These cards show abysmal write performance when an 8K unaligned write crosses an 8K barrier. This is the manufacturer recommendation.
Change-Id: I527fc44ff968f7ead61328bf655515fdf27b9f48 Signed-off-by: Andrei Warkentin <[email protected]> --- drivers/mmc/card/Kconfig | 7 +++++++ drivers/mmc/card/block-quirks.c | 30 ++++++++++++++++++++++++++++++ 2 files changed, 37 insertions(+), 0 deletions(-) diff --git a/drivers/mmc/card/Kconfig b/drivers/mmc/card/Kconfig index 063fa16..6d9f391 100644 --- a/drivers/mmc/card/Kconfig +++ b/drivers/mmc/card/Kconfig @@ -21,6 +21,13 @@ config MMC_BLOCK_QUIRKS help Say Y here to enable various workarounds for known cards. +config MMC_BLOCK_QUIRK_TOSHIBA_MMC32G + tristate "Toshiba MMC32G quirks" + depends on MMC_BLOCK_QUIRKS + default n + help + Say Y if you have a Toshiba MMC32G eMMC card. + config MMC_BLOCK_BOUNCE bool "Use bounce buffer for simple hosts" depends on MMC_BLOCK diff --git a/drivers/mmc/card/block-quirks.c b/drivers/mmc/card/block-quirks.c index ceae70c..8aea551 100644 --- a/drivers/mmc/card/block-quirks.c +++ b/drivers/mmc/card/block-quirks.c @@ -109,3 +109,33 @@ int mmc_blk_quirk_register(struct mmc_blk_quirk *quirk, bool is_mmc) rb_insert_color(&quirk->rb_node, &quirk_tree[!is_mmc]); return 0; } + +#ifdef CONFIG_MMC_BLOCK_QUIRK_TOSHIBA_MMC32G +static int toshiba_mmc32g(struct mmc_blk_data *md, struct mmc_card *card) +{ + printk(KERN_INFO "Applying Toshiba MMC32G workarounds\n"); + + /* Page size 8K, this card doesn't like unaligned writes + across 8K boundary. */ + md->write_align_size = 8192; + return 0; +} + +static struct mmc_blk_quirk toshiba_mmc32g_q = { + .name = "MMC32G", + .manfid = 0x11, + .oemid = 0x0100, + + /* Any date, any revision, */ + .rev_start = 0, + .rev_end = (u64) -1, + .probe = toshiba_mmc32g +}; + +int __init mmc_toshiba_quirks(void) +{ + return mmc_blk_quirk_register(&toshiba_mmc32g_q, true); +} + +device_initcall(mmc_toshiba_quirks); +#endif /* CONFIG_MMC_BLOCK_QUIRK_TOSHIBA_MMC32G */ -- 1.7.0.4 -- To unsubscribe from this list: send the line "unsubscribe linux-mmc" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
