linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Wu Fengguang <fengguang.wu@intel.com>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
	Wu Fengguang <fengguang.wu@intel.com>,
	Hugh Dickins <hugh.dickins@tiscali.co.uk>,
	Nick Piggin <npiggin@suse.de>, Andi Kleen <andi@firstfloor.org>,
	"riel@redhat.com" <riel@redhat.com>,
	"chris.mason@oracle.com" <chris.mason@oracle.com>,
	"linux-mm@kvack.org" <linux-mm@kvack.org>
Subject: [PATCH 4/5] HWPOISON: report sticky EIO for poisoned file
Date: Thu, 11 Jun 2009 22:22:43 +0800	[thread overview]
Message-ID: <20090611144430.813191526@intel.com> (raw)
In-Reply-To: 20090611142239.192891591@intel.com

[-- Attachment #1: hwpoison-more-sticky-eio.patch --]
[-- Type: text/plain, Size: 2787 bytes --]

This makes the EIO reports on write(), fsync(), or the NFS close()
sticky enough. The only way to get rid of it may be

	echo 3 > /proc/sys/vm/drop_caches

Note that the impacted process will only be killed if it mapped the page.
XXX
via read()/write()/fsync() instead of memory mapped reads/writes, simply
because it's very hard to find them.

Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
---
 include/linux/pagemap.h |   13 +++++++++++++
 mm/filemap.c            |   11 +++++++++++
 mm/memory-failure.c     |    2 +-
 3 files changed, 25 insertions(+), 1 deletion(-)

--- sound-2.6.orig/include/linux/pagemap.h
+++ sound-2.6/include/linux/pagemap.h
@@ -23,6 +23,7 @@ enum mapping_flags {
 	AS_ENOSPC	= __GFP_BITS_SHIFT + 1,	/* ENOSPC on async write */
 	AS_MM_ALL_LOCKS	= __GFP_BITS_SHIFT + 2,	/* under mm_take_all_locks() */
 	AS_UNEVICTABLE	= __GFP_BITS_SHIFT + 3,	/* e.g., ramdisk, SHM_LOCK */
+	AS_HWPOISON	= __GFP_BITS_SHIFT + 4,	/* hardware memory corruption */
 };
 
 static inline void mapping_set_error(struct address_space *mapping, int error)
@@ -52,6 +53,18 @@ static inline int mapping_unevictable(st
 	return !!mapping;
 }
 
+#ifdef CONFIG_MEMORY_FAILURE
+static inline int mapping_hwpoison(struct address_space *mapping)
+{
+	return test_bit(AS_HWPOISON, &mapping->flags);
+}
+#else
+static inline int mapping_hwpoison(struct address_space *mapping)
+{
+	return 0;
+}
+#endif
+
 static inline gfp_t mapping_gfp_mask(struct address_space * mapping)
 {
 	return (__force gfp_t)mapping->flags & __GFP_BITS_MASK;
--- sound-2.6.orig/mm/filemap.c
+++ sound-2.6/mm/filemap.c
@@ -302,6 +302,8 @@ int wait_on_page_writeback_range(struct 
 		ret = -ENOSPC;
 	if (test_and_clear_bit(AS_EIO, &mapping->flags))
 		ret = -EIO;
+	if (mapping_hwpoison(mapping))
+		ret = -EIO;
 
 	return ret;
 }
@@ -460,6 +462,15 @@ int add_to_page_cache_locked(struct page
 
 	VM_BUG_ON(!PageLocked(page));
 
+	/*
+	 * Hardware corrupted page will be removed from mapping,
+	 * so we want to deny (possibly) reloading the old data.
+	 */
+	if (unlikely(mapping_hwpoison(mapping))) {
+		error = -EIO;
+		goto out;
+	}
+
 	error = mem_cgroup_cache_charge(page, current->mm,
 					gfp_mask & GFP_RECLAIM_MASK);
 	if (error)
--- sound-2.6.orig/mm/memory-failure.c
+++ sound-2.6/mm/memory-failure.c
@@ -184,7 +184,7 @@ static int me_pagecache_dirty(struct pag
 		 * the first EIO, but we're not worse than other parts
 		 * of the kernel.
 		 */
-		mapping_set_error(mapping, EIO);
+		set_bit(AS_HWPOISON, &mapping->flags);
 	}
 
 	return me_pagecache_clean(p, pfn);

-- 

--
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>

  parent reply	other threads:[~2009-06-11 14:52 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-06-11 14:22 [PATCH 0/5] [RFC] HWPOISON incremental fixes Wu Fengguang
2009-06-11 14:22 ` [PATCH 1/5] HWPOISON: define VM_FAULT_HWPOISON to 0 when feature is disabled Wu Fengguang
2009-06-11 15:44   ` Rik van Riel
2009-06-12 10:00   ` Andi Kleen
2009-06-12 13:15     ` Wu Fengguang
2009-06-12 11:22   ` Ingo Molnar
2009-06-12 12:57     ` Wu Fengguang
2009-06-12 13:17       ` Ingo Molnar
2009-06-12 13:33         ` Wu Fengguang
2009-06-12 15:36           ` Ingo Molnar
2009-06-12 16:14             ` Wu Fengguang
2009-06-12 18:07               ` Alan Cox
2009-06-12 17:55             ` Theodore Tso
2009-06-12 13:58         ` Andi Kleen
2009-06-12 15:28         ` Linus Torvalds
2009-06-12 15:35           ` Ingo Molnar
2009-06-12 16:05             ` Rik van Riel
2009-06-12 16:37             ` H. Peter Anvin
2009-06-12 16:48               ` Ingo Molnar
2009-06-15  7:04               ` Nick Piggin
2009-06-15  6:52             ` Nick Piggin
2009-06-16 20:27               ` Russ Anderson
2009-06-17  7:51                 ` Nick Piggin
2009-06-12 15:45         ` Ingo Molnar
2009-06-12 16:12           ` Linus Torvalds
2009-06-11 14:22 ` [PATCH 2/5] HWPOISON: fix tasklist_lock/anon_vma locking order Wu Fengguang
2009-06-11 15:59   ` Rik van Riel
2009-06-12 10:03   ` Andi Kleen
2009-06-12 10:07     ` Nick Piggin
2009-06-12 13:27     ` Wu Fengguang
2009-06-12 14:04       ` Wu Fengguang
2009-06-11 14:22 ` [PATCH 3/5] HWPOISON: remove early kill option for now Wu Fengguang
2009-06-11 16:06   ` Rik van Riel
2009-06-12  9:59   ` Andi Kleen
2009-06-11 14:22 ` Wu Fengguang [this message]
2009-06-11 16:31   ` [PATCH 4/5] HWPOISON: report sticky EIO for poisoned file Rik van Riel
2009-06-12 10:07   ` Andi Kleen
2009-06-12 13:41     ` Wu Fengguang
2009-06-11 14:22 ` [PATCH 5/5] HWPOISON: use the safer invalidate page for possible metadata pages Wu Fengguang
2009-06-11 16:36   ` Rik van Riel
2009-06-12 10:56 ` [PATCH 0/5] [RFC] HWPOISON incremental fixes Andi Kleen
2009-06-12 13:59   ` Wu Fengguang

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=20090611144430.813191526@intel.com \
    --to=fengguang.wu@intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=andi@firstfloor.org \
    --cc=chris.mason@oracle.com \
    --cc=hugh.dickins@tiscali.co.uk \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=npiggin@suse.de \
    --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: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).