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

[Arjan van de Ven]

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

almost any allocation all over the kernel can in theory block ;)
I'd say be pragmatic in this and avoid the obvious pagecache blocking,
but ignore the rest, it'll be rare and if it's there, of short duration.
Userland can get rescheduled anyway at any time for brief periods

Attachment: signature.asc
Description: This is a digitally signed message part