Skip to contentSkip to conversation

i2c: i801: Drop useless masking in i801_access

PATCH Applied Jean Delvare 3 participants 8 messages Mar 1, 2022 – Mar 3, 2022

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/2 i2c: i801: Drop useless masking in i801_access

I went all the way back to kernel 2.6.12 and that masking was never needed. I suppose it was there in anticipation of software PEC support, but that was never added to the driver (and never will be, as this is made obsolete by hardware PEC).

I'm also removing initialization to 0, which is not needed either, and would prevent the compiler from reporting an actual usage of uninitialized variables.

---
 drivers/i2c/busses/i2c-i801.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/i2c/busses/i2c-i801.c +2 -2
@@ -792,7 +792,7 @@ static s32 i801_access(struct i2c_adapte
 {
 	int hwpec;
 	int block = 0;
-	int ret = 0, xact = 0;
+	int ret, xact;
 	struct i801_priv *priv = i2c_get_adapdata(adap);
 
 	mutex_lock(&priv->acpi_lock);
@@ -904,7 +904,7 @@ static s32 i801_access(struct i2c_adapte
 	if ((read_write == I2C_SMBUS_WRITE) || (xact == I801_QUICK))
 		goto out;
 
-	switch (xact & 0x7f) {
+	switch (xact) {
 	case I801_BYTE:	/* Result put in SMBHSTDAT0 */
 	case I801_BYTE_DATA:
 		data->byte = inb_p(SMBHSTDAT0(priv));
Signature · 2 lines
Jean Delvare
SUSE L3 Support
#1 lore raw

PATCH 2/2 i2c: i801: Add support for the Process Call command

The Process Call command is implemented by the hardware since the very first Intel 82801 chipset, and trivial to support. Oscar Romero Matamala from the Georgia Institute of Technology told me it is needed for an experiment his team is working on at the moment, so let's just add support for it.

---
Disclaimer: untested code. I don't have any device at hand which supports the Process Call command so I just can't test it. If anyone has a chance, please test. If not, I hope Oscar will be able to test it soon.

 drivers/i2c/busses/i2c-i801.c |   12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)
drivers/i2c/busses/i2c-i801.c +11 -1
@@ -166,7 +166,7 @@
 #define I801_BYTE		0x04
 #define I801_BYTE_DATA		0x08
 #define I801_WORD_DATA		0x0C
-#define I801_PROC_CALL		0x10	/* unimplemented */
+#define I801_PROC_CALL		0x10
 #define I801_BLOCK_DATA		0x14
 #define I801_I2C_BLOCK_DATA	0x18	/* ICH5 and later */
 #define I801_BLOCK_PROC_CALL	0x1C
@@ -838,6 +838,14 @@ static s32 i801_access(struct i2c_adapte
 		}
 		xact = I801_WORD_DATA;
 		break;
+	case I2C_SMBUS_PROC_CALL:
+		outb_p((addr & 0x7f) << 1, SMBHSTADD(priv));
+		outb_p(command, SMBHSTCMD(priv));
+		outb_p(data->word & 0xff, SMBHSTDAT0(priv));
+		outb_p((data->word & 0xff00) >> 8, SMBHSTDAT1(priv));
+		xact = I801_PROC_CALL;
+		read_write = I2C_SMBUS_READ;
+		break;
 	case I2C_SMBUS_BLOCK_DATA:
 		outb_p(((addr & 0x7f) << 1) | (read_write & 0x01),
 		       SMBHSTADD(priv));
@@ -910,6 +918,7 @@ static s32 i801_access(struct i2c_adapte
 		data->byte = inb_p(SMBHSTDAT0(priv));
 		break;
 	case I801_WORD_DATA:
+	case I801_PROC_CALL:
 		data->word = inb_p(SMBHSTDAT0(priv)) +
 			     (inb_p(SMBHSTDAT1(priv)) << 8);
 		break;
@@ -935,6 +944,7 @@ static u32 i801_func(struct i2c_adapter
 
