[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20160217132903.767990400@linutronix.de>
Date: Wed, 17 Feb 2016 13:47:31 -0000
From: Thomas Gleixner <tglx@...utronix.de>
To: LKML <linux-kernel@...r.kernel.org>
Cc: Peter Zijlstra <peterz@...radead.org>,
Ingo Molnar <mingo@...nel.org>, Borislav Petkov <bp@...en8.de>,
Stephane Eranian <eranian@...gle.com>,
Harish Chegondi <harish.chegondi@...el.com>,
Kan Liang <kan.liang@...el.com>,
Andi Kleen <andi.kleen@...el.com>
Subject: [patch 00/11] x86/perf/intel_uncore: Cleanup and enhancements
While working on the hotplug rewrite I stumbled over the uncore drivers. The
intel_uncore driver particular is a complete trainwreck:
- Lacks any form of proper error handling. Most errors are simply ignored.
- Leaks memory and hardware state in case of failures
- Tries to mimick a per cpu machinery for a facility which is strictly per
package. That is implemented with convoluted alloc/free dancing during cpu
hotplug with magic loops over the online cpus
The series cleans up the mess
- Implement proper error handling
- Switch to a per package storage model
- Make MSR and PCI independent as far as it goes
- Allow it to build as a module
Thanks,
tglx
---
arch/x86/Kconfig | 14
arch/x86/include/asm/topology.h | 3
arch/x86/kernel/cpu/Makefile | 3
arch/x86/kernel/cpu/perf_event_intel_uncore.c | 587 +++++++++-----------
arch/x86/kernel/cpu/perf_event_intel_uncore.h | 24
arch/x86/kernel/cpu/perf_event_intel_uncore_nhmex.c | 6
arch/x86/kernel/cpu/perf_event_intel_uncore_snb.c | 13
arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c | 57 +
lib/cpumask.c | 1
9 files changed, 400 insertions(+), 308 deletions(-)
Powered by blists - more mailing lists