[U-Boot] [PATCH v1 1/2] bfin: Remove spi dma function in bf5xx.
From: Scott Jiang scott.jiang.linux@gmail.com
BF5xx rx dma causes spi flash random read error. Accually spi controller has problems both on tx and rx dma. So remove spi dma support in u-boot.
Signed-off-by: Scott Jiang scott.jiang.linux@gmail.com Signed-off-by: Sonic Zhang sonic.zhang@analog.com --- drivers/spi/bfin_spi.c | 103 +----------------------------------------------- 1 files changed, 1 insertions(+), 102 deletions(-)
diff --git a/drivers/spi/bfin_spi.c b/drivers/spi/bfin_spi.c index e080bec..fff384a 100644 --- a/drivers/spi/bfin_spi.c +++ b/drivers/spi/bfin_spi.c @@ -13,7 +13,6 @@ #include <spi.h>
#include <asm/blackfin.h> -#include <asm/dma.h> #include <asm/gpio.h> #include <asm/portmux.h> #include <asm/mach-common/bits/spi.h> @@ -244,106 +243,10 @@ void spi_release_bus(struct spi_slave *slave) SSYNC(); }
-#ifdef __ADSPBF54x__ -# define SPI_DMA_BASE DMA4_NEXT_DESC_PTR -#elif defined(__ADSPBF533__) || defined(__ADSPBF532__) || defined(__ADSPBF531__) || \ - defined(__ADSPBF538__) || defined(__ADSPBF539__) -# define SPI_DMA_BASE DMA5_NEXT_DESC_PTR -#elif defined(__ADSPBF561__) -# define SPI_DMA_BASE DMA2_4_NEXT_DESC_PTR -#elif defined(__ADSPBF537__) || defined(__ADSPBF536__) || defined(__ADSPBF534__) || \ - defined(__ADSPBF52x__) || defined(__ADSPBF51x__) -# define SPI_DMA_BASE DMA7_NEXT_DESC_PTR -# elif defined(__ADSPBF50x__) -# define SPI_DMA_BASE DMA6_NEXT_DESC_PTR -#else -# error "Please provide SPI DMA channel defines" -#endif -static volatile struct dma_register *dma = (void *)SPI_DMA_BASE; - #ifndef CONFIG_BFIN_SPI_IDLE_VAL # define CONFIG_BFIN_SPI_IDLE_VAL 0xff #endif
-#ifdef CONFIG_BFIN_SPI_NO_DMA -# define SPI_DMA 0 -#else -# define SPI_DMA 1 -#endif - -static int spi_dma_xfer(struct bfin_spi_slave *bss, const u8 *tx, u8 *rx, - uint bytes) -{ - int ret = -1; - u16 ndsize, spi_config, dma_config; - struct dmasg dmasg[2]; - const u8 *buf; - - if (tx) { - debug("%s: doing half duplex TX\n", __func__); - buf = tx; - spi_config = TDBR_DMA; - dma_config = 0; - } else { - debug("%s: doing half duplex RX\n", __func__); - buf = rx; - spi_config = RDBR_DMA; - dma_config = WNR; - } - - dmasg[0].start_addr = (unsigned long)buf; - dmasg[0].x_modify = 1; - dma_config |= WDSIZE_8 | DMAEN; - if (bytes <= 65536) { - blackfin_dcache_flush_invalidate_range(buf, buf + bytes); - ndsize = NDSIZE_5; - dmasg[0].cfg = NDSIZE_0 | dma_config | FLOW_STOP | DI_EN; - dmasg[0].x_count = bytes; - } else { - blackfin_dcache_flush_invalidate_range(buf, buf + 65536 - 1); - ndsize = NDSIZE_7; - dmasg[0].cfg = NDSIZE_5 | dma_config | FLOW_ARRAY | DMA2D; - dmasg[0].x_count = 0; /* 2^16 */ - dmasg[0].y_count = bytes >> 16; /* count / 2^16 */ - dmasg[0].y_modify = 1; - dmasg[1].start_addr = (unsigned long)(buf + (bytes & ~0xFFFF)); - dmasg[1].cfg = NDSIZE_0 | dma_config | FLOW_STOP | DI_EN; - dmasg[1].x_count = bytes & 0xFFFF; /* count % 2^16 */ - dmasg[1].x_modify = 1; - } - - dma->cfg = 0; - dma->irq_status = DMA_DONE | DMA_ERR; - dma->curr_desc_ptr = dmasg; - write_SPI_CTL(bss, (bss->ctl & ~TDBR_CORE)); - write_SPI_STAT(bss, -1); - SSYNC(); - - write_SPI_TDBR(bss, CONFIG_BFIN_SPI_IDLE_VAL); - dma->cfg = ndsize | FLOW_ARRAY | DMAEN; - write_SPI_CTL(bss, (bss->ctl & ~TDBR_CORE) | spi_config); - SSYNC(); - - /* - * We already invalidated the first 64k, - * now while we just wait invalidate the remaining part. - * Its not likely that the DMA is going to overtake - */ - if (bytes > 65536) - blackfin_dcache_flush_invalidate_range(buf + 65536, buf + bytes); - - while (!