[PATCH v11 1/2] drm/amdgpu: add debugfs for reset registers list

[PATCH v11 1/2] drm/amdgpu: add debugfs for reset registers list

[Somalapuram Amaranath]

List of register populated for dump collection during the GPU reset.

Signed-off-by: Somalapuram Amaranath <Amaranath.Somalapuram@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h         |  4 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 81 +++++++++++++++++++++
 2 files changed, 85 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index b85b67a88a3d..6e35f2c4c869 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1097,6 +1097,10 @@ struct amdgpu_device {
 
 	struct amdgpu_reset_control     *reset_cntl;
 	uint32_t                        ip_versions[HW_ID_MAX][HWIP_MAX_INSTANCE];
+
+	/* reset dump register */
+	uint32_t			*reset_dump_reg_list;
+	int                             num_regs;
 };
 
 static inline struct amdgpu_device *drm_to_adev(struct drm_device *ddev)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index 164d6a9e9fbb..df6d9fb69657 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -1609,6 +1609,85 @@ DEFINE_DEBUGFS_ATTRIBUTE(fops_ib_preempt, NULL,
 DEFINE_DEBUGFS_ATTRIBUTE(fops_sclk_set, NULL,
 			amdgpu_debugfs_sclk_set, "%llu\n");
 
+static ssize_t amdgpu_reset_dump_register_list_read(struct file *f,
+				char __user *buf, size_t size, loff_t *pos)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)file_inode(f)->i_private;
+	char reg_offset[11];
+	int i, ret, len = 0;
+
+	if (*pos)
+		return 0;
+
+	ret = down_read_killable(&adev->reset_sem);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < adev->num_regs; i++) {
+		sprintf(reg_offset, "0x%x\n", adev->reset_dump_reg_list[i]);
+		up_read(&adev->reset_sem);
+		ret = copy_to_user(buf + len, reg_offset, strlen(reg_offset));
+		if (ret)
+			return ret;
+
+		len += strlen(reg_offset);
+		ret = down_read_killable(&adev->reset_sem);
+		if (ret)
+			return ret;
+	}
+
+	up_read(&adev->reset_sem);
+	if (ret)
+		return ret;
+
+	*pos += len;
+
+	return len;
+}
+
+static ssize_t amdgpu_reset_dump_register_list_write(struct file *f,
+			const char __user *buf, size_t size, loff_t *pos)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)file_inode(f)->i_private;
+	char reg_offset[12];
+	uint32_t *tmp;
+	int ret, i = 0, len = 0;
+
+	do {
+		memset(reg_offset, 0, 12);
+		ret = copy_from_user(reg_offset, buf + len, min(11, ((int)size-len)));
+		if (ret)
+			goto error_free;
+
+		tmp = krealloc_array(tmp, i + 1, sizeof(uint32_t), GFP_KERNEL);
+		if (sscanf(reg_offset, "%X %n", &tmp[i], &ret) != 1)
+			goto error_free;
+
+		len += ret;
+		i++;
+	} while (len < size);
+
+	ret = down_write_killable(&adev->reset_sem);
+	if (ret)
+		goto error_free;
+
+	swap(adev->reset_dump_reg_list, tmp);
+	adev->num_regs = i;
+	up_write(&adev->reset_sem);
+	ret = size;
+
+error_free:
+	kfree(tmp);
+	return ret;
+}
+
+static const struct file_operations amdgpu_reset_dump_register_list = {
+	.owner = THIS_MODULE,
+	.read = amdgpu_reset_dump_register_list_read,
+	.write = amdgpu_reset_dump_register_list_write,
+	.llseek = default_llseek
+};
+
 int amdgpu_debugfs_init(struct amdgpu_device *adev)
 {
 	struct dentry *root = adev_to_drm(adev)->primary->debugfs_root;
@@ -1672,6 +1751,8 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev)
 			    &amdgpu_debugfs_test_ib_fops);
 	debugfs_create_file("amdgpu_vm_info", 0444, root, adev,
 			    &amdgpu_debugfs_vm_info_fops);
+	debugfs_create_file("amdgpu_reset_dump_register_list", 0644, root, adev,
+			    &amdgpu_reset_dump_register_list);
 
 	adev->debugfs_vbios_blob.data = adev->bios;
 	adev->debugfs_vbios_blob.size = adev->bios_size;
-- 
2.25.1

[PATCH v11 2/2] drm/amdgpu: add reset register dump trace on GPU

[Somalapuram Amaranath]

Dump the list of register values to trace event on GPU reset.

Signed-off-by: Somalapuram Amaranath <Amaranath.Somalapuram@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 17 +++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h  | 16 ++++++++++++++++
 2 files changed, 33 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 1e651b959141..7c48fd716adb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -4534,6 +4534,22 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
 	return r;
 }
 
+static int amdgpu_reset_reg_dumps(struct amdgpu_device *adev)
+{
+	uint32_t reg_value;
+	int i;
+
+	lockdep_assert_held(&adev->reset_sem);
+	dump_stack();
+
+	for (i = 0; i < adev->num_regs; i++) {
+		reg_value = RREG32(adev->reset_dump_reg_list[i]);
+		trace_amdgpu_reset_reg_dumps(adev->reset_dump_reg_list[i], reg_value);
+	}
+
+	return 0;
+}
+
 int amdgpu_do_asic_reset(struct list_head *device_list_handle,
 			 struct amdgpu_reset_context *reset_context)
 {
@@ -4544,6 +4560,7 @@ int amdgpu_do_asic_reset(struct list_head *device_list_handle,
 	/* Try reset handler method first */
 	tmp_adev = list_first_entry(device_list_handle, struct amdgpu_device,
 				    reset_list);
+	amdgpu_reset_reg_dumps(tmp_adev);
 	r = amdgpu_reset_perform_reset(tmp_adev, reset_context);
 	/* If reset handler not implemented, continue; otherwise return */
 	if (r == -ENOSYS)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
index d855cb53c7e0..b9637925e85c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
@@ -537,6 +537,22 @@ TRACE_EVENT(amdgpu_ib_pipe_sync,
 		      __entry->seqno)
 );
 
+TRACE_EVENT(amdgpu_reset_reg_dumps,
+	    TP_PROTO(uint32_t address, uint32_t value),
+	    TP_ARGS(address, value),
+	    TP_STRUCT__entry(
+			     __field(uint32_t, address)
+			     __field(uint32_t, value)
+			     ),
+	    TP_fast_assign(
+			   __entry->address = address;
+			   __entry->value = value;
+			   ),
+	    TP_printk("amdgpu register dump 0x%x: 0x%x",
+		      __entry->address,
+		      __entry->value)
+);
+
 #undef AMDGPU_JOB_GET_TIMELINE_NAME
 #endif
 
-- 
2.25.1

Re: [PATCH v11 1/2] drm/amdgpu: add debugfs for reset registers list

[Christian König]

Am 22.02.22 um 15:37 schrieb Somalapuram Amaranath:
> List of register populated for dump collection during the GPU reset.
>
> Signed-off-by: Somalapuram Amaranath <Amaranath.Somalapuram@amd.com>
> ---
>   drivers/gpu/drm/amd/amdgpu/amdgpu.h         |  4 +
>   drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 81 +++++++++++++++++++++
>   2 files changed, 85 insertions(+)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
> index b85b67a88a3d..6e35f2c4c869 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
> @@ -1097,6 +1097,10 @@ struct amdgpu_device {
>   
>   	struct amdgpu_reset_control     *reset_cntl;
>   	uint32_t                        ip_versions[HW_ID_MAX][HWIP_MAX_INSTANCE];
> +
> +	/* reset dump register */
> +	uint32_t			*reset_dump_reg_list;
> +	int                             num_regs;
>   };
>   
>   static inline struct amdgpu_device *drm_to_adev(struct drm_device *ddev)
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
> index 164d6a9e9fbb..df6d9fb69657 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
> @@ -1609,6 +1609,85 @@ DEFINE_DEBUGFS_ATTRIBUTE(fops_ib_preempt, NULL,
>   DEFINE_DEBUGFS_ATTRIBUTE(fops_sclk_set, NULL,
>   			amdgpu_debugfs_sclk_set, "%llu\n");
>   
> +static ssize_t amdgpu_reset_dump_register_list_read(struct file *f,
> +				char __user *buf, size_t size, loff_t *pos)
> +{
> +	struct amdgpu_device *adev = (struct amdgpu_device *)file_inode(f)->i_private;
> +	char reg_offset[11];
> +	int i, ret, len = 0;
> +
> +	if (*pos)
> +		return 0;
> +
> +	ret = down_read_killable(&adev->reset_sem);
> +	if (ret)
> +		return ret;
> +
> +	for (i = 0; i < adev->num_regs; i++) {
> +		sprintf(reg_offset, "0x%x\n", adev->reset_dump_reg_list[i]);
> +		up_read(&adev->reset_sem);
> +		ret = copy_to_user(buf + len, reg_offset, strlen(reg_offset));
> +		if (ret)
> +			return ret;

You need to return -EFAULT here, not the return value of copy_to_user() 
cause that are the number of bytes which couldn't be copied.

> +
> +		len += strlen(reg_offset);
> +		ret = down_read_killable(&adev->reset_sem);
> +		if (ret)
> +			return ret;
> +	}
> +
> +	up_read(&adev->reset_sem);
> +	if (ret)
> +		return ret;
> +
> +	*pos += len;
> +
> +	return len;
> +}
> +
> +static ssize_t amdgpu_reset_dump_register_list_write(struct file *f,
> +			const char __user *buf, size_t size, loff_t *pos)
> +{
> +	struct amdgpu_device *adev = (struct amdgpu_device *)file_inode(f)->i_private;
> +	char reg_offset[12];
> +	uint32_t *tmp;
> +	int ret, i = 0, len = 0;
> +
> +	do {
> +		memset(reg_offset, 0, 12);
> +		ret = copy_from_user(reg_offset, buf + len, min(11, ((int)size-len)));
> +		if (ret)
> +			goto error_free;

Same here.

> +
> +		tmp = krealloc_array(tmp, i + 1, sizeof(uint32_t), GFP_KERNEL);
> +		if (sscanf(reg_offset, "%X %n", &tmp[i], &ret) != 1)
> +			goto error_free;
> +
> +		len += ret;
> +		i++;
> +	} while (len < size);
> +
> +	ret = down_write_killable(&adev->reset_sem);
> +	if (ret)
> +		goto error_free;
> +
> +	swap(adev->reset_dump_reg_list, tmp);
> +	adev->num_regs = i;
> +	up_write(&adev->reset_sem);
> +	ret = size;
> +
> +error_free:
> +	kfree(tmp);
> +	return ret;

> +}
> +
> +static const struct file_operations amdgpu_reset_dump_register_list = {
> +	.owner = THIS_MODULE,
> +	.read = amdgpu_reset_dump_register_list_read,
> +	.write = amdgpu_reset_dump_register_list_write,
> +	.llseek = default_llseek
> +};
> +
>   int amdgpu_debugfs_init(struct amdgpu_device *adev)
>   {
>   	struct dentry *root = adev_to_drm(adev)->primary->debugfs_root;
> @@ -1672,6 +1751,8 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev)
>   			    &amdgpu_debugfs_test_ib_fops);
>   	debugfs_create_file("amdgpu_vm_info", 0444, root, adev,
>   			    &amdgpu_debugfs_vm_info_fops);
> +	debugfs_create_file("amdgpu_reset_dump_register_list", 0644, root, adev,
> +			    &amdgpu_reset_dump_register_list);
>   
>   	adev->debugfs_vbios_blob.data = adev->bios;
>   	adev->debugfs_vbios_blob.size = adev->bios_size;

Re: [PATCH v11 2/2] drm/amdgpu: add reset register dump trace on GPU

[Andrey Grodzovsky]

Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>

Andrey

On 2022-02-22 09:37, Somalapuram Amaranath wrote:
> Dump the list of register values to trace event on GPU reset.
>
> Signed-off-by: Somalapuram Amaranath <Amaranath.Somalapuram@amd.com>
> ---
>   drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 17 +++++++++++++++++
>   drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h  | 16 ++++++++++++++++
>   2 files changed, 33 insertions(+)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> index 1e651b959141..7c48fd716adb 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> @@ -4534,6 +4534,22 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
>   	return r;
>   }
>   
> +static int amdgpu_reset_reg_dumps(struct amdgpu_device *adev)
> +{
> +	uint32_t reg_value;
> +	int i;
> +
> +	lockdep_assert_held(&adev->reset_sem);
> +	dump_stack();
> +
> +	for (i = 0; i < adev->num_regs; i++) {
> +		reg_value = RREG32(adev->reset_dump_reg_list[i]);
> +		trace_amdgpu_reset_reg_dumps(adev->reset_dump_reg_list[i], reg_value);
> +	}
> +
> +	return 0;
> +}
> +
>   int amdgpu_do_asic_reset(struct list_head *device_list_handle,
>   			 struct amdgpu_reset_context *reset_context)
>   {
> @@ -4544,6 +4560,7 @@ int amdgpu_do_asic_reset(struct list_head *device_list_handle,
>   	/* Try reset handler method first */
>   	tmp_adev = list_first_entry(device_list_handle, struct amdgpu_device,
>   				    reset_list);
> +	amdgpu_reset_reg_dumps(tmp_adev);
>   	r = amdgpu_reset_perform_reset(tmp_adev, reset_context);
>   	/* If reset handler not implemented, continue; otherwise return */
>   	if (r == -ENOSYS)
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
> index d855cb53c7e0..b9637925e85c 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
> @@ -537,6 +537,22 @@ TRACE_EVENT(amdgpu_ib_pipe_sync,
>   		      __entry->seqno)
>   );
>   
> +TRACE_EVENT(amdgpu_reset_reg_dumps,
> +	    TP_PROTO(uint32_t address, uint32_t value),
> +	    TP_ARGS(address, value),
> +	    TP_STRUCT__entry(
> +			     __field(uint32_t, address)
> +			     __field(uint32_t, value)
> +			     ),
> +	    TP_fast_assign(
> +			   __entry->address = address;
> +			   __entry->value = value;
> +			   ),
> +	    TP_printk("amdgpu register dump 0x%x: 0x%x",
> +		      __entry->address,
> +		      __entry->value)
> +);
> +
>   #undef AMDGPU_JOB_GET_TIMELINE_NAME
>   #endif
>