It is waiting udelay(50) x 1024 (= 50ms) for status check
in worst case, but it is overkill.
And we shouldn't use udelay() for 50us
(linux/Documentation/timers/timers-howto.txt)
Waiting maximum udelay(5) x 1024 (= 5ms) is very enough
for status check.
This patch fixes these issue.
Reported-by: Hiromitsu Yamasaki <hiromitsu.yamasaki.ym@renesas.com>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Current ssi.c is assuming below 2 things to probing HDMI node.
1) remote node is including "hdmi0" or "hdmi1" in node name
2) remote_ep->full_name is including full path name
But, these assumptions are broken by below
1) Node names should not use numerical suffixes
commit 6b5ac2f1cb ("arm64: dts: renesas: r8a7795: Drop bogus HDMI
node names suffixes")
2) node full_name no longer include full path name
commit a7e4cfb0a7 ("of/fdt: only store the device node basename
in full_name")
Because of these reasons, ssi.c can't probe HDMI on current kernel.
This patch probes HDMI0/1 by using its address.
Note is that we need to keep updating for this address for future
generation chip.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Nguyen Viet Dung <dung.nguyen.aj@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
SSI/SRC have under/over flow error handling, and its status is useful
for debuging. But sometimes it might be too much message,
and it might blocks necessity other information.
To avoid such situation, basically this patch indicates interrupt
status debug message if DEBUG was defined, but it will be suppressed
if RSND_DEBUG_NO_IRQ_STATUS was defined.
Reported-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
SSI had shared counting pointer position method between PIO/DMA mode
before. But now DMA mode is using DMAEngine feature to get it.
Thus, this counting pointer position method is needed for only PIO mode.
We don't need to share code anymore.
This patch names PIO related functions as rsnd_ssi_pio_xxx(), and
merged/cleanuped each feature.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Current rsnd driver is judging 16bit/24bit data by using
runtime->sample_bits, but it is indicating physical size,
not format size. This is confusable code.
This patch uses snd_pcm_format_width() to be more correct code.
Tested-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
period_pos can always be calculated by byte_pos and
byte_per_period, there is no reason to maintain this
variable in rsnd_dai_stream.
This patch removes period_pos from rsnd_ssi and calculates
next_period_byte with consideration of actual byte_pos value.
Signed-off-by: Jiada Wang <jiada_wang@mentor.com>
Acked-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Currently there is race condition between set of byte_pos and wrap
it around when new buffer starts. If .pointer is called in-between
it will result in inconsistent pointer position be returned
from .pointer callback.
This patch increments buffer pointer atomically to avoid this issue.
Signed-off-by: Jiada Wang <jiada_wang@mentor.com>
Reviewed-by: Takashi Sakamoto <takashi.sakamoto@miraclelinux.com>
Acked-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
SSI parent mod might be NULL. ssi_parent_mod() needs to care
about it. Otherwise, it uses negative shift.
This patch fixes it.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
When stop case, it was Playback, it need to check all data were
completely sent. But in Capture case, it might not receive data
anymore. SSISR::DIRQ check is not need for Capture case.
Reported-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
SSI is using rsnd_ssi_flags_xxx() macro to control flags.
But it is useful macro not only for SSI. This patch replace it
to more generic rsnd_flags_xxx().
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
The current device tree representation of the R-Car SSI assumes that they
are numbered consecutively, starting from 0. Alas, this is not the case
with the R8A77995 (D3) SoC which SSI1/SSI2 aren't present. In order to
keep the existing device trees working, I'm suggesting to use a disabled
node for SSI0/SSI1. Teach the SSI probe to just skip disabled nodes.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Each module's dai callback function availability is controlled
by mod->status. For example "always called", "call once".
In .probe/.remove case, it needs to be called always, because
.probe will call xxx_attach() function on .probe, especially
if platform is using MIXer.
For example, below case, MIX0/DVC0/SSI0 needs to be called twice.
playback = <&src0 &ctu02 &mix0 &dvc0 &ssi0>;
playback = <&src2 &ctu03 &mix0 &dvc0 &ssi0>;
But in this case, SSI0 will call request_irq() twice.
This patch add new RSND_SSI_PROBED flag and control it
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Current rsnd_ssi_mode_flags() is not useful. This patch adds
new rsnd_ssi_flags_has/add() macro for it.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
commit 701172dca1 ("ASoC: rsnd: don't use devm_request_irq() for SSI")
exchanged devm_request_irq() to request_irq() for SSI, because SSI will
fallback into PIO mode if DMA doesn't work.
But, because of it, PIO mode needed to call free_irq() when removing,
not only DMA mode.
This patch call free_irq() both PIO/DMA. Otherwise, rsnd IRQ handler
will be increased if user tried many unbind/bind.
