* intel-iommu sysfs oops.
@ 2017-02-23 19:44 Dave Jones
2017-02-23 23:30 ` Joerg Roedel
0 siblings, 1 reply; 5+ messages in thread
From: Dave Jones @ 2017-02-23 19:44 UTC (permalink / raw)
To: Linux Kernel; +Cc: jroedel, dwmw2, iommu
cat /sys/devices/virtual/iommu/dmar0/intel-iommu/version
Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC
CPU: 2 PID: 1488 Comm: cat Not tainted 4.10.0-think+ #5
task: ffff8804ee440040 task.stack: ffffc90000d48000
RIP: 0010:intel_iommu_show_version+0x13/0x40
RSP: 0018:ffffc90000d4bcf0 EFLAGS: 00010286
RAX: ffff8804ede80008 RBX: ffffffff81ec2c80 RCX: 0000000000000000
RDX: 0000000000000000 RSI: ffffffff81ec2c80 RDI: ffff88017fc2d5d0
RBP: ffffc90000d4bcf0 R08: 0000000000000000 R09: ffff8804ede80008
R10: 00000000068b13bb R11: 0000000000000006 R12: 0000000000001000
R13: ffffffff81a96f10 R14: ffff8804fdd131b8 R15: ffff8804ede80008
FS: 00007feb8ad5e700(0000) GS:ffff880507c00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000000 CR3: 00000004ee49a000 CR4: 00000000001406e0
Call Trace:
dev_attr_show+0x20/0x50
? sysfs_file_ops+0x46/0x70
sysfs_kf_seq_show+0xb7/0x110
kernfs_seq_show+0x26/0x30
seq_read+0x129/0x4a0
kernfs_fop_read+0x13b/0x1a0
__vfs_read+0x37/0x140
? __context_tracking_exit.part.5+0x82/0x150
vfs_read+0xab/0x180
SyS_read+0x58/0xc0
do_syscall_64+0x61/0x170
entry_SYSCALL64_slow_path+0x25/0x25
RIP: 0033:0x7feb8a8875a0
RSP: 002b:00007fff8dece608 EFLAGS: 00000246
ORIG_RAX: 0000000000000000
RAX: ffffffffffffffda RBX: 0000000000020000 RCX: 00007feb8a8875a0
RDX: 0000000000020000 RSI: 00007feb8aba2000 RDI: 0000000000000003
RBP: 00007feb8aba2000 R08: ffffffffffffffff R09: 0000000000000000
R10: 000000000000037b R11: 0000000000000246 R12: 00007feb8aba2000
R13: 0000000000000003 R14: 0000000000020000 R15: 0000000000001000
$ dmesg | grep -i dmar
[ 0.000000] ACPI: DMAR 0x00000000D479B0B8 0000B8 (v01 LENOVO TC-FB 00001B30 INTL 00000001)
[ 0.190199] DMAR: Host address width 39
[ 0.190208] DMAR: DRHD base: 0x000000fed90000 flags: 0x0
[ 0.190235] DMAR: dmar0: reg_base_addr fed90000 ver 1:0 cap c0000020660462 ecap f0101a
[ 0.190250] DMAR: DRHD base: 0x000000fed91000 flags: 0x1
[ 0.190272] DMAR: dmar1: reg_base_addr fed91000 ver 1:0 cap d2008020660462 ecap f010da
[ 0.190286] DMAR: RMRR base: 0x000000d5e6c000 end: 0x000000d5e92fff
[ 0.190302] DMAR: RMRR base: 0x000000d7000000 end: 0x000000df1fffff
[ 0.190318] DMAR-IR: IOAPIC id 8 under DRHD base 0xfed91000 IOMMU 1
[ 0.190330] DMAR-IR: HPET id 0 under DRHD base 0xfed91000
[ 0.190341] DMAR-IR: Queued invalidation will be enabled to support x2apic and Intr-remapping.
