Reviewed-By: Kent Russell <kent.russell@xxxxxxx>
Kent
-----Original Message-----
From: amd-gfx <amd-gfx-bounces@xxxxxxxxxxxxxxxxxxxxx> On Behalf Of Kim, Jonathan
Sent: Tuesday, July 16, 2019 1:12 PM
To: amd-gfx@xxxxxxxxxxxxxxxxxxxxx
Subject: RE: [PATCH 3/3] drm/amdgpu: adding xgmi error monitoring
Note patch 1 sent to internal amd mailing list for review.
-----Original Message-----
From: Kim, Jonathan <Jonathan.Kim@xxxxxxx>
Sent: Tuesday, July 16, 2019 1:09 PM
To: amd-gfx@xxxxxxxxxxxxxxxxxxxxx
Cc: Kim, Jonathan <Jonathan.Kim@xxxxxxx>; Kim, Jonathan <Jonathan.Kim@xxxxxxx>
Subject: [PATCH 3/3] drm/amdgpu: adding xgmi error monitoring
monitor xgmi errors via mc pie status through fica registers.
Change-Id: Id80b6c2f635a294afe343cf55a03902e9a1787a5
Signed-off-by: Jonathan Kim <Jonathan.Kim@xxxxxxx>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c | 38 ++++++++++++++++++++++--
1 file changed, 36 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
index d11eba09eadd..4b87fda15ac5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
@@ -25,7 +25,7 @@
#include "amdgpu.h"
#include "amdgpu_xgmi.h"
#include "amdgpu_smu.h"
-
+#include "df/df_3_6_offset.h"
static DEFINE_MUTEX(xgmi_mutex);
@@ -131,9 +131,37 @@ static ssize_t amdgpu_xgmi_show_device_id(struct device *dev,
}
+#define AMDGPU_XGMI_SET_FICAA(o) ((o) | 0x456801)
+static ssize_t amdgpu_xgmi_show_error(struct device *dev,
+ struct device_attribute *attr,
+ char *buf)
+{
+ struct drm_device *ddev = dev_get_drvdata(dev);
+ struct amdgpu_device *adev = ddev->dev_private;
+ uint32_t ficaa_pie_ctl_in, ficaa_pie_status_in;
+ uint64_t fica_out;
+ unsigned int error_count = 0;
+
+ ficaa_pie_ctl_in = AMDGPU_XGMI_SET_FICAA(0x200);
+ ficaa_pie_status_in = AMDGPU_XGMI_SET_FICAA(0x208);
-static DEVICE_ATTR(xgmi_device_id, S_IRUGO, amdgpu_xgmi_show_device_id, NULL);
+ fica_out = adev->df_funcs->get_fica(adev, ficaa_pie_ctl_in);
+ if (fica_out != 0x1f)
+ pr_err("xGMI error counters not enabled!\n");
+
+ fica_out = adev->df_funcs->get_fica(adev, ficaa_pie_status_in);
+
+ if ((fica_out & 0xffff) == 2)
+ error_count = ((fica_out >> 62) & 0x1) + (fica_out >> 63);
+ adev->df_funcs->set_fica(adev, ficaa_pie_status_in, 0, 0);
+
+ return snprintf(buf, PAGE_SIZE, "%d\n", error_count); }
+
+
+static DEVICE_ATTR(xgmi_device_id, S_IRUGO, amdgpu_xgmi_show_device_id,
+NULL); static DEVICE_ATTR(xgmi_error, S_IRUGO, amdgpu_xgmi_show_error,
+NULL);
static int amdgpu_xgmi_sysfs_add_dev_info(struct amdgpu_device *adev,
struct amdgpu_hive_info *hive)
@@ -148,6 +176,12 @@ static int amdgpu_xgmi_sysfs_add_dev_info(struct amdgpu_device *adev,
return ret;
}
+ /* Create xgmi error file */
+ ret = device_create_file(adev->dev, &dev_attr_xgmi_error);
+ if (ret)
+ pr_err("failed to create xgmi_error\n");
+
+
/* Create sysfs link to hive info folder on the first device */
if (adev != hive->adev) {
ret = sysfs_create_link(&adev->dev->kobj, hive->kobj,
--
2.17.1
_______________________________________________
amd-gfx mailing list
amd-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
_______________________________________________
amd-gfx mailing list
amd-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