Fixes: 701172dca1 ("ASoC: rsnd: don't use devm_request_irq() for SSI")
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
commit 1f8754d4da ("ASoC: rsnd: don't call free_irq() on
Parent SSI") fixed Parent SSI duplicate free_irq().
But on Renesas Sound, not only Parent SSI but also Multi SSI
have same issue.
This patch avoid duplicate free_irq() if it was not pure SSI.
Fixes: 1f8754d4da ("ASoC: rsnd: don't call free_irq() on Parent SSI")
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
In case of SSI0 playback, SSI1 capture, SSI0 might be shared for
clock output if clock master mode.
Current rsnd driver had been assumed that SSI clock contiguous
output which is needed for SSI parent needs SSICR::EN (SSI module
enable) bit.
But, this bit controls data input/output, not for clock.
Clock contiguous output needs SSICR : FORCE, SCKD, SWSD,
and SSIWSR : CONT. Not SSICR : EN.
Because of this wrong assumption, and insufficient control, on current
code, for example, if it starts SSI0(playback) -> SSI1(capture) order,
SSI0 SSICR::EN bit will temporarily be 0.
It causes playback side underrun error. This is bug.
We can reproduce this issue with SSI+SRC (without DVC), and capture
during playback operation.
This patch fixup current (wrong) assumption, and control SSICR::EN bit
correctly.
Reported-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
For example SSI0/SSI1 case, SSI1 can share pin with SSI0.
And then, SSI1 needs SSI0's clock.
This clock controlling is very picky and difficult to understand
in current code.
This patch adds new rsnd_ssi_can_output_clk() macro,
the code will be more understandable.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Make it the same style as other functions
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
for_each_child_of_node performs an of_node_get on each iteration, so a
jump out of the loop requires an of_node_put.
The semantic patch that fixes this problem is as follows
(http://coccinelle.lip6.fr):
// <smpl>
@@
local idexpression n;
expression e,e1;
identifier l;
@@
for_each_child_of_node(e1,n) {
...
(
of_node_put(n);
|
e = n
|
+ of_node_put(n);
? goto l;
)
...
}
...
l: ... when != n
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Mark Brown <broonie@kernel.org>
This patch fixes this WARNING
sound/soc/sh/rcar/ssi.c:285:5-14: WARNING: Unsigned expression\
compared with zero: main_rate < 0
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Current Renesas sound driver is assuming that all Sampling rate and
channles are possible to use, but these are depends on inputed clock
and SSI connection situation.
For example, if it is using 1 SSI, enabled TDM mode and has 12288000
input clock, 2ch output can support until 192000Hz, but 6ch output can
support until 64000Hz, 8ch can support 48000Hz.
To control these situation correctly, it needs to support
hw_constraints / refine feature.
To support such feature, this patch adds new rsnd_ssi_clk_query().
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Current Renesas sound driver has slots and slots_num in
struct rsnd_dai, but these are very un-understandable naming
(It had named from TDM slots).
In this driver, the "slots" means total usable channels, and
"stot_num" means SSI lane number if Multi SSI was used.
To more understandable code, this patch renames "slots" to
"max_channels", and "slots_num" to "ssi_lane", and replaces related
functions name.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Current rsnd driver is sharing pointer related code between
PIO / DMA. But, it is used only PIO mode now, no longer needed.
This patch cleanup these.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Current rsnd driver updates pointer when DMA transfer was finished
in DMA transfer mode. But PulseAudio requests more accurate
pointer update when timer mode.
This patch consider about DMA transfer residue and update more
accurate pointer.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
rsnd_ssi_non_ops is never used. Let's remove it
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
commit 90431eb49b ("ASoC: rsnd: don't use PDTA bit for 24bit on SSI")
fixups 24bit mode data alignment, but PIO was not cared.
This patch fixes PIO mode 24bit data alignment
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Renesas R-Car Gen3 can output HDMI sound if SSIU/SSI are connected to
R-Car built-in HDMI device (R-Car Gen3 built-in HDMI device will be
controlled by DRM/KMS driver).
If SSIx was connected to HDMI0/1 on DT, SSI driver will detect it
automatically by this patch.
Note is that now Renesas R-Car sound driver is assuming that it is
using OF-graph base simple card for HDMI sound.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
If SSI uses shared pin, some SSI will be used as parent SSI.
Then, normal SSI's remove and Parent SSI's remove
(these are same SSI) will be called when unbind or remove timing.
In this case, free_irq() will be called twice.
This patch solve this issue.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Reported-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Current SSI uses PDTA bit which indicates data that Input/Output
data are Right-Aligned. But, 24bit sound should be Left-Aligned
in this HW. Because Linux is using Right-Aligned data, and HW uses
Left-Aligned data, current 24bit data is missing lower 8bit.
