Re: [fs] 140402bab8: stress-ng.splice.ops_per_sec -100.0% regression

2 messages, 2 authors, 2020-07-01 · open the first message on its own page

Re: [fs] 140402bab8: stress-ng.splice.ops_per_sec -100.0% regression

From: Christoph Hellwig <hch@lst.de>
Date: 2020-07-01 12:13:54

FYI, this is because stress-nh tests splice using /dev/null.  Which
happens to actually have the iter ops, but doesn't have explicit
splice_read operation.

Re: [fs] 140402bab8: stress-ng.splice.ops_per_sec -100.0% regression

From: Linus Torvalds <torvalds@linux-foundation.org>
Date: 2020-07-01 20:32:29

On Wed, Jul 1, 2020 at 5:13 AM Christoph Hellwig [off-list ref] wrote:
FYI, this is because stress-nh tests splice using /dev/null.  Which
happens to actually have the iter ops, but doesn't have explicit
splice_read operation.
Heh. I guess a splice op for /dev/null should be fairly trivial to implement..

               Linus
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help