* extend mwrap dump to include frees/allocations
@ 2019-10-08 22:52 Sam Saffron
2019-11-02 2:12 ` Eric Wong
0 siblings, 1 reply; 4+ messages in thread
From: Sam Saffron @ 2019-10-08 22:52 UTC (permalink / raw)
To: mwrap-public
(allocations-frees) * (total_allocated / allocations) is a reasonable
estimate for leak per call site.
Currently mwrap dump does not include this info, I was thinking we
default sort this down by estimated leak size and included the missing
bits on info into the dump report?
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: extend mwrap dump to include frees/allocations
2019-10-08 22:52 extend mwrap dump to include frees/allocations Sam Saffron
@ 2019-11-02 2:12 ` Eric Wong
2019-11-03 21:55 ` Sam Saffron
0 siblings, 1 reply; 4+ messages in thread
From: Eric Wong @ 2019-11-02 2:12 UTC (permalink / raw)
To: Sam Saffron; +Cc: mwrap-public
Sam Saffron <sam.saffron@gmail.com> wrote:
> (allocations-frees) * (total_allocated / allocations) is a reasonable
> estimate for leak per call site.
>
> Currently mwrap dump does not include this info, I was thinking we
> default sort this down by estimated leak size and included the missing
> bits on info into the dump report?
Do you mean Mwrap.dump the method or via ENV dump_path: destructor?
(or both?). It's been a while since I looked at this code(*)
It'd have to only be enabled via an optional flags arg for
compatibility with existing code.
Sorting the dump output would be expensive for big processes
because of the temporary memory required. I think that's best
be left for the consumer to do in another thread/process.
(*) Fwiw, I've started learning XS and porting it to Perl5, and
hoping to maintain feature parity: <https://80x24.org/mwrap-perl/>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: extend mwrap dump to include frees/allocations
2019-11-02 2:12 ` Eric Wong
@ 2019-11-03 21:55 ` Sam Saffron
2021-10-28 6:11 ` Eric Wong
0 siblings, 1 reply; 4+ messages in thread
From: Sam Saffron @ 2019-11-03 21:55 UTC (permalink / raw)
To: Eric Wong; +Cc: mwrap-public
Yes I was thinking about the method (not the ENV) but I guess both
work. I follow about the sorting concern, maybe we use a flag for the
sorting and another for the new format?
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: extend mwrap dump to include frees/allocations
2019-11-03 21:55 ` Sam Saffron
@ 2021-10-28 6:11 ` Eric Wong
0 siblings, 0 replies; 4+ messages in thread
From: Eric Wong @ 2021-10-28 6:11 UTC (permalink / raw)
To: Sam Saffron; +Cc: mwrap-public
Sam Saffron <sam.saffron@gmail.com> wrote:
> Yes I was thinking about the method (not the ENV) but I guess both
> work. I follow about the sorting concern, maybe we use a flag for the
> sorting and another for the new format?
Now (~2 years later :P) I'm thinking about dumping SQL text
suitable for loading into SQLite which opens up far more
sorting + viewing options...
Not sure if this project is something you or anybody else is
still interested in...
Anyways, pretty much all work is going into the Perl version
before (possibly) being ported to Ruby since I actually write
new code in Perl: https://80x24.org/mwrap-perl.git/
https://80x24.org/mwrap-perl/
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2021-10-28 6:11 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-10-08 22:52 extend mwrap dump to include frees/allocations Sam Saffron
2019-11-02 2:12 ` Eric Wong
2019-11-03 21:55 ` Sam Saffron
2021-10-28 6:11 ` Eric Wong
Code repositories for project(s) associated with this public inbox
https://80x24.org/mwrap.git/
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).