public inbox for libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: Adhemerval Zanella Netto <adhemerval.zanella@linaro.org>
To: Siddhesh Poyarekar <siddhesh@gotplt.org>,
	libc-alpha@sourceware.org, Florian Weimer <fweimer@redhat.com>
Subject: Re: [PATCH 1/2] elf: Remove glibc.rtld.dynamic_sort tunable
Date: Wed, 22 Mar 2023 15:46:27 -0300	[thread overview]
Message-ID: <ee911490-1797-0eb2-19e2-e617100b82cc@linaro.org> (raw)
In-Reply-To: <50f4d048-bdde-fccc-fdd2-edd7b67994ee@gotplt.org>



On 22/03/23 15:40, Siddhesh Poyarekar wrote:
> On 2023-03-22 13:51, Adhemerval Zanella Netto wrote:
>>
>>
>> On 22/03/23 14:08, Siddhesh Poyarekar wrote:
>>> On 2023-03-13 15:06, Adhemerval Zanella wrote:
>>>> And make the DFS algorithm (added on 2.35) the de-facto implementation.
>>>
>>> Isn't this orthogonal to the proposal to remove --enable-tunables?
>>
>> Yes, the DFS is already the default one.  I meant that now there is no way to
>> select the one algorithm.  Maybe use the commit messagE:
>>
>>    And make the old sort algorithm no longer available.
> 
> OK, I'll stick to reviewing the --enable-tunables patch then, assuming that it can go in independently of this one.  I feel much less comfortable acking this one since I haven't been closely involved in sorting algorithm selection.  I get the sense that maybe it makes sense to wait a couple more releases before removing the old sort algorithm but that's not based on any specific knowledge, just an instinct.

Well, without tunables the old algorithm can not be selected.  So the second
part does depend of this.

The old algorithm was kept mainly to provide some fallback compatibility to
binaries that does not work correctly with newer one; but the default one
is already in 3 release, it is faster, and so far I have not heard of any
drawnback. 

  reply	other threads:[~2023-03-22 18:46 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-03-13 19:06 [PATCH 0/2] Make tunable a default feature Adhemerval Zanella
2023-03-13 19:06 ` [PATCH 1/2] elf: Remove glibc.rtld.dynamic_sort tunable Adhemerval Zanella
2023-03-22 17:08   ` Siddhesh Poyarekar
2023-03-22 17:51     ` Adhemerval Zanella Netto
2023-03-22 18:40       ` Siddhesh Poyarekar
2023-03-22 18:46         ` Adhemerval Zanella Netto [this message]
2023-03-22 18:52           ` Siddhesh Poyarekar
2023-03-22 19:11             ` Adhemerval Zanella Netto
2023-03-13 19:06 ` [PATCH 2/2] Remove --enable-tunables configure option Adhemerval Zanella
2023-03-22 16:25   ` Siddhesh Poyarekar
2023-03-14 18:02 ` [PATCH 0/2] Make tunable a default feature Siddhesh Poyarekar
2023-03-15 20:05   ` Adhemerval Zanella Netto
2023-03-17 11:15     ` Siddhesh Poyarekar
2023-03-20 16:59       ` Adhemerval Zanella Netto
2023-03-20 23:15         ` Siddhesh Poyarekar
2023-03-22 14:40           ` Adhemerval Zanella Netto
2023-03-22 15:21             ` Siddhesh Poyarekar
2023-03-22 17:58               ` Adhemerval Zanella Netto

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=ee911490-1797-0eb2-19e2-e617100b82cc@linaro.org \
    --to=adhemerval.zanella@linaro.org \
    --cc=fweimer@redhat.com \
    --cc=libc-alpha@sourceware.org \
    --cc=siddhesh@gotplt.org \
    /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).