ACK: [SRU][X][PATCH 0/1] Fix for tools/perf build failure (LP: #1858798)

[Colin Ian King]

On 08/01/2020 14:59, Kleber Sacilotto de Souza wrote:
> BugLink: https://bugs.launchpad.net/bugs/1858798
> 
> [Impact]
> xenial/linux after the update to 4.4.208 upstream stable fails to build
> tools/perf on environments that don't have libunwind installed.
> 
> https://launchpad.net/~canonical-kernel-team/+archive/ubuntu/ppa/+build/18530346/+files/buildlog_ubuntu-xenial-powerpc.linux_4.4.0-172.201_BUILDING.txt.gz
> 
> [...]
> builtin-report.c: In function ‘report__setup_sample_type’:
> builtin-report.c:289:6: error: ‘dwarf_callchain_users’ undeclared (first use in this function)
>   if (dwarf_callchain_users) {
>       ^
> [...]
> 
> This is caused by the following commit from linux-4.4.y:
> 
> commit faece3af80729763bb28449fa803d12c962571bd
> Author: Jin Yao <yao.jin at linux.intel.com>
> Date: Fri Oct 11 10:21:22 2019 +0800
> 
>     perf report: Add warning when libunwind not compiled in
> 
>     [ Upstream commit 800d3f561659b5436f8c57e7c26dd1f6928b5615 ]
> 
> [Fix]
> Revert the mentioned commit. The variable 'dwarf_callchain_users' was
> introduced by eabad8c6856f "perf unwind: Do not look just at the global
> callchain_param.record_mode" which as of now hasn't been backported to
> stable linux-4.4.y.
> 
> [Test]
> Build the kernel on PPA for powerpc, ppc64el and s390x.
> 
> [Regression Potential]
> Null. This commit is supposed to only add a warning and is breaking the
> build for some architectures, so reverting it will not cause any regression
> for the users.
> 
> Kleber Sacilotto de Souza (1):
>   Revert "perf report: Add warning when libunwind not compiled in"
> 
>  tools/perf/builtin-report.c | 7 -------
>  1 file changed, 7 deletions(-)
> 

Seems reasonable to me. I doubt we will miss that warning message.

Acked-by: Colin Ian King <colin.king at canonical.com>