On 2018/3/29 12:21, Harish Jenny K N wrote:
On Friday 23 March 2018 07:08 PM, Shawn Lin wrote:
On 2018/3/23 16:39, Harish Jenny K N wrote:
From: Diwakar Sharma <diwakar.sharma@xxxxxxxxxxxx>
This patch adds a quirk to report "SDR50" mode on IMX side
even if support of SDR104 was received from Marvell 8887
WiFi + Bluetooth chip side, as Marvell does not recommend
to use SDIO at the speed of higher than 150MHz >
Signed-off-by: Diwakar Sharma <diwakar.sharma@xxxxxxxxxxxx>
Signed-off-by: Harish Jenny K N <harish_kandiga@xxxxxxxxxx>
---
drivers/mmc/core/card.h | 5 +++++
drivers/mmc/core/quirks.h | 3 +++
drivers/mmc/core/sdio.c | 23 +++++++++++++++++++++++
include/linux/mmc/card.h | 2 ++
include/linux/mmc/sdio_ids.h | 1 +
5 files changed, 34 insertions(+)
diff --git a/drivers/mmc/core/card.h b/drivers/mmc/core/card.h
index 9c821ee..e6c7dad 100644
--- a/drivers/mmc/core/card.h
+++ b/drivers/mmc/core/card.h
@@ -221,4 +221,9 @@ static inline int mmc_card_broken_hpi(const struct mmc_card *c)
return c->quirks & MMC_QUIRK_BROKEN_HPI;
}
+static inline int mmc_card_broken_uhs(const struct mmc_card *c)
+{
+ return c->quirks & MMC_QUIRK_BROKEN_UHS;
+}
+
#endif
diff --git a/drivers/mmc/core/quirks.h b/drivers/mmc/core/quirks.h
index 5153577..575b6f7 100644
--- a/drivers/mmc/core/quirks.h
+++ b/drivers/mmc/core/quirks.h
@@ -132,6 +132,9 @@
SDIO_FIXUP(SDIO_VENDOR_ID_MARVELL, SDIO_DEVICE_ID_MARVELL_8797_F0,
add_quirk, MMC_QUIRK_BROKEN_IRQ_POLLING),
+ SDIO_FIXUP(SDIO_VENDOR_ID_MARVELL, SDIO_DEVICE_ID_MARVELL_8887WLAN,
+ add_quirk, MMC_QUIRK_BROKEN_UHS),
+
END_FIXUP
};
diff --git a/drivers/mmc/core/sdio.c b/drivers/mmc/core/sdio.c
index c599a62..9ed8b31 100644
--- a/drivers/mmc/core/sdio.c
+++ b/drivers/mmc/core/sdio.c
@@ -566,6 +566,9 @@ static int mmc_sdio_init_card(struct mmc_host *host, u32 ocr,
int retries = 10;
u32 rocr = 0;
u32 ocr_card = ocr;
+ int cccr_vsn;
+ unsigned char data;
+ int uhs;
WARN_ON(!host->claimed);
@@ -740,6 +743,26 @@ static int mmc_sdio_init_card(struct mmc_host *host, u32 ocr,
card->ocr = ocr_card;
mmc_fixup_device(card, sdio_fixup_methods);
+ uhs = ocr & R4_18V_PRESENT;
+ err = mmc_io_rw_direct(card, 0, 0, SDIO_CCCR_CCCR, 0, &data);
+ if (err)
+ goto remove;
+
+ cccr_vsn = data & 0x0f;
+ if (cccr_vsn >= SDIO_CCCR_REV_3_00 && uhs) {
+ if (mmc_host_uhs(card->host) && mmc_card_broken_uhs(card)) {
+ /*
+ * The MMC device reports a non-working
+ * maximum clock rate, workaround is to
+ * force the MMC host controller to report
+ * a lower clock rate than supported.
+ */
+ card->sw_caps.sd3_bus_mode &= ~SD_MODE_UHS_DDR50;
+ card->sw_caps.sd3_bus_mode &= ~SD_MODE_UHS_SDR104;
+ card->sw_caps.sd3_bus_mode |= SD_MODE_UHS_SDR50;
so SDR50 isn't UHS mode? Could you just remove the sd-uhs-sdr104,sd-
uhs-ddr50 from your host's cap if using this wifi? Or set max-frequency
to 150M with sd-uhs-sdr104 mode?
Changing host caps does not seem to be correct way ( the host->caps represent the capabilities of the actual host controller used by the card. This shouldn't be changing around IMHO. Please correct me if I am wrong.)
Might not be true, for instance, you have MMC_CAP_UHS_SDR104 support
for the controller, but you don't have a 3.3/1.8 adaptive vqmmc, so
you can't add MMC_CAP_UHS_SDR104. That being said, we shouldn't only
consider the host.
I will send v2 patch by reducing the max-frequency to 150MHz in SDR104 Mode.
Thanks,
Harish Jenny K N
--
To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
