Re: [patch rfc] towards supporting O_NONBLOCK on regular files

[Jeff Moyer <jmoyer@redhat.com>]

==> Regarding Re: [patch rfc] towards supporting O_NONBLOCK on regular files; "Stephen C. Tweedie" <sct@redhat.com> adds:

sct> Hi, On Thu, 2004-10-07 at 11:12, Marcelo Tosatti wrote:

>> Oh yes, theres also the indirect blocks which we might need to read from
>> disk.

sct> Right.

>> Now the question is, how strict should the O_NONBLOCK implementation be
>> in reference to "not blocking" ?

sct> Well, I suspect that depends on the application.  But if you've got an
sct> app that really wants to make sure its hot path is as fast as possible
sct> (eg. a high-throughput server multiplexing disk IO and networking
sct> through a single event loop), then ideally the app would want to punt
sct> any blocking disk IO to another thread.

sct> So it's a matter of significant extra programing for a small extra
sct> reduction in app blocking potential.

sct> I think it's worth getting this right in the long term, though.
sct> Getting readahead of indirect blocks right has other benefits too ---
sct> eg. we may be able to fix the situation where we end up trying to read
sct> indirect blocks before we've even submitted the IO for the previous
sct> data blocks, breaking the IO pipeline ordering.

So for the short term, are you an advocate of the patch posted?

-Jeff