Race condition or something when fetching local files?

Race condition or something when fetching local files?

[Martin Jansa]

[-- Attachment #1: Type: text/plain, Size: 10459 bytes --]

I have noticed that even local patches are stored in downloads dir, 
but today I've noticed that wrong file was used later in do_patch.
I have latest oe-core/meta-oe + latest bitbake.

Pulseaudio was building here for first time

downloads/buildfix.patch is from previous build of xf86-video-intel-2.16.0
$ head downloads/buildfix.patch
Newer mesa doesn't provide dri support, just dri2. The intel driver
...

pulseaudio has right one:
 $ head /OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/buildfix.patch
Derived from OE by Dongxiao Xu <dongxiao.xu@intel.com>

Upstream-Status: Pending

Index: pulseaudio-0.9.11/src/pulsecore/atomic.h
...

but while building pulseaudio it's using wrong downloads/buildfix.patch

$ bitbake -e -b /OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio_0.9.23.bb | grep ^FILESPATH | sed "s/:/\n/"
FILESPATH="/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23-r5/linux-gnueabi
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23-r5/arm
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23-r5/build-linux
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23-r5/pn-pulseaudio
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23-r5/nokia900
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23-r5/
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23-r5/shr
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23-r5/forcevariable
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23-r5/libc-glibc
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23-r5/
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/linux-gnueabi
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/arm
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/build-linux
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/pn-pulseaudio
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/nokia900
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/shr
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/forcevariable
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/libc-glibc
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/linux-gnueabi
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/arm
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/build-linux
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/pn-pulseaudio
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/nokia900
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/shr
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/forcevariable
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/libc-glibc
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/linux-gnueabi
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/arm
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/build-linux
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/pn-pulseaudio
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/nokia900
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/shr
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/forcevariable
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/libc-glibc
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio-0.9.23/
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/linux-gnueabi
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/arm
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/build-linux
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/pn-pulseaudio
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/nokia900
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/shr
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/forcevariable
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/libc-glibc
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio/
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/files/linux-gnueabi
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/files/arm
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/files/build-linux
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/files/pn-pulseaudio
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/files/nokia900
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/files/
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/files/shr
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/files/forcevariable
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/files/libc-glibc
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/files/
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/linux-gnueabi
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/arm
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/build-linux
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pn-pulseaudio
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/nokia900
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/shr
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/forcevariable
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/libc-glibc
/OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/"

-c cleansstate doesn't help

rm -rf downloads/buildfix.patch* and probably also -c cleanall _does_ help and do_patch finished now OK.


log from first build:
NOTE: Running task 6777 of 7835 (ID: 6734, /OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio_0.9.23.bb, do_fetch)
NOTE: package pulseaudio-0.9.23-r5: task do_fetch: Started
NOTE: package pulseaudio-0.9.23-r5: task do_fetch: Succeeded
NOTE: Running task 6778 of 7835 (ID: 6730, /OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio_0.9.23.bb, do_unpack)
NOTE: package pulseaudio-0.9.23-r5: task do_unpack: Started
NOTE: package pulseaudio-0.9.23-r5: task do_unpack: Succeeded
NOTE: Running task 6779 of 7835 (ID: 6731, /OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio_0.9.23.bb, do_patch)
NOTE: package pulseaudio-0.9.23-r5: task do_patch: Started
ERROR: Error executing a python function in /OE/shr-core/openembedded-core/meta/recipes-multimedia/pulseaudio/pulseaudio_0.9.23.bb:
CmdError: Command Error: exit status: 1  Output:
Applying patch buildfix.patch
can't find file to patch at input line 13
Perhaps you used the wrong -p or --strip option?
The text leading up to this was:
--------------------------
|Newer mesa doesn't provide dri support, just dri2. The intel driver
|hasn't quite caught up and whilst the configure checks work, the
|global includes have not and need manually tweaking.
|
|RP 2011/10/12
|
|Upstream Status: Superceeded
|
|Index: xf86-video-intel-2.16.0/src/intel.h
|===================================================================
|--- xf86-video-intel-2.16.0.orig/src/intel.h   2011-10-12 20:36:18.587187994 +0100
|+++ xf86-video-intel-2.16.0/src/intel.h        2011-10-12 20:36:30.717188011 +0100
--------------------------
No file to patch.  Skipping patch.
1 out of 1 hunk ignored
Patch buildfix.patch does not apply (enforce with -f)

ERROR: The stack trace of python calls that resulted in this exception/failure was:
ERROR:   File "patch_do_patch", line 136, in <module>
ERROR:
ERROR:   File "patch_do_patch", line 133, in patch_do_patch
ERROR:
ERROR:   File "/OE/shr-core/openembedded-core/meta/lib/oe/patch.py", line 338, in Resolve
ERROR:     raise sys.exc_value
ERROR:
ERROR: The code that was being executed was:
ERROR:      0132:                       raise bb.build.FuncFailed(str(sys.exc_value))
ERROR:      0133:               resolver.Resolve()
ERROR:      0134:
ERROR:      0135:
ERROR:  *** 0136:patch_do_patch(d)
ERROR:      0137:
ERROR: (file: 'patch_do_patch', lineno: 136, function: <module>)
ERROR:      0129:                       patchset.Import({"file":local, "remote":url, "strippath": striplevel}, True)
ERROR:      0130:               except Exception:
ERROR:      0131:                       import sys
ERROR:      0132:                       raise bb.build.FuncFailed(str(sys.exc_value))
ERROR:  *** 0133:               resolver.Resolve()
ERROR:      0134:
ERROR:      0135:
ERROR:      0136:patch_do_patch(d)
ERROR:      0137:
ERROR: (file: 'patch_do_patch', lineno: 133, function: patch_do_patch)
ERROR: Function 'patch_do_patch' failed
ERROR: Logfile of failure stored in: /OE/shr-core/tmp/work/armv7a-vfp-neon-oe-linux-gnueabi/pulseaudio-0.9.23-r5/temp/log.do_patch.28119


-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]

Re: Race condition or something when fetching local files?

[Khem Raj]

On Tue, Oct 18, 2011 at 11:35 AM, Martin Jansa <martin.jansa@gmail.com> wrote:
> I have noticed that even local patches are stored in downloads dir,
> but today I've noticed that wrong file was used later in do_patch.
> I have latest oe-core/meta-oe + latest bitbake.
>
> Pulseaudio was building here for first time
>
> downloads/buildfix.patch is from previous build of xf86-video-intel-2.16.0
> $ head downloads/buildfix.patch

looking in bitbake lib/bb/fetch2/local.py it seems it searches DL_DIR
before other
search paths. cleansstate might not have information that the patch
was downloaded
and not applied from local metadata after the recipe is updated.

Re: Race condition or something when fetching local files?

[Martin Jansa]

[-- Attachment #1: Type: text/plain, Size: 1118 bytes --]

On Tue, Oct 18, 2011 at 01:09:43PM -0700, Khem Raj wrote:
> On Tue, Oct 18, 2011 at 11:35 AM, Martin Jansa <martin.jansa@gmail.com> wrote:
> > I have noticed that even local patches are stored in downloads dir,
> > but today I've noticed that wrong file was used later in do_patch.
> > I have latest oe-core/meta-oe + latest bitbake.
> >
> > Pulseaudio was building here for first time
> >
> > downloads/buildfix.patch is from previous build of xf86-video-intel-2.16.0
> > $ head downloads/buildfix.patch
> 
> looking in bitbake lib/bb/fetch2/local.py it seems it searches DL_DIR
> before other

Which would be wrong for sure if I have
foo_1.0.bb with SRC_URI = "file://bar.patch"

foo-1.0/om-gta02/bar.patch
foo-1.0/nokia900/bar.patch

Then I don't want to use bar.patch from downloads/bar.patch where it's
from om-gta02 build when I'm building nokia900 now.

Regards,

> search paths. cleansstate might not have information that the patch
> was downloaded
> and not applied from local metadata after the recipe is updated.


-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]

Re: Race condition or something when fetching local files?

