From: Sergey Kaplun via Tarantool-patches <tarantool-patches@dev.tarantool.org>
To: Sergey Bronnikov <sergeyb@tarantool.org>
Cc: Sergey Bronnikov <estetus@gmail.com>,
tarantool-patches@dev.tarantool.org
Subject: Re: [Tarantool-patches] [PATCH luajit][v2] Always close profiler output file.
Date: Tue, 4 Mar 2025 18:40:49 +0300 [thread overview]
Message-ID: <Z8cfAQZreaGMs76J@root> (raw)
In-Reply-To: <c15aae3b-d4f4-4469-953a-727e6a1b303a@tarantool.org>
Hi, Sergey!
Thanks for the fixes, LGTM.
On 27.02.25, Sergey Bronnikov wrote:
> Hi, Sergey,
>
> thanks for review!
>
> Changes applied and force-pushed.
>
> Sergey
>
> On 26.02.2025 14:58, Sergey Kaplun wrote:
<snipped>
> >> What is a rationale for this?
> > As I said before, this is an additional possible demonstration of the
> > bug. I see nothing bad to add it as well, since it simulates a little
> > bit different workload -- flushing the output before we finish the
> > process or if we unload the module. I am not insisting on it, though.
>
> How often users unload profiler's module? For me, this case looks artificial
The following module [1] is used to reload all Lua modules (including
`jit.p`). It was used for any Lua package update or code modification.
[1]: https://github.com/moonlibs/package-reload
Also, it is not only about unloading the module but also about finishing
the process too (as you first discovered). This looks like a real use
case. This variant of the test is just simpler than creating a child
process and checking its result files.
> and the final goal for us is not a demonstration, but covering a code
> touched by the patch,
>
> it is done by proposed test.
>
> (I'm really confused that on review I learn more and more new
> requirements to the patches
It is not a requirement, just a suggestion. See the last line of my
comment. I'm OK with reproducing the issue only since it is not the
vital part of the code.
>
> like adding all available testcases that cover a patch or that code
> should follow C89 or
>
> that using etc.)
>
> The bug fixed by the patch is quite minor, it will not break production
> and will not kill humans etc.
>
> I believe a single testcase and time that we both spend on doing
> backport, test and two iterations
>
> of review is more than enough for this patch. Moreover, jit.p is not
> used by Tarantool users,
>
> we provide our own profilers to them.
So, just ignore it then.
>
>
> >> This is a helper script for LuaJIT profiler and according to our
> >> backporting rules
> >>
> >> test must cover a backported patch. This rule is fulfilled by my test.
> > <snipped>
> >
--
Best regards,
Sergey Kaplun
next prev parent reply other threads:[~2025-03-04 15:40 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-02-12 12:20 Sergey Bronnikov via Tarantool-patches
2025-02-13 15:31 ` Sergey Kaplun via Tarantool-patches
2025-02-25 8:58 ` Sergey Bronnikov via Tarantool-patches
2025-02-26 11:58 ` Sergey Kaplun via Tarantool-patches
2025-02-27 13:07 ` Sergey Bronnikov via Tarantool-patches
2025-03-04 15:40 ` Sergey Kaplun via Tarantool-patches [this message]
2025-03-05 14:13 ` Sergey Bronnikov via Tarantool-patches
2025-03-05 14:58 ` Sergey Kaplun via Tarantool-patches
2025-03-05 15:09 ` Sergey Bronnikov via Tarantool-patches
2025-03-12 11:12 ` Sergey Kaplun via Tarantool-patches
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=Z8cfAQZreaGMs76J@root \
--to=tarantool-patches@dev.tarantool.org \
--cc=estetus@gmail.com \
--cc=sergeyb@tarantool.org \
--cc=skaplun@tarantool.org \
--subject='Re: [Tarantool-patches] [PATCH luajit][v2] Always close profiler output file.' \
/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
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox