[U-Boot] [PATCH] cmd_i2c: Provide option for bulk 'i2c write' in one transaction
I2C chips do exist that require a write of some multi-byte data to occur in a single bus transaction (aka atomic transfer), otherwise either the write does not come into effect at all, or normal operation of internal circuitry cannot be guaranteed. The current implementation of the 'i2c write' command (transfer of multiple bytes from a memory buffer) in fact performs a separate transaction for each byte to be written and thus cannot support such types of I2C slave devices.
This patch provides an alternative by allowing 'i2c write' to execute the write transfer of the given number of bytes in a single bus transaction if CONFIG_SYS_I2C_BULK_WRITE is defined in the board header (otherwise the old method shall compile).
Signed-off-by: Lubomir Popov l-popov@ti.com --- common/cmd_i2c.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)
diff --git a/common/cmd_i2c.c b/common/cmd_i2c.c index 3a75f94..7116458 100644 --- a/common/cmd_i2c.c +++ b/common/cmd_i2c.c @@ -280,10 +280,22 @@ static int do_i2c_write(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[ return cmd_usage(cmdtp);
/* - * Length is the number of objects, not number of bytes. + * Length is the number of bytes. */ length = simple_strtoul(argv[4], NULL, 16);
+#if defined(CONFIG_SYS_I2C_BULK_WRITE) + /* + * Write all bytes in a single I2C transaction. If the target + * device is an EEPROM, it is your responsibility to not cross + * a page bounady. + */ + if (i2c_write(chip, devaddr, alen, memaddr, length) != 0) { + puts("Error writing to the chip.\n"); + return 1; + } +#else + /* Perform <length> separate write transactions of one byte each */ while (length-- > 0) { if (i2c_write(chip, devaddr++, alen, memaddr++, 1) != 0) { puts("Error writing to the chip.\n"); @@ -296,6 +308,7 @@ static int do_i2c_write(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[ udelay(11000); #endif } +#endif return 0; }
Hi,
On 24 November 2014 at 09:00, Lubomir Popov lpopov@mm-sol.com wrote:
I2C chips do exist that require a write of some multi-byte data to occur in a single bus transaction (aka atomic transfer), otherwise either the write does not come into effect at all, or normal operation of internal circuitry cannot be guaranteed. The current implementation of the 'i2c write' command (transfer of multiple bytes from a memory buffer) in fact performs a separate transaction for each byte to be written and thus cannot support such types of I2C slave devices.
This patch provides an alternative by allowing 'i2c write' to execute the write transfer of the given number of bytes in a single bus transaction if CONFIG_SYS_I2C_BULK_WRITE is defined in the board header (otherwise the old method shall compile).
Signed-off-by: Lubomir Popov l-popov@ti.com
common/cmd_i2c.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)
diff --git a/common/cmd_i2c.c b/common/cmd_i2c.c index 3a75f94..7116458 100644 --- a/common/cmd_i2c.c +++ b/common/cmd_i2c.c @@ -280,10 +280,22 @@ static int do_i2c_write(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[ return cmd_usage(cmdtp);
/*
* Length is the number of objects, not number of bytes.
* Length is the number of bytes. */ length = simple_strtoul(argv[4], NULL, 16);+#if defined(CONFIG_SYS_I2C_BULK_WRITE)
/** Write all bytes in a single I2C transaction. If the target* device is an EEPROM, it is your responsibility to not cross* a page bounady.*/if (i2c_write(chip, devaddr, alen, memaddr, length) != 0) {puts("Error writing to the chip.\n");return 1;}+#else
/* Perform <length> separate write transactions of one byte each */ while (length-- > 0) { if (i2c_write(chip, devaddr++, alen, memaddr++, 1) != 0) { puts("Error writing to the chip.\n");@@ -296,6 +308,7 @@ static int do_i2c_write(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[ udelay(11000); #endif
This should really be an option/flag on the command. We might want to do different things depending on the situation.
