Skip to contentSkip to conversation

drm/xe/vsec: Support BMG devices

PATCH Superseded Michael J. Ruhl 3 participants 9 messages Aug 12, 2024 – Aug 15, 2024

Subsystems (3)

v11 of 12 Superseded Latest is v12 →

lore
Share & tools
lore
#0 lore raw

PATCH v11 1/1 drm/xe/vsec: Support BMG devices

The Battlemage (BMG) discrete graphics card supports the Platform, Monitoring Technology (PMT) feature directly on the primary PCI device.

Utilize the PMT callback API to add support for the BMG devices.

---

v10:
Use the mutex guard feature
Use the xe_pm_runtime_get_if_active() return value correctly Add r/b from Ilpo
Deferred DG2 patches for the future
v11:
add include for cleanup.h

 drivers/gpu/drm/xe/Makefile          |   1 +
 drivers/gpu/drm/xe/xe_device.c       |   5 +
 drivers/gpu/drm/xe/xe_device_types.h |   6 +
 drivers/gpu/drm/xe/xe_vsec.c         | 222 +++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_vsec.h         |  13 ++
 5 files changed, 247 insertions(+)
 create mode 100644 drivers/gpu/drm/xe/xe_vsec.c
 create mode 100644 drivers/gpu/drm/xe/xe_vsec.h
drivers/gpu/drm/xe/Makefile +1 -0
index 628c245c4822..f4c6761dc2e7 100644
@@ -129,6 +129,7 @@ xe-y += xe_bb.o \
 	xe_vm.o \
 	xe_vram.o \
 	xe_vram_freq.o \
+	xe_vsec.o \
 	xe_wait_user_fence.o \
 	xe_wa.o \
 	xe_wopcm.o
drivers/gpu/drm/xe/xe_device.c +5 -0
index 76109415eba6..a7c759c98560 100644
@@ -53,6 +53,7 @@
 #include "xe_ttm_sys_mgr.h"
 #include "xe_vm.h"
 #include "xe_vram.h"
+#include "xe_vsec.h"
 #include "xe_wait_user_fence.h"
 
 static int xe_file_open(struct drm_device *dev, struct drm_file *file)
@@ -317,6 +318,8 @@ struct xe_device *xe_device_create(struct pci_dev *pdev,
 		goto err;
 	}
 
+	drmm_mutex_init(&xe->drm, &xe->pmt.lock);
+
 	err = xe_display_create(xe);
 	if (WARN_ON(err))
 		goto err;
@@ -692,6 +695,8 @@ int xe_device_probe(struct xe_device *xe)
 	for_each_gt(gt, xe, id)
 		xe_gt_sanitize_freq(gt);
 
+	xe_vsec_init(xe);
+
 	return devm_add_action_or_reset(xe->drm.dev, xe_device_sanitize, xe);
 
 err_fini_display:
drivers/gpu/drm/xe/xe_device_types.h +6 -0
index 3bca6d344744..448a92148081 100644
@@ -451,6 +451,12 @@ struct xe_device {
 		struct mutex lock;
 	} d3cold;
 
+	/** @pmt: Support the PMT driver callback interface */
+	struct {
+		/** @pmt.lock: protect access for telemetry data */
+		struct mutex lock;
+	} pmt;
+
 	/**
 	 * @pm_callback_task: Track the active task that is running in either
 	 * the runtime_suspend or runtime_resume callbacks.
drivers/gpu/drm/xe/xe_vsec.c new file +222 -0

222 lines not on this page · raw message

drivers/gpu/drm/xe/xe_vsec.h new file +13 -0
new file mode 100644
index 000000000000..3fd29a21cad6
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright © 2022 - 2024 Intel Corporation
+ */
+
+#ifndef _XE_VSEC_H_
+#define _XE_VSEC_H_
+
+struct xe_device;
+
+void xe_vsec_init(struct xe_device *xe);
+
+#endif
Signature · 1 line
2.44.0
#1 lore raw

✗ CI.Patch_applied: failure for drm/xe/vsec: Support BMG devices (rev3)

== Series Details ==

== Series Details ==

Series: drm/xe/vsec: Support BMG devices (rev3)
URL : https://patchwork.freedesktop.org/series/137199/
State : failure

== Summary ==

=== Applying kernel patches on branch 'drm-tip' with base: ===
Base commit: 4f95bd933f4d drm-tip: 2024y-08m-12d-17h-20m-47s UTC integration manifest === git am output follows ===
error: patch failed: drivers/gpu/drm/xe/xe_device.c:53
error: drivers/gpu/drm/xe/xe_device.c: patch does not apply
hint: Use 'git am --show-current-patch=diff' to see the failed patch
hint: When you have resolved this problem, run "git am --continue".
hint: If you prefer to skip this patch, run "git am --skip" instead.
hint: To restore the original branch and stop patching, run "git am --abort".
hint: Disable this message with "git config advice.mergeConflict false"
Applying: drm/xe/vsec: Support BMG devices
Patch failed at 0001 drm/xe/vsec: Support BMG devices

#2 lore raw
In reply to Michael J. Ruhl
Quoted 6 lines · Michael J. Ruhl

The Battlemage (BMG) discrete graphics card supports the Platform, Monitoring Technology (PMT) feature directly on the primary PCI device.

Utilize the PMT callback API to add support for the BMG devices.

...

Quoted 7 lines
drivers/gpu/drm/xe/xe_vsec.c new file +7 -0
+#include <linux/bitfield.h>
+#include <linux/bits.h>
+#include <linux/cleanup.h>
+#include <linux/intel_vsec.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/pci.h>

...

drivers/gpu/drm/xe/xe_vsec.c line 27

Quoted 7 lines
drivers/gpu/drm/xe/xe_vsec.c new file +6 -0
+#define SOC_BASE		0x280000
+
+#define BMG_PMT_BASE		0xDB000
+#define BMG_DISCOVERY_OFFSET	(SOC_BASE + BMG_PMT_BASE)
 
+#define BMG_TELEMETRY_BASE	0xE0000
+#define BMG_TELEMETRY_OFFSET	(SOC_BASE + BMG_TELEMETRY_BASE)

This looks like double indirection.
Wouldn't suffix _BASE_OFFSET be better for PMT and TELEMETRY cases?

...

Quoted 1 line

+#define BMG_DEVICE_ID 0xE2F8

Is this defined in any specification? I mean is the format the same as PCI device ID?

...

Quoted 1 line

+#define GFX_BAR 0

Do you need a separate definition for this?

...

drivers/gpu/drm/xe/xe_vsec.c line 88

Quoted 4 lines
drivers/gpu/drm/xe/xe_vsec.c new file +4 -0
+enum record_id {
+	PUNIT,
+	OOBMSM_0,
+	OOBMSM_1

Trailing comma?

drivers/gpu/drm/xe/xe_vsec.c line 94

Quoted 6 lines
drivers/gpu/drm/xe/xe_vsec.c new file +6 -0
+};
+
+enum capability {
+	CRASHLOG,
+	TELEMETRY,
+	WATCHER

Ditto?

Quoted 1 line

+};

...

Quoted 10 lines
drivers/gpu/drm/xe/xe_vsec.c new file +10 -0
+	switch (record_id) {
+	case PUNIT:
+		*index = 0;
+		if (cap_type == TELEMETRY)
+			*offset = PUNIT_TELEMETRY_OFFSET;
+		else
+			*offset = PUNIT_WATCHER_OFFSET;
+		break;
+
+	case OOBMSM_0:

drivers/gpu/drm/xe/xe_vsec.c line 135

Quoted 10 lines
drivers/gpu/drm/xe/xe_vsec.c new file +10 -0
+		*index = 1;
+		if (cap_type == WATCHER)
+			*offset = OOBMSM_0_WATCHER_OFFSET;
+		break;
+
+	case OOBMSM_1:
+		*index = 1;
+		if (cap_type == TELEMETRY)
+			*offset = OOBMSM_1_TELEMETRY_OFFSET;
+		break;

default case?

Quoted 1 line
(quoted diff) +1 -0
+	}

...

Quoted 18 lines
drivers/gpu/drm/xe/xe_vsec.c new file +18 -0
+static int xe_pmt_telem_read(struct pci_dev *pdev, u32 guid, u64 *data, u32 count)
+{
+	struct xe_device *xe = pdev_to_xe_device(pdev);
+	void __iomem *telem_addr = xe->mmio.regs + BMG_TELEMETRY_OFFSET;
+	u32 mem_region;
+	u32 offset;
+	int ret;
+
+	ret = guid_decode(guid, &mem_region, &offset);
+	if (ret)
+		return ret;
+
+	telem_addr += offset;
+
+	guard(mutex)(&xe->pmt.lock);
+
+	/* indicate that we are not at an appropriate power level */
+	if (!xe_pm_runtime_get_if_active(xe))

drivers/gpu/drm/xe/xe_vsec.c line 167

Quoted 10 lines
drivers/gpu/drm/xe/xe_vsec.c new file +9 -0
+		return -ENODATA;
+
+	/* set SoC re-mapper index register based on GUID memory region */
+	xe_mmio_rmw32(xe->tiles[0].primary_gt, SG_REMAP_INDEX1, SG_REMAP_BITS,
+		      FIELD_PREP(SG_REMAP_BITS, mem_region));
+
+	memcpy_fromio(data, telem_addr, count);
 
+	ret = count;
+	xe_pm_runtime_put(xe);

Does this have a side effect on count? If yes, a comment, if no, you may return count directly.

Quoted 2 lines
(quoted diff) +2 -0
+	return ret;
+}
Signature · 2 lines
With Best Regards,
Andy Shevchenko
#3 lore raw
Quoted 34 lines

-----Original Message-----
From: Andy Shevchenko <[email protected]>
Sent: Tuesday, August 13, 2024 10:11 AM
To: Ruhl, Michael J <[email protected]>
Cc: [email protected]; [email protected]; [email protected]; [email protected]; Brost, Matthew <[email protected]>; [email protected]; Vivi, Rodrigo <[email protected]>
Subject: Re: [PATCH v11] drm/xe/vsec: Support BMG devices

Quoted 5 lines · Michael J. Ruhl

The Battlemage (BMG) discrete graphics card supports the Platform, Monitoring Technology (PMT) feature directly on the primary PCI device.

Utilize the PMT callback API to add support for the BMG devices.

...

Quoted 7 lines
drivers/gpu/drm/xe/xe_vsec.c new file +7 -0
+#include <linux/bitfield.h>
+#include <linux/bits.h>
+#include <linux/cleanup.h>
+#include <linux/intel_vsec.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/pci.h>

...

Quoted 4 lines
drivers/gpu/drm/xe/xe_vsec.c new file +4 -0
+#define SOC_BASE		0x280000
+
+#define BMG_PMT_BASE		0xDB000
+#define BMG_DISCOVERY_OFFSET	(SOC_BASE + BMG_PMT_BASE)
Quoted 5 lines

drivers/gpu/drm/xe/xe_vsec.c line 27

Quoted 2 lines
drivers/gpu/drm/xe/xe_vsec.c new file +2 -0
+#define BMG_TELEMETRY_BASE	0xE0000
+#define BMG_TELEMETRY_OFFSET	(SOC_BASE + BMG_TELEMETRY_BASE)

This looks like double indirection.
Wouldn't suffix _BASE_OFFSET be better for PMT and TELEMETRY cases?

I am not sure I understand.

Are you saying rename BMG_PMT_BASE to BMG_PMT_BASE_OFFSET?

Quoted 7 lines

...

Quoted 1 line

+#define BMG_DEVICE_ID 0xE2F8

Is this defined in any specification? I mean is the format the same as PCI device ID?

I think that this is defined in BMG PMT yaml definition. It is provide in the PMT discovery data, so it is defined by the specific device.

Quoted 5 lines

...

Quoted 1 line

+#define GFX_BAR 0

Do you need a separate definition for this?

Guess not. Will remove. 😊

Quoted 9 lines

...

drivers/gpu/drm/xe/xe_vsec.c line 88

Quoted 4 lines
drivers/gpu/drm/xe/xe_vsec.c new file +4 -0
+enum record_id {
+	PUNIT,
+	OOBMSM_0,
+	OOBMSM_1

Trailing comma?

Quoted 8 lines

drivers/gpu/drm/xe/xe_vsec.c line 94

Quoted 6 lines
drivers/gpu/drm/xe/xe_vsec.c new file +6 -0
+};
+
+enum capability {
+	CRASHLOG,
+	TELEMETRY,
+	WATCHER

Ditto?

Will update.

Quoted 18 lines
Quoted 1 line

+};

...

