[PATCH 0/6] SD/MMC driver for MX25/35/51

[Eric Bénard]

Hi Richard,

Le 18/10/2010 10:48, Zhu Richard-R65037 a écrit :
> I tested your serial patches with Eric's platform related codes on i.MX51 BBG3.0 board.
> Two phenomena:
> The debug msg "mmc0: mmc_rescan: trying to init card at 400000 Hz..." are dumped out repeatedly.
> ROOT: There are pr_info codes in the mmc_rescan func.
>
fixed by this patch :
http://git.kernel.org/?p=linux/kernel/git/cjb/mmc.git;a=commit;h=0ca6d53ec81a13263f6643372b2a5abf7818e8fa

> The other:
> Encounter the Timeout and CRC error when the sandisk 4G SDHC card are used as the ROOTFS storage media
> When I boot up the BBG board.
>
> It seems that the "SDHCI_QUIRK_BROKEN_TIMEOUT_VAL" is mandatory required at my side when I did the tests
>   on BBG3.0 board, and the i.MX51 board can support the "MULTI-BLOCK".
>
you seems to be right, the attached patch seems to fix it (tested on our board 
using a Linaro rootfs on a Transcend SDHC card and I didn't get errors after 
applying this patch but I'm stopped by udev errors now with this rootfs).

diff --git a/drivers/mmc/host/sdhci-esdhc-imx.c 
b/drivers/mmc/host/sdhci-esdhc-imx.c
index 2e9cca1..e8f7048 100644
--- a/drivers/mmc/host/sdhci-esdhc-imx.c
+++ b/drivers/mmc/host/sdhci-esdhc-imx.c
@@ -134,7 +134,8 @@ static struct sdhci_ops sdhci_esdhc_ops = {

  struct sdhci_pltfm_data sdhci_esdhc_imx_pdata = {
         .quirks = ESDHC_DEFAULT_QUIRKS | SDHCI_QUIRK_NO_MULTIBLOCK
-                       | SDHCI_QUIRK_BROKEN_ADMA,
+                       | SDHCI_QUIRK_BROKEN_ADMA
+                       | SDHCI_QUIRK_BROKEN_TIMEOUT_VAL,
         /* ADMA has issues. Might be fixable */
         /* NO_MULTIBLOCK might be MX35 only (Errata: ENGcm07207) */
         .ops = &sdhci_esdhc_ops,


> What's kinds of tests and environment you did at your side?
>
Environment : our i.MX515 based boards.
Tests (until now) : mount + a few read/write.
We have a nand flash on board so our rootfs is on it so I didn't test a rootfs 
until this morning.

Eric