On Mon, May 25, 2026 at 07:51:16PM +0800, Fenghua Yu wrote: > Hi, Richard, > > On 5/25/26 19:39, Richard Cheng wrote: > > aarch64 has no vendor_id in /proc/cpuinfo, so detect_vendor() returns 0 > > and arch_supports_noncont_cat() falls through to "return false". > > L3_NONCONT_CAT spuriously fails on every ARM MPAM platform. > > > > Define ARCH_ARM, short-circuite detect_vendor() to it on aarch64, and > > add it to the AMD/Hygon alwasy-supports early-out in > > arch_support_noncont_cat(). > > > > aarch64 has many implementers (ARM 0x41, NVIDIA 0x43, etc...) but MPAM > > mandates non-contiguous CPBM uniformly, so per-implementer bits is not > > needed here. > > > > Signed-off-by: Richard Cheng <[email protected]> > > I see you have three patches for selftest. Seems they are together. But they > are not assembled in a series like [PATCH 1/3], [PATCH 2/3], [PATCH 3/3]. > > Could you please send them in a series? > > Thanks. > > -Fenghua
Hi Fenghua, Thanks, I'll organize them in a serie and send a v2. Best regards, Richard Cheng.
