On 4/8/25 21:57, Catalin Marinas wrote:
On Tue, Apr 08, 2025 at 03:34:51PM +0530, Anshuman Khandual wrote:
On 4/8/25 15:23, Greg KH wrote:
On Tue, Apr 08, 2025 at 03:08:52PM +0530, Anshuman Khandual wrote:
This series adds fine grained trap control in EL2 required for FEAT_PMUv3p9 registers like PMICNTR_EL0, PMICFILTR_EL0, and PMUACR_EL1 which are already being used in the kernel. This is required to prevent their EL1 access trap into EL2.
The following commits that enabled access into FEAT_PMUv3p9 registers have already been merged upstream from 6.13 onwards.
d8226d8cfbaf ("perf: arm_pmuv3: Add support for Armv9.4 PMU instruction counter") 0bbff9ed8165 ("perf/arm_pmuv3: Add PMUv3.9 per counter EL0 access control")
The sysreg patches in this series are required for the final patch which fixes the actual problem.
But you aren't going to fix the 6.14.y tree? We can't take patches that skip newer stable releases for obvious reasons.
And 6.13.y is only going to be alive for a few more days, is there some specific reason this is needed now for 6.13.y?
I have also sent same series for 6.14 stable version as well. It will be great to have these patches applied both on 6.13 as well 6.14. Thank you.
TBH, 6.13 is end of life soon, so not sure it's worth carrying those patches. Do you have a reason for this Anshuman?
Not particularly for 6.13.
We can just skip these from 6.13 if that is not going to be around for long. I was under the impression that all maintained stable releases starting from where this problem got introduced upto now, need to be covered for this back porting.