From: Richard Purdie <richard.purdie@linuxfoundation.org>
To: Koen Kooi <koen@dominion.thruhere.net>
Cc: openembedded-core@lists.openembedded.org,
Koen Kooi <koen.kooi@linaro.org>
Subject: Re: [PATCHv4] git: fix perl binding installation
Date: Fri, 09 May 2014 11:16:07 +0100 [thread overview]
Message-ID: <1399630567.31891.84.camel@ted> (raw)
In-Reply-To: <51490BD6-0BCB-4422-9620-8D4FFAD3791F@dominion.thruhere.net>
On Fri, 2014-05-09 at 11:16 +0200, Koen Kooi wrote:
> Op 9 mei 2014, om 10:50 heeft Koen Kooi <koen@dominion.thruhere.net> het volgende geschreven:
>
> >
> > Op 9 mei 2014, om 08:38 heeft Saul Wold <sgw@linux.intel.com> het volgende geschreven:
> >
> >> On 05/08/2014 10:59 PM, Koen Kooi wrote:
> >>>
> >>> Op 8 mei 2014, om 19:30 heeft Koen Kooi <koen@dominion.thruhere.net> het volgende geschreven:
> >>>
> >>>>
> >>>> Op 8 mei 2014, om 18:37 heeft Saul Wold <sgw@linux.intel.com> het volgende geschreven:
> >>>>
> >>>>> On 05/02/2014 01:36 AM, Koen Kooi wrote:
> >>>>>> Git.pm wasn't ending up in the package because do_install removed it
> >>>>>> with a misleading comment about multilib. Workaround the problem so
> >>>>>> that Git.pm ends up in the correct dir *and* doesn't get deleted.
> >>>>>>
> >>>>>
> >>>>> Seems to be a problem as RP was concerned with:
> >>>>>
> >>>>>
> >>>>>> /srv/ssd/builds/world/tmp/sysroots/x86_64-linux/usr/bin/git difftool
> >>>>>> Can't locate Error.pm in @INC (@INC contains: /srv/ssd/builds/world/tmp/sysroots/x86_64-linux/usr/lib/perl-native/perl/5.14.3 /srv/ssd/builds/world/tmp/sysroots/x86_64-linux/usr/lib/perl-native/perl//5.14.3 /srv/ssd/builds/world/tmp/sysroots/x86_64-linux/usr/lib/perl-native/perl/ /srv/ssd/machines/meta-intel-iot-devkit/build/tmp/sysroots/x86_64-linux/usr/lib/perl-native/perl/5.14.3 .) at /srv/ssd/builds/world/tmp/sysroots/x86_64-linux/usr/lib/git/git-core/git-difftool line 17.
> >>>>>> BEGIN failed--compilation aborted at /srv/ssd/builds/world/tmp/sysroots/x86_64-linux/usr/lib/git/git-core/git-difftool line 17.
> >>>>>
> >>>>> Doing a find in my sysroot:
> >>>>>
> >>>>> /srv/ssd/builds/world/tmp/sysroots/x86_64-linux/usr/lib/perl-native/perl/5.14.3/CPANPLUS/Error.pm
> >>>>> /srv/ssd/builds/world/tmp/sysroots/x86_64-linux/usr/lib/perl/5.14.3/Error.pm
> >>>>
> >>>> Drat, I checked for buildpaths and didn't find anything. I'll retest and send a new version hopefully tomorrow.
> >>>
> >>> I need to read errors better :) Do you have http://patches.openembedded.org/patch/70635/ applied? And did 'difftool' work before this? None of the perl tools worked for me due to error.pm missing.
> >>>
> >> Apparently not, this must have gotten lost during the release crazies.
> >>
> >> Can you please resend a rebased patch set against master with both of these tested using the ' target.
> >
> > What is 'git-replacement-native' and how do I test it?
>
> OK, I figured it out, I remove TMPDIR and sstate-cache, then did:
>
> bitbake git-replacement-native
> bitbake linaro-image-minimal
>
> That blew up with:
>
> RROR: Function failed: Fetcher failure: Fetch command failed with exit code 128, output:
> Cloning into '/build/linaro/build/build/tmp-eglibc/work/x86_64-linux/prelink-native/1.0+gitAUTOINC+6822ec76aa-r0/git'...
> git-upload-pack: error while loading shared libraries: libcrypto.so.1.0.0: cannot open shared object file: No such file or directory
> fatal: Could not read from remote repository.
>
> Please make sure you have the correct access rights
> and the repository exists.
>
> ERROR: Logfile of failure stored in: /build/linaro/build/build/tmp-eglibc/work/x86_64-linux/prelink-native/1.0+gitAUTOINC+6822ec76aa-r0/temp/log.do_unpack.23257
> Log data follows:
> | DEBUG: Executing python function do_unpack
> | DEBUG: Executing python function base_do_unpack
> | DEBUG: Running export PATH="/build/linaro/build/openembedded-core/scripts/native-intercept:/build/linaro/build/openembedded-core/scripts:/build/linaro/build/build/tmp-eglibc/sysroots/x86_64-linux/usr/bin/x86_64-linux:/build/linaro/build
> ld/linaro/build/build/tmp-eglibc/sysroots/x86_64-linux/sbin:/build/linaro/build/build/tmp-eglibc/sysroots/x86_64-linux/bin:/build/linaro/build/openembedded-core/scripts:/build/linaro/build/openembedded-core/bitbake/bin:/usr/lib64/qt-3.3/b
> -s -n /build/linaro/build/build/downloads/git2/git.yoctoproject.org.prelink-cross.git/ /build/linaro/build/build/tmp-eglibc/work/x86_64-linux/prelink-native/1.0+gitAUTOINC+6822ec76aa-r0/git/
> | DEBUG: Python function base_do_unpack finished
> | DEBUG: Python function do_unpack finished
> | ERROR: Function failed: Fetcher failure: Fetch command failed with exit code 128, output:
> | Cloning into '/build/linaro/build/build/tmp-eglibc/work/x86_64-linux/prelink-native/1.0+gitAUTOINC+6822ec76aa-r0/git'...
> | git-upload-pack: error while loading shared libraries: libcrypto.so.1.0.0: cannot open shared object file: No such file or directory
> | fatal: Could not read from remote repository.
> |
> | Please make sure you have the correct access rights
> | and the repository exists.
> |
> ERROR: Task 117 (virtual:native:/build/linaro/build/openembedded-core/meta/recipes-devtools/prelink/prelink_git.bb, do_unpack) failed with exit code '1'
> NOTE: Tasks Summary: Attempted 265 tasks of which 124 didn't need to be rerun and 1 failed.
> Waiting for 0 running tasks to finish:
>
> Summary: 1 task failed:
> virtual:native:/build/linaro/build/openembedded-core/meta/recipes-devtools/prelink/prelink_git.bb, do_unpack
> Summary: There were 3 WARNING messages shown.
> Summary: There was 1 ERROR message shown, returning a non-zero exit code.
>
> real 4m49.390s
> user 8m30.627s
> sys 1m37.368s
>
> Libcrypto does seem to be in sysroot:
>
> [koen@thinkpad build]$ find build/ -name "libcrypto.so.1.0.0"
> build/tmp-eglibc/sysroots/x86_64-linux/lib/libcrypto.so.1.0.0
>
> I don't think this is due to my perl changes, though.
With the native binaries, we encode relative paths to libdir into RPATH.
I'm guessing but I think your "mv xxx" in do_install messes up the
RPATHs in the binaries/libraries.
The fix is probably to make that only happen in target builds
(${CLASS_OVERRIDE} == "class-target" or something like that).
but like I said, I'm guessing.
Cheers,
Richard
next prev parent reply other threads:[~2014-05-09 10:16 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-05-02 8:36 [PATCHv4] git: fix perl binding installation Koen Kooi
2014-05-08 16:37 ` Saul Wold
2014-05-08 17:30 ` Koen Kooi
2014-05-09 5:59 ` Koen Kooi
2014-05-09 6:38 ` Saul Wold
2014-05-09 8:50 ` Koen Kooi
2014-05-09 9:16 ` Koen Kooi
2014-05-09 10:16 ` Richard Purdie [this message]
[not found] ` <CAP0W-Rn5PU7NxYWCKZ59R_oqHDfkRt3xEayhzfhyP-1D6OVtjw@mail.gmail.com>
[not found] ` <1399631039.31891.85.camel@ted>
[not found] ` <CAP0W-R=dx1GNAfXkR-SDyB6JxYBYuiD7BSZ20KiB-UZQBmkrgw@mail.gmail.com>
2014-05-12 8:15 ` Koen Kooi
2014-05-12 9:16 ` Richard Purdie
2014-05-12 10:08 ` Richard Purdie
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=1399630567.31891.84.camel@ted \
--to=richard.purdie@linuxfoundation.org \
--cc=koen.kooi@linaro.org \
--cc=koen@dominion.thruhere.net \
--cc=openembedded-core@lists.openembedded.org \
/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
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox