From 3d2606f42984613d324ad3047cf503bcddc3880a Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Fri, 20 May 2011 09:46:54 +0200
Subject: [PATCH 1/3] oprofile, x86: Enable preemption during pci device setup
in IBS init
IBS initialization is a mix of per-core register access and per-node
pci device setup. Register access should be pinned to the cpu, but pci
setup must run with preemption enabled.
This patch better separates the code into non-/preemptible sections
and fixes sleeping with preemption disabled. See bug message below.
Fixes also freeing the eilvt entry by introducing put_eilvt().
BUG: sleeping function called from invalid context at mm/slub.c:824
in_atomic(): 1, irqs_disabled(): 0, pid: 32357, name: modprobe
INFO: lockdep is turned off.
Pid: 32357, comm: modprobe Not tainted 2.6.39-rc7+ #14
Call Trace:
[<ffffffff8104bdc8>] __might_sleep+0x112/0x117
[<ffffffff81129693>] kmem_cache_alloc_trace+0x4b/0xe7
[<ffffffff81278f14>] kzalloc.constprop.0+0x29/0x2b
[<ffffffff81278f4c>] pci_get_subsys+0x36/0x78
[<ffffffff81022689>] ? setup_APIC_eilvt+0xfb/0x139
[<ffffffff81278fa4>] pci_get_device+0x16/0x18
[<ffffffffa06c8b5d>] op_amd_init+0xd3/0x211 [oprofile]
[<ffffffffa064d000>] ? 0xffffffffa064cfff
[<ffffffffa064d298>] op_nmi_init+0x21e/0x26a [oprofile]
[<ffffffffa064d062>] oprofile_arch_init+0xe/0x26 [oprofile]
[<ffffffffa064d010>] oprofile_init+0x10/0x42 [oprofile]
[<ffffffff81002099>] do_one_initcall+0x7f/0x13a
[<ffffffff81096524>] sys_init_module+0x132/0x281
[<ffffffff814cc682>] system_call_fastpath+0x16/0x1b
Reported-by: Dave Jones <davej@redhat.com>
Cc: <stable@kernel.org> [2.6.37.x]
Signed-off-by: Robert Richter <robert.richter@amd.com>
---
arch/x86/oprofile/op_model_amd.c | 97 ++++++++++++++++++--------------
1 file changed, 55 insertions(+), 42 deletions(-)
diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index c3b8e24f2b16..9fd8a567fe1e 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -316,16 +316,23 @@ static void op_amd_stop_ibs(void)
wrmsrl(MSR_AMD64_IBSOPCTL, 0);
}
-static inline int eilvt_is_available(int offset)
+static inline int get_eilvt(int offset)
{
- /* check if we may assign a vector */
return !setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 1);
}
+static inline int put_eilvt(int offset)
+{
+ return !setup_APIC_eilvt(offset, 0, 0, 1);
+}
+
static inline int ibs_eilvt_valid(void)
{
int offset;
u64 val;
+ int valid = 0;
+
+ preempt_disable();
rdmsrl(MSR_AMD64_IBSCTL, val);
offset = val & IBSCTL_LVT_OFFSET_MASK;
@@ -333,16 +340,20 @@ static inline int ibs_eilvt_valid(void)
if (!(val & IBSCTL_LVT_OFFSET_VALID)) {
pr_err(FW_BUG "cpu %d, invalid IBS interrupt offset %d (MSR%08X=0x%016llx)\n",
smp_processor_id(), offset, MSR_AMD64_IBSCTL, val);
- return 0;
+ goto out;
}
- if (!eilvt_is_available(offset)) {
+ if (!get_eilvt(offset)) {
pr_err(FW_BUG "cpu %d, IBS interrupt offset %d not available (MSR%08X=0x%016llx)\n",
smp_processor_id(), offset, MSR_AMD64_IBSCTL, val);
- return 0;
+ goto out;
}
- return 1;
+ valid = 1;
+out:
+ preempt_enable();
+
+ return valid;
}
static inline int get_ibs_offset(void)
@@ -600,67 +611,69 @@ static int setup_ibs_ctl(int ibs_eilvt_off)
static int force_ibs_eilvt_setup(void)
{
- int i;
+ int offset;
int ret;
- /* find the next free available EILVT entry */
- for (i = 1; i < 4; i++) {
- if (!eilvt_is_available(i))
- continue;
- ret = setup_ibs_ctl(i);
- if (ret)
- return ret;
- pr_err(FW_BUG "using offset %d for IBS interrupts\n", i);
- return 0;
+ /*
+ * find the next free available EILVT entry, skip offset 0,
+ * pin search to this cpu
+ */
+ preempt_disable();
+ for (offset = 1; offset < APIC_EILVT_NR_MAX; offset++) {
+ if (get_eilvt(offset))
+ break;
+ }
+ preempt_enable();
+
+ if (offset == APIC_EILVT_NR_MAX) {
+ printk(KERN_DEBUG "No EILVT entry available\n");
+ return -EBUSY;
}
- printk(KERN_DEBUG "No EILVT entry available\n");
-
- return -EBUSY;
-}
-
-static int __init_ibs_nmi(void)
-{
- int ret;
-
- if (ibs_eilvt_valid())
- return 0;
-
- ret = force_ibs_eilvt_setup();
+ ret = setup_ibs_ctl(offset);
if (ret)
- return ret;
+ goto out;
- if (!ibs_eilvt_valid())
- return -EFAULT;
+ if (!ibs_eilvt_valid()) {
+ ret = -EFAULT;
+ goto out;
+ }
+ pr_err(FW_BUG "using offset %d for IBS interrupts\n", offset);
pr_err(FW_BUG "workaround enabled for IBS LVT offset\n");
return 0;
+out:
+ preempt_disable();
+ put_eilvt(offset);
+ preempt_enable();
+ return ret;
}
/*
* check and reserve APIC extended interrupt LVT offset for IBS if
* available
- *
- * init_ibs() preforms implicitly cpu-local operations, so pin this
- * thread to its current CPU
*/
static void init_ibs(void)
{
- preempt_disable();
-
ibs_caps = get_ibs_caps();
+
if (!ibs_caps)
+ return;
+
+ if (ibs_eilvt_valid())
goto out;
- if (__init_ibs_nmi() < 0)
- ibs_caps = 0;
- else
- printk(KERN_INFO "oprofile: AMD IBS detected (0x%08x)\n", ibs_caps);
+ if (!force_ibs_eilvt_setup())
+ goto out;
+
+ /* Failed to setup ibs */
+ ibs_caps = 0;
+ return;
out:
- preempt_enable();
+ printk(KERN_INFO "oprofile: AMD IBS detected (0x%08x)\n", ibs_caps);
}
static int (*create_arch_files)(struct super_block *sb, struct dentry *root);
From d819437156fd99da61d4e1402b2dbfc5cc472265 Mon Sep 17 00:00:00 2001
From: Eric B Munson <emunson@mgebm.net>
Date: Mon, 23 May 2011 10:22:40 -0400
Subject: [PATCH 2/3] oprofile, powerpc: Handle events that raise an exception
without overflowing
Commit 0837e3242c73566fc1c0196b4ec61779c25ffc93 fixes a situation on POWER7
where events can roll back if a specualtive event doesn't actually complete.
This can raise a performance monitor exception. We need to catch this to ensure
that we reset the PMC. In all cases the PMC will be less than 256 cycles from
overflow.
This patch lifts Anton's fix for the problem in perf and applies it to oprofile
as well.
Signed-off-by: Eric B Munson <emunson@mgebm.net>
Cc: <stable@kernel.org> # as far back as it applies cleanly
Tested-by: Maynard Johnson <maynardj@us.ibm.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
---
arch/powerpc/oprofile/op_model_power4.c | 24 +++++++++++++++++++++++-
1 file changed, 23 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/oprofile/op_model_power4.c b/arch/powerpc/oprofile/op_model_power4.c
index 8ee51a252cf1..e6bec74be131 100644
--- a/arch/powerpc/oprofile/op_model_power4.c
+++ b/arch/powerpc/oprofile/op_model_power4.c
@@ -261,6 +261,28 @@ static int get_kernel(unsigned long pc, unsigned long mmcra)
return is_kernel;
}
+static bool pmc_overflow(unsigned long val)
+{
+ if ((int)val < 0)
+ return true;
+
+ /*
+ * Events on POWER7 can roll back if a speculative event doesn't
+ * eventually complete. Unfortunately in some rare cases they will
+ * raise a performance monitor exception. We need to catch this to
+ * ensure we reset the PMC. In all cases the PMC will be 256 or less
+ * cycles from overflow.
+ *
+ * We only do this if the first pass fails to find any overflowing
+ * PMCs because a user might set a period of less than 256 and we
+ * don't want to mistakenly reset them.
+ */
+ if (__is_processor(PV_POWER7) && ((0x80000000 - val) <= 256))
+ return true;
+
+ return false;
+}
+
static void power4_handle_interrupt(struct pt_regs *regs,
struct op_counter_config *ctr)
{
@@ -281,7 +303,7 @@ static void power4_handle_interrupt(struct pt_regs *regs,
for (i = 0; i < cur_cpu_spec->num_pmcs; ++i) {
val = classic_ctr_read(i);
- if (val < 0) {
+ if (pmc_overflow(val)) {
if (oprofile_running && ctr[i].enabled) {
oprofile_add_ext_sample(pc, regs, i, is_kernel);
classic_ctr_write(i, reset_value[i]);
From b76a06e08d94b2a63e47837dfe46bbbf0a3af6c2 Mon Sep 17 00:00:00 2001
From: Anton Blanchard <anton@samba.org>
Date: Sun, 8 May 2011 19:32:36 -0400
Subject: [PATCH 3/3] oprofile: Use linux/mutex.h
The oprofile code is still including asm/mutex.h instead of
linux/mutex.h.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Robert Richter <robert.richter@amd.com>
---
drivers/oprofile/event_buffer.h | 2 +-
drivers/oprofile/oprof.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/oprofile/event_buffer.h b/drivers/oprofile/event_buffer.h
index 4e70749f8d16..a8d5bb3cba89 100644
--- a/drivers/oprofile/event_buffer.h
+++ b/drivers/oprofile/event_buffer.h
@@ -11,7 +11,7 @@
#define EVENT_BUFFER_H
#include <linux/types.h>
-#include <asm/mutex.h>
+#include <linux/mutex.h>
int alloc_event_buffer(void);
diff --git a/drivers/oprofile/oprof.c b/drivers/oprofile/oprof.c
index f9bda64fcd1b..dccd8636095c 100644
--- a/drivers/oprofile/oprof.c
+++ b/drivers/oprofile/oprof.c
@@ -14,7 +14,7 @@
#include <linux/moduleparam.h>
#include <linux/workqueue.h>
#include <linux/time.h>
-#include <asm/mutex.h>
+#include <linux/mutex.h>
#include "oprof.h"
#include "event_buffer.h"