e\nCc: David Rientjes \u003crientjes@google.com\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Andi Kleen \u003cak@suse.de\u003e\nCc: Keir Fraser \u003ckeir@xensource.com\u003e\nCc: Jan Beulich \u003cjbeulich@novell.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b0cb1a19d05b8ea8611a9ef48a17fe417f1832e6", "tree": "895fe53e8dc4fc59d05b963ac079f6ff759ad0fb", "parents": [ "6c8dca5d53f95009d4fff00195bf38f277dc4366" ], "author": { "name": "Rafael J. Wysocki", "email": "rjw@sisk.pl", "time": "Sun Jul 29 23:24:36 2007 +0200" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Sun Jul 29 16:45:38 2007 -0700" }, "message": "Replace CONFIG_SOFTWARE_SUSPEND with CONFIG_HIBERNATION\n\nReplace CONFIG_SOFTWARE_SUSPEND with CONFIG_HIBERNATION to avoid\nconfusion (among other things, with CONFIG_SUSPEND introduced in the\nnext patch).\n\nSigned-off-by: Rafael J. Wysocki \u003crjw@sisk.pl\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "2a7326b5bbafac4c96bcdb944b2a773593030b96", "tree": "25bc49eadea73cf2133198963d1baf3f5def7316", "parents": [ "831441862956fffa17b9801db37e6ea1650b0f69" ], "author": { "name": "Christoph Lameter", "email": "clameter@sgi.com", "time": "Tue Jul 17 04:03:37 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue Jul 17 10:23:02 2007 -0700" }, "message": "CONFIG_BOUNCE to avoid useless inclusion of bounce buffer logic\n\nThe bounce buffer logic is included on systems that do not need it. If a\nsystem does not have zones like ZONE_DMA and ZONE_HIGHMEM that can lead to\nthe use of bounce buffers then there is no need to reserve memory pools etc\netc. This is true f.e. for SGI Altix.\n\nAlso nicifies the Makefile and gets rid of the tricky \"and\" there.\n\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nAcked-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b91cba52e9b7b3f1c0037908a192d93a869ca9e5", "tree": "bbce7f323c8f52b308af5a152673a75b3e445360", "parents": [ "98283bb49c6c8c070ebde9f47489d3e9a83c1323", "e509ac4bbc661052dc73a2e8138800ba77d4ecb9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Mon Jul 16 10:32:02 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Mon Jul 16 10:32:02 2007 -0700" }, "message": "Merge master.kernel.org:/pub/scm/linux/kernel/git/lethal/sh-2.6\n\n* master.kernel.org:/pub/scm/linux/kernel/git/lethal/sh-2.6: (68 commits)\n sh: sh-rtc support for SH7709.\n sh: Revert __xdiv64_32 size change.\n sh: Update r7785rp defconfig.\n sh: Export div symbols for GCC 4.2 and ST GCC.\n sh: fix race in parallel out-of-tree build\n sh: Kill off dead mach.c for hp6xx.\n sh: hd64461.h cleanup and added comments.\n sh: Update the alignment when 4K stacks are used.\n sh: Add a .bss.page_aligned section for 4K stacks.\n sh: Don\u0027t let SH-4A clobber SH-4 CFLAGS.\n sh: Add parport stub for SuperIO ports.\n sh: Drop -Wa,-dsp for DSP tuning.\n sh: Update dreamcast defconfig.\n fb: pvr2fb: A few more __devinit annotations for PCI.\n fb: pvr2fb: Fix up section mismatch warnings.\n sh: Select IPR-IRQ for SH7091.\n sh: Correct __xdiv64_32/div64_32 return value size.\n sh: Fix timer-tmu build for SH-3.\n sh: Add cpu and mach links to CLEAN_FILES.\n sh: Preliminary support for the SH-X3 CPU.\n ...\n" }, { "commit": "f057eac0d7ad967138390a9dd7fd8267e1e39d19", "tree": "96e951adb2934ee4495edda09f94c67c02fcf5ab", "parents": [ "693783817a79d8619335e2bf1a33de73cf189864" ], "author": { "name": "Stephen Rothwell", "email": "sfr@canb.auug.org.au", "time": "Sun Jul 15 23:40:05 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Mon Jul 16 09:05:42 2007 -0700" }, "message": "Introduce CONFIG_VIRT_TO_BUS\n\nMake some offending drivers depend on it and set CONFIG_ARCH_NO_VIRT_TO_BUS\nfor ppc64 so that we don\u0027t build those drivers.\n\nThis gets PowerPC allmodconfig and allyesconfig much closer to building.\n\nSigned-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nCc: Al Viro \u003cviro@ftp.linux.org.uk\u003e\nAcked-by: David Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "33d63bd83bf9aa6b662a376a96b825acba721e8f", "tree": "469eadf218a6e435b06eb151c88340c0adb7b83e", "parents": [ "05a117847b43d44f336bbf272a1063661431a5e5" ], "author": { "name": "Paul Mundt", "email": "lethal@linux-sh.org", "time": "Thu Jun 07 11:32:52 2007 +0900" }, "committer": { "name": "Paul Mundt", "email": "lethal@hera.kernel.org", "time": "Fri Jun 08 02:43:51 2007 +0000" }, "message": "sh: memory hot-add for sparsemem users support.\n\nThis enables simple hotplug support for sparsemem users. Presently\nthis only permits memory being added in to node 0 on ZONE_NORMAL.\n\nSigned-off-by: Paul Mundt \u003clethal@linux-sh.org\u003e\n" }, { "commit": "6c645ac72582bacb85b90a1cf88e81a13045aba4", "tree": "a6101de4bd5426ebfb06b097749251ff6f9e98b2", "parents": [ "e827f20f1d34e91fbbb0df4674ddd8c3aad517da" ], "author": { "name": "Paul Mundt", "email": "lethal@linux-sh.org", "time": "Mon May 14 09:55:35 2007 +0900" }, "committer": { "name": "Paul Mundt", "email": "lethal@linux-sh.org", "time": "Mon May 14 09:55:35 2007 +0900" }, "message": "sh64: generic quicklist support.\n\nSigned-off-by: Paul Mundt \u003clethal@linux-sh.org\u003e\n" }, { "commit": "5f8c9908f200b775a3d6c345bc6f3e928e2426a9", "tree": "0e7077bdc8fef01c54845b0ed5f524bb31324350", "parents": [ "36f021b579d195cdc5fa6f3e2bab198b4bf70643" ], "author": { "name": "Paul Mundt", "email": "lethal@linux-sh.org", "time": "Tue May 08 11:55:21 2007 +0900" }, "committer": { "name": "Paul Mundt", "email": "lethal@hera.kernel.org", "time": "Wed May 09 01:35:00 2007 +0000" }, "message": "sh: generic quicklist support.