- 12 Nov, 2010 1 commit
-
-
Mark Brown authored
The value makes no odds and it makes life easier with caches. Signed-off-by:
Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by:
Liam Girdwood <lrg@slimlogic.co.uk>
-
- 11 Nov, 2010 8 commits
-
-
Mark Brown authored
-
Dimitris Papastamos authored
This patch adds support for rbtree compression when storing the register cache. It does this by not adding any uninitialized registers (those whose value is 0). If any of those registers is written with a nonzero value they get added into the rbtree. Consider a sample device with a large sparse register map. The register indices are between [0, 0x31ff]. An array of 12800 registers is thus created each of which is 2 bytes. This results in a 25kB region. This array normally lives outside soc-core, normally in the driver itself. The original soc-core code would kmemdup this region resulting in 50kB total memory. When using the rbtree compression technique and __devinitconst on the original array the figures are as follows. For this typical device, you might have 100 initialized registers, that is registers that are nonzero by default. We build an rbtree with 100 nodes, each of which is 24 bytes. This results in ~2kB of memory. Assuming that the target arch can freeup the memory used by the initial __devinitconst array, we end up using about ~2kB bytes of actual memory. The memory footprint will increase as uninitialized registers get written and thus new nodes created in the rbtree. In practice, most of those registers are never changed. If the target arch can't freeup the __devinitconst array, we end up using a total of ~27kB. The difference between the rbtree and the LZO caching techniques, is that if using the LZO technique the size of the cache will increase slower as more uninitialized registers get changed. Signed-off-by:
Dimitris Papastamos <dp@opensource.wolfsonmicro.com> Acked-by:
-
Dimitris Papastamos authored
This patch adds support for LZO compression when storing the register cache. The initial register defaults cache is marked as __devinitconst and the only change required for a driver to use LZO compression is to set the compress_type member in codec->driver to SND_SOC_LZO_COMPRESSION. For a typical device whose register map would normally occupy 25kB or 50kB by using the LZO compression technique, one can get down to ~5-7kB. There might be a performance penalty associated with each individual read/write due to decompressing/compressing the underlying cache, however that should not be noticeable. These memory benefits depend on whether the target architecture can get rid of the memory occupied by the original register defaults cache which is marked as __devinitconst. Nevertheless there will be some memory gain even if the target architecture can't get rid of the original register map, this should be around ~30-32kB instead of 50kB. Signed-off-by:
-
Dimitris Papastamos authored
This patch introduces the new caching API and migrates the old caching interface into the new one. The flat register caching technique does not use compression at all and it is equivalent to the old caching technique. One can still access codec->reg_cache directly but this is not advised as that will not be portable across different caching strategies. None of the existing drivers need to be changed to adapt to this caching technique. There should be no noticeable overhead associated with using the new caching API. Signed-off-by:
-
Mark Brown authored
Trace events for DAPM allow us to monitor the performance and behaviour of DAPM with logging which can be built into the kernel permanantly, is more suited to automated analysis and display and less likely to suffer interference from other logging activity. Currently trace events are generated for: - Start and stop of DAPM processing - Start and stop of bias level changes - Power decisions for widgets - Widget event execution start and stop giving some view as to what is happening and where latencies occur. Actual changes in widget power can be seen via the register write trace in soc-core. Signed-off-by:
-
Mark Brown authored
The trace subsystem provides a convenient way of instrumenting the kernel which can be left on all the time with extremely low impact on the system unlike prints to the kernel log which can be very spammy. Begin adding support for instrumenting ASoC via this interface by adding trace for the register access primitives. Signed-off-by:
-
Mark Brown authored
Make the DAPM sequence execution look a bit nicer by factoring out the code to invoke an event into a single function since it's all the same pretty much. Signed-off-by:
-
Ryan Mallon authored
Add Kconfig dependency on AT91_PROGRAMMABLE_CLOCKS for the Atmel SoC audio SAM9G20-EK and PlayPaq boards. Fixes link errors on missing clk_set_parent and clk_set_rate when building without AT91_PROGRAMMABLE_CLOCKS. Signed-off-by:
Ryan Mallon <ryan@bluewatersys.com> Acked-by:
Geoffrey Wossum <gwossum@acm.org> Acked-by:
-
- 10 Nov, 2010 5 commits
-
-
Mark Brown authored
-
Mark Brown authored
Ensure that whatever ran before us leaves the WM835x with a sane default audio interface configuration as we do not override the companding, loopback or tristate settings and do not reset the chip at startup (as it is a PMIC). Reported-by:
Keiji Mitsuhisa <Keiji.Mitsuhisa@wolfsonmicro.com> Signed-off-by:
-
Mark Brown authored
The WM8350 driver was using some custom constants to interpret the direction of the MCLK signal which had the opposite values to those used as standard by the ASoC core, causing confusion in machine drivers such as the 1133-EV1 board. Reported-by:
Tommy Zhu <Tommy.Zhu@wolfsonmicro.com> Signed-off-by:
-
Jarkko Nikula authored
Prints from pop_dbg are enabled when dapm_pop_time != 0. Convert it to use dev_info so that parent device of DAPM context is printed. Signed-off-by:
Jarkko Nikula <jhnikula@gmail.com> Acked-by:
-
Jarkko Nikula authored
Switch printk and pr_ prints to dev_ variants. It is helpful to see parent device of DAPM context especially when there are multiple DAPM contexts (codecs currently). This is mostly simple conversion. Exceptions are in snd_soc_dapm_set_pin that prints also pin state, uniform "dapm: unknown pin" error prints from snd_soc_dapm_set_pin, snd_soc_dapm_force_enable_pin and snd_soc_dapm_ignore_suspend, and pop_dbg which is converted by an another patch. Signed-off-by:
-
- 09 Nov, 2010 1 commit
-
-
Axel Lin authored
Add soc_init_card_debugfs and soc_cleanup_card_debugfs functions to fix below error. CC sound/soc/soc-core.o sound/soc/soc-core.c: In function 'soc_probe': sound/soc/soc-core.c:1689: error: implicit declaration of function 'soc_init_card_debugfs' sound/soc/soc-core.c: In function 'soc_remove': sound/soc/soc-core.c:1718: error: implicit declaration of function 'soc_cleanup_card_debugfs' make[2]: *** [sound/soc/soc-core.o] Error 1 make[1]: *** [sound/soc] Error 2 make: *** [sound] Error 2 Signed-off-by:
Axel Lin <axel.lin@gmail.com> Acked-by:
-
- 08 Nov, 2010 5 commits
-
-
Mark Brown authored
-
Dimitris Papastamos authored
There is no need to use '&' in this case. Either way, if a is an array of some type, then a == &a == &a[0]. Signed-off-by:
-
Dimitris Papastamos authored
Instead of dereferencing a NULL function pointer and falling apart use BUG_ON() for any unimplemented hw_read() calls. Signed-off-by:
-
Marek Belisko authored
When make mini2440_defconfig compilation end with undefined references to DMA functions. There was missing selection for S3C2410_DMA when compile ASoC audio for S3C24xx CPU. Tested on mini2440 board. Signed-off-by:
Marek Belisko <marek.belisko@gmail.com> Acked-by:
-
Axel Lin authored
Signed-off-by:
-
- 06 Nov, 2010 8 commits
-
-
Mark Brown authored
-
Dimitris Papastamos authored
The WM8770 is a high performance, multi-channel audio codec. The WM8770 is ideal for surround sound processing applications for home hi-fi, automotive and other audio visual equipment. Signed-off-by:
-
Jarkko Nikula authored
Based on discussion the dapm_pop_time in debugsfs should be per card rather than per device. Single pop time value for entire card is cleaner when the DAPM sequencing is extended to cross-device paths. debugfs/asoc/{card->name}/{codec dir}/dapm_pop_time -> debugfs/asoc/{card->name}/dapm_pop_time Signed-off-by: -
Jarkko Nikula authored
Make use of sound card debugfs directory and move codec directories under the parent card debugfs directory. debugfs/asoc/{codec dir} -> debugfs/asoc/{card->name}/{codec dir}. Signed-off-by: -
Jarkko Nikula authored
There will be need to have sound card specific debugfs entries. This patch introduces a new debugfs/asoc/{card->name}/ directory but does not add yet any entries there. Signed-off-by: -
Liam Girdwood authored
Decoupling Dynamic Audio Power Management (DAPM) from codec devices is required when developing ASoC further. Such as for other ASoC components to have DAPM widgets or when extending DAPM to handle cross-device paths. This patch decouples DAPM related variables from struct snd_soc_codec and moves them to new struct snd_soc_dapm_context that is used to encapsulate DAPM context of a device. ASoC core and API of DAPM functions are modified to use DAPM context instead of codec. This patch does not change current functionality and a large part of changes come because of structure and internal API changes. Core implementation is from Liam Girdwood <lrg@slimlogic.co.uk> with some minor core changes, codecs and machine driver conversions from Jarkko Nikula <jhnikula@gmail.com>. Signed-off-by:
-
Dimitris Papastamos authored
The member reg_cache is not used at all and therefore it should be removed. This member was usually needed for older versions of ASoC that did not handle caching automatically and had to be done in the driver itself. Signed-off-by:
-
Mark Brown authored
When doing anything with the system, especially DAPM, we need to hold the CODEC mutex. Signed-off-by:
-
- 05 Nov, 2010 7 commits
-
-
Sascha Hauer authored
Whether we can do mono or not depends on the codec. No need to limit this in the ssi driver. Signed-off-by:
Sascha Hauer <s.hauer@pengutronix.de> Acked-by:
-
Mark Brown authored
-
Sascha Hauer authored
We have different codecs on the pcm038 (ac97 wm9712 and mc13783). To make alsactl restore work correctly these should have different names. Signed-off-by:
-
Sascha Hauer authored
Signed-off-by:
-
Sascha Hauer authored
We have two different transfer methods on i.MX: FIQ and DMA. Since the merge of the ASoC multicomponent support the DMA device is lost. Add it again. Also, imx_ssi_dai_probe has to be called for !AC97 aswell. Signed-off-by:
-
Sascha Hauer authored
Signed-off-by:
-
Sascha Hauer authored
Signed-off-by:
-
- 03 Nov, 2010 5 commits
-
-
Mark Brown authored
Ensure that the system does not suspend while we process a WM8962 jack event by using pm_wakeup_event() to block the suspend while we're waiting for the jack to settle. Use a slightly longer timeout than the jack waits to allow for other stuff to take over and delays in scheduling. Signed-off-by:
-
Mark Brown authored
We now have trace in the ASoC core so we don't need to our own trace in the driver. Signed-off-by:
-
Dimitris Papastamos authored
No need to print the register-value pair again, as we've already hooked snd_soc_write() for that matter. Signed-off-by:
-
Mark Brown authored
Facilitating adding trace type stuff. For a first pass add some dev_dbg() statements into them. Signed-off-by:
-
Mark Brown authored
Allow the standard soc-jack GPIO based jack handling to handle the use of GPIOs which may sleep (such as those on GPIO expanders) by converting the code to use request_any_context_irq(). Signed-off-by:
-