[Denys Dmytriyenko]

On Tue, Oct 18, 2011 at 10:17:08PM +0200, Martin Jansa wrote:
> On Tue, Oct 18, 2011 at 01:09:43PM -0700, Khem Raj wrote:
> > On Tue, Oct 18, 2011 at 11:35 AM, Martin Jansa <martin.jansa@gmail.com> wrote:
> > > I have noticed that even local patches are stored in downloads dir,
> > > but today I've noticed that wrong file was used later in do_patch.
> > > I have latest oe-core/meta-oe + latest bitbake.
> > >
> > > Pulseaudio was building here for first time
> > >
> > > downloads/buildfix.patch is from previous build of xf86-video-intel-2.16.0
> > > $ head downloads/buildfix.patch
> > 
> > looking in bitbake lib/bb/fetch2/local.py it seems it searches DL_DIR
> > before other
> 
> Which would be wrong for sure if I have
> foo_1.0.bb with SRC_URI = "file://bar.patch"
> 
> foo-1.0/om-gta02/bar.patch
> foo-1.0/nokia900/bar.patch
> 
> Then I don't want to use bar.patch from downloads/bar.patch where it's
> from om-gta02 build when I'm building nokia900 now.

Was there any resolution on this one?

-- 
Denys

Re: Race condition or something when fetching local files?

[Martin Jansa]

[-- Attachment #1: Type: text/plain, Size: 1223 bytes --]

On Thu, Oct 20, 2011 at 08:47:21PM -0400, Denys Dmytriyenko wrote:
> On Tue, Oct 18, 2011 at 10:17:08PM +0200, Martin Jansa wrote:
> > On Tue, Oct 18, 2011 at 01:09:43PM -0700, Khem Raj wrote:
> > > On Tue, Oct 18, 2011 at 11:35 AM, Martin Jansa <martin.jansa@gmail.com> wrote:
> > > > I have noticed that even local patches are stored in downloads dir,
> > > > but today I've noticed that wrong file was used later in do_patch.
> > > > I have latest oe-core/meta-oe + latest bitbake.
> > > >
> > > > Pulseaudio was building here for first time
> > > >
> > > > downloads/buildfix.patch is from previous build of xf86-video-intel-2.16.0
> > > > $ head downloads/buildfix.patch
> > > 
> > > looking in bitbake lib/bb/fetch2/local.py it seems it searches DL_DIR
> > > before other
> > 
> > Which would be wrong for sure if I have
> > foo_1.0.bb with SRC_URI = "file://bar.patch"
> > 
> > foo-1.0/om-gta02/bar.patch
> > foo-1.0/nokia900/bar.patch
> > 
> > Then I don't want to use bar.patch from downloads/bar.patch where it's
> > from om-gta02 build when I'm building nokia900 now.
> 
> Was there any resolution on this one?

NO afaik

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]

Re: Race condition or something when fetching local files?

[Martin Jansa]

[-- Attachment #1: Type: text/plain, Size: 1302 bytes --]

On Tue, Oct 18, 2011 at 10:17:08PM +0200, Martin Jansa wrote:
> On Tue, Oct 18, 2011 at 01:09:43PM -0700, Khem Raj wrote:
> > On Tue, Oct 18, 2011 at 11:35 AM, Martin Jansa <martin.jansa@gmail.com> wrote:
> > > I have noticed that even local patches are stored in downloads dir,
> > > but today I've noticed that wrong file was used later in do_patch.
> > > I have latest oe-core/meta-oe + latest bitbake.
> > >
> > > Pulseaudio was building here for first time
> > >
> > > downloads/buildfix.patch is from previous build of xf86-video-intel-2.16.0
> > > $ head downloads/buildfix.patch
> > 
> > looking in bitbake lib/bb/fetch2/local.py it seems it searches DL_DIR
> > before other
> 
> Which would be wrong for sure if I have
> foo_1.0.bb with SRC_URI = "file://bar.patch"
> 
> foo-1.0/om-gta02/bar.patch
> foo-1.0/nokia900/bar.patch
> 
> Then I don't want to use bar.patch from downloads/bar.patch where it's
> from om-gta02 build when I'm building nokia900 now.
> 
> Regards,
> 
> > search paths. cleansstate might not have information that the patch
> > was downloaded
> > and not applied from local metadata after the recipe is updated.

Reported in
http://bugzilla.pokylinux.org/show_bug.cgi?id=1710

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]