\n\nThis moves SH over to the generic quicklists. As per x86_64,\nwe have special mappings for the PGDs, so these go on their\nown list..\n\nSigned-off-by: Paul Mundt \u003clethal@linux-sh.org\u003e\n" }, { "commit": "6225e93735acaa09865bce746958f1046c2e0bc3", "tree": "c741862fbd9f6a1fa350b08debfcfb159bb8bf71", "parents": [ "c09d87517298fd01543739ba26987645deb4e6a9" ], "author": { "name": "Christoph Lameter", "email": "clameter@sgi.com", "time": "Sun May 06 14:49:50 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Mon May 07 12:12:54 2007 -0700" }, "message": "Quicklists for page table pages\n\nOn x86_64 this cuts allocation overhead for page table pages down to a\nfraction (kernel compile / editing load. TSC based measurement of times spend\nin each function):\n\nno quicklist\n\npte_alloc 1569048 4.3s(401ns/2.7us/179.7us)\npmd_alloc 780988 2.1s(337ns/2.7us/86.1us)\npud_alloc 780072 2.2s(424ns/2.8us/300.6us)\npgd_alloc 260022 1s(920ns/4us/263.1us)\n\nquicklist:\n\npte_alloc 452436 573.4ms(8ns/1.3us/121.1us)\npmd_alloc 196204 174.5ms(7ns/889ns/46.1us)\npud_alloc 195688 172.4ms(7ns/881ns/151.3us)\npgd_alloc 65228 9.8ms(8ns/150ns/6.1us)\n\npgd allocations are the most complex and there we see the most dramatic\nimprovement (may be we can cut down the amount of pgds cached somewhat?). But\neven the pte allocations still see a doubling of performance.\n\n1. Proven code from the IA64 arch.\n\n\tThe method used here has been fine tuned for years and\n\tis NUMA aware. It is based on the knowledge that accesses\n\tto page table pages are sparse in nature. Taking a page\n\toff the freelists instead of allocating a zeroed pages\n\tallows a reduction of number of cachelines touched\n\tin addition to getting rid of the slab overhead. So\n\tperformance improves. This is particularly useful if pgds\n\tcontain standard mappings. We can save on the teardown\n\tand setup of such a page if we have some on the quicklists.\n\tThis includes avoiding lists operations that are otherwise\n\tnecessary on alloc and free to track pgds.\n\n2. Light weight alternative to use slab to manage page size pages\n\n\tSlab overhead is significant and even page allocator use\n\tis pretty heavy weight. The use of a per cpu quicklist\n\tmeans that we touch only two cachelines for an allocation.\n\tThere is no need to access the page_struct (unless arch code\n\tneeds to fiddle around with it). So the fast past just\n\tmeans bringing in one cacheline at the beginning of the\n\tpage. That same cacheline may then be used to store the\n\tpage table entry. Or a second cacheline may be used\n\tif the page table entry is not in the first cacheline of\n\tthe page. The current code will zero the page which means\n\ttouching 32 cachelines (assuming 128 byte). We get down\n\tfrom 32 to 2 cachelines in the fast path.\n\n3. x86_64 gets lightweight page table page management.\n\n\tThis will allow x86_64 arch code to faster repopulate pgds\n\tand other page table entries. The list operations for pgds\n\tare reduced in the same way as for i386 to the point where\n\ta pgd is allocated from the page allocator and when it is\n\tfreed back to the page allocator. A pgd can pass through\n\tthe quicklists without having to be reinitialized.\n\n64 Consolidation of code from multiple arches\n\n\tSo far arches have their own implementation of quicklist\n\tmanagement. This patch moves that feature into the core allowing\n\tan easier maintenance and consistent management of quicklists.\n\nPage table pages have the characteristics that they are typically zero or in a\nknown state when they are freed. This is usually the exactly same state as\nneeded after allocation. So it makes sense to build a list of freed page\ntable pages and then consume the pages already in use first. Those pages have\nalready been initialized correctly (thus no need to zero them) and are likely\nalready cached in such a way that the MMU can use them most effectively. Page\ntable pages are used in a sparse way so zeroing them on allocation is not too\nuseful.\n\nSuch an implementation already exits for ia64. Howver, that implementation\ndid not support constructors and destructors as needed by i386 / x86_64. It\nalso only supported a single quicklist. The implementation here has\nconstructor and destructor support as well as the ability for an arch to\nspecify how many quicklists are needed.\n\nQuicklists are defined by an arch defining CONFIG_QUICKLIST. If more than one\nquicklist is necessary then we can define NR_QUICK for additional lists. F.e.\n i386 needs two and thus has\n\nconfig NR_QUICK\n\tint\n\tdefault 2\n\nIf an arch has requested quicklist support then pages can be allocated\nfrom the quicklist (or from the page allocator if the quicklist is\nempty) via:\n\nquicklist_alloc(\u003cquicklist-nr\u003e, \u003cgfpflags\u003e, \u003cconstructor\u003e)\n\nPage table pages can be freed using:\n\nquicklist_free(\u003cquicklist-nr\u003e, \u003cdestructor\u003e, \u003cpage\u003e)\n\nPages must have a definite state after allocation and before\nthey are freed. If no constructor is specified then pages\nwill be zeroed on allocation and must be zeroed before they are\nfreed.\n\nIf a constructor is used then the constructor will establish\na definite page state. F.e. the i386 and x86_64 pgd constructors\nestablish certain mappings.\n\nConstructors and destructors can also be used to track the pages.\ni386 and x86_64 use a list of pgds in order to be able to dynamically\nupdate standard mappings.\n\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nCc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\nCc: Andi Kleen \u003cak@suse.de\u003e\nCc: \"Luck, Tony\" \u003ctony.luck@intel.com\u003e\nCc: William Lee Irwin III \u003cwli@holomorphy.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "5ac6da669e2476dbdac89b357b05b5a79bc5b657", "tree": "45b16221dd4f246595638e5fb6f629ddda96ff33", "parents": [ "339ba9b15df58199b9783a23af234e947ec9e6ba" ], "author": { "name": "Christoph Lameter", "email": "clameter@sgi.com", "time": "Sat Feb 10 01:43:14 2007 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Sun Feb 11 10:51:19 2007 -0800" }, "message": "[PATCH] Set CONFIG_ZONE_DMA for arches with GENERIC_ISA_DMA\n\nAs Andi pointed out: CONFIG_GENERIC_ISA_DMA only disables the ISA DMA\nchannel management. Other functionality may still expect GFP_DMA to\nprovide memory below 16M. So we need to make sure that CONFIG_ZONE_DMA is\nset independent of CONFIG_GENERIC_ISA_DMA. Undo the modifications to\nmm/Kconfig where we made ZONE_DMA dependent on GENERIC_ISA_DMA and set\ntheses explicitly in each arches Kconfig.\n\nReviews must occur for each arch in order to determine if ZONE_DMA can be\nswitched off. It can only be switched off if we know that all devices\nsupported by a platform are capable of performing DMA transfers to all of\nmemory (Some arches already support this: uml, avr32, sh sh64, parisc and\nIA64/Altix).\n\nIn order to switch ZONE_DMA off conditionally, one would have to establish\na scheme by which one can assure that no drivers are enabled that are only\ncapable of doing I/O to a part of memory, or one needs to provide an\nalternate means of performing an allocation from a specific range of memory\n(like provided by alloc_pages_range()) and insure that all drivers use that\ncall. In that case the arches alloc_dma_coherent() may need to be modified\nto call alloc_pages_range() instead of relying on GFP_DMA.\n\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "4b51d66989218aad731a721b5b28c79bf5388c09", "tree": "8ff7acbd219f699c20c2f1fd201ffb3db5a64062", "parents": [ "66701b1499a3ff11882c8c4aef36e8eac86e17b1" ], "author": { "name": "Christoph Lameter", "email": "clameter@sgi.com", "time": "Sat Feb 10 01:43:10 2007 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Sun Feb 11 10:51:18 2007 -0800" }, "message": "[PATCH] optional ZONE_DMA: optional ZONE_DMA in the VM\n\nMake ZONE_DMA optional in core code.\n\n- ifdef all code for ZONE_DMA and related definitions following the example\n for ZONE_DMA32 and ZONE_HIGHMEM.\n\n- Without ZONE_DMA, ZONE_HIGHMEM and ZONE_DMA32 we get to a ZONES_SHIFT of\n 0.\n\n- Modify the VM statistics to work correctly without a DMA zone.\n\n- Modify slab to not create DMA slabs if there is no ZONE_DMA.\n\n[akpm@osdl.org: cleanup]\n[jdike@addtoit.com: build fix]\n[apw@shadowen.org: Simplify calculation of the number of bits we need for ZONES_SHIFT]\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nCc: Andi Kleen \u003cak@suse.de\u003e\nCc: \"Luck, Tony\" \u003ctony.luck@intel.com\u003e\nCc: Kyle McMartin \u003ckyle@mcmartin.ca\u003e\nCc: Matthew Wilcox \u003cwilly@debian.org\u003e\nCc: James Bottomley \u003cJames.Bottomley@steeleye.com\u003e\nCc: Paul Mundt \u003clethal@linux-sh.org\u003e\nSigned-off-by: Andy Whitcroft \u003capw@shadowen.org\u003e\nSigned-off-by: Jeff Dike \u003cjdike@addtoit.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "66701b1499a3ff11882c8c4aef36e8eac86e17b1", "tree": "7900ced6b590c3cd939bcdc92355ff0a14f856e3", "parents": [ "6267276f3fdda9ad0d5ca451bdcbdf42b802d64b" ], "author": { "name": "Christoph Lameter", "email": "clameter@sgi.com", "time": "Sat Feb 10 01:43:09 2007 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Sun Feb 11 10:51:18 2007 -0800" }, "message": "[PATCH] optional ZONE_DMA: introduce CONFIG_ZONE_DMA\n\nThis patch simply defines CONFIG_ZONE_DMA for all arches. We later do special\nthings with CONFIG_ZONE_DMA after the VM and an arch are prepared to work\nwithout ZONE_DMA.\n\nCONFIG_ZONE_DMA can be defined in two ways depending on how an architecture\nhandles ISA DMA.\n\nFirst if CONFIG_GENERIC_ISA_DMA is set by the arch then we know that the arch\nneeds ZONE_DMA because ISA DMA devices are supported. We can catch this in\nmm/Kconfig and do not need to modify arch code.\n\nSecond, arches may use ZONE_DMA in an unknown way. We set CONFIG_ZONE_DMA for\nall arches that do not set CONFIG_GENERIC_ISA_DMA in order to insure backwards\ncompatibility. The arches may later undefine ZONE_DMA if their arch code has\nbeen verified to not depend on ZONE_DMA.\n\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nCc: Andi Kleen \u003cak@suse.de\u003e\nCc: \"Luck, Tony\" \u003ctony.luck@intel.com\u003e\nCc: Kyle McMartin \u003ckyle@mcmartin.ca\u003e\nCc: Matthew Wilcox \u003cwilly@debian.org\u003e\nCc: James Bottomley \u003cJames.Bottomley@steeleye.com\u003e\nCc: Paul Mundt \u003clethal@linux-sh.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "84eb8d0608af1576175307ed8fb3c8fde329e579", "tree": "13f3223e6907ebe309328d460fe51ce7ac346ae1", "parents": [ "992caacf1141b31e94540eb31e0540e3da3a5e25" ], "author": { "name": "Matt LaPlante", "email": "kernel1@cyberdogtech.com", "time": "Tue Oct 03 22:53:09 2006 +0200" }, "committer": { "name": "Adrian Bunk", "email": "bunk@stusta.de", "time": "Tue Oct 03 22:53:09 2006 +0200" }, "message": "Fix \"can not\" in Documentation and Kconfig\n\nRandy brought it to my attention that in proper english \"can not\" should always\nbe written \"cannot\". I donot see any reason to argue, even if I mightnot\nunderstand why this rule exists. This patch fixes \"can not\" in several\nDocumentation files as well as three Kconfigs.\n\nSigned-off-by: Matt LaPlante \u003ckernel1@cyberdogtech.com\u003e\nAcked-by: Randy Dunlap \u003crdunlap@xenotime.net\u003e\nSigned-off-by: Adrian Bunk \u003cbunk@stusta.de\u003e\n" }, { "commit": "44c09201a4178e08ed1c8cc37e7aea0683888f0a", "tree": "2b8a859ef668b24cc7c41331d29357979e07c364", "parents": [ "095096038d637c477ef3c1b674612bcbc4d60c2d" ], "author": { "name": "Matt LaPlante", "email": "kernel1@cyberdogtech.com", "time": "Tue Oct 03 22:34:14 2006 +0200" }, "committer": { "name": "Adrian Bunk", "email": "bunk@stusta.de", "time": "Tue Oct 03 22:34:14 2006 +0200" }, "message": "more misc typo fixes\n\nSigned-off-by: Adrian Bunk \u003cbunk@stusta.de\u003e\n" }, { "commit": "ec69acbb1191df671ff8e07c8e146619a5c53f70", "tree": "431aea50c8b614686543342c114ce36465b9dbc2", "parents": [ "f28c5edc06ecd8068b38b7662ad19f4d20d741af" ], "author": { "name": "Keith Mannthey", "email": "kmannth@us.ibm.com", "time": "Sat Sep 30 23:27:05 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Oct 01 00:39:18 2006 -0700" }, "message": "[PATCH] hot-add-mem x86_64: Kconfig changes\n\nCreate Kconfig namespace for MEMORY_HOTPLUG_RESERVE and MEMORY_HOTPLUG_SPARSE.\n This is needed to create a disticiton between the 2 paths. Selecting the\nhigh level opiton of MEMORY_HOTPLUG will get you MEMORY_HOTPLUG_SPARSE if you\nhave sparsemem enabled or MEMORY_HOTPLUG_RESERVE if you are x86_64 with\ndiscontig and ACPI numa support.\n\nSigned-off-by: Keith Mannthey \u003ckmannth@us.ibm.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Andi Kleen \u003cak@muc.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "1903ac54f8536b11478e4f01c339e10b538f59e0", "tree": "ff5410f0539ab4aa09f964fa1d0c6dc26c614dc2", "parents": [ "47c2a3aa4475d27073dd3c7e183fcc13f495c8f5", "87937472ff8e34ad5c7b798a8a52e4368af216df" ], "author": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Thu Jun 29 10:49:17 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Thu Jun 29 10:49:17 2006 -0700" }, "message": "Merge master.kernel.org:/pub/scm/linux/kernel/git/gregkh/pci-2.6\n\n* master.kernel.org:/pub/scm/linux/kernel/git/gregkh/pci-2.6:\n [PATCH] i386: export memory more than 4G through /proc/iomem\n [PATCH] 64bit Resource: finally enable 64bit resource sizes\n [PATCH] 64bit Resource: convert a few remaining drivers to use resource_size_t where needed\n [PATCH] 64bit resource: change pnp core to use resource_size_t\n [PATCH] 64bit resource: change pci core and arch code to use resource_size_t\n [PATCH] 64bit resource: change resource core to use resource_size_t\n [PATCH] 64bit resource: introduce resource_size_t for the start and end of struct resource\n [PATCH] 64bit resource: fix up printks for resources in misc drivers\n [PATCH] 64bit resource: fix up printks for resources in arch and core code\n [PATCH] 64bit resource: fix up printks for resources in pcmcia drivers\n [PATCH] 64bit resource: fix up printks for resources in video drivers\n [PATCH] 64bit resource: fix up printks for resources in ide drivers\n [PATCH] 64bit resource: fix up printks for resources in mtd drivers\n [PATCH] 64bit resource: fix up printks for resources in pci core and hotplug drivers\n [PATCH] 64bit resource: fix up printks for resources in networks drivers\n [PATCH] 64bit resource: fix up printks for resources in sound drivers\n [PATCH] 64bit resource: C99 changes for struct resource declarations\n\nFixed up trivial conflict in drivers/ide/pci/cmd64x.c (the printk that\nwas changed by the 64-bit resources had been deleted in the meantime ;)\n" }, { "commit": "cc57637b0b015fb5d70dbbec740de516d33af07d", "tree": "eafbfdf6e058194b46f0fe8dcb0e0c5de5b6fd6d", "parents": [ "81b0c8713385ce1b1b9058e916edcf9561ad76d6" ], "author": { "name": "Yasunori Goto", "email": "y-goto@jp.fujitsu.com", "time": "Thu Jun 29 02:24:27 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Thu Jun 29 10:26:20 2006 -0700" }, "message": "[PATCH] solve config broken: undefined reference to `online_page\u0027\n\nMemory hotplug code of i386 adds memory to only highmem. So, if\nCONFIG_HIGHMEM is not set, CONFIG_MEMORY_HOTPLUG shouldn\u0027t be set.\nOtherwise, it causes compile error.\n\nIn addition, many architecture can\u0027t use memory hotplug feature yet. So, I\nintroduce CONFIG_ARCH_ENABLE_MEMORY_HOTPLUG.\n\nSigned-off-by: Yasunori Goto \u003cy-goto@jp.fujitsu.com\u003e\nCc: \u003cstable@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "1f04bbd2d396a701c5af2e5b92bad896c2550c16", "tree": "ec84c99b1259361bc0878ebe78f1acadb30dad25", "parents": [ "2842f11419704f8707fffc82e10d2263427fc130" ], "author": { "name": "Yasunori Goto", "email": "y-goto@jp.fujitsu.com", "time": "Tue Jun 27 02:53:37 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Tue Jun 27 17:32:36 2006 -0700" }, "message": "[PATCH] sparc64: support sparsemem and !memory hotplug\n\nFix \"undefined reference to `arch_add_memory\u0027\" on sparc64 allmodconfig.\n\nsparc64 doesn\u0027t support memory hotplug. But we want it to support\nsparsemem.\n\nSigned-off-by: Yasunori Goto \u003cy-goto@jp.fujitsu.com\u003e\nCc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "6550e07f41ce8473ed684dac54fbfbd42183ffda", "tree": "d8ed394175785e9b28333d7ae130b725e176dfb5", "parents": [ "2427ddd8fae2febe3f5ac1ba76b092541304d9f0" ], "author": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Mon Jun 12 17:11:31 2006 -0700" }, "committer": { "name": "Greg Kroah-Hartman", "email": "gregkh@suse.de", "time": "Tue Jun 27 09:24:00 2006 -0700" }, "message": "[PATCH] 64bit Resource: finally enable 64bit resource sizes\n\nIntroduce the Kconfig entry and actually switch to a 64bit value, if\nwanted, for resource_size_t.\n\nBased on a patch series originally from Vivek Goyal \u003cvgoyal@in.ibm.com\u003e\n\nCc: Vivek Goyal \u003cvgoyal@in.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@suse.de\u003e\n" }, { "commit": "6c5240ae7f48c83fcaa8e24fa63e7eb09aba5651", "tree": "fede2324f4348701e60758d7f894aae4b09cdc9a", "parents": [ "d75a0fcda2cfc71b50e16dc89e0c32c57d427e85" ], "author": { "name": "Christoph Lameter", "email": "clameter@sgi.com", "time": "Fri Jun 23 02:03:37 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Jun 23 07:42:50 2006 -0700" }, "message": "[PATCH] Swapless page migration: modify core logic\n\nUse the migration entries for page migration\n\nThis modifies the migration code to use the new migration entries. It now\nbecomes possible to migrate anonymous pages without having to add a swap\nentry.\n\nWe add a couple of new functions to replace migration entries with the proper\nptes.\n\nWe cannot take the tree_lock for migrating anonymous pages anymore. However,\nwe know that we hold the only remaining reference to the page when the page\ncount reaches 1.\n\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "d784124cfe9377c1a24d8efba31401f81c7c11f9", "tree": "d4d46868e79dcab59fc589b11d4f0207b41f63b0", "parents": [ "0718dc2a82c865ca75975acabaf984057f9fd488" ], "author": { "name": "Christoph Lameter", "email": "clameter@sgi.com", "time": "Sat Mar 25 03:06:48 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sat Mar 25 08:22:50 2006 -0800" }, "message": "[PATCH] mm: make page migration dependent on swap and NUMA\n\nThe page migration code could function without NUMA but we currently have\nno users for the non-NUMA case.\n\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nCc: Adrian Bunk \u003cbunk@stusta.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "b20a35035f983f4ac7e29c4a68f30e43510007e0", "tree": "fdf090ddddbcc275349f62f71adc98649e2c683b", "parents": [ "442295c94bf650221af3ef20fc68fa3e93876818" ], "author": { "name": "Christoph Lameter", "email": "clameter@sgi.com", "time": "Wed Mar 22 00:09:12 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Mar 22 07:54:06 2006 -0800" }, "message": "[PATCH] page migration reorg\n\nCentralize the page migration functions in anticipation of additional\ntinkering. Creates a new file mm/migrate.c\n\n1. Extract buffer_migrate_page() from fs/buffer.c\n\n2. Extract central migration code from vmscan.c\n\n3. Extract some components from mempolicy.c\n\n4. Export pageout() and remove_from_swap() from vmscan.c\n\n5. Make it possible to configure NUMA systems without page migration\n and non-NUMA systems with page migration.\n\nI had to so some #ifdeffing in mempolicy.c that may need a cleanup.\n\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "7cbe34cf86c673503b177ff47cfa2c7030dabb50", "tree": "9b39d7e8f11fed68242d1cb1f0c85dfcf96e3250", "parents": [ "49d2e9cc4544369635cd6f4ef6d5bb0f757079a7" ], "author": { "name": "Christoph Lameter", "email": "clameter@engr.sgi.com", "time": "Sun Jan 08 01:00:49 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Jan 08 20:12:41 2006 -0800" }, "message": "[PATCH] Swap Migration V5: Add CONFIG_MIGRATION for page migration support\n\nInclude page migration if the system is NUMA or having a memory model that\nallows distinct areas of memory (SPARSEMEM, DISCONTIGMEM).\n\nAnd:\n- Only include lru_add_drain_per_cpu if building for an SMP system.\n\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "c898ec16e83331abde39118e22e9e38335bbb950", "tree": "bad725f27cb14e50663d39b8f36b1b79ab88fa80", "parents": [ "215c3409eed16c89b6d11ea1126bd9d4f36b9afd" ], "author": { "name": "Anton Blanchard", "email": "anton@samba.org", "time": "Fri Jan 06 00:12:07 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Fri Jan 06 08:33:37 2006 -0800" }, "message": "[PATCH] allow flatmem to be disabled when only sparsemem is implemented\n\nOn architectures that implement sparsemem but not discontigmem we want to\nbe able to hide the flatmem option in some cases. On ppc64 for example,\nwhen we select NUMA we must not select flatmem.\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\nSigned-off-by: Andy Whitcroft \u003capw@shadowen.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "7b6ac9dffe6f4dd8776908b234ac1410ed15f112", "tree": "1e30fde299519eb964899be79560599e04a32a66", "parents": [ "c101e77301877086e6f977fcfb140d1cbbe23fd5" ], "author": { "name": "Hugh Dickins", "email": "hugh@veritas.com", "time": "Wed Nov 23 13:37:37 2005 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Nov 23 16:08:38 2005 -0800" }, "message": "[PATCH] mm: update split ptlock Kconfig\n\nCloser attention to the arithmetic shows that neither ppc64 nor sparc really\nuses one page for multiple page tables: how on earth could they, while\npte_alloc_one returns just a struct page pointer, with no offset?