* [PATCH v6 00/13] mm: Teach memory_failure() about ZONE_DEVICE pages
@ 2018-07-14 4:49 Dan Williams
2018-07-14 4:50 ` [PATCH v6 06/13] mm, dev_pagemap: Do not clear ->mapping on final put Dan Williams
2018-07-19 17:57 ` [PATCH v6 00/13] mm: Teach memory_failure() about ZONE_DEVICE pages Dave Jiang
0 siblings, 2 replies; 6+ messages in thread
From: Dan Williams @ 2018-07-14 4:49 UTC (permalink / raw)
To: linux-nvdimm
Cc: linux-edac, Tony Luck, Borislav Petkov, stable, Jan Kara,
H. Peter Anvin, x86, Thomas Gleixner, Christoph Hellwig,
Ross Zwisler, Ingo Molnar, Michal Hocko, Matthew Wilcox,
Naoya Horiguchi, Jérôme Glisse, Andrew Morton,
Souptick Joarder, linux-fsdevel, linux-mm, linux-kernel
Changes since v5 [1]:
* Move put_page() before memory_failure() in madvise_inject_error()
(Naoya)
* The previous change uncovered a latent bug / broken assumption in
__put_devmap_managed_page(). We need to preserve page->mapping for
dax pages when they go idle.
* Rename mapping_size() to dev_pagemap_mapping_size() (Naoya)
* Catch and fail attempts to soft-offline dax pages (Naoya)
* Collect Naoya's ack on "mm, memory_failure: Collect mapping size in
collect_procs()"
[1]: https://lists.01.org/pipermail/linux-nvdimm/2018-July/016682.html
---
As it stands, memory_failure() gets thoroughly confused by dev_pagemap
backed mappings. The recovery code has specific enabling for several
possible page states and needs new enabling to handle poison in dax
mappings.
In order to support reliable reverse mapping of user space addresses:
1/ Add new locking in the memory_failure() rmap path to prevent races
that would typically be handled by the page lock.
2/ Since dev_pagemap pages are hidden from the page allocator and the
"compound page" accounting machinery, add a mechanism to determine the
size of the mapping that encompasses a given poisoned pfn.
3/ Given pmem errors can be repaired, change the speculatively accessed
poison protection, mce_unmap_kpfn(), to be reversible and otherwise
allow ongoing access from the kernel.
A side effect of this enabling is that MADV_HWPOISON becomes usable for
dax mappings, however the primary motivation is to allow the system to
survive userspace consumption of hardware-poison via dax. Specifically
the current behavior is:
mce: Uncorrected hardware memory error in user-access at af34214200
{1}[Hardware Error]: It has been corrected by h/w and requires no further action
mce: [Hardware Error]: Machine check events logged
{1}[Hardware Error]: event severity: corrected
Memory failure: 0xaf34214: reserved kernel page still referenced by 1 users
[..]
Memory failure: 0xaf34214: recovery action for reserved kernel page: Failed
mce: Memory error not recovered
<reboot>
...and with these changes:
Injecting memory failure for pfn 0x20cb00 at process virtual address 0x7f763dd00000
Memory failure: 0x20cb00: Killing dax-pmd:5421 due to hardware memory corruption
Memory failure: 0x20cb00: recovery action for dax page: Recovered
Given all the cross dependencies I propose taking this through
nvdimm.git with acks from Naoya, x86/core, x86/RAS, and of course dax
folks.
---
Dan Williams (13):
device-dax: Convert to vmf_insert_mixed and vm_fault_t
device-dax: Enable page_mapping()
device-dax: Set page->index
filesystem-dax: Set page->index
mm, madvise_inject_error: Disable MADV_SOFT_OFFLINE for ZONE_DEVICE pages
mm, dev_pagemap: Do not clear ->mapping on final put
mm, madvise_inject_error: Let memory_failure() optionally take a page reference
mm, memory_failure: Collect mapping size in collect_procs()
filesystem-dax: Introduce dax_lock_mapping_entry()
mm, memory_failure: Teach memory_failure() about dev_pagemap pages
x86/mm/pat: Prepare {reserve,free}_memtype() for "decoy" addresses
x86/memory_failure: Introduce {set,clear}_mce_nospec()
libnvdimm, pmem: Restore page attributes when clearing errors
arch/x86/include/asm/set_memory.h | 42 ++++++
arch/x86/kernel/cpu/mcheck/mce-internal.h | 15 --
arch/x86/kernel/cpu/mcheck/mce.c | 38 -----
arch/x86/mm/pat.c | 16 ++
drivers/dax/device.c | 75 +++++++---
drivers/nvdimm/pmem.c | 26 ++++
drivers/nvdimm/pmem.h | 13 ++
fs/dax.c | 125 ++++++++++++++++-
include/linux/dax.h | 13 ++
include/linux/huge_mm.h | 5 -
include/linux/mm.h | 1
include/linux/set_memory.h | 14 ++
kernel/memremap.c | 1
mm/hmm.c | 2
mm/huge_memory.c | 4 -
mm/madvise.c | 16 ++
mm/memory-failure.c | 210 +++++++++++++++++++++++------
17 files changed, 481 insertions(+), 135 deletions(-)
^ permalink raw reply [flat|nested] 6+ messages in thread* [PATCH v6 06/13] mm, dev_pagemap: Do not clear ->mapping on final put
2018-07-14 4:49 [PATCH v6 00/13] mm: Teach memory_failure() about ZONE_DEVICE pages Dan Williams
@ 2018-07-14 4:50 ` Dan Williams
2018-07-23 16:12 ` Dave Jiang
2018-07-19 17:57 ` [PATCH v6 00/13] mm: Teach memory_failure() about ZONE_DEVICE pages Dave Jiang
1 sibling, 1 reply; 6+ messages in thread
From: Dan Williams @ 2018-07-14 4:50 UTC (permalink / raw)
To: linux-nvdimm
Cc: stable, Jan Kara, Jérôme Glisse, Andrew Morton,
Ross Zwisler, hch, linux-fsdevel, linux-mm, linux-kernel
MEMORY_DEVICE_FS_DAX relies on typical page semantics whereby ->mapping
is only ever cleared by truncation, not final put.
Without this fix dax pages may forget their mapping association at the
end of every page pin event.
Move this atypical behavior that HMM wants into the HMM ->page_free()
callback.
Cc: <stable@vger.kernel.org>
Cc: Jan Kara <jack@suse.cz>
Cc: Jérôme Glisse <jglisse@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Fixes: d2c997c0f145 ("fs, dax: use page->mapping...")
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
kernel/memremap.c | 1 -
mm/hmm.c | 2 ++
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/kernel/memremap.c b/kernel/memremap.c
index 5857267a4af5..62603634a1d2 100644
--- a/kernel/memremap.c
+++ b/kernel/memremap.c
@@ -339,7 +339,6 @@ void __put_devmap_managed_page(struct page *page)
__ClearPageActive(page);
__ClearPageWaiters(page);
- page->mapping = NULL;
mem_cgroup_uncharge(page);
page->pgmap->page_free(page, page->pgmap->data);
diff --git a/mm/hmm.c b/mm/hmm.c
index de7b6bf77201..f9d1d89dec4d 100644
--- a/mm/hmm.c
+++ b/mm/hmm.c
@@ -963,6 +963,8 @@ static void hmm_devmem_free(struct page *page, void *data)
{
struct hmm_devmem *devmem = data;
+ page->mapping = NULL;
+
devmem->ops->free(devmem, page);
}
^ permalink raw reply related [flat|nested] 6+ messages in thread* Re: [PATCH v6 06/13] mm, dev_pagemap: Do not clear ->mapping on final put
2018-07-14 4:50 ` [PATCH v6 06/13] mm, dev_pagemap: Do not clear ->mapping on final put Dan Williams
@ 2018-07-23 16:12 ` Dave Jiang
2018-07-23 16:23 ` Jerome Glisse
0 siblings, 1 reply; 6+ messages in thread
From: Dave Jiang @ 2018-07-23 16:12 UTC (permalink / raw)
To: Jérôme Glisse
Cc: Dan Williams, linux-nvdimm, Jan Kara, linux-kernel, stable,
linux-mm, Jérôme Glisse, linux-fsdevel, Andrew Morton,
hch
Jerome,
Is it possible to get an ack for this? Thanks!
On 07/13/2018 09:50 PM, Dan Williams wrote:
> MEMORY_DEVICE_FS_DAX relies on typical page semantics whereby ->mapping
> is only ever cleared by truncation, not final put.
>
> Without this fix dax pages may forget their mapping association at the
> end of every page pin event.
>
> Move this atypical behavior that HMM wants into the HMM ->page_free()
> callback.
>
> Cc: <stable@vger.kernel.org>
> Cc: Jan Kara <jack@suse.cz>
> Cc: Jérôme Glisse <jglisse@redhat.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
> Fixes: d2c997c0f145 ("fs, dax: use page->mapping...")
> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
> ---
> kernel/memremap.c | 1 -
> mm/hmm.c | 2 ++
> 2 files changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/kernel/memremap.c b/kernel/memremap.c
> index 5857267a4af5..62603634a1d2 100644
> --- a/kernel/memremap.c
> +++ b/kernel/memremap.c
> @@ -339,7 +339,6 @@ void __put_devmap_managed_page(struct page *page)
> __ClearPageActive(page);
> __ClearPageWaiters(page);
>
> - page->mapping = NULL;
> mem_cgroup_uncharge(page);
>
> page->pgmap->page_free(page, page->pgmap->data);
> diff --git a/mm/hmm.c b/mm/hmm.c
> index de7b6bf77201..f9d1d89dec4d 100644
> --- a/mm/hmm.c
> +++ b/mm/hmm.c
> @@ -963,6 +963,8 @@ static void hmm_devmem_free(struct page *page, void *data)
> {
> struct hmm_devmem *devmem = data;
>
> + page->mapping = NULL;
> +
> devmem->ops->free(devmem, page);
> }
>
>
> _______________________________________________
> Linux-nvdimm mailing list
> Linux-nvdimm@lists.01.org
> https://lists.01.org/mailman/listinfo/linux-nvdimm
>
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [PATCH v6 06/13] mm, dev_pagemap: Do not clear ->mapping on final put
2018-07-23 16:12 ` Dave Jiang
@ 2018-07-23 16:23 ` Jerome Glisse
0 siblings, 0 replies; 6+ messages in thread
From: Jerome Glisse @ 2018-07-23 16:23 UTC (permalink / raw)
To: Dave Jiang
Cc: Dan Williams, linux-nvdimm, Jan Kara, linux-kernel, stable,
linux-mm, linux-fsdevel, Andrew Morton, hch
On Mon, Jul 23, 2018 at 09:12:06AM -0700, Dave Jiang wrote:
> Jerome,
> Is it possible to get an ack for this? Thanks!
>
> On 07/13/2018 09:50 PM, Dan Williams wrote:
> > MEMORY_DEVICE_FS_DAX relies on typical page semantics whereby ->mapping
> > is only ever cleared by truncation, not final put.
> >
> > Without this fix dax pages may forget their mapping association at the
> > end of every page pin event.
> >
> > Move this atypical behavior that HMM wants into the HMM ->page_free()
> > callback.
> >
> > Cc: <stable@vger.kernel.org>
> > Cc: Jan Kara <jack@suse.cz>
> > Cc: Andrew Morton <akpm@linux-foundation.org>
> > Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
> > Fixes: d2c997c0f145 ("fs, dax: use page->mapping...")
> > Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Acked-by: J�r�me Glisse <jglisse@redhat.com>
> > ---
> > kernel/memremap.c | 1 -
> > mm/hmm.c | 2 ++
> > 2 files changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/kernel/memremap.c b/kernel/memremap.c
> > index 5857267a4af5..62603634a1d2 100644
> > --- a/kernel/memremap.c
> > +++ b/kernel/memremap.c
> > @@ -339,7 +339,6 @@ void __put_devmap_managed_page(struct page *page)
> > __ClearPageActive(page);
> > __ClearPageWaiters(page);
> >
> > - page->mapping = NULL;
> > mem_cgroup_uncharge(page);
> >
> > page->pgmap->page_free(page, page->pgmap->data);
> > diff --git a/mm/hmm.c b/mm/hmm.c
> > index de7b6bf77201..f9d1d89dec4d 100644
> > --- a/mm/hmm.c
> > +++ b/mm/hmm.c
> > @@ -963,6 +963,8 @@ static void hmm_devmem_free(struct page *page, void *data)
> > {
> > struct hmm_devmem *devmem = data;
> >
> > + page->mapping = NULL;
> > +
> > devmem->ops->free(devmem, page);
> > }
> >
> >
> > _______________________________________________
> > Linux-nvdimm mailing list
> > Linux-nvdimm@lists.01.org
> > https://lists.01.org/mailman/listinfo/linux-nvdimm
> >
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v6 00/13] mm: Teach memory_failure() about ZONE_DEVICE pages
2018-07-14 4:49 [PATCH v6 00/13] mm: Teach memory_failure() about ZONE_DEVICE pages Dan Williams
2018-07-14 4:50 ` [PATCH v6 06/13] mm, dev_pagemap: Do not clear ->mapping on final put Dan Williams
@ 2018-07-19 17:57 ` Dave Jiang
2018-07-24 7:39 ` Ingo Molnar
1 sibling, 1 reply; 6+ messages in thread
From: Dave Jiang @ 2018-07-19 17:57 UTC (permalink / raw)
To: Ingo Molnar
Cc: Dan Williams, linux-nvdimm, Tony Luck, Jan Kara, Naoya Horiguchi,
linux-kernel, x86, Michal Hocko, Andrew Morton, stable,
Souptick Joarder, linux-mm, Jérôme Glisse,
Borislav Petkov, Matthew Wilcox, H. Peter Anvin, linux-fsdevel,
Thomas Gleixner, Christoph Hellwig, linux-edac
Ingo,
Is it possible to ack the x86 bits in this patch series? I'm hoping to
get this pulled through the libnvdimm tree for 4.19. Thanks!
On 07/13/2018 09:49 PM, Dan Williams wrote:
> Changes since v5 [1]:
> * Move put_page() before memory_failure() in madvise_inject_error()
> (Naoya)
> * The previous change uncovered a latent bug / broken assumption in
> __put_devmap_managed_page(). We need to preserve page->mapping for
> dax pages when they go idle.
> * Rename mapping_size() to dev_pagemap_mapping_size() (Naoya)
> * Catch and fail attempts to soft-offline dax pages (Naoya)
> * Collect Naoya's ack on "mm, memory_failure: Collect mapping size in
> collect_procs()"
>
> [1]: https://lists.01.org/pipermail/linux-nvdimm/2018-July/016682.html
>
> ---
>
> As it stands, memory_failure() gets thoroughly confused by dev_pagemap
> backed mappings. The recovery code has specific enabling for several
> possible page states and needs new enabling to handle poison in dax
> mappings.
>
> In order to support reliable reverse mapping of user space addresses:
>
> 1/ Add new locking in the memory_failure() rmap path to prevent races
> that would typically be handled by the page lock.
>
> 2/ Since dev_pagemap pages are hidden from the page allocator and the
> "compound page" accounting machinery, add a mechanism to determine the
> size of the mapping that encompasses a given poisoned pfn.
>
> 3/ Given pmem errors can be repaired, change the speculatively accessed
> poison protection, mce_unmap_kpfn(), to be reversible and otherwise
> allow ongoing access from the kernel.
>
> A side effect of this enabling is that MADV_HWPOISON becomes usable for
> dax mappings, however the primary motivation is to allow the system to
> survive userspace consumption of hardware-poison via dax. Specifically
> the current behavior is:
>
> mce: Uncorrected hardware memory error in user-access at af34214200
> {1}[Hardware Error]: It has been corrected by h/w and requires no further action
> mce: [Hardware Error]: Machine check events logged
> {1}[Hardware Error]: event severity: corrected
> Memory failure: 0xaf34214: reserved kernel page still referenced by 1 users
> [..]
> Memory failure: 0xaf34214: recovery action for reserved kernel page: Failed
> mce: Memory error not recovered
> <reboot>
>
> ...and with these changes:
>
> Injecting memory failure for pfn 0x20cb00 at process virtual address 0x7f763dd00000
> Memory failure: 0x20cb00: Killing dax-pmd:5421 due to hardware memory corruption
> Memory failure: 0x20cb00: recovery action for dax page: Recovered
>
> Given all the cross dependencies I propose taking this through
> nvdimm.git with acks from Naoya, x86/core, x86/RAS, and of course dax
> folks.
