Hi

On 2017/3/19 8:45, Bean Huo (beanhuo) wrote:
This patch fixes the issue that mmc_blk_issue_rq still
flushes cache when eMMC cache has already been off
through user space tool, such as mmc-utils.


I did a quick test and see the case you refer to, so

Tested-by: Shawn Lin <shawn....@rock-chips.com>
Reviewed-by: Shawn Lin <shawn....@rock-chips.com>

The reason is that card->ext_csd.cache_ctrl isn't reset.

Signed-off-by: beanhuo <bean...@micron.com>
---
 drivers/mmc/core/block.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c
index 1621fa0..fb3635ac 100644
--- a/drivers/mmc/core/block.c
+++ b/drivers/mmc/core/block.c
@@ -64,6 +64,7 @@ MODULE_ALIAS("mmc:block");
 #define MMC_BLK_TIMEOUT_MS  (10 * 60 * 1000)        /* 10 minute timeout */
 #define MMC_SANITIZE_REQ_TIMEOUT 240000
 #define MMC_EXTRACT_INDEX_FROM_ARG(x) ((x & 0x00FF0000) >> 16)
+#define MMC_EXTRACT_VALUE_FROM_ARG(x) ((x & 0x0000FF00) >> 8)

 #define mmc_req_rel_wr(req)    ((req->cmd_flags & REQ_FUA) && \
                                  (rq_data_dir(req) == WRITE))
@@ -535,6 +536,14 @@ static int __mmc_blk_ioctl_cmd(struct mmc_card *card, 
struct mmc_blk_data *md,
                return data.error;
        }

+       if ((MMC_EXTRACT_INDEX_FROM_ARG(cmd.arg) == EXT_CSD_CACHE_CTRL) &&
+           (cmd.opcode == MMC_SWITCH) && (card->ext_csd.cache_size > 0)) {
+               if (MMC_EXTRACT_VALUE_FROM_ARG(cmd.arg) & 1)
+                       card->ext_csd.cache_ctrl = 1;
+               else
+                       card->ext_csd.cache_ctrl = 0;
+       }
+
        /*
         * According to the SD specs, some commands require a delay after
         * issuing the command.


Reply via email to