\n\nWell, arm26 manages it by returning a pte_t pointer cast to a struct page\npointer, harumph, then compensating in its pmd_populate. But arm26 is never\nSMP, so it\u0027s not a problem for split ptlock either.\n\nAnd the PA-RISC situation has been recently improved: CONFIG_PA20 works\nwithout the 16-byte alignment which inflated its spinlock_t. But the current\nunion of spinlock_t with private does make the 7xxx struct page significantly\nlarger, even without debug, so disable its split ptlock.\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "2d4b95f06062d590aef8e44d42cec27b1828119f", "tree": "7beb1c976d3e08a9ef9cb37fa342e2b4f35e6aab", "parents": [ "732ee21f2894819781766a0cd88e32bdd630d11e" ], "author": { "name": "Hugh Dickins", "email": "hugh@veritas.com", "time": "Mon Nov 07 00:57:57 2005 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Mon Nov 07 07:53:23 2005 -0800" }, "message": "[PATCH] Suppress split ptlock on arches which may use one page for multiple page tables\n\nSuppress split ptlock on arches which may use one page for multiple page\ntables. Reconsider what better to do (particularly on ppc64) later on.\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "3947be1969a9ce455ec30f60ef51efb10e4323d1", "tree": "0b4b3b4c268beb7aa88cb685cce48b6bb5053c47", "parents": [ "bdc8cb984576ab5b550c8b24c6fa111a873503e3" ], "author": { "name": "Dave Hansen", "email": "haveblue@us.ibm.com", "time": "Sat Oct 29 18:16:54 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sat Oct 29 21:40:44 2005 -0700" }, "message": "[PATCH] memory hotplug: sysfs and add/remove functions\n\nThis adds generic memory add/remove and supporting functions for memory\nhotplug into a new file as well as a memory hotplug kernel config option.\n\nIndividual architecture patches will follow.\n\nFor now, disable memory hotplug when swsusp is enabled. There\u0027s a lot of\nchurn there right now. We\u0027ll fix it up properly once it calms down.\n\nSigned-off-by: Matt Tolentino \u003cmatthew.e.tolentino@intel.com\u003e\nSigned-off-by: Dave Hansen \u003chaveblue@us.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "4c21e2f2441dc5fbb957b030333f5a3f2d02dea7", "tree": "1f76d33bb1d76221c6424bc5fed080a4f91349a6", "parents": [ "b38c6845b695141259019e2b7c0fe6c32a6e720d" ], "author": { "name": "Hugh Dickins", "email": "hugh@veritas.com", "time": "Sat Oct 29 18:16:40 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sat Oct 29 21:40:42 2005 -0700" }, "message": "[PATCH] mm: split page table lock\n\nChristoph Lameter demonstrated very poor scalability on the SGI 512-way, with\na many-threaded application which concurrently initializes different parts of\na large anonymous area.\n\nThis patch corrects that, by using a separate spinlock per page table page, to\nguard the page table entries in that page, instead of using the mm\u0027s single\npage_table_lock. (But even then, page_table_lock is still used to guard page\ntable allocation, and anon_vma allocation.)\n\nIn this implementation, the spinlock is tucked inside the struct page of the\npage table page: with a BUILD_BUG_ON in case it overflows - which it would in\nthe case of 32-bit PA-RISC with spinlock debugging enabled.\n\nSplitting the lock is not quite for free: another cacheline access. Ideally,\nI suppose we would use split ptlock only for multi-threaded processes on\nmulti-cpu machines; but deciding that dynamically would have its own costs.\nSo for now enable it by config, at some number of cpus - since the Kconfig\nlanguage doesn\u0027t support inequalities, let preprocessor compare that with\nNR_CPUS. But I don\u0027t think it\u0027s worth being user-configurable: for good\ntesting of both split and unsplit configs, split now at 4 cpus, and perhaps\nchange that to 8 later.\n\nThere is a benefit even for singly threaded processes: kswapd can be attacking\none part of the mm while another part is busy faulting.\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "f3519f91942f2b43942400348c16d63fe9327f04", "tree": "893fa3d0a6a936f0e3123539e3ea196a71006a07", "parents": [ "b9491ac835829e6a34e2bbaa8adad261c71bf990" ], "author": { "name": "Dave Hansen", "email": "haveblue@us.ibm.com", "time": "Fri Sep 16 19:27:54 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sat Sep 17 11:50:01 2005 -0700" }, "message": "[PATCH] fix mm/Kconfig spelling\n\nSigned-off-by: Dave Hansen \u003chaveblue@us.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "3e347261a80b57df792ab9464b5f0ed59add53a8", "tree": "047b35e0f9ec82b3beeff882a9af6292a500097c", "parents": [ "802f192e4a600f7ef84ca25c8b818c8830acef5a" ], "author": { "name": "Bob Picco", "email": "bob.picco@hp.com", "time": "Sat Sep 03 15:54:28 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@evo.osdl.org", "time": "Mon Sep 05 00:05:38 2005 -0700" }, "message": "[PATCH] sparsemem extreme implementation\n\nWith cleanups from Dave Hansen \u003chaveblue@us.ibm.com\u003e\n\nSPARSEMEM_EXTREME makes mem_section a one dimensional array of pointers to\nmem_sections. This two level layout scheme is able to achieve smaller\nmemory requirements for SPARSEMEM with the tradeoff of an additional shift\nand load when fetching the memory section. The current SPARSEMEM\nimplementation is a one dimensional array of mem_sections which is the\ndefault SPARSEMEM configuration. The patch attempts isolates the\nimplementation details of the physical layout of the sparsemem section\narray.