[alsa-devel] [PATCH 6/7] ASoC: S3C64XX: Pass I2S base address from platform data
From: Jassi Brar jassi.brar@samsung.com
Instead of having I2Sv2 core to figure out the base address of I2Sv3 controller pass the address directly.
Signed-off-by: Jassi Brar jassi.brar@samsung.com --- sound/soc/s3c24xx/s3c64xx-i2s.c | 10 +++++++++- 1 files changed, 9 insertions(+), 1 deletions(-)
diff --git a/sound/soc/s3c24xx/s3c64xx-i2s.c b/sound/soc/s3c24xx/s3c64xx-i2s.c index 751a1b6..9cf62c2 100644 --- a/sound/soc/s3c24xx/s3c64xx-i2s.c +++ b/sound/soc/s3c24xx/s3c64xx-i2s.c @@ -166,6 +166,13 @@ static __devinit int s3c64xx_iis_dev_probe(struct platform_device *pdev) dev_err(&pdev->dev, "Unable to get I2S SFR address\n"); return -ENXIO; } + + if (!request_mem_region(res->start, resource_size(res), + "s3c64xx-i2s")) { + dev_err(&pdev->dev, "Unable to request SFR region\n"); + return -EBUSY; + } + i2s->dma_capture->dma_addr = res->start + S3C2412_IISRXD; i2s->dma_playback->dma_addr = res->start + S3C2412_IISTXD;
@@ -191,7 +198,8 @@ static __devinit int s3c64xx_iis_dev_probe(struct platform_device *pdev)
clk_enable(i2s->iis_cclk);
- ret = s3c_i2sv2_probe(pdev, dai, i2s, 0); + ret = s3c_i2sv2_probe(pdev, dai, i2s, + i2s->dma_playback->dma_addr - S3C2412_IISTXD); if (ret) goto err_clk;
On Wed, Dec 09, 2009 at 01:29:54PM +0900, jassisinghbrar@gmail.com wrote:
- if (!request_mem_region(res->start, resource_size(res),
"s3c64xx-i2s")) {dev_err(&pdev->dev, "Unable to request SFR region\n");return -EBUSY;- }
This feels like it should be in the patch that requested the resource rather than here. Probably ought to have a version that does the request in 2.6.33 too if we can...
@@ -191,7 +198,8 @@ static __devinit int s3c64xx_iis_dev_probe(struct platform_device *pdev)
clk_enable(i2s->iis_cclk);
- ret = s3c_i2sv2_probe(pdev, dai, i2s, 0);
- ret = s3c_i2sv2_probe(pdev, dai, i2s,
i2s->dma_playback->dma_addr - S3C2412_IISTXD);
A comment explaining why we're doing the subtraction wouldn't hurt; the code is OK.
On Wed, Dec 9, 2009 at 8:09 PM, Mark Brown broonie@opensource.wolfsonmicro.com wrote:
On Wed, Dec 09, 2009 at 01:29:54PM +0900, jassisinghbrar@gmail.com wrote:
- if (!request_mem_region(res->start, resource_size(res),
- "s3c64xx-i2s")) {
- dev_err(&pdev->dev, "Unable to request SFR region\n");
- return -EBUSY;
- }
This feels like it should be in the patch that requested the resource rather than here. Probably ought to have a version that does the request in 2.6.33 too if we can...
that patch is only for setting DMA channel and dst addresses, so we request MEM resource there.
@@ -191,7 +198,8 @@ static __devinit int s3c64xx_iis_dev_probe(struct platform_device *pdev)
clk_enable(i2s->iis_cclk);
- ret = s3c_i2sv2_probe(pdev, dai, i2s, 0);
- ret = s3c_i2sv2_probe(pdev, dai, i2s,
- i2s->dma_playback->dma_addr - S3C2412_IISTXD);
if (ret) goto err_clk;
A comment explaining why we're doing the subtraction wouldn't hurt; the code is OK.
its just that i didn't want to reorder the code and MEM resource was requested into a temporary variable 'res' not immediately before this call. At this point res->start should do too though.
On Wed, Dec 09, 2009 at 08:14:24PM +0900, jassi brar wrote:
On Wed, Dec 9, 2009 at 8:09 PM, Mark Brown
- if (!request_mem_region(res->start, resource_size(res),
- "s3c64xx-i2s")) {
- dev_err(&pdev->dev, "Unable to request SFR region\n");
- return -EBUSY;
- }
This feels like it should be in the patch that requested the resource rather than here. Probably ought to have a version that does the request in 2.6.33 too if we can...
that patch is only for setting DMA channel and dst addresses, so we request MEM resource there.
I'm not sure I follow? In any case the real issue here is that looking at this change and it's hard to see how it's immediately related to the subject of the patch; it's not about how the data is passed and somewhat distant from the code which does do the passing.
A comment explaining why we're doing the subtraction wouldn't hurt; the code is OK.
its just that i didn't want to reorder the code and MEM resource was requested into a temporary variable 'res' not immediately before this call. At this point res->start should do too though.
Yes, I saw - all I'm saying is that putting a note in the code
participants (3)
-
jassi brar -
jassisinghbrar＠gmail.com
-
Mark Brown