This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH] Converted benchmark to benchtest.
- From: OndÅej BÃlka <neleai at seznam dot cz>
- To: Siddhesh Poyarekar <siddhesh at redhat dot com>
- Cc: Torvald Riegel <triegel at redhat dot com>, Carlos O'Donell <carlos at redhat dot com>, Andi Kleen <andi at firstfloor dot org>, libc-alpha at sourceware dot org
- Date: Thu, 5 Jun 2014 22:02:14 +0200
- Subject: Re: [PATCH] Converted benchmark to benchtest.
- Authentication-results: sourceware.org; auth=none
- References: <53497694 dot 4040901 at redhat dot com> <20140412194007 dot GA1284 at domone dot podge> <1397402907 dot 10643 dot 19303 dot camel at triegel dot csb> <534B8B38 dot 8090402 at redhat dot com> <8761m9iekn dot fsf at tassilo dot jf dot intel dot com> <87mwfifiks dot fsf at tassilo dot jf dot intel dot com> <53603208 dot 8060205 at redhat dot com> <1399033983 dot 32485 dot 5900 dot camel at triegel dot csb> <20140503104831 dot GA16118 at domone dot podge> <20140605123701 dot GF9145 at spoyarek dot pnq dot redhat dot com>
On Thu, Jun 05, 2014 at 06:07:01PM +0530, Siddhesh Poyarekar wrote:
> On Sat, May 03, 2014 at 12:48:31PM +0200, OndÅej BÃlka wrote:
> > Its nice to have but orthogonal on proposed change, so it should not
> > block this patch
>
> This looks like something that could easily use the $bench-inputs
> format since you're really just looking for mean times. I haven't
> checked to see if a failed branch would be expensive compared to the
> lock/unlock cycle we're testing, but I'm guessing that it should get
> cached and hence not have an impact since we don't interleave the
> rdlock and wrlock calls.
>
I assume that you asked about branch misprediction, that should not be
problem here, there are two ways how that it could happen. First one is
that it was not cache but then this is in cold path so it is not that
interesting. Second possibility is that we took different path previous
time, this is also not that interesting because previous call went to
slow path which was order of magnitude slower so its contribution is
tiny.
A third case where misprediction can happen is that there are parallel
readers. That would need a different benchmark.