\n\nSPARSEMEM_EXTREME requires bootmem to be functioning at the time of\nmemory_present() calls. This is not always feasible, so architectures\nwhich do not need it may allocate everything statically by using\nSPARSEMEM_STATIC.\n\nSigned-off-by: Andy Whitcroft \u003capw@shadowen.org\u003e\nSigned-off-by: Bob Picco \u003cbob.picco@hp.com\u003e\nSigned-off-by: Dave Hansen \u003chaveblue@us.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "802f192e4a600f7ef84ca25c8b818c8830acef5a", "tree": "51e9a6ed164e6a2d8741af510c3954ad79bf19af", "parents": [ "0216f86dafb389c0ad97529fd45e64e883298cfd" ], "author": { "name": "Bob Picco", "email": "bob.picco@hp.com", "time": "Sat Sep 03 15:54:26 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@evo.osdl.org", "time": "Mon Sep 05 00:05:38 2005 -0700" }, "message": "[PATCH] SPARSEMEM EXTREME\n\nA new option for SPARSEMEM is ARCH_SPARSEMEM_EXTREME. Architecture\nplatforms with a very sparse physical address space would likely want to\nselect this option. For those architecture platforms that don\u0027t select the\noption, the code generated is equivalent to SPARSEMEM currently in -mm.\nI\u0027ll be posting a patch on ia64 ml which uses this new SPARSEMEM feature.\n\nARCH_SPARSEMEM_EXTREME makes mem_section a one dimensional array of\npointers to mem_sections. This two level layout scheme is able to achieve\nsmaller memory requirements for SPARSEMEM with the tradeoff of an\nadditional shift and load when fetching the memory section. The current\nSPARSEMEM -mm implementation is a one dimensional array of mem_sections\nwhich is the default SPARSEMEM configuration. The patch attempts isolates\nthe implementation details of the physical layout of the sparsemem section\narray.\n\nARCH_SPARSEMEM_EXTREME depends on 64BIT and is by default boolean false.\n\nI\u0027ve boot tested under aim load ia64 configured for ARCH_SPARSEMEM_EXTREME.\n I\u0027ve also boot tested a 4 way Opteron machine with !ARCH_SPARSEMEM_EXTREME\nand tested with aim.\n\nSigned-off-by: Andy Whitcroft \u003capw@shadowen.org\u003e\nSigned-off-by: Bob Picco \u003cbob.picco@hp.com\u003e\nSigned-off-by: Dave Hansen \u003chaveblue@us.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "d41dee369bff3b9dcb6328d4d822926c28cc2594", "tree": "a0405f3b7af3ebca21838a7d427bd75a067bf850", "parents": [ "af705362ab6018071310c5fcd436a6b457517d5f" ], "author": { "name": "Andy Whitcroft", "email": "apw@shadowen.org", "time": "Thu Jun 23 00:07:54 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:04 2005 -0700" }, "message": "[PATCH] sparsemem memory model\n\nSparsemem abstracts the use of discontiguous mem_maps[]. This kind of\nmem_map[] is needed by discontiguous memory machines (like in the old\nCONFIG_DISCONTIGMEM case) as well as memory hotplug systems. Sparsemem\nreplaces DISCONTIGMEM when enabled, and it is hoped that it can eventually\nbecome a complete replacement.\n\nA significant advantage over DISCONTIGMEM is that it\u0027s completely separated\nfrom CONFIG_NUMA. When producing this patch, it became apparent in that NUMA\nand DISCONTIG are often confused.\n\nAnother advantage is that sparse doesn\u0027t require each NUMA node\u0027s ranges to be\ncontiguous. It can handle overlapping ranges between nodes with no problems,\nwhere DISCONTIGMEM currently throws away that memory.\n\nSparsemem uses an array to provide different pfn_to_page() translations for\neach SECTION_SIZE area of physical memory. This is what allows the mem_map[]\nto be chopped up.\n\nIn order to do quick pfn_to_page() operations, the section number of the page\nis encoded in page-\u003eflags. Part of the sparsemem infrastructure enables\nsharing of these bits more dynamically (at compile-time) between the\npage_zone() and sparsemem operations. However, on 32-bit architectures, the\nnumber of bits is quite limited, and may require growing the size of the\npage-\u003eflags type in certain conditions. Several things might force this to\noccur: a decrease in the SECTION_SIZE (if you want to hotplug smaller areas of\nmemory), an increase in the physical address space, or an increase in the\nnumber of used page-\u003eflags.\n\nOne thing to note is that, once sparsemem is present, the NUMA node\ninformation no longer needs to be stored in the page-\u003eflags. It might provide\nspeed increases on certain platforms and will be stored there if there is\nroom. But, if out of room, an alternate (theoretically slower) mechanism is\nused.\n\nThis patch introduces CONFIG_FLATMEM. It is used in almost all cases where\nthere used to be an #ifndef DISCONTIG, because SPARSEMEM and DISCONTIGMEM\noften have to compile out the same areas of code.\n\nSigned-off-by: Andy Whitcroft \u003capw@shadowen.org\u003e\nSigned-off-by: Dave Hansen \u003chaveblue@us.ibm.com\u003e\nSigned-off-by: Martin Bligh \u003cmbligh@aracnet.com\u003e\nSigned-off-by: Adrian Bunk \u003cbunk@stusta.de\u003e\nSigned-off-by: Yasunori Goto \u003cy-goto@jp.fujitsu.com\u003e\nSigned-off-by: Bob Picco \u003cbob.picco@hp.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "af705362ab6018071310c5fcd436a6b457517d5f", "tree": "429e505558ee98e19b08ade8d02eedc3fee2478b", "parents": [ "b159d43fbf7eaaac6ecc647f51cf4257332db47b" ], "author": { "name": "Andy Whitcroft", "email": "apw@shadowen.org", "time": "Thu Jun 23 00:07:53 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:04 2005 -0700" }, "message": "[PATCH] generify memory present\n\nAllow architectures to indicate that they will be providing hooks to indice\ninstalled memory areas, memory_present(). Provide prototypes for the i386\nimplementation.