boards: xenvm: remove incorrect condition for Kconfig heap values #81790
Conversation
zephyrbot
added
size: XS
firscity
force-pushed
the
board_xenvm_typo_fix
branch
from
77cf4e9 to
fab0cd6
Compare
|
@firscity it looks like the CI failures might be legit? |
|
@kartben yeah, thank you |
firscity
dismissed stale reviews from nordicjm, kartben, lorc, and luca-fancellu
via
9e1692b
zephyrbot
requested review from
andyross,
ceolin,
cfriedt,
dcpleung,
gmarull,
nashif,
npitre,
peter-mitsis and
tbursztyka
tests/kernel/device/testcase.yaml
Outdated
| @@ -7,7 +7,7 @@ tests: | |||
| kernel.device: | |||
| integration_platforms: | |||
| - native_sim | |||
| platform_exclude: xenvm | |||
| platform_exclude: xenvm xenvm/xenvm/gicv3 | |||
There was a problem hiding this comment.
Make all of these multiline please, this way diffs will be cleaner going forward
There was a problem hiding this comment.
Thanks for suggestion, done.
firscity
force-pushed
the
board_xenvm_typo_fix
branch
from
9e1692b to
aea83bc
Compare
|
Thanks everyone for review and approves. For everyone previously reviewed: Also I have no idea why |
|
Hi @rlubos, as I can see, you are maintainer of |
This has been fixed in main. Please rebase to make the error go away:) |
firscity
force-pushed
the
board_xenvm_typo_fix
branch
from
aea83bc to
1553ef5
Compare
|
@kartben yeah, thank you for information! Now everything looks good, @povergoing take a look please |
|
@povergoing ping |
There was a problem hiding this comment.
I think you need another maintainer, @povergoing might not be active at the moment
Right. They should probably remove themselves as a maintainer then, based on their recent inactivity. Any other collaborator willing to step up as maintainers? |
xenvm Kconfig contained incorrect name for board parameter. It led to
build issues - heap size was set incorrectly. Since whole file is
already placed under right Kconfig condition ("if BOARD_XENVM"), remove
incorrect parameter at all.
This issue was introduced by commit 8dc3f85 ("hwmv2: Introduce
Hardware model version 2 and convert devices") due to the typo.
Signed-off-by: Dmytro Firsov <[email protected]>
The issue with mentioned test for GICv3 version of xenvm virtual board is the same as for regular xenvm - device tree overlay for this test overrides #address-cells and #size-cells property for original device tree file (from 0x2 to 0x1), which leads to incorrect DT parsing by actual Xen drivers. This causes build errors, so test should be skipped for GICv3 platform too. Same issue for regular xenvm was fixed by commit 40fe366 ("tests: kernel: exclude xenvm from device tests"). Issue for GICv3 appeared after migrating to HWMv2. Signed-off-by: Dmytro Firsov <[email protected]>
firscity
dismissed stale reviews from cfriedt, luca-fancellu, kartben, and lorc
via
dea2263
firscity
force-pushed
the
board_xenvm_typo_fix
branch
2 times, most recently
from
dea2263 to
8ade5c5
Compare
|
@kartben thank you for your support, rebased |
xenvm Kconfig contained incorrect name for board parameter. It led to build issues - heap size was set incorrectly. Since whole file is already placed under right Kconfig condition ("if BOARD_XENVM"), remove incorret parameter at all.
This issue was introduced by commit 8dc3f85 during migration to hwmv2 due to typo.