Hi,
On 30-03-15 11:14, Ulf Hansson wrote:
On 29 March 2015 at 20:09, Hans de Goede <hdegoede@xxxxxxxxxx> wrote:
The eMMC on a tablet I've will stop working / communicating as soon as
the kernel executes:
mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
EXT_CSD_HPI_MGMT, 1,
card->ext_csd.generic_cmd6_time);
There seems to be no way to reliable identify eMMC-s which have a broken
hpi implementation, but at least for eMMC's which are soldered onto a board
we can work around this by specifying that hpi is broken in devicetree.
Seems like a reasonable approach!
Signed-off-by: Hans de Goede <hdegoede@xxxxxxxxxx>
---
Documentation/devicetree/bindings/mmc/mmc-card.txt | 31 ++++++++++++++++++++++
drivers/mmc/core/mmc.c | 10 ++++++-
2 files changed, 40 insertions(+), 1 deletion(-)
create mode 100644 Documentation/devicetree/bindings/mmc/mmc-card.txt
diff --git a/Documentation/devicetree/bindings/mmc/mmc-card.txt b/Documentation/devicetree/bindings/mmc/mmc-card.txt
new file mode 100644
index 0000000..0cc67fb
--- /dev/null
+++ b/Documentation/devicetree/bindings/mmc/mmc-card.txt
@@ -0,0 +1,31 @@
+mmc-card / eMMC bindings
+------------------------
+
+This documents describes the devicetree bindings for a mmc-host controller
+child node describing a mmc-card / an eMMC, see "Use of Function subnodes"
+in mmc.txt
+
+Required properties:
+-compatible : Must be "mmc-card"
+-reg : Must be <0>
Instead of using a compatible, why can we just use "reg = <0>"?
Yes we could do that, but using a compatible is sort of the standard way
to ensure that the contents of the node match the same binding rules as
the code parsing the node. e.g. i2c child nodes also have a compatible string,
as does the oob irq info for broadcom sdio wifi cards, which is the one
existing user of mmc host child nodes I'm aware of.
Regards,
Hans
+
+Optional properties:
+-broken-hpi : Use this to indicate that the mmc-card has a broken hpi
+ implementation, and that hpi should not be used
+
+Example:
+
+&mmc2 {
+ pinctrl-names = "default";
+ pinctrl-0 = <&mmc2_pins_a>;
+ vmmc-supply = <®_vcc3v3>;
+ bus-width = <8>;
+ non-removable;
+ status = "okay";
+
+ mmmcard: mmccard@0 {
+ reg = <0>;
+ compatible = "mmc-card";
+ broken-hpi;
+ };
+};
diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c
index 1d41e85..8d18920 100644
--- a/drivers/mmc/core/mmc.c
+++ b/drivers/mmc/core/mmc.c
@@ -11,6 +11,7 @@
*/
#include <linux/err.h>
+#include <linux/of.h>
#include <linux/slab.h>
#include <linux/stat.h>
#include <linux/pm_runtime.h>
@@ -336,6 +337,8 @@ static int mmc_decode_ext_csd(struct mmc_card *card, u8 *ext_csd)
{
int err = 0, idx;
unsigned int part_size;
+ struct device_node *np = card->dev.of_node;
+ bool broken_hpi = false;
/* Version is coded in the CSD_STRUCTURE byte in the EXT_CSD register */
card->ext_csd.raw_ext_csd_structure = ext_csd[EXT_CSD_STRUCTURE];
@@ -349,6 +352,9 @@ static int mmc_decode_ext_csd(struct mmc_card *card, u8 *ext_csd)
}
}
+ if (np && of_device_is_compatible(np, "mmc-card"))
+ broken_hpi = of_property_read_bool(np, "broken-hpi");
+
/*
* The EXT_CSD format is meant to be forward compatible. As long
* as CSD_STRUCTURE does not change, all values for EXT_CSD_REV
@@ -494,7 +500,7 @@ static int mmc_decode_ext_csd(struct mmc_card *card, u8 *ext_csd)
}
/* check whether the eMMC card supports HPI */
- if (ext_csd[EXT_CSD_HPI_FEATURES] & 0x1) {
+ if (!broken_hpi && (ext_csd[EXT_CSD_HPI_FEATURES] & 0x1)) {
card->ext_csd.hpi = 1;
if (ext_csd[EXT_CSD_HPI_FEATURES] & 0x2)
card->ext_csd.hpi_cmd = MMC_STOP_TRANSMISSION;
@@ -1251,6 +1257,8 @@ static int mmc_init_card(struct mmc_host *host, u32 ocr,
card->ocr = ocr;
card->type = MMC_TYPE_MMC;
card->rca = 1;
+ /* Also gets done by mmc_add_card, but we need early access */
+ card->dev.of_node = mmc_of_find_child_device(host, 0);
memcpy(card->raw_cid, cid, sizeof(card->raw_cid));
}
--
2.3.4
Kind regards
Uffe
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
