From: Joonsoo Kim <iamjoonsoo.kim@lge.com> To: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com> Cc: Andrew Morton <akpm@linux-foundation.org>, Rik van Riel <riel@redhat.com>, Mel Gorman <mgorman@suse.de>, Michal Hocko <mhocko@suse.cz>, KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>, Hugh Dickins <hughd@google.com>, Davidlohr Bueso <davidlohr.bueso@hp.com>, David Gibson <david@gibson.dropbear.id.au>, linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 7/9] mm, hugetlb: add VM_NORESERVE check in vma_has_reserves() Date: Tue, 16 Jul 2013 16:12:17 +0900 [thread overview] Message-ID: <20130716071216.GC30116@lge.com> (raw) In-Reply-To: <874nbvhx90.fsf@linux.vnet.ibm.com> On Tue, Jul 16, 2013 at 11:17:23AM +0530, Aneesh Kumar K.V wrote: > Joonsoo Kim <iamjoonsoo.kim@lge.com> writes: > > > On Mon, Jul 15, 2013 at 08:41:12PM +0530, Aneesh Kumar K.V wrote: > >> Joonsoo Kim <iamjoonsoo.kim@lge.com> writes: > >> > >> > If we map the region with MAP_NORESERVE and MAP_SHARED, > >> > we can skip to check reserve counting and eventually we cannot be ensured > >> > to allocate a huge page in fault time. > >> > With following example code, you can easily find this situation. > >> > > >> > Assume 2MB, nr_hugepages = 100 > >> > > >> > fd = hugetlbfs_unlinked_fd(); > >> > if (fd < 0) > >> > return 1; > >> > > >> > size = 200 * MB; > >> > flag = MAP_SHARED; > >> > p = mmap(NULL, size, PROT_READ|PROT_WRITE, flag, fd, 0); > >> > if (p == MAP_FAILED) { > >> > fprintf(stderr, "mmap() failed: %s\n", strerror(errno)); > >> > return -1; > >> > } > >> > > >> > size = 2 * MB; > >> > flag = MAP_ANONYMOUS | MAP_SHARED | MAP_HUGETLB | MAP_NORESERVE; > >> > p = mmap(NULL, size, PROT_READ|PROT_WRITE, flag, -1, 0); > >> > if (p == MAP_FAILED) { > >> > fprintf(stderr, "mmap() failed: %s\n", strerror(errno)); > >> > } > >> > p[0] = '0'; > >> > sleep(10); > >> > > >> > During executing sleep(10), run 'cat /proc/meminfo' on another process. > >> > You'll find a mentioned problem. > >> > > >> > Solution is simple. We should check VM_NORESERVE in vma_has_reserves(). > >> > This prevent to use a pre-allocated huge page if free count is under > >> > the reserve count. > >> > >> You have a problem with this patch, which i guess you are fixing in > >> patch 9. Consider two process > >> > >> a) MAP_SHARED on fd > >> b) MAP_SHARED | MAP_NORESERVE on fd > >> > >> We should allow the (b) to access the page even if VM_NORESERVE is set > >> and we are out of reserve space . > > > > I can't get your point. > > Please elaborate more on this. > > > One process mmap with MAP_SHARED and another one with MAP_SHARED | MAP_NORESERVE > Now the first process will result in reserving the pages from the hugtlb > pool. Now if the second process try to dequeue huge page and we don't > have free space we will fail because > > vma_has_reservers will now return zero because VM_NORESERVE is set > and we can have (h->free_huge_pages - h->resv_huge_pages) == 0; I think that this behavior is correct, because a user who mapped with VM_NORESERVE should not think their allocation always succeed. With patch 9, he can be ensured to succeed, but I think it is side-effect. > The below hunk in your patch 9 