lkml.org 
[lkml]   [2019]   [Nov]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
Subject[tip: perf/core] perf/x86/intel/pt: Factor out pt_config_start()
The following commit has been merged into the perf/core branch of tip:

Commit-ID: 8e105a1fc2a02d78698834974083c980d2e5b513
Gitweb: https://git.kernel.org/tip/8e105a1fc2a02d78698834974083c980d2e5b513
Author: Alexander Shishkin <alexander.shishkin@linux.intel.com>
AuthorDate: Fri, 25 Oct 2019 17:08:34 +03:00
Committer: Ingo Molnar <mingo@kernel.org>
CommitterDate: Wed, 13 Nov 2019 11:06:15 +01:00

perf/x86/intel/pt: Factor out pt_config_start()

PT trace is now enabled at the bottom of the event configuration
function that takes care of all configuration bits related to a given
event, including the address filter update. This is only needed where
the event configuration changes, that is, in ->add()/->start().

In the interrupt path we can use a lighter version that keeps the
configuration intact, since it hasn't changed, and only flips the
enable bit.

Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Cc: adrian.hunter@intel.com
Cc: mathieu.poirier@linaro.org
Link: https://lkml.kernel.org/r/20191025140835.53665-3-alexander.shishkin@linux.intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
arch/x86/events/intel/pt.c | 22 ++++++++++++++++------
1 file changed, 16 insertions(+), 6 deletions(-)

diff --git a/arch/x86/events/intel/pt.c b/arch/x86/events/intel/pt.c
index 05e43d0..170f3b4 100644
--- a/arch/x86/events/intel/pt.c
+++ b/arch/x86/events/intel/pt.c
@@ -397,6 +397,20 @@ static bool pt_event_valid(struct perf_event *event)
* These all are cpu affine and operate on a local PT
*/

+static void pt_config_start(struct perf_event *event)
+{
+ struct pt *pt = this_cpu_ptr(&pt_ctx);
+ u64 ctl = event->hw.config;
+
+ ctl |= RTIT_CTL_TRACEEN;
+ if (READ_ONCE(pt->vmx_on))
+ perf_aux_output_flag(&pt->handle, PERF_AUX_FLAG_PARTIAL);
+ else
+ wrmsrl(MSR_IA32_RTIT_CTL, ctl);
+
+ WRITE_ONCE(event->hw.config, ctl);
+}
+
/* Address ranges and their corresponding msr configuration registers */
static const struct pt_address_range {
unsigned long msr_a;
@@ -468,7 +482,6 @@ static u64 pt_config_filters(struct perf_event *event)

static void pt_config(struct perf_event *event)
{
- struct pt *pt = this_cpu_ptr(&pt_ctx);
u64 reg;

/* First round: clear STATUS, in particular the PSB byte counter. */
@@ -501,10 +514,7 @@ static void pt_config(struct perf_event *event)
reg |= (event->attr.config & PT_CONFIG_MASK);

event->hw.config = reg;
- if (READ_ONCE(pt->vmx_on))
- perf_aux_output_flag(&pt->handle, PERF_AUX_FLAG_PARTIAL);
- else
- wrmsrl(MSR_IA32_RTIT_CTL, reg);
+ pt_config_start(event);
}

static void pt_config_stop(struct perf_event *event)
@@ -1381,7 +1391,7 @@ void intel_pt_interrupt(void)

pt_config_buffer(topa_to_page(buf->cur)->table, buf->cur_idx,
buf->output_off);
- pt_config(event);
+ pt_config_start(event);
}
}

\
 
 \ /
  Last update: 2019-11-13 11:57    [W:0.229 / U:1.092 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site