The code refactoring to move the WC page handling into the common
memalloc layer caused a breakage for HD-audio HDMI.  I overlooked that
the driver is using the SG-buffer, which isn't covered by the patch.

This patch adds the mmap workaround for WC pages to SG-buffer
handler.  A caveat is that it falls back to the default handler by
returning an error after setting the pgprot, so it won't work in all
cases but merely for PCM (which is currently the only use case).

Fixes: 623c1010 ("ALSA: memalloc: Fix pgprot for WC mmap on x86")
Reported-and-tested-by: Andy Lavr <andy.lavr@gmail.com>
Link: https://lore.kernel.org/r/20210808080034.20337-1-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

The recent refactoring of es1688 object creation with the use of
devres caused a problem with the non-PnP probe of GUS driver, as it
tries to probe multiple times with different parameters  That is, this
object needs the immediate resource release and the devres doesn't fit
for it.

This patch reverts partially the commit for restoring the classic
resource management for es1688 object.

Fixes: 1bb11c1c ("ALSA: es1688: Allocate resources with device-managed APIs")
Reported-by: kernel test robot <oliver.sang@intel.com>
Link: https://lore.kernel.org/r/20210805032513.GA30485@xsang-OptiPlex-9020
Link: https://lore.kernel.org/r/20210805062148.30951-1-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

Commit 6cdc01eb ("ALSA: pci/korg1212: remove 'set but not used'
warnings") missed one __maybe_unused, add to enable make W=1 sound/
compilation with x86_64 allmodconfig
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20210804185442.274057-1-pierre-louis.bossart@linux.intel.comSigned-off-by: Takashi Iwai <tiwai@suse.de>

In the recent fix commit eda80d7c ("ALSA: memalloc: Fix regression
with SNDRV_DMA_TYPE_CONTINUOUS"), we replaced the pfn argument of the
remap_page_pfn() call from the uninitialized dmab->addr.  It was the
right fix, but it'd be more generic if we actually initialize
dmab->area for SNDRV_DMA_TYPE_CONTINOUS, too.  e.g. the field is used
in the common snd_sgbuf_get_addr(), too.

This patch adds the initialization of addr field and does revert of
the previous change to refer to it again in the mmap call.

Link: https://lore.kernel.org/r/20210804074125.8170-1-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

We have a special handling of WC pages on x86, and it's currently
specific to HD-audio.  The last forgotten piece was the pgprot setup
for the mmap with WC pages.

This patch moves the pgprot setup for WC pages from HD-audio-specific
mmap callback to the common helper code.  It allows us to remove the
superfluous mmap callback in HD-audio and its prepare_mmap
redirection.

Link: https://lore.kernel.org/r/20210804061329.29265-1-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

This patch simplifies the buffer pre-allocation code of sprd driver
with the standard managed buffer helper.  It uses the newly
introduced fixed-size buffer allocation helper.
Acked-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20210802072815.13551-16-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

This patch simplifies the buffer pre-allocation code of qcom qdsp6
driver with the standard managed buffer helper.  It uses the newly
introduced fixed-size buffer allocation helper.

Cc: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Cc: Banajit Goswami <bgoswami@codeaurora.org>
Acked-by: Mark Brown <broonie@kernel.org>
Tested-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20210802072815.13551-15-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

This patch simplifies the buffer pre-allocation code of qcom lpass
driver with the standard managed buffer helper.  It uses the newly
introduced fixed-size buffer allocation helper.

Cc: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Cc: Banajit Goswami <bgoswami@codeaurora.org>
Acked-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20210802072815.13551-14-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

This patch simplifies the buffer pre-allocation code of fsl mpc5200
driver with the standard managed buffer helper.  It uses the newly
introduced fixed-size buffer allocation helper.
Acked-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20210802072815.13551-13-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

This patch simplifies the buffer pre-allocation code of fsl_dma
driver with the standard managed buffer helper.  It uses the newly
introduced fixed-size buffer allocation helper.

Cc: Nicolin Chen <nicoleotsuka@gmail.com>
Cc: Xiubo Li <Xiubo.Lee@gmail.com>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Shengjiu Wang <shengjiu.wang@gmail.com>
Acked-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20210802072815.13551-12-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

This patch simplifies the buffer pre-allocation code of fsl_asrc_dma
driver with the standard managed buffer helper.  It uses the newly
introduced fixed-size buffer allocation helper.

Cc: Nicolin Chen <nicoleotsuka@gmail.com>
Cc: Xiubo Li <Xiubo.Lee@gmail.com>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Shengjiu Wang <shengjiu.wang@gmail.com>
Acked-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20210802072815.13551-11-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

As the standard buffer allocation helper supports WC pages now, we can
convert imx-pcm-rpmsg driver to use that.  This allows us to remove
lots of superfluous code.
Acked-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20210802072815.13551-10-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

As the standard buffer allocation helper supports WC pages now, we can
convert imx-pcm-rpmsg driver to use that.  This allows us to remove
lots of superfluous code.

