From: markus.t.metzger@intel.com
To: mingo@elte.hu, tglx@linutronix.de, hpa@zytor.com
Cc: markus.t.metzger@gmail.com, roland@redhat.com,
eranian@googlemail.com, oleg@redhat.com, juan.villacis@intel.com,
ak@linux.jf.intel.com, linux-kernel@vger.kernel.org,
Markus Metzger <markus.t.metzger@intel.com>
Subject: [patch 04/18] x86, bts: wait until traced task has been scheduled out
Date: Thu, 02 Apr 2009 16:54:59 +0200 [thread overview]
Message-ID: <20090402145709.422226000@intel.com> (raw)
In-Reply-To: 20090402145455.597376000@intel.com
[-- Attachment #1: tip.master.ds.wait_task_inactive.patch --]
[-- Type: text/plain, Size: 2412 bytes --]
In order to stop branch tracing for a running task, we need to first
clear the branch tracing control bits before we may free the tracing
buffer.
If the traced task is running, the cpu might still trace that task
after the branch trace control bits have cleared.
Wait until the traced task has been scheduled out before proceeding.
A similar problem affects the task debug store context. We first remove
the context, then we need to wait until the task has been scheduled
out before we can free the context memory.
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
---
arch/x86/kernel/ds.c | 40 40 + 0 - 0 !
1 file changed, 40 insertions(+)
Index: b/arch/x86/kernel/ds.c
===================================================================
--- a/arch/x86/kernel/ds.c
+++ b/arch/x86/kernel/ds.c
@@ -250,6 +250,40 @@ static DEFINE_PER_CPU(struct ds_context
#define system_context per_cpu(system_context_array, smp_processor_id())
+/*
+ * Wait for the traced task to unschedule.
+ *
+ * This guarantees that the bts trace configuration has been
+ * synchronized with the cpu executing the task.
+ */
+static void wait_to_unschedule(struct task_struct *task)
+{
+ unsigned long nvcsw;
+ unsigned long nivcsw;
+
+ if (!task)
+ return;
+
+ if (task == current)
+ return;
+
+ nvcsw = task->nvcsw;
+ nivcsw = task->nivcsw;
+ for (;;) {
+ if (!task_is_running(task))
+ break;
+ /*
+ * The switch count is incremented before the actual
+ * context switch. We thus wait for two switches to be
+ * sure at least one completed.
+ */
+ if ((task->nvcsw - nvcsw) > 1)
+ break;
+ if ((task->nivcsw - nivcsw) > 1)
+ break;
+ }
+}
+
static inline struct ds_context *ds_get_context(struct task_struct *task)
{
struct ds_context **p_context =
@@ -321,6 +355,9 @@ static inline void ds_put_context(struct
spin_unlock_irqrestore(&ds_lock, irq);
+ /* The context might still be in use for context switching. */
+ wait_to_unschedule(context->task);
+
kfree(context);
}
@@ -789,6 +826,9 @@ void ds_release_bts(struct bts_tracer *t
WARN_ON_ONCE(tracer->ds.context->bts_master != tracer);
tracer->ds.context->bts_master = NULL;
+ /* Make sure tracing stopped and the tracer is not in use. */
+ wait_to_unschedule(tracer->ds.context->task);
+
put_tracer(tracer->ds.context->task);
ds_put_context(tracer->ds.context);
--
[-- Attachment #2: Type: text/plain, Size: 656 bytes --]
---------------------------------------------------------------------
Intel GmbH
Dornacher Strasse 1
85622 Feldkirchen/Muenchen Germany
Sitz der Gesellschaft: Feldkirchen bei Muenchen
Geschaeftsfuehrer: Douglas Lusk, Peter Gleissner, Hannes Schwaderer
Registergericht: Muenchen HRB 47456 Ust.-IdNr.
VAT Registration No.: DE129385895
Citibank Frankfurt (BLZ 502 109 00) 600119052
This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.
next prev parent reply other threads:[~2009-04-02 14:58 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-04-02 14:54 [patch 00/18] x86, bts, ptrace, hw-branch-tracer: fixes and cleanups markus.t.metzger
2009-04-02 14:54 ` [patch 01/18] x86, bts: fix race when bts tracer is removed markus.t.metzger
2009-04-02 18:45 ` Ingo Molnar
2009-04-03 6:19 ` Metzger, Markus T
2009-04-03 7:17 ` Metzger, Markus T
2009-04-03 11:30 ` Ingo Molnar
2009-04-03 11:29 ` Ingo Molnar
2009-04-02 14:54 ` [patch 02/18] sched: add task_is_running() fucntion to sched.h markus.t.metzger
2009-04-02 14:54 ` [patch 03/18] x86, ptrace, bts: defer branch trace stopping markus.t.metzger
2009-04-02 18:40 ` Ingo Molnar
2009-04-02 14:54 ` markus.t.metzger [this message]
2009-04-02 19:17 ` [patch 04/18] x86, bts: wait until traced task has been scheduled out Ingo Molnar
2009-04-03 6:22 ` Metzger, Markus T
2009-04-03 11:32 ` Ingo Molnar
2009-04-03 11:36 ` Metzger, Markus T
2009-04-02 14:55 ` [patch 05/18] x86, bts: fix race between per-task and per-cpu branch tracing markus.t.metzger
2009-04-02 14:55 ` [patch 06/18] x86, debugctlmsr: add _on_cpu variants to debugctlmsr functions markus.t.metzger
2009-04-02 14:55 ` [patch 07/18] x86, bts, hw-branch-tracer: add _noirq variants to the debug store interface markus.t.metzger
2009-04-02 14:55 ` [patch 08/18] x86, hw-branch-tracer: allocate selftest iterator on heap markus.t.metzger
2009-04-02 14:55 ` [patch 09/18] x86, ds: fix compiler warning markus.t.metzger
2009-04-02 14:55 ` [patch 10/18] x86, ds: fix bounds check in ds selftest markus.t.metzger
2009-04-02 14:55 ` [patch 11/18] x86, ds: selftest each cpu markus.t.metzger
2009-04-02 14:55 ` [patch 12/18] x86, ds: add task tracing selftest markus.t.metzger
2009-04-02 14:55 ` [patch 13/18] x86, ds: add leakage warning markus.t.metzger
2009-04-02 19:27 ` Ingo Molnar
2009-04-03 6:42 ` Metzger, Markus T
2009-04-02 14:55 ` [patch 14/18] x86, ds: use single debug store cpu configuration markus.t.metzger
2009-04-02 19:29 ` Ingo Molnar
2009-04-03 6:46 ` Metzger, Markus T
2009-04-02 14:55 ` [patch 15/18] x86, ptrace: remove duplicate functionality markus.t.metzger
2009-04-02 14:55 ` [patch 16/18] x86, ds: dont use TIF_DEBUGCTLMSR markus.t.metzger
2009-04-02 14:55 ` [patch 17/18] x86, ds: fix bad ds_reset_pebs() markus.t.metzger
2009-04-02 14:55 ` [patch 18/18] x86, ds: support Core i7 markus.t.metzger
2009-04-02 19:22 ` [patch 00/18] x86, bts, ptrace, hw-branch-tracer: fixes and cleanups Ingo Molnar
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=20090402145709.422226000@intel.com \
--to=markus.t.metzger@intel.com \
--cc=ak@linux.jf.intel.com \
--cc=eranian@googlemail.com \
--cc=hpa@zytor.com \
--cc=juan.villacis@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=markus.t.metzger@gmail.com \
--cc=mingo@elte.hu \
--cc=oleg@redhat.com \
--cc=roland@redhat.com \
--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