All of lore.kernel.org
 help / color / mirror / Atom feed
From: Nikanth Karthikesan <knikanth@suse.de>
To: linux-mm@kvack.org, Andrew Morton <akpm@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org, Matt Mackall <mpm@selenic.com>,
	balbir@linux.vnet.ibm.com, rguenther@novell.com, matz@novell.com
Subject: [PATCH] After swapout/swapin private dirty mappings become clean
Date: Tue, 14 Sep 2010 16:40:55 +0530	[thread overview]
Message-ID: <201009141640.55650.knikanth@suse.de> (raw)

/proc/$pid/smaps broken: After swapout/swapin private dirty mappings become
clean.

When a page with private file mapping becomes dirty, the vma will be in both
i_mmap tree and anon_vma list. The /proc/$pid/smaps will account these pages
as dirty and backed by the file.

But when those dirty pages gets swapped out, and when they are read back from
swap, they would be marked as clean, as it should be, as they are part of swap
cache now.

But the /proc/$pid/smaps would report the vma as a mapping of a file and it is
clean. The pages are actually in same state i.e., dirty with respect to file
still, but which was once reported as dirty is now being reported as clean to
user-space.

This confuses tools like gdb which uses this information. Those tools think
that those pages were never modified and it creates problem when they create
dumps.

The file mapping of the vma also cannot be broken as pages never read earlier,
will still have to come from the file. Just that those dirty pages have become
clean anonymous pages.

During swaping in, restoring the exact state as dirty file-backed pages before
swapout would be useless, as there in no real bug. Breaking the vma with only
anonymous pages as seperate vmas unnecessary may not be a good thing as well.
So let us just export the information that a file-backed vma has anonymous
dirty pages.

Export this information in smaps by prepending file-names with "[anon]+", when
some of the pages in a file backed vma become anonymous.

Signed-off-by: Nikanth Karthikesan <knikanth@suse.de>

---

diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 439fc1f..68f9806 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -242,6 +242,8 @@ static void show_map_vma(struct seq_file *m, struct vm_area_struct *vma)
 	 */
 	if (file) {
 		pad_len_spaces(m, len);
+		if (vma->anon_vma)
+			seq_puts(m, "[anon]+");
 		seq_path(m, &file->f_path, "\n");
 	} else {
 		const char *name = arch_vma_name(vma);



WARNING: multiple messages have this Message-ID (diff)
From: Nikanth Karthikesan <knikanth@suse.de>
To: linux-mm@kvack.org, Andrew Morton <akpm@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org, Matt Mackall <mpm@selenic.com>,
	balbir@linux.vnet.ibm.com, rguenther@novell.com, matz@novell.com
Subject: [PATCH] After swapout/swapin private dirty mappings become clean
Date: Tue, 14 Sep 2010 16:40:55 +0530	[thread overview]
Message-ID: <201009141640.55650.knikanth@suse.de> (raw)

/proc/$pid/smaps broken: After swapout/swapin private dirty mappings become
clean.

When a page with private file mapping becomes dirty, the vma will be in both
i_mmap tree and anon_vma list. The /proc/$pid/smaps will account these pages
as dirty and backed by the file.

But when those dirty pages gets swapped out, and when they are read back from
swap, they would be marked as clean, as it should be, as they are part of swap
cache now.

But the /proc/$pid/smaps would report the vma as a mapping of a file and it is
clean. The pages are actually in same state i.e., dirty with respect to file
still, but which was once reported as dirty is now being reported as clean to
user-space.

This confuses tools like gdb which uses this information. Those tools think
that those pages were never modified and it creates problem when they create
dumps.

The file mapping of the vma also cannot be broken as pages never read earlier,
will still have to come from the file. Just that those dirty pages have become
clean anonymous pages.

During swaping in, restoring the exact state as dirty file-backed pages before
swapout would be useless, as there in no real bug. Breaking the vma with only
anonymous pages as seperate vmas unnecessary may not be a good thing as well.
So let us just export the information that a file-backed vma has anonymous
dirty pages.

Export this information in smaps by prepending file-names with "[anon]+", when
some of the pages in a file backed vma become anonymous.

Signed-off-by: Nikanth Karthikesan <knikanth@suse.de>

---

diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 439fc1f..68f9806 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -242,6 +242,8 @@ static void show_map_vma(struct seq_file *m, struct vm_area_struct *vma)
 	 */
 	if (file) {
 		pad_len_spaces(m, len);
+		if (vma->anon_vma)
+			seq_puts(m, "[anon]+");
 		seq_path(m, &file->f_path, "\n");
 	} else {
 		const char *name = arch_vma_name(vma);


--
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>

             reply	other threads:[~2010-09-14 11:08 UTC|newest]

Thread overview: 86+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-09-14 11:10 Nikanth Karthikesan [this message]
2010-09-14 11:10 ` [PATCH] After swapout/swapin private dirty mappings become clean Nikanth Karthikesan
2010-09-14 11:33 ` Richard Guenther
2010-09-14 11:33   ` Richard Guenther
2010-09-14 17:12   ` Nikanth Karthikesan
2010-09-14 17:12     ` Nikanth Karthikesan
2010-09-14 17:14     ` [PATCH v2] After swapout/swapin private dirty mappings are reported clean in smaps Nikanth Karthikesan
2010-09-14 17:14       ` Nikanth Karthikesan
2010-09-15  0:26       ` KOSAKI Motohiro
2010-09-15  0:26         ` KOSAKI Motohiro
2010-09-15  4:38         ` Nikanth Karthikesan
2010-09-15  4:38           ` Nikanth Karthikesan
2010-09-15  4:48           ` KOSAKI Motohiro
2010-09-15  4:48             ` KOSAKI Motohiro
2010-09-15  5:04             ` Nikanth Karthikesan
2010-09-15  5:04               ` Nikanth Karthikesan
2010-09-15  5:20               ` KOSAKI Motohiro
2010-09-15  5:20                 ` KOSAKI Motohiro
2010-09-15  6:31                 ` Nikanth Karthikesan
2010-09-15  6:31                   ` Nikanth Karthikesan
2010-09-15 14:09                   ` Balbir Singh
2010-09-15 14:09                     ` Balbir Singh
2010-09-15 14:14                     ` Richard Guenther
2010-09-15 14:14                       ` Richard Guenther
2010-09-15 14:46                       ` Matt Mackall
2010-09-15 14:46                         ` Matt Mackall
2010-09-15 14:53                         ` Richard Guenther
2010-09-15 14:53                           ` Richard Guenther
2010-09-15 17:24                           ` Matt Mackall
2010-09-15 17:24                             ` Matt Mackall
2010-09-15 19:08                             ` Richard Guenther
2010-09-15 19:08                               ` Richard Guenther
2010-09-15 19:18                             ` Hugh Dickins
2010-09-15 19:18                               ` Hugh Dickins
2010-09-15 19:46                               ` Matt Mackall
2010-09-15 19:46                                 ` Matt Mackall
2010-09-15 19:53                                 ` Richard Guenther
2010-09-15 19:53                                   ` Richard Guenther
2010-09-15 21:47                                 ` Hugh Dickins
2010-09-15 21:47                                   ` Hugh Dickins
2010-09-16  3:26                                   ` [PATCH] Export amount of anonymous memory in a mapping via smaps Nikanth Karthikesan
2010-09-16  3:26                                     ` Nikanth Karthikesan
2010-09-16  3:52                                     ` KOSAKI Motohiro
2010-09-16  3:52                                       ` KOSAKI Motohiro
2010-09-16  6:04                                       ` [PATCH] Document the new Anonymous field in smaps Nikanth Karthikesan
2010-09-16  6:04                                         ` Nikanth Karthikesan
2010-09-16  6:34                                         ` [PATCH] smaps: fix dirty pages accounting KOSAKI Motohiro
2010-09-16  6:34                                           ` KOSAKI Motohiro
2010-09-16 16:56                                           ` Hugh Dickins
2010-09-16 16:56                                             ` Hugh Dickins
2010-09-16 16:50                                         ` [PATCH] Document the new Anonymous field in smaps Hugh Dickins
2010-09-17  6:04                                           ` [PATCH v2] " Nikanth Karthikesan
2010-09-17  6:04                                             ` Nikanth Karthikesan
2010-09-20  7:11                                             ` Hugh Dickins
2010-09-20 19:24                                               ` Matt Mackall
2010-09-20 19:24                                                 ` Matt Mackall
2010-09-16 16:40                                     ` [PATCH] Export amount of anonymous memory in a mapping via smaps Hugh Dickins
2010-09-16 16:40                                       ` Hugh Dickins
2010-09-15 17:41                       ` [PATCH v2] After swapout/swapin private dirty mappings are reported clean in smaps Balbir Singh
2010-09-15 17:41                         ` Balbir Singh
2010-09-19 17:37                       ` Nikanth Karthikesan
2010-09-19 17:37                         ` Nikanth Karthikesan
2010-09-19 17:38                         ` [PATCH] Document /proc/pid/pagemap in Documentation/filesystems/proc.txt Nikanth Karthikesan
2010-09-19 17:38                           ` Nikanth Karthikesan
2010-09-20 21:27                           ` Matt Mackall
2010-09-20 21:27                             ` Matt Mackall
2010-09-20  5:24                         ` [PATCH v2] After swapout/swapin private dirty mappings are reported clean in smaps Nikanth Karthikesan
2010-09-20  5:24                           ` Nikanth Karthikesan
2010-09-20 14:30                         ` Richard Guenther
2010-09-20 14:30                           ` Richard Guenther
2010-09-15  0:24 ` [PATCH] After swapout/swapin private dirty mappings become clean KOSAKI Motohiro
2010-09-15  0:24   ` KOSAKI Motohiro
2010-09-15  4:37   ` Nikanth Karthikesan
2010-09-15  4:37     ` Nikanth Karthikesan
2010-09-15  4:46     ` KOSAKI Motohiro
2010-09-15  4:46       ` KOSAKI Motohiro
2010-09-15  5:00       ` Nikanth Karthikesan
2010-09-15  5:00         ` Nikanth Karthikesan
2010-09-15  5:15         ` KOSAKI Motohiro
2010-09-15  5:15           ` KOSAKI Motohiro
2010-09-15  6:29           ` Nikanth Karthikesan
2010-09-15  6:29             ` Nikanth Karthikesan
2010-09-15  8:40         ` Richard Guenther
2010-09-15  8:40           ` Richard Guenther
2010-09-16  1:29           ` KOSAKI Motohiro
2010-09-16  1:29             ` KOSAKI Motohiro

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=201009141640.55650.knikanth@suse.de \
    --to=knikanth@suse.de \
    --cc=akpm@linux-foundation.org \
    --cc=balbir@linux.vnet.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=matz@novell.com \
    --cc=mpm@selenic.com \
    --cc=rguenther@novell.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: link
Be 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.