Cc: Nicolin Chen <nicoleotsuka@gmail.com>
Cc: Xiubo Li <Xiubo.Lee@gmail.com>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Shengjiu Wang <shengjiu.wang@gmail.com>
Acked-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20210802072815.13551-9-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

As the standard buffer allocation helper supports WC pages now, we can
convert imx-pcm-fiq driver to use that.  This allows us to remove lots
of superfluous code.

Cc: Nicolin Chen <nicoleotsuka@gmail.com>
Cc: Xiubo Li <Xiubo.Lee@gmail.com>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Shengjiu Wang <shengjiu.wang@gmail.com>
Acked-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20210802072815.13551-8-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

As the standard buffer allocation helper supports WC pages now, we can
convert bcm driver to use that.  This allows us to remove lots of
superfluous code.
Acked-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20210802072815.13551-7-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

Now with the recent addition of WC buffer allocation support, we can
use the standard PCM buffer allocation helpers for pxa2xx drivers.
This allows us to remove lots of superfluous code.

Cc: Daniel Mack <daniel@zonque.org>
Cc: Haojian Zhuang <haojian.zhuang@gmail.com>
Cc: Robert Jarzmik <robert.jarzmik@free.fr>
Acked-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20210802072815.13551-6-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

There are the generic DMA API calls for allocating and managing the
pages with the write-combined attribute.  Let's use them for all
architectures but x86; x86 still needs the special handling to
override the page attributes.
Acked-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20210802072815.13551-5-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

A few drivers want to have rather the exact buffer preallocation at
the driver probe time and keep using it for the whole operations
without allowing dynamic buffer allocation.  For satisfying the
demands, this patch extends the managed buffer allocation API
slightly.

Namely, when 0 is passed to max argument of the allocation helper
functions snd_pcm_set_managed_buffer*(), it treats as if the fixed
size allocation of the given size.  If the pre-allocation fails in
this mode, the function returns now -ENOMEM.  Otherwise, i.e. max
argument is non-zero, the function never returns -ENOMEM but tries to
fall back to the smaller chunks and allows the dynamic allocation
later -- which is still the default behavior until now.

For more intuitive use, also two new helpers are added for handling
the fixed size buffer allocation, too: snd_pcm_set_fixed_buffer() and
snd_pcm_set_fixed_buffer_all().
Acked-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20210802072815.13551-4-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

SNDRV_DMA_TYPE_DEV_UC and SNDRV_DMA_TYPE_DEV_UC_SG are incorrectly
named as if they were for the uncached memory, while actually we set
the pages as write-combined.  Rename them to reflect the right
attribute.
Acked-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20210802072815.13551-3-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

Return the pointer directly from alloc ops instead of setting
dmab->area at each place.  It simplifies the code a bit.
Acked-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20210802072815.13551-2-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

Don't populate array names_to_check on the stack but instead it
static.  Makes the object code smaller by 56 bytes.  Also clean
up checkpatch warning by adding extra const for names_to_check
and pointer s.

Before:
    text    data     bss     dec     hex filename
  103512   34380       0  137892   21aa4 ./sound/usb/mixer.o

After:
    text    data     bss     dec     hex filename
  103264   34572       0  137836   21a6c ./sound/usb/mixer.o

(gcc version 10.2.0)
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Link: https://lore.kernel.org/r/20210803122839.7143-1-colin.king@canonical.comSigned-off-by: Takashi Iwai <tiwai@suse.de>

Just briefly described about the new option.

Link: https://lore.kernel.org/r/20210729074404.19728-6-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

This patch adds a new module option, quirk_flags, for allowing user to
try some additional device-specific quirk behavior more easily.
When this option is set to non-zero, it overrides the quirk_flags, and
the specific workaround is applied.

Link: https://lore.kernel.org/r/20210729074404.19728-5-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

The generic DSD raw detection is based on the known allow list, and we
can integrate it into quirk_flags, too.

Link: https://lore.kernel.org/r/20210729074404.19728-4-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

The mixer code has a flag ignore_ctl_error for ignoring the errors
returned from the device wrt mixer accesses, and this is set from the
entries in mixer_maps.c, as well as ignore_ctl_error module option.
Those can be well integrated into the new quirk_flags field, too.

Link: https://lore.kernel.org/r/20210729074404.19728-3-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

The auto-suspend suppression workaround for Lenovo machines are
handled in quirks-table.h.  Now it's more easier to handle with
quirk_flags.

