Skip to contentSkip to conversation

drm/i915/uc: Extra info notice about FW version mis-match vs overrides

PATCH In review [email protected] 2 participants 7 messages Nov 21, 2019 – Dec 11, 2019

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 drm/i915/uc: Extra info notice about FW version mis-match vs overrides

From: John Harrison <[email protected]>

If a FW override is present then a version mis-match is actually ignored. The warning message was still being printed, though. Which could confuse people by implying that the load had failed due to the mis-match when actually something else had failed.

This patch adds an extra message to say that the mis-match is being ignored if an override is present.

---
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +3 -0
index 66a30ab7044a..c1ae807b07ae 100644
@@ -361,6 +361,9 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw, struct drm_i915_private *i915)
 			err = -ENOEXEC;
 			goto fail;
 		}
+
+		dev_notice(dev, "%s firmware %s: Firmware override so ignoring version mis-match\n",
+			   intel_uc_fw_type_repr(uc_fw->type), uc_fw->path);
 	}
 
 	obj = i915_gem_object_create_shmem_from_data(i915, fw->data, fw->size);
Signature · 6 lines
2.21.0.5.gaeb582a983

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#1 lore raw
Quoted 4 lines · On Thu, 21 Nov 2019 01:27:23 +0100, <[email protected]> wrote:

From: John Harrison <[email protected]>

If a FW override is present then a version mis-match is actually ignored. The warning message was still being printed, though. Which

It wasn't a "warning", just "notice"

Quoted 2 lines

could confuse people by implying that the load had failed due to the mis-match when actually something else had failed.

The mis-match still might be a reason why something else failed. If there is possible confusing, it's likely due to lack of or incomplete message from this other failure point. So we should make sure that all failure points correctly indicate the failure reason to avoid confusing. Do you recall what this other confusing failure was?

Quoted 14 lines

This patch adds an extra message to say that the mis-match is being ignored if an override is present.

---
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +0 -0
index 66a30ab7044a..c1ae807b07ae 100644

drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw, · line 368

Quoted 8 lines
drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +3 -0
@@ -361,6 +361,9 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw,  
 struct drm_i915_private *i915)
 			err = -ENOEXEC;
 			goto fail;
 		}
+
+		dev_notice(dev, "%s firmware %s: Firmware override so ignoring  
 version mis-match\n",
+			   intel_uc_fw_type_repr(uc_fw->type), uc_fw->path);

If you still want to include clear statement about mis-match being ignored, then maybe instead of adding new message it could be combined with old one:

        if (uc_fw->major_ver_found != uc_fw->major_ver_wanted ||
            uc_fw->minor_ver_found < uc_fw->minor_ver_wanted) {
-               dev_notice(dev, "%s firmware %s: unexpected version: %u.%u != %u.%u\n",
+               dev_notice(dev, "%s firmware %s: %s: %u.%u != %u.%u\n",
                           intel_uc_fw_type_repr(uc_fw->type), uc_fw->path,
+                          intel_uc_fw_is_overridden(uc_fw) ?
+                          "ignoring unexpected version" : "wrong version",
                           uc_fw->major_ver_found, uc_fw->minor_ver_found,
                           uc_fw->major_ver_wanted,  
uc_fw->minor_ver_wanted);
                if (!intel_uc_fw_is_overridden(uc_fw)) {

Michal
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#2 lore raw
Quoted 9 lines · Michal Wajdeczko
Quoted 4 lines · On Thu, 21 Nov 2019 01:27:23 +0100, <[email protected]> wrote:

From: John Harrison <[email protected]>

If a FW override is present then a version mis-match is actually ignored. The warning message was still being printed, though. Which

It wasn't a "warning", just "notice"

Quoted 8 lines
Quoted 2 lines

could confuse people by implying that the load had failed due to the mis-match when actually something else had failed.

The mis-match still might be a reason why something else failed. If there is possible confusing, it's likely due to lack of or incomplete message from this other failure point. So we should make sure that all failure points correctly indicate the failure reason to avoid confusing. Do you recall what this other confusing failure was?

Sorry, bad commit message comment. The point was that you could successfully load the GuC FW but then something entirely unrelated fails (with or without appropriate error message). However, the first apparent failure in dmesg is the GuC version mis-match. Therefore a user (or even developer) might assume that all subsequent issues are caused by the FW mismatch causing the GuC to not load at all and hence not investigate any later messages until this first one is fixed/understood.

So the point is just to avoid people wasting time investigating something that is not actually an error.

Quoted 10 lines
Quoted 10 lines

This patch adds an extra message to say that the mis-match is being ignored if an override is present.

---
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 3 +++

 1 file changed, 3 insertions(+)
Quoted 17 lines

drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw, · line 368

Quoted 13 lines
drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +3 -0
index 66a30ab7044a..c1ae807b07ae 100644
@@ -361,6 +361,9 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw, 
 struct drm_i915_private *i915)
              err = -ENOEXEC;
              goto fail;
          }
+
+        dev_notice(dev, "%s firmware %s: Firmware override so 
 ignoring version mis-match\n",
+               intel_uc_fw_type_repr(uc_fw->type), uc_fw->path);

If you still want to include clear statement about mis-match being ignored, then maybe instead of adding new message it could be combined with old one:

Or maybe just move the mis-match notice into the 'goto fail' section and not print anything at all in the case of an override. On the grounds that if someone is specifying an override then it is almost certainly because the default version is not what they want. So yes, it obviously is going to be mis-match.

Quoted 13 lines

if (uc_fw->major_ver_found != uc_fw->major_ver_wanted || uc_fw->minor_ver_found < uc_fw->minor_ver_wanted) {
-               dev_notice(dev, "%s firmware %s: unexpected version:
%u.%u != %u.%u\n",
+               dev_notice(dev, "%s firmware %s: %s: %u.%u != %u.%u\n", intel_uc_fw_type_repr(uc_fw->type), uc_fw->path, +                          intel_uc_fw_is_overridden(uc_fw) ?
+                          "ignoring unexpected version" : "wrong version",
                           uc_fw->major_ver_found, uc_fw->minor_ver_found, uc_fw->major_ver_wanted, uc_fw->minor_ver_wanted); if (!intel_uc_fw_is_overridden(uc_fw)) {

Quoted 1 line

Michal

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#3 lore raw
Quoted 13 lines · John Harrison
Quoted 12 lines · Michal Wajdeczko
Quoted 4 lines · On Thu, 21 Nov 2019 01:27:23 +0100, <[email protected]> wrote:

From: John Harrison <[email protected]>

If a FW override is present then a version mis-match is actually ignored. The warning message was still being printed, though. Which

It wasn't a "warning", just "notice"

Quoted 2 lines

could confuse people by implying that the load had failed due to the mis-match when actually something else had failed.

Quoted 7 lines
Quoted 5 lines

The mis-match still might be a reason why something else failed. If there is possible confusing, it's likely due to lack of or incomplete message from this other failure point. So we should make sure that all failure points correctly indicate the failure reason to avoid confusing. Do you recall what this other confusing failure was?

Sorry, bad commit message comment. The point was that you could successfully load the GuC FW but then something entirely unrelated fails (with or without appropriate error message). However, the first apparent failure in dmesg is the GuC version mis-match. Therefore a user (or even

again, it wasn't error message, just a notice to clearly indicate that fw version being used is not matching tested and guaranteed configuration.

Quoted 1 line

developer) might assume that all subsequent issues are caused by the FW mismatch causing the GuC to not load at all and hence not investigate

We fail to fetch mismatched fw only if user is trying to cheat with us (by renaming mismatched fw blob to expected filename). In case of override, fetch will continue, but we may fail due to hard to predict ABI incompatibility/breakage.

Quoted 1 line

any later messages until this first one is fixed/understood.

If there are any later errors, we have to account that these might be caused by mismatched fw, otherwise, we would use that other fw as default one.

Quoted 2 lines

So the point is just to avoid people wasting time investigating something that is not actually an error.

Ordinary user is never expected to see this notice.

Users/developers that override fw path must expect this notice. There is a still risk of potential issues caused by loading mismatched firmware that will not have proper error message.

Quoted 29 lines
Quoted 12 lines
Quoted 12 lines

This patch adds an extra message to say that the mis-match is being ignored if an override is present.

