From: Jiri Olsa <jolsa@redhat.com>
To: linux-kernel@vger.kernel.org
Cc: Frederic Weisbecker <fweisbec@gmail.com>,
Ingo Molnar <mingo@elte.hu>, Paul Mackerras <paulus@samba.org>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Arnaldo Carvalho de Melo <acme@redhat.com>,
Jiri Olsa <jolsa@redhat.com>
Subject: [PATCH 2/2] perf: Simplify the sample's user regs/stack retrieval
Date: Thu, 4 Oct 2012 12:38:45 +0200 [thread overview]
Message-ID: <1349347125-5770-3-git-send-email-jolsa@redhat.com> (raw)
In-Reply-To: <1349347125-5770-1-git-send-email-jolsa@redhat.com>
The perf_sample_regs_user function now returns result of the user
register retrieval, which simplifies surrounding code a bit.
The perf_output_sample_ustack now combines initial dump size
output for both (stack dump un/available) cases.
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
---
kernel/events/core.c | 64 ++++++++++++++++++++++++-------------------------
1 files changed, 31 insertions(+), 33 deletions(-)
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 71329d6..e392dcb 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -3783,11 +3783,15 @@ perf_output_sample_regs(struct perf_output_handle *handle,
__weak void
arch_sample_regs_user_fixup(struct perf_regs_user *regs_user, int kernel) { }
-static void perf_sample_regs_user(struct perf_regs_user *regs_user,
+static int perf_sample_regs_user(struct perf_regs_user *regs_user,
struct pt_regs *regs)
{
int kernel = !user_mode(regs);
+ /* Already defined. */
+ if (regs_user->abi)
+ return 1;
+
if (kernel) {
if (current->mm)
regs = task_pt_regs(current);
@@ -3800,6 +3804,8 @@ static void perf_sample_regs_user(struct perf_regs_user *regs_user,
regs_user->abi = perf_reg_abi(current);
arch_sample_regs_user_fixup(regs_user, kernel);
}
+
+ return regs ? 1 : 0;
}
/*
@@ -3825,10 +3831,6 @@ perf_sample_ustack_size(u16 stack_size, u16 header_size,
{
u64 task_size;
- /* No regs, no stack pointer, no dump. */
- if (!regs)
- return 0;
-
/*
* Check if we fit in with the requested stack size into the:
* - TASK_SIZE
@@ -3862,29 +3864,26 @@ static void
perf_output_sample_ustack(struct perf_output_handle *handle, u64 dump_size,
struct pt_regs *regs)
{
- /* Case of a kernel thread, nothing to dump */
- if (!regs) {
- u64 size = 0;
- perf_output_put(handle, size);
- } else {
+ /*
+ * We dump:
+ * static size
+ * - the size requested by user or the best one we can fit
+ * in to the sample max size
+ * - zero (and final data) if there's nothing to dump
+ * data
+ * - user stack dump data
+ * dynamic size
+ * - the actual dumped size
+ */
+
+ /* Static size. */
+ perf_output_put(handle, dump_size);
+
+ if (dump_size) {
unsigned long sp;
unsigned int rem;
u64 dyn_size;
- /*
- * We dump:
- * static size
- * - the size requested by user or the best one we can fit
- * in to the sample max size
- * data
- * - user stack dump data
- * dynamic size
- * - the actual dumped size
- */
-
- /* Static size. */
- perf_output_put(handle, dump_size);
-
/* Data. */
sp = perf_user_stack_pointer(regs);
rem = __output_copy_user(handle, (void *) sp, dump_size);
@@ -4235,9 +4234,7 @@ void perf_prepare_sample(struct perf_event_header *header,
/* regs dump ABI info */
int size = sizeof(u64);
- perf_sample_regs_user(&data->regs_user, regs);
-
- if (data->regs_user.regs) {
+ if (perf_sample_regs_user(&data->regs_user, regs)) {
u64 mask = event->attr.sample_regs_user;
size += hweight64(mask) * sizeof(u64);
}
@@ -4253,14 +4250,15 @@ void perf_prepare_sample(struct perf_event_header *header,
* up the rest of the sample size.
*/
struct perf_regs_user *uregs = &data->regs_user;
- u16 stack_size = event->attr.sample_stack_user;
+ u64 sample_stack_user = event->attr.sample_stack_user;
+ u16 stack_size = 0;
u16 size = sizeof(u64);
- if (!uregs->abi)
- perf_sample_regs_user(uregs, regs);
-
- stack_size = perf_sample_ustack_size(stack_size, header->size,
- uregs->regs);
+ if (perf_sample_regs_user(uregs, regs)) {
+ stack_size = perf_sample_ustack_size(sample_stack_user,
+ header->size,
+ uregs->regs);
+ }
/*
* If there is something to dump, add space for the dump
--
1.7.7.6
prev parent reply other threads:[~2012-10-04 10:39 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-10-01 17:31 [PATCH] perf x86_64: Fix rsp register for system call fast path Jiri Olsa
2012-10-02 10:44 ` Peter Zijlstra
2012-10-02 14:58 ` [PATCHv2] " Jiri Olsa
2012-10-02 15:49 ` Frederic Weisbecker
2012-10-02 16:06 ` Jiri Olsa
2012-10-02 16:16 ` Frederic Weisbecker
2012-10-03 12:29 ` [PATCHv3] " Jiri Olsa
2012-10-03 12:35 ` Frederic Weisbecker
2012-10-03 13:13 ` [PATCHv4] " Jiri Olsa
2012-10-03 13:22 ` Peter Zijlstra
2012-10-03 13:30 ` Jiri Olsa
2012-10-04 10:38 ` [PATH 0/2] perf: x86_64 rsp related changes Jiri Olsa
2012-10-04 10:38 ` [PATCH 1/2] perf x86_64: Fix rsp register for system call fast path Jiri Olsa
2012-10-04 10:38 ` Jiri Olsa [this message]
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=1349347125-5770-3-git-send-email-jolsa@redhat.com \
--to=jolsa@redhat.com \
--cc=a.p.zijlstra@chello.nl \
--cc=acme@redhat.com \
--cc=fweisbec@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=paulus@samba.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
all inboxes | Powered by JetHome®