Re: [oe-commits] [openembedded-core] 11/18: busybox: Backport patches to support removal of __NR_clock_gettime

Re: [oe-commits] [openembedded-core] 11/18: busybox: Backport patches to support removal of __NR_clock_gettime

[Pierre-Jean Texier]

Hello All,

Le 15/02/2020 à 11:28, git@git.openembedded.org a écrit :
>      busybox: Backport patches to support removal of __NR_clock_gettime
>      
>      This helps compiling with musl on 32bit arches now that musl has
>      switched to 64bit time_t
>      
>      Signed-off-by: Khem Raj<raj.khem@gmail.com>
>      Signed-off-by: Richard Purdie<richard.purdie@linuxfoundation.org>
> ---
>   ...all-wrappers-around-clock_gettime-closes-.patch | 120 +++++++++++++++++++++
>   ...-date-Use-64-prefix-syscall-if-we-have-to.patch |  53 +++++++++
>   ...-time-Use-64-prefix-syscall-if-we-have-to.patch |  43 ++++++++
>   ...runsv-Use-64-prefix-syscall-if-we-have-to.patch |  46 ++++++++
>   meta/recipes-core/busybox/busybox_1.31.1.bb        |   4 +
>   5 files changed, 266 insertions(+)

this commit causes the following build issue :

  ERROR: busybox-1.31.1-r0 do_patch: Command Error: 'quilt --quiltrc 
/home/runner/work/kas-files/kas-files/build/tmp/work/cortexa7t2hf-neon-vfpv4-poky-linux-gnueabi/busybox/1.31.1-r0/recipe-sysroot-native/etc/quiltrc 
push' exited with 0  Output:
Applying patch 
0001-Remove-syscall-wrappers-around-clock_gettime-closes-.patch
patching file Makefile.flags
patching file coreutils/date.c
Hunk #2 FAILED at 270.
1 out of 2 hunks FAILED -- rejects in file coreutils/date.c
patching file libbb/time.c
patching file runit/runsv.c
Patch 0001-Remove-syscall-wrappers-around-clock_gettime-closes-.patch 
does not apply (enforce with -f)
ERROR: Logfile of failure stored in: 
/home/runner/work/kas-files/kas-files/build/tmp/work/cortexa7t2hf-neon-vfpv4-poky-linux-gnueabi/busybox/1.31.1-r0/temp/log.do_patch.62122
ERROR: Task 
(/home/runner/work/kas-files/kas-files/poky/meta/recipes-core/busybox/busybox_1.31.1.bb:do_patch) 
failed with exit code '1'

Tested on top of poky master (commit 
7f2af8bd795ace46828952a7812f0bbdf26529a [1])

[1] - https://github.com/texierp/kas-files/runs/447729194

Thanks !

Regards
-- 
Pierre-Jean Texier
Embedded Linux Engineer
https://koncepto.io

Re: [oe-commits] [openembedded-core] 11/18: busybox: Backport patches to support removal of __NR_clock_gettime

[Martin Jansa]

Isn't it caused by some other layer prepending some busybox patch causing
this? Can you check with bitbake -e?

I'm not seeing this one in my builds (and I was just refreshing some of our
webOS OSE patches to apply cleanly on 1.31.1)

Regards,

On Sat, Feb 15, 2020 at 5:30 PM Pierre-Jean Texier via Openembedded-commits
<openembedded-commits@lists.openembedded.org> wrote:

> Hello All,
>
> Le 15/02/2020 à 11:28, git@git.openembedded.org a écrit :
> >      busybox: Backport patches to support removal of __NR_clock_gettime
> >
> >      This helps compiling with musl on 32bit arches now that musl has
> >      switched to 64bit time_t
> >
> >      Signed-off-by: Khem Raj<raj.khem@gmail.com>
> >      Signed-off-by: Richard Purdie<richard.purdie@linuxfoundation.org>
> > ---
> >   ...all-wrappers-around-clock_gettime-closes-.patch | 120
> +++++++++++++++++++++
> >   ...-date-Use-64-prefix-syscall-if-we-have-to.patch |  53 +++++++++
> >   ...-time-Use-64-prefix-syscall-if-we-have-to.patch |  43 ++++++++
> >   ...runsv-Use-64-prefix-syscall-if-we-have-to.patch |  46 ++++++++
> >   meta/recipes-core/busybox/busybox_1.31.1.bb        |   4 +
> >   5 files changed, 266 insertions(+)
>
> this commit causes the following build issue :
>
>   ERROR: busybox-1.31.1-r0 do_patch: Command Error: 'quilt --quiltrc
> /home/runner/work/kas-files/kas-files/build/tmp/work/cortexa7t2hf-neon-vfpv4-poky-linux-gnueabi/busybox/1.31.1-r0/recipe-sysroot-native/etc/quiltrc
>
> push' exited with 0  Output:
> Applying patch
> 0001-Remove-syscall-wrappers-around-clock_gettime-closes-.patch
> patching file Makefile.flags
> patching file coreutils/date.c
> Hunk #2 FAILED at 270.
> 1 out of 2 hunks FAILED -- rejects in file coreutils/date.c
> patching file libbb/time.c
> patching file runit/runsv.c
> Patch 0001-Remove-syscall-wrappers-around-clock_gettime-closes-.patch
> does not apply (enforce with -f)
> ERROR: Logfile of failure stored in:
>
> /home/runner/work/kas-files/kas-files/build/tmp/work/cortexa7t2hf-neon-vfpv4-poky-linux-gnueabi/busybox/1.31.1-r0/temp/log.do_patch.62122
> ERROR: Task
>
> (/home/runner/work/kas-files/kas-files/poky/meta/recipes-core/busybox/busybox_1.31.1.bb:do_patch)
>
> failed with exit code '1'
>
> Tested on top of poky master (commit
> 7f2af8bd795ace46828952a7812f0bbdf26529a [1])
>
> [1] - https://github.com/texierp/kas-files/runs/447729194
>
> Thanks !
>
> Regards
> --
> Pierre-Jean Texier
> Embedded Linux Engineer
> https://koncepto.io
> --
> _______________________________________________
> Openembedded-commits mailing list
> Openembedded-commits@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-commits
>

Re: [oe-commits] [openembedded-core] 11/18: busybox: Backport patches to support removal of __NR_clock_gettime

[Pierre-Jean Texier]

Martin,

Le 15/02/2020 à 17:38, Martin Jansa a écrit :
> Isn't it caused by some other layer prepending some busybox patch causing
> this? Can you check with bitbake -e?
> 
> I'm not seeing this one in my builds (and I was just refreshing some of our
> webOS OSE patches to apply cleanly on 1.31.1)

You're indeed right :(, after a quick check, it seems the build failure
comes from the meta-riscv.

Thanks, and sorry for the noise.

Regards,
-- 
Pierre-Jean Texier
Embedded Linux Engineer
https://koncepto.io

Re: [oe-commits] [openembedded-core] 11/18: busybox: Backport patches to support removal of __NR_clock_gettime

[Khem Raj]

On Sat, Feb 15, 2020 at 8:46 AM Pierre-Jean Texier <pjtexier@koncepto.io> wrote:
>
> Martin,
>
> Le 15/02/2020 à 17:38, Martin Jansa a écrit :
> > Isn't it caused by some other layer prepending some busybox patch causing
> > this? Can you check with bitbake -e?
> >
> > I'm not seeing this one in my builds (and I was just refreshing some of our
> > webOS OSE patches to apply cleanly on 1.31.1)
>
> You're indeed right :(, after a quick check, it seems the build failure
> comes from the meta-riscv.
>
> Thanks, and sorry for the noise.
>

some of these patches were in meta-riscv which is fixed now.

> Regards,
> --
> Pierre-Jean Texier
> Embedded Linux Engineer
> https://koncepto.io