[31/31] HWPOISON: Add a madvise() injector for soft page offlining
in [Kernel]
|
Prev: pcmcia: fix 'driver ... did not release config properly' warning
Next: [PATCH] x86: Fix rebooting on Dell Precision WorkStation T7400
From: Michael Kerrisk on 19 Jun 2010 08:40 Hi Andi, On Tue, Dec 8, 2009 at 11:16 PM, Andi Kleen <andi(a)firstfloor.org> wrote: > > Process based injection is much easier to handle for test programs, > who can first bring a page into a specific state and then test. > So add a new MADV_SOFT_OFFLINE to soft offline a page, similar > to the existing hard offline injector. I see that this made its way into 2.6.33. Could you write a short piece on it for the madvise.2 man page? Thanks, Michael > Signed-off-by: Andi Kleen <ak(a)linux.intel.com> > > --- > �include/asm-generic/mman-common.h | � �1 + > �mm/madvise.c � � � � � � � � � � �| � 15 ++++++++++++--- > �2 files changed, 13 insertions(+), 3 deletions(-) > > Index: linux/include/asm-generic/mman-common.h > =================================================================== > --- linux.orig/include/asm-generic/mman-common.h > +++ linux/include/asm-generic/mman-common.h > @@ -35,6 +35,7 @@ > �#define MADV_DONTFORK �10 � � � � � � �/* don't inherit across fork */ > �#define MADV_DOFORK � �11 � � � � � � �/* do inherit across fork */ > �#define MADV_HWPOISON �100 � � � � � � /* poison a page for testing */ > +#define MADV_SOFT_OFFLINE 101 � � � � �/* soft offline page for testing */ > > �#define MADV_MERGEABLE � 12 � � � � � �/* KSM may merge identical pages */ > �#define MADV_UNMERGEABLE 13 � � � � � �/* KSM may not merge identical pages */ > Index: linux/mm/madvise.c > =================================================================== > --- linux.orig/mm/madvise.c > +++ linux/mm/madvise.c > @@ -9,6 +9,7 @@ > �#include <linux/pagemap.h> > �#include <linux/syscalls.h> > �#include <linux/mempolicy.h> > +#include <linux/page-isolation.h> > �#include <linux/hugetlb.h> > �#include <linux/sched.h> > �#include <linux/ksm.h> > @@ -222,7 +223,7 @@ static long madvise_remove(struct vm_are > �/* > �* Error injection support for memory error handling. > �*/ > -static int madvise_hwpoison(unsigned long start, unsigned long end) > +static int madvise_hwpoison(int bhv, unsigned long start, unsigned long end) > �{ > � � � �int ret = 0; > > @@ -233,6 +234,14 @@ static int madvise_hwpoison(unsigned lon > � � � � � � � �int ret = get_user_pages_fast(start, 1, 0, &p); > � � � � � � � �if (ret != 1) > � � � � � � � � � � � �return ret; > + � � � � � � � if (bhv == MADV_SOFT_OFFLINE) { > + � � � � � � � � � � � printk(KERN_INFO "Soft offlining page %lx at %lx\n", > + � � � � � � � � � � � � � � � page_to_pfn(p), start); > + � � � � � � � � � � � ret = soft_offline_page(p, MF_COUNT_INCREASED); > + � � � � � � � � � � � if (ret) > + � � � � � � � � � � � � � � � break; > + � � � � � � � � � � � continue; > + � � � � � � � } > � � � � � � � �printk(KERN_INFO "Injecting memory failure for page %lx at %lx\n", > � � � � � � � � � � � page_to_pfn(p), start); > � � � � � � � �/* Ignore return value for now */ > @@ -333,8 +342,8 @@ SYSCALL_DEFINE3(madvise, unsigned long, > � � � �size_t len; > > �#ifdef CONFIG_MEMORY_FAILURE > - � � � if (behavior == MADV_HWPOISON) > - � � � � � � � return madvise_hwpoison(start, start+len_in); > + � � � if (behavior == MADV_HWPOISON || behavior == MADV_SOFT_OFFLINE) > + � � � � � � � return madvise_hwpoison(behavior, start, start+len_in); > �#endif > � � � �if (!madvise_behavior_valid(behavior)) > � � � � � � � �return error; > > -- > To unsubscribe, send a message with 'unsubscribe linux-mm' in > the body to majordomo(a)kvack.org. �For more info on Linux MM, > see: http://www.linux-mm.org/ . > Don't email: <a href=mailto:"dont(a)kvack.org"> email(a)kvack.org </a> > -- Michael Kerrisk Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/ Author of "The Linux Programming Interface", http://blog.man7.org/ -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Michael Kerrisk on 19 Jun 2010 09:30 Hi Andi, Thanks for this. Some comments below. On Sat, Jun 19, 2010 at 3:20 PM, Andi Kleen <andi(a)firstfloor.org> wrote: > On Sat, Jun 19, 2010 at 02:36:28PM +0200, Michael Kerrisk wrote: >> Hi Andi, >> >> On Tue, Dec 8, 2009 at 11:16 PM, Andi Kleen <andi(a)firstfloor.org> wrote: >> > >> > Process based injection is much easier to handle for test programs, >> > who can first bring a page into a specific state and then test. >> > So add a new MADV_SOFT_OFFLINE to soft offline a page, similar >> > to the existing hard offline injector. >> >> I see that this made its way into 2.6.33. Could you write a short >> piece on it for the madvise.2 man page? > > Also fixed the previous snippet slightly. (thanks) > commit edb43354f0ffc04bf4f23f01261f9ea9f43e0d3d > Author: Andi Kleen <ak(a)linux.intel.com> > Date: � Sat Jun 19 15:19:28 2010 +0200 > > � �MADV_SOFT_OFFLINE > > � �Signed-off-by: Andi Kleen <ak(a)linux.intel.com> > > diff --git a/man2/madvise.2 b/man2/madvise.2 > index db29feb..9dccd97 100644 > --- a/man2/madvise.2 > +++ b/man2/madvise.2 > @@ -154,7 +154,15 @@ processes. > �This operation may result in the calling process receiving a > �.B SIGBUS > �and the page being unmapped. > -This feature is intended for memory testing. > +This feature is intended for testing of memory error handling code. > +This feature is only available if the kernel was configured with > +.BR CONFIG_MEMORY_FAILURE . > +.TP > +.BR MADV_SOFT_OFFLINE " (Since Linux 2.6.33) > +Soft offline a page. This will result in the memory of the page > +being copied to a new page and original page be offlined. The operation Can you explain the term "offlined" please. > +should be transparent to the calling process. Does "should be transparent" mean "is normally invisible"? Thanks, Michael > +This feature is intended for testing of memory error handling code. > �This feature is only available if the kernel was configured with > �.BR CONFIG_MEMORY_FAILURE . > �.TP > > -- Michael Kerrisk Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/ Author of "The Linux Programming Interface" http://blog.man7.org/ -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Michael Kerrisk on 19 Jun 2010 09:50 Hi Andi, On Sat, Jun 19, 2010 at 3:30 PM, Andi Kleen <andi(a)firstfloor.org> wrote: > On Sat, Jun 19, 2010 at 03:25:16PM +0200, Michael Kerrisk wrote: >> Hi Andi, >> >> Thanks for this. Some comments below. >> >> On Sat, Jun 19, 2010 at 3:20 PM, Andi Kleen <andi(a)firstfloor.org> wrote: >> > On Sat, Jun 19, 2010 at 02:36:28PM +0200, Michael Kerrisk wrote: >> >> Hi Andi, >> >> >> >> On Tue, Dec 8, 2009 at 11:16 PM, Andi Kleen <andi(a)firstfloor.org> wrote: >> >> > >> >> > Process based injection is much easier to handle for test programs, >> >> > who can first bring a page into a specific state and then test. >> >> > So add a new MADV_SOFT_OFFLINE to soft offline a page, similar >> >> > to the existing hard offline injector. >> >> >> >> I see that this made its way into 2.6.33. Could you write a short >> >> piece on it for the madvise.2 man page? >> > >> > Also fixed the previous snippet slightly. >> >> (thanks) >> >> > commit edb43354f0ffc04bf4f23f01261f9ea9f43e0d3d >> > Author: Andi Kleen <ak(a)linux.intel.com> >> > Date: � Sat Jun 19 15:19:28 2010 +0200 >> > >> > � �MADV_SOFT_OFFLINE >> > >> > � �Signed-off-by: Andi Kleen <ak(a)linux.intel.com> >> > >> > diff --git a/man2/madvise.2 b/man2/madvise.2 >> > index db29feb..9dccd97 100644 >> > --- a/man2/madvise.2 >> > +++ b/man2/madvise.2 >> > @@ -154,7 +154,15 @@ processes. >> > �This operation may result in the calling process receiving a >> > �.B SIGBUS >> > �and the page being unmapped. >> > -This feature is intended for memory testing. >> > +This feature is intended for testing of memory error handling code. >> > +This feature is only available if the kernel was configured with >> > +.BR CONFIG_MEMORY_FAILURE . >> > +.TP >> > +.BR MADV_SOFT_OFFLINE " (Since Linux 2.6.33) >> > +Soft offline a page. This will result in the memory of the page >> > +being copied to a new page and original page be offlined. The operation >> >> Can you explain the term "offlined" please. > > The memory is not used anymore and taken out of normal > memory management (until unpoisoned) Is there a userspace operation to unpoison (i.e., reverse MADV_SOFT_OFFLINE)? I ask because I wondered if there is something additional to be documented. > and the "HardwareCorrupted:" counter in /proc/meminfo increases > > (don't put the later in, I'm thinking about changing that) Okay. >> >> > +should be transparent to the calling process. >> >> Does "should be transparent" mean "is normally invisible"? > > Yes. It's similar to being swapped out and swapped in again. Okay. Thanks, Michael -- Michael Kerrisk Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/ Author of "The Linux Programming Interface" http://blog.man7.org/ -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Michael Kerrisk on 19 Jun 2010 10:20 Hi Andi, On Sat, Jun 19, 2010 at 4:09 PM, Andi Kleen <andi(a)firstfloor.org> wrote: > On Sat, Jun 19, 2010 at 03:43:28PM +0200, Michael Kerrisk wrote: >> Is there a userspace operation to unpoison (i.e., reverse MADV_SOFT_OFFLINE)? > > Yes, but it's only a debugfs interface currently. Okay -- thanks. >> I ask because I wondered if there is something additional to be documented. > > I don't think debugfs needs manpages atm. Okay. I edited your text somewhat. Could you please review the below. Cheers, Michael ..TP ..BR MADV_SOFT_OFFLINE " (Since Linux 2.6.33) Soft offline the pages in the range specified by ..I addr and ..IR length . This memory of each page in the specified range is copied to a new page, and the original page is offlined (i.e., no longer used, and taken out of normal memory management). The effect of the ..B MADV_SOFT_OFFLINE operation is normally invisible to (i.e., does not change the semantics of) the calling process. This feature is intended for testing of memory error-handling code; it is only available if the kernel was configured with ..BR CONFIG_MEMORY_FAILURE . -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Michael Kerrisk on 20 Jun 2010 02:30
Hi Andi, On Sat, Jun 19, 2010 at 9:52 PM, Andi Kleen <andi(a)firstfloor.org> wrote: >> .TP >> .BR MADV_SOFT_OFFLINE " (Since Linux 2.6.33) >> Soft offline the pages in the range specified by >> .I addr >> and >> .IR length . >> This memory of each page in the specified range is copied to a new page, > > Actually there are some cases where it's also dropped if it's cached page. > > Perhaps better would be something more fuzzy like > > "the contents are preserved" The problem to me is that this gets so fuzzy that it's hard to understand the meaning (I imagine many readers will ask: "What does it mean that the contents are preserved"?). Would you be able to come up with a wording that is a little miore detailed? >> and the original page is offlined >> (i.e., no longer used, and taken out of normal memory management). Thanks, Michael -- Michael Kerrisk Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/ Author of "The Linux Programming Interface" http://blog.man7.org/ -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |