Merge tag 'safesetid-5.13' of git://github.com/micah-morton/linux

[linux.git] / drivers / gpu / drm / amd / amdgpu / amdgpu_ras.c

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 0541196ae1ed80c9358e01fd4b74cddf187db207..b0d2fc9454caadb0d5e5410a63022406c01ac835 100644 (file)
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -114,7 +114,7 @@ static int amdgpu_reserve_page_direct(struct amdgpu_device *adev, uint64_t addre
 
        if (amdgpu_ras_check_bad_page(adev, address)) {
                dev_warn(adev->dev,
 
        if (amdgpu_ras_check_bad_page(adev, address)) {
                dev_warn(adev->dev,

-                        "RAS WARN: 0x%llx has been marked as bad page!\n",
+                        "RAS WARN: 0x%llx has already been marked as bad page!\n",

                         address);
                return 0;
        }
                         address);
                return 0;
        }


@@ -221,18 +221,17 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f,
                op = 1;
        else if (sscanf(str, "inject %32s %8s", block_name, err) == 2)
                op = 2;
                op = 1;
        else if (sscanf(str, "inject %32s %8s", block_name, err) == 2)
                op = 2;

-       else if (sscanf(str, "retire_page") == 0)
+       else if (strstr(str, "retire_page") != NULL)

                op = 3;
        else if (str[0] && str[1] && str[2] && str[3])
                /* ascii string, but commands are not matched. */
                return -EINVAL;
 
        if (op != -1) {
                op = 3;
        else if (str[0] && str[1] && str[2] && str[3])
                /* ascii string, but commands are not matched. */
                return -EINVAL;
 
        if (op != -1) {

-

                if (op == 3) {
                if (op == 3) {

-                       if (sscanf(str, "%*s %llu", &address) != 1)
-                               if (sscanf(str, "%*s 0x%llx", &address) != 1)
-                                       return -EINVAL;
+                       if (sscanf(str, "%*s 0x%llx", &address) != 1 &&
+                           sscanf(str, "%*s %llu", &address) != 1)
+                               return -EINVAL;

 
                        data->op = op;
                        data->inject.address = address;
 
                        data->op = op;
                        data->inject.address = address;


@@ -255,11 +254,11 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f,
                data->op = op;
 
                if (op == 2) {
                data->op = op;
 
                if (op == 2) {

-                       if (sscanf(str, "%*s %*s %*s %u %llu %llu",
-                                               &sub_block, &address, &value) != 3)
-                               if (sscanf(str, "%*s %*s %*s 0x%x 0x%llx 0x%llx",
-                                                       &sub_block, &address, &value) != 3)
-                                       return -EINVAL;
+                       if (sscanf(str, "%*s %*s %*s 0x%x 0x%llx 0x%llx",
+                                  &sub_block, &address, &value) != 3 &&
+                           sscanf(str, "%*s %*s %*s %u %llu %llu",
+                                  &sub_block, &address, &value) != 3)
+                               return -EINVAL;

                        data->head.sub_block_index = sub_block;
                        data->inject.address = address;
                        data->inject.value = value;
                        data->head.sub_block_index = sub_block;
                        data->inject.address = address;
                        data->inject.value = value;


@@ -278,7 +277,7 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f,
 /**
  * DOC: AMDGPU RAS debugfs control interface
  *
 /**
  * DOC: AMDGPU RAS debugfs control interface
  *

- * It accepts struct ras_debug_if who has two members.
+ * The control interface accepts struct ras_debug_if which has two members.

  *
  * First member: ras_debug_if::head or ras_debug_if::inject.
  *
  *
  * First member: ras_debug_if::head or ras_debug_if::inject.
  *


@@ -303,32 +302,33 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f,
  *
  * How to use the interface?
  *
  *
  * How to use the interface?
  *

- * Programs
+ * In a program

  *
  *

- * Copy the struct ras_debug_if in your codes and initialize it.
- * Write the struct to the control node.
+ * Copy the struct ras_debug_if in your code and initialize it.
+ * Write the struct to the control interface.

  *
  *

- * Shells
+ * From shell

  *
  * .. code-block:: bash
  *
  *
  * .. code-block:: bash
  *

- *     echo op block [error [sub_block address value]] > .../ras/ras_ctrl
+ *     echo "disable <block>" > /sys/kernel/debug/dri/<N>/ras/ras_ctrl
+ *     echo "enable  <block> <error>" > /sys/kernel/debug/dri/<N>/ras/ras_ctrl
+ *     echo "inject  <block> <error> <sub-block> <address> <value> > /sys/kernel/debug/dri/<N>/ras/ras_ctrl

  *
  *

- * Parameters:
+ * Where N, is the card which you want to affect.

  *
  *

- * op: disable, enable, inject
- *     disable: only block is needed
- *     enable: block and error are needed
- *     inject: error, address, value are needed
- * block: umc, sdma, gfx, .........
+ * "disable" requires only the block.
+ * "enable" requires the block and error type.
+ * "inject" requires the block, error type, address, and value.
+ * The block is one of: umc, sdma, gfx, etc.

  *     see ras_block_string[] for details
  *     see ras_block_string[] for details

- * error: ue, ce
- *     ue: multi_uncorrectable
- *     ce: single_correctable
- * sub_block:
- *     sub block index, pass 0 if there is no sub block
+ * The error type is one of: ue, ce, where,
+ *     ue is multi-uncorrectable
+ *     ce is single-correctable
+ * The sub-block is a the sub-block index, pass 0 if there is no sub-block.
+ * The address and value are hexadecimal numbers, leading 0x is optional.

  *
  *

- * here are some examples for bash commands:
+ * For instance,

  *
  * .. code-block:: bash
  *
  *
  * .. code-block:: bash
  *


@@ -336,17 +336,17 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f,
  *     echo inject umc ce 0 0 0 > /sys/kernel/debug/dri/0/ras/ras_ctrl
  *     echo disable umc > /sys/kernel/debug/dri/0/ras/ras_ctrl
  *
  *     echo inject umc ce 0 0 0 > /sys/kernel/debug/dri/0/ras/ras_ctrl
  *     echo disable umc > /sys/kernel/debug/dri/0/ras/ras_ctrl
  *

- * How to check the result?
+ * How to check the result of the operation?

  *
  *

- * For disable/enable, please check ras features at
+ * To check disable/enable, see "ras" features at,

  * /sys/class/drm/card[0/1/2...]/device/ras/features
  *
  * /sys/class/drm/card[0/1/2...]/device/ras/features
  *

- * For inject, please check corresponding err count at
- * /sys/class/drm/card[0/1/2...]/device/ras/[gfx/sdma/...]_err_count
+ * To check inject, see the corresponding error count at,
+ * /sys/class/drm/card[0/1/2...]/device/ras/[gfx|sdma|umc|...]_err_count

  *
  * .. note::
  *     Operations are only allowed on blocks which are supported.
  *
  * .. note::
  *     Operations are only allowed on blocks which are supported.

- *     Please check ras mask at /sys/module/amdgpu/parameters/ras_mask
+ *     Check the "ras" mask at /sys/module/amdgpu/parameters/ras_mask

  *     to see which blocks support RAS on a particular asic.
  *
  */
  *     to see which blocks support RAS on a particular asic.
  *
  */


@@ -367,11 +367,9 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, const char __user *
        if (ret)
                return -EINVAL;
 
        if (ret)
                return -EINVAL;
 

-       if (data.op == 3)
-       {
+       if (data.op == 3) {

                ret = amdgpu_reserve_page_direct(adev, data.inject.address);
                ret = amdgpu_reserve_page_direct(adev, data.inject.address);

-
-               if (ret)
+               if (!ret)

                        return size;
                else
                        return ret;
                        return size;
                else
                        return ret;


@@ -503,6 +501,12 @@ static ssize_t amdgpu_ras_sysfs_read(struct device *dev,
        if (amdgpu_ras_query_error_status(obj->adev, &info))
                return -EINVAL;
 
        if (amdgpu_ras_query_error_status(obj->adev, &info))
                return -EINVAL;
 

+
+       if (obj->adev->asic_type == CHIP_ALDEBARAN) {
+               if (amdgpu_ras_reset_error_status(obj->adev, info.head.block))
+                       DRM_WARN("Failed to reset error counter and error status");
+       }
+

        return sysfs_emit(buf, "%s: %lu\n%s: %lu\n", "ue", info.ue_count,
                          "ce", info.ce_count);
 }
        return sysfs_emit(buf, "%s: %lu\n%s: %lu\n", "ue", info.ue_count,
                          "ce", info.ce_count);
 }


@@ -1269,6 +1273,8 @@ static struct dentry *amdgpu_ras_debugfs_create_ctrl_node(struct amdgpu_device *
                            &amdgpu_ras_debugfs_ctrl_ops);
        debugfs_create_file("ras_eeprom_reset", S_IWUGO | S_IRUGO, dir, adev,
                            &amdgpu_ras_debugfs_eeprom_ops);
                            &amdgpu_ras_debugfs_ctrl_ops);
        debugfs_create_file("ras_eeprom_reset", S_IWUGO | S_IRUGO, dir, adev,
                            &amdgpu_ras_debugfs_eeprom_ops);

+       debugfs_create_u32("bad_page_cnt_threshold", 0444, dir,
+                          &con->bad_page_cnt_threshold);

 
        /*
         * After one uncorrectable error happens, usually GPU recovery will
 
        /*
         * After one uncorrectable error happens, usually GPU recovery will