[v1,2/3] hugetlbfs: improve read HWPOISON hugepage
Commit Message
When a hugepage contains HWPOISON pages, read() fails to read any byte
of the hugepage and returns -EIO, although many bytes in the HWPOISON
hugepage are readable.
Improve this by allowing hugetlbfs_read_iter returns as many bytes as
possible. For a requested range [offset, offset + len) that contains
HWPOISON page, return [offset, first HWPOISON page addr); the next read
attempt will fail and return -EIO.
Signed-off-by: Jiaqi Yan <jiaqiyan@google.com>
---
fs/hugetlbfs/inode.c | 62 +++++++++++++++++++++++++++++++++++++++-----
1 file changed, 56 insertions(+), 6 deletions(-)
Comments
On 05/17/23 16:09, Jiaqi Yan wrote:
> When a hugepage contains HWPOISON pages, read() fails to read any byte
> of the hugepage and returns -EIO, although many bytes in the HWPOISON
> hugepage are readable.
>
> Improve this by allowing hugetlbfs_read_iter returns as many bytes as
> possible. For a requested range [offset, offset + len) that contains
> HWPOISON page, return [offset, first HWPOISON page addr); the next read
> attempt will fail and return -EIO.
>
> Signed-off-by: Jiaqi Yan <jiaqiyan@google.com>
> ---
> fs/hugetlbfs/inode.c | 62 +++++++++++++++++++++++++++++++++++++++-----
> 1 file changed, 56 insertions(+), 6 deletions(-)
>
> diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
> index ecfdfb2529a3..1baa08ec679f 100644
> --- a/fs/hugetlbfs/inode.c
> +++ b/fs/hugetlbfs/inode.c
> @@ -282,6 +282,46 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
> }
> #endif
>
> +/*
> + * Someone wants to read @bytes from a HWPOISON hugetlb @page from @offset.
> + * Returns the maximum number of bytes one can read without touching the 1st raw
> + * HWPOISON subpage.
> + *
> + * The implementation borrows the iteration logic from copy_page_to_iter*.
> + */
> +static size_t adjust_range_hwpoison(struct page *page, size_t offset, size_t bytes)
> +{
> + size_t n = 0;
> + size_t res = 0;
> + struct folio *folio = page_folio(page);
> +
> + folio_lock(folio);
What is the reason for taking folio_lock?
> +
> + /* First subpage to start the loop. */
> + page += offset / PAGE_SIZE;
> + offset %= PAGE_SIZE;
> + while (1) {
> + if (find_raw_hwp_page(folio, page) != NULL)
> + break;
> +
> + /* Safe to read n bytes without touching HWPOISON subpage. */
> + n = min(bytes, (size_t)PAGE_SIZE - offset);
> + res += n;
> + bytes -= n;
> + if (!bytes || !n)
> + break;
> + offset += n;
> + if (offset == PAGE_SIZE) {
> + page++;
> + offset = 0;
> + }
> + }
> +
> + folio_unlock(folio);
> +
> + return res;
> +}
> +
> /*
> * Support for read() - Find the page attached to f_mapping and copy out the
> * data. This provides functionality similar to filemap_read().
> @@ -300,7 +340,7 @@ static ssize_t hugetlbfs_read_iter(struct kiocb *iocb, struct iov_iter *to)
>
> while (iov_iter_count(to)) {
> struct page *page;
> - size_t nr, copied;
> + size_t nr, copied, want;
>
> /* nr is the maximum number of bytes to copy from this page */
> nr = huge_page_size(h);
> @@ -328,16 +368,26 @@ static ssize_t hugetlbfs_read_iter(struct kiocb *iocb, struct iov_iter *to)
> } else {
> unlock_page(page);
>
> - if (PageHWPoison(page)) {
> - put_page(page);
> - retval = -EIO;
> - break;
> + if (!PageHWPoison(page))
> + want = nr;
> + else {
> + /*
> + * Adjust how many bytes safe to read without
> + * touching the 1st raw HWPOISON subpage after
> + * offset.
