[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <440141.1692179410@warthog.procyon.org.uk>
Date: Wed, 16 Aug 2023 10:50:10 +0100
From: David Howells <dhowells@...hat.com>
To: David Laight <David.Laight@...LAB.COM>
Cc: dhowells@...hat.com,
Linus Torvalds <torvalds@...ux-foundation.org>,
Alexander Viro <viro@...iv.linux.org.uk>,
Jens Axboe <axboe@...nel.dk>, Christoph Hellwig <hch@....de>,
Christian Brauner <christian@...uner.io>,
Matthew Wilcox <willy@...radead.org>,
"jlayton@...nel.org" <jlayton@...nel.org>,
"linux-block@...r.kernel.org" <linux-block@...r.kernel.org>,
"linux-fsdevel@...r.kernel.org" <linux-fsdevel@...r.kernel.org>,
"linux-mm@...ck.org" <linux-mm@...ck.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [RFC PATCH v2] iov_iter: Convert iterate*() to inline funcs
David Laight <David.Laight@...LAB.COM> wrote:
> It is harder to compare because of some of the random name changes.
I wouldn't say 'random' exactly, but if you prefer, some of the name changing
can be split out into a separate patch. The macros are kind of the worst
since they picked up variable names from the callers.
> The version of the source I found seems to pass priv2 to functions
> that don't use it?
That can't be avoided if I convert everything to inline functions and function
pointers - but the optimiser can get rid of it where it can inline the step
function.
I tried passing the iterator to the step functions instead, but that just made
things bigger. memcpy_from_iter_mc() is interesting to deal with. I would
prefer to deal with it in the caller so we only do the check once, but that
might mean duplicating the caller.
Actually, ->copy_mc is only set in once place, dump_emit_page() in coredump.c,
and only on a bvec iterator, so I could probably do a special function just
for that that calls iterate_bvec() rather than iterate_and_advance2() and then
make _copy_from_iter() just use memcpy_from_iter() and get rid of
iov_iter::copy_mc entirely.
> Since the functions aren't inlined you get the cost of passing
> the parameters.
> This seems to affect the common cases.
Actually, in v2, the action functions for common cases are marked
__always_inline and do get fully inlined and the code in some paths actually
ends up slightly smaller.
> Is that all left over from a version that passed function pointers
> (with the hope they'd be inlined?).
> Just directly inlining the simple copies should help.
I did that in v2 for things like memcpy_to_iter() and memcpy_from_iter().
> I rather hope the should_fail_usercopy() and instrument_copy_xxx()
> calls are usually either absent or, at most, nops.
Okay - it's probably worth marking those too, then.
> This all seems to have a lot fewer options than last time I looked.
I'm not sure what you mean by 'a lot fewer options'?
> Is it worth optimising the KVEC case with a single buffer?
You mean an equivalent of UBUF? Maybe. There are probably a whole bunch of
netfs places that do single-kvec writes, though I'm trying to convert these
over to bvec arrays, combining them with their data, and MSG_SPLICE_PAGES.
David
Powered by blists - more mailing lists