* [PATCH v2] fs/hugetlbfs/inode.c: mm/memory-failure.c: fix hugetlbfs hwpoison handling
@ 2024-01-12 18:08 Sidhartha Kumar
2024-01-15 7:49 ` Muhammad Usama Anjum
` (3 more replies)
0 siblings, 4 replies; 6+ messages in thread
From: Sidhartha Kumar @ 2024-01-12 18:08 UTC (permalink / raw)
To: linux-kernel, linux-mm
Cc: akpm, usama.anjum, muchun.song, jiaqiyan, willy, linmiaohe,
naoya.horiguchi, shy828301, jthoughton, jane.chu, Sidhartha Kumar,
stable
has_extra_refcount() makes the assumption that the page cache adds a ref
count of 1 and subtracts this in the extra_pins case. Commit a08c7193e4f1
(mm/filemap: remove hugetlb special casing in filemap.c) modifies
__filemap_add_folio() by calling folio_ref_add(folio, nr); for all cases
(including hugtetlb) where nr is the number of pages in the folio. We
should adjust the number of references coming from the page cache by
subtracing the number of pages rather than 1.
In hugetlbfs_read_iter(), folio_test_has_hwpoisoned() is testing the wrong
flag as, in the hugetlb case, memory-failure code calls
folio_test_set_hwpoison() to indicate poison. folio_test_hwpoison() is the
correct function to test for that flag.
After these fixes, the hugetlb hwpoison read selftest passes all cases.
Fixes: a08c7193e4f1 ("mm/filemap: remove hugetlb special casing in filemap.c")
Closes: https://lore.kernel.org/linux-mm/20230713001833.3778937-1-jiaqiyan@google.com/T/#m8e1469119e5b831bbd05d495f96b842e4a1c5519
Cc: <stable@vger.kernel.org> # 6.7+
Signed-off-by: Sidhartha Kumar <sidhartha.kumar@oracle.com>
Reported-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
---
v1 -> v2:
move ref_count adjustment to if(extra_pins) block as that represents
ref counts from the page cache per Miaohe Lin.
fs/hugetlbfs/inode.c | 2 +-
mm/memory-failure.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index 36132c9125f9..3a248e4f7e93 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -340,7 +340,7 @@ static ssize_t hugetlbfs_read_iter(struct kiocb *iocb, struct iov_iter *to)
} else {
folio_unlock(folio);
- if (!folio_test_has_hwpoisoned(folio))
+ if (!folio_test_hwpoison(folio))
want = nr;
else {
/*
diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index d8c853b35dbb..ef7ae73b65bd 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -976,7 +976,7 @@ static bool has_extra_refcount(struct page_state *ps, struct page *p,
int count = page_count(p) - 1;
if (extra_pins)
- count -= 1;
+ count -= folio_nr_pages(page_folio(p));
if (count > 0) {
pr_err("%#lx: %s still referenced by %d users\n",
--
2.31.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH v2] fs/hugetlbfs/inode.c: mm/memory-failure.c: fix hugetlbfs hwpoison handling
2024-01-12 18:08 [PATCH v2] fs/hugetlbfs/inode.c: mm/memory-failure.c: fix hugetlbfs hwpoison handling Sidhartha Kumar
@ 2024-01-15 7:49 ` Muhammad Usama Anjum
2024-01-15 12:02 ` Muchun Song
` (2 subsequent siblings)
3 siblings, 0 replies; 6+ messages in thread
From: Muhammad Usama Anjum @ 2024-01-15 7:49 UTC (permalink / raw)
To: Sidhartha Kumar, linux-kernel, linux-mm
Cc: Muhammad Usama Anjum, akpm, muchun.song, jiaqiyan, willy,
linmiaohe, naoya.horiguchi, shy828301, jthoughton, jane.chu,
stable, kernel@collabora.com
On 1/12/24 11:08 PM, Sidhartha Kumar wrote:
> has_extra_refcount() makes the assumption that the page cache adds a ref
> count of 1 and subtracts this in the extra_pins case. Commit a08c7193e4f1
> (mm/filemap: remove hugetlb special casing in filemap.c) modifies
> __filemap_add_folio() by calling folio_ref_add(folio, nr); for all cases
> (including hugtetlb) where nr is the number of pages in the folio. We
> should adjust the number of references coming from the page cache by
> subtracing the number of pages rather than 1.
