Re: [PATCH 5.15 000/567] 5.15.99-rc1 review

[Daniel Díaz]

Hello!

On 07/03/23 10:55, Greg Kroah-Hartman wrote:
> This is the start of the stable review cycle for the 5.15.99 release.
> There are 567 patches in this series, all will be posted as a response
> to this one.  If anyone has any issues with these being applied, please
> let me know.
>
> Responses should be made by Thu, 09 Mar 2023 16:57:34 +0000.
> Anything received after that time might be too late.
>
> The whole patch series can be found in one patch at:
> 	https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.15.99-rc1.gz
> or in the git tree and branch at:
> 	git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.15.y
> and the diffstat can be found below.
>
> thanks,
>
> greg k-h

We see Perf failing to compile on: arm, arm64, i386, x86_64, under OpenEmbedded (GCC 11.3) when building for the following machines:
* Dragonboard 410c (arm64)
* Dragonboard 845c (arm64)
* Juno (arm64)
* X15 (arm)
* intel-core2-32 (i386)
* intel-corei7-64 (x86_64)

Error:
-----8<-----
util/intel-pt-decoder/intel-pt-decoder.c: In function 'intel_pt_eptw_lookahead_cb':
util/intel-pt-decoder/intel-pt-decoder.c:1445:7: error: 'INTEL_PT_CFE' undeclared (first use in this function); did you mean 'INTEL_PT_CBR'?
 1445 |  case INTEL_PT_CFE:
      |       ^~~~~~~~~~~~
      |       INTEL_PT_CBR
util/intel-pt-decoder/intel-pt-decoder.c:1445:7: note: each undeclared identifier is reported only once for each function it appears in
util/intel-pt-decoder/intel-pt-decoder.c:1446:7: error: 'INTEL_PT_CFE_IP' undeclared (first use in this function); did you mean 'INTEL_PT_BEP_IP'?
 1446 |  case INTEL_PT_CFE_IP:
      |       ^~~~~~~~~~~~~~~
      |       INTEL_PT_BEP_IP
util/intel-pt-decoder/intel-pt-decoder.c:1447:7: error: 'INTEL_PT_EVD' undeclared (first use in this function); did you mean 'INTEL_PT_OVF'?
 1447 |  case INTEL_PT_EVD:
      |       ^~~~~~~~~~~~
      |       INTEL_PT_OVF
----->8-----

Additionally, we see this same problem outside of OpenEmbedded but only on arm (32-bits), because in all other cases it fails to build at an earlier point with the already reported problem of init_disassemble_info() [1][2], so it's very likely to be there too on all architectures.

[1] https://lore.kernel.org/all/CAEUSe7-vBpHrbEy+eQrNZ_LTeqHpn2eQEr3C7cmfNYjK1YL4Ww@xxxxxxxxxxxxxx/
[2] https://lore.kernel.org/all/e6e2df31-6327-f2ad-3049-0cbfa214ae5c@xxxxxxxxxx/

Greetings!

Daniel Díaz
daniel.diaz@xxxxxxxxxx