* Short read handling inconsistency for IOSQE_IO_LINK
@ 2020-09-25 9:37 Carter Li 李通洲
0 siblings, 0 replies; only message in thread
From: Carter Li 李通洲 @ 2020-09-25 9:37 UTC (permalink / raw)
To: io-uring
Hi all,
Short read of IORING_OP_READ in a link chain is considered an error, because it
will break READ-WRITE chain, which is good.
The problems are:
1. Short read of IORING_OP_RECV in a link chain is *NOT* considered an error,
which is odd and not consistant with IORING_OP_READ. Is it a bug or by design?
2. Short read of IORING_OP_SPLICE in a link chain is considered an error. It is
consistant with IORING_OP_READ, but not very useful because the link
1) splice(fd1, NULL, pfd1, NULL, -1, 0)
2) splice(pfd2, NULL, fd2, NULL, -1, 0)
is a valid operation since the kernel knows how much data can be read in a pipe.
Any thoughts?
Regards,
Carter
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2020-09-25 9:53 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2020-09-25 9:37 Short read handling inconsistency for IOSQE_IO_LINK Carter Li 李通洲
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox