https://bugs.winehq.org/show_bug.cgi?id=57016
Bug ID: 57016 Summary: staging patchinstall.py fails on security wine_todo Product: Wine-staging Version: 9.13 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs@winehq.org Reporter: zlice@crtdrift.us CC: leslie_alistair@hotmail.com, z.figura12@gmail.com Distribution: ---
Created attachment 76859 --> https://bugs.winehq.org/attachment.cgi?id=76859 current_broken_diff
wine-staging-9.13/patches/server-Stored_ACLs/0004-server-Temporarily-store-the-full-security-descripto.patch
https://bugs.winehq.org/show_bug.cgi?id=57016
--- Comment #1 from zlice zlice@crtdrift.us --- Created attachment 76860 --> https://bugs.winehq.org/attachment.cgi?id=76860 updated_working_diff
https://bugs.winehq.org/show_bug.cgi?id=57016
--- Comment #2 from zlice zlice@crtdrift.us --- attachments made me not get to the actual issue...
staging/patchinstall.py -all
will fail trying to patch for security tests.
wine-staging-9.13/patches/server-Stored_ACLs/0004-server-Temporarily-store-the-full-security-descripto.patch
Can not find the lines in dlls/advapi32/tests/security.c which seem to be off and need updated.
https://bugs.winehq.org/show_bug.cgi?id=57016
aurb arusanu.bu@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |arusanu.bu@gmail.com
--- Comment #3 from aurb arusanu.bu@gmail.com --- I have a similar issue with this specific patch mentioned here but in a strange way. In my desktop environment, bash/KDE, the patch applies successfully with no hiccups. However, the patching fails in a chroot environment and I haven't figure out why yet.
https://bugs.winehq.org/show_bug.cgi?id=57016
--- Comment #4 from Zeb Figura z.figura12@gmail.com --- I can't reproduce this. Can you please past the error message you see?
https://bugs.winehq.org/show_bug.cgi?id=57016
--- Comment #5 from zlice zlice@crtdrift.us --- Created attachment 76964 --> https://bugs.winehq.org/attachment.cgi?id=76964 patch_fail
This is done in a chroot technically with Void's build system.
But the `dlls/advapi32/tests/security.c` file as originally attached, does not have the same line numbers as the patches in wine's staging patch dir it seems.
https://bugs.winehq.org/show_bug.cgi?id=57016
--- Comment #6 from Zeb Figura z.figura12@gmail.com --- (In reply to zlice from comment #5)
Created attachment 76964 [details] patch_fail
This is done in a chroot technically with Void's build system.
But the `dlls/advapi32/tests/security.c` file as originally attached, does not have the same line numbers as the patches in wine's staging patch dir it seems.
This seems to be part of some broader build system, given the "post_patch". That kind of thing really needs to be stated up front. Can you reproduce the failure using commands directly?
Also, there's something odd about the output, it's interleaved in a nonsensical way. Is patchinstall.py somehow being run in parallel?
https://bugs.winehq.org/show_bug.cgi?id=57016
--- Comment #7 from aurb arusanu.bu@gmail.com --- Created attachment 76965 --> https://bugs.winehq.org/attachment.cgi?id=76965 full error patching log
(In reply to Zeb Figura from comment #4)
I can't reproduce this. Can you please past the error message you see?
There is not much output similar to comment #5, here is the gist of it:
+ ./staging/patchinstall.py DESTDIR=./ --all patching dlls/ntdll/loader.c .... patching dlls/advapi32/tests/security.c 1 out of 4 hunks FAILED -- saving rejects to file dlls/advapi32/tests/security.c.rej [PATCH] ERR: Textual patch did not apply, aborting. .... /home/aurb/rpmbuild/BUILD/wine-9.15/staging/../patches/gitapply.sh -d ./ < /home/aurb/rpmbuild/BUILD/wine-9.15/staging/../patches/server-Stored_ACLs/0004-server-Temporarily-store-the-full-security-descripto.patch Failed to apply patch server-Stored_ACLs/0004-server-Temporarily-store-the-full-security-descripto.patch error: Bad exit status from /home/aurb/rpmbuild/tmp/rpm-tmp.6yu3QR (%prep)
https://bugs.winehq.org/show_bug.cgi?id=57016
--- Comment #8 from aurb arusanu.bu@gmail.com --- Created attachment 76966 --> https://bugs.winehq.org/attachment.cgi?id=76966 rejected bit of the patch
The failing patch part.
https://bugs.winehq.org/show_bug.cgi?id=57016
--- Comment #9 from zlice zlice@crtdrift.us --- (In reply to Zeb Figura from comment #6)
(In reply to zlice from comment #5)
Created attachment 76964 [details] patch_fail
This is done in a chroot technically with Void's build system.
But the `dlls/advapi32/tests/security.c` file as originally attached, does not have the same line numbers as the patches in wine's staging patch dir it seems.
This seems to be part of some broader build system, given the "post_patch". That kind of thing really needs to be stated up front. Can you reproduce the failure using commands directly?
Also, there's something odd about the output, it's interleaved in a nonsensical way. Is patchinstall.py somehow being run in parallel?
It does work extracting manually. It looks like gitapply.sh has a `nogit` var that doesn't get set. Setting this to 1 by default will result in the same error on `dlls/advapi32/tests/security.c`
https://bugs.winehq.org/show_bug.cgi?id=57016
--- Comment #10 from Zeb Figura z.figura12@gmail.com --- Thanks, that helps.
Can we just require git at build time, rather than spending time debugging this? Is that a blocker for any of these environments?
https://bugs.winehq.org/show_bug.cgi?id=57016
--- Comment #11 from zlice zlice@crtdrift.us --- I have no clue what keeps happening, commenting and logging in seem to be just doing whatever they want on firefox.
Adding 'git' to the build deps allows patchinstall.py and gitapply.sh to run. The original issue isn't exactly accurate and the suggested diff-patches are probably not needed. Good enough for a work around to install git to chroot.
https://bugs.winehq.org/show_bug.cgi?id=57016
--- Comment #12 from zlice zlice@crtdrift.us --- (In reply to Zeb Figura from comment #10)
Thanks, that helps.
Can we just require git at build time, rather than spending time debugging this? Is that a blocker for any of these environments?
Yes, this is what I added. Not sure if it should be fixed otherwise but that's good enough. Clearly no one using Void has built staging in a while since this was just brought up the other week/day :)
https://bugs.winehq.org/show_bug.cgi?id=57016
--- Comment #13 from aurb arusanu.bu@gmail.com --- Adding git works for me too. Great, thanks.
-
WineHQ Bugzilla