Skip to contentSkip to conversation

mmc: clarifications on host.post_req()

PATCH New Per Forlin 4 participants 7 messages Aug 29, 2011 – Sep 21, 2011

Subsystems (2)
MMC, SD and SDIO Arm primecell mmci PL180/1 driver
lore
Share & tools
lore

Per Forlin

author
#0 lore raw

cover mmc: clarifications on host.post_req()

Fixes for 3.1.
This patchset doesn't fix any bugs in 3.1 but it improves the documentation in order to prevent new bugs.

Per Forlin (2):
mmc: core: clarify how to use post_req in case of errors
mmc: mmci: simplify err check in mmci_post_request

 drivers/mmc/core/core.c  |    6 ++++++
 drivers/mmc/host/mmci.c  |    2 +-
 include/linux/mmc/host.h |    3 +++
 3 files changed, 10 insertions(+), 1 deletions(-)
Signature · 1 line
1.7.4.1
author
#1 lore raw

PATCH 1/2 mmc: core: clarify how to use post_req in case of errors

The err condition in post_req() is set to undo a call made to pre_req() that hasn't been started yet.
The err condition is not set if an MMC request returns error.

Signed-off-by Per Forlin <[email protected]>
---
 drivers/mmc/core/core.c  |    6 ++++++
 include/linux/mmc/host.h |    3 +++
 2 files changed, 9 insertions(+), 0 deletions(-)
drivers/mmc/core/core.c +6 -0
index 91a0a74..542aa06 100644
@@ -279,8 +279,14 @@ struct mmc_async_req *mmc_start_req(struct mmc_host *host,
 		mmc_wait_for_req_done(host, host->areq->mrq);
 		err = host->areq->err_check(host->card, host->areq);
 		if (err) {
+			/* post process the completed failed request */
 			mmc_post_req(host, host->areq->mrq, 0);
 			if (areq)
+				/*
+				 * Cancel the new prepared request, because
+				 * it can't run until the failed
+				 * request has been properly handled.
+				 */
 				mmc_post_req(host, areq->mrq, -EINVAL);
 
 			host->areq = NULL;
include/linux/mmc/host.h +3 -0
index 1d09562..7896fb4 100644
@@ -108,6 +108,9 @@ struct mmc_host_ops {
 	 * It is optional for the host to implement pre_req and post_req in
 	 * order to support double buffering of requests (prepare one
 	 * request while another request is active).
+	 * pre_req() must always be followed by a post_req().
+	 * To undo a call made to pre_req(), call post_req() with
+	 * a nonzero err condition.
 	 */
 	void	(*post_req)(struct mmc_host *host, struct mmc_request *req,
 			    int err);
Signature · 1 line
1.7.4.1
#2 lore raw

PATCH 1/2 mmc: core: clarify how to use post_req in case of errors

Quoted 5 lines · Per Forlin

The err condition in post_req() is set to undo a call made to pre_req() that hasn't been started yet.
The err condition is not set if an MMC request returns error.

Signed-off-by Per Forlin <[email protected]>

Comments look correct so:

Acked-by Linus Walleij <[email protected]>

Thanks,
Linus Walleij

author
#3 lore raw

PATCH 2/2 mmc: mmci: simplify err check in mmci_post_request

In reply to Per Forlin

The error condition indicates that mmci_post_request() should cleanup after the mmci_pre_request(). In this case the resources allocated by device_prep_slave_sg() are freed by calling dmaengine_terminate_all(). dma_unmap_sg() should always be performed if the host_cookie is set.

Signed-off-by Per Forlin <[email protected]>
---
 drivers/mmc/host/mmci.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
drivers/mmc/host/mmci.c +1 -1
index 56e9a41..40e4c05 100644
@@ -529,7 +529,7 @@ static void mmci_post_request(struct mmc_host *mmc, struct mmc_request *mrq,
 	if (chan) {
 		if (err)
 			dmaengine_terminate_all(chan);
-		if (err || data->host_cookie)
+		if (data->host_cookie)
 			dma_unmap_sg(mmc_dev(host->mmc), data->sg,
 				     data->sg_len, dir);
 		mrq->data->host_cookie = 0;
Signature · 1 line
1.7.4.1
#4 lore raw

PATCH 2/2 mmc: mmci: simplify err check in mmci_post_request

Quoted 6 lines · Per Forlin

The error condition indicates that mmci_post_request() should cleanup after the mmci_pre_request(). In this case the resources allocated by device_prep_slave_sg() are freed by calling dmaengine_terminate_all(). dma_unmap_sg() should always be performed if the host_cookie is set.

Signed-off-by Per Forlin <[email protected]>
Acked-by Linus Walleij <[email protected]>

Thanks,
Linus Walleij

#5 lore raw

cover mmc: clarifications on host.post_req()

In reply to Per Forlin

Hi Per,

Quoted 3 lines · Per Forlin

Fixes for 3.1.
This patchset doesn't fix any bugs in 3.1 but it improves the documentation in order to prevent new bugs.

I've merged this into mmc-next now, but I'm planning on sending it for 3.2 instead of 3.1 at this point; hope that's okay.

Thanks,

- Chris.

Signature · 2 lines
Chris Ball   <[email protected]>   <http://printf.net/>
One Laptop Per Child
#6 lore raw

cover mmc: clarifications on host.post_req()

Quoted 10 lines · Chris Ball

Hi Per,

Quoted 3 lines · Per Forlin

Fixes for 3.1.
This patchset doesn't fix any bugs in 3.1 but it improves the documentation in order to prevent new bugs.

I've merged this into mmc-next now, but I'm planning on sending it for 3.2 instead of 3.1 at this point; hope that's okay.

Fine with me.

Thanks,
Per