\n\nSigned-off-by: Andy Whitcroft \u003capw@shadowen.org\u003e\nSigned-off-by: Dave Hansen \u003chaveblue@us.ibm.com\u003e\nSigned-off-by: Martin Bligh \u003cmbligh@aracnet.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "785dcd44b60ec8ede76fed0af54333ab5f3e848c", "tree": "d928787cd93c608f24a6366d0990cd962a5a0c16", "parents": [ "e1785e85b9c81c67b581b511ee4efac6c81e9edb" ], "author": { "name": "Dave Hansen", "email": "haveblue@us.ibm.com", "time": "Thu Jun 23 00:07:50 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:03 2005 -0700" }, "message": "[PATCH] mm/Kconfig: give DISCONTIG more help text\n\nThis gives DISCONTIGMEM a bit more help text to explain what it does, not just\nwhen to choose it.\n\nSigned-off-by: Dave Hansen \u003chaveblue@us.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "e1785e85b9c81c67b581b511ee4efac6c81e9edb", "tree": "97d0470fec528f9c995674abd39c02c36ec2d110", "parents": [ "074ccf8016b61f4b40066f8d737ab31e17a6afd1" ], "author": { "name": "Dave Hansen", "email": "haveblue@us.ibm.com", "time": "Thu Jun 23 00:07:49 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:03 2005 -0700" }, "message": "[PATCH] mm/Kconfig: hide \"Memory Model\" selection menu\n\nI got some feedback from users who think that the new \"Memory Model\" menu is a\nlittle invasive. This patch will hide that menu, except when\nCONFIG_EXPERIMENTAL is enabled *or* when an individual architecture wants it.\n\nAn individual arch may want to enable it because they\u0027ve removed their\narch-specific DISCONTIG prompt in favor of the mm/Kconfig one.\n\nSigned-off-by: Dave Hansen \u003chaveblue@us.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "44d0f805c77902a22dda244fd092b4567066b2b9", "tree": "62732a526b6ade8b9cc0f403130ddf4d4b4fdf46", "parents": [ "93b7504e3e6c1d98586854806e51bea329ea3aa9" ], "author": { "name": "Dave Hansen", "email": "haveblue@us.ibm.com", "time": "Thu Jun 23 00:07:47 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:03 2005 -0700" }, "message": "[PATCH] sparsemem: fix minor \"defaults\" issue in mm/Kconfig\n\nThe following patch applies on top of 2.6.12-rc2-mm1. It fixes a minor\nuser interaction issue, and an early reference to SPARSEMEM.\n\nThis \"choice\" menu would always default to FLATMEM, as it was listed first.\n Move it to the end so that the other defaults have a chance first.\n\nSigned-off-by: Dave Hansen \u003chaveblue@us.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "93b7504e3e6c1d98586854806e51bea329ea3aa9", "tree": "7b0d6f3a6214960daf3136f8c418178405521c07", "parents": [ "0e19243e9a19ef8e5994852671bd06bb51630811" ], "author": { "name": "Dave Hansen", "email": "haveblue@us.ibm.com", "time": "Thu Jun 23 00:07:47 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:03 2005 -0700" }, "message": "[PATCH] Introduce new Kconfig option for NUMA or DISCONTIG\n\nThere is some confusion that arose when working on SPARSEMEM patch between\nwhat is needed for DISCONTIG vs. NUMA.\n\nMultiple pg_data_t\u0027s are needed for DISCONTIGMEM or NUMA, independently.\nAll of the current NUMA implementations require an implementation of\nDISCONTIG. Because of this, quite a lot of code which is really needed for\nNUMA is actually under DISCONTIG #ifdefs. For SPARSEMEM, we changed some\nof these #ifdefs to CONFIG_NUMA, but that broke the DISCONTIG\u003dy and NUMA\u003dn\ncase.\n\nIntroducing this new NEED_MULTIPLE_NODES config option allows code that is\nneeded for both NUMA or DISCONTIG to be separated out from code that is\nspecific to DISCONTIG.\n\nOne great advantage of this approach is that it doesn\u0027t require every\narchitecture to be converted over. All of the current implementations\nshould \"just work\", only the ones implementing SPARSEMEM will have to be\nfixed up.\n\nThe change to free_area_init() makes it work inside, or out of the new\nconfig option.\n\nSigned-off-by: Dave Hansen \u003chaveblue@us.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "3a9da7655d2d5b7f790a370328cf093440c80496", "tree": "29fd39ee7dce0cfb4abc104a8492c22ff7f06111", "parents": [ "5b505b90b2d54e526cc8d123bdef3b98c9f0bbc6" ], "author": { "name": "Dave Hansen", "email": "haveblue@us.ibm.com", "time": "Thu Jun 23 00:07:42 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Thu Jun 23 09:45:01 2005 -0700" }, "message": "[PATCH] create mm/Kconfig for arch-independent memory options\n\nWith sparsemem being introduced, we need a central place for new\nmemory-related .config options: mm/Kconfig. This allows us to remove many\nof the duplicated arch-specific options.\n\nThe new option, CONFIG_FLATMEM, is there to enable us to detangle NUMA and\nDISCONTIGMEM. This is a requirement for sparsemem because sparsemem uses\nthe NUMA code without the presence of DISCONTIGMEM. The sparsemem patches\nuse CONFIG_FLATMEM in generic code, so this patch is a requirement before\napplying them.\n\nAlmost all places that used to do \u0027#ifndef CONFIG_DISCONTIGMEM\u0027 should use\n\u0027#ifdef CONFIG_FLATMEM\u0027 instead.\n\nSigned-off-by: Andy Whitcroft \u003capw@shadowen.org\u003e\nSigned-off-by: Dave Hansen \u003chaveblue@us.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" } ] }