Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 642044 - sys-libs/glibc-2.26-r4 - some patches apply to ChangeLog only
Summary: sys-libs/glibc-2.26-r4 - some patches apply to ChangeLog only
Status: RESOLVED WORKSFORME
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: Current packages (show other bugs)
Hardware: All Linux
: Normal normal (vote)
Assignee: Gentoo Toolchain Maintainers
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2017-12-22 14:35 UTC by Ulenrich
Modified: 2017-12-23 20:54 UTC (History)
1 user (show)

See Also:
Package list:
Runtime testing required: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Ulenrich 2017-12-22 14:35:33 UTC
I did manually look look into the patches for the coming r4 glibc (not keyworded yet). Thereby seeing following patches only include some Changelog messages without doing any to the actual glibc code:
---
0111_all_Don-t-use-IFUNC-resolver-for-longjmp-or-system-in-li.patch
0112_all_Fix-s390-version-of-pt-longjmp.c.patch
0113_all_Add-test-for-bug-21041.patch
0118_all_Add-test-for-bug-21041.patch
0129_all_Update-NEWS-and-ChangeLog-for-CVE-2017-15671.patch
0130_all_glob-Fix-buffer-overflow-during-GLOB_TILDE-unescapin.patch
0132_all_Update-NEWS-to-add-CVE-2017-15804-entry.patch
0135_all_ia64-Add-ipc_priv.h-header-to-set-__IPC_64-to-zero.patch
0136_all_posix-Fix-mmap-for-m68k-and-ia64-BZ-21908.patch
0137_all_ia64-Fix-thread-stack-allocation-permission-set-BZ-2.patch
---
Is there something gone wrong?
Comment 1 Ulenrich 2017-12-22 16:40:44 UTC
I emerged sys-libs/glibc-2.26-r4 and rebooted my machine without error:
Gentoo~unstable amd64 linux-4.14.8 with sys-libs/glibc-2.26-r4
works.

But still 
I especially ask myself what about these two missing patches:

0111_all_Don-t-use-IFUNC-resolver-for-longjmp-or-system-in-li.patch
0130_all_glob-Fix-buffer-overflow-during-GLOB_TILDE-unescapin.patch
Comment 2 Sergei Trofimovich (RETIRED) gentoo-dev 2017-12-22 19:17:39 UTC
My wild guess would be that creation/deletion of files is not handled correctly by a new script preparing patchsets.
Comment 3 Sergei Trofimovich (RETIRED) gentoo-dev 2017-12-22 19:47:27 UTC
Actually patches themselves are incompletely cheryr-picked.
Comment 4 Andreas K. Hüttel archtester gentoo-dev 2017-12-23 20:54:52 UTC
TL;DR: As far as I can see everything is working just fine and exactly as expected...

The explanation:

1) Patches only touching ChangeLog

When I backport a security fix into our patchset that is in upstream master, but not in the release branch (e.g. upstream 2.26) yet, I have to drop the chunk with the changelog addition (since it does not fit to our state of the changelog). When afterwards *upstream* backports the fix as well, it enters the upstream 2.26 branch with an adapted changelog text. And when I afterwards rebase the upstream 2.26 fixes onto our gentoo/2.26 branch, git recognizes that the code part of the commit is already there and drops it, leaving a commit that only touches the changelog...

2) Missing patch numbers

We have a couple of files in our gentoo branches in scripts/gentoo/extras. For example the openrc init script for nscd... The extras subdirectory is added into the patchset tar directly, so it would make no sense to have its contents in the patches as well. The patchset is generated with "git format-patch", but all commits that contain "[no patch]" at the start of the commit message are removed...

I hope this clarifies things. :)

PS. https://github.com/gentoo/glibc/blob/gentoo/2.25/scripts/gentoo/make-tarball.sh