From: Robert Richter <robert.richter@amd.com>
To: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>,
LKML <linux-kernel@vger.kernel.org>,
oprofile-list <oprofile-list@lists.sourceforge.net>,
Robert Richter <robert.richter@amd.com>,
Thomas Gleixner <tglx@linutronix.de>,
Andi Kleen <andi@firstfloor.org>
Subject: [PATCH 3/9] oprofile, perf, x86: introduce new functions to reserve perfctrs by index
Date: Thu, 4 Mar 2010 16:22:05 +0100 [thread overview]
Message-ID: <1267716131-17908-4-git-send-email-robert.richter@amd.com> (raw)
In-Reply-To: <1267716131-17908-1-git-send-email-robert.richter@amd.com>
Current perfctr reservation code allocates single pmu msrs. The msr
addresses may differ depending on the model and offset calculation is
necessary. This can be easier implemented by reserving a counter by
its index only.
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Robert Richter <robert.richter@amd.com>
---
arch/x86/include/asm/nmi.h | 2 ++
arch/x86/kernel/cpu/perfctr-watchdog.c | 29 ++++++++++++++++++++---------
arch/x86/oprofile/op_model_amd.c | 4 ++--
arch/x86/oprofile/op_model_ppro.c | 4 ++--
4 files changed, 26 insertions(+), 13 deletions(-)
diff --git a/arch/x86/include/asm/nmi.h b/arch/x86/include/asm/nmi.h
index 93da9c3..f5db47d 100644
--- a/arch/x86/include/asm/nmi.h
+++ b/arch/x86/include/asm/nmi.h
@@ -19,6 +19,8 @@ extern void die_nmi(char *str, struct pt_regs *regs, int do_panic);
extern int check_nmi_watchdog(void);
extern int nmi_watchdog_enabled;
extern int avail_to_resrv_perfctr_nmi_bit(unsigned int);
+extern int reserve_perfctr(unsigned int);
+extern void release_perfctr(unsigned int);
extern int reserve_perfctr_nmi(unsigned int);
extern void release_perfctr_nmi(unsigned int);
extern int reserve_evntsel_nmi(unsigned int);
diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c
index 5c129ee..0868b8b 100644
--- a/arch/x86/kernel/cpu/perfctr-watchdog.c
+++ b/arch/x86/kernel/cpu/perfctr-watchdog.c
@@ -117,11 +117,8 @@ int avail_to_resrv_perfctr_nmi_bit(unsigned int counter)
}
EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi_bit);
-int reserve_perfctr_nmi(unsigned int msr)
+int reserve_perfctr(unsigned int counter)
{
- unsigned int counter;
-
- counter = nmi_perfctr_msr_to_bit(msr);
/* register not managed by the allocator? */
if (counter > NMI_MAX_COUNTER_BITS)
return 1;
@@ -130,19 +127,33 @@ int reserve_perfctr_nmi(unsigned int msr)
return 1;
return 0;
}
-EXPORT_SYMBOL(reserve_perfctr_nmi);
+EXPORT_SYMBOL(reserve_perfctr);
-void release_perfctr_nmi(unsigned int msr)
+void release_perfctr(unsigned int counter)
{
- unsigned int counter;
-
- counter = nmi_perfctr_msr_to_bit(msr);
/* register not managed by the allocator? */
if (counter > NMI_MAX_COUNTER_BITS)
return;
clear_bit(counter, perfctr_nmi_owner);
}
+EXPORT_SYMBOL(release_perfctr);
+
+int reserve_perfctr_nmi(unsigned int msr)
+{
+ unsigned int counter;
+ counter = nmi_perfctr_msr_to_bit(msr);
+ return reserve_perfctr(counter);
+}
+EXPORT_SYMBOL(reserve_perfctr_nmi);
+
+void release_perfctr_nmi(unsigned int msr)
+{
+ unsigned int counter;
+
+ counter = nmi_perfctr_msr_to_bit(msr);
+ release_perfctr(counter);
+}
EXPORT_SYMBOL(release_perfctr_nmi);
