Skip to contentSkip to conversation

perf/x86/intel/uncore: Fix oops when counting IMC uncore events on some TGL

PATCH Superseded [email protected] 3 participants 11 messages May 28, 2020

Subsystems (2)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/3 perf/x86/intel/uncore: Fix oops when counting IMC uncore events on some TGL

From: Kan Liang <[email protected]>

When counting IMC uncore events on some TGL machines, an oops will be triggered.

  [ 393.101262] BUG: unable to handle page fault for address:
  ffffb45200e15858
  [ 393.101269] #PF: supervisor read access in kernel mode
  [ 393.101271] #PF: error_code(0x0000) - not-present page

Current perf uncore driver still use the IMC MAP SIZE inherited from SNB, which is 0x6000.
However, the offset of IMC uncore counters is larger than 0x6000, e.g. 0xd8a0.

Enlarge the IMC MAP SIZE for TGL to 0xe000.

Fixes fdb64822443e ("perf/x86: Add Intel Tiger Lake uncore support")

---

No change since V1

 arch/x86/events/intel/uncore_snb.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
arch/x86/events/intel/uncore_snb.c +2 -1
index 3de1065..1038e9f 100644
@@ -1085,6 +1085,7 @@ static struct pci_dev *tgl_uncore_get_mc_dev(void)
 }
 
 #define TGL_UNCORE_MMIO_IMC_MEM_OFFSET		0x10000
+#define TGL_UNCORE_PCI_IMC_MAP_SIZE		0xe000
 
 static void tgl_uncore_imc_freerunning_init_box(struct intel_uncore_box *box)
 {
@@ -1112,7 +1113,7 @@ static void tgl_uncore_imc_freerunning_init_box(struct intel_uncore_box *box)
 	addr |= ((resource_size_t)mch_bar << 32);
 #endif
 
-	box->io_addr = ioremap(addr, SNB_UNCORE_PCI_IMC_MAP_SIZE);
+	box->io_addr = ioremap(addr, TGL_UNCORE_PCI_IMC_MAP_SIZE);
 }
 
 static struct intel_uncore_ops tgl_uncore_imc_freerunning_ops = {
Signature · 1 line
2.7.4
#1 lore raw

PATCH v2 2/3 perf/x86/intel/uncore: Record the size of mapped area

From: Kan Liang <[email protected]>

Perf cannot validate an address before the actual access to MMIO space of some uncore units, e.g. IMC on TGL. Accessing an invalid address, which exceeds mapped area, can trigger oops.

Perf never records the size of mapped area. Generic functions, e.g. uncore_mmio_read_counter(), cannot get the correct size for address validation.

Add mmio_map_size in intel_uncore_type to record the size of mapped area. Also sanity check the size before ioremap.

---

New patch

 arch/x86/events/intel/uncore.h       |  1 +
 arch/x86/events/intel/uncore_snb.c   | 20 ++++++++++++++++++--
 arch/x86/events/intel/uncore_snbep.c | 13 ++++++++++++-
 3 files changed, 31 insertions(+), 3 deletions(-)
arch/x86/events/intel/uncore.h +1 -0
index 0da4a46..c2e5725 100644
@@ -61,6 +61,7 @@ struct intel_uncore_type {
 		unsigned msr_offset;
 		unsigned mmio_offset;
 	};
+	unsigned mmio_map_size;
 	unsigned num_shared_regs:8;
 	unsigned single_fixed:1;
 	unsigned pair_ctr_ctl:1;
arch/x86/events/intel/uncore_snb.c +18 -2
index 1038e9f..52bcb69 100644
@@ -415,6 +415,7 @@ static const struct attribute_group snb_uncore_imc_format_group = {
 
 static void snb_uncore_imc_init_box(struct intel_uncore_box *box)
 {
+	struct intel_uncore_type *type = box->pmu->type;
 	struct pci_dev *pdev = box->pci_dev;
 	int where = SNB_UNCORE_PCI_IMC_BAR_OFFSET;
 	resource_size_t addr;
@@ -430,7 +431,13 @@ static void snb_uncore_imc_init_box(struct intel_uncore_box *box)
 
 	addr &= ~(PAGE_SIZE - 1);
 
-	box->io_addr = ioremap(addr, SNB_UNCORE_PCI_IMC_MAP_SIZE);
+	if (!type->mmio_map_size) {
+		pr_warn("perf uncore: Cannot ioremap for %s. Size of map area is 0.\n",
+			type->name);
+		return;
+	}
+
+	box->io_addr = ioremap(addr, type->mmio_map_size);
 	box->hrtimer_duration = UNCORE_SNB_IMC_HRTIMER_INTERVAL;
 }
 
@@ -586,6 +593,7 @@ static struct intel_uncore_type snb_uncore_imc = {
 	.num_counters   = 2,
 	.num_boxes	= 1,
 	.num_freerunning_types	= SNB_PCI_UNCORE_IMC_FREERUNNING_TYPE_MAX,
+	.mmio_map_size	= SNB_UNCORE_PCI_IMC_MAP_SIZE,
 	.freerunning	= snb_uncore_imc_freerunning,
 	.event_descs	= snb_uncore_imc_events,
 	.format_group	= &snb_uncore_imc_format_group,
@@ -1091,6 +1099,7 @@ static void tgl_uncore_imc_freerunning_init_box(struct intel_uncore_box *box)
 {
 	struct pci_dev *pdev = tgl_uncore_get_mc_dev();
 	struct intel_uncore_pmu *pmu = box->pmu;
+	struct intel_uncore_type *type = pmu->type;
 	resource_size_t addr;
 	u32 mch_bar;
 
@@ -1113,7 +1122,13 @@ static void tgl_uncore_imc_freerunning_init_box(struct intel_uncore_box *box)
 	addr |= ((resource_size_t)mch_bar << 32);
 #endif
 
-	box->io_addr = ioremap(addr, TGL_UNCORE_PCI_IMC_MAP_SIZE);
+	if (!type->mmio_map_size) {
+		pr_warn("perf uncore: Cannot ioremap for %s. Size of map area is 0.\n",
+			type->name);
+		return;
+	}
+
+	box->io_addr = ioremap(addr, type->mmio_map_size);
 }
 
 static struct intel_uncore_ops tgl_uncore_imc_freerunning_ops = {
@@ -1139,6 +1154,7 @@ static struct intel_uncore_type tgl_uncore_imc_free_running = {
 	.num_counters		= 3,
 	.num_boxes		= 2,
 	.num_freerunning_types	= TGL_MMIO_UNCORE_IMC_FREERUNNING_TYPE_MAX,
+	.mmio_map_size		= TGL_UNCORE_PCI_IMC_MAP_SIZE,
 	.freerunning		= tgl_uncore_imc_freerunning,
 	.ops			= &tgl_uncore_imc_freerunning_ops,
 	.event_descs		= tgl_uncore_imc_events,
arch/x86/events/intel/uncore_snbep.c +12 -1
index 07652fa..801b662 100644
@@ -4421,6 +4421,7 @@ static void __snr_uncore_mmio_init_box(struct intel_uncore_box *box,
 				       unsigned int box_ctl, int mem_offset)
 {
 	struct pci_dev *pdev = snr_uncore_get_mc_dev(box->dieid);
+	struct intel_uncore_type *type = box->pmu->type;
 	resource_size_t addr;
 	u32 pci_dword;
 
@@ -4435,7 +4436,13 @@ static void __snr_uncore_mmio_init_box(struct intel_uncore_box *box,
 
 	addr += box_ctl;
 
-	box->io_addr = ioremap(addr, SNR_IMC_MMIO_SIZE);
+	if (!type->mmio_map_size) {
+		pr_warn("perf uncore: Cannot ioremap for %s. Size of map area is 0.\n",
+			type->name);
+		return;
+	}
+
+	box->io_addr = ioremap(addr, type->mmio_map_size);
 	if (!box->io_addr)
 		return;
 