In fact with driver model's I2C implementation, currently in review, this logic of splitting things up can be handled in the uclass layer:
http://patchwork.ozlabs.org/patch/414066/
Regards, Simon
Hi Simon,
Hi,
On 24 November 2014 at 09:00, Lubomir Popov lpopov@mm-sol.com wrote:
I2C chips do exist that require a write of some multi-byte data to occur in a single bus transaction (aka atomic transfer), otherwise either the write does not come into effect at all, or normal operation of internal circuitry cannot be guaranteed. The current implementation of the 'i2c write' command (transfer of multiple bytes from a memory buffer) in fact performs a separate transaction for each byte to be written and thus cannot support such types of I2C slave devices.
This patch provides an alternative by allowing 'i2c write' to execute the write transfer of the given number of bytes in a single bus transaction if CONFIG_SYS_I2C_BULK_WRITE is defined in the board header (otherwise the old method shall compile).
Signed-off-by: Lubomir Popov l-popov@ti.com
common/cmd_i2c.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)
diff --git a/common/cmd_i2c.c b/common/cmd_i2c.c index 3a75f94..7116458 100644 --- a/common/cmd_i2c.c +++ b/common/cmd_i2c.c @@ -280,10 +280,22 @@ static int do_i2c_write(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[ return cmd_usage(cmdtp);
/*
* Length is the number of objects, not number of bytes.
* Length is the number of bytes. */ length = simple_strtoul(argv[4], NULL, 16);+#if defined(CONFIG_SYS_I2C_BULK_WRITE)
/** Write all bytes in a single I2C transaction. If the target* device is an EEPROM, it is your responsibility to not cross* a page bounady.*/if (i2c_write(chip, devaddr, alen, memaddr, length) != 0) {puts("Error writing to the chip.\n");return 1;}+#else
/* Perform <length> separate write transactions of one byte each */ while (length-- > 0) { if (i2c_write(chip, devaddr++, alen, memaddr++, 1) != 0) { puts("Error writing to the chip.\n");@@ -296,6 +308,7 @@ static int do_i2c_write(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[ udelay(11000); #endif
This should really be an option/flag on the command. We might want to do different things depending on the situation.
Agree. I thought about this initially, even about adding a separate command, but then decided that the human device using 'i2c write' with or without the config option would be intelligent enough to know what is being needed and done. From a chip compatibility perspective this is not an issue, since for byte-wise transfers the 'i2c mw' and 'i2c mm' commands still do exist. So I went the easy way :-).
In fact with driver model's I2C implementation, currently in review, this logic of splitting things up can be handled in the uclass layer:
Right, in the DM context a command line option would be the better solution. I shall see to revising the patch in this respect.
Regards, Simon
Best regards, Lubo
Hi Simon,
[snip]
while (length-- > 0) {if (i2c_write(chip, devaddr++, alen, memaddr++, 1) != 0)return i2c_report_err(-1, I2C_ERR_WRITE);/*
- No write delay with FRAM devices.
*/
Should this be indented?
#if !defined(CONFIG_SYS_I2C_FRAM)
udelay(11000);
udelay(11000);#endif
}
[snip]
Which indent do you mean? The udelay() is within the while loop, so it should. The comment above your note - I kept it as is, aligned with the #if/#endif.
Regards, Lubo
+Heiko
Hi,
On 26 November 2014 at 12:37, Lubomir Popov lpopov@mm-sol.com wrote:
Hi Simon,
[snip]
while (length-- > 0) {if (i2c_write(chip, devaddr++, alen, memaddr++, 1) != 0)return i2c_report_err(-1, I2C_ERR_WRITE);/*
- No write delay with FRAM devices.
*/
Should this be indented?
#if !defined(CONFIG_SYS_I2C_FRAM)
udelay(11000);
udelay(11000);#endif
}[snip]
Which indent do you mean? The udelay() is within the while loop, so it should. The comment above your note - I kept it as is, aligned with the #if/#endif.
It was hard to tell in my emailed, but yes i see, you have kept it where it was, which is fine.
I added Heiko (I2C maintainer) to cc. As mentioned, but in more detail, with driver model 'i2c flags 2' will select this behaviour for any i2c access, not just for this command. See u-boot-dm/i2c-working2 file i2c-uclass.c.
Regards, Simon
Hello Lubomir,
Am 24.11.2014 17:00, schrieb Lubomir Popov:
I2C chips do exist that require a write of some multi-byte data to occur in a single bus transaction (aka atomic transfer), otherwise either the write does not come into effect at all, or normal operation of internal circuitry cannot be guaranteed. The current implementation of the 'i2c write' command (transfer of multiple bytes from a memory buffer) in fact performs a separate transaction for each byte to be written and thus cannot support such types of I2C slave devices.
