| Message ID | 20230712-kunit-arm64-cpu-max-v1-1-4892fe50f40e@kernel.org (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | 1b0975ee3bdd3eb19a47371c26fd7ef8f7f6b599 |
| Delegated to: | Brendan Higgins |
| Headers | show |
| Series | kunit: qemu_configs: Enable all architectural features for arm64 | expand |
On Thu, 13 Jul 2023 at 04:23, Mark Brown <broonie@kernel.org> wrote: > > While it probably doesn't make a huge difference given the current KUnit > coverage we will get the best coverage of arm64 architecture features if > we specify -cpu=max rather than picking a specific CPU, this will include > all architecture features that qemu supports including many which have not > yet made it into physical implementations. > > Due to performance issues emulating the architected pointer authentication > algorithm it is recommended to use the implementation defined algorithm > that qemu has instead, this should make no meaningful difference to the > coverage and will run the tests faster. > > Signed-off-by: Mark Brown <broonie@kernel.org> > --- Looks good and works for me here, thanks. Reviewed-by: David Gow <davidgow@google.com> Cheers, -- David > tools/testing/kunit/qemu_configs/arm64.py | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/tools/testing/kunit/qemu_configs/arm64.py b/tools/testing/kunit/qemu_configs/arm64.py > index 67d04064f785..d3ff27024755 100644 > --- a/tools/testing/kunit/qemu_configs/arm64.py > +++ b/tools/testing/kunit/qemu_configs/arm64.py > @@ -9,4 +9,4 @@ CONFIG_SERIAL_AMBA_PL011_CONSOLE=y''', > qemu_arch='aarch64', > kernel_path='arch/arm64/boot/Image.gz', > kernel_command_line='console=ttyAMA0', > - extra_qemu_params=['-machine', 'virt', '-cpu', 'cortex-a57']) > + extra_qemu_params=['-machine', 'virt', '-cpu', 'max,pauth-impdef=on']) > > --- > base-commit: 06c2afb862f9da8dc5efa4b6076a0e48c3fbaaa5 > change-id: 20230702-kunit-arm64-cpu-max-7e3aa5f02fb2 > > Best regards, > -- > Mark Brown <broonie@kernel.org> >
diff --git a/tools/testing/kunit/qemu_configs/arm64.py b/tools/testing/kunit/qemu_configs/arm64.py index 67d04064f785..d3ff27024755 100644 --- a/tools/testing/kunit/qemu_configs/arm64.py +++ b/tools/testing/kunit/qemu_configs/arm64.py @@ -9,4 +9,4 @@ CONFIG_SERIAL_AMBA_PL011_CONSOLE=y''', qemu_arch='aarch64', kernel_path='arch/arm64/boot/Image.gz', kernel_command_line='console=ttyAMA0', - extra_qemu_params=['-machine', 'virt', '-cpu', 'cortex-a57']) + extra_qemu_params=['-machine', 'virt', '-cpu', 'max,pauth-impdef=on'])
While it probably doesn't make a huge difference given the current KUnit coverage we will get the best coverage of arm64 architecture features if we specify -cpu=max rather than picking a specific CPU, this will include all architecture features that qemu supports including many which have not yet made it into physical implementations. Due to performance issues emulating the architected pointer authentication algorithm it is recommended to use the implementation defined algorithm that qemu has instead, this should make no meaningful difference to the coverage and will run the tests faster. Signed-off-by: Mark Brown <broonie@kernel.org> --- tools/testing/kunit/qemu_configs/arm64.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- base-commit: 06c2afb862f9da8dc5efa4b6076a0e48c3fbaaa5 change-id: 20230702-kunit-arm64-cpu-max-7e3aa5f02fb2 Best regards,