On Mon, Oct 05, 2026, [email protected] wrote: > > Fixes: c46234ebb4d1 ("tls: RX path for ktls") > > [Severity: Medium] > Is c46234ebb4d1 the right commit to blame for this stall? It looks like > the previous patch in this series added the poll-then-splice sleep: > "tls: consume empty data records in tls_sw_splice_read()".
The tag is right for what the patch changes. tls_sw_splice_read() has ignored the socket's O_NONBLOCK since c46234ebb4d1, and a splice on a nonblocking socket with no record queued sleeps without any help from 3/8. The second paragraph of the patch description is wrong for a tree without 3/8, though. There, a caller that polls and then splices an empty record gets the false EOF, not a sleep. The stall that paragraph describes needs the retry 3/8 adds. > Could this patch go before "tls: consume empty data records in > tls_sw_splice_read()" in the series? Yes. v3 will move this patch first and reword that paragraph. -- Chuck Lever (Come to NFS bake-a-thon! https://nfsv4bat.org)