---
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 3 +++
 1 file changed, 3 insertions(+)
Quoted 17 lines

drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw, · line 368

Quoted 13 lines
drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +3 -0
index 66a30ab7044a..c1ae807b07ae 100644
@@ -361,6 +361,9 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw,  
 struct drm_i915_private *i915)
             err = -ENOEXEC;
             goto fail;
         }
+
+        dev_notice(dev, "%s firmware %s: Firmware override so  
 ignoring version mis-match\n",
+               intel_uc_fw_type_repr(uc_fw->type), uc_fw->path);

If you still want to include clear statement about mis-match being ignored, then maybe instead of adding new message it could be combined with old one:

Quoted 1 line

Or maybe just move the mis-match notice into the 'goto fail' section and not print anything at all in the case of an override. On the grounds that if someone is specifying an override then it is almost certainly because the default version is not what they want. So yes, it obviously is going to be mis-match.

You can safely (and silently) override fw according to fw version compatibility schema: same major and newer minor are OK to use.

This notice message was capturing cases when you try to bend the rules:
use mismatched major (as it is unlikely to work, since it indicates no backward compatibility on ABI level, otherwise major version should stay the same) or use of legacy minor (possibly will also not work due to missing feature that is expected/used by driver).

Michal

Quoted 18 lines
Quoted 18 lines

if (uc_fw->major_ver_found != uc_fw->major_ver_wanted ||

            uc_fw->minor_ver_found < uc_fw->minor_ver_wanted) {
-               dev_notice(dev, "%s firmware %s: unexpected version:  
%u.%u != %u.%u\n",
+               dev_notice(dev, "%s firmware %s: %s: %u.%u != %u.%u\n",
                           intel_uc_fw_type_repr(uc_fw->type),  

uc_fw->path,

(quoted diff) +2 -0
+                          intel_uc_fw_is_overridden(uc_fw) ?
+                          "ignoring unexpected version" : "wrong  

version",

                           uc_fw->major_ver_found,  
uc_fw->minor_ver_found,
                           uc_fw->major_ver_wanted,  
uc_fw->minor_ver_wanted);
                if (!intel_uc_fw_is_overridden(uc_fw)) {
Quoted 1 line
Quoted 1 line

Michal

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#4 lore raw

PATCH 1/1 drm/i915/uc: Don't complain about FW versions when overridden

In reply to [email protected]

From: John Harrison <[email protected]>

If a FW override is present then a version mis-match is actually ignored. The warning notice was still being printed, though. Which could confuse people by implying that the load had failed when it had actually succeeded. Given that the whole point of the override is to use different versions of FW, there isn't really much point reporting the mis-match.

So, only print the notice when actually failing the load and avoid any potential confusion.

v2: Original patch added a new 'ignore the previous notice' notice. Now it just suppresses the existing notice. Review feedback from Michal W.

CC Michal Wajdeczko <[email protected]>
---
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)
drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +5 -6
index 66a30ab7044a..aa1b7ad02b56 100644
@@ -351,16 +351,15 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw, struct drm_i915_private *i915)
 	uc_fw->minor_ver_found = FIELD_GET(CSS_SW_VERSION_UC_MINOR,
 					   css->sw_version);
 
-	if (uc_fw->major_ver_found != uc_fw->major_ver_wanted ||
-	    uc_fw->minor_ver_found < uc_fw->minor_ver_wanted) {
+	if ((uc_fw->major_ver_found != uc_fw->major_ver_wanted ||
+	     uc_fw->minor_ver_found < uc_fw->minor_ver_wanted) &&
+	    !intel_uc_fw_is_overridden(uc_fw)) {
 		dev_notice(dev, "%s firmware %s: unexpected version: %u.%u != %u.%u\n",
 			   intel_uc_fw_type_repr(uc_fw->type), uc_fw->path,
 			   uc_fw->major_ver_found, uc_fw->minor_ver_found,
 			   uc_fw->major_ver_wanted, uc_fw->minor_ver_wanted);
-		if (!intel_uc_fw_is_overridden(uc_fw)) {
-			err = -ENOEXEC;
-			goto fail;
-		}
+		err = -ENOEXEC;
+		goto fail;
 	}
 
 	obj = i915_gem_object_create_shmem_from_data(i915, fw->data, fw->size);
Signature · 6 lines
2.21.0.5.gaeb582a983

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
#5 lore raw

Re: [Intel-gfx] [PATCH] drm/i915/uc: Don't complain about FW versions when overridden

Quoted 9 lines · On Sat, 07 Dec 2019 02:01:55 +0100, <[email protected]> wrote:

From: John Harrison <[email protected]>

If a FW override is present then a version mis-match is actually ignored. The warning notice was still being printed, though. Which could confuse people by implying that the load had failed when it had actually succeeded. Given that the whole point of the override is to use different versions of FW, there isn't really much point reporting the mis-match.

Quoted 6 lines

So, only print the notice when actually failing the load and avoid any potential confusion.

v2: Original patch added a new 'ignore the previous notice' notice. Now it just suppresses the existing notice. Review feedback from Michal W.

This is not exactly from my previous feedback. I'm not sold that suppressing message at fw fetch step about loading incompatible fw (that may later fail to load) is a good idea. Note that loaded fw version will be only printed when we successfully load it (see intel_uc_init_hw). In case of any fw load failure (which is now more likely to be caused by fw/drv version mismatch) we will not print which fw version we attempted to load (except fw blob name, which is not guaranteed to follow major.minor naming schema).

Michal

Quoted 23 lines
CC Michal Wajdeczko <[email protected]>
---
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)
drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +3 -2
index 66a30ab7044a..aa1b7ad02b56 100644
@@ -351,16 +351,15 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw,  
 struct drm_i915_private *i915)
 	uc_fw->minor_ver_found = FIELD_GET(CSS_SW_VERSION_UC_MINOR,
 					   css->sw_version);
-	if (uc_fw->major_ver_found != uc_fw->major_ver_wanted ||
-	    uc_fw->minor_ver_found < uc_fw->minor_ver_wanted) {
+	if ((uc_fw->major_ver_found != uc_fw->major_ver_wanted ||
+	     uc_fw->minor_ver_found < uc_fw->minor_ver_wanted) &&
+	    !intel_uc_fw_is_overridden(uc_fw)) {
 		dev_notice(dev, "%s firmware %s: unexpected version: %u.%u !=  
 %u.%u\n",
 			   intel_uc_fw_type_repr(uc_fw->type), uc_fw->path,

drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw, · line 365

Quoted 10 lines
drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +2 -4
 			   uc_fw->major_ver_found, uc_fw->minor_ver_found,
 			   uc_fw->major_ver_wanted, uc_fw->minor_ver_wanted);
-		if (!intel_uc_fw_is_overridden(uc_fw)) {
-			err = -ENOEXEC;
-			goto fail;
-		}
+		err = -ENOEXEC;
+		goto fail;
 	}
 	obj = i915_gem_object_create_shmem_from_data(i915, fw->data, fw->size);

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#6 lore raw

PATCH v3 1/1 drm/i915/uc: More useful FW version mis-match notice

In reply to [email protected]

From: John Harrison <[email protected]>

If a FW override is present then a version mis-match is actually ignored. The warning notice was still being printed, though. Which could confuse people by implying that the load had failed when it had actually succeeded. So add an extra tag to the notice to say whether the mis-match caused a failure to load (normal case) or was ignored (override case).

So, only print the notice when actually failing the load and avoid any potential confusion.

v2: Original patch added a new 'ignore the previous notice' notice. Now it just suppresses the existing notice. Review feedback from Michal W.
v3: Always print the notice, but add a tag about override or failure. Review feedback from Michal W.

CC Michal Wajdeczko <[email protected]>
---
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +4 -2
index 66a30ab7044a..deae1e89de0e 100644
@@ -353,10 +353,12 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw, struct drm_i915_private *i915)
 
 	if (uc_fw->major_ver_found != uc_fw->major_ver_wanted ||
 	    uc_fw->minor_ver_found < uc_fw->minor_ver_wanted) {
-		dev_notice(dev, "%s firmware %s: unexpected version: %u.%u != %u.%u\n",
+		dev_notice(dev, "%s firmware %s: unexpected version: %u.%u != %u.%u, %s\n",
 			   intel_uc_fw_type_repr(uc_fw->type), uc_fw->path,
 			   uc_fw->major_ver_found, uc_fw->minor_ver_found,
-			   uc_fw->major_ver_wanted, uc_fw->minor_ver_wanted);
+			   uc_fw->major_ver_wanted, uc_fw->minor_ver_wanted,
+			   intel_uc_fw_is_overridden(uc_fw) ?
+			   "ignoring" : "aborting");
 		if (!intel_uc_fw_is_overridden(uc_fw)) {
 			err = -ENOEXEC;
 			goto fail;
Signature · 6 lines
2.21.0.5.gaeb582a983

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx