perf, trace: Fix IRQ-disable removal from perf/tracepoint interaction
in [Kernel]
|
Prev: [PATCH 4/5] superblock: add filesystem shrinker operations
Next: [Patch 1/4] Allow arch-specific cleanup before breakpoint unregistration
From: Peter Zijlstra on 25 May 2010 05:10 On Tue, 2010-05-25 at 08:06 +0000, tip-bot for Peter Zijlstra wrote: > Commit-ID: 23ef672d1e8236e6fe606a39963ba2d576972d84 > Gitweb: http://git.kernel.org/tip/23ef672d1e8236e6fe606a39963ba2d576972d84 > Author: Peter Zijlstra <peterz(a)infradead.org> > AuthorDate: Sun, 23 May 2010 20:16:27 +0200 > Committer: Ingo Molnar <mingo(a)elte.hu> > CommitDate: Tue, 25 May 2010 09:02:08 +0200 > > perf, trace: Fix IRQ-disable removal from perf/tracepoint interaction > > On x86 perf_arch_fetch_caller_regs() stores pt_regs::flags, which is the > same as used in local_save_flags(). > > So to avoid another local_save_flags() for the ftrace code, I wanted to > reuse it, and totally forgot the !x86 side of things. > > [ Possibly we could remove the local_save_flags() from the > perf_arch_fetch_caller_regs() as I'm not sure we actually use > pt_regs::flags (Frederic?) and leave the ftrace thing to use > local_save_flags(). ] > > Signed-off-by: Peter Zijlstra <a.p.zijlstra(a)chello.nl> > Acked-by: Paul Mackerras <paulus(a)samba.org> > Cc: Arnaldo Carvalho de Melo <acme(a)redhat.com> > Cc: Frederic Weisbecker <fweisbec(a)gmail.com> > Cc: Mike Galbraith <efault(a)gmx.de> > Cc: Steven Rostedt <rostedt(a)goodmis.org> > LKML-Reference: <20100524042958.GA10628(a)drongo> > Signed-off-by: Ingo Molnar <mingo(a)elte.hu> Could you replace this with: --- Subject: perf, trace: Fix !x86 build issue From: Peter Zijlstra <a.p.zijlstra(a)chello.nl> Date: Tue May 25 09:22:38 CEST 2010 Patch b7e2ecef92 (perf, trace: Optimize tracepoints by removing IRQ-disable from perf/tracepoint interaction) made the unfortunate mistake of assuming the world is x86 only, correct this. The problem was that perf_fetch_caller_regs() did local_save_flags() into regs->flags, and I re-used that to remove another local_save_flags(), forgetting !x86 doesn't have regs->flags. Do the reverse, remove the local_save_flags() from perf_fetch_caller_regs() and let the ftrace site do the local_save_flags() instead. Signed-off-by: Peter Zijlstra <a.p.zijlstra(a)chello.nl> Acked-by: Paul Mackerras <paulus(a)samba.org> --- arch/x86/kernel/cpu/perf_event.c | 6 +++++- kernel/trace/trace_event_perf.c | 4 +++- 2 files changed, 8 insertions(+), 2 deletions(-) Index: linux-2.6/arch/x86/kernel/cpu/perf_event.c =================================================================== --- linux-2.6.orig/arch/x86/kernel/cpu/perf_event.c +++ linux-2.6/arch/x86/kernel/cpu/perf_event.c @@ -1717,7 +1717,11 @@ void perf_arch_fetch_caller_regs(struct */ regs->bp = rewind_frame_pointer(skip + 1); regs->cs = __KERNEL_CS; - local_save_flags(regs->flags); + /* + * We abuse bit 3 to pass exact information, see perf_misc_flags + * and the comment with PERF_EFLAGS_EXACT. + */ + regs->flags = 0; } unsigned long perf_instruction_pointer(struct pt_regs *regs) Index: linux-2.6/kernel/trace/trace_event_perf.c =================================================================== --- linux-2.6.orig/kernel/trace/trace_event_perf.c +++ linux-2.6/kernel/trace/trace_event_perf.c @@ -157,6 +157,7 @@ __kprobes void *perf_trace_buf_prepare(i struct pt_regs *regs, int *rctxp) { struct trace_entry *entry; + unsigned long flags; char *raw_data; int pc; @@ -174,7 +175,8 @@ __kprobes void *perf_trace_buf_prepare(i memset(&raw_data[size - sizeof(u64)], 0, sizeof(u64)); entry = (struct trace_entry *)raw_data; - tracing_generic_entry_update(entry, regs->flags, pc); + local_save_flags(flags); + tracing_generic_entry_update(entry, flags, pc); entry->type = type; return raw_data; -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |