[PATCH 2/5] man: fixups for io_uring_prep_cmd_discard.3
Christoph Hellwig <[email protected]> Wed, 15 Jul 2026 08:39:29 +0200
| Newsgroups | org.kernel.vger.io-uring |
|---|---|
| Message-ID | <[email protected]> |
Mention the ioctl this is modelled after instead of the io_uring cmd used to implement io_uring_prep_cmd_discard.3, and drop an incorrect plural for page cache. Signed-off-by: Christoph Hellwig <[email protected]> --- man/io_uring_prep_cmd_discard.3 | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/man/io_uring_prep_cmd_discard.3 b/man/io_uring_prep_cmd_discard.3 index 97786726f03b..349df86ff9c9 100644 --- a/man/io_uring_prep_cmd_discard.3 +++ b/man/io_uring_prep_cmd_discard.3 @@ -26,14 +26,14 @@ to start discarding at the specified .IR offset . -The command is an asynchronous equivalent of -.B BLOCK_URING_CMD_DISCARD +The command is an asynchronous equivalent of the +.B BLKDISCARD ioctl with a few differences. It allows multiple parallel discards, and it does not exclude concurrent writes and reads. As a result, it may lead to races for the data on the disk, if the application has IO inflight for the same ranges that the discard operates on. It's the user's responsibility to account for that. -Furthermore, only best efforts are done to invalidate page caches. The user has -to make sure that no other inflight requests are modifying or reading the +Furthermore, only best efforts are done to invalidate the page cache. The user +has to make sure that no other inflight requests are modifying or reading the range(s). If that is the case, it might result in stale page cache and data inconsistencies. -- 2.53.0