Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1707753 > unrolled thread
| Started by | Vikas Shivappa <vikas.shivappa@linux.intel.com> |
|---|---|
| First post | 2017-08-09 20:50 +0200 |
| Last post | 2017-08-09 20:50 +0200 |
| Articles | 2 — 1 participant |
Back to article view | Back to linux.kernel
[PATCH 0/3 V3] Cqm3 based on resctrl fs with MBM support Vikas Shivappa <vikas.shivappa@linux.intel.com> - 2017-08-09 20:50 +0200
[PATCH 1/3] x86/intel_rdt/cqm: Add fix to clear the default RMID during hotcpu Vikas Shivappa <vikas.shivappa@linux.intel.com> - 2017-08-09 20:50 +0200
| From | Vikas Shivappa <vikas.shivappa@linux.intel.com> |
|---|---|
| Date | 2017-08-09 20:50 +0200 |
| Subject | [PATCH 0/3 V3] Cqm3 based on resctrl fs with MBM support |
| Message-ID | <ucEdj-2Fa-7@gated-at.bofh.it> |
Sending some changes to the V2 version based on Thomas feedback that the worst case limbo processing in IPI was still not acceptable and to improve the cache line access during scheduling. This also includes a fix during hot cpu processing which was reported. Patches are based on the tip/x86/cache (where the V2 was merged) Vikas Shivappa (3): x86/intel_rdt/cqm: Add fix to clear the default RMID during hotcpu x86/intel_rdt: Modify the intel_pqr_state for better performance x86/intel_rdt/cqm: Improve limbo list processing arch/x86/include/asm/intel_rdt_sched.h | 30 +++-- arch/x86/kernel/cpu/intel_rdt.c | 34 +++-- arch/x86/kernel/cpu/intel_rdt.h | 10 ++ arch/x86/kernel/cpu/intel_rdt_monitor.c | 209 +++++++++++++------------------ arch/x86/kernel/cpu/intel_rdt_rdtgroup.c | 10 +- 5 files changed, 146 insertions(+), 147 deletions(-) -- 1.9.1
[toc] | [next] | [standalone]
| From | Vikas Shivappa <vikas.shivappa@linux.intel.com> |
|---|---|
| Date | 2017-08-09 20:50 +0200 |
| Subject | [PATCH 1/3] x86/intel_rdt/cqm: Add fix to clear the default RMID during hotcpu |
| Message-ID | <ucEdj-2Fa-11@gated-at.bofh.it> |
| In reply to | #1707753 |
The user configured per cpu default RMID is not cleared during cpu hotplug. This may lead to incorrect RMID values after a cpu goes offline and again comes back online. Clear the per cpu default RMID during cpu offline and online handling. Reported-by: Prakyha Sai Praneeth <sai.praneeth.prakhya@intel.com> Signed-off-by: Vikas Shivappa <vikas.shivappa@linux.intel.com> --- arch/x86/kernel/cpu/intel_rdt.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/kernel/cpu/intel_rdt.c b/arch/x86/kernel/cpu/intel_rdt.c index da4f389..4b9edb2 100644 --- a/arch/x86/kernel/cpu/intel_rdt.c +++ b/arch/x86/kernel/cpu/intel_rdt.c @@ -551,6 +551,7 @@ static void clear_closid_rmid(int cpu) struct intel_pqr_state *state = this_cpu_ptr(&pqr_state); per_cpu(rdt_cpu_default.closid, cpu) = 0; + per_cpu(rdt_cpu_default.rmid, cpu) = 0; state->closid = 0; state->rmid = 0; wrmsr(IA32_PQR_ASSOC, 0, 0); -- 1.9.1
[toc] | [prev] | [standalone]
Back to top | Article view | linux.kernel
csiph-web