[ 0.190634] DMAR-IR: Enabled IRQ remapping in x2apic mode
[ 1.050338] DMAR: No ATSR found
[ 1.050717] DMAR: dmar0: Using Queued invalidation
[ 1.050739] DMAR: dmar1: Using Queued invalidation
[ 1.050799] DMAR: Setting RMRR:
[ 1.050938] DMAR: Setting identity map for device 0000:00:02.0 [0xd7000000 - 0xdf1fffff]
[ 1.051763] DMAR: Setting identity map for device 0000:00:14.0 [0xd5e6c000 - 0xd5e92fff]
[ 1.051911] DMAR: Setting identity map for device 0000:00:1a.0 [0xd5e6c000 - 0xd5e92fff]
[ 1.052054] DMAR: Setting identity map for device 0000:00:1d.0 [0xd5e6c000 - 0xd5e92fff]
[ 1.052083] DMAR: Prepare 0-16MiB unity mapping for LPC
[ 1.052200] DMAR: Setting identity map for device 0000:00:1f.0 [0x0 - 0xffffff]
[ 1.052331] DMAR: Intel(R) Virtualization Technology for Directed I/O
[ 1.220475] [drm] DMAR active, disabling use of stolen memory
[ 1.653051] calling dmar_free_unused_resources+0x0/0xcf @ 1
[ 1.653054] initcall dmar_free_unused_resources+0x0/0xcf returned 0 after 1 usecs
$ dmesg | grep -i iommu
[ 0.190318] DMAR-IR: IOAPIC id 8 under DRHD base 0xfed91000 IOMMU 1
[ 0.568456] calling iommu_init+0x0/0x2b @ 1
[ 0.568470] initcall iommu_init+0x0/0x2b returned 0 after 0 usecs
[ 0.570137] calling iommu_dev_init+0x0/0x19 @ 1
[ 0.570158] initcall iommu_dev_init+0x0/0x19 returned 0 after 0 usecs
[ 1.050256] calling pci_iommu_init+0x0/0x3c @ 1
[ 1.052700] iommu: Adding device 0000:00:00.0 to group 0
[ 1.052767] iommu: Adding device 0000:00:02.0 to group 1
[ 1.052833] iommu: Adding device 0000:00:03.0 to group 2
[ 1.052895] iommu: Adding device 0000:00:14.0 to group 3
[ 1.052965] iommu: Adding device 0000:00:16.0 to group 4
[ 1.053026] iommu: Adding device 0000:00:19.0 to group 5
[ 1.053093] iommu: Adding device 0000:00:1a.0 to group 6
[ 1.053157] iommu: Adding device 0000:00:1b.0 to group 7
[ 1.053220] iommu: Adding device 0000:00:1c.0 to group 8
[ 1.053286] iommu: Adding device 0000:00:1c.3 to group 9
[ 1.053348] iommu: Adding device 0000:00:1d.0 to group 10
[ 1.053443] iommu: Adding device 0000:00:1f.0 to group 11
[ 1.053487] iommu: Adding device 0000:00:1f.2 to group 11
[ 1.053529] iommu: Adding device 0000:00:1f.3 to group 11
[ 1.053599] iommu: Adding device 0000:02:00.0 to group 12
[ 1.065133] initcall pci_iommu_init+0x0/0x3c returned 0 after 14515 usecs
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: intel-iommu sysfs oops.
@ 2017-02-23 23:30 ` Joerg Roedel
0 siblings, 0 replies; 5+ messages in thread
From: Joerg Roedel @ 2017-02-23 23:30 UTC (permalink / raw)
To: Dave Jones, Linux Kernel, dwmw2, iommu
Hi Dave,
On Thu, Feb 23, 2017 at 02:44:06PM -0500, Dave Jones wrote:
> cat /sys/devices/virtual/iommu/dmar0/intel-iommu/version
>
> Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC
Thanks for the report, the problem reproduces easily here. The diff
below fixes the issue on Intel, AMD has the same problem and needs a
similar fix I'll do shortly.
