mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: mingo@kernel.org, peterz@infradead.org
Cc: vincent.weaver@maine.edu, eranian@google.com, jolsa@redhat.com,
	linux-kernel@vger.kernel.org
Subject: [PATCH 05/10] perf/x86: Simplify dynamic constraint code somewhat
Date: Thu, 21 May 2015 13:17:15 +0200	[thread overview]
Message-ID: <20150521111933.002158797@infradead.org> (raw)
In-Reply-To: <20150521111710.475482798@infradead.org>

[-- Attachment #1: peterz-pmu-sched-4.patch --]
[-- Type: text/plain, Size: 2788 bytes --]

Instead of using cx after the dynamic allocation, put all cx inside
the dynamic allocation block and use c outside of it.

Also use direct assignment to copy the structure; let the compiler
figure it out.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
 arch/x86/kernel/cpu/perf_event_intel.c |   21 ++++++++++-----------
 1 file changed, 10 insertions(+), 11 deletions(-)

--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -1970,7 +1970,6 @@ static struct event_constraint *
 intel_get_excl_constraints(struct cpu_hw_events *cpuc, struct perf_event *event,
 			   int idx, struct event_constraint *c)
 {
-	struct event_constraint *cx;
 	struct intel_excl_cntrs *excl_cntrs = cpuc->excl_cntrs;
 	struct intel_excl_states *xlo;
 	int tid = cpuc->excl_thread_id;
@@ -1989,8 +1988,6 @@ intel_get_excl_constraints(struct cpu_hw
 	if (!excl_cntrs)
 		return c;
 
-	cx = c;
-
 	/*
 	 * because we modify the constraint, we need
 	 * to make a copy. Static constraints come
@@ -2000,6 +1997,7 @@ intel_get_excl_constraints(struct cpu_hw
 	 * been cloned (marked dynamic)
 	 */
 	if (!(c->flags & PERF_X86_EVENT_DYNAMIC)) {
+		struct event_constraint *cx;
 
 		/* sanity check */
 		if (idx < 0)
@@ -2014,13 +2012,14 @@ intel_get_excl_constraints(struct cpu_hw
 		 * initialize dynamic constraint
 		 * with static constraint
 		 */
-		memcpy(cx, c, sizeof(*cx));
+		*cx = *c;
 
 		/*
 		 * mark constraint as dynamic, so we
 		 * can free it later on
 		 */
 		cx->flags |= PERF_X86_EVENT_DYNAMIC;
+		c = cx;
 	}
 
 	/*
@@ -2049,37 +2048,37 @@ intel_get_excl_constraints(struct cpu_hw
 	 * SHARED   : sibling counter measuring non-exclusive event
 	 * UNUSED   : sibling counter unused
 	 */
-	for_each_set_bit(i, cx->idxmsk, X86_PMC_IDX_MAX) {
+	for_each_set_bit(i, c->idxmsk, X86_PMC_IDX_MAX) {
 		/*
 		 * exclusive event in sibling counter
 		 * our corresponding counter cannot be used
 		 * regardless of our event
 		 */
 		if (xlo->state[i] == INTEL_EXCL_EXCLUSIVE)
-			__clear_bit(i, cx->idxmsk);
+			__clear_bit(i, c->idxmsk);
 		/*
 		 * if measuring an exclusive event, sibling
 		 * measuring non-exclusive, then counter cannot
 		 * be used
 		 */
 		if (is_excl && xlo->state[i] == INTEL_EXCL_SHARED)
-			__clear_bit(i, cx->idxmsk);
+			__clear_bit(i, c->idxmsk);
 	}
 
 	/*
 	 * recompute actual bit weight for scheduling algorithm
 	 */
-	cx->weight = hweight64(cx->idxmsk64);
+	c->weight = hweight64(c->idxmsk64);
 
 	/*
 	 * if we return an empty mask, then switch
 	 * back to static empty constraint to avoid
 	 * the cost of freeing later on
 	 */
-	if (cx->weight == 0)
-		cx = &emptyconstraint;
+	if (c->weight == 0)
+		c = &emptyconstraint;
 
-	return cx;
+	return c;
 }
 
 static struct event_constraint *



  parent reply	other threads:[~2015-05-21 11:23 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-05-21 11:17 [PATCH 00/10] Various x86 pmu scheduling patches Peter Zijlstra
2015-05-21 11:17 ` [PATCH 01/10] perf,x86: Fix event/group validation Peter Zijlstra
2015-05-21 12:35   ` Stephane Eranian
2015-05-21 12:56     ` Peter Zijlstra
2015-05-21 13:07       ` Stephane Eranian
2015-05-21 13:09         ` Peter Zijlstra
2015-05-21 13:18           ` Stephane Eranian
2015-05-21 13:20             ` Peter Zijlstra
2015-05-21 13:27               ` Stephane Eranian
2015-05-21 13:29                 ` Peter Zijlstra
2015-05-21 13:36                   ` Stephane Eranian
2015-05-21 14:03                     ` Peter Zijlstra
2015-05-21 15:11                       ` Stephane Eranian
2015-05-22  6:49                         ` Ingo Molnar
2015-05-22  9:26                           ` Stephane Eranian
2015-05-22  9:46                             ` Ingo Molnar
2015-05-21 14:53   ` Peter Zijlstra
2015-05-21 15:42     ` Stephane Eranian
2015-08-21 20:31   ` Sasha Levin
2015-09-10  4:48     ` Sasha Levin
2015-09-10  8:54     ` Stephane Eranian
2015-09-10 10:01       ` Peter Zijlstra
2015-05-21 11:17 ` [PATCH 02/10] perf/x86: Improve HT workaround GP counter constraint Peter Zijlstra
2015-05-22 10:04   ` Stephane Eranian
2015-05-22 11:21     ` Peter Zijlstra
2015-05-22 11:24       ` Stephane Eranian
2015-05-22 11:28       ` Peter Zijlstra
2015-05-22 12:35         ` Stephane Eranian
2015-05-22 12:53           ` Peter Zijlstra
2015-05-22 12:55             ` Stephane Eranian
2015-05-22 12:59               ` Peter Zijlstra
2015-05-22 13:05                 ` Stephane Eranian
2015-05-22 13:07                   ` Stephane Eranian
2015-05-22 13:25                     ` Peter Zijlstra
2015-05-22 13:29                       ` Stephane Eranian
2015-05-22 13:36                         ` Peter Zijlstra
2015-05-22 13:40                           ` Stephane Eranian
2015-05-22 13:48                             ` Peter Zijlstra
2015-05-23  8:26                               ` Ingo Molnar
2015-05-22 13:25                   ` Peter Zijlstra
2015-05-22 13:10                 ` Peter Zijlstra
2015-05-21 11:17 ` [PATCH 03/10] perf/x86: Correct local vs remote sibling state Peter Zijlstra
2015-05-21 13:31   ` Stephane Eranian
2015-05-21 14:10     ` Peter Zijlstra
2015-05-21 11:17 ` [PATCH 04/10] perf/x86: Use lockdep Peter Zijlstra
2015-05-21 11:17 ` Peter Zijlstra [this message]
2015-05-21 11:17 ` [PATCH 06/10] perf/x86: Make WARNs consistent Peter Zijlstra
2015-05-21 11:17 ` [PATCH 07/10] perf/x86: Move intel_commit_scheduling() Peter Zijlstra
2015-05-21 11:17 ` [PATCH 08/10] perf/x86: Remove pointless tests Peter Zijlstra
2015-05-21 13:24   ` Stephane Eranian
2015-05-21 11:17 ` [PATCH 09/10] perf/x86: Remove intel_excl_states::init_state Peter Zijlstra
2015-05-21 13:39   ` Stephane Eranian
2015-05-21 14:12     ` Peter Zijlstra
2015-05-21 11:17 ` [PATCH 10/10] perf,x86: Simplify logic Peter Zijlstra
2015-05-21 11:48 ` [PATCH 00/10] Various x86 pmu scheduling patches Stephane Eranian
2015-05-21 12:53   ` 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=20150521111933.002158797@infradead.org \
    --to=peterz@infradead.org \
    --cc=eranian@google.com \
    --cc=jolsa@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=vincent.weaver@maine.edu \
    /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

all inboxes | Powered by JetHome®