summaryrefslogtreecommitdiff
path: root/mm
diff options
context:
space:
mode:
authorHugh Dickins <hughd@google.com>2022-04-14 19:13:27 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2022-04-15 14:49:54 -0700
commit1bdec44b1eee32e311b44b5b06144bb7d9b33938 (patch)
tree67349dd5de85a8db47093ac128cea601f2b628eb /mm
parent7fbd166a8f2d697c3e2b4c8432d33253f00266b3 (diff)
tmpfs: fix regressions from wider use of ZERO_PAGE
Chuck Lever reported fsx-based xfstests generic 075 091 112 127 failing when 5.18-rc1 NFS server exports tmpfs: bisected to recent tmpfs change. Whilst nfsd_splice_action() does contain some questionable handling of repeated pages, and Chuck was able to work around there, history from Mark Hemment makes clear that there might be similar dangers elsewhere: it was not a good idea for me to pass ZERO_PAGE down to unknown actors. Revert shmem_file_read_iter() to using ZERO_PAGE for holes only when iter_is_iovec(); in other cases, use the more natural iov_iter_zero() instead of copy_page_to_iter(). We would use iov_iter_zero() throughout, but the x86 clear_user() is not nearly so well optimized as copy to user (dd of 1T sparse tmpfs file takes 57 seconds rather than 44 seconds). And now pagecache_init() does not need to SetPageUptodate(ZERO_PAGE(0)): which had caused boot failure on arm noMMU STM32F7 and STM32H7 boards Link: https://lkml.kernel.org/r/9a978571-8648-e830-5735-1f4748ce2e30@google.com Fixes: 56a8c8eb1eaf ("tmpfs: do not allocate pages on read") Signed-off-by: Hugh Dickins <hughd@google.com> Reported-by: Patrice CHOTARD <patrice.chotard@foss.st.com> Reported-by: Chuck Lever III <chuck.lever@oracle.com> Tested-by: Chuck Lever III <chuck.lever@oracle.com> Cc: Mark Hemment <markhemm@googlemail.com> Cc: Patrice CHOTARD <patrice.chotard@foss.st.com> Cc: Mikulas Patocka <mpatocka@redhat.com> Cc: Lukas Czerner <lczerner@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Cc: "Darrick J. Wong" <djwong@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'mm')
-rw-r--r--mm/filemap.c6
-rw-r--r--mm/shmem.c31
2 files changed, 20 insertions, 17 deletions
diff --git a/mm/filemap.c b/mm/filemap.c
index 3a5ffb5587cd..9a1eef6c5d35 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1063,12 +1063,6 @@ void __init pagecache_init(void)
init_waitqueue_head(&folio_wait_table[i]);
page_writeback_init();
-
- /*
- * tmpfs uses the ZERO_PAGE for reading holes: it is up-to-date,
- * and splice's page_cache_pipe_buf_confirm() needs to see that.
- */
- SetPageUptodate(ZERO_PAGE(0));
}
/*
diff --git a/mm/shmem.c b/mm/shmem.c
index 529c9ad3e926..4b2fea33158e 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -2513,7 +2513,6 @@ static ssize_t shmem_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
pgoff_t end_index;
unsigned long nr, ret;
loff_t i_size = i_size_read(inode);
- bool got_page;
end_index = i_size >> PAGE_SHIFT;
if (index > end_index)
@@ -2570,24 +2569,34 @@ static ssize_t shmem_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
*/
if (!offset)
mark_page_accessed(page);
- got_page = true;
+ /*
+ * Ok, we have the page, and it's up-to-date, so
+ * now we can copy it to user space...
+ */
+ ret = copy_page_to_iter(page, offset, nr, to);
+ put_page(page);
+
+ } else if (iter_is_iovec(to)) {
+ /*
+ * Copy to user tends to be so well optimized, but
+ * clear_user() not so much, that it is noticeably
+ * faster to copy the zero page instead of clearing.
+ */
+ ret = copy_page_to_iter(ZERO_PAGE(0), offset, nr, to);
} else {
- page = ZERO_PAGE(0);
- got_page = false;
+ /*
+ * But submitting the same page twice in a row to
+ * splice() - or others? - can result in confusion:
+ * so don't attempt that optimization on pipes etc.
+ */
+ ret = iov_iter_zero(nr, to);
}
- /*
- * Ok, we have the page, and it's up-to-date, so
- * now we can copy it to user space...
- */
- ret = copy_page_to_iter(page, offset, nr, to);
retval += ret;
offset += ret;
index += offset >> PAGE_SHIFT;
offset &= ~PAGE_MASK;
- if (got_page)
- put_page(page);
if (!iov_iter_count(to))
break;
if (ret < nr) {