int reserve_evntsel_nmi(unsigned int msr)
diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index 79e79a6..905995d 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -134,7 +134,7 @@ static void op_amd_fill_in_addresses(struct op_msrs * const msrs)
int i;
for (i = 0; i < NUM_COUNTERS; i++) {
- if (reserve_perfctr_nmi(MSR_K7_PERFCTR0 + i)) {
+ if (reserve_perfctr(i)) {
msrs->counters[i].addr = MSR_K7_PERFCTR0 + i;
msrs->controls[i].addr = MSR_K7_EVNTSEL0 + i;
}
@@ -428,7 +428,7 @@ static void op_amd_shutdown(struct op_msrs const * const msrs)
for (i = 0; i < NUM_COUNTERS; ++i) {
if (msrs->counters[i].addr)
- release_perfctr_nmi(MSR_K7_PERFCTR0 + i);
+ release_perfctr(i);
}
}
diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index d6abdf6..a838ee4 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -35,7 +35,7 @@ static void ppro_fill_in_addresses(struct op_msrs * const msrs)
int i;
for (i = 0; i < num_counters; i++) {
- if (!reserve_perfctr_nmi(MSR_P6_PERFCTR0 + i)) {
+ if (reserve_perfctr(i)) {
msrs->counters[i].addr = MSR_P6_PERFCTR0 + i;
msrs->controls[i].addr = MSR_P6_EVNTSEL0 + i;
}
@@ -192,7 +192,7 @@ static void ppro_shutdown(struct op_msrs const * const msrs)
for (i = 0; i < num_counters; ++i) {
if (msrs->counters[i].addr)
- release_perfctr_nmi(MSR_P6_PERFCTR0 + i);
+ release_perfctr(i);
}
if (reset_value) {
kfree(reset_value);
--
1.7.0
next prev parent reply other threads:[~2010-03-04 16:49 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-03-04 15:22 [PATCH 0/9] oprofile, perf, x86: introduce new functions to reserve perfctrs Robert Richter
2010-03-04 15:22 ` [PATCH 1/9] perf, x86: reduce number of CONFIG_X86_LOCAL_APIC macros Robert Richter
2010-03-04 15:22 ` [PATCH 2/9] oprofile, perf, x86: do not allocate evntsel counter msr Robert Richter
2010-03-04 15:22 ` Robert Richter [this message]
2010-03-20 5:45 ` [PATCH 3/9] oprofile, perf, x86: introduce new functions to reserve perfctrs by index Andi Kleen
2010-03-25 15:52 ` Robert Richter
2010-03-25 19:33 ` Andi Kleen
2010-03-04 15:22 ` [PATCH 4/9] tsc, x86: use new perfctr reservation functions in tsc code Robert Richter
2010-03-04 15:22 ` [PATCH 5/9] perf, x86: use new perfctr reservation functions in perf code Robert Richter
2010-03-04 15:22 ` [PATCH 6/9] oprofile/x86: rework error handler in nmi_setup() Robert Richter
2010-03-04 15:22 ` [PATCH 7/9] oprofile/x86: return -EBUSY if counters are already reserved Robert Richter
2010-03-04 15:22 ` [PATCH 8/9] oprofile/x86: group IBS code Robert Richter
2010-03-04 15:22 ` [PATCH 9/9] oprofile/x86: implement perfctr reservation for IBS Robert Richter
2010-03-04 17:59 ` [PATCH 0/9] oprofile, perf, x86: introduce new functions to reserve perfctrs Peter Zijlstra
2010-03-11 11:48 ` Peter Zijlstra
2010-03-11 12:47 ` Ingo Molnar
2010-03-11 15:45 ` Robert Richter
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1267716131-17908-4-git-send-email-robert.richter@amd.com \
--to=robert.richter@amd.com \
--cc=andi@firstfloor.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=oprofile-list@lists.sourceforge.net \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
Powered by JetHome