handles that > > + if (vma->vm_flags & VM_NORESERVE) { > + /* > + * This address is already reserved by other process(chg == 0), > + * so, we should decreament reserved count. Without > + * decreamenting, reserve count is remained after releasing > + * inode, because this allocated page will go into page cache > + * and is regarded as coming from reserved pool in releasing > + * step. Currently, we don't have any other solution to deal > + * with this situation properly, so add work-around here. > + */ > + if (vma->vm_flags & VM_MAYSHARE && chg == 0) > + return 1; > + else > + return 0; > + } > > so may be both of these should be folded ? I think that these patches should not be folded, because these handle two separate issues. Reserve count mismatch issue mentioned in patch 9 is not introduced by patch 7. Thanks. > > -aneesh > > -- > To unsubscribe, send a message with 'unsubscribe linux-mm' in > the body to majordomo@kvack.org. For more info on Linux MM, > see: http://www.linux-mm.org/ . > Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: Joonsoo Kim <iamjoonsoo.kim@lge.com> To: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com> Cc: Andrew Morton <akpm@linux-foundation.org>, Rik van Riel <riel@redhat.com>, Mel Gorman <mgorman@suse.de>, Michal Hocko <mhocko@suse.cz>, KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>, Hugh Dickins <hughd@google.com>, Davidlohr Bueso <davidlohr.bueso@hp.com>, David Gibson <david@gibson.dropbear.id.au>, linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 7/9] mm, hugetlb: add VM_NORESERVE check in vma_has_reserves() Date: Tue, 16 Jul 2013 16:12:17 +0900 [thread overview] Message-ID: <20130716071216.GC30116@lge.com> (raw) In-Reply-To: <874nbvhx90.fsf@linux.vnet.ibm.com> On Tue, Jul 16, 2013 at 11:17:23AM +0530, Aneesh Kumar K.V wrote: > Joonsoo Kim <iamjoonsoo.kim@lge.com> writes: > > > On Mon, Jul 15, 2013 at 08:41:12PM +0530, Aneesh Kumar K.V wrote: > >> Joonsoo Kim <iamjoonsoo.kim@lge.com> writes: > >> > >> > If we map the region with MAP_NORESERVE and MAP_SHARED, > >> > we can skip to check reserve counting and eventually we cannot be ensured > >> > to allocate a huge page in fault time. > >> > With following example code, you can easily find this situation. > >> > > >> > Assume 2MB, nr_hugepages = 100 > >> > > >> > fd = hugetlbfs_unlinked_fd(); > >> > if (fd < 0) > >> > return 1; > >> > > >> > size = 200 * MB; > >> > flag = MAP_SHARED; > >> > p = mmap(NULL, size, PROT_READ|PROT_WRITE, flag, fd, 0); > >> > if (p == MAP_FAILED) { > >> > fprintf(stderr, "mmap() failed: %s\n", strerror(errno)); > >> > return -1; > >> > } > >> > > >> > size = 2 * MB; > >> > flag = MAP_ANONYMOUS | MAP_SHARED | MAP_HUGETLB | MAP_NORESERVE; > >> > p = mmap(NULL, size, PROT_READ|PROT_WRITE, flag, -1, 0); > >> > if (p == MAP_FAILED) { > >> > fprintf(stderr, "mmap() failed: %s\n", strerror(errno)); > >> > } > >> > p[0] = '0'; > >> > sleep(10); > >> > > >> > During executing sleep(10), run 'cat /proc/meminfo' on another process. > >> > You'll find a mentioned problem. > >> > > >> > Solution is simple. We should check VM_NORESERVE in vma_has_reserves(). > >> > This prevent to use a pre-allocated huge page if free count is under > >> > the reserve count. > >> > >> You have a problem with this patch, which i guess you are fixing in > >> patch 9. Consider two process > >> > >> a) MAP_SHARED on fd > >> b) MAP_SHARED | MAP_NORESERVE on fd > >> > >> We should allow the (b) to access the page even if VM_NORESERVE is set > >> and we are out of reserve space . > > > > I can't get your point. > > Please elaborate more on this. > > > One process mmap with MAP_SHARED and another one with MAP_SHARED | MAP_NORESERVE > Now the first process will result in reserving the pages from the hugtlb > pool. Now if the second process try to dequeue huge page and we don't > have free space we will fail because > > vma_has_reservers will now return zero because VM_NORESERVE is set > and we can have (h->free_huge_pages - h->resv_huge_pages) == 0; I think that this behavior is correct, because a user who mapped with VM_NORESERVE should not think their allocation always succeed. With patch 9, he can be ensured to succeed, but I think it is side-effect. > The below hunk in your patch 9 handles that > > + if (vma->vm_flags & VM_NORESERVE) { > + /* > + * This address is already reserved by other process(chg == 0), > + * so, we should decreament reserved count. Without > + * decreamenting, reserve count is remained after releasing > + * inode, because this allocated page will go into page cache > + * and is regarded as coming from reserved pool in releasing > + * step. Currently, we don't have any other solution to deal > + * with this situation properly, so add work-around here. > + */ > + if (vma->vm_flags & VM_MAYSHARE && chg == 0) > + return 1; > + else > + return 0; > + } > > so may be both of these should be folded ? I think that these patches should not be folded, because these handle two separate issues. Reserve count mismatch issue mentioned in patch 9 is not introduced by patch 7. Thanks. > > -aneesh > > -- > To unsubscribe, send a message with 'unsubscribe linux-mm' in > the body to majordomo@kvack.org. For more info on Linux MM, > see: http://www.linux-mm.org/ . > Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a> -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2013-07-16 7:12 UTC|newest] Thread overview: 92+ messages / expand[flat|nested] mbox.gz Atom feed top 2013-07-15 9:52 [PATCH 0/9] mm, hugetlb: clean-up and possible bug fix Joonsoo Kim 2013-07-15 9:52 ` Joonsoo Kim 2013-07-15 9:52 ` [PATCH 1/9] mm, hugetlb: move up the code which check availability of free huge page Joonsoo Kim 2013-07-15 9:52 ` Joonsoo Kim 2013-07-15 14:01 ` Aneesh Kumar K.V 2013-07-15 14:01 ` Aneesh Kumar K.V 2013-07-16 1:16 ` Joonsoo Kim 2013-07-16 1:16 ` Joonsoo Kim 2013-07-16 3:36 ` Aneesh Kumar K.V 2013-07-16 3:36 ` Aneesh Kumar K.V 2013-07-16 5:10 ` Joonsoo Kim 2013-07-16 5:10 ` Joonsoo Kim 2013-07-22 14:45 ` Michal Hocko 2013-07-22 14:45 ` Michal Hocko 2013-07-15 9:52 ` [PATCH 2/9] mm, hugetlb: trivial commenting fix Joonsoo Kim 2013-07-15 9:52 ` Joonsoo Kim 2013-07-15 13:12 ` Hillf Danton 2013-07-15 13:12 ` Hillf Danton 2013-07-15 14:02 ` Aneesh Kumar K.V 2013-07-15 14:02 ` Aneesh Kumar K.V 2013-07-22 14:46 ` Michal Hocko 2013-07-22 14:46 ` Michal Hocko 2013-07-15 9:52 ` [PATCH 3/9] mm, hugetlb: clean-up alloc_huge_page() Joonsoo Kim 2013-07-15 9:52 ` Joonsoo Kim 2013-07-22 14:51 ` Michal Hocko 2013-07-22 14:51 ` Michal Hocko 2013-07-23 7:29 ` Joonsoo Kim 2013-07-23 7:29 ` Joonsoo Kim 2013-07-15 9:52 ` [PATCH 4/9] mm, hugetlb: fix and clean-up node iteration code to alloc or free Joonsoo Kim 2013-07-15 9:52 ` Joonsoo Kim 2013-07-15 14:27 ` Aneesh Kumar K.V 2013-07-15 14:27 ` Aneesh Kumar K.V 2013-07-16 1:41 ` Joonsoo Kim 2013-07-16 1:41 ` Joonsoo Kim 2013-07-17 2:00 ` Jianguo Wu 2013-07-17 2:00 ` Jianguo Wu 2013-07-18 6:46 ` Joonsoo Kim 2013-07-18 6:46 ` Joonsoo Kim 2013-07-15 9:52 ` [PATCH 5/9] mm, hugetlb: remove redundant list_empty check in gather_surplus_pages() Joonsoo Kim 2013-07-15 9:52 ` Joonsoo Kim 2013-07-15 14:31 ` Aneesh Kumar K.V 2013-07-15 14:31 ` Aneesh Kumar K.V 2013-07-16 1:42 ` Joonsoo Kim 2013-07-16 1:42 ` Joonsoo Kim 2013-07-22 14:55 ` Michal Hocko 2013-07-22 14:55 ` Michal Hocko 2013-07-15 9:52 ` [PATCH 6/9] mm, hugetlb: do not use a page in page cache for cow optimization Joonsoo Kim 2013-07-15 9:52 ` Joonsoo Kim 2013-07-15 13:55 ` Aneesh Kumar K.V 2013-07-15 13:55 ` Aneesh Kumar K.V 2013-07-16 1:56 ` Joonsoo Kim 2013-07-16 1:56 ` Joonsoo Kim 2013-07-17 8:55 ` Wanpeng Li 2013-07-17 8:55 ` Wanpeng Li 2013-07-15 9:52 ` [PATCH 7/9] mm, hugetlb: add VM_NORESERVE check in vma_has_reserves() Joonsoo Kim 2013-07-15 9:52 ` Joonsoo Kim 2013-07-15 14:48 ` Aneesh Kumar K.V 2013-07-15 14:48 ` Aneesh Kumar K.V 2013-07-15 15:11 ` Aneesh Kumar K.V 2013-07-15 15:11 ` Aneesh Kumar K.V 2013-07-16 2:12 ` Joonsoo Kim 2013-07-16 2:12 ` Joonsoo Kim 2013-07-16 5:47 ` Aneesh Kumar K.V 2013-07-16 5:47 ` Aneesh Kumar K.V 2013-07-16 7:12 ` Joonsoo Kim [this message] 2013-07-16 7:12 ` Joonsoo Kim 2013-07-18 2:03 ` Wanpeng Li 2013-07-18 2:03 ` Wanpeng Li 2013-07-15 9:52 ` [PATCH 8/9] mm, hugetlb: remove decrement_hugepage_resv_vma() Joonsoo Kim 2013-07-15 9:52 ` Joonsoo Kim 2013-07-15 14:50 ` Aneesh Kumar K.V 2013-07-15 14:50 ` Aneesh Kumar K.V 2013-07-17 9:31 ` Wanpeng Li 2013-07-17 9:31 ` Wanpeng Li 2013-07-15 9:52 ` [PATCH 9/9] mm, hugetlb: decrement reserve count if VM_NORESERVE alloc page cache Joonsoo Kim 2013-07-15 9:52 ` Joonsoo Kim 2013-07-15 15:11 ` Aneesh Kumar K.V 2013-07-15 15:11 ` Aneesh Kumar K.V 2013-07-18 2:02 ` Wanpeng Li 2013-07-18 2:02 ` Wanpeng Li 2013-07-15 14:10 ` [PATCH 0/9] mm, hugetlb: clean-up and possible bug fix Aneesh Kumar K.V 2013-07-15 14:10 ` Aneesh Kumar K.V 2013-07-16 1:10 ` Joonsoo Kim 2013-07-16 1:10 ` Joonsoo Kim 2013-07-16 1:27 ` Sam Ben 2013-07-16 1:27 ` Sam Ben 2013-07-16 1:45 ` Joonsoo Kim 2013-07-16 1:45 ` Joonsoo Kim 2013-07-16 1:55 ` Sam Ben 2013-07-16 1:55 ` Sam Ben 2013-07-16 2:14 ` Joonsoo Kim 2013-07-16 2:14 ` Joonsoo Kim
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20130716071216.GC30116@lge.com \ --to=iamjoonsoo.kim@lge.com \ --cc=akpm@linux-foundation.org \ --cc=aneesh.kumar@linux.vnet.ibm.com \ --cc=david@gibson.dropbear.id.au \ --cc=davidlohr.bueso@hp.com \ --cc=hughd@google.com \ --cc=kamezawa.hiroyu@jp.fujitsu.com \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=mgorman@suse.de \ --cc=mhocko@suse.cz \ --cc=riel@redhat.com \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.