summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2016-02-20ASoC: rsnd: judge work SSI in runtimeKuninori Morimoto
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>
2016-02-20ASoC: rsnd: judge multi SSI in runtimeKuninori Morimoto
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>
2016-02-20ASoC: rsnd: move rsnd_ssi_irq() positionKuninori Morimoto
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>
2016-02-20ASoC: rsnd: Parent SSI attach is not needed if not clock masterKuninori Morimoto
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>
2016-02-20ASoC: rsnd: tidyup SSI init/start sequenceKuninori Morimoto
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>
2016-02-20ASoC: rsnd: add missing .irq callback for DMAKuninori Morimoto
commit b5b442abd9d5 ("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>
2016-02-20ASoC: rsnd: fixup forever loop bug on SSIKuninori Morimoto
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>
2016-02-20ASoC: rsnd: SSI function parameter uses "mod" instead of "ssi"Kuninori Morimoto
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>
2016-02-20ASoC: rsnd: indicates status failed SSIKuninori Morimoto
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-02-20ASoC: rsnd: setup SRC_ROUTE_MODE0 before SRC_SRCIRKuninori Morimoto
SRC_ROUTE_MODE0 determines whether to use SRC. Thus, it should be setup before SRC_SRCIR. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-02-20ASoC: rsnd: rename RSND_DVC_CHANNELS to RSND_MAX_CHANNELSKuninori Morimoto
The channels number is not only for DVC. Let's rename it. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-02-09ASoC: rsnd: rsnd_ssi_is_multi_slave() macro uses rsnd_ssi_multi_slaves()Kuninori Morimoto
b4c83b171 ("ASoC: rsnd: add Multi channel support") added Multi channel support, and current rsnd_ssi_is_multi_slave()'s check method is !SSI equals SSIM1/2/3. But, SSI parent also hit to this macro. Because of this reason, some stream which needs SSI parent clock can't work correctly. This patch uses rsnd_ssi_multi_slaves() to solve this issue. This issue was reported by Dung. 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>
2016-01-27ASoC: rsnd: add .irq callbackKuninori Morimoto
Current rsnd driver has .init/.start/.stop/.quit callbacks, and it needs many IPs (SRC/CTU/MUX/DVC/CMD/SSIU/SSI). Because of these relationship, it might get unnecessary error IRQ when start/stop. This patch adds new .irq callback and control IRQ enable/disable timing to avoid it. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-27ASoC: rsnd: don't auto-recover when under/over run errorKuninori Morimoto
Renesas R-Car sound needs recovery (= restart) when under/over run error occurred, and current driver tries it on under/over run error handler automatically. But this recovery should be handled by userland, not kernel. This patch stops XRUN when under/over run error occur, and will leave the recovery of HW in userland. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-22ASoC: rsnd: disable SRC.out/in in same timeKuninori Morimoto
commit b761bf272bce ("ASoC: rsnd: disable SRC.out only when stop timing") disabled SRC.out/in in different timing, but was based on picky HW information. Now, we have confirmed that we can disable both in the same time. This patch do it. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-22ASoC: rsnd: tidyup Playback/Capture sequenceKuninori Morimoto
Based on datasheet sequence Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-22ASoC: rsnd: ctu: settings matches to datasheetKuninori Morimoto
Current CTU settings was rough. This patch makes it match to datasheet. But do nothing at this point. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-22ASoC: rsnd: ctu: add rsnd_ctu_halt()Kuninori Morimoto
Based on datasheet process Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-22ASoC: rsnd: ctu: add rsnd_mix_activation()Kuninori Morimoto
Based on datasheet Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-22ASoC: rsnd: attach Audio-DMAC-periperi correctlyKuninori Morimoto
SSI/SRC will try to attach DMAC as Audio-DMAC or Audio-DMAC-periperi. It is fixed IP, but will be attached to each streams as different module in case of MUX (= multi sound path will be merged). This patch solves this issue. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-22ASoC: rsnd: each mod has status again for CTU/MUX supportKuninori Morimoto
SSI will be used as normal SSI or as clock parent SSI. Therefor, rsnd driver wants to control SSI and parent SSI separately. Otherwise it can't use Playback/Capture in the same time. And it has been done by c2dc47d5cf("ASoC: rsnd: rsnd_dai_stream has each mod's status insted of rsnd_mod") before. OTOH, rsnd driver doesn't want to control CTU/MUX/DVC/SSIU/SSI in separately. Otherwise, these will be re-initialized during playing if MUX merges 2 sounds. Because of these picky reasons, this patch re-defines status on each mod, and add new parent_ssi_status on rsnd_dai_stream. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-22ASoC: rsnd: select each SRC correctly for CMD data pathKuninori Morimoto
To select CMD data patch, it should use correct SRC from each stream in MUX case. But current code is selecting SRC from fixed stream. This patch solves this issue. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-22ASoC: rsnd: don't overwrite io on rsnd_cmd_init()Kuninori Morimoto
Current rsnd_cmd_init() overwrites "io" which will be used end of this function. This patch solved this issue. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-22ASoC: rsnd: always call probe/remove for MUXKuninori Morimoto
CTU/MUX/DVC/SSIU/SSI/Audio-DMAC-periperi might be used under multipath. So, probe()/remove() need to be called multiple times. This patch allows it. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-22ASoC: rsnd: try to connect connected mod is not errorKuninori Morimoto
If system uses CTU/MUX, CTU/MUX/DVC will try to connect same CMD to system, but it is not error in this case. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-22ASoC: rsnd: don't update status if rsnd_mod_call() doesn't matchKuninori Morimoto
If system uses CTU/MUX, and if probe error happened, it will try to call rsnd_dai_call(remove, ...). Then, MUX/DVC/SSIU/SSI might be called without calling rsnd_dai_call(probe, ...). Then, each mod status might be un-matched. It doesn't call un-matched remove function by this patch. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-22ASoC: rsnd: add debug message for rsnd_mod_call()Kuninori Morimoto
rsnd_mod_call() tries to call each IP's relevant function. But it is difficult to understand which function returned error. This patch adds debug message for this purpose Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-22ASoC: rsnd: use rsnd_mod_init() for ADGKuninori Morimoto
Current ADG doesn't use rsnd_mod_init(), but this limitation is no longer necessary. Let's use common rsnd_mod_init() Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-22ASoC: rsnd: remove unsed *parentKuninori Morimoto
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-01-06ASoC: rsnd: precedence error in rsnd_ssiu_init()Dan Carpenter
The bitwise OR has higher precedence than ?: so the val2 was always set to 0x2. Fixes: b4c83b171557 ('ASoC: rsnd: add Multi channel support') Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-30ASoC: rsnd: fix usrcnt decrementing bugAndrzej Hajda
Field usrcnt is unsigned so it cannot be lesser than zero. The patch fixes the check, moves it to the beginning of the function and changes return value to -EIO in case of usercnt error. The problem has been detected using proposed semantic patch scripts/coccinelle/tests/unsigned_lesser_than_zero.cocci [1]. [1]: http://permalink.gmane.org/gmane.linux.kernel/2038576 Signed-off-by: Andrzej Hajda <a.hajda@samsung.com> Acked-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-22ASoC: rsnd: add Multi channel supportKuninori Morimoto
This patch adds Multi channel support on Renesas R-Car sound. This patch is tested on Salvator-X board, but it can't use Multi channel, because supported format is different between codec chip and R-Car. Thus, it was tested on board which doesn't mount codec chip, with oscilloscope. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Acked-by: Rob Herring <robh@kernel.org> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-18ASoC: rsnd: add missing DT example for Simple Card with TDMKuninori Morimoto
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-18ASoC: rsnd: add missing DT example for Simple CardKuninori Morimoto
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-18ASoC: rsnd: add rsnd_parse_connect_common() and remove complex macroKuninori Morimoto
Current rsnd driver is using complex macro to parse DAI connection. This patch adds new rsnd_parse_connect_common() and replace current macro to it. This is prepare for multi channel support Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-18ASoC: rsnd: add rsnd_set_slot() / rsnd_get_slot_num()Kuninori Morimoto
TDM will use 6 or 8 slots on 1 SSI, and Multi channel will use 6 or 8 slots on few SSI (each SSI uses 2 slots). Thus, this adds new slot control functions which can be prepare for Multi channel support. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-18ASoC: rsnd: tidyup rsnd_get_slot_xxx() namingKuninori Morimoto
