Re: tools/build: fixdep versus tools/lib/bpf

["Wangnan (F)" <wangnan0@huawei.com>]

On 2015/10/29 5:01, Arnaldo Carvalho de Melo wrote:
> Em Wed, Oct 28, 2015 at 09:44:50PM +0100, Jiri Olsa escreveu:
>> On Wed, Oct 28, 2015 at 05:37:52PM -0300, Arnaldo Carvalho de Melo wrote:
>>> Em Wed, Oct 28, 2015 at 09:13:52PM +0100, Jiri Olsa escreveu:
>>>> On Wed, Oct 28, 2015 at 01:30:40PM -0300, Arnaldo Carvalho de Melo wrote:
>>>>> Hi Jiri, Wang,
>>>>>
>>>>> 	I'm getting these while doing 'make -C tools/perf build-test':
>>>>>
>>>>>    LD       fixdep-in.o
>>>>>    LINK     fixdep
>>>>> /bin/sh: /home/acme/git/linux/tools/build/fixdep: Permission denied
>>>>> make[6]: *** [bpf.o] Error 1
>>>>> make[5]: *** [libbpf-in.o] Error 2
>>>>> make[4]: *** [/home/acme/git/linux/tools/lib/bpf/libbpf.a] Error 2
>>>>> make[4]: *** Waiting for unfinished jobs....
>>>>>
>>>>>
>>>>> It happens at different tests, i.e. seems like a race somewhere in the
>>>>> build system, can you take a look? It happens with my perf/ebpf branch.

Never see this message before. Tried make build-test, unable to reproduce
in my environment. Maybe my machine is not fast enough?

>>>> could not reproduce, but looks like attached patch should help
>>> I'll test this now, i.e. make it go thru a 'make -C tools/perf
>>> build-test'.
>>>
>>> In the interest of speeding up things, please provide an explanation of
>>> why this should be applied, so that I can add it to the changeset log.
>>>
>>> Thanks a bunch!
>>>
>> The fixdep tool needs to be built as the first binary.
>> Libraries are built in paralel, so each of them needs
>> to depend on fixdep target.
>
> I really need a faster machine, you provided the answer at this point:
>
> - make_tags_O: cd . && make -f Makefile O=/tmp/tmp.xfRz6THR6o
>    DESTDIR=/tmp/tmp.y0nuN0Fr9n tags
> - make_cscope_O: cd . && make -f Makefile O=/tmp/tmp.O6phQXHU4z
>    DESTDIR=/tmp/tmp.5mdMeF1pH2 cscope
> - tarpkg: ./tests/perf-targz-src-pkg .
> - make -C <kernelsrc> tools/perf
> - make -C <kernelsrc>/tools perf
>
> Almost there :-)

So the problem is gone? Thanks to Jiri, and I'll add this to my local tree,
as a separated patch first.

Thank you.

> - Arnaldo