>
> ---
>
> Dan Williams (13):
> device-dax: Convert to vmf_insert_mixed and vm_fault_t
> device-dax: Enable page_mapping()
> device-dax: Set page->index
> filesystem-dax: Set page->index
> mm, madvise_inject_error: Disable MADV_SOFT_OFFLINE for ZONE_DEVICE pages
> mm, dev_pagemap: Do not clear ->mapping on final put
> mm, madvise_inject_error: Let memory_failure() optionally take a page reference
> mm, memory_failure: Collect mapping size in collect_procs()
> filesystem-dax: Introduce dax_lock_mapping_entry()
> mm, memory_failure: Teach memory_failure() about dev_pagemap pages
> x86/mm/pat: Prepare {reserve,free}_memtype() for "decoy" addresses
> x86/memory_failure: Introduce {set,clear}_mce_nospec()
> libnvdimm, pmem: Restore page attributes when clearing errors
>
>
> arch/x86/include/asm/set_memory.h | 42 ++++++
> arch/x86/kernel/cpu/mcheck/mce-internal.h | 15 --
> arch/x86/kernel/cpu/mcheck/mce.c | 38 -----
> arch/x86/mm/pat.c | 16 ++
> drivers/dax/device.c | 75 +++++++---
> drivers/nvdimm/pmem.c | 26 ++++
> drivers/nvdimm/pmem.h | 13 ++
> fs/dax.c | 125 ++++++++++++++++-
> include/linux/dax.h | 13 ++
> include/linux/huge_mm.h | 5 -
> include/linux/mm.h | 1
> include/linux/set_memory.h | 14 ++
> kernel/memremap.c | 1
> mm/hmm.c | 2
> mm/huge_memory.c | 4 -
> mm/madvise.c | 16 ++
> mm/memory-failure.c | 210 +++++++++++++++++++++++------
> 17 files changed, 481 insertions(+), 135 deletions(-)
> _______________________________________________
> Linux-nvdimm mailing list
> Linux-nvdimm@lists.01.org
> https://lists.01.org/mailman/listinfo/linux-nvdimm
>
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [PATCH v6 00/13] mm: Teach memory_failure() about ZONE_DEVICE pages
2018-07-19 17:57 ` [PATCH v6 00/13] mm: Teach memory_failure() about ZONE_DEVICE pages Dave Jiang
@ 2018-07-24 7:39 ` Ingo Molnar
0 siblings, 0 replies; 6+ messages in thread
From: Ingo Molnar @ 2018-07-24 7:39 UTC (permalink / raw)
To: Dave Jiang
Cc: Ingo Molnar, Dan Williams, linux-nvdimm, Tony Luck, Jan Kara,
Naoya Horiguchi, linux-kernel, x86, Michal Hocko, Andrew Morton,
stable, Souptick Joarder, linux-mm, Jérôme Glisse,
Borislav Petkov, Matthew Wilcox, H. Peter Anvin, linux-fsdevel,
Thomas Gleixner, Christoph Hellwig, linux-edac
* Dave Jiang <dave.jiang@intel.com> wrote:
> Ingo,
> Is it possible to ack the x86 bits in this patch series? I'm hoping to
> get this pulled through the libnvdimm tree for 4.19. Thanks!
With the minor typo fixed in the first patch, both patches are looking good to me:
Acked-by: Ingo Molnar <mingo@kernel.org>
Assuming that it gets properly tested, etc.
Thanks,
Ingo
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2018-07-24 7:39 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2018-07-14 4:49 [PATCH v6 00/13] mm: Teach memory_failure() about ZONE_DEVICE pages Dan Williams
2018-07-14 4:50 ` [PATCH v6 06/13] mm, dev_pagemap: Do not clear ->mapping on final put Dan Williams
2018-07-23 16:12 ` Dave Jiang
2018-07-23 16:23 ` Jerome Glisse
2018-07-19 17:57 ` [PATCH v6 00/13] mm: Teach memory_failure() about ZONE_DEVICE pages Dave Jiang
2018-07-24 7:39 ` Ingo Molnar
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).