2009-03-16 11:27:39

by Wolfgang Mües

[permalink] [raw]
Subject: [PATCH 3/7 revised] mmc_spi: adjust for delayed data token response

From: Wolfgang Muees <[email protected]>

o some cards are not able to send the data token in time, but
miss the time frame for some bits(!). So synchronize to the
start of the token.

Signed-off-by: Wolfgang Muees <[email protected]>

---

This is the revised patch #3, to apply after the already accepted
patches #1,2 and #5. It contains the bugfix and a better explanation
of what is going on here.

diff -uprN 2_6_29_rc7_patch125/drivers/mmc/host/mmc_spi.c 2_6_29_rc7_patch1253/drivers/mmc/host/mmc_spi.c
--- 2_6_29_rc7_patch125/drivers/mmc/host/mmc_spi.c 2009-03-11 13:00:06.000000000 +0100
+++ 2_6_29_rc7_patch1253/drivers/mmc/host/mmc_spi.c 2009-03-16 12:15:36.000000000 +0100
@@ -612,6 +612,7 @@ mmc_spi_writeblock(struct mmc_spi_host *
struct spi_device *spi = host->spi;
int status, i;
struct scratch *scratch = host->data;
+ u32 pattern;

if (host->mmc->use_spi_crc)
scratch->crc_val = cpu_to_be16(
@@ -639,8 +640,27 @@ mmc_spi_writeblock(struct mmc_spi_host *
* doesn't necessarily tell whether the write operation succeeded;
* it just says if the transmission was ok and whether *earlier*
* writes succeeded; see the standard.
- */
- switch (SPI_MMC_RESPONSE_CODE(scratch->status[0])) {
+ *
+ * In practice, there are (even modern SDHC-)cards which are late
+ * in sending the response, and miss the time frame by a few bits,
+ * so we have to cope with this situation and check the response
+ * bit-by-bit. Arggh!!!
+ */
+ pattern = scratch->status[0] << 24;
+ pattern |= scratch->status[1] << 16;
+ pattern |= scratch->status[2] << 8;
+ pattern |= scratch->status[3];
+
+ /* First 3 bit of pattern are undefined */
+ pattern |= 0xE0000000;
+
+ /* left-adjust to leading 0 bit */
+ while (pattern & 0x80000000)
+ pattern <<= 1;
+ /* right-adjust for pattern matching. Code is in bit 4..0 now. */
+ pattern >>= 27;
+
+ switch (pattern) {
case SPI_RESPONSE_ACCEPTED:
status = 0;
break;
@@ -671,8 +691,9 @@ mmc_spi_writeblock(struct mmc_spi_host *
/* Return when not busy. If we didn't collect that status yet,
* we'll need some more I/O.
*/
- for (i = 1; i < sizeof(scratch->status); i++) {
- if (scratch->status[i] != 0)
+ for (i = 4; i < sizeof(scratch->status); i++) {
+ /* card is non-busy if the most recent bit is 1 */
+ if (scratch->status[i] & 0x01)
return 0;
}
return mmc_spi_wait_unbusy(host, timeout);

---
regards

i. A. Wolfgang M?es
--
Auerswald GmbH & Co. KG
Hardware Development
Telefon: +49 (0)5306 9219 0
Telefax: +49 (0)5306 9219 94
E-Mail: [email protected]
Web: http://www.auerswald.de
?
--------------------------------------------------------------
Auerswald GmbH & Co. KG, Vor den Grash?fen 1, 38162 Cremlingen
Registriert beim AG Braunschweig HRA 13289
p.h.G Auerswald Gesch?ftsf?hrungsges. mbH
Registriert beim AG Braunschweig HRB 7463
Gesch?ftsf?hrer: Dipl-Ing. Gerhard Auerswald


2009-03-16 19:38:17

by David Brownell

[permalink] [raw]
Subject: Re: [PATCH 3/7 revised] mmc_spi: adjust for delayed data token response

On Monday 16 March 2009, Wolfgang M?es wrote:
> From: Wolfgang Muees <[email protected]>
>
> o some cards are not able to send the data token in time, but
> miss the time frame for some bits(!). So synchronize to the
> start of the token.
>
> Signed-off-by: Wolfgang Muees <[email protected]>

Acked-by: David Brownell <[email protected]>


> ---
>
> This is the revised patch #3, to apply after the already accepted
> patches #1,2 and #5. It contains the bugfix and a better explanation
> of what is going on here.
>
> diff -uprN 2_6_29_rc7_patch125/drivers/mmc/host/mmc_spi.c 2_6_29_rc7_patch1253/drivers/mmc/host/mmc_spi.c
> --- 2_6_29_rc7_patch125/drivers/mmc/host/mmc_spi.c 2009-03-11 13:00:06.000000000 +0100
> +++ 2_6_29_rc7_patch1253/drivers/mmc/host/mmc_spi.c 2009-03-16 12:15:36.000000000 +0100
> @@ -612,6 +612,7 @@ mmc_spi_writeblock(struct mmc_spi_host *
> struct spi_device *spi = host->spi;
> int status, i;
> struct scratch *scratch = host->data;
> + u32 pattern;
>
> if (host->mmc->use_spi_crc)
> scratch->crc_val = cpu_to_be16(
> @@ -639,8 +640,27 @@ mmc_spi_writeblock(struct mmc_spi_host *
> * doesn't necessarily tell whether the write operation succeeded;
> * it just says if the transmission was ok and whether *earlier*
> * writes succeeded; see the standard.
> - */
> - switch (SPI_MMC_RESPONSE_CODE(scratch->status[0])) {
> + *
> + * In practice, there are (even modern SDHC-)cards which are late
> + * in sending the response, and miss the time frame by a few bits,
> + * so we have to cope with this situation and check the response
> + * bit-by-bit. Arggh!!!
> + */
> + pattern = scratch->status[0] << 24;
> + pattern |= scratch->status[1] << 16;
> + pattern |= scratch->status[2] << 8;
> + pattern |= scratch->status[3];
> +
> + /* First 3 bit of pattern are undefined */
> + pattern |= 0xE0000000;
> +
> + /* left-adjust to leading 0 bit */
> + while (pattern & 0x80000000)
> + pattern <<= 1;
> + /* right-adjust for pattern matching. Code is in bit 4..0 now. */
> + pattern >>= 27;
> +
> + switch (pattern) {
> case SPI_RESPONSE_ACCEPTED:
> status = 0;
> break;
> @@ -671,8 +691,9 @@ mmc_spi_writeblock(struct mmc_spi_host *
> /* Return when not busy. If we didn't collect that status yet,
> * we'll need some more I/O.
> */
> - for (i = 1; i < sizeof(scratch->status); i++) {
> - if (scratch->status[i] != 0)
> + for (i = 4; i < sizeof(scratch->status); i++) {
> + /* card is non-busy if the most recent bit is 1 */
> + if (scratch->status[i] & 0x01)
> return 0;
> }
> return mmc_spi_wait_unbusy(host, timeout);
>
> ---
> regards
>
> i. A. Wolfgang M?es
> --
> Auerswald GmbH & Co. KG
> Hardware Development
> Telefon: +49 (0)5306 9219 0
> Telefax: +49 (0)5306 9219 94
> E-Mail: [email protected]
> Web: http://www.auerswald.de
> ?
> --------------------------------------------------------------
> Auerswald GmbH & Co. KG, Vor den Grash?fen 1, 38162 Cremlingen
> Registriert beim AG Braunschweig HRA 13289
> p.h.G Auerswald Gesch?ftsf?hrungsges. mbH
> Registriert beim AG Braunschweig HRB 7463
> Gesch?ftsf?hrer: Dipl-Ing. Gerhard Auerswald
>
>


2009-03-22 14:47:39

by Pierre Ossman

[permalink] [raw]
Subject: Re: [PATCH 3/7 revised] mmc_spi: adjust for delayed data token response

On Mon, 16 Mar 2009 12:23:03 +0100
Wolfgang Mües <[email protected]> wrote:

> From: Wolfgang Muees <[email protected]>
>
> o some cards are not able to send the data token in time, but
> miss the time frame for some bits(!). So synchronize to the
> start of the token.
>
> Signed-off-by: Wolfgang Muees <[email protected]>
>

Queued.

--
-- Pierre Ossman

WARNING: This correspondence is being monitored by the
Swedish government. Make sure your server uses encryption
for SMTP traffic and consider using PGP for end-to-end
encryption.


Attachments:
signature.asc (198.00 B)