Quoted 14 lines
drivers/gpu/drm/xe/xe_vsec.c new file +14 -0
+	switch (record_id) {
+	case PUNIT:
+		*index = 0;
+		if (cap_type == TELEMETRY)
+			*offset = PUNIT_TELEMETRY_OFFSET;
+		else
+			*offset = PUNIT_WATCHER_OFFSET;
+		break;
+
+	case OOBMSM_0:
+		*index = 1;
+		if (cap_type == WATCHER)
+			*offset = OOBMSM_0_WATCHER_OFFSET;
+		break;
Quoted 8 lines

drivers/gpu/drm/xe/xe_vsec.c line 135

Quoted 6 lines
drivers/gpu/drm/xe/xe_vsec.c new file +6 -0
+
+	case OOBMSM_1:
+		*index = 1;
+		if (cap_type == TELEMETRY)
+			*offset = OOBMSM_1_TELEMETRY_OFFSET;
+		break;

default case?

I validate the record_id and cap_type values at the beginning of the function, so default would be redundant.

The goal was to validate, then set data.

So adding the default will remove the record_id check from the if. Do you prefer that path?

Quoted 32 lines
Quoted 1 line
(quoted diff) +1 -0
+	}

...

Review

Quoted 4 lines
(quoted diff) +4 -0
+static int xe_pmt_telem_read(struct pci_dev *pdev, u32 guid, u64
+*data, u32 count) {
+	struct xe_device *xe = pdev_to_xe_device(pdev);
+	void __iomem *telem_addr = xe->mmio.regs +

BMG_TELEMETRY_OFFSET;

Quoted 8 lines
drivers/gpu/drm/xe/xe_vsec.c new file +8 -0
+	u32 mem_region;
+	u32 offset;
+	int ret;
+
+	ret = guid_decode(guid, &mem_region, &offset);
+	if (ret)
+		return ret;
+

drivers/gpu/drm/xe/xe_vsec.c

Quoted 10 lines
drivers/gpu/drm/xe/xe_vsec.c new file +10 -0
+	telem_addr += offset;
+
+	guard(mutex)(&xe->pmt.lock);
+
+	/* indicate that we are not at an appropriate power level */
+	if (!xe_pm_runtime_get_if_active(xe))
+		return -ENODATA;
+
+	/* set SoC re-mapper index register based on GUID memory region */
+	xe_mmio_rmw32(xe->tiles[0].primary_gt, SG_REMAP_INDEX1,

SG_REMAP_BITS,

Quoted 3 lines
drivers/gpu/drm/xe/xe_vsec.c new file +3 -0
+		      FIELD_PREP(SG_REMAP_BITS, mem_region));
+
+	memcpy_fromio(data, telem_addr, count);
Quoted 5 lines

drivers/gpu/drm/xe/xe_vsec.c line 167

Quoted 2 lines
drivers/gpu/drm/xe/xe_vsec.c new file +2 -0
+	ret = count;
+	xe_pm_runtime_put(xe);

Does this have a side effect on count? If yes, a comment, if no, you may return count directly.

It does not. I was looking at that yesterday (when I updated with the guard()) and thought about updating. I will update.

Thank you for your comments!

M

Quoted 8 lines

Review

Quoted 3 lines
(quoted diff) +2 -0
+	return ret;
+}

--
With Best Regards,
Andy Shevchenko

#4 lore raw
Quoted 3 lines · Michael J
Quoted 3 lines

From: Andy Shevchenko <[email protected]>
Sent: Tuesday, August 13, 2024 10:11 AM
On Mon, Aug 12, 2024 at 04:04:22PM -0400, Michael J. Ruhl wrote:

...

Quoted 11 lines
Quoted 10 lines

drivers/gpu/drm/xe/xe_vsec.c line 27

Quoted 7 lines
drivers/gpu/drm/xe/xe_vsec.c new file +6 -0
+#define SOC_BASE		0x280000
+
+#define BMG_PMT_BASE		0xDB000
+#define BMG_DISCOVERY_OFFSET	(SOC_BASE + BMG_PMT_BASE)
 
+#define BMG_TELEMETRY_BASE	0xE0000
+#define BMG_TELEMETRY_OFFSET	(SOC_BASE + BMG_TELEMETRY_BASE)

This looks like double indirection.
Wouldn't suffix _BASE_OFFSET be better for PMT and TELEMETRY cases?

Quoted 3 lines

I am not sure I understand.

Are you saying rename BMG_PMT_BASE to BMG_PMT_BASE_OFFSET?

Yes. Same for BMG_TELEMETRY_.

...

Quoted 7 lines
Quoted 4 lines
Quoted 1 line

+#define BMG_DEVICE_ID 0xE2F8

Is this defined in any specification? I mean is the format the same as PCI device ID?

I think that this is defined in BMG PMT yaml definition. It is provide in the PMT discovery data, so it is defined by the specific device.

Is there any documentation / specification about this? Can it be UUID or 64-bit number or other format? _Where_ is this being specified?

...

Quoted 23 lines
Quoted 14 lines
Quoted 14 lines
drivers/gpu/drm/xe/xe_vsec.c new file +14 -0
+	switch (record_id) {
+	case PUNIT:
+		*index = 0;
+		if (cap_type == TELEMETRY)
+			*offset = PUNIT_TELEMETRY_OFFSET;
+		else
+			*offset = PUNIT_WATCHER_OFFSET;
+		break;
+
+	case OOBMSM_0:
+		*index = 1;
+		if (cap_type == WATCHER)
+			*offset = OOBMSM_0_WATCHER_OFFSET;
+		break;
Quoted 8 lines

drivers/gpu/drm/xe/xe_vsec.c line 135

Quoted 6 lines
drivers/gpu/drm/xe/xe_vsec.c new file +6 -0
+
+	case OOBMSM_1:
+		*index = 1;
+		if (cap_type == TELEMETRY)
+			*offset = OOBMSM_1_TELEMETRY_OFFSET;
+		break;

default case?

Quoted 7 lines

I validate the record_id and cap_type values at the beginning of the function, so default would be redundant.

The goal was to validate, then set data.

So adding the default will remove the record_id check from the if. Do you prefer that path?

Yes.

Quoted 1 line
Quoted 1 line
Quoted 1 line
(quoted diff) +1 -0
+	}
Signature · 2 lines
With Best Regards,
Andy Shevchenko
#5 lore raw
Quoted 27 lines

-----Original Message-----
From: Andy Shevchenko <[email protected]>
Sent: Wednesday, August 14, 2024 9:56 AM
To: Ruhl, Michael J <[email protected]>
Cc: [email protected]; [email protected]; [email protected]; [email protected]; Brost, Matthew <[email protected]>; [email protected]; Vivi, Rodrigo <[email protected]>
Subject: Re: [PATCH v11] drm/xe/vsec: Support BMG devices

Quoted 3 lines · Michael J
Quoted 3 lines

From: Andy Shevchenko <[email protected]>
Sent: Tuesday, August 13, 2024 10:11 AM On Mon, Aug 12, 2024 at 04:04:22PM -0400, Michael J. Ruhl wrote:

...

Quoted 4 lines
Quoted 4 lines

drivers/gpu/drm/xe/xe_vsec.c

Quoted 4 lines
drivers/gpu/drm/xe/xe_vsec.c new file +4 -0
+#define SOC_BASE		0x280000
+
+#define BMG_PMT_BASE		0xDB000
+#define BMG_DISCOVERY_OFFSET	(SOC_BASE +

BMG_PMT_BASE)

Quoted 3 lines
Quoted 3 lines

Review

Quoted 3 lines
(quoted diff) +2 -0
+#define BMG_TELEMETRY_BASE	0xE0000
+#define BMG_TELEMETRY_OFFSET	(SOC_BASE +

BMG_TELEMETRY_BASE)

Quoted 1 line
Quoted 1 line
Quoted 8 lines
Quoted 6 lines
Quoted 2 lines

This looks like double indirection.
Wouldn't suffix _BASE_OFFSET be better for PMT and TELEMETRY cases?

I am not sure I understand.

Are you saying rename BMG_PMT_BASE to BMG_PMT_BASE_OFFSET?

Yes. Same for BMG_TELEMETRY_.

Got it. I will update.

Quoted 7 lines

...

Quoted 5 lines
Quoted 4 lines
Quoted 1 line

+#define BMG_DEVICE_ID 0xE2F8

Is this defined in any specification? I mean is the format the same as PCI device ID?

Quoted 6 lines
Quoted 2 lines

I think that this is defined in BMG PMT yaml definition. It is provide in the PMT discovery data, so it is defined by the specific device.

Is there any documentation / specification about this? Can it be UUID or 64-bit number or other format? _Where_ is this being specified?

The GUID is defined by the YAML file associated with the PMT device. In this case 16 bits are a device ID.

From the cover letter of the PMT patch set (Intel Platform Monitoring Technology):

-
The GUID uniquely identifies the register space of any monitor data exposed by the capability. The GUID is associated with an XML file from the vendor that describes the mapping of the register space along with properties of the monitor data.
--

I was told that this was the value to use for this specific device/feature.

It is specified internally. Not sure if there is any "documentation" available beyond that.

Quoted 28 lines

...

Quoted 26 lines
Quoted 14 lines
Quoted 14 lines
drivers/gpu/drm/xe/xe_vsec.c new file +14 -0
+	switch (record_id) {
+	case PUNIT:
+		*index = 0;
+		if (cap_type == TELEMETRY)
+			*offset = PUNIT_TELEMETRY_OFFSET;
+		else
+			*offset = PUNIT_WATCHER_OFFSET;
+		break;
+
+	case OOBMSM_0:
+		*index = 1;
+		if (cap_type == WATCHER)
+			*offset = OOBMSM_0_WATCHER_OFFSET;
+		break;
Quoted 8 lines

drivers/gpu/drm/xe/xe_vsec.c line 135

Quoted 6 lines
drivers/gpu/drm/xe/xe_vsec.c new file +6 -0
+
+	case OOBMSM_1:
+		*index = 1;
+		if (cap_type == TELEMETRY)
+			*offset = OOBMSM_1_TELEMETRY_OFFSET;
+		break;

default case?

I validate the record_id and cap_type values at the beginning of the function, so default would be redundant.

Quoted 6 lines
Quoted 4 lines

The goal was to validate, then set data.

So adding the default will remove the record_id check from the if. Do you prefer that path?

Yes.

Ok, I will update.

Thanks!

Mike

Quoted 6 lines
Quoted 1 line
Quoted 1 line

Review

Quoted 1 line
(quoted diff) +1 -0
+	}

--
With Best Regards,
Andy Shevchenko

#6 lore raw
Quoted 6 lines · Michael J
Quoted 6 lines

From: Andy Shevchenko <[email protected]>
Sent: Wednesday, August 14, 2024 9:56 AM

Quoted 3 lines · Michael J
Quoted 3 lines

From: Andy Shevchenko <[email protected]>
Sent: Tuesday, August 13, 2024 10:11 AM On Mon, Aug 12, 2024 at 04:04:22PM -0400, Michael J. Ruhl wrote:

...

Quoted 24 lines
Quoted 11 lines
Quoted 7 lines
Quoted 4 lines
Quoted 1 line

+#define BMG_DEVICE_ID 0xE2F8

Is this defined in any specification? I mean is the format the same as PCI device ID?

I think that this is defined in BMG PMT yaml definition. It is provide in the PMT discovery data, so it is defined by the specific device.

Is there any documentation / specification about this? Can it be UUID or 64-bit number or other format? _Where_ is this being specified?

The GUID is defined by the YAML file associated with the PMT device. In this case 16 bits are a device ID.

From the cover letter of the PMT patch set (Intel Platform Monitoring Technology):

-
The GUID uniquely identifies the register space of any monitor data exposed by the capability. The GUID is associated with an XML file from the vendor that describes the mapping of the register space along with properties of the monitor data.
--

Quoted 4 lines

I was told that this was the value to use for this specific device/feature.

It is specified internally. Not sure if there is any "documentation" available beyond that.

The YAML is *not* the specification. Do we have one that I can access to? And I asked not about GUID, I asked about ID.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#7 lore raw
Quoted 47 lines

-----Original Message-----
From: Andy Shevchenko <[email protected]>
Sent: Wednesday, August 14, 2024 2:41 PM
To: Ruhl, Michael J <[email protected]>
Cc: [email protected]; [email protected]; [email protected]; [email protected]; Brost, Matthew <[email protected]>; [email protected]; Vivi, Rodrigo <[email protected]>
Subject: Re: [PATCH v11] drm/xe/vsec: Support BMG devices

Quoted 6 lines · Michael J
Quoted 6 lines

From: Andy Shevchenko <[email protected]>
Sent: Wednesday, August 14, 2024 9:56 AM On Tue, Aug 13, 2024 at

Quoted 3 lines · Michael J
Quoted 3 lines

From: Andy Shevchenko <[email protected]>
Sent: Tuesday, August 13, 2024 10:11 AM On Mon, Aug 12, 2024 at 04:04:22PM -0400, Michael J. Ruhl wrote:

...