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index f5e02f8..ec3f6c8 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -4730,11 +4730,16 @@ static int intel_iommu_cpu_dead(unsigned int cpu)
return 0;
}
+static struct intel_iommu *dev_to_intel_iommu(struct device *dev)
+{
+ return container_of(dev, struct intel_iommu, iommu.dev);
+}
+
static ssize_t intel_iommu_show_version(struct device *dev,
struct device_attribute *attr,
char *buf)
{
- struct intel_iommu *iommu = dev_get_drvdata(dev);
+ struct intel_iommu *iommu = dev_to_intel_iommu(dev);
u32 ver = readl(iommu->reg + DMAR_VER_REG);
return sprintf(buf, "%d:%d\n",
DMAR_VER_MAJOR(ver), DMAR_VER_MINOR(ver));
@@ -4745,7 +4750,7 @@ static ssize_t intel_iommu_show_address(struct device *dev,
struct device_attribute *attr,
char *buf)
{
- struct intel_iommu *iommu = dev_get_drvdata(dev);
+ struct intel_iommu *iommu = dev_to_intel_iommu(dev);
return sprintf(buf, "%llx\n", iommu->reg_phys);
}
static DEVICE_ATTR(address, S_IRUGO, intel_iommu_show_address, NULL);
@@ -4754,7 +4759,7 @@ static ssize_t intel_iommu_show_cap(struct device *dev,
struct device_attribute *attr,
char *buf)
{
- struct intel_iommu *iommu = dev_get_drvdata(dev);
+ struct intel_iommu *iommu = dev_to_intel_iommu(dev);
return sprintf(buf, "%llx\n", iommu->cap);
}
static DEVICE_ATTR(cap, S_IRUGO, intel_iommu_show_cap, NULL);
@@ -4763,7 +4768,7 @@ static ssize_t intel_iommu_show_ecap(struct device *dev,
struct device_attribute *attr,
char *buf)
{
- struct intel_iommu *iommu = dev_get_drvdata(dev);
+ struct intel_iommu *iommu = dev_to_intel_iommu(dev);
return sprintf(buf, "%llx\n", iommu->ecap);
}
static DEVICE_ATTR(ecap, S_IRUGO, intel_iommu_show_ecap, NULL);
@@ -4772,7 +4777,7 @@ static ssize_t intel_iommu_show_ndoms(struct device *dev,
struct device_attribute *attr,
char *buf)
{
- struct intel_iommu *iommu = dev_get_drvdata(dev);
+ struct intel_iommu *iommu = dev_to_intel_iommu(dev);
return sprintf(buf, "%ld\n", cap_ndoms(iommu->cap));
}
static DEVICE_ATTR(domains_supported, S_IRUGO, intel_iommu_show_ndoms, NULL);
@@ -4781,7 +4786,7 @@ static ssize_t intel_iommu_show_ndoms_used(struct device *dev,
struct device_attribute *attr,
char *buf)
{
- struct intel_iommu *iommu = dev_get_drvdata(dev);
+ struct intel_iommu *iommu = dev_to_intel_iommu(dev);
return sprintf(buf, "%d\n", bitmap_weight(iommu->domain_ids,
cap_ndoms(iommu->cap)));
}
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: intel-iommu sysfs oops.
@ 2017-02-23 23:30 ` Joerg Roedel
0 siblings, 0 replies; 5+ messages in thread
From: Joerg Roedel @ 2017-02-23 23:30 UTC (permalink / raw)
To: Dave Jones, Linux Kernel, dwmw2-wEGCiKHe2LqWVfeAwA7xHQ,
iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA
Hi Dave,
On Thu, Feb 23, 2017 at 02:44:06PM -0500, Dave Jones wrote:
> cat /sys/devices/virtual/iommu/dmar0/intel-iommu/version
>
> Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC
Thanks for the report, the problem reproduces easily here. The diff
below fixes the issue on Intel, AMD has the same problem and needs a
similar fix I'll do shortly.
