| Message ID | 20220707220436.4105443-1-mail@conchuod.ie (mailing list archive) |
|---|---|
| Headers | show |
| Series | Fix RISC-V's arch-topology reporting | expand |
On 07/07/2022 23:04, Conor Dooley wrote: > From: Conor Dooley <conor.dooley@microchip.com> > > Hey all, > It's my first time messing around with arch/ code at all, let alone > more than one arch, so forgive me if I have screwed up how to do a > migration like this. > > The goal here is the fix the incorrectly reported arch topology on > RISC-V which seems to have been broken since it was added. > cpu, package and thread IDs are all currently reported as -1, so tools > like lstopo think systems have multiple threads on the same core when > this is not true: > https://github.com/open-mpi/hwloc/issues/536 > > Because I want to backport that fix, I implemented store_cpu_topology > for RISC-V before migrating to the generic version which seems to have > just complicated things, but was Sudeep's preferred way of doing it. > > Palmer, I have not marked the first patch as RFC because I would like > it to be taken as a fix for 5.19-rc(late) independently of the rest if > possible. Apparently I forgot to swap RFC for PATCH on it, but point still stands.
From: Conor Dooley <conor.dooley@microchip.com> Hey all, It's my first time messing around with arch/ code at all, let alone more than one arch, so forgive me if I have screwed up how to do a migration like this. The goal here is the fix the incorrectly reported arch topology on RISC-V which seems to have been broken since it was added. cpu, package and thread IDs are all currently reported as -1, so tools like lstopo think systems have multiple threads on the same core when this is not true: https://github.com/open-mpi/hwloc/issues/536 Because I want to backport that fix, I implemented store_cpu_topology for RISC-V before migrating to the generic version which seems to have just complicated things, but was Sudeep's preferred way of doing it. Palmer, I have not marked the first patch as RFC because I would like it to be taken as a fix for 5.19-rc(late) independently of the rest if possible. The rest of the series is RFC b/c I don't know what I am doing ;) I only built tested for arm64, so hopefully it is not broken when used. Thanks, Conor. Conor Dooley (4): riscv: arch-topology: fix default topology reporting arch-topology: add a default implementation of store_cpu_topology() riscv: arch-topology: move riscv to the generic store_cpu_topology() arm64: arch-topology move arm64 to the generic store_cpu_topology() arch/arm64/kernel/smp.c | 16 ++++++++++++-- arch/arm64/kernel/topology.c | 40 ----------------------------------- arch/riscv/Kconfig | 2 +- arch/riscv/kernel/smpboot.c | 4 +++- drivers/base/arch_topology.c | 19 +++++++++++++++++ include/linux/arch_topology.h | 1 + 6 files changed, 38 insertions(+), 44 deletions(-) base-commit: b6f1f2fa2bddd69ff46a190b8120bd440fd50563