On Tue, 13 Sep 2022 14:04:41 +0200
Janusz Krzysztofik <janusz.krzysztofik@xxxxxxxxxxxxxxx> wrote:
> As soon as we first time open a drm device to be exercised, we identify
> its chipset, store that information, and perform some basic device
> specific checks. We also store a filter that matches the device bus
> address to make sure we will exercise still one and the same device after
> each driver unbind-rebind / device unplug-rediscover operation.
>
> However, each time before we select device specific health checks, whether
> before or after a driver unbind-rebind / device unplug-rediscover
> operation, we identify the device chipset again. Besides being more
> expensive, that approach could also result in using different set of
> health checks should device filters not work as expected.
>
> Always use the stored chipset type instead of identifying it now and
> again.
>
> Signed-off-by: Janusz Krzysztofik <janusz.krzysztofik@xxxxxxxxxxxxxxx>
Makes sense to me, once we have patch 3/4 applied. So,
Reviewed-by: Mauro Carvalho Chehab <mchehab@xxxxxxxxxx>
> ---
> tests/core_hotunplug.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/tests/core_hotunplug.c b/tests/core_hotunplug.c
> index 29cb26ab12..7c7ed09e63 100644
> --- a/tests/core_hotunplug.c
> +++ b/tests/core_hotunplug.c
> @@ -400,7 +400,7 @@ static void node_healthcheck(struct hotunplug *priv, unsigned flags)
> if (closed) /* store fd for cleanup if not dirty */
> priv->fd.drm_hc = fd_drm;
>
> - if (is_i915_device(fd_drm)) {
> + if (priv->chipset == DRIVER_INTEL) {
> /* don't report library failed asserts as healthcheck failure */
> priv->failure = "Unrecoverable test failure";
> if (local_i915_healthcheck(fd_drm, "") &&