>
> In hugetlbfs_read_iter(), folio_test_has_hwpoisoned() is testing the wrong
> flag as, in the hugetlb case, memory-failure code calls
> folio_test_set_hwpoison() to indicate poison. folio_test_hwpoison() is the
> correct function to test for that flag.
>
> After these fixes, the hugetlb hwpoison read selftest passes all cases.
>
> Fixes: a08c7193e4f1 ("mm/filemap: remove hugetlb special casing in filemap.c")
> Closes: https://lore.kernel.org/linux-mm/20230713001833.3778937-1-jiaqiyan@google.com/T/#m8e1469119e5b831bbd05d495f96b842e4a1c5519
> Cc: <stable@vger.kernel.org> # 6.7+
> Signed-off-by: Sidhartha Kumar <sidhartha.kumar@oracle.com>
> Reported-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Tested-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
> ---
>
> v1 -> v2:
> move ref_count adjustment to if(extra_pins) block as that represents
> ref counts from the page cache per Miaohe Lin.
>
> fs/hugetlbfs/inode.c | 2 +-
> mm/memory-failure.c | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
> index 36132c9125f9..3a248e4f7e93 100644
> --- a/fs/hugetlbfs/inode.c
> +++ b/fs/hugetlbfs/inode.c
> @@ -340,7 +340,7 @@ static ssize_t hugetlbfs_read_iter(struct kiocb *iocb, struct iov_iter *to)
> } else {
> folio_unlock(folio);
>
> - if (!folio_test_has_hwpoisoned(folio))
> + if (!folio_test_hwpoison(folio))
> want = nr;
> else {
> /*
> diff --git a/mm/memory-failure.c b/mm/memory-failure.c
> index d8c853b35dbb..ef7ae73b65bd 100644
> --- a/mm/memory-failure.c
> +++ b/mm/memory-failure.c
> @@ -976,7 +976,7 @@ static bool has_extra_refcount(struct page_state *ps, struct page *p,
> int count = page_count(p) - 1;
>
> if (extra_pins)
> - count -= 1;
> + count -= folio_nr_pages(page_folio(p));
>
> if (count > 0) {
> pr_err("%#lx: %s still referenced by %d users\n",
--
BR,
Muhammad Usama Anjum
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2] fs/hugetlbfs/inode.c: mm/memory-failure.c: fix hugetlbfs hwpoison handling
2024-01-12 18:08 [PATCH v2] fs/hugetlbfs/inode.c: mm/memory-failure.c: fix hugetlbfs hwpoison handling Sidhartha Kumar
2024-01-15 7:49 ` Muhammad Usama Anjum
@ 2024-01-15 12:02 ` Muchun Song
2024-01-15 15:08 ` Matthew Wilcox
2024-01-16 2:36 ` Miaohe Lin
2024-01-16 2:43 ` Muchun Song
3 siblings, 1 reply; 6+ messages in thread
From: Muchun Song @ 2024-01-15 12:02 UTC (permalink / raw)
To: Sidhartha Kumar, linux-kernel, linux-mm
Cc: akpm, usama.anjum, jiaqiyan, willy, linmiaohe, naoya.horiguchi,
shy828301, jthoughton, jane.chu, stable
On 2024/1/13 02:08, Sidhartha Kumar wrote:
> has_extra_refcount() makes the assumption that the page cache adds a ref
> count of 1 and subtracts this in the extra_pins case. Commit a08c7193e4f1
> (mm/filemap: remove hugetlb special casing in filemap.c) modifies
> __filemap_add_folio() by calling folio_ref_add(folio, nr); for all cases
> (including hugtetlb) where nr is the number of pages in the folio. We
> should adjust the number of references coming from the page cache by
> subtracing the number of pages rather than 1.
>
> In hugetlbfs_read_iter(), folio_test_has_hwpoisoned() is testing the wrong
> flag as, in the hugetlb case, memory-failure code calls
> folio_test_set_hwpoison() to indicate poison. folio_test_hwpoison() is the
> correct function to test for that flag.