This patch provides an alternative by allowing 'i2c write' to execute the write transfer of the given number of bytes in a single bus transaction if CONFIG_SYS_I2C_BULK_WRITE is defined in the board header (otherwise the old method shall compile).
Signed-off-by: Lubomir Popov l-popov@ti.com
common/cmd_i2c.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)
Could you rebase your patch against current mainline, please? As we have now DM in i2c subsystem your patch does not apply clean anymore ... thanks!
bye, Heiko
diff --git a/common/cmd_i2c.c b/common/cmd_i2c.c index 3a75f94..7116458 100644 --- a/common/cmd_i2c.c +++ b/common/cmd_i2c.c @@ -280,10 +280,22 @@ static int do_i2c_write(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[ return cmd_usage(cmdtp);
/*
* Length is the number of objects, not number of bytes.
* Length is the number of bytes.+#if defined(CONFIG_SYS_I2C_BULK_WRITE)
- /*
* Write all bytes in a single I2C transaction. If the target* device is an EEPROM, it is your responsibility to not cross* a page bounady.*/- if (i2c_write(chip, devaddr, alen, memaddr, length) != 0) {
puts("Error writing to the chip.\n");return 1;- }
+#else
- /* Perform <length> separate write transactions of one byte each */ while (length-- > 0) { if (i2c_write(chip, devaddr++, alen, memaddr++, 1) != 0) { puts("Error writing to the chip.\n");
@@ -296,6 +308,7 @@ static int do_i2c_write(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[ udelay(11000); #endif } +#endif return 0; }
Hi Heiko,
Hello Lubomir,
Am 24.11.2014 17:00, schrieb Lubomir Popov:
I2C chips do exist that require a write of some multi-byte data to occur in a single bus transaction (aka atomic transfer), otherwise either the write does not come into effect at all, or normal operation of internal circuitry cannot be guaranteed. The current implementation of the 'i2c write' command (transfer of multiple bytes from a memory buffer) in fact performs a separate transaction for each byte to be written and thus cannot support such types of I2C slave devices.
This patch provides an alternative by allowing 'i2c write' to execute the write transfer of the given number of bytes in a single bus transaction if CONFIG_SYS_I2C_BULK_WRITE is defined in the board header (otherwise the old method shall compile).
Signed-off-by: Lubomir Popov l-popov@ti.com
common/cmd_i2c.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)
Could you rebase your patch against current mainline, please? As we have now DM in i2c subsystem your patch does not apply clean anymore ... thanks!
It looks like you are referring V1 of the patch here. The correct version is V2 (http://patchwork.ozlabs.org/patch/415117/). If it doesn't apply as well, I shall try to find some time to fix it.
Thanks, Lubo
bye, Heiko
diff --git a/common/cmd_i2c.c b/common/cmd_i2c.c index 3a75f94..7116458 100644 --- a/common/cmd_i2c.c +++ b/common/cmd_i2c.c @@ -280,10 +280,22 @@ static int do_i2c_write(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[ return cmd_usage(cmdtp);
/*
* Length is the number of objects, not number of bytes.
* Length is the number of bytes.+#if defined(CONFIG_SYS_I2C_BULK_WRITE)
- /*
* Write all bytes in a single I2C transaction. If the target* device is an EEPROM, it is your responsibility to not cross* a page bounady.*/- if (i2c_write(chip, devaddr, alen, memaddr, length) != 0) {
puts("Error writing to the chip.\n");return 1;- }
+#else
- /* Perform <length> separate write transactions of one byte each */ while (length-- > 0) { if (i2c_write(chip, devaddr++, alen, memaddr++, 1) != 0) { puts("Error writing to the chip.\n");
@@ -296,6 +308,7 @@ static int do_i2c_write(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[ udelay(11000); #endif } +#endif return 0; }
-- DENX Software Engineering GmbH, Managing Director: Wolfgang Denk HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Hello Lubomir,
Am 28.01.2015 09:32, schrieb Lubomir Popov:
Hi Heiko,
Hello Lubomir,
Am 24.11.2014 17:00, schrieb Lubomir Popov:
I2C chips do exist that require a write of some multi-byte data to occur in a single bus transaction (aka atomic transfer), otherwise either the write does not come into effect at all, or normal operation of internal circuitry cannot be guaranteed. The current implementation of the 'i2c write' command (transfer of multiple bytes from a memory buffer) in fact performs a separate transaction for each byte to be written and thus cannot support such types of I2C slave devices.
This patch provides an alternative by allowing 'i2c write' to execute the write transfer of the given number of bytes in a single bus transaction if CONFIG_SYS_I2C_BULK_WRITE is defined in the board header (otherwise the old method shall compile).
Signed-off-by: Lubomir Popov l-popov@ti.com
common/cmd_i2c.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)
Could you rebase your patch against current mainline, please? As we have now DM in i2c subsystem your patch does not apply clean anymore ... thanks!
It looks like you are referring V1 of the patch here. The correct version is V2 (http://patchwork.ozlabs.org/patch/415117/). If it doesn't apply as well, I shall try to find some time to fix it.
Sorry, was the wrong EMail I replied ... but the v2 does not apply to current mainline. If you can fix it, that would be great.
Thanks!
bye, Heiko
Thanks, Lubo
bye, Heiko
diff --git a/common/cmd_i2c.c b/common/cmd_i2c.c index 3a75f94..7116458 100644 --- a/common/cmd_i2c.c +++ b/common/cmd_i2c.c @@ -280,10 +280,22 @@ static int do_i2c_write(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[ return cmd_usage(cmdtp);
/*
* Length is the number of objects, not number of bytes.
* Length is the number of bytes.+#if defined(CONFIG_SYS_I2C_BULK_WRITE)
- /*
* Write all bytes in a single I2C transaction. If the target* device is an EEPROM, it is your responsibility to not cross* a page bounady.*/- if (i2c_write(chip, devaddr, alen, memaddr, length) != 0) {
puts("Error writing to the chip.\n");return 1;- }
+#else
- /* Perform <length> separate write transactions of one byte each */ while (length-- > 0) { if (i2c_write(chip, devaddr++, alen, memaddr++, 1) != 0) { puts("Error writing to the chip.\n");
@@ -296,6 +308,7 @@ static int do_i2c_write(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[ udelay(11000); #endif } +#endif return 0; }
-- DENX Software Engineering GmbH, Managing Director: Wolfgang Denk HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Hi Heiko,
Hello Lubomir,
Am 28.01.2015 09:32, schrieb Lubomir Popov:
Hi Heiko,
Hello Lubomir,
Am 24.11.2014 17:00, schrieb Lubomir Popov:
I2C chips do exist that require a write of some multi-byte data to occur in a single bus transaction (aka atomic transfer), otherwise either the write does not come into effect at all, or normal operation of internal circuitry cannot be guaranteed. The current implementation of the 'i2c write' command (transfer of multiple bytes from a memory buffer) in fact performs a separate transaction for each byte to be written and thus cannot support such types of I2C slave devices.
This patch provides an alternative by allowing 'i2c write' to execute the write transfer of the given number of bytes in a single bus transaction if CONFIG_SYS_I2C_BULK_WRITE is defined in the board header (otherwise the old method shall compile).
Signed-off-by: Lubomir Popov l-popov@ti.com
common/cmd_i2c.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)
Could you rebase your patch against current mainline, please? As we have now DM in i2c subsystem your patch does not apply clean anymore ... thanks!
It looks like you are referring V1 of the patch here. The correct version is V2 (http://patchwork.ozlabs.org/patch/415117/). If it doesn't apply as well, I shall try to find some time to fix it.
Sorry, was the wrong EMail I replied ... but the v2 does not apply to current mainline. If you can fix it, that would be great.
OK. Do you have a deadline to finish the I2C updates? I'm really very busy right now and am not sure if I shall manage with this before the end of the week, but can try...
Regards, Lubo
Thanks!
bye, Heiko
Thanks, Lubo
bye, Heiko
diff --git a/common/cmd_i2c.c b/common/cmd_i2c.c index 3a75f94..7116458 100644 --- a/common/cmd_i2c.c +++ b/common/cmd_i2c.c @@ -280,10 +280,22 @@ static int do_i2c_write(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[ return cmd_usage(cmdtp);
/*
* Length is the number of objects, not number of bytes.
* Length is the number of bytes.+#if defined(CONFIG_SYS_I2C_BULK_WRITE)
- /*
* Write all bytes in a single I2C transaction. If the target* device is an EEPROM, it is your responsibility to not cross* a page bounady.*/- if (i2c_write(chip, devaddr, alen, memaddr, length) != 0) {
puts("Error writing to the chip.\n");return 1;- }
+#else
- /* Perform <length> separate write transactions of one byte each */ while (length-- > 0) { if (i2c_write(chip, devaddr++, alen, memaddr++, 1) != 0) { puts("Error writing to the chip.\n");
@@ -296,6 +308,7 @@ static int do_i2c_write(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[ udelay(11000); #endif } +#endif return 0; }
-- DENX Software Engineering GmbH, Managing Director: Wolfgang Denk HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
-- DENX Software Engineering GmbH, Managing Director: Wolfgang Denk HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Hello Lubomir,
Am 28.01.2015 10:00, schrieb Lubomir Popov:
Hi Heiko,
Hello Lubomir,
Am 28.01.2015 09:32, schrieb Lubomir Popov:
Hi Heiko,
Hello Lubomir,
Am 24.11.2014 17:00, schrieb Lubomir Popov:
I2C chips do exist that require a write of some multi-byte data to occur in a single bus transaction (aka atomic transfer), otherwise either the write does not come into effect at all, or normal operation of internal circuitry cannot be guaranteed. The current implementation of the 'i2c write' command (transfer of multiple bytes from a memory buffer) in fact performs a separate transaction for each byte to be written and thus cannot support such types of I2C slave devices.
This patch provides an alternative by allowing 'i2c write' to execute the write transfer of the given number of bytes in a single bus transaction if CONFIG_SYS_I2C_BULK_WRITE is defined in the board header (otherwise the old method shall compile).
Signed-off-by: Lubomir Popov l-popov@ti.com
common/cmd_i2c.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-)Could you rebase your patch against current mainline, please? As we have now DM in i2c subsystem your patch does not apply clean anymore ... thanks!
It looks like you are referring V1 of the patch here. The correct version is V2 (http://patchwork.ozlabs.org/patch/415117/). If it doesn't apply as well, I shall try to find some time to fix it.
Sorry, was the wrong EMail I replied ... but the v2 does not apply to current mainline. If you can fix it, that would be great.
OK. Do you have a deadline to finish the I2C updates? I'm really very busy right now and am not sure if I shall manage with this before the end of the week, but can try...
No, I have no deadline for your patch, as you posted it before the current merge window opened ... but It would be nice to get it in, so we may have tester before the next release ...
Thanks!
bye, Heiko
Regards, Lubo
Thanks!
bye, Heiko
Thanks, Lubo
bye, Heiko
diff --git a/common/cmd_i2c.c b/common/cmd_i2c.c index 3a75f94..7116458 100644 --- a/common/cmd_i2c.c +++ b/common/cmd_i2c.c @@ -280,10 +280,22 @@ static int do_i2c_write(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[ return cmd_usage(cmdtp);
/*
* Length is the number of objects, not number of bytes.
* Length is the number of bytes. */ length = simple_strtoul(argv[4], NULL, 16);+#if defined(CONFIG_SYS_I2C_BULK_WRITE)
- /*
* Write all bytes in a single I2C transaction. If the target* device is an EEPROM, it is your responsibility to not cross* a page bounady.*/- if (i2c_write(chip, devaddr, alen, memaddr, length) != 0) {
puts("Error writing to the chip.\n");return 1;- }
+#else
- /* Perform <length> separate write transactions of one byte each */ while (length-- > 0) { if (i2c_write(chip, devaddr++, alen, memaddr++, 1) != 0) { puts("Error writing to the chip.\n");
@@ -296,6 +308,7 @@ static int do_i2c_write(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[ udelay(11000); #endif } +#endif return 0; }
-- DENX Software Engineering GmbH, Managing Director: Wolfgang Denk HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
-- DENX Software Engineering GmbH, Managing Director: Wolfgang Denk HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
participants (3)
-
Heiko Schocher -
Lubomir Popov -
Simon Glass