This series fixes kcontrol put callback in some of the Tegra drivers
which are used on platforms based on Tegra210 and later. The callback
is expected to return 1 whenever the HW update is done.
ADMAIF, I2S, DMIC and DSPK drivers are updated as part of it.
This idea is suggested by Jaroslav. Similar suggestion came from
Mark during review of series [0] and drivers under those series
were taken care. This series takes care of remaining drivers.
I have added 'Suggested-by" tags accordingly.
[0] https://lore.kernel.org/linux-arm-kernel/[email protected]/
Sameer Pujar (4):
ASoC: tegra: Fix kcontrol put callback in ADMAIF
ASoC: tegra: Fix kcontrol put callback in I2S
ASoC: tegra: Fix kcontrol put callback in DMIC
ASoC: tegra: Fix kcontrol put callback in DSPK
sound/soc/tegra/tegra186_dspk.c | 4 +++-
sound/soc/tegra/tegra210_admaif.c | 4 +++-
sound/soc/tegra/tegra210_dmic.c | 4 +++-
sound/soc/tegra/tegra210_i2s.c | 4 +++-
4 files changed, 12 insertions(+), 4 deletions(-)
--
2.7.4
On 03. 11. 21 6:04, Sameer Pujar wrote:
> This series fixes kcontrol put callback in some of the Tegra drivers
> which are used on platforms based on Tegra210 and later. The callback
> is expected to return 1 whenever the HW update is done.
The logic is a little bit another. The 1 means change, so you should compare
the register / cached value and return 1 only when a different value is set.
Your implementation invokes the change events for duplicate updates.
Jaroslav
--
Jaroslav Kysela <[email protected]>
Linux Sound Maintainer; ALSA Project; Red Hat, Inc.
On 11/3/2021 2:20 PM, Jaroslav Kysela wrote:
> External email: Use caution opening links or attachments
>
>
> On 03. 11. 21 6:04, Sameer Pujar wrote:
>> This series fixes kcontrol put callback in some of the Tegra drivers
>> which are used on platforms based on Tegra210 and later. The callback
>> is expected to return 1 whenever the HW update is done.
>
> The logic is a little bit another. The 1 means change, so you should
> compare
> the register / cached value and return 1 only when a different value
> is set.
> Your implementation invokes the change events for duplicate updates.
Sorry, I did not think the about duplicate updates. Thanks for the
suggestion.