>
> After these fixes, the hugetlb hwpoison read selftest passes all cases.
>
> Fixes: a08c7193e4f1 ("mm/filemap: remove hugetlb special casing in filemap.c")
> Closes: https://lore.kernel.org/linux-mm/20230713001833.3778937-1-jiaqiyan@google.com/T/#m8e1469119e5b831bbd05d495f96b842e4a1c5519
> Cc: <stable@vger.kernel.org> # 6.7+
> Signed-off-by: Sidhartha Kumar <sidhartha.kumar@oracle.com>
> Reported-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
> ---
>
> v1 -> v2:
> move ref_count adjustment to if(extra_pins) block as that represents
> ref counts from the page cache per Miaohe Lin.
>
> fs/hugetlbfs/inode.c | 2 +-
> mm/memory-failure.c | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
> index 36132c9125f9..3a248e4f7e93 100644
> --- a/fs/hugetlbfs/inode.c
> +++ b/fs/hugetlbfs/inode.c
> @@ -340,7 +340,7 @@ static ssize_t hugetlbfs_read_iter(struct kiocb *iocb, struct iov_iter *to)
> } else {
> folio_unlock(folio);
>
> - if (!folio_test_has_hwpoisoned(folio))
> + if (!folio_test_hwpoison(folio))
> want = nr;
> else {
> /*
> diff --git a/mm/memory-failure.c b/mm/memory-failure.c
> index d8c853b35dbb..ef7ae73b65bd 100644
> --- a/mm/memory-failure.c
> +++ b/mm/memory-failure.c
> @@ -976,7 +976,7 @@ static bool has_extra_refcount(struct page_state *ps, struct page *p,
> int count = page_count(p) - 1;
>
> if (extra_pins)
> - count -= 1;
> + count -= folio_nr_pages(page_folio(p));
IIUC, It seems than this also fix shmem of THP, because
has_extra_refcount() is called
from me_pagecache_clean(), right?
>
> if (count > 0) {
> pr_err("%#lx: %s still referenced by %d users\n",
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2] fs/hugetlbfs/inode.c: mm/memory-failure.c: fix hugetlbfs hwpoison handling
2024-01-15 12:02 ` Muchun Song
@ 2024-01-15 15:08 ` Matthew Wilcox
0 siblings, 0 replies; 6+ messages in thread
From: Matthew Wilcox @ 2024-01-15 15:08 UTC (permalink / raw)
To: Muchun Song
Cc: Sidhartha Kumar, linux-kernel, linux-mm, akpm, usama.anjum,
jiaqiyan, linmiaohe, naoya.horiguchi, shy828301, jthoughton,
jane.chu, stable
On Mon, Jan 15, 2024 at 08:02:51PM +0800, Muchun Song wrote:
> On 2024/1/13 02:08, Sidhartha Kumar wrote:
> > has_extra_refcount() makes the assumption that the page cache adds a ref
> > count of 1 and subtracts this in the extra_pins case. Commit a08c7193e4f1
> > (mm/filemap: remove hugetlb special casing in filemap.c) modifies
> > __filemap_add_folio() by calling folio_ref_add(folio, nr); for all cases
> > (including hugtetlb) where nr is the number of pages in the folio. We
> > should adjust the number of references coming from the page cache by
> > subtracing the number of pages rather than 1.
> >
> > In hugetlbfs_read_iter(), folio_test_has_hwpoisoned() is testing the wrong
> > flag as, in the hugetlb case, memory-failure code calls
> > folio_test_set_hwpoison() to indicate poison. folio_test_hwpoison() is the
> > correct function to test for that flag.
> >
> > After these fixes, the hugetlb hwpoison read selftest passes all cases.
> >
> > Fixes: a08c7193e4f1 ("mm/filemap: remove hugetlb special casing in filemap.c")
> > Closes: https://lore.kernel.org/linux-mm/20230713001833.3778937-1-jiaqiyan@google.com/T/#m8e1469119e5b831bbd05d495f96b842e4a1c5519
> > Cc: <stable@vger.kernel.org> # 6.7+
> > Signed-off-by: Sidhartha Kumar <sidhartha.kumar@oracle.com>
> > Reported-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
> > ---
> >
> > v1 -> v2:
> > move ref_count adjustment to if(extra_pins) block as that represents
> > ref counts from the page cache per Miaohe Lin.
> >
> > fs/hugetlbfs/inode.c | 2 +-
> > mm/memory-failure.c | 2 +-
> > 2 files changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
> > index 36132c9125f9..3a248e4f7e93 100644
> > --- a/fs/hugetlbfs/inode.c
> > +++ b/fs/hugetlbfs/inode.c
> > @@ -340,7 +340,7 @@ static ssize_t hugetlbfs_read_iter(struct kiocb *iocb, struct iov_iter *to)
> > } else {
> > folio_unlock(folio);
> > - if (!folio_test_has_hwpoisoned(folio))
> > + if (!folio_test_hwpoison(folio))
> > want = nr;
> > else {
> > /*
> > diff --git a/mm/memory-failure.c b/mm/memory-failure.c
> > index d8c853b35dbb..ef7ae73b65bd 100644
> > --- a/mm/memory-failure.c
> > +++ b/mm/memory-failure.c
> > @@ -976,7 +976,7 @@ static bool has_extra_refcount(struct page_state *ps, struct page *p,
> > int count = page_count(p) - 1;
> > if (extra_pins)
> > - count -= 1;
> > + count -= folio_nr_pages(page_folio(p));
>
> IIUC, It seems than this also fix shmem of THP, because has_extra_refcount()
> is called
> from me_pagecache_clean(), right?
Usually we split THPs before we get to this point, but we always poison
an entire hugetlb page. This should probably change, but it's not high
on my list of things to do.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2] fs/hugetlbfs/inode.c: mm/memory-failure.c: fix hugetlbfs hwpoison handling
2024-01-12 18:08 [PATCH v2] fs/hugetlbfs/inode.c: mm/memory-failure.c: fix hugetlbfs hwpoison handling Sidhartha Kumar
2024-01-15 7:49 ` Muhammad Usama Anjum
2024-01-15 12:02 ` Muchun Song
@ 2024-01-16 2:36 ` Miaohe Lin
2024-01-16 2:43 ` Muchun Song
3 siblings, 0 replies; 6+ messages in thread
From: Miaohe Lin @ 2024-01-16 2:36 UTC (permalink / raw)
To: Sidhartha Kumar, linux-kernel, linux-mm
Cc: akpm, usama.anjum, muchun.song, jiaqiyan, willy, naoya.horiguchi,
shy828301, jthoughton, jane.chu, stable
On 2024/1/13 2:08, Sidhartha Kumar wrote:
> has_extra_refcount() makes the assumption that the page cache adds a ref
> count of 1 and subtracts this in the extra_pins case. Commit a08c7193e4f1
> (mm/filemap: remove hugetlb special casing in filemap.c) modifies
> __filemap_add_folio() by calling folio_ref_add(folio, nr); for all cases
> (including hugtetlb) where nr is the number of pages in the folio. We
> should adjust the number of references coming from the page cache by
> subtracing the number of pages rather than 1.
>
> In hugetlbfs_read_iter(), folio_test_has_hwpoisoned() is testing the wrong
> flag as, in the hugetlb case, memory-failure code calls
> folio_test_set_hwpoison() to indicate poison. folio_test_hwpoison() is the
> correct function to test for that flag.
>
> After these fixes, the hugetlb hwpoison read selftest passes all cases.
>
> Fixes: a08c7193e4f1 ("mm/filemap: remove hugetlb special casing in filemap.c")
> Closes: https://lore.kernel.org/linux-mm/20230713001833.3778937-1-jiaqiyan@google.com/T/#m8e1469119e5b831bbd05d495f96b842e4a1c5519
> Cc: <stable@vger.kernel.org> # 6.7+
> Signed-off-by: Sidhartha Kumar <sidhartha.kumar@oracle.com>
> Reported-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
> ---
>
> v1 -> v2:
> move ref_count adjustment to if(extra_pins) block as that represents
> ref counts from the page cache per Miaohe Lin.
Thanks for your update of patch.
