This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Pre-release review of glibc 2.17: 0 P1 bugs, 14 P2 bugs, 0 P3bugs.


From: "Carlos O'Donell" <carlos@systemhalted.org>
Date: Thu, 29 Nov 2012 10:49:59 -0500

> I've gone through the bugs marked with milestone 2.17.
> 
> There are 14 P2 bugs in the tracker with milestone 2.17.

Thanks for doing this.

> The following 2 issues should go in immediately:
> 
> "Document __chk_fail ABI in test case"
> http://sourceware.org/bugzilla/show_bug.cgi?id=14197
> - Comment fixup should go in immediately we delayed it last time
> - It's less of an issue if this goes in ASAP.

I've commited Florian's patch.

> "hppa: Fix two integer to pointer warnings."
> http://sourceware.org/bugzilla/show_bug.cgi?id=6677
> - Testing this fix now.
> - Restricted to hppa.

I assume you will commit this when your testing is done.

> The following bug was something Roland said was "unacceptable" and
> should be fixed.
> 
> "_{POSIX_V7|_POSIX_V6|XBS5}_ILP32_OFFBIG are wrong for -mx32 build"
> http://sourceware.org/bugzilla/show_bug.cgi?id=14200
> - No patch and no review.
> - This is too subtle for me to understand.

HJ seems to have posted a patch and therefore I think Roland and he
need to sort this out.

> The other 11 should have their milestone bumped to indicate
> we are not accepting them for 2.17:
> 
> "QoI regression: strstr() slowed from O(n) to O(n^2) on SSE4 machines"
> http://sourceware.org/bugzilla/show_bug.cgi?id=12100
> - Patch exists.
> - IMO this is too risky to go in for 2.17.

I think a patch posted in June should not need to wait for yet another
release.  I'm very disappointed that this change has rotted for so
long, after someone put in the effort to fix a regression that is 6
releases old.

> "fclose violates POSIX 2008 on seekable input streams"
> http://sourceware.org/bugzilla/show_bug.cgi?id=12724
> - No patch exists, and too risky.

Agreed.  It's troubling to see that a major system component like Ruby
seems to be running into problems because of this.

> "assertion error in res_query.c"
> http://sourceware.org/bugzilla/show_bug.cgi?id=13013
> - Patch exists.
> - Insufficient review.

I think I could review this patch and that it would be worth trying to
get it into the release.

> "Building libc.pot has a lot of warnings."
> http://sourceware.org/bugzilla/show_bug.cgi?id=13853
> - Patches exist but need serious review.
> - Risky to change these scripts without more time.

Agreed, too risky.

> "Malloc can deadlock in retry paths"
> http://sourceware.org/bugzilla/show_bug.cgi?id=13939
> - Insufficient review.
> - Risky.

Agreed, risky.  But I'd really hate to be the user who hits that
bug unknowingly.

> "Cleanup reversal of sunrpc obsoletion code"
> http://sourceware.org/bugzilla/show_bug.cgi?id=14084
> - Low priority.
> - Hack working well.

Agreed.

> "Clean up shlib-versions"
> http://sourceware.org/bugzilla/show_bug.cgi?id=14171
> - Sweeping change not valid for freeze.

Agreed.

> "--localedir arg to configure is ignored"
> http://sourceware.org/bugzilla/show_bug.cgi?id=14259
> - No update from patch author.

Agreed.

> "fprintf() function is multithread-unsafe"
> http://sourceware.org/bugzilla/show_bug.cgi?id=14267
> - Not enough review to go into 2.17.
> - This is an honest to goodness real problem, but we need more serious review.

I think the patch could be resonably verified and the change installed,
it's not that overly complicated in my opinion.

> "Removal of sysdeps/x86_64/fpu/s_sincos.S causes regressions"
> http://sourceware.org/bugzilla/show_bug.cgi?id=14412
> - No patches.
> - Real problem though.

I wish we had a usable fix for this one, but we don't so...

> "power: memset is broken on powerpc 405"
> http://sourceware.org/bugzilla/show_bug.cgi?id=14595
> - Patch on a branch.
> - No review or testing yet.

The patch is actually installed in the tree already, and therefore
this bug should be closed.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]