To fix this issue, this patch removes PDTA bit, and shift 8bit
in necessary module
Reported-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
R-Car Datasheet is indicating "SSICR.CKDV = 000 is invalid when
SSIWSR.WS_MODE = 1 or SSIWSR.CONT = 1".
Current driver will set CONT, thus, we shouldn't use CKDV = 000.
This patch fixup it.
Reported-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
ssi->usrcnt will be updated on snd_soc_dai_ops::trigger,
but snd_pcm_ops::hw_params will be called *before* it.
Thus, ssi->usrcnt is still 0 when 1st call.
rsnd_ssi_hw_params() needs to check its called count, this means
trigger should be if (ssi->usrcnt) instead of if (ssi->usrcnt > 1).
Reported-by: Nguyen Viet Dung <nv-dung@jinso.co.jp>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
DMA mod is now connected to stream via rsnd_dai_connect().
This means DMA mod can use .remove for its clearance.
rsnd_dma_detach() is no longer needed.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
SSI will use DMA mode, and migh be fallback to PIO mode.
Using devm_request_irq() makes its operation more complex when
it fallbacks to PIO mode.
Let's use manual request_irq()/free_irq()
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Current Renesas Sound driver is based on DeviceTree, and no one is
using this driver from non DT. Non-DT support is no longer needed.
Let's remove it.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
commit 1a5658c213 ("ASoC: rsnd: count .probe/.remove for
rsnd_mod_call()") solved multi-resource-free issue, by putting
.probe/.remove under count control. But,it breaks sound mixing case
(if it was used under DPCM). In such case, it uses MIXn/DVCn/SSIn,
and these should be always probed.
This patch reverted above patch, and solved the same issue by
modifing _rsnd_kctrl_remove() function.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Use of_property_read_bool to check for the existence of a property.
The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)
// <smpl>
@@
expression e1,e2;
statement S2,S1;
@@
- if (of_get_property(e1,e2,NULL))
+ if (of_property_read_bool(e1,e2))
S1 else S2
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Mark Brown <broonie@kernel.org>
SRC will convert rate, and then, CMD and SSI want to know its
rate (= SRC.in / SRC.out) for each purpose.
Current driver is supporting only Playback, but SRC+Capture support
needs more flexibility.
This patch adds rsnd_src_get_in/out_rate() for it.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Current SSI is supporting Normal SSI/Multi mode SSI/TDM mode SSI
and its behavior is based on input channels.
This input channel might be converted by CTU,
and SSI needs to be Multi SSI mode / TDM SSI mode if 6ch input
EX) 6ch input, CTU for 2ch, playback
6ch 6ch 2ch 2ch 2ch 2ch
-> SRC -> CTU -> MIX -> DVC -> SSIU -> SSI
EX) 6ch input, no CTU, Multi SSI, playback
6ch 6ch 6ch 6ch 6ch 2ch
-> SRC -> CTU -> MIX -> DVC -> SSIU -> SSI0/SSI1/SSI2
Current driver is using rsnd_get_adinr_chan() / rsnd_get_slot_width()
for this purpose, but it is complicated enough without meaning.
This patch adds new rsnd_runtime_channel_xxx() which is caring
CTU/Multi SSI.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Current rsnd supports multi SSI (maximum 4 SSI for 8ch),
and, it should determine whether using each SSI or not in runtime.
All SSIs are not used even if there are 4 SSI in case of stereo.
Current driver setups un-used SSI in such case. It is no problem,
but not needed. This patch judges it.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Current rsnd supports multi SSI (maximum 4 SSI for 8ch),
and, it should determine whether using each SSI or not in runtime.
Current judgement is vague, and had broken by
c308abe45e2("ASoC: rsnd: rsnd_ssi_is_multi_slave() macro uses
rsnd_ssi_multi_slaves()")
This patch makes clean it, and solve this issue.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
prepare for runtime judging for SSI work
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Parent SSI is needed if it is PIN sharing and clock master,
otherwise, not needed. But, whether clockk master is judged on
.set_fmt, thus, it can't call rsnd_ssi_parent_attach() on .probe.
Now, .pcm_new will be called after .set_fmt, so this patch reuses it
at this point.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
SSI want to have SSIWSR settings and SSICR settings without EN bit
when init, and SSICR EN bit only when start timing.
Otherwise, SSI output signal might be unstable.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
commit b5b442abd9 ("ASoC: rsnd: add .irq callback") added .irq callback
but SSI DMA is missing it. This patch adds it.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
commit b5b442abd9 ("ASoC: rsnd: add .irq callback")
added .irq support, and it cares both parent SSI and normal SSI.
But it should care only normal SSI. Otherwise SSI might be
forever loop if SSI is used as both parent SSI and normal SSI
(= 2 users), and if under/over run error happen. Because irq disable
do nothing in such case. This patch solve this issue.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
To reduce confusion, SSI uses "mod" instead of "ssi"
as function parameter
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Kuninori Morimoto