>
> fs/hugetlbfs/inode.c | 2 +-
> mm/memory-failure.c | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
> index 36132c9125f9..3a248e4f7e93 100644
> --- a/fs/hugetlbfs/inode.c
> +++ b/fs/hugetlbfs/inode.c
> @@ -340,7 +340,7 @@ static ssize_t hugetlbfs_read_iter(struct kiocb *iocb, struct iov_iter *to)
> } else {
> folio_unlock(folio);
>
> - if (!folio_test_has_hwpoisoned(folio))
> + if (!folio_test_hwpoison(folio))
> want = nr;
> else {
> /*
> diff --git a/mm/memory-failure.c b/mm/memory-failure.c
> index d8c853b35dbb..ef7ae73b65bd 100644
> --- a/mm/memory-failure.c
> +++ b/mm/memory-failure.c
> @@ -976,7 +976,7 @@ static bool has_extra_refcount(struct page_state *ps, struct page *p,
> int count = page_count(p) - 1;
>
> if (extra_pins)
> - count -= 1;
> + count -= folio_nr_pages(page_folio(p));
I think this should be the right solution. @extra_pins indicates the extra page refcnt from page cache.
Acked-by: Miaohe Lin <linmiaohe@huawei.com>
Thanks.
>
> if (count > 0) {
> pr_err("%#lx: %s still referenced by %d users\n",
>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2] fs/hugetlbfs/inode.c: mm/memory-failure.c: fix hugetlbfs hwpoison handling
2024-01-12 18:08 [PATCH v2] fs/hugetlbfs/inode.c: mm/memory-failure.c: fix hugetlbfs hwpoison handling Sidhartha Kumar
` (2 preceding siblings ...)
2024-01-16 2:36 ` Miaohe Lin
@ 2024-01-16 2:43 ` Muchun Song
3 siblings, 0 replies; 6+ messages in thread
From: Muchun Song @ 2024-01-16 2:43 UTC (permalink / raw)
To: Sidhartha Kumar
Cc: LKML, Linux-MM, Andrew Morton, usama.anjum, Jiaqi Yan,
Matthew Wilcox (Oracle), Miaohe Lin, Naoya Horiguchi, Yang Shi,
James Houghton, jane.chu, stable
> On Jan 13, 2024, at 02:08, Sidhartha Kumar <sidhartha.kumar@oracle.com> wrote:
>
> has_extra_refcount() makes the assumption that the page cache adds a ref
> count of 1 and subtracts this in the extra_pins case. Commit a08c7193e4f1
> (mm/filemap: remove hugetlb special casing in filemap.c) modifies
> __filemap_add_folio() by calling folio_ref_add(folio, nr); for all cases
> (including hugtetlb) where nr is the number of pages in the folio. We
> should adjust the number of references coming from the page cache by
> subtracing the number of pages rather than 1.
>
> In hugetlbfs_read_iter(), folio_test_has_hwpoisoned() is testing the wrong
> flag as, in the hugetlb case, memory-failure code calls
> folio_test_set_hwpoison() to indicate poison. folio_test_hwpoison() is the
> correct function to test for that flag.
>
> After these fixes, the hugetlb hwpoison read selftest passes all cases.
>
> Fixes: a08c7193e4f1 ("mm/filemap: remove hugetlb special casing in filemap.c")
> Closes: https://lore.kernel.org/linux-mm/20230713001833.3778937-1-jiaqiyan@google.com/T/#m8e1469119e5b831bbd05d495f96b842e4a1c5519
> Cc: <stable@vger.kernel.org> # 6.7+
> Signed-off-by: Sidhartha Kumar <sidhartha.kumar@oracle.com>
> Reported-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Acked-by: Muchun Song <muchun.song@linux.dev>
Thanks.
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2024-01-16 2:43 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-01-12 18:08 [PATCH v2] fs/hugetlbfs/inode.c: mm/memory-failure.c: fix hugetlbfs hwpoison handling Sidhartha Kumar
2024-01-15 7:49 ` Muhammad Usama Anjum
2024-01-15 12:02 ` Muchun Song
2024-01-15 15:08 ` Matthew Wilcox
2024-01-16 2:36 ` Miaohe Lin
2024-01-16 2:43 ` Muchun Song
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).