linux-api.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* selective block polling and preadv2/pwritev2 revisited
@ 2015-12-24 14:14 Christoph Hellwig
  2015-12-24 14:14 ` [PATCH 4/6] vfs: add the RWF_HIPRI flag for preadv2/pwritev2 Christoph Hellwig
                   ` (3 more replies)
  0 siblings, 4 replies; 12+ messages in thread
From: Christoph Hellwig @ 2015-12-24 14:14 UTC (permalink / raw)
  To: viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn, axboe-b10kYP2dOMg
  Cc: milosz-B5zB6C1i6pkAvxtiuMwx3w,
	linux-fsdevel-u79uwXL29TY76Z2rM5mHXA,
	linux-block-u79uwXL29TY76Z2rM5mHXA,
	linux-api-u79uwXL29TY76Z2rM5mHXA

This series allows to selectively enable/disable polling for completions
in the block layer [1] on a per-I/O basis.  For this it resurrects the
preadv2/pwritev2 syscalls that Milosz prepared a while ago (and which
are much simpler now due to VFS changes that happened in the meantime).
That approach also had a man page update prepared, which I will resubmit
with the current flags once this series makes it in.

Polling for block I/O is important to reduce the latency on flash and
post-flash storage technologies.  On the fastest NVMe controller I have
access to it almost halves latencies from over 7 microseconds to about 4
microseonds.  But it only is usesful if we actually care for the latency
of this particular I/O, and generally is a waste if enabled for all I/O
to a given device.  This series uses the per-I/O flags in preadv2/pwritev2
to control this behavior.  The alternative would be a new O_* flag set
at open time or using fcntl, but this is still to corse-grained for some
applications and we're starting to run out out of open flags.

Note that there are plenty of other use cases for preadv2/pwritev2 as well,
but I'd like to concentrate on this one for now.  Example are: non-blocking
reads (the original purpose), per-I/O O_SYNC, user space support for T10
DIF/DIX applications tags and probably some more.

[1] only supported for NVMe at the moment.

^ permalink raw reply	[flat|nested] 12+ messages in thread

end of thread, other threads:[~2016-01-07 14:41 UTC | newest]

Thread overview: 12+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-12-24 14:14 selective block polling and preadv2/pwritev2 revisited Christoph Hellwig
2015-12-24 14:14 ` [PATCH 4/6] vfs: add the RWF_HIPRI flag for preadv2/pwritev2 Christoph Hellwig
2015-12-24 14:14 ` [PATCH 5/6] direct-io: only use block polling if explicitly requested Christoph Hellwig
2015-12-24 14:14 ` [PATCH 6/6] blk-mq: enable polling support by default Christoph Hellwig
     [not found] ` <1450966464-6847-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
2015-12-24 14:14   ` [PATCH 1/6] vfs: pass a flags argument to vfs_readv/vfs_writev Christoph Hellwig
2015-12-24 14:14   ` [PATCH 2/6] vfs: vfs: Define new syscalls preadv2,pwritev2 Christoph Hellwig
2015-12-24 14:14   ` [PATCH 3/6] x86: wire up preadv2 and pwritev2 Christoph Hellwig
2016-01-04 14:58   ` selective block polling and preadv2/pwritev2 revisited Sagi Grimberg
     [not found]     ` <568A889E.4020204-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org>
2016-01-04 16:39       ` Christoph Hellwig
     [not found]         ` <20160104163949.GA17409-jcswGhMUV9g@public.gmane.org>
2016-01-06 17:01           ` Sagi Grimberg
2016-01-06 22:49             ` Martin K. Petersen
     [not found]               ` <yq1h9iqiary.fsf-+q57XtR/GgMb6DWv4sQWN6xOck334EZe@public.gmane.org>
2016-01-07 14:41                 ` Sagi Grimberg

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).