(dma->irq_status & DMA_DONE)) - if (ctrlc()) - goto done; - - dma->cfg = 0; - - ret = 0; - done: - write_SPI_CTL(bss, bss->ctl); - return ret; -} - static int spi_pio_xfer(struct bfin_spi_slave *bss, const u8 *tx, u8 *rx, uint bytes) { @@ -395,11 +298,7 @@ int spi_xfer(struct spi_slave *slave, unsigned int bitlen, const void *dout, if (flags & SPI_XFER_BEGIN) spi_cs_activate(slave);
- /* TX DMA doesn't work quite right */ - if (SPI_DMA && bytes > 6 && (!tx /*|| !rx*/)) - ret = spi_dma_xfer(bss, tx, rx, bytes); - else - ret = spi_pio_xfer(bss, tx, rx, bytes); + ret = spi_pio_xfer(bss, tx, rx, bytes);
done: if (flags & SPI_XFER_END)
From: Scott Jiang scott.jiang.linux@gmail.com
There may be dirty data in RDBR, so we should discard invalid data. This operation also clears RXS bit in STAT register.
Signed-off-by: Scott Jiang scott.jiang.linux@gmail.com Signed-off-by: Sonic Zhang sonic.zhang@analog.com --- drivers/spi/bfin_spi.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/drivers/spi/bfin_spi.c b/drivers/spi/bfin_spi.c index fff384a..1845e4f 100644 --- a/drivers/spi/bfin_spi.c +++ b/drivers/spi/bfin_spi.c @@ -250,6 +250,8 @@ void spi_release_bus(struct spi_slave *slave) static int spi_pio_xfer(struct bfin_spi_slave *bss, const u8 *tx, u8 *rx, uint bytes) { + /* discard invalid data and clear RXS */ + read_SPI_RDBR(bss); /* todo: take advantage of hardware fifos */ while (bytes--) { u8 value = (tx ? *tx++ : CONFIG_BFIN_SPI_IDLE_VAL);
On Monday 04 March 2013 02:20:08 Sonic Zhang wrote:
From: Scott Jiang scott.jiang.linux@gmail.com
BF5xx rx dma causes spi flash random read error. Accually spi controller has problems both on tx and rx dma. So remove spi dma support in u-boot.
this is wrong, and imo, unnecessary. it's wrong because using DMA gains a lot of speed increases, and works for in many cases (i haven't seen cases where it didn't work, but i haven't been actively developing in the last year of course). it's unnecessary because there's already a CONFIG_xxx option to disable it if your platform is having problems and you can't figure things out, and don't need the speed gains. -mike
Mike,
test case is [#6887], if you save data and read out to compare it sometimes fail. I agree that we can keep this patch, but it introduces problems to test engineer. This is a random error, she is difficult to know which board and which case we should enable this function.
2013/3/12 Mike Frysinger vapier@gentoo.org:
On Monday 04 March 2013 02:20:08 Sonic Zhang wrote:
From: Scott Jiang scott.jiang.linux@gmail.com
BF5xx rx dma causes spi flash random read error. Accually spi controller has problems both on tx and rx dma. So remove spi dma support in u-boot.
this is wrong, and imo, unnecessary. it's wrong because using DMA gains a lot of speed increases, and works for in many cases (i haven't seen cases where it didn't work, but i haven't been actively developing in the last year of course). it's unnecessary because there's already a CONFIG_xxx option to disable it if your platform is having problems and you can't figure things out, and don't need the speed gains. -mike
participants (3)
-
Mike Frysinger -
Scott Jiang -
Sonic Zhang