@@ -4530,6 +4537,7 @@ static struct intel_uncore_type snr_uncore_imc = {
 	.event_mask	= SNBEP_PMON_RAW_EVENT_MASK,
 	.box_ctl	= SNR_IMC_MMIO_PMON_BOX_CTL,
 	.mmio_offset	= SNR_IMC_MMIO_OFFSET,
+	.mmio_map_size	= SNR_IMC_MMIO_SIZE,
 	.ops		= &snr_uncore_mmio_ops,
 	.format_group	= &skx_uncore_format_group,
 };
@@ -4570,6 +4578,7 @@ static struct intel_uncore_type snr_uncore_imc_free_running = {
 	.num_counters		= 3,
 	.num_boxes		= 1,
 	.num_freerunning_types	= SNR_IMC_FREERUNNING_TYPE_MAX,
+	.mmio_map_size		= SNR_IMC_MMIO_SIZE,
 	.freerunning		= snr_imc_freerunning,
 	.ops			= &snr_uncore_imc_freerunning_ops,
 	.event_descs		= snr_uncore_imc_freerunning_events,
@@ -4987,6 +4996,7 @@ static struct intel_uncore_type icx_uncore_imc = {
 	.event_mask	= SNBEP_PMON_RAW_EVENT_MASK,
 	.box_ctl	= SNR_IMC_MMIO_PMON_BOX_CTL,
 	.mmio_offset	= SNR_IMC_MMIO_OFFSET,
+	.mmio_map_size	= SNR_IMC_MMIO_SIZE,
 	.ops		= &icx_uncore_mmio_ops,
 	.format_group	= &skx_uncore_format_group,
 };
@@ -5044,6 +5054,7 @@ static struct intel_uncore_type icx_uncore_imc_free_running = {
 	.num_counters		= 5,
 	.num_boxes		= 4,
 	.num_freerunning_types	= ICX_IMC_FREERUNNING_TYPE_MAX,
+	.mmio_map_size		= SNR_IMC_MMIO_SIZE,
 	.freerunning		= icx_imc_freerunning,
 	.ops			= &icx_uncore_imc_freerunning_ops,
 	.event_descs		= icx_uncore_imc_freerunning_events,
Signature · 1 line
2.7.4
#2 lore raw

Re: [PATCH V2 2/3] perf/x86/intel/uncore: Record the size of mapped area

arch/x86/events/intel/uncore_snb.c static void snb_uncore_imc_init_box(struct intel_uncore_box *box) · line 438

Quoted 6 lines · On Thu, May 28, 2020 at 06:15:26AM -0700, [email protected] wrote:
arch/x86/events/intel/uncore_snb.c +5 -1
-	box->io_addr = ioremap(addr, SNB_UNCORE_PCI_IMC_MAP_SIZE);
+	if (!type->mmio_map_size) {
+		pr_warn("perf uncore: Cannot ioremap for %s. Size of map area is 0.\n",
+			type->name);
+		return;
+	}

Is that likely that the size is 0?

In any case you have to test the return value of ioremap. So I would rather test the address for 0 than the size.

Similar in other code below.

I thought you were going to add a range check based on the size to the actual access. Did I miss it in the patch?

-Andi

author
#3 lore raw

Re: [PATCH V2 2/3] perf/x86/intel/uncore: Record the size of mapped area

Quoted 9 lines · Andi Kleen

arch/x86/events/intel/uncore_snb.c static void snb_uncore_imc_init_box(struct intel_uncore_box *box) · line 438

Quoted 6 lines · On Thu, May 28, 2020 at 06:15:26AM -0700, [email protected] wrote:
arch/x86/events/intel/uncore_snb.c +5 -1
-	box->io_addr = ioremap(addr, SNB_UNCORE_PCI_IMC_MAP_SIZE);
+	if (!type->mmio_map_size) {
+		pr_warn("perf uncore: Cannot ioremap for %s. Size of map area is 0.\n",
+			type->name);
+		return;
+	}

Is that likely that the size is 0?

In case someone forgets to set mmio_map_size.

Quoted 3 lines

In any case you have to test the return value of ioremap. So I would rather test the address for 0 than the size.

The box->io_addr is checked now, but there is no warning message.
I will remove the check for mmio_map_size, and add a warning message for the check of box->io_addr.

Thanks,
Kan

#4 lore raw

PATCH v2 3/3 perf/x86/intel/uncore: Validate MMIO address before accessing

In reply to [email protected]

From: Kan Liang <[email protected]>

An oops will be triggered, if perf tries to access an invalid address which exceeds the mapped area.

Check the address before the actual access to MMIO sapce of an uncore unit.

---
 arch/x86/events/intel/uncore.c       |  3 +++
 arch/x86/events/intel/uncore.h       | 12 ++++++++++++
 arch/x86/events/intel/uncore_snbep.c |  6 ++++++
 3 files changed, 21 insertions(+)
arch/x86/events/intel/uncore.c +3 -0
index cf76d66..284f8e7 100644
@@ -132,6 +132,9 @@ u64 uncore_mmio_read_counter(struct intel_uncore_box *box,
 	if (!box->io_addr)
 		return 0;
 
+	if (!is_valid_mmio_offset(box, event->hw.event_base))
+		return 0;
+
 	return readq(box->io_addr + event->hw.event_base);
 }
 
arch/x86/events/intel/uncore.h +12 -0
index c2e5725..672e9e1 100644
@@ -197,6 +197,18 @@ static inline bool uncore_pmc_freerunning(int idx)
 	return idx == UNCORE_PMC_IDX_FREERUNNING;
 }
 
