From: Vineet Gupta <Vineet.Gupta1@synopsys.com>
To: <linux-snps-arc@lists.infradead.org>
Cc: <linux-kernel@vger.kernel.org>,
Peter Zijlstra <peterz@infradead.org>,
Vineet Gupta <Vineet.Gupta1@synopsys.com>
Subject: [PATCH 1/4] ARCv2: perf: tweak overflow interrupt
Date: Tue, 7 Nov 2017 14:13:01 -0800 [thread overview]
Message-ID: <1510092784-19942-2-git-send-email-vgupta@synopsys.com> (raw)
In-Reply-To: <1510092784-19942-1-git-send-email-vgupta@synopsys.com>
Current perf ISR loops thru all 32 counters, checking for each if it
caused the interrupt. Instead only loop thru counters which actually
interrupted (typically 1).
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
---
arch/arc/kernel/perf_event.c | 24 ++++++++++++++----------
1 file changed, 14 insertions(+), 10 deletions(-)
diff --git a/arch/arc/kernel/perf_event.c b/arch/arc/kernel/perf_event.c
index 2ce24e74f879..0eaa132a2c90 100644
--- a/arch/arc/kernel/perf_event.c
+++ b/arch/arc/kernel/perf_event.c
@@ -377,21 +377,22 @@ static irqreturn_t arc_pmu_intr(int irq, void *dev)
struct perf_sample_data data;
struct arc_pmu_cpu *pmu_cpu = this_cpu_ptr(&arc_pmu_cpu);
struct pt_regs *regs;
- int active_ints;
+ unsigned int active_ints;
int idx;
arc_pmu_disable(&arc_pmu->pmu);
active_ints = read_aux_reg(ARC_REG_PCT_INT_ACT);
+ if (!active_ints)
+ goto done;
regs = get_irq_regs();
- for (idx = 0; idx < arc_pmu->n_counters; idx++) {
- struct perf_event *event = pmu_cpu->act_counter[idx];
+ do {
+ struct perf_event *event;
struct hw_perf_event *hwc;
- if (!(active_ints & (1 << idx)))
- continue;
+ idx = __ffs(active_ints);
/* Reset interrupt flag by writing of 1 */
write_aux_reg(ARC_REG_PCT_INT_ACT, 1 << idx);
@@ -404,19 +405,22 @@ static irqreturn_t arc_pmu_intr(int irq, void *dev)
write_aux_reg(ARC_REG_PCT_INT_CTRL,
read_aux_reg(ARC_REG_PCT_INT_CTRL) | (1 << idx));
+ event = pmu_cpu->act_counter[idx];
hwc = &event->hw;
WARN_ON_ONCE(hwc->idx != idx);
arc_perf_event_update(event, &event->hw, event->hw.idx);
perf_sample_data_init(&data, 0, hwc->last_period);
- if (!arc_pmu_event_set_period(event))
- continue;
+ if (arc_pmu_event_set_period(event)) {
+ if (perf_event_overflow(event, &data, regs))
+ arc_pmu_stop(event, 0);
+ }
- if (perf_event_overflow(event, &data, regs))
- arc_pmu_stop(event, 0);
- }
+ active_ints &= ~(1U << idx);
+ } while (active_ints);
+done:
arc_pmu_enable(&arc_pmu->pmu);
return IRQ_HANDLED;
--
2.7.4
next prev parent reply other threads:[~2017-11-07 22:14 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-11-07 22:13 [PATCH 0/4] ARC perf updates Vineet Gupta
2017-11-07 22:13 ` Vineet Gupta [this message]
2017-11-07 22:13 ` [PATCH 2/4] ARCv2: perf: optimize given that num counters <= 32 Vineet Gupta
2017-11-07 22:13 ` [PATCH 3/4] ARC: perf: avoid vmalloc backed mmap Vineet Gupta
2017-11-07 22:13 ` [PATCH 4/4] ARCv2: entry: Reduce perf intr return path Vineet Gupta
2017-11-14 10:28 ` Peter Zijlstra
2017-11-14 23:01 ` Vineet Gupta
2017-11-15 10:18 ` Peter Zijlstra
2017-11-17 23:42 ` Vineet Gupta
2017-11-21 23:26 ` Vineet Gupta
2017-11-13 21:30 ` [PATCH 0/4] ARC perf updates Vineet Gupta
2017-11-21 22:09 ` Vineet Gupta
2017-11-21 23:13 ` 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=1510092784-19942-2-git-send-email-vgupta@synopsys.com \
--to=vineet.gupta1@synopsys.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-snps-arc@lists.infradead.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).