LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Pekka Paalanen <pq@iki.fi>
To: Ingo Molnar <mingo@elte.hu>
Cc: Pekka Paalanen <pq@iki.fi>, linux-kernel@vger.kernel.org
Subject: [PATCH 2/4] x86 mmiotrace: fix relay-buffer-full flag for SMP
Date: Tue, 5 Feb 2008 22:30:46 +0200	[thread overview]
Message-ID: <20080205223046.668309be@daedalus.pq.iki.fi> (raw)
In-Reply-To: <20080205222807.7b35ef2b@daedalus.pq.iki.fi>

Relay has per-cpu buffers, but mmiotrace was using only a single flag
for detecting buffer full/not-full transitions. The new code makes
this per-cpu and actually counts missed events.

Signed-off-by: Pekka Paalanen <pq@iki.fi>
---
 arch/x86/kernel/mmiotrace/mmio-mod.c |   26 ++++++++++++++++----------
 1 files changed, 16 insertions(+), 10 deletions(-)

diff --git a/arch/x86/kernel/mmiotrace/mmio-mod.c b/arch/x86/kernel/mmiotrace/mmio-mod.c
index 82ae920..f492b65 100644
--- a/arch/x86/kernel/mmiotrace/mmio-mod.c
+++ b/arch/x86/kernel/mmiotrace/mmio-mod.c
@@ -29,6 +29,7 @@
 #include <asm/pgtable.h>
 #include <linux/mmiotrace.h>
 #include <asm/e820.h> /* for ISA_START_ADDRESS */
+#include <asm/atomic.h>
 
 #include "kmmio.h"
 #include "pf_in.h"
@@ -47,9 +48,13 @@ struct trap_reason {
 	int active_traces;
 };
 
+/* Accessed per-cpu. */
 static struct trap_reason pf_reason[NR_CPUS];
 static struct mm_io_header_rw cpu_trace[NR_CPUS];
 
+/* Access to this is not per-cpu. */
+static atomic_t dropped[NR_CPUS];
+
 static struct file_operations mmio_fops = {
 	.owner = THIS_MODULE,
 };
@@ -57,7 +62,6 @@ static struct file_operations mmio_fops = {
 static const size_t subbuf_size = 256*1024;
 static struct rchan *chan;
 static struct dentry *dir;
-static int suspended;      /* XXX should this be per cpu? */
 static struct proc_dir_entry *proc_marker_file;
 
 /* module parameters */
@@ -269,19 +273,21 @@ static void post(struct kmmio_probe *p, unsigned long condition,
 static int subbuf_start_handler(struct rchan_buf *buf, void *subbuf,
 					void *prev_subbuf, size_t prev_padding)
 {
+	unsigned int cpu = buf->cpu;
+	atomic_t *drop = &dropped[cpu];
+	int count;
 	if (relay_buf_full(buf)) {
-		if (!suspended) {
-			suspended = 1;
-			printk(KERN_ERR MODULE_NAME
-						": cpu %d buffer full!!!\n",
-						smp_processor_id());
+		if (atomic_inc_return(drop) == 1) {
+			printk(KERN_ERR MODULE_NAME ": cpu %d buffer full!\n",
+									cpu);
 		}
 		return 0;
-	} else if (suspended) {
-		suspended = 0;
+	} else if ((count = atomic_read(drop))) {
 		printk(KERN_ERR MODULE_NAME
-					": cpu %d buffer no longer full.\n",
-					smp_processor_id());
+					": cpu %d buffer no longer full, "
+					"missed %d events.\n",
+					cpu, count);
+		atomic_sub(count, drop);
 	}
 
 	return 1;
-- 
1.5.3.7


  reply	other threads:[~2008-02-05 20:30 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-01-27 16:52 [PATCH] x86: Add a list for custom page fault handlers Pekka Paalanen
2008-01-27 17:55 ` [RFC PATCH] x86: mmiotrace - trace memory mapped IO Pekka Paalanen
2008-01-30 22:39   ` Pekka Paalanen
2008-01-27 19:29 ` [PATCH] x86: Add a list for custom page fault handlers Ingo Molnar
2008-01-27 21:03 ` Peter Zijlstra
2008-01-30  2:28 ` Harvey Harrison
2008-01-30  2:34   ` Harvey Harrison
2008-01-30 18:08     ` Pekka Paalanen
2008-01-31 15:07       ` Ingo Molnar
2008-01-31 16:02         ` [PATCH v2] " Pekka Paalanen
2008-01-31 16:15           ` Arjan van de Ven
2008-02-03  6:55             ` Pekka Paalanen
2008-02-03  7:03               ` Ingo Molnar
2008-02-03 21:40                 ` Pekka Paalanen
2008-02-05 20:28                 ` [PATCH 1/4] x86 mmiotrace: use lookup_address() Pekka Paalanen
2008-02-05 20:30                   ` Pekka Paalanen [this message]
2008-02-05 20:44                     ` [PATCH 2/4] x86 mmiotrace: fix relay-buffer-full flag for SMP Eric Dumazet
2008-02-05 21:14                       ` Pekka Paalanen
2008-02-05 21:35                         ` Eric Dumazet
2008-02-09 17:53                           ` [PATCH] x86 mmiotrace: Use percpu instead of arrays Pekka Paalanen
2008-02-05 20:31                   ` [PATCH 3/4] x86 mmiotrace: comment about user space ABI Pekka Paalanen
2008-02-05 20:39                   ` [PATCH 4/4] x86 mmiotrace: move files into arch/x86/mm/ Pekka Paalanen
2008-02-06  3:02                     ` Randy Dunlap
2008-02-09 11:21                       ` Pekka Paalanen
2008-02-07 12:53                     ` Ingo Molnar
2008-02-07 12:56                       ` Christoph Hellwig
2008-02-09 17:52                         ` [RFC PATCH] x86: explicit call to mmiotrace in do_page_fault() Pekka Paalanen
2008-02-09 18:01                           ` Arjan van de Ven
2008-02-09 18:23                             ` Pekka Paalanen
2008-02-09 18:56                               ` Pekka Enberg
2008-02-09 19:11                                 ` Pekka Paalanen
2008-02-09 19:19                                   ` Pekka Enberg
2008-02-09 18:39                             ` Peter Zijlstra
2008-02-09 18:39                           ` Peter Zijlstra
2008-02-10 18:05                             ` [RFC PATCH v2] " Pekka Paalanen
2008-02-11  2:12                               ` Pavel Roskin
2008-02-11 18:04                                 ` Pekka Paalanen
2008-02-06  5:00                   ` [PATCH 1/4] x86 mmiotrace: use lookup_address() Christoph Hellwig
2008-02-07 12:52                     ` Ingo Molnar
2008-01-31 16:16           ` [RFC PATCH v2] x86: mmiotrace - trace memory mapped IO Pekka Paalanen
2008-01-31 16:29             ` Arjan van de Ven
2008-02-03  7:21               ` Pekka Paalanen
2008-01-30 18:20 ` [PATCH] x86: Add a list for custom page fault handlers Arjan van de Ven

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=20080205223046.668309be@daedalus.pq.iki.fi \
    --to=pq@iki.fi \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --subject='Re: [PATCH 2/4] x86 mmiotrace: fix relay-buffer-full flag for SMP' \
    /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

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).