Quoted 8 lines
Quoted 8 lines
Quoted 7 lines
Quoted 4 lines
Quoted 1 line

+#define BMG_DEVICE_ID 0xE2F8

Is this defined in any specification? I mean is the format the same as PCI device ID?

I think that this is defined in BMG PMT yaml definition. It is provide in the PMT discovery data, so it is defined by the specific device.

Quoted 8 lines
Quoted 3 lines

Is there any documentation / specification about this? Can it be UUID or 64-bit number or other format? _Where_ is this being specified?

The GUID is defined by the YAML file associated with the PMT device. In this case 16 bits are a device ID.

From the cover letter of the PMT patch set (Intel Platform Monitoring

Technology):

Quoted 8 lines

-
The GUID uniquely identifies the register space of any monitor data exposed by the capability. The GUID is associated with an XML file from the vendor that describes the mapping of the register space along with properties of the monitor data.
--

Quoted 7 lines
Quoted 4 lines

I was told that this was the value to use for this specific device/feature.

It is specified internally. Not sure if there is any "documentation" available beyond that.

The YAML is *not* the specification. Do we have one that I can access to? And I asked not about GUID, I asked about ID.

Andy,

For the BMG device, the device ID is defined as part PMT GUID, and will be defined by the BMG PMT YAML specification.

So this is a vendor defined value.

Need to do some testing, and then I will re-post the patch.

Thanks!

M

Quoted 4 lines

--
With Best Regards,
Andy Shevchenko

#8 lore raw
Quoted 10 lines · Michael J
Quoted 10 lines

-----Original Message-----
From: Andy Shevchenko <[email protected]>
Sent: Wednesday, August 14, 2024 2:41 PM

Quoted 6 lines · Michael J
Quoted 6 lines

From: Andy Shevchenko <[email protected]>
Sent: Wednesday, August 14, 2024 9:56 AM On Tue, Aug 13, 2024 at

Quoted 3 lines · Michael J
Quoted 3 lines

From: Andy Shevchenko <[email protected]>
Sent: Tuesday, August 13, 2024 10:11 AM On Mon, Aug 12, 2024 at 04:04:22PM -0400, Michael J. Ruhl wrote:

...

Quoted 33 lines
Quoted 25 lines
Quoted 8 lines
Quoted 8 lines
Quoted 7 lines
Quoted 4 lines
Quoted 1 line

+#define BMG_DEVICE_ID 0xE2F8

Is this defined in any specification? I mean is the format the same as PCI device ID?

I think that this is defined in BMG PMT yaml definition. It is provide in the PMT discovery data, so it is defined by the specific device.

Quoted 8 lines
Quoted 3 lines

Is there any documentation / specification about this? Can it be UUID or 64-bit number or other format? _Where_ is this being specified?

The GUID is defined by the YAML file associated with the PMT device. In this case 16 bits are a device ID.

From the cover letter of the PMT patch set (Intel Platform Monitoring

Technology):

Quoted 8 lines

-
The GUID uniquely identifies the register space of any monitor data exposed by the capability. The GUID is associated with an XML file from the vendor that describes the mapping of the register space along with properties of the monitor data.
--

Quoted 7 lines
Quoted 4 lines

I was told that this was the value to use for this specific device/feature.

It is specified internally. Not sure if there is any "documentation" available beyond that.

The YAML is *not* the specification. Do we have one that I can access to? And I asked not about GUID, I asked about ID.

Quoted 6 lines

Andy,

For the BMG device, the device ID is defined as part PMT GUID, and will be defined by the BMG PMT YAML specification.

So this is a vendor defined value.

Okay, thank you for clarification.

Quoted 1 line

Need to do some testing, and then I will re-post the patch.

Signature · 2 lines
With Best Regards,
Andy Shevchenko