This is the mail archive of the
mailing list for the glibc project.
Re: BENCHTESTS: memmove benchtests do not test overlapping memory regions
- From: Steve Ellcey <sellcey at marvell dot com>
- To: "anton dot youdkevitch at bell-sw dot com" <anton dot youdkevitch at bell-sw dot com>, "zackw at panix dot com" <zackw at panix dot com>
- Cc: "libc-alpha at sourceware dot org" <libc-alpha at sourceware dot org>
- Date: Mon, 11 Feb 2019 16:29:31 +0000
- Subject: Re: BENCHTESTS: memmove benchtests do not test overlapping memory regions
- References: <CAKCAbMiqe9RbdN=mrr-ea41+OR=bCN97hbO93P5Xar8Yf1S2Tg@mail.gmail.com>
On Fri, 2019-02-08 at 11:41 -0500, Zack Weinberg wrote:
> On Fri, Feb 8, 2019 at 10:51 AM Anton Youdkevitch
> <email@example.com> wrote:
> > The current testing framework is using two buffers
> > for all sorts of string benchmarks. Considering how
> > the buffers get allocated (using mmap with the first
> > argument set to 0) they can never overlap.
> > So, the memmove benchtest actually tests memcpy
> > (with a little additional startup) instead of memmove.
> > I would like to adapt memmove benchset to cover the
> > case of overlapping memory regions as well.
> Yes, that would be a welcome improvement. It's probably going to be
> big enough that you will need to file contributor paperwork with the
> FSF: please read and follow the instructions at
Anton is doing this work under a contract that was signed with Cavium
and Cavium has a copyright assignment on file. Cavium was recently
purchased by Marvell but I believe Marvell has a copyright assignment
as well. I don't have a fencepost account anymore so I can't check
that myself, but that is my understanding.