2021-06-25 21:29:58

by Mike Tipton

Subject: [PATCH 1/4] interconnect: Zero initial BW after sync-state

The initial BW values may be used by providers to enforce floors. Zero
these values after sync-state so that providers know when to stop
enforcing them.

Fixes: b1d681d8d324 ("interconnect: Add sync state support")
Signed-off-by: Mike Tipton <[email protected]>
---
drivers/interconnect/core.c | 2 ++
1 file changed, 2 insertions(+)

diff --git a/drivers/interconnect/core.c b/drivers/interconnect/core.c
index 8a1e70e00876..945121e18b5c 100644
--- a/drivers/interconnect/core.c
+++ b/drivers/interconnect/core.c
@@ -1106,6 +1106,8 @@ void icc_sync_state(struct device *dev)
dev_dbg(p->dev, "interconnect provider is in synced state\n");
list_for_each_entry(n, &p->nodes, node_list) {
if (n->init_avg || n->init_peak) {
+ n->init_avg = 0;
+ n->init_peak = 0;
aggregate_requests(n);
p->set(n, n);
}
--
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project

2021-07-01 17:44:33

by Odelu Kukatla

Subject: Re: [PATCH 1/4] interconnect: Zero initial BW after sync-state

On 2021-06-26 02:58, Mike Tipton wrote:
> The initial BW values may be used by providers to enforce floors. Zero
> these values after sync-state so that providers know when to stop
> enforcing them.
>
> Fixes: b1d681d8d324 ("interconnect: Add sync state support")
> Signed-off-by: Mike Tipton <[email protected]>
> ---
> drivers/interconnect/core.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/drivers/interconnect/core.c b/drivers/interconnect/core.c
> index 8a1e70e00876..945121e18b5c 100644
> --- a/drivers/interconnect/core.c
> +++ b/drivers/interconnect/core.c
> @@ -1106,6 +1106,8 @@ void icc_sync_state(struct device *dev)
> dev_dbg(p->dev, "interconnect provider is in synced state\n");
> list_for_each_entry(n, &p->nodes, node_list) {
> if (n->init_avg || n->init_peak) {
> + n->init_avg = 0;
> + n->init_peak = 0;
nit: It is good to reset init/floor levels back to zero, but we don't
need to do this as we have sync_state flag to let providers know when to
stop enforcing.
> aggregate_requests(n);
> p->set(n, n);
> }

2021-07-12 15:43:31

by Mike Tipton

Subject: Re: [PATCH 1/4] interconnect: Zero initial BW after sync-state

On 7/1/2021 9:56 AM, [email protected] wrote:
> On 2021-06-26 02:58, Mike Tipton wrote:
>> The initial BW values may be used by providers to enforce floors. Zero
>> these values after sync-state so that providers know when to stop
>> enforcing them.
>>
>> Fixes: b1d681d8d324 ("interconnect: Add sync state support")
>> Signed-off-by: Mike Tipton <[email protected]>
>> ---
>>  drivers/interconnect/core.c | 2 ++
>>  1 file changed, 2 insertions(+)
>>
>> diff --git a/drivers/interconnect/core.c b/drivers/interconnect/core.c
>> index 8a1e70e00876..945121e18b5c 100644
>> --- a/drivers/interconnect/core.c
>> +++ b/drivers/interconnect/core.c
>> @@ -1106,6 +1106,8 @@ void icc_sync_state(struct device *dev)
>>          dev_dbg(p->dev, "interconnect provider is in synced state\n");
>>          list_for_each_entry(n, &p->nodes, node_list) {
>>              if (n->init_avg || n->init_peak) {
>> +                n->init_avg = 0;
>> +                n->init_peak = 0;
> nit: It is good to reset init/floor levels back to zero, but we don't
> need to do this as we have sync_state flag to let providers know when to
> stop enforcing.

The synced_state variable is static to this file. It's not exposed to
providers. In fact, we could entirely remove synced_state with this
patch since it's unnecessary after zeroing the initial floors.

>>                  aggregate_requests(n);
>>                  p->set(n, n);
>>              }