 	return I2C_FUNC_SMBUS_QUICK | I2C_FUNC_SMBUS_BYTE |
 	       I2C_FUNC_SMBUS_BYTE_DATA | I2C_FUNC_SMBUS_WORD_DATA |
+	       I2C_FUNC_SMBUS_PROC_CALL |
 	       I2C_FUNC_SMBUS_BLOCK_DATA | I2C_FUNC_SMBUS_WRITE_I2C_BLOCK |
 	       ((priv->features & FEATURE_SMBUS_PEC) ? I2C_FUNC_SMBUS_PEC : 0) |
 	       ((priv->features & FEATURE_BLOCK_PROC) ?
Signature · 2 lines
Jean Delvare
SUSE L3 Support
Reviewed-by on v1 2/2
#2 lore raw

Re: [PATCH 2/2] i2c: i801: Add support for the Process Call command

Quoted 7 lines · Jean Delvare

The Process Call command is implemented by the hardware since the very first Intel 82801 chipset, and trivial to support. Oscar Romero Matamala from the Georgia Institute of Technology told me it is needed for an experiment his team is working on at the moment, so let's just add support for it.

Quoted 9 lines

---
Disclaimer: untested code. I don't have any device at hand which supports the Process Call command so I just can't test it. If anyone has a chance, please test. If not, I hope Oscar will be able to test it soon.

  drivers/i2c/busses/i2c-i801.c |   12 +++++++++++-
  1 file changed, 11 insertions(+), 1 deletion(-)

As far as I can see this follows how it is depicted in the specification.

Applied
#3 lore raw
In reply to Jean Delvare
Quoted 10 lines · Jean Delvare

I went all the way back to kernel 2.6.12 and that masking was never needed. I suppose it was there in anticipation of software PEC support, but that was never added to the driver (and never will be, as this is made obsolete by hardware PEC).

I'm also removing initialization to 0, which is not needed either, and would prevent the compiler from reporting an actual usage of uninitialized variables.

Applied to for-next, thanks!

Applied
#4 lore raw
In reply to Jean Delvare
Quoted 10 lines · Jean Delvare

I went all the way back to kernel 2.6.12 and that masking was never needed. I suppose it was there in anticipation of software PEC support, but that was never added to the driver (and never will be, as this is made obsolete by hardware PEC).

I'm also removing initialization to 0, which is not needed either, and would prevent the compiler from reporting an actual usage of uninitialized variables.

Applied to for-next, thanks!

Jean, there are still some patches pending for i801, mainly from Heiner and an interesting one from Hector. Do you have time for these?

http://patchwork.ozlabs.org/project/linux-i2c/list/?series=&submitter=&state=&q=i801&archive=&delegate=

#5 lore raw

Hi Wolfram,

Quoted 10 lines · Wolfram Sang
Quoted 9 lines · Jean Delvare

I went all the way back to kernel 2.6.12 and that masking was never needed. I suppose it was there in anticipation of software PEC support, but that was never added to the driver (and never will be, as this is made obsolete by hardware PEC).

I'm also removing initialization to 0, which is not needed either, and would prevent the compiler from reporting an actual usage of uninitialized variables.

Quoted 8 lines
Quoted 1 line

Applied to for-next, thanks!

Jean, there are still some patches pending for i801, mainly from Heiner and an interesting one from Hector. Do you have time for these?

http://patchwork.ozlabs.org/project/linux-i2c/list/?series=&submitter=&state=&q=i801&archive=&delegate=

Yes, they are on my to-do list and not forgotten. Just I had a fairly busy January and February. Plus I am currently investigating 2 i2c-i801 issues that have been reported to me privately, and I'd rather get them fixed first before we continue with driver cleanups.

Stay tuned,

Signature · 2 lines
Jean Delvare
SUSE L3 Support
#6 lore raw
Quoted 5 lines

Yes, they are on my to-do list and not forgotten. Just I had a fairly busy January and February. Plus I am currently investigating 2 i2c-i801 issues that have been reported to me privately, and I'd rather get them fixed first before we continue with driver cleanups.

OK, thanks for the heads up!

Reviewed-by on v1 1/2
#7 lore raw
In reply to Jean Delvare
Quoted 5 lines · Jean Delvare