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index f5e02f8..ec3f6c8 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -4730,11 +4730,16 @@ static int intel_iommu_cpu_dead(unsigned int cpu)
return 0;
}
+static struct intel_iommu *dev_to_intel_iommu(struct device *dev)
+{
+ return container_of(dev, struct intel_iommu, iommu.dev);
+}
+
static ssize_t intel_iommu_show_version(struct device *dev,
struct device_attribute *attr,
char *buf)
{
- struct intel_iommu *iommu = dev_get_drvdata(dev);
+ struct intel_iommu *iommu = dev_to_intel_iommu(dev);
u32 ver = readl(iommu->reg + DMAR_VER_REG);
return sprintf(buf, "%d:%d\n",
DMAR_VER_MAJOR(ver), DMAR_VER_MINOR(ver));
@@ -4745,7 +4750,7 @@ static ssize_t intel_iommu_show_address(struct device *dev,
struct device_attribute *attr,
char *buf)
{
- struct intel_iommu *iommu = dev_get_drvdata(dev);
+ struct intel_iommu *iommu = dev_to_intel_iommu(dev);
return sprintf(buf, "%llx\n", iommu->reg_phys);
}
static DEVICE_ATTR(address, S_IRUGO, intel_iommu_show_address, NULL);
@@ -4754,7 +4759,7 @@ static ssize_t intel_iommu_show_cap(struct device *dev,
struct device_attribute *attr,
char *buf)
{
- struct intel_iommu *iommu = dev_get_drvdata(dev);
+ struct intel_iommu *iommu = dev_to_intel_iommu(dev);
return sprintf(buf, "%llx\n", iommu->cap);
}
static DEVICE_ATTR(cap, S_IRUGO, intel_iommu_show_cap, NULL);
@@ -4763,7 +4768,7 @@ static ssize_t intel_iommu_show_ecap(struct device *dev,
struct device_attribute *attr,
char *buf)
{
- struct intel_iommu *iommu = dev_get_drvdata(dev);
+ struct intel_iommu *iommu = dev_to_intel_iommu(dev);
return sprintf(buf, "%llx\n", iommu->ecap);
}
static DEVICE_ATTR(ecap, S_IRUGO, intel_iommu_show_ecap, NULL);
@@ -4772,7 +4777,7 @@ static ssize_t intel_iommu_show_ndoms(struct device *dev,
struct device_attribute *attr,
char *buf)
{
- struct intel_iommu *iommu = dev_get_drvdata(dev);
+ struct intel_iommu *iommu = dev_to_intel_iommu(dev);
return sprintf(buf, "%ld\n", cap_ndoms(iommu->cap));
}
static DEVICE_ATTR(domains_supported, S_IRUGO, intel_iommu_show_ndoms, NULL);
@@ -4781,7 +4786,7 @@ static ssize_t intel_iommu_show_ndoms_used(struct device *dev,
struct device_attribute *attr,
char *buf)
{
- struct intel_iommu *iommu = dev_get_drvdata(dev);
+ struct intel_iommu *iommu = dev_to_intel_iommu(dev);
return sprintf(buf, "%d\n", bitmap_weight(iommu->domain_ids,
cap_ndoms(iommu->cap)));
}
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: intel-iommu sysfs oops.
@ 2017-02-24 21:28 ` Dave Jones
0 siblings, 0 replies; 5+ messages in thread
From: Dave Jones @ 2017-02-24 21:28 UTC (permalink / raw)
To: Joerg Roedel; +Cc: Linux Kernel, dwmw2, iommu
On Fri, Feb 24, 2017 at 12:30:38AM +0100, Joerg Roedel wrote:
> Hi Dave,
>
> On Thu, Feb 23, 2017 at 02:44:06PM -0500, Dave Jones wrote:
> > cat /sys/devices/virtual/iommu/dmar0/intel-iommu/version
> >
> > Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC
>
> Thanks for the report, the problem reproduces easily here. The diff
> below fixes the issue on Intel, AMD has the same problem and needs a
> similar fix I'll do shortly.
Seems to do the job. Thanks for the quick fix Joerg!
Dave
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: intel-iommu sysfs oops.
@ 2017-02-24 21:28 ` Dave Jones
0 siblings, 0 replies; 5+ messages in thread
From: Dave Jones @ 2017-02-24 21:28 UTC (permalink / raw)
To: Joerg Roedel
Cc: iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA,
dwmw2-wEGCiKHe2LqWVfeAwA7xHQ, Linux Kernel
On Fri, Feb 24, 2017 at 12:30:38AM +0100, Joerg Roedel wrote:
> Hi Dave,
>
> On Thu, Feb 23, 2017 at 02:44:06PM -0500, Dave Jones wrote:
> > cat /sys/devices/virtual/iommu/dmar0/intel-iommu/version
> >
> > Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC
>
> Thanks for the report, the problem reproduces easily here. The diff
> below fixes the issue on Intel, AMD has the same problem and needs a
> similar fix I'll do shortly.
Seems to do the job. Thanks for the quick fix Joerg!
Dave
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2017-02-24 21:28 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-02-23 19:44 intel-iommu sysfs oops Dave Jones
2017-02-23 23:30 ` Joerg Roedel
2017-02-23 23:30 ` Joerg Roedel
2017-02-24 21:28 ` Dave Jones
2017-02-24 21:28 ` Dave Jones
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.