On 2/5/2026 9:00 PM, Raag Jadav wrote:
On Mon, Feb 02, 2026 at 12:14:00PM +0530, Riana Tauro wrote:
PVC supports GT error reporting via vector registers along with
error status register. Add support to report these errors and
update respective counters. Incase of Subslice error reported
by vector register, process the error status register
for applicable bits.
The counter is embedded in the xe drm ras structure and is
exposed to the userspace using the drm_ras generic netlink
interface.
$ sudo ynl --family drm_ras --do query-error-counter --json
We usually add '\' at the end for wrapping commands so that they're easy
to apply directly (and same for all other patches where applicable).
'{"node-id":0, "error-id":1}'
Ditto.
Will fix this
{'error-id': 1, 'error-name': 'core-compute', 'error-value': 0}
Co-developed-by: Himal Prasad Ghimiray <[email protected]>
Signed-off-by: Himal Prasad Ghimiray <[email protected]>
Signed-off-by: Riana Tauro <[email protected]>
---
v2: Add ID's and names as uAPI (Rodrigo)
Add documentation
Modify commit message
v3: remove 'error' from counters
use drmm_kcalloc
add a for_each for severity
differentitate error classes and severity in UAPI(Raag)
Use correctable and uncorrectable in uapi (Pratik / Aravind)
v4: modify enums in UAPI
improve comments
add bounds check in handler
add error mask macro (Raag)
use atomic_t
add null pointer checks
---
drivers/gpu/drm/xe/regs/xe_hw_error_regs.h | 62 ++++++-
drivers/gpu/drm/xe/xe_hw_error.c | 199 +++++++++++++++++++--
2 files changed, 241 insertions(+), 20 deletions(-)
diff --git a/drivers/gpu/drm/xe/regs/xe_hw_error_regs.h
b/drivers/gpu/drm/xe/regs/xe_hw_error_regs.h
index c146b9ef44eb..17982a335941 100644
--- a/drivers/gpu/drm/xe/regs/xe_hw_error_regs.h
+++ b/drivers/gpu/drm/xe/regs/xe_hw_error_regs.h
@@ -6,15 +6,59 @@
#ifndef _XE_HW_ERROR_REGS_H_
#define _XE_HW_ERROR_REGS_H_
-#define HEC_UNCORR_ERR_STATUS(base) XE_REG((base) + 0x118)
-#define UNCORR_FW_REPORTED_ERR BIT(6)
+#define HEC_UNCORR_ERR_STATUS(base) XE_REG((base) + 0x118)
+#define UNCORR_FW_REPORTED_ERR REG_BIT(6)
-#define HEC_UNCORR_FW_ERR_DW0(base) XE_REG((base) + 0x124)
+#define HEC_UNCORR_FW_ERR_DW0(base) XE_REG((base) + 0x124)
+
+#define ERR_STAT_GT_COR 0x100160
+#define EU_GRF_COR_ERR REG_BIT(15)
+#define EU_IC_COR_ERR REG_BIT(14)
+#define SLM_COR_ERR REG_BIT(13)
+#define GUC_COR_ERR REG_BIT(1)
+
+#define ERR_STAT_GT_NONFATAL 0x100164
+#define ERR_STAT_GT_FATAL 0x100168
+#define EU_GRF_FAT_ERR REG_BIT(15)
+#define SLM_FAT_ERR REG_BIT(13)
+#define GUC_FAT_ERR REG_BIT(6)
+#define FPU_FAT_ERR REG_BIT(3)
+
+#define ERR_STAT_GT_REG(x) XE_REG(_PICK_EVEN((x), \
+
ERR_STAT_GT_COR, \
+
ERR_STAT_GT_NONFATAL))
+
+#define PVC_COR_ERR_MASK (GUC_COR_ERR |
SLM_COR_ERR | \
+ EU_IC_COR_ERR |
EU_GRF_COR_ERR)
+
+#define PVC_FAT_ERR_MASK (FPU_FAT_ERR |
GUC_FAT_ERR | \
+ EU_GRF_FAT_ERR |
SLM_FAT_ERR)
Nit: Whitespace please!
alignment?
+#define DEV_ERR_STAT_NONFATAL 0x100178
+#define DEV_ERR_STAT_CORRECTABLE 0x10017c
+#define DEV_ERR_STAT_REG(x) XE_REG(_PICK_EVEN((x), \
+
DEV_ERR_STAT_CORRECTABLE, \
+
DEV_ERR_STAT_NONFATAL))
I know it was already like this but how does this evaluate for FATAL?
#define _PICK_EVEN(__index, __a, __b) ((__a) + (__index) * ((__b) - (__a)))
(index, 0x10017c, 0x100178) = (0x10017c + index * (0x100178 - 0x10017c));
0 = 0x10017c
1 = 0x100178
2 = 0x100174
+#define XE_CSC_ERROR 17
+#define XE_GT_ERROR 0
+
+#define ERR_STAT_GT_FATAL_VECTOR_0 0x100260
+#define ERR_STAT_GT_FATAL_VECTOR_1 0x100264
+
+#define ERR_STAT_GT_FATAL_VECTOR_REG(x)
XE_REG(_PICK_EVEN((x), \
+
ERR_STAT_GT_FATAL_VECTOR_0, \
+
ERR_STAT_GT_FATAL_VECTOR_1))
+
+#define ERR_STAT_GT_COR_VECTOR_0 0x1002a0
+#define ERR_STAT_GT_COR_VECTOR_1 0x1002a4
+
+#define ERR_STAT_GT_COR_VECTOR_REG(x) XE_REG(_PICK_EVEN((x), \
+
ERR_STAT_GT_COR_VECTOR_0, \
+
ERR_STAT_GT_COR_VECTOR_1))
+
+#define ERR_STAT_GT_VECTOR_REG(hw_err, x) (hw_err ==
HARDWARE_ERROR_CORRECTABLE ? \
+
ERR_STAT_GT_COR_VECTOR_REG(x) : \
+
ERR_STAT_GT_FATAL_VECTOR_REG(x))
Ditto for whitespace.
-#define DEV_ERR_STAT_NONFATAL 0x100178
-#define DEV_ERR_STAT_CORRECTABLE 0x10017c
-#define DEV_ERR_STAT_REG(x) XE_REG(_PICK_EVEN((x), \
-
DEV_ERR_STAT_CORRECTABLE, \
-
DEV_ERR_STAT_NONFATAL))
-#define XE_CSC_ERROR BIT(17)
#endif
diff --git a/drivers/gpu/drm/xe/xe_hw_error.c b/drivers/gpu/drm/xe/xe_hw_error.c
index 2019aaaa1ebe..ff31fb322c8a 100644
--- a/drivers/gpu/drm/xe/xe_hw_error.c
+++ b/drivers/gpu/drm/xe/xe_hw_error.c
@@ -3,6 +3,7 @@
* Copyright © 2025 Intel Corporation
*/
+#include <linux/bitmap.h>
#include <linux/fault-inject.h>
#include "regs/xe_gsc_regs.h"
@@ -15,7 +16,13 @@
#include "xe_mmio.h"
#include "xe_survivability_mode.h"
-#define HEC_UNCORR_FW_ERR_BITS 4
+#define GT_HW_ERROR_MAX_ERR_BITS 16
+#define HEC_UNCORR_FW_ERR_BITS 4
+#define XE_RAS_REG_SIZE 32
+
+#define PVC_ERROR_MASK_SET(hw_err, err_bit) \
+ ((hw_err == HARDWARE_ERROR_CORRECTABLE) ? (BIT(err_bit) &
PVC_COR_ERR_MASK) : \
+ (BIT(err_bit) & PVC_FAT_ERR_MASK))
I'd write this as below and move it to xe_hw_error_regs.h
This is not specific to register selection or defining bits. It's
related to mask. So .c should be the right place
#define PVC_COR_ERR_MASK_SET(err_bit) (PVC_COR_ERR_MASK &
REG_BIT(err_bit))
#define PVC_FAT_ERR_MASK_SET(err_bit) (PVC_FAT_ERR_MASK &
REG_BIT(err_bit))
#define PVC_ERR_MASK_SET(hw_err, err_bit) ((hw_err ==
HARDWARE_ERROR_CORRECTABLE) ? \
PVC_COR_ERR_MASK_SET(err_bit) : \
PVC_FAT_ERR_MASK_SET(err_bit)
...
+static void gt_hw_error_handler(struct xe_tile *tile, const enum
hardware_error hw_err,
+ u32 error_id)
+{
+ const enum drm_xe_ras_error_severity severity =
hw_err_to_severity(hw_err);
+ struct xe_device *xe = tile_to_xe(tile);
+ struct xe_drm_ras *ras = &xe->ras;
+ struct xe_drm_ras_counter *info = ras->info[severity];
+ struct xe_mmio *mmio = &tile->mmio;
+ unsigned long err_stat = 0;
+ int i, len;
+
+ if (xe->info.platform != XE_PVC)
+ return;
+
+ if (!info)
+ return;
Since info allocation is not related to hardware, we shouldn't even be
at this point without it. So let's not hide bugs and fail probe instead.
yes currently it is supported only on PVC. I can remove this here as
there is a PVC check but cannot remove the one suggested below.
Thanks
Riana
+ if (hw_err == HARDWARE_ERROR_NONFATAL) {
+ atomic_inc(&info[error_id].counter);
+ log_hw_error(tile, info[error_id].name, severity);
+ return;
+ }
...
static void hw_error_source_handler(struct xe_tile *tile, const enum
hardware_error hw_err)
{
const enum drm_xe_ras_error_severity severity =
hw_err_to_severity(hw_err);
const char *severity_str = error_severity[severity];
struct xe_device *xe = tile_to_xe(tile);
- unsigned long flags;
- u32 err_src;
+ struct xe_drm_ras *ras = &xe->ras;
+ struct xe_drm_ras_counter *info = ras->info[severity];
+ unsigned long flags, err_src;
+ u32 err_bit;
- if (xe->info.platform != XE_BATTLEMAGE)
+ if (!IS_DGFX(xe))
return;
spin_lock_irqsave(&xe->irq.lock, flags);
@@ -108,11 +242,53 @@ static void hw_error_source_handler(struct xe_tile *tile,
const enum hardware_er
goto unlock;
}
- if (err_src & XE_CSC_ERROR)
+ /*
+ * On encountering CSC firmware errors, the graphics device becomes
unrecoverable
+ * so return immediately on error. The only way to recover from these
errors is
+ * firmware flash. The device will enter Runtime Survivability mode
when such
+ * errors are detected.
+ */
+ if (err_src & XE_CSC_ERROR) {
csc_hw_error_handler(tile, hw_err);
+ goto clear_reg;
+ }
- xe_mmio_write32(&tile->mmio, DEV_ERR_STAT_REG(hw_err), err_src);
+ if (!info)
+ goto clear_reg;
Same as above.
Raag
