[ Upstream commit 0669631dbccd41cf3ca7aa70213fcd8bb41c4b38 ]

The STM32 SAI driver do not set the clock strobing bit (CKSTR) for DSP_A,
DSP_B and LEFT_J formats, causing data to be sampled on the wrong BCLK
edge when SND_SOC_DAIFMT_NB_NF is used.

Per ALSA convention, NB_NF requires sampling on the rising BCLK edge.
The STM32MP25 SAI reference manual states that CKSTR=1 is required for
signals received by the SAI to be sampled on the SCK rising edge.
Without setting CKSTR=1, the SAI samples on the falling edge, violating
the NB_NF convention. For comparison, the NXP FSL SAI driver correctly
sets FSL_SAI_CR2_BCP for DSP_A, DSP_B and LEFT_J, consistent with its
I2S handling.

This patch adds SAI_XCR1_CKSTR for DSP_A, DSP_B and LEFT_J in
stm32_sai_set_dai_fmt which was verified empirically with a cs47l35 codec.
RIGHT_J (LSB) is not investigated and addressed by this patch.

Note: the STM32 I2S driver (stm32_i2s_set_dai_fmt) may have the same issue
for DSP_A mode, as I2S_CGFR_CKPOL is not set. This has not been verified
and is left for a separate investigation.

Signed-off-by: Tomasz Merta <tommerta@gmail.com>
Link: https://patch.msgid.link/20260408084056.20588-1-tommerta@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

Merge series from Johan Hovold <johan@kernel.org>:

This series fixes device and OF node reference leaks during probe and
a clock prepare imbalance on probe failures.

Included is a related cleanup of an error path.

Use an error label for deregistering also the PCM DMA for consistency.

Signed-off-by: Johan Hovold <johan@kernel.org>
Reviewed-by: olivier moysan <olivier.moysan@foss.st.com>
Link: https://patch.msgid.link/20251124104908.15754-5-johan@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

The reference taken to the sync provider OF node when probing the
platform device is currently only dropped if the set_sync() callback
fails during DAI probe.

Make sure to drop the reference on platform probe failures (e.g. probe
deferral) and on driver unbind.

This also avoids a potential use-after-free in case the DAI is ever
reprobed without first rebinding the platform driver.

Fixes: 5914d285f6b7 ("ASoC: stm32: sai: Add synchronization support")
Fixes: d4180b4c02e7 ("ASoC: stm32: sai: fix set_sync service")
Cc: Olivier Moysan <olivier.moysan@st.com>
Cc: stable@vger.kernel.org      # 4.16: d4180b4c02e7
Signed-off-by: Johan Hovold <johan@kernel.org>
Reviewed-by: olivier moysan <olivier.moysan@foss.st.com>
Link: https://patch.msgid.link/20251124104908.15754-4-johan@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

Make sure to unprepare the parent clock also on probe failures (e.g.
probe deferral).

Fixes: a14bf98c045b ("ASoC: stm32: sai: fix possible circular locking")
Cc: stable@vger.kernel.org	# 5.5
Cc: Olivier Moysan <olivier.moysan@st.com>
Signed-off-by: Johan Hovold <johan@kernel.org>
Reviewed-by: olivier moysan <olivier.moysan@foss.st.com>
Link: https://patch.msgid.link/20251124104908.15754-3-johan@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

Make sure to drop the reference taken when looking up the sync provider
device and its driver data during DAI probe on probe failures and on
unbind.

Note that holding a reference to a device does not prevent its driver
data from going away so there is no point in keeping the reference.

Fixes: 7dd0d835582f ("ASoC: stm32: sai: simplify sync modes management")
Fixes: 1c3816a19487 ("ASoC: stm32: sai: add missing put_device()")
Cc: stable@vger.kernel.org	# 4.16: 1c3816a19487
Cc: olivier moysan <olivier.moysan@st.com>
Cc: Wen Yang <yellowriver2010@hotmail.com>
Signed-off-by: Johan Hovold <johan@kernel.org>
Reviewed-by: olivier moysan <olivier.moysan@foss.st.com>
Link: https://patch.msgid.link/20251124104908.15754-2-johan@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

In the past %pK was preferable to %p as it would not leak raw pointer
values into the kernel log.
Since commit ad67b74d2469 ("printk: hash addresses printed with %p")
the regular %p has been improved to avoid this issue.
Furthermore, restricted pointers ("%pK") were never meant to be used
through printk(). They can still unintentionally leak raw pointers or
acquire sleeping locks in atomic contexts.

Switch to the regular pointer formatting which is safer and
easier to reason about.

Signed-off-by: Thomas Weißschuh <thomas.weissschuh@linutronix.de>
Link: https://patch.msgid.link/20251106-restricted-pointers-stm-v1-1-7d03eb5020ba@linutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>

The mclk direction now needs to be specified in endpoint node with
"system-clock-direction-out" property. However some calls to the
set_sysclk callback, related to CPU DAI clock, result in unbalanced
calls to clock API.
The set_sysclk callback in STM32 SAI driver is intended only for mclk
management. So it is relevant to ensure that calls to set_sysclk are
related to mclk only.
Since the master clock is handled only at runtime, skip the calls to
set_sysclk in the initialization phase.

Signed-off-by: Olivier Moysan <olivier.moysan@foss.st.com>
Link: https://patch.msgid.link/20250916123118.84175-1-olivier.moysan@foss.st.com
Signed-off-by: Mark Brown <broonie@kernel.org>

calc_clk_div() will only return a non-zero value (-EINVAL)
in case of error. On the other hand, req->rate is an unsigned long.
It seems quite odd that req->rate would be assigned a negative value,
which is clearly not a rate, and success would be returned.

Reinstate previous logic, which would just return error.

Fixes: afd529d74002 ("ASoC: stm: stm32_i2s: convert from round_rate() to determine_rate()")
Link: https://scan7.scan.coverity.com/#/project-view/53936/11354?selectedIssue=1647702

Signed-off-by: Sergio Perez Gonzalez <sperezglz@gmail.com>
Link: https://patch.msgid.link/20250729020052.404617-1-sperezglz@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series from Brian Masney <bmasney@redhat.com>:

The round_rate() clk ops is deprecated in the clk framework in favor
of the determine_rate() clk ops, so let's go ahead and convert the
drivers in the rtc subsystem using the Coccinelle semantic patch
posted below. I did a few minor cosmetic cleanups of the code in a
few cases.

Coccinelle semantic patch:

    virtual patch

    // Look up the current name of the round_rate function
    @ has_round_rate @
    identifier round_rate_name =~ ".*_round_rate";
    identifier hw_param, rate_param, parent_rate_param;
    @@

    long round_rate_name(struct clk_hw *hw_param, unsigned long rate_param,
                  unsigned long *parent_rate_param)
    {
    	...
    }

    // Rename the route_rate function name to determine_rate()
    @ script:python generate_name depends on has_round_rate @
    round_rate_name << has_round_rate.round_rate_name;
    new_name;
    @@

    coccinelle.new_name = round_rate_name.replace("_round_rate", "_determine_rate")

    // Change rate to req->rate; also change occurrences of 'return XXX'.
    @ chg_rate depends on generate_name @
    identifier has_round_rate.round_rate_name;
    identifier has_round_rate.hw_param;
    identifier has_round_rate.rate_param;
    identifier has_round_rate.parent_rate_param;
    identifier ERR =~ "E.*";
    expression E;
    @@

    long round_rate_name(struct clk_hw *hw_param, unsigned long rate_param,
                  unsigned long *parent_rate_param)
    {
    <...
    (
    -return -ERR;
    +return -ERR;
    |
    - return rate_param;
    + return 0;
    |
    - return E;
    + req->rate = E;
    +
    + return 0;
    |
    - rate_param
    + req->rate
    )
    ...>
    }

    // Coccinelle only transforms the first occurrence of the rate parameter
    // Run a second time. FIXME: Is there a better way to do this?
    @ chg_rate2 depends on generate_name @
    identifier has_round_rate.round_rate_name;
    identifier has_round_rate.hw_param;
    identifier has_round_rate.rate_param;
    identifier has_round_rate.parent_rate_param;
    @@

    long round_rate_name(struct clk_hw *hw_param, unsigned long rate_param,
                  unsigned long *parent_rate_param)
    {
    <...
    - rate_param
    + req->rate
    ...>
    }

    // Change parent_rate to req->best_parent_rate
    @ chg_parent_rate depends on generate_name @
    identifier has_round_rate.round_rate_name;
    identifier has_round_rate.hw_param;
    identifier has_round_rate.rate_param;
    identifier has_round_rate.parent_rate_param;
    @@

    long round_rate_name(struct clk_hw *hw_param, unsigned long rate_param,
                  unsigned long *parent_rate_param)
    {
    <...
    (
    - *parent_rate_param
    + req->best_parent_rate
    |
    - parent_rate_param
    + &req->best_parent_rate
    )
    ...>
    }

    // Convert the function definition from round_rate() to determine_rate()
    @ func_definition depends on chg_rate @
    identifier has_round_rate.round_rate_name;
    identifier has_round_rate.hw_param;
    identifier has_round_rate.rate_param;
    identifier has_round_rate.parent_rate_param;
    identifier generate_name.new_name;
    @@

    - long round_rate_name(struct clk_hw *hw_param, unsigned long rate_param,
    -               unsigned long *parent_rate_param)
    + int new_name(struct clk_hw *hw, struct clk_rate_request *req)
    {
        ...
    }

    // Update the ops from round_rate() to determine_rate()
    @ ops depends on func_definition @
    identifier has_round_rate.round_rate_name;
    identifier generate_name.new_name;
    @@

    {
        ...,
    -   .round_rate = round_rate_name,
    +   .determine_rate = new_name,
        ...,
    }

Note that I used coccinelle 1.2 instead of 1.3 since the newer version
adds unnecessary braces as described in this post.
https://lore.kernel.org/cocci/67642477-5f3e-4b2a-914d-579a54f48cbd@intel.com/