Link: https://lore.kernel.org/r/20210729074404.19728-2-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

The rate validation at the device probe is applied only to the
specific devices (currently only for MOTU devices), and this check can
be moved to quirk_flags gracefully, too.

Link: https://lore.kernel.org/r/20210729074404.19728-1-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

Yet another delay is applied at switching the interface.  This can be
moved to quirk_flags, too.

Link: https://lore.kernel.org/r/20210729073855.19043-10-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

We apply some delay for the control messages on certain devices as a
workaround, and this can be moved into the quirk_flags as well.
Currently there are three different delay periods (1ms, 5ms and 20ms),
so three different quirk bits are assigned for them.

Link: https://lore.kernel.org/r/20210729073855.19043-9-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

The devices with ITF-USB DSD type are listed in another function, and
this can be integrated into the quirk_flags easily.

Link: https://lore.kernel.org/r/20210729073855.19043-8-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

There are a couple of device-specific quirks in the clock setup code,
and those can be moved gracefully to quirk_flags, too.

Link: https://lore.kernel.org/r/20210729073855.19043-7-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

The snd_usb_audio.playback_first flag is used by the implicit feedback
mode handling, and this can be also moved to quirk_flags.

Link: https://lore.kernel.org/r/20210729073855.19043-6-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

There is another quirk for the transfer, and that's currently specific
to Zoom R16/24, handled in create_standard_audio_quirk().  Let's move
this also to the new quirk_flags.

Link: https://lore.kernel.org/r/20210729073855.19043-5-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

The txfr_quirk field was meant for aligning the transfer, and it's set
for certain devices in quirks-table.h.  Now we can move that stuff
also to the new quirk_flags gracefully, and reduce the quirks-table.h
entries (that are exposed to module device table).

As the quirks-table.h entries are also with the name string override,
provide the corresponding entries to the usb_audio_names[] table,
too.

Link: https://lore.kernel.org/r/20210729073855.19043-4-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

The devices that can have media-controller API entries are currently
specified via tables in quirks-table.h, as a part of descriptor
override.  This can fit better to the new quirk_flags, as we just need
a matching with the given ID and create the MC entries accordingly.

Link: https://lore.kernel.org/r/20210729073855.19043-3-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

As more and more device-specific workarounds came up and gathered in
various places, it becomes harder to manage.  Now it's time to clean
up and collect workarounds more consistently and make them more easily
applicable.

This patch is the first step for that: a new field quirk_flags is
introduced in snd_usb_audio struct to contain the bit flags for
various device-specific quirks.  Those are separate one from the
quirks in quirks-table.h; the quirks-table.h entries are for more
intrusive stuff that needs the descriptor override, while the new
quirk_flags is for easier ones that are tied with the vendor:product
IDs.

In this patch, as the first example, we convert the list of devices
and vendors to ignore GET_SAMPLE_RATE, formerly defined in
snb_usb_get_sample_rate_quirk().

Link: https://lore.kernel.org/r/20210729073855.19043-2-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

The recent code refactoring made the mmap of continuous pages to be
done via the own helper snd_dma_continuous_mmap() with
remap_pfn_range().  There I overlooked that dmab->addr isn't set for
the allocation with SNDRV_DMA_TYPE_CONTINUOUS.  This resulted always
in an error at mmap with this buffer type on the system such as
Intel SST Baytrail driver.

This patch fixes the regression by passing the correct address.

Fixes: 30b7ba69 ("ALSA: core: Add continuous and vmalloc mmap ops")
Reported-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/8d6674da-7d7b-803e-acc9-7de6cb1223fa@redhat.com
Link: https://lore.kernel.org/r/20210801113801.31290-1-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

At the transition to the devres-managed card release, we've put the
check of double-free at trigger_card_release().  But this wasn't
enough, as the code path calls snd_card_free() again, and it would
lead to the doubly snd_card_free() calls.

Actually the v1 patch was correct to handle this, but I forgot that
corner case and moved the check to the more obvious place as I thought
it's clearer.  But, as usual, devils live in details.

This patch corrects the check of the double-free to the right place,
with a bit more comments.

Fixes: e8ad415b ("ALSA: core: Add managed card creation")
Link: https://lore.kernel.org/r/20210731083446.26680-1-tiwai@suse.deSigned-off-by: Takashi Iwai <tiwai@suse.de>

Prefer using '"%s...", __func__' to get current function's name in
an output message.
Signed-off-by: Jason Wang <wangborong@cdjrlc.com>
Link: https://lore.kernel.org/r/20210731052647.141046-1-wangborong@cdjrlc.comSigned-off-by: Takashi Iwai <tiwai@suse.de>