rsnd_get_slot_rdai() returns total slots (it returns 6 if total 6 channels) , and rsnd_get_slot_extend() returns extended SSI width (it returns 8 if total 6 channels). This will be used on SSI multi channel support too (It will return 2 if total 6 channels with 3 SSI). But, it is using confusable naming. This patch changes rsnd_get_slot_rdai() -> rsnd_get_slot(), rsnd_get_slot_extend() -> rsnd_get_slot_width() Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-18ASoC: rsnd: remove rsnd_get_slot_runtime()Kuninori Morimoto
Current Renesas sound driver is using rsnd_get_slot_runtime(), but it is same as runtime->channels. This patch removes rsnd_get_slot_runtime() Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-18ASoC: rsnd: SSI/SSIU use rsnd_get_slot_extend() to check TDMKuninori Morimoto
Current SSI/SSIU are using rsnd_get_slot_runtime() to check TDM, but using rsnd_get_slot_extend() is more sane. This patch fix it up Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-18ASoC: rsnd: rsnd_dai_connect() returns error if it connect to existing modKuninori Morimoto
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-18ASoC: rsnd: tidyup debug print position on rsnd_dma_attach()Kuninori Morimoto
It can't output corrent dma name *before* rsnd_mod_init(). It goes to *after* rsnd_mod_init() by this patch Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-18ASoC: rsnd: tidyup return value of rsnd_get_adinr_bit()Kuninori Morimoto
Renesas sound driver has rsnd_get_adinr_bit/chan() functions. It is assuming _bit() returns ADINR :: OTBL, and _chan() returns ADINR :: CHNUM. Current _bit() returns both OTBL and CHNUM. This patch fixup it. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-18ASoC: rsnd: fixup SSIU control timingKuninori Morimoto
SSIU should be controlled after SSI. This patch fix up it Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-17ASoC: rsrc-card: tidyup dai format for DPCMKuninori Morimoto
rsrc-card is DPCM supported version of simple-card. Thus it has similar DT format. OTOH, snd_soc_dai_link requests cpu/codec, but one of them will be snd-soc-dummy in DPCM case, and DPCM requests frontend/backend dai_link. This means it might have multi backend/codec. And, SND_SOC_DAIFMT_xxx is based on "codec". Because of these difference, current rsrc card can't detect correct dai_fmt. This patch detect correct dai fmt from 1st "codec". Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-17ASoC: rsrc-card: Remove support for setting differing DAI formatsKuninori Morimoto
1efb53a220 ("ASoC: simple-card: Remove support for setting differing DAI formats") removed set_fmt support from simple-card. rsrc-card follows same style, because it is based on simple-card. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-17ASoC: rsrc-card: enable to use tdm_slot on DTKuninori Morimoto
Renesas sound driver will use tdm slot on TDM Multi Mode support. This patch enables tdm slot on rsrc card driver on DT. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-15rcar: mux: Avoid use of ret uninitialisedMark Brown
We use ret as the return value from the rsnd_mix_probe() but if there are no child nodes and no errors then we will never initialize ret leading to build warnings. Ensure ret is initialized before we iterate over the child nodes to avoid this. Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-15rcar: dvc: Avoid use of ret uninitialisedMark Brown
We use ret as the return value from the rsnd_dvc_probe() but if there are no child nodes and no errors then we will never initialize ret leading to build warnings. Ensure ret is initialized before we iterate over the child nodes to avoid this. Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-15rcar: ctu: Avoid use of ret uninitialisedMark Brown
We use ret as the return value from the rsnd_ctu_probe() but if there are no child nodes and no errors then we will never initialize ret leading to build warnings. Ensure ret is initialized before we iterate over the child nodes to avoid this. Signed-off-by: Mark Brown <broonie@kernel.org>
2015-12-08ASoC: rsnd: tidyup data align position for captureKuninori Morimoto
L/R channel data has been treated as inverted on R-Car sound 16bit mode, Thus, 4689032b1("ASoC: rsnd: tidyup data align position") tidyuped data align position. But it couldn't care about capture case. This patch cares both playback/capture Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org>
generated by cgit (git 2.34.1) at 2025-03-18 11:08:09 +0000