> + */
> + want = adjust_range_hwpoison(page, offset, nr);
> + if (want == 0) {
> + put_page(page);
> + retval = -EIO;
> + break;
> + }
> }
>
> /*
> * We have the page, copy it to user space buffer.
> */
> - copied = copy_page_to_iter(page, offset, nr, to);
> + copied = copy_page_to_iter(page, offset, want, to);
> put_page(page);
> }
> offset += copied;
> --
> 2.40.1.606.ga4b1b128d6-goog
>
Code looks fine, just wondering about that folio_lock.
On Thu, May 18, 2023 at 3:18 PM Mike Kravetz <mike.kravetz@oracle.com> wrote:
>
> On 05/17/23 16:09, Jiaqi Yan wrote:
> > When a hugepage contains HWPOISON pages, read() fails to read any byte
> > of the hugepage and returns -EIO, although many bytes in the HWPOISON
> > hugepage are readable.
> >
> > Improve this by allowing hugetlbfs_read_iter returns as many bytes as
> > possible. For a requested range [offset, offset + len) that contains
> > HWPOISON page, return [offset, first HWPOISON page addr); the next read
> > attempt will fail and return -EIO.
> >
> > Signed-off-by: Jiaqi Yan <jiaqiyan@google.com>
> > ---
> > fs/hugetlbfs/inode.c | 62 +++++++++++++++++++++++++++++++++++++++-----
> > 1 file changed, 56 insertions(+), 6 deletions(-)
> >
> > diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
> > index ecfdfb2529a3..1baa08ec679f 100644
> > --- a/fs/hugetlbfs/inode.c
> > +++ b/fs/hugetlbfs/inode.c
> > @@ -282,6 +282,46 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
> > }
> > #endif
> >
> > +/*
> > + * Someone wants to read @bytes from a HWPOISON hugetlb @page from @offset.
> > + * Returns the maximum number of bytes one can read without touching the 1st raw
> > + * HWPOISON subpage.
> > + *
> > + * The implementation borrows the iteration logic from copy_page_to_iter*.
> > + */
> > +static size_t adjust_range_hwpoison(struct page *page, size_t offset, size_t bytes)
> > +{
> > + size_t n = 0;
> > + size_t res = 0;
> > + struct folio *folio = page_folio(page);
> > +
> > + folio_lock(folio);
>
> What is the reason for taking folio_lock?
I intended to make this routine (mostly find_raw_hwp_page) to be
serialized with folio_clear_hugetlb_hwpoison() and
hwpoison_user_mappings() in try_memory_failure_hugetlb(). They don't
directly affect the raw_hwp_list. I can remove the lock in v2.
>
> > +
> > + /* First subpage to start the loop. */
> > + page += offset / PAGE_SIZE;
> > + offset %= PAGE_SIZE;
> > + while (1) {
> > + if (find_raw_hwp_page(folio, page) != NULL)
> > + break;
> > +
> > + /* Safe to read n bytes without touching HWPOISON subpage. */
> > + n = min(bytes, (size_t)PAGE_SIZE - offset);
> > + res += n;
> > + bytes -= n;
> > + if (!bytes || !n)
> > + break;
> > + offset += n;
> > + if (offset == PAGE_SIZE) {
> > + page++;
> > + offset = 0;
> > + }
> > + }
> > +
> > + folio_unlock(folio);
> > +
> > + return res;
> > +}
> > +
> > /*
> > * Support for read() - Find the page attached to f_mapping and copy out the
> > * data. This provides functionality similar to filemap_read().
> > @@ -300,7 +340,7 @@ static ssize_t hugetlbfs_read_iter(struct kiocb *iocb, struct iov_iter *to)
> >
> > while (iov_iter_count(to)) {
> > struct page *page;
> > - size_t nr, copied;
> > + size_t nr, copied, want;
> >
> > /* nr is the maximum number of bytes to copy from this page */
> > nr = huge_page_size(h);
> > @@ -328,16 +368,26 @@ static ssize_t hugetlbfs_read_iter(struct kiocb *iocb, struct iov_iter *to)
> > } else {
> > unlock_page(page);
> >
> > - if (PageHWPoison(page)) {
> > - put_page(page);
> > - retval = -EIO;
> > - break;
> > + if (!PageHWPoison(page))
> > + want = nr;
> > + else {
> > + /*
> > + * Adjust how many bytes safe to read without
> > + * touching the 1st raw HWPOISON subpage after
> > + * offset.
> > + */
> > + want = adjust_range_hwpoison(page, offset, nr);
> > + if (want == 0) {
> > + put_page(page);
> > + retval = -EIO;
> > + break;
> > + }
> > }
> >
> > /*
> > * We have the page, copy it to user space buffer.
> > */
> > - copied = copy_page_to_iter(page, offset, nr, to);
> > + copied = copy_page_to_iter(page, offset, want, to);
> > put_page(page);
> > }
> > offset += copied;
> > --
> > 2.40.1.606.ga4b1b128d6-goog
> >
>
> Code looks fine, just wondering about that folio_lock.
> --
> Mike Kravetz
@@ -282,6 +282,46 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
}
#endif
+/*
+ * Someone wants to read @bytes from a HWPOISON hugetlb @page from @offset.
+ * Returns the maximum number of bytes one can read without touching the 1st raw
+ * HWPOISON subpage.
+ *
+ * The implementation borrows the iteration logic from copy_page_to_iter*.
+ */
+static size_t adjust_range_hwpoison(struct page *page, size_t offset, size_t bytes)
+{
+ size_t n = 0;
+ size_t res = 0;
+ struct folio *folio = page_folio(page);
+
+ folio_lock(folio);
+
+ /* First subpage to start the loop. */
+ page += offset / PAGE_SIZE;
+ offset %= PAGE_SIZE;
+ while (1) {
+ if (find_raw_hwp_page(folio, page) != NULL)
+ break;
+
+ /* Safe to read n bytes without touching HWPOISON subpage. */
+ n = min(bytes, (size_t)PAGE_SIZE - offset);
+ res += n;
+ bytes -= n;
+ if (!bytes || !n)
+ break;
+ offset += n;
+ if (offset == PAGE_SIZE) {
+ page++;
+ offset = 0;
+ }
+ }
+
+ folio_unlock(folio);
+
+ return res;
+}
+
/*
* Support for read() - Find the page attached to f_mapping and copy out the
* data. This provides functionality similar to filemap_read().
@@ -300,7 +340,7 @@ static ssize_t hugetlbfs_read_iter(struct kiocb *iocb, struct iov_iter *to)
while (iov_iter_count(to)) {
struct page *page;
- size_t nr, copied;
+ size_t nr, copied, want;
/* nr is the maximum number of bytes to copy from this page */
nr = huge_page_size(h);
@@ -328,16 +368,26 @@ static ssize_t hugetlbfs_read_iter(struct kiocb *iocb, struct iov_iter *to)
} else {
unlock_page(page);
- if (PageHWPoison(page)) {
- put_page(page);
- retval = -EIO;
- break;
+ if (!PageHWPoison(page))
+ want = nr;
+ else {
+ /*
+ * Adjust how many bytes safe to read without
+ * touching the 1st raw HWPOISON subpage after
+ * offset.
+ */
+ want = adjust_range_hwpoison(page, offset, nr);
+ if (want == 0) {
+ put_page(page);
+ retval = -EIO;
+ break;
+ }
}
/*
* We have the page, copy it to user space buffer.
*/
- copied = copy_page_to_iter(page, offset, nr, to);
+ copied = copy_page_to_iter(page, offset, want, to);
put_page(page);
}
offset += copied;