Skip to contentSkip to conversation

tests/device_reset: Skip if no drm devices are preset

PATCH In review Nirmoy Das 4 participants 7 messages Mar 12, 2024

lore
Share & tools
lore

Nirmoy Das

author
#0 lore raw

PATCH 1/1 tests/device_reset: Skip if no drm devices are preset

Skip the test if no drm devices are preset.

v2: fix a typo.

---
 tests/device_reset.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tests/device_reset.c +1 -1
index 817cbc285..73531d696 100644
@@ -198,7 +198,7 @@ static void init_device_fds(struct device_fds *dev)
 	 * a device file descriptor open for exit handler use.
 	 */
 	dev->fds.dev = __drm_open_driver(DRIVER_ANY);
-	igt_assert_fd(dev->fds.dev);
+	igt_require(dev->fds.dev != -1);
 	if (is_i915_device(dev->fds.dev))
 		igt_require_gem(dev->fds.dev);
 
Signature · 1 line
2.42.0
#1 lore raw

Hi Nirmoy,

Quoted 1 line · On Tuesday, 12 March 2024 11:07:57 CET Nirmoy Das wrote:

Skip the test if no drm devices are preset.

Did you mean present? (also in the subject line)

Quoted 14 lines

v2: fix a typo.

---
 tests/device_reset.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tests/device_reset.c +0 -0
index 817cbc285..73531d696 100644

tests/device_reset.c static void init_device_fds(struct device_fds *dev) · line 201

Quoted 5 lines
tests/device_reset.c +1 -1
@@ -198,7 +198,7 @@ static void init_device_fds(struct device_fds *dev)
 	 * a device file descriptor open for exit handler use.
 	 */
 	dev->fds.dev = __drm_open_driver(DRIVER_ANY);
-	igt_assert_fd(dev->fds.dev);
+	igt_require(dev->fds.dev != -1);

What if __drm_open_driver() returns (unlikely) a negative value other than -1? In IGT we used to verify validity of a file descriptor by checking if its value is not negative, i.e., >= 0 (see e.g. lib/igt_core:igt_assert_fd(), lib/drmtest.c:drm_open_driver()). Please follow the same convention.

Thanks,
Janusz

Quoted 4 lines
tests/device_reset.c +0 -0
 	if (is_i915_device(dev->fds.dev))
 		igt_require_gem(dev->fds.dev);
 
author
#2 lore raw
Quoted 5 lines · Janusz Krzysztofik

Hi Nirmoy,

Quoted 1 line · On Tuesday, 12 March 2024 11:07:57 CET Nirmoy Das wrote:

Skip the test if no drm devices are preset.

Did you mean present? (also in the subject line)

Right, another typo

Quoted 9 lines
Quoted 9 lines

v2: fix a typo.

---
  tests/device_reset.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)
Quoted 11 lines

tests/device_reset.c static void init_device_fds(struct device_fds *dev) · line 201

Quoted 10 lines
tests/device_reset.c +1 -1
index 817cbc285..73531d696 100644
@@ -198,7 +198,7 @@ static void init_device_fds(struct device_fds *dev)
  	 * a device file descriptor open for exit handler use.
  	 */
  	dev->fds.dev = __drm_open_driver(DRIVER_ANY);
-	igt_assert_fd(dev->fds.dev);
+	igt_require(dev->fds.dev != -1);

What if __drm_open_driver() returns (unlikely) a negative value other than -1?

__drm_open_driverdoc says: "An open DRM fd or -1 on error" so that won't happen.

Quoted 4 lines

In IGT we used to verify validity of a file descriptor by checking if its value is not negative, i.e., >= 0 (see e.g. lib/igt_core:igt_assert_fd(), lib/drmtest.c:drm_open_driver()). Please follow the same convention.

I can do s/igt_assert_fd/igt_require_fd

Thanks,

Nirmoy

Quoted 10 lines

Thanks,
Janusz

Quoted 4 lines
tests/device_reset.c +0 -0
  	if (is_i915_device(dev->fds.dev))
  		igt_require_gem(dev->fds.dev);
 
#3 lore raw
Quoted 28 lines · On Tuesday, 12 March 2024 11:54:32 CET Nirmoy Das wrote:
Quoted 5 lines · Janusz Krzysztofik

Hi Nirmoy,

Quoted 1 line · On Tuesday, 12 March 2024 11:07:57 CET Nirmoy Das wrote:

Skip the test if no drm devices are preset.

Did you mean present? (also in the subject line)

Right, another typo

Quoted 20 lines
Quoted 14 lines

v2: fix a typo.

---
  tests/device_reset.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)
tests/device_reset.c +0 -0
index 817cbc285..73531d696 100644

tests/device_reset.c static void init_device_fds(struct device_fds *dev) · line 201

Quoted 5 lines
tests/device_reset.c +1 -1
@@ -198,7 +198,7 @@ static void init_device_fds(struct device_fds *dev)
  	 * a device file descriptor open for exit handler use.
  	 */
  	dev->fds.dev = __drm_open_driver(DRIVER_ANY);
-	igt_assert_fd(dev->fds.dev);
+	igt_require(dev->fds.dev != -1);

What if __drm_open_driver() returns (unlikely) a negative value other than

-1?

Quoted 2 lines

__drm_open_driverdoc says: "An open DRM fd or -1 on error" so that won't

happen.

That won't happen as long as there are no bugs in __drm_open_driver().

If you don't want to skip on (fd < -1) then let's fail, but we should never succeed in that unlikely case, I believe.

Thanks,
Janusz

Quoted 22 lines
Quoted 5 lines

In IGT we used to verify validity of a file descriptor by checking if its value is not negative, i.e., >= 0 (see e.g. lib/igt_core:igt_assert_fd(), lib/drmtest.c:drm_open_driver()). Please follow the same convention.

I can do s/igt_assert_fd/igt_require_fd

Thanks,

Nirmoy

Quoted 10 lines

Thanks,
Janusz

Quoted 4 lines
tests/device_reset.c +0 -0
  	if (is_i915_device(dev->fds.dev))
  		igt_require_gem(dev->fds.dev);
 
#4 lore raw
Quoted 27 lines · Janusz Krzysztofik
Quoted 26 lines · On Tuesday, 12 March 2024 11:54:32 CET Nirmoy Das wrote:
Quoted 5 lines · Janusz Krzysztofik

Hi Nirmoy,

Quoted 1 line · On Tuesday, 12 March 2024 11:07:57 CET Nirmoy Das wrote:

Skip the test if no drm devices are preset.

Did you mean present? (also in the subject line)

Right, another typo

Quoted 19 lines
Quoted 13 lines

v2: fix a typo.

---
   tests/device_reset.c | 2 +-
   1 file changed, 1 insertion(+), 1 deletion(-)
tests/device_reset.c +0 -0
index 817cbc285..73531d696 100644

tests/device_reset.c static void init_device_fds(struct device_fds *dev) · line 201

Quoted 5 lines
tests/device_reset.c +1 -1
@@ -198,7 +198,7 @@ static void init_device_fds(struct device_fds *dev)
   	 * a device file descriptor open for exit handler use.
   	 */
   	dev->fds.dev = __drm_open_driver(DRIVER_ANY);
-	igt_assert_fd(dev->fds.dev);
+	igt_require(dev->fds.dev != -1);

What if __drm_open_driver() returns (unlikely) a negative value other than

Quoted 8 lines

-1?

Quoted 1 line

__drm_open_driverdoc says: "An open DRM fd or -1 on error" so that won't

happen.

That won't happen as long as there are no bugs in __drm_open_driver().

If you don't want to skip on (fd < -1) then let's fail, but we should never succeed in that unlikely case, I believe.

No, we should take documented return value as a contract and shouldn't over worry about the function

breaking it.

Regards,

Nirmoy

Quoted 26 lines

Thanks,
Janusz

Quoted 19 lines
Quoted 4 lines

In IGT we used to verify validity of a file descriptor by checking if its value is not negative, i.e., >= 0 (see e.g. lib/igt_core:igt_assert_fd(), lib/drmtest.c:drm_open_driver()). Please follow the same convention.

I can do s/igt_assert_fd/igt_require_fd

Thanks,

Nirmoy

Quoted 9 lines

Thanks,
Janusz

Quoted 4 lines
tests/device_reset.c +0 -0
   	if (is_i915_device(dev->fds.dev))
   		igt_require_gem(dev->fds.dev);
 
#5 lore raw

✓ CI.xeBAT: success for tests/device_reset: Skip if no drm devices are preset (rev3)

In reply to Nirmoy Das
== Series Details ==

== Series Details ==

Series: tests/device_reset: Skip if no drm devices are preset (rev3)
URL : https://patchwork.freedesktop.org/series/130992/
State : success

== Summary ==

CI Bug Log - changes from XEIGT_7755_BAT -> XEIGTPW_10813_BAT
====================================================

Summary
-------

**SUCCESS**

No regressions found.

Participating hosts (4 -> 4)
------------------------------

No changes in participating hosts

Changes
-------

No changes found

Build changes
-------------

* IGT: IGT_7755 -> IGTPW_10813

IGTPW_10813: 10813
IGT_7755: 7755
xe-929-61fa386186cb302cf26e29aef76d5b1a6ffe57ad: 61fa386186cb302cf26e29aef76d5b1a6ffe57ad

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10813/index.html

#6 lore raw

✗ Fi.CI.BAT: failure for tests/device_reset: Skip if no drm devices are preset (rev3)

In reply to Nirmoy Das
== Series Details ==

== Series Details ==

Series: tests/device_reset: Skip if no drm devices are preset (rev3)
URL : https://patchwork.freedesktop.org/series/130992/
State : failure

== Summary ==

CI Bug Log - changes from IGT_7755 -> IGTPW_10813
====================================================

Summary
-------

**FAILURE**

Serious unknown changes coming with IGTPW_10813 absolutely need to be verified manually.

If you think the reported changes have nothing to do with the changes
introduced in IGTPW_10813, please notify your bug team ([email protected]) to allow them to document this new failure mode, which will reduce false positives in CI.

External URL: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10813/index.html

Participating hosts (34 -> 29)
------------------------------

Missing (5): bat-dg2-8 fi-snb-2520m fi-glk-j4005 bat-rpls-3 bat-jsl-1

Possible new issues
-------------------

Here are the unknown changes that may have been introduced in IGTPW_10813:

### IGT changes ###

#### Possible regressions ####

* igt@kms_pm_rpm@basic-rte:
- bat-arls-2: [PASS][1] -> [ABORT][2]
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_7755/bat-arls-2/igt@[email protected]
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10813/bat-arls-2/igt@[email protected]

Known issues
------------

Here are the changes found in IGTPW_10813 that come from known issues:

### IGT changes ###

#### Issues hit ####

* igt@gem_huc_copy@huc-copy:
- fi-cfl-8109u: NOTRUN -> [SKIP][3] ([i915#2190])
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10813/fi-cfl-8109u/igt@[email protected]

* igt@gem_lmem_swapping@basic@lmem0:
- bat-dg2-14: [PASS][4] -> [FAIL][5] ([i915#10378])
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_7755/bat-dg2-14/igt@gem_lmem_swapping@[email protected]
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10813/bat-dg2-14/igt@gem_lmem_swapping@[email protected]

* igt@gem_lmem_swapping@verify-random:
- fi-cfl-8109u: NOTRUN -> [SKIP][6] ([i915#4613]) +3 other tests skip
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10813/fi-cfl-8109u/igt@[email protected]

* igt@i915_pm_rpm@module-reload:
- fi-blb-e6850: NOTRUN -> [SKIP][7] +32 other tests skip
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10813/fi-blb-e6850/igt@[email protected]

* igt@i915_selftest@live@perf:
- bat-dg2-9: NOTRUN -> [ABORT][8] ([i915#10366])
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10813/bat-dg2-9/igt@i915_selftest@[email protected]

* igt@kms_pm_backlight@basic-brightness:
- fi-cfl-8109u: NOTRUN -> [SKIP][9] +11 other tests skip
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10813/fi-cfl-8109u/igt@[email protected]

#### Possible fixes ####

* igt@i915_selftest@live@dmabuf:
- bat-dg2-9: [ABORT][10] ([i915#10366]) -> [PASS][11]
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_7755/bat-dg2-9/igt@i915_selftest@[email protected]
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10813/bat-dg2-9/igt@i915_selftest@[email protected]

* igt@i915_selftest@live@guc_multi_lrc:
- bat-atsm-1: [ABORT][12] ([i915#10366]) -> [PASS][13]
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/IGT_7755/bat-atsm-1/igt@i915_selftest@live@guc_multi_lrc.html
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10813/bat-atsm-1/igt@i915_selftest@live@guc_multi_lrc.html

[i915#10366]: https://gitlab.freedesktop.org/drm/intel/issues/10366
[i915#10378]: https://gitlab.freedesktop.org/drm/intel/issues/10378
[i915#2190]: https://gitlab.freedesktop.org/drm/intel/issues/2190
[i915#4613]: https://gitlab.freedesktop.org/drm/intel/issues/4613

Build changes
-------------

* CI: CI-20190529 -> None
* IGT: IGT_7755 -> IGTPW_10813

CI-20190529: 20190529
CI_DRM_14420: 61fa386186cb302cf26e29aef76d5b1a6ffe57ad @ git://anongit.freedesktop.org/gfx-ci/linux
IGTPW_10813: 10813
IGT_7755: 7755

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10813/index.html