On 27-01-21, 20:04, Bjorn Andersson wrote:
> a6xx_gmu_stop() calls dev_pm_opp_set_opp() with NULL as opp in order to
> drop its bandwidth request, which was valid with dev_pm_opp_set_bw().
> But after the transition to dev_pm_opp_set_opp() this leads to a NULL
> dereference before jumping into _set_opp(), which does disable the
> vote as expected.
>
> Fixes: a0d67b94e2ef ("opp: Implement dev_pm_opp_set_opp()")
> Signed-off-by: Bjorn Andersson <bjorn.andersson@xxxxxxxxxx>
> ---
> drivers/opp/core.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/opp/core.c b/drivers/opp/core.c
> index 583bb1274df9..3ff05f40e443 100644
> --- a/drivers/opp/core.c
> +++ b/drivers/opp/core.c
> @@ -1157,7 +1157,7 @@ int dev_pm_opp_set_opp(struct device *dev, struct dev_pm_opp *opp)
> return PTR_ERR(opp_table);
> }
>
> - ret = _set_opp(dev, opp_table, opp, opp->rate);
> + ret = _set_opp(dev, opp_table, opp, opp ? opp->rate : 0);
> dev_pm_opp_put_opp_table(opp_table);
>
> return ret;
I specially handled this case with care and still missed this :(
I have updated the original patch to fix it, thanks.
--
viresh
