| Message ID | 1364998113-13428-4-git-send-email-amarendra.xt@samsung.com |
|---|---|
| State | New |
| Headers | show |
Hi Amar, I'm not sure that need to enable the auto-stop command feature. why do you enable this feature? Is there any benefit? Is it related with read/write issue? Best Regards, Jaehoon Chung On 04/03/2013 11:08 PM, Amar wrote: > This patch enumerates dwmci and set auto stop command during > dwmci initialisation. > EMMC read/write is not happening in current implementation > due to improper fifo size computation. Hence modified the fifo size > computation to resolve EMMC read write issues. > > Signed-off-by: Amar <amarendra.xt@samsung.com> > --- > Changes since V1: > 1)Created the macros RX_WMARK_SHIFT and RX_WMARK_MASK in header file. > > Changes since V2: > 1)Updation of commit message and resubmition of proper patch set. > > Changes since V3: > 1)Updated to use the macro DWMCI_CTRL_SEND_AS_CCSD instead of > the hard coded value (1 << 10). > > Changes since V4: > 1)Updated the function dwmci_send_cmd() to use get_timer() instead > of using mdelay(1). > > Changes since V5: > 1)Updated in response to review comments. > > Changes since V6: > No change. > > Changes since V7: > 1)Updated the function dwmci_setup_bus() to return 0 if (freq == 0). > This is to avoid the run time exception "raise:Signal # 8 caught". > > drivers/mmc/dw_mmc.c | 28 ++++++++++++++++++---------- > 1 file changed, 18 insertions(+), 10 deletions(-) > > diff --git a/drivers/mmc/dw_mmc.c b/drivers/mmc/dw_mmc.c > index 4070d4e..963a515 100644 > --- a/drivers/mmc/dw_mmc.c > +++ b/drivers/mmc/dw_mmc.c > @@ -129,13 +129,13 @@ static int dwmci_send_cmd(struct mmc *mmc, struct mmc_cmd *cmd, > unsigned int timeout = 100000; > u32 retry = 10000; > u32 mask, ctrl; > + ulong start = get_timer(0); > > while (dwmci_readl(host, DWMCI_STATUS) & DWMCI_BUSY) { > - if (timeout == 0) { > + if (get_timer(start) > timeout) { > printf("Timeout on data busy\n"); > return TIMEOUT; > } > - timeout--; > } > > dwmci_writel(host, DWMCI_RINTSTS, DWMCI_INTMSK_ALL); > @@ -143,7 +143,6 @@ static int dwmci_send_cmd(struct mmc *mmc, struct mmc_cmd *cmd, > if (data) > dwmci_prepare_data(host, data); > > - > dwmci_writel(host, DWMCI_CMDARG, cmd->cmdarg); > > if (data) > @@ -231,9 +230,8 @@ static int dwmci_setup_bus(struct dwmci_host *host, u32 freq) > int timeout = 10000; > unsigned long sclk; > > - if (freq == host->clock) > + if ((freq == host->clock) || (freq == 0)) > return 0; > - > /* > * If host->mmc_clk didn't define, > * then assume that host->bus_hz is source clock value. > @@ -314,7 +312,7 @@ static void dwmci_set_ios(struct mmc *mmc) > static int dwmci_init(struct mmc *mmc) > { > struct dwmci_host *host = (struct dwmci_host *)mmc->priv; > - u32 fifo_size, fifoth_val; > + u32 fifo_size, fifoth_val, ier; > > dwmci_writel(host, DWMCI_PWREN, 1); > > @@ -323,6 +321,14 @@ static int dwmci_init(struct mmc *mmc) > return -1; > } > > + /* Enumerate at 400KHz */ > + dwmci_setup_bus(host, mmc->f_min); > + > + /* Set auto stop command */ > + ier = dwmci_readl(host, DWMCI_CTRL); > + ier |= DWMCI_CTRL_SEND_AS_CCSD; > + dwmci_writel(host, DWMCI_CTRL, ier); > + > dwmci_writel(host, DWMCI_RINTSTS, 0xFFFFFFFF); > dwmci_writel(host, DWMCI_INTMASK, 0); > > @@ -332,11 +338,13 @@ static int dwmci_init(struct mmc *mmc) > dwmci_writel(host, DWMCI_BMOD, 1); > > fifo_size = dwmci_readl(host, DWMCI_FIFOTH); > - if (host->fifoth_val) > + fifo_size = ((fifo_size & RX_WMARK_MASK) >> RX_WMARK_SHIFT) + 1; > + if (host->fifoth_val) { > fifoth_val = host->fifoth_val; > - else > - fifoth_val = MSIZE(0x2) | RX_WMARK(fifo_size/2 -1) | > - TX_WMARK(fifo_size/2); > + } else { > + fifoth_val = MSIZE(0x2) | RX_WMARK(fifo_size / 2 - 1) | > + TX_WMARK(fifo_size / 2); > + } > dwmci_writel(host, DWMCI_FIFOTH, fifoth_val); > > dwmci_writel(host, DWMCI_CLKENA, 0); >
Hi Jaehoon, Auto-stop command is not related to read/write issue. Today, I commented the 'auto-stop command' enable code and tested, SD/EMMC read write and EMMC boot work fine. Hence enabling Auto-stop command is not required. I shall update the same in next patchset. Thanks & Regards Amarendra Reddy On 9 April 2013 16:19, Jaehoon Chung <jh80.chung@samsung.com> wrote: > Hi Amar, > > I'm not sure that need to enable the auto-stop command feature. > why do you enable this feature? Is there any benefit? > Is it related with read/write issue? > > Best Regards, > Jaehoon Chung > > On 04/03/2013 11:08 PM, Amar wrote: > > This patch enumerates dwmci and set auto stop command during > > dwmci initialisation. > > EMMC read/write is not happening in current implementation > > due to improper fifo size computation. Hence modified the fifo size > > computation to resolve EMMC read write issues. > > > > Signed-off-by: Amar <amarendra.xt@samsung.com> > > --- > > Changes since V1: > > 1)Created the macros RX_WMARK_SHIFT and RX_WMARK_MASK in header > file. > > > > Changes since V2: > > 1)Updation of commit message and resubmition of proper patch set. > > > > Changes since V3: > > 1)Updated to use the macro DWMCI_CTRL_SEND_AS_CCSD instead of > > the hard coded value (1 << 10). > > > > Changes since V4: > > 1)Updated the function dwmci_send_cmd() to use get_timer() instead > > of using mdelay(1). > > > > Changes since V5: > > 1)Updated in response to review comments. > > > > Changes since V6: > > No change. > > > > Changes since V7: > > 1)Updated the function dwmci_setup_bus() to return 0 if (freq == > 0). > > This is to avoid the run time exception "raise:Signal # 8 caught". > > > > drivers/mmc/dw_mmc.c | 28 ++++++++++++++++++---------- > > 1 file changed, 18 insertions(+), 10 deletions(-) > > > > diff --git a/drivers/mmc/dw_mmc.c b/drivers/mmc/dw_mmc.c > > index 4070d4e..963a515 100644 > > --- a/drivers/mmc/dw_mmc.c > > +++ b/drivers/mmc/dw_mmc.c > > @@ -129,13 +129,13 @@ static int dwmci_send_cmd(struct mmc *mmc, struct > mmc_cmd *cmd, > > unsigned int timeout = 100000; > > u32 retry = 10000; > > u32 mask, ctrl; > > + ulong start = get_timer(0); > > > > while (dwmci_readl(host, DWMCI_STATUS) & DWMCI_BUSY) { > > - if (timeout == 0) { > > + if (get_timer(start) > timeout) { > > printf("Timeout on data busy\n"); > > return TIMEOUT; > > } > > - timeout--; > > } > > > > dwmci_writel(host, DWMCI_RINTSTS, DWMCI_INTMSK_ALL); > > @@ -143,7 +143,6 @@ static int dwmci_send_cmd(struct mmc *mmc, struct > mmc_cmd *cmd, > > if (data) > > dwmci_prepare_data(host, data); > > > > - > > dwmci_writel(host, DWMCI_CMDARG, cmd->cmdarg); > > > > if (data) > > @@ -231,9 +230,8 @@ static int dwmci_setup_bus(struct dwmci_host *host, > u32 freq) > > int timeout = 10000; > > unsigned long sclk; > > > > - if (freq == host->clock) > > + if ((freq == host->clock) || (freq == 0)) > > return 0; > > - > > /* > > * If host->mmc_clk didn't define, > > * then assume that host->bus_hz is source clock value. > > @@ -314,7 +312,7 @@ static void dwmci_set_ios(struct mmc *mmc) > > static int dwmci_init(struct mmc *mmc) > > { > > struct dwmci_host *host = (struct dwmci_host *)mmc->priv; > > - u32 fifo_size, fifoth_val; > > + u32 fifo_size, fifoth_val, ier; > > > > dwmci_writel(host, DWMCI_PWREN, 1); > > > > @@ -323,6 +321,14 @@ static int dwmci_init(struct mmc *mmc) > > return -1; > > } > > > > + /* Enumerate at 400KHz */ > > + dwmci_setup_bus(host, mmc->f_min); > > + > > + /* Set auto stop command */ > > + ier = dwmci_readl(host, DWMCI_CTRL); > > + ier |= DWMCI_CTRL_SEND_AS_CCSD; > > + dwmci_writel(host, DWMCI_CTRL, ier); > > + > > dwmci_writel(host, DWMCI_RINTSTS, 0xFFFFFFFF); > > dwmci_writel(host, DWMCI_INTMASK, 0); > > > > @@ -332,11 +338,13 @@ static int dwmci_init(struct mmc *mmc) > > dwmci_writel(host, DWMCI_BMOD, 1); > > > > fifo_size = dwmci_readl(host, DWMCI_FIFOTH); > > - if (host->fifoth_val) > > + fifo_size = ((fifo_size & RX_WMARK_MASK) >> RX_WMARK_SHIFT) + 1; > > + if (host->fifoth_val) { > > fifoth_val = host->fifoth_val; > > - else > > - fifoth_val = MSIZE(0x2) | RX_WMARK(fifo_size/2 -1) | > > - TX_WMARK(fifo_size/2); > > + } else { > > + fifoth_val = MSIZE(0x2) | RX_WMARK(fifo_size / 2 - 1) | > > + TX_WMARK(fifo_size / 2); > > + } > > dwmci_writel(host, DWMCI_FIFOTH, fifoth_val); > > > > dwmci_writel(host, DWMCI_CLKENA, 0); > > > > _______________________________________________ > U-Boot mailing list > U-Boot@lists.denx.de > http://lists.denx.de/mailman/listinfo/u-boot >
diff --git a/drivers/mmc/dw_mmc.c b/drivers/mmc/dw_mmc.c index 4070d4e..963a515 100644 --- a/drivers/mmc/dw_mmc.c +++ b/drivers/mmc/dw_mmc.c @@ -129,13 +129,13 @@ static int dwmci_send_cmd(struct mmc *mmc, struct mmc_cmd *cmd, unsigned int timeout = 100000; u32 retry = 10000; u32 mask, ctrl; + ulong start = get_timer(0); while (dwmci_readl(host, DWMCI_STATUS) & DWMCI_BUSY) { - if (timeout == 0) { + if (get_timer(start) > timeout) { printf("Timeout on data busy\n"); return TIMEOUT; } - timeout--; } dwmci_writel(host, DWMCI_RINTSTS, DWMCI_INTMSK_ALL); @@ -143,7 +143,6 @@ static int dwmci_send_cmd(struct mmc *mmc, struct mmc_cmd *cmd, if (data) dwmci_prepare_data(host, data); - dwmci_writel(host, DWMCI_CMDARG, cmd->cmdarg); if (data) @@ -231,9 +230,8 @@ static int dwmci_setup_bus(struct dwmci_host *host, u32 freq) int timeout = 10000; unsigned long sclk; - if (freq == host->clock) + if ((freq == host->clock) || (freq == 0)) return 0; - /* * If host->mmc_clk didn't define, * then assume that host->bus_hz is source clock value. @@ -314,7 +312,7 @@ static void dwmci_set_ios(struct mmc *mmc) static int dwmci_init(struct mmc *mmc) { struct dwmci_host *host = (struct dwmci_host *)mmc->priv; - u32 fifo_size, fifoth_val; + u32 fifo_size, fifoth_val, ier; dwmci_writel(host, DWMCI_PWREN, 1); @@ -323,6 +321,14 @@ static int dwmci_init(struct mmc *mmc) return -1; } + /* Enumerate at 400KHz */ + dwmci_setup_bus(host, mmc->f_min); + + /* Set auto stop command */ + ier = dwmci_readl(host, DWMCI_CTRL); + ier |= DWMCI_CTRL_SEND_AS_CCSD; + dwmci_writel(host, DWMCI_CTRL, ier); + dwmci_writel(host, DWMCI_RINTSTS, 0xFFFFFFFF); dwmci_writel(host, DWMCI_INTMASK, 0); @@ -332,11 +338,13 @@ static int dwmci_init(struct mmc *mmc) dwmci_writel(host, DWMCI_BMOD, 1); fifo_size = dwmci_readl(host, DWMCI_FIFOTH); - if (host->fifoth_val) + fifo_size = ((fifo_size & RX_WMARK_MASK) >> RX_WMARK_SHIFT) + 1; + if (host->fifoth_val) { fifoth_val = host->fifoth_val; - else - fifoth_val = MSIZE(0x2) | RX_WMARK(fifo_size/2 -1) | - TX_WMARK(fifo_size/2); + } else { + fifoth_val = MSIZE(0x2) | RX_WMARK(fifo_size / 2 - 1) | + TX_WMARK(fifo_size / 2); + } dwmci_writel(host, DWMCI_FIFOTH, fifoth_val); dwmci_writel(host, DWMCI_CLKENA, 0);
This patch enumerates dwmci and set auto stop command during dwmci initialisation. EMMC read/write is not happening in current implementation due to improper fifo size computation. Hence modified the fifo size computation to resolve EMMC read write issues. Signed-off-by: Amar <amarendra.xt@samsung.com> --- Changes since V1: 1)Created the macros RX_WMARK_SHIFT and RX_WMARK_MASK in header file. Changes since V2: 1)Updation of commit message and resubmition of proper patch set. Changes since V3: 1)Updated to use the macro DWMCI_CTRL_SEND_AS_CCSD instead of the hard coded value (1 << 10). Changes since V4: 1)Updated the function dwmci_send_cmd() to use get_timer() instead of using mdelay(1). Changes since V5: 1)Updated in response to review comments. Changes since V6: No change. Changes since V7: 1)Updated the function dwmci_setup_bus() to return 0 if (freq == 0). This is to avoid the run time exception "raise:Signal # 8 caught". drivers/mmc/dw_mmc.c | 28 ++++++++++++++++++---------- 1 file changed, 18 insertions(+), 10 deletions(-)