From: Robert Richter <robert.richter@amd.com>
To: Ingo Molnar <mingo@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
Stephane Eranian <eranian@google.com>,
LKML <linux-kernel@vger.kernel.org>,
Robert Richter <robert.richter@amd.com>
Subject: [PATCH 07/10] perf, amd: Generalize northbridge constraints code for family 15h
Date: Tue, 19 Jun 2012 20:10:45 +0200 [thread overview]
Message-ID: <1340129448-8690-8-git-send-email-robert.richter@amd.com> (raw)
In-Reply-To: <1340129448-8690-1-git-send-email-robert.richter@amd.com>
Generalize northbridge constraints code for family 15h and use this
code there too. The main difference is the use of a counter mask for
northbridge counters. Families others than 15h use unconstraints mask
(any generic counter) while on family 15h we have 4 separate nb
counters. We get the nb counter bit mask from the constraint that we
get for a nb event. Still, the nb counters are not yet enabled. See
next patch for this.
Signed-off-by: Robert Richter <robert.richter@amd.com>
---
arch/x86/kernel/cpu/perf_event_amd.c | 59 +++++++++++++++++++++++-----------
1 files changed, 40 insertions(+), 19 deletions(-)
diff --git a/arch/x86/kernel/cpu/perf_event_amd.c b/arch/x86/kernel/cpu/perf_event_amd.c
index 27b2806..e538512 100644
--- a/arch/x86/kernel/cpu/perf_event_amd.c
+++ b/arch/x86/kernel/cpu/perf_event_amd.c
@@ -188,20 +188,13 @@ static inline int amd_has_nb(struct cpu_hw_events *cpuc)
return nb && nb->nb_id != -1;
}
-static void amd_put_event_constraints(struct cpu_hw_events *cpuc,
- struct perf_event *event)
+static void amd_put_nb_event_constraints(struct cpu_hw_events *cpuc,
+ struct perf_event *event)
{
- struct hw_perf_event *hwc = &event->hw;
struct amd_nb *nb = cpuc->amd_nb;
int i;
/*
- * only care about NB events
- */
- if (!(amd_has_nb(cpuc) && amd_is_nb_event(hwc)))
- return;
-
- /*
* need to scan whole list because event may not have
* been assigned during scheduling
*
@@ -247,12 +240,13 @@ static void amd_put_event_constraints(struct cpu_hw_events *cpuc,
*
* Given that resources are allocated (cmpxchg), they must be
* eventually freed for others to use. This is accomplished by
- * calling amd_put_event_constraints().
+ * calling amd_put_nb_event_constraints().
*
* Non NB events are not impacted by this restriction.
*/
static struct event_constraint *
-amd_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
+__amd_get_nb_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event,
+ struct event_constraint *c)
{
struct hw_perf_event *hwc = &event->hw;
struct amd_nb *nb = cpuc->amd_nb;
@@ -260,12 +254,6 @@ amd_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
int idx, new = -1;
/*
- * if not NB event or no NB, then no constraints
- */
- if (!(amd_has_nb(cpuc) && amd_is_nb_event(hwc)))
- return &unconstrained;
-
- /*
* detect if already present, if so reuse
*
* cannot merge with actual allocation
@@ -275,7 +263,7 @@ amd_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
* because of successive calls to x86_schedule_events() from
* hw_perf_group_sched_in() without hw_perf_enable()
*/
- for_each_generic_counter(idx) {
+ for_each_set_bit(idx, c->idxmsk, X86_PMC_IDX_MAX) {
if (new == -1 || hwc->idx == idx)
/* assign free slot, prefer hwc->idx */
old = cmpxchg(nb->owners + idx, NULL, event);
@@ -481,7 +469,7 @@ static struct event_constraint amd_f15_PMC50 = EVENT_CONSTRAINT(0, 0x3F, 0);
static struct event_constraint amd_f15_PMC53 = EVENT_CONSTRAINT(0, 0x38, 0);
static struct event_constraint *
-amd_get_event_constraints_f15h(struct cpu_hw_events *cpuc, struct perf_event *event)
+__amd_get_event_constraints_f15h(struct cpu_hw_events *cpuc, struct perf_event *event)
{
struct hw_perf_event *hwc = &event->hw;
unsigned int event_code = amd_get_event_code(hwc);
@@ -552,6 +540,39 @@ amd_get_event_constraints_f15h(struct cpu_hw_events *cpuc, struct perf_event *ev
}
}
+static struct event_constraint *
+amd_get_nb_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event,
+ struct event_constraint *c)
+{
+ /*
+ * only care about NB events
+ */
+ if (!(amd_has_nb(cpuc) && amd_is_nb_event(&event->hw)))
+ return c;
+
+ return __amd_get_nb_event_constraints(cpuc, event, c);
+}
+
+static struct event_constraint *
+amd_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
+{
+ return amd_get_nb_event_constraints(cpuc, event, &unconstrained);
+}
+
+static struct event_constraint *
+amd_get_event_constraints_f15h(struct cpu_hw_events *cpuc, struct perf_event *event)
+{
+ struct event_constraint *c = __amd_get_event_constraints_f15h(cpuc, event);
+ return amd_get_nb_event_constraints(cpuc, event, c);
+}
+
+static void amd_put_event_constraints(struct cpu_hw_events *cpuc,
+ struct perf_event *event)
+{
+ if (amd_has_nb(cpuc) && amd_is_nb_event(&event->hw))
+ amd_put_nb_event_constraints(cpuc, event);
+}
+
static __initconst const struct x86_pmu amd_pmu = {
.name = "AMD",
.handle_irq = x86_pmu_handle_irq,
--
1.7.8.4
next prev parent reply other threads:[~2012-06-19 18:11 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-19 18:10 [PATCH 00/10] perf, x86: Add northbridge counter support for AMD family 15h Robert Richter
2012-06-19 18:10 ` [PATCH 01/10] perf, amd: Rework northbridge event constraints handler Robert Richter
2012-06-19 18:10 ` [PATCH 02/10] perf, x86: Rework counter reservation code Robert Richter
2012-06-19 18:10 ` [PATCH 03/10] perf, x86: Use bitmasks for generic counters Robert Richter
2012-06-19 18:10 ` [PATCH 04/10] perf, x86: Rename Intel specific macros Robert Richter
2012-06-19 18:10 ` [PATCH 05/10] perf, x86: Move Intel specific code to intel_pmu_init() Robert Richter
2012-06-20 9:36 ` Peter Zijlstra
2012-06-20 14:22 ` Robert Richter
2012-06-19 18:10 ` [PATCH 06/10] perf, amd: Unify AMD's generic and family 15h pmus Robert Richter
2012-06-19 18:10 ` Robert Richter [this message]
2012-06-19 18:10 ` [PATCH 08/10] perf, amd: Enable northbridge counters on family 15h Robert Richter
2012-06-19 18:10 ` [PATCH 09/10] perf, x86: Improve debug output in check_hw_exists() Robert Richter
2012-06-19 18:10 ` [PATCH 10/10] perf, amd: Check northbridge event config value Robert Richter
2012-06-20 8:36 ` [PATCH 00/10] perf, x86: Add northbridge counter support for AMD family 15h Stephane Eranian
2012-06-20 8:54 ` Peter Zijlstra
[not found] ` <CABPqkBS9hRxKLsecVK+AgRue6oqTtAg4=0Dpd5Z2VwAUja50fw@mail.gmail.com>
2012-06-20 9:29 ` Robert Richter
2012-06-20 9:38 ` Peter Zijlstra
2012-06-20 10:00 ` Robert Richter
2012-06-20 10:16 ` Peter Zijlstra
2012-06-20 12:29 ` Robert Richter
2012-06-20 15:54 ` Peter Zijlstra
2012-06-20 16:08 ` Peter Zijlstra
2012-06-20 16:21 ` Stephane Eranian
2012-06-20 10:46 ` Stephane Eranian
2012-06-20 12:41 ` Robert Richter
2012-06-20 9:41 ` Peter Zijlstra
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=1340129448-8690-8-git-send-email-robert.richter@amd.com \
--to=robert.richter@amd.com \
--cc=eranian@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=peterz@infradead.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.