linux-kernel - [PATCH RESEND] mmc: sdio: rw extended, use a sg entry for each block

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <1316598039-30238-1-git-send-email-nicolas.ferre@atmel.com>
Date:	Wed, 21 Sep 2011 11:40:39 +0200
From:	Nicolas Ferre <nicolas.ferre@...el.com>
To:	cjb@...top.org, linux-mmc@...r.kernel.org
Cc:	linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
	nicolas.pitre@...aro.org, per.forlin@...aro.org,
	Nicolas Ferre <nicolas.ferre@...el.com>
Subject: [PATCH RESEND] mmc: sdio: rw extended, use a sg entry for each block

While using io multiple blocks operations, change the way that sg is built:
use one sg entry for each block instead of aggregating the whole buffer
in a single sg entry.
Using a single sg entry for a multiple block command may lead to
misunderstanding between the sd/mmc and the DMA controllers. In fact, the
knowledge of the block length will allow both controllers to optimize burst
sizes on internal bus while dealing with those data.

Use a sg table to store start addresses of blocks within the data buffer.

Signed-off-by: Nicolas Ferre <nicolas.ferre@...el.com>
---
 drivers/mmc/core/sdio_ops.c |   38 +++++++++++++++++++++++++++++---------
 1 files changed, 29 insertions(+), 9 deletions(-)

diff --git a/drivers/mmc/core/sdio_ops.c b/drivers/mmc/core/sdio_ops.c
index f087d87..aea6978 100644
--- a/drivers/mmc/core/sdio_ops.c
+++ b/drivers/mmc/core/sdio_ops.c
@@ -124,7 +124,7 @@ int mmc_io_rw_extended(struct mmc_card *card, int write, unsigned fn,
 	struct mmc_request mrq = {0};
 	struct mmc_command cmd = {0};
 	struct mmc_data data = {0};
-	struct scatterlist sg;
+	struct sg_table sgt;
 
 	BUG_ON(!card);
 	BUG_ON(fn > 7);
@@ -144,24 +144,44 @@ int mmc_io_rw_extended(struct mmc_card *card, int write, unsigned fn,
 	cmd.arg |= fn << 28;
 	cmd.arg |= incr_addr ? 0x04000000 : 0x00000000;
 	cmd.arg |= addr << 9;
-	if (blocks == 1 && blksz <= 512)
-		cmd.arg |= (blksz == 512) ? 0 : blksz;	/* byte mode */
-	else
-		cmd.arg |= 0x08000000 | blocks;		/* block mode */
+	if (blocks == 1 && blksz <= 512) {
+		/* byte mode */
+		struct scatterlist sg;
+
+		cmd.arg |= (blksz == 512) ? 0 : blksz;
+		sg_init_one(&sg, buf, blksz * blocks);
+
+		data.sg = &sg;
+		data.sg_len = 1;
+	} else {
+		/* block mode */
+		struct scatterlist *sg_ptr;
+		int i;
+
+		cmd.arg |= 0x08000000 | blocks;
+		if (sg_alloc_table(&sgt, blocks, GFP_KERNEL))
+			return -ENOMEM;
+		for_each_sg(sgt.sgl, sg_ptr, sgt.nents, i) {
+			sg_set_buf(sg_ptr, buf + i * blksz, blksz);
+		}
+
+		data.sg = sgt.sgl;
+		data.sg_len = sgt.nents;
+	}
+
 	cmd.flags = MMC_RSP_SPI_R5 | MMC_RSP_R5 | MMC_CMD_ADTC;
 
 	data.blksz = blksz;
 	data.blocks = blocks;
 	data.flags = write ? MMC_DATA_WRITE : MMC_DATA_READ;
-	data.sg = &sg;
-	data.sg_len = 1;
-
-	sg_init_one(&sg, buf, blksz * blocks);
 
 	mmc_set_data_timeout(&data, card);
 
 	mmc_wait_for_req(card->host, &mrq);
 
+	if (blocks != 1 || blksz > 512)
+		sg_free_table(&sgt);
+
 	if (cmd.error)
 		return cmd.error;
 	if (data.error)
-- 
1.7.3

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/