+static inline bool is_valid_mmio_offset(struct intel_uncore_box *box,
+					unsigned long offset)
+{
+	if (box->pmu->type->mmio_map_size > offset)
+		return true;
+
+	pr_warn_once("perf uncore: Access invalid address of %s.\n",
+		     box->pmu->type->name);
+
+	return false;
+}
+
 static inline
 unsigned int uncore_mmio_box_ctl(struct intel_uncore_box *box)
 {
arch/x86/events/intel/uncore_snbep.c +6 -0
index 801b662..9c7a641 100644
@@ -4487,6 +4487,9 @@ static void snr_uncore_mmio_enable_event(struct intel_uncore_box *box,
 	if (!box->io_addr)
 		return;
 
+	if (!is_valid_mmio_offset(box, hwc->config_base))
+		return;
+
 	writel(hwc->config | SNBEP_PMON_CTL_EN,
 	       box->io_addr + hwc->config_base);
 }
@@ -4499,6 +4502,9 @@ static void snr_uncore_mmio_disable_event(struct intel_uncore_box *box,
 	if (!box->io_addr)
 		return;
 
+	if (!is_valid_mmio_offset(box, hwc->config_base))
+		return;
+
 	writel(hwc->config, box->io_addr + hwc->config_base);
 }
 
Signature · 1 line
2.7.4
#5 lore raw

Re: [PATCH V2 3/3] perf/x86/intel/uncore: Validate MMIO address before accessing

Quoted 7 lines · On Thu, May 28, 2020 at 06:15:27AM -0700, [email protected] wrote:

From: Kan Liang <[email protected]>

An oops will be triggered, if perf tries to access an invalid address which exceeds the mapped area.

Check the address before the actual access to MMIO sapce of an uncore unit.

Ah ok the range check is here

Quoted 14 lines
---
 arch/x86/events/intel/uncore.c       |  3 +++
 arch/x86/events/intel/uncore.h       | 12 ++++++++++++
 arch/x86/events/intel/uncore_snbep.c |  6 ++++++
 3 files changed, 21 insertions(+)
arch/x86/events/intel/uncore.c +0 -0
index cf76d66..284f8e7 100644

arch/x86/events/intel/uncore.c u64 uncore_mmio_read_counter(struct intel_uncore_box *box, · line 136

Quoted 5 lines
arch/x86/events/intel/uncore.c +2 -0
@@ -132,6 +132,9 @@ u64 uncore_mmio_read_counter(struct intel_uncore_box *box,
 	if (!box->io_addr)
 		return 0;
 
+	if (!is_valid_mmio_offset(box, event->hw.event_base))
+		return 0;

Is this function used somewhere else? Otherwise it should be added together with its users.

-Andi

author
#6 lore raw

Re: [PATCH V2 3/3] perf/x86/intel/uncore: Validate MMIO address before accessing

Quoted 28 lines · Andi Kleen
Quoted 7 lines · On Thu, May 28, 2020 at 06:15:27AM -0700, [email protected] wrote:

From: Kan Liang <[email protected]>

An oops will be triggered, if perf tries to access an invalid address which exceeds the mapped area.

Check the address before the actual access to MMIO sapce of an uncore unit.

Ah ok the range check is here

Quoted 17 lines
---
  arch/x86/events/intel/uncore.c       |  3 +++
  arch/x86/events/intel/uncore.h       | 12 ++++++++++++
  arch/x86/events/intel/uncore_snbep.c |  6 ++++++
  3 files changed, 21 insertions(+)
arch/x86/events/intel/uncore.c +0 -0
index cf76d66..284f8e7 100644
@@ -132,6 +132,9 @@ u64 uncore_mmio_read_counter(struct intel_uncore_box *box,
  	if (!box->io_addr)
  		return 0;
 
Quoted 6 lines

arch/x86/events/intel/uncore.c u64 uncore_mmio_read_counter(struct intel_uncore_box *box, · line 136

Quoted 2 lines
arch/x86/events/intel/uncore.c +2 -0
+	if (!is_valid_mmio_offset(box, event->hw.event_base))
+		return 0;

Is this function used somewhere else? Otherwise it should be added together with its users.

Yes, it's generic function. Current MMIO uncore units invoke it to read counter.

Thanks,
Kan

#7 lore raw

RE: [PATCH V2 3/3] perf/x86/intel/uncore: Validate MMIO address before accessing

In reply to [email protected]

From: [email protected]

Quoted 1 line

Sent: 28 May 2020 14:15

...

arch/x86/events/intel/uncore.h static inline bool uncore_pmc_freerunning(int idx) · line 201

Quoted 2 lines
arch/x86/events/intel/uncore.h +2 -0
+static inline bool is_valid_mmio_offset(struct intel_uncore_box *box,
+					unsigned long offset)

You need a better name, needs to start 'uncore_' and 'mmio' probably isn't right either.

arch/x86/events/intel/uncore.h static inline bool uncore_pmc_freerunning(int idx) · line 204

Quoted 3 lines
arch/x86/events/intel/uncore.h +3 -0
+{
+	if (box->pmu->type->mmio_map_size > offset)
+		return true;

Swap over.
Conditionals always read best if 'variable op constant'. So you want:

	if (offset < box->pmu->type->mmio_map_size)
		return true;

arch/x86/events/intel/uncore.h static inline bool uncore_pmc_freerunning(int idx) · line 207

Quoted 3 lines
arch/x86/events/intel/uncore.h +3 -0
+
+	pr_warn_once("perf uncore: Access invalid address of %s.\n",
+		     box->pmu->type->name);

Pretty hard to debug without the invalid offset.

I've no idea what the code is supposed to be doing though.

David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK Registration No: 1397386 (Wales)

author
#8 lore raw

Re: [PATCH V2 3/3] perf/x86/intel/uncore: Validate MMIO address before accessing

Quoted 9 lines · David Laight

From: [email protected]

Quoted 1 line

Sent: 28 May 2020 14:15

...

arch/x86/events/intel/uncore.h static inline bool uncore_pmc_freerunning(int idx) · line 201

Quoted 2 lines
arch/x86/events/intel/uncore.h +2 -0
+static inline bool is_valid_mmio_offset(struct intel_uncore_box *box,
+					unsigned long offset)

You need a better name, needs to start 'uncore_' and 'mmio' probably isn't right either.

Sure

Quoted 10 lines

arch/x86/events/intel/uncore.h static inline bool uncore_pmc_freerunning(int idx) · line 204

Quoted 3 lines
arch/x86/events/intel/uncore.h +3 -0
+{
+	if (box->pmu->type->mmio_map_size > offset)
+		return true;

Swap over.
Conditionals always read best if 'variable op constant'. So you want:

	if (offset < box->pmu->type->mmio_map_size)
		return true;

Sure

Quoted 6 lines

arch/x86/events/intel/uncore.h static inline bool uncore_pmc_freerunning(int idx) · line 207

Quoted 3 lines
arch/x86/events/intel/uncore.h +3 -0
+
+	pr_warn_once("perf uncore: Access invalid address of %s.\n",
+		     box->pmu->type->name);

Pretty hard to debug without the invalid offset.

I will dump the box->io_addr and offset for debugging.

Thanks,
Kan

Quoted 8 lines

I've no idea what the code is supposed to be doing though.

David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK Registration No: 1397386 (Wales)

#9 lore raw

Re: [PATCH V2 3/3] perf/x86/intel/uncore: Validate MMIO address before accessing

Quoted 8 lines
Quoted 6 lines

arch/x86/events/intel/uncore.h static inline bool uncore_pmc_freerunning(int idx) · line 207

Quoted 3 lines
arch/x86/events/intel/uncore.h +3 -0
+
+	pr_warn_once("perf uncore: Access invalid address of %s.\n",
+		     box->pmu->type->name);

Pretty hard to debug without the invalid offset.

I will dump the box->io_addr and offset for debugging.

Please don't overengineer.

-Andi

author
#10 lore raw

Re: [PATCH V2 3/3] perf/x86/intel/uncore: Validate MMIO address before accessing

Quoted 11 lines · Andi Kleen
Quoted 8 lines
Quoted 6 lines

arch/x86/events/intel/uncore.h static inline bool uncore_pmc_freerunning(int idx) · line 207

Quoted 3 lines
arch/x86/events/intel/uncore.h +3 -0
+
+	pr_warn_once("perf uncore: Access invalid address of %s.\n",
+		     box->pmu->type->name);

Pretty hard to debug without the invalid offset.

I will dump the box->io_addr and offset for debugging.

Please don't overengineer.

OK. Will only dump the invalid offset.

Thanks,
Kan