* [PATCH] mmc: sdio: optimization for process_sdio_pending_irq
@ 2011-12-09 12:29 Sangwook Lee
2011-12-09 14:17 ` Nicolas Pitre
0 siblings, 1 reply; 2+ messages in thread
From: Sangwook Lee @ 2011-12-09 12:29 UTC (permalink / raw)
To: cjb, linux-mmc
Cc: nicolas.pitre, per.forlin, linux-kernel, patches, Sangwook Lee
Skip fillng in mmc_command struct for CCCR_INTx
everytime sdio_irq_thread (process_sdio_pending_irqs)
calls mmc_io_rw_direct.
Signed-off-by: Sangwook Lee <sangwook.lee@linaro.org>
---
drivers/mmc/core/sdio_irq.c | 2 +-
drivers/mmc/core/sdio_ops.c | 40 ++++++++++++++++++++++++++++++++++++++++
drivers/mmc/core/sdio_ops.h | 1 +
3 files changed, 42 insertions(+), 1 deletions(-)
diff --git a/drivers/mmc/core/sdio_irq.c b/drivers/mmc/core/sdio_irq.c
index 68f81b9..67f782a 100644
--- a/drivers/mmc/core/sdio_irq.c
+++ b/drivers/mmc/core/sdio_irq.c
@@ -44,7 +44,7 @@ static int process_sdio_pending_irqs(struct mmc_card *card)
return 1;
}
- ret = mmc_io_rw_direct(card, 0, 0, SDIO_CCCR_INTx, 0, &pending);
+ ret = mmc_io_rw_direct_irq(card, &pending);
if (ret) {
pr_debug("%s: error %d reading SDIO_CCCR_INTx\n",
mmc_card_id(card), ret);
diff --git a/drivers/mmc/core/sdio_ops.c b/drivers/mmc/core/sdio_ops.c
index b0517cc..c80fc32 100644
--- a/drivers/mmc/core/sdio_ops.c
+++ b/drivers/mmc/core/sdio_ops.c
@@ -111,6 +111,46 @@ static int mmc_io_rw_direct_host(struct mmc_host *host, int write, unsigned fn,
return 0;
}
+static struct mmc_command sdio_intx_cmd = {
+ .opcode = SD_IO_RW_DIRECT,
+ .arg = SDIO_CCCR_INTx << 9,
+ .flags = MMC_RSP_SPI_R5 | MMC_RSP_R5 | MMC_CMD_AC,
+};
+
+int mmc_io_rw_direct_irq(struct mmc_card *card, u8 *out)
+{
+ int err;
+ struct mmc_host *host ;
+
+ BUG_ON(!card);
+ host = card->host;
+ BUG_ON(!host);
+
+ err = mmc_wait_for_cmd(host, &sdio_intx_cmd, 0);
+ if (err)
+ return err;
+
+ if (mmc_host_is_spi(host)) {
+ /* host driver already reported errors */
+ } else {
+ if (sdio_intx_cmd.resp[0] & R5_ERROR)
+ return -EIO;
+ if (sdio_intx_cmd.resp[0] & R5_FUNCTION_NUMBER)
+ return -EINVAL;
+ if (sdio_intx_cmd.resp[0] & R5_OUT_OF_RANGE)
+ return -ERANGE;
+ }
+
+ if (out) {
+ if (mmc_host_is_spi(host))
+ *out = (sdio_intx_cmd.resp[0] >> 8) & 0xFF;
+ else
+ *out = sdio_intx_cmd.resp[0] & 0xFF;
+ }
+
+ return 0;
+}
+
int mmc_io_rw_direct(struct mmc_card *card, int write, unsigned fn,
unsigned addr, u8 in, u8 *out)
{
diff --git a/drivers/mmc/core/sdio_ops.h b/drivers/mmc/core/sdio_ops.h
index 12a4d3a..5fe8ad6 100644
--- a/drivers/mmc/core/sdio_ops.h
+++ b/drivers/mmc/core/sdio_ops.h
@@ -15,6 +15,7 @@
int mmc_send_io_op_cond(struct mmc_host *host, u32 ocr, u32 *rocr);
int mmc_io_rw_direct(struct mmc_card *card, int write, unsigned fn,
unsigned addr, u8 in, u8* out);
+int mmc_io_rw_direct_irq(struct mmc_card *card, u8 *out);
int mmc_io_rw_extended(struct mmc_card *card, int write, unsigned fn,
unsigned addr, int incr_addr, u8 *buf, unsigned blocks, unsigned blksz);
int sdio_reset(struct mmc_host *host);
--
1.7.4.1
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [PATCH] mmc: sdio: optimization for process_sdio_pending_irq
2011-12-09 12:29 [PATCH] mmc: sdio: optimization for process_sdio_pending_irq Sangwook Lee
@ 2011-12-09 14:17 ` Nicolas Pitre
0 siblings, 0 replies; 2+ messages in thread
From: Nicolas Pitre @ 2011-12-09 14:17 UTC (permalink / raw)
To: Sangwook Lee; +Cc: cjb, linux-mmc, per.forlin, linux-kernel, patches
On Fri, 9 Dec 2011, Sangwook Lee wrote:
> Skip fillng in mmc_command struct for CCCR_INTx
> everytime sdio_irq_thread (process_sdio_pending_irqs)
> calls mmc_io_rw_direct.
>
> Signed-off-by: Sangwook Lee <sangwook.lee@linaro.org>
Is there a real advantage doing this?
While you might be saving on a few stores, you are increasing code size
and therefore i-cache miss probabilities. So this is not clear to me
this is actually a gain.
> ---
> drivers/mmc/core/sdio_irq.c | 2 +-
> drivers/mmc/core/sdio_ops.c | 40 ++++++++++++++++++++++++++++++++++++++++
> drivers/mmc/core/sdio_ops.h | 1 +
> 3 files changed, 42 insertions(+), 1 deletions(-)
>
> diff --git a/drivers/mmc/core/sdio_irq.c b/drivers/mmc/core/sdio_irq.c
> index 68f81b9..67f782a 100644
> --- a/drivers/mmc/core/sdio_irq.c
> +++ b/drivers/mmc/core/sdio_irq.c
> @@ -44,7 +44,7 @@ static int process_sdio_pending_irqs(struct mmc_card *card)
> return 1;
> }
>
> - ret = mmc_io_rw_direct(card, 0, 0, SDIO_CCCR_INTx, 0, &pending);
> + ret = mmc_io_rw_direct_irq(card, &pending);
> if (ret) {
> pr_debug("%s: error %d reading SDIO_CCCR_INTx\n",
> mmc_card_id(card), ret);
> diff --git a/drivers/mmc/core/sdio_ops.c b/drivers/mmc/core/sdio_ops.c
> index b0517cc..c80fc32 100644
> --- a/drivers/mmc/core/sdio_ops.c
> +++ b/drivers/mmc/core/sdio_ops.c
> @@ -111,6 +111,46 @@ static int mmc_io_rw_direct_host(struct mmc_host *host, int write, unsigned fn,
> return 0;
> }
>
> +static struct mmc_command sdio_intx_cmd = {
> + .opcode = SD_IO_RW_DIRECT,
> + .arg = SDIO_CCCR_INTx << 9,
> + .flags = MMC_RSP_SPI_R5 | MMC_RSP_R5 | MMC_CMD_AC,
> +};
> +
> +int mmc_io_rw_direct_irq(struct mmc_card *card, u8 *out)
> +{
> + int err;
> + struct mmc_host *host ;
> +
> + BUG_ON(!card);
> + host = card->host;
> + BUG_ON(!host);
> +
> + err = mmc_wait_for_cmd(host, &sdio_intx_cmd, 0);
> + if (err)
> + return err;
> +
> + if (mmc_host_is_spi(host)) {
> + /* host driver already reported errors */
> + } else {
> + if (sdio_intx_cmd.resp[0] & R5_ERROR)
> + return -EIO;
> + if (sdio_intx_cmd.resp[0] & R5_FUNCTION_NUMBER)
> + return -EINVAL;
> + if (sdio_intx_cmd.resp[0] & R5_OUT_OF_RANGE)
> + return -ERANGE;
> + }
> +
> + if (out) {
> + if (mmc_host_is_spi(host))
> + *out = (sdio_intx_cmd.resp[0] >> 8) & 0xFF;
> + else
> + *out = sdio_intx_cmd.resp[0] & 0xFF;
> + }
> +
> + return 0;
> +}
> +
> int mmc_io_rw_direct(struct mmc_card *card, int write, unsigned fn,
> unsigned addr, u8 in, u8 *out)
> {
> diff --git a/drivers/mmc/core/sdio_ops.h b/drivers/mmc/core/sdio_ops.h
> index 12a4d3a..5fe8ad6 100644
> --- a/drivers/mmc/core/sdio_ops.h
> +++ b/drivers/mmc/core/sdio_ops.h
> @@ -15,6 +15,7 @@
> int mmc_send_io_op_cond(struct mmc_host *host, u32 ocr, u32 *rocr);
> int mmc_io_rw_direct(struct mmc_card *card, int write, unsigned fn,
> unsigned addr, u8 in, u8* out);
> +int mmc_io_rw_direct_irq(struct mmc_card *card, u8 *out);
> int mmc_io_rw_extended(struct mmc_card *card, int write, unsigned fn,
> unsigned addr, int incr_addr, u8 *buf, unsigned blocks, unsigned blksz);
> int sdio_reset(struct mmc_host *host);
> --
> 1.7.4.1
>
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2011-12-09 14:17 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2011-12-09 12:29 [PATCH] mmc: sdio: optimization for process_sdio_pending_irq Sangwook Lee
2011-12-09 14:17 ` Nicolas Pitre
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®