public inbox for libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: Siddhesh Poyarekar <siddhesh@gotplt.org>
To: Florian Weimer <fweimer@redhat.com>,
	Szabolcs Nagy via Libc-alpha <libc-alpha@sourceware.org>
Cc: Szabolcs Nagy <Szabolcs.Nagy@arm.com>,
	Siddhesh Poyarekar <siddhesh@redhat.com>,
	Wilco Dijkstra <Wilco.Dijkstra@arm.com>
Subject: Re: [PATCH] AArch64: Add hp-timing.h
Date: Wed, 30 Jun 2021 15:03:05 +0530	[thread overview]
Message-ID: <3117c71b-4354-219e-bff8-af3858da48c3@gotplt.org> (raw)
In-Reply-To: <878s2rk9z2.fsf@oldenburg.str.redhat.com>

On 6/30/21 2:25 PM, Florian Weimer via Libc-alpha wrote:
> * Szabolcs Nagy via Libc-alpha:
> 
>> The 06/28/2021 11:11, Wilco Dijkstra wrote:
>>> Add hp-timing.h using the cntcvt_el0 counter. Return timing in nanoseconds
>>> so it is fully compatible with generic hp-timing. Don't set HP_TIMING_INLINE
>>> in the dynamic linker since very old kernels may not handle cntcvt correctly.
>>
>> can you please add a comment here that this
>> is only used in internal benchmarks and not
>> in installed glibc binaries. with that it's
>> ok to commit.
> 
> This commit would not be entirely accurate.  We install and ship the
> benchtests, so they need to use a stable ISA (and kernel ABI).

That doesn't seem like a reasonable limitation for benchtests.

> If the benchtests are like the testsuite and are not intended to be
> installed and distributed, we probably need to make downstream changes.

There's value in having the benchtest binaries in the distribution too, 
so maybe we need to figure out a way to do both, i.e. not be bound by 
ABI restrictions and at the same time, be deployable to do performance 
comparison.

Siddhesh

  reply	other threads:[~2021-06-30  9:33 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-28 10:11 Wilco Dijkstra
2021-06-28 10:20 ` Florian Weimer
2021-06-28 10:53   ` Wilco Dijkstra
2021-06-28 11:06     ` Florian Weimer
2021-06-28 11:35       ` Wilco Dijkstra
2021-06-28 11:39         ` Florian Weimer
2021-06-28 11:49       ` Szabolcs Nagy
2021-06-28 11:55         ` Adhemerval Zanella
2021-06-28 12:10         ` Wilco Dijkstra
2021-06-28 13:15           ` Adhemerval Zanella
2021-06-28 14:24           ` Wilco Dijkstra
2021-06-28 14:46             ` Adhemerval Zanella
2021-06-28 19:16               ` Wilco Dijkstra
2021-06-30  8:46 ` Szabolcs Nagy
2021-06-30  8:55   ` Florian Weimer
2021-06-30  9:33     ` Siddhesh Poyarekar [this message]
2021-07-01 10:21     ` Wilco Dijkstra
2021-07-01 13:48       ` Szabolcs Nagy
2021-07-01 15:33         ` Wilco Dijkstra
2021-07-06 11:33       ` Florian Weimer

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=3117c71b-4354-219e-bff8-af3858da48c3@gotplt.org \
    --to=siddhesh@gotplt.org \
    --cc=Szabolcs.Nagy@arm.com \
    --cc=Wilco.Dijkstra@arm.com \
    --cc=fweimer@redhat.com \
    --cc=libc-alpha@sourceware.org \
    --cc=siddhesh@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).