Re: [PATCH] clk: qcom: enable ALWAYS_ON for titan_top_gdsc

[Brian Masney <bmasney@redhat.com>]

Hi Vladimir,

On Fri, Jun 26, 2026 at 07:54:39PM +0300, Vladimir Zapolskiy wrote:
> On 6/26/26 19:32, Konrad Dybcio wrote:
> > On 6/26/26 6:26 PM, Brian Masney wrote:
> > > With the introduction of sync_state support in the clk and pmdomain
> > > subsystems, the following warning happens when the unused clocks are
> > > shutdown in camcc-sc8280xp:
> > 
> > Stuck at _on_ sounds wrong.. does clk_ignore_unused / removing
> > the sync state from the clock part only resolve this? There may
> > be a clock dependency for the TITAN_TOP_GDSC that we're failing to
> > describe
> > 
> 
> Let me remind that there is a series, which solves this problem in a proper
> and elegant way, it would be nice to get the review tags though:
> 
> [1] https://lore.kernel.org/linux-clk/20260331140142.892579-1-vladimir.zapolskiy@linaro.org

As I mentioned in my other message, this series didn't address my issue.
sc8280xp wasn't included in your series, so I had to make the following
change in my tree:

diff --git a/drivers/clk/qcom/camcc-sc8280xp.c b/drivers/clk/qcom/camcc-sc8280xp.c
index 18f5a3eb313e1..4724ea54c5f67 100644
--- a/drivers/clk/qcom/camcc-sc8280xp.c
+++ b/drivers/clk/qcom/camcc-sc8280xp.c
@@ -2995,6 +2995,7 @@ static const struct qcom_cc_desc camcc_sc8280xp_desc = {
        .num_resets = ARRAY_SIZE(camcc_sc8280xp_resets),
        .gdscs = camcc_sc8280xp_gdscs,
        .num_gdscs = ARRAY_SIZE(camcc_sc8280xp_gdscs),
+       .cc_gdsc = &titan_top_gdsc,
 };
 
 static const struct of_device_id camcc_sc8280xp_match_table[] = {

Brian