...

Commit 4d90f6cd9813 breaks parallel builds in 2020.01-rc series. This fixes the issue, reverting it also fixes it. Not sure which is the best route.

Signed-off-by: Peter Robinson pbrobinson@gmail.com

Makefile | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/Makefile b/Makefile index 8024baaddf..4330c2f3a4 100644 --- a/Makefile +++ b/Makefile @@ -1895,14 +1895,14 @@ checkarmreloc: u-boot false; \ fi

-tools/version.h: include/version.h +version.h: include/version.h $(call if_changed,copy)

-envtools: scripts_basic $(version_h) $(timestamp_h) tools/version.h +envtools: scripts_basic $(version_h) $(timestamp_h) version.h $(Q)$(MAKE) $(build)=tools/env

tools-only: export TOOLS_ONLY=y -tools-only: scripts_basic $(version_h) $(timestamp_h) tools/version.h +tools-only: scripts_basic $(version_h) $(timestamp_h) version.h $(Q)$(MAKE) $(build)=tools

tools-all: export HOST_TOOLS_ALL=y

It's a heck of a build machine you're throwing this at I assume. Can

I can reproduce it on my laptop when doing a local rpm build, the full logs output on the build system is in the link below. I'm trying to work out if a specific build flag causes the issue.

https://kojipkgs.fedoraproject.org//work/tasks/467/39430467/build.log

you explain a bit more on how this is still the right set of dependencies? Assuming you haven't hooked up Azure so you can get a CI

TBH I'm not sure if does but all the binaries are there and they appear to work OK with testing.

run of the Windows build, I'll take care of that as part of testing this in general, thanks.

I've no idea how to do a Windows build let alone one in Azure, sorry.

Peter

On Wed, Dec 4, 2019 at 8:37 PM Tom Rini trini@konsulko.com wrote:

On Wed, Dec 04, 2019 at 03:53:01PM +0000, Peter Robinson wrote:

...

...

...

Commit 4d90f6cd9813 breaks parallel builds in 2020.01-rc series. This fixes the issue, reverting it also fixes it. Not sure which is the best route.

Signed-off-by: Peter Robinson pbrobinson@gmail.com

Makefile | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/Makefile b/Makefile index 8024baaddf..4330c2f3a4 100644 --- a/Makefile +++ b/Makefile @@ -1895,14 +1895,14 @@ checkarmreloc: u-boot false; \ fi

-tools/version.h: include/version.h +version.h: include/version.h $(call if_changed,copy)

-envtools: scripts_basic $(version_h) $(timestamp_h) tools/version.h +envtools: scripts_basic $(version_h) $(timestamp_h) version.h $(Q)$(MAKE) $(build)=tools/env

tools-only: export TOOLS_ONLY=y -tools-only: scripts_basic $(version_h) $(timestamp_h) tools/version.h +tools-only: scripts_basic $(version_h) $(timestamp_h) version.h $(Q)$(MAKE) $(build)=tools

tools-all: export HOST_TOOLS_ALL=y

It's a heck of a build machine you're throwing this at I assume. Can

I can reproduce it on my laptop when doing a local rpm build, the full logs output on the build system is in the link below. I'm trying to work out if a specific build flag causes the issue.

https://kojipkgs.fedoraproject.org//work/tasks/467/39430467/build.log

Ah, interesting. So the tools directory doesn't exist yet, in this case. I'm not sure why it does when we're building via buildman. Can you try this:

That works for my local build, just trying on the build system but you can add my tested by, thanks for the proper fix, my make fu is limited.

Tested-by: Peter Robinson pbrobinson@gmail.com

diff --git a/Makefile b/Makefile index 0ee0c701022a..960926caad45 100644 --- a/Makefile +++ b/Makefile @@ -1889,6 +1889,7 @@ checkarmreloc: u-boot fi

tools/version.h: include/version.h

•   $(Q)mkdir -p $(dir $@)
    $(call if_changed,copy)
  
  

envtools: scripts_basic $(version_h) $(timestamp_h) tools/version.h

-- Tom