From: Matthew Wilcox <matthew.r.wilcox@intel.com> To: linux-kernel@vger.kernel.org, linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, willy@linux.intel.com Cc: Matthew Wilcox <matthew.r.wilcox@intel.com> Subject: [PATCH v6 22/22] dax: Add reporting of major faults Date: Tue, 25 Feb 2014 09:18:38 -0500 [thread overview] Message-ID: <1393337918-28265-23-git-send-email-matthew.r.wilcox@intel.com> (raw) In-Reply-To: <1393337918-28265-1-git-send-email-matthew.r.wilcox@intel.com> If we have to call get_block with the create argument set to 1, then the filesystem almost certainly had to zero the block. which is an I/O, which should be reported as a major fault. Note that major faults on DAX files happen for different reasons than major faults on non-DAX files. DAX files behave as if everything except file holes is already cached. That's all the more reason to report major faults when we do have to do I/O; it may be a valuable resource for sysadmins trying to diagnose performance problems. Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com> --- fs/dax.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/fs/dax.c b/fs/dax.c index cdc8012..79a67c5 100644 --- a/fs/dax.c +++ b/fs/dax.c @@ -20,10 +20,12 @@ #include <linux/fs.h> #include <linux/genhd.h> #include <linux/highmem.h> +#include <linux/memcontrol.h> #include <linux/mm.h> #include <linux/mutex.h> #include <linux/sched.h> #include <linux/uio.h> +#include <linux/vmstat.h> int dax_clear_blocks(struct inode *inode, sector_t block, long size) { @@ -286,6 +288,7 @@ static int do_dax_fault(struct vm_area_struct *vma, struct vm_fault *vmf, pgoff_t size; unsigned long pfn; int error; + int major = 0; size = (i_size_read(inode) + PAGE_SIZE - 1) >> PAGE_SHIFT; if (vmf->pgoff >= size) @@ -301,6 +304,9 @@ static int do_dax_fault(struct vm_area_struct *vma, struct vm_fault *vmf, if (!buffer_written(&bh) && !vmf->cow_page) { if (vmf->flags & FAULT_FLAG_WRITE) { error = get_block(inode, block, &bh, 1); + count_vm_event(PGMAJFAULT); + mem_cgroup_count_vm_event(vma->vm_mm, PGMAJFAULT); + major = VM_FAULT_MAJOR; if (error || bh.b_size < PAGE_SIZE) return VM_FAULT_SIGBUS; } else { @@ -332,7 +338,7 @@ static int do_dax_fault(struct vm_area_struct *vma, struct vm_fault *vmf, /* -EBUSY is fine, somebody else faulted on the same PTE */ if (error != -EBUSY) BUG_ON(error); - return VM_FAULT_NOPAGE; + return VM_FAULT_NOPAGE | major; } /** -- 1.8.5.3
WARNING: multiple messages have this Message-ID (diff)
From: Matthew Wilcox <matthew.r.wilcox@intel.com> To: linux-kernel@vger.kernel.org, linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, willy@linux.intel.com Cc: Matthew Wilcox <matthew.r.wilcox@intel.com> Subject: [PATCH v6 22/22] dax: Add reporting of major faults Date: Tue, 25 Feb 2014 09:18:38 -0500 [thread overview] Message-ID: <1393337918-28265-23-git-send-email-matthew.r.wilcox@intel.com> (raw) In-Reply-To: <1393337918-28265-1-git-send-email-matthew.r.wilcox@intel.com> If we have to call get_block with the create argument set to 1, then the filesystem almost certainly had to zero the block. which is an I/O, which should be reported as a major fault. Note that major faults on DAX files happen for different reasons than major faults on non-DAX files. DAX files behave as if everything except file holes is already cached. That's all the more reason to report major faults when we do have to do I/O; it may be a valuable resource for sysadmins trying to diagnose performance problems. Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com> --- fs/dax.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/fs/dax.c b/fs/dax.c index cdc8012..79a67c5 100644 --- a/fs/dax.c +++ b/fs/dax.c @@ -20,10 +20,12 @@ #include <linux/fs.h> #include <linux/genhd.h> #include <linux/highmem.h> +#include <linux/memcontrol.h> #include <linux/mm.h> #include <linux/mutex.h> #include <linux/sched.h> #include <linux/uio.h> +#include <linux/vmstat.h> int dax_clear_blocks(struct inode *inode, sector_t block, long size) { @@ -286,6 +288,7 @@ static int do_dax_fault(struct vm_area_struct *vma, struct vm_fault *vmf, pgoff_t size; unsigned long pfn; int error; + int major = 0; size = (i_size_read(inode) + PAGE_SIZE - 1) >> PAGE_SHIFT; if (vmf->pgoff >= size) @@ -301,6 +304,9 @@ static int do_dax_fault(struct vm_area_struct *vma, struct vm_fault *vmf, if (!buffer_written(&bh) && !vmf->cow_page) { if (vmf->flags & FAULT_FLAG_WRITE) { error = get_block(inode, block, &bh, 1); + count_vm_event(PGMAJFAULT); + mem_cgroup_count_vm_event(vma->vm_mm, PGMAJFAULT); + major = VM_FAULT_MAJOR; if (error || bh.b_size < PAGE_SIZE) return VM_FAULT_SIGBUS; } else { @@ -332,7 +338,7 @@ static int do_dax_fault(struct vm_area_struct *vma, struct vm_fault *vmf, /* -EBUSY is fine, somebody else faulted on the same PTE */ if (error != -EBUSY) BUG_ON(error); - return VM_FAULT_NOPAGE; + return VM_FAULT_NOPAGE | major; } /** -- 1.8.5.3 -- 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:[~2014-02-25 14:21 UTC|newest] Thread overview: 79+ messages / expand[flat|nested] mbox.gz Atom feed top 2014-02-25 14:18 [PATCH v6 00/22] Support ext4 on NV-DIMMs Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 01/22] Fix XIP fault vs truncate race Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 02/22] Allow page fault handlers to perform the COW Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 03/22] axonram: Fix bug in direct_access Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 04/22] Change direct_access calling convention Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 05/22] Introduce IS_DAX(inode) Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 06/22] Replace XIP read and write with DAX I/O Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-03-11 0:32 ` Toshi Kani 2014-03-11 0:32 ` Toshi Kani 2014-03-11 12:53 ` Matthew Wilcox 2014-03-11 12:53 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 07/22] Replace the XIP page fault handler with the DAX page fault handler Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-28 17:49 ` Toshi Kani 2014-02-28 17:49 ` Toshi Kani 2014-02-28 20:20 ` Matthew Wilcox 2014-02-28 20:20 ` Matthew Wilcox 2014-02-28 22:18 ` Toshi Kani 2014-02-28 22:18 ` Toshi Kani 2014-02-28 22:18 ` Toshi Kani 2014-03-02 23:30 ` Dave Chinner 2014-03-02 23:30 ` Dave Chinner 2014-03-03 23:07 ` Ross Zwisler 2014-03-03 23:07 ` Ross Zwisler 2014-03-04 0:56 ` Dave Chinner 2014-03-04 0:56 ` Dave Chinner 2014-03-20 19:38 ` Matthew Wilcox 2014-03-20 19:38 ` Matthew Wilcox 2014-03-20 23:55 ` Dave Chinner 2014-03-20 23:55 ` Dave Chinner 2014-02-25 14:18 ` [PATCH v6 08/22] Replace xip_truncate_page with dax_truncate_page Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 09/22] Remove mm/filemap_xip.c Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 10/22] Remove get_xip_mem Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 11/22] Replace ext2_clear_xip_target with dax_clear_blocks Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 12/22] ext2: Remove ext2_xip_verify_sb() Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 13/22] ext2: Remove ext2_use_xip Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 14/22] ext2: Remove xip.c and xip.h Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 15/22] Remove CONFIG_EXT2_FS_XIP and rename CONFIG_FS_XIP to CONFIG_FS_DAX Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 16/22] ext2: Remove ext2_aops_xip Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 17/22] Get rid of most mentions of XIP in ext2 Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 18/22] xip: Add xip_zero_page_range Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 19/22] ext4: Make ext4_block_zero_page_range static Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 20/22] ext4: Add DAX functionality Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` [PATCH v6 21/22] ext4: Fix typos Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox 2014-02-25 14:18 ` Matthew Wilcox [this message] 2014-02-25 14:18 ` [PATCH v6 22/22] dax: Add reporting of major faults Matthew Wilcox 2014-02-26 15:07 ` [PATCH v6 23/22] Bugfixes Matthew Wilcox 2014-02-26 15:07 ` Matthew Wilcox 2014-02-27 14:01 ` [PATCH v6 00/22] Support ext4 on NV-DIMMs Florian Weimer 2014-02-27 14:01 ` Florian Weimer 2014-02-27 16:29 ` Matthew Wilcox 2014-02-27 16:29 ` Matthew Wilcox 2014-02-27 16:36 ` Florian Weimer 2014-02-27 16:36 ` Florian Weimer 2014-03-02 8:22 ` Pavel Machek 2014-03-02 8:22 ` Pavel Machek [not found] ` <CF4DEE22.25C8F%matthew.r.wilcox@intel.com> 2014-03-18 18:45 ` [PATCH v6 20/22] ext4: Add DAX functionality Ross Zwisler 2014-03-18 18:45 ` Ross Zwisler
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=1393337918-28265-23-git-send-email-matthew.r.wilcox@intel.com \ --to=matthew.r.wilcox@intel.com \ --cc=linux-fsdevel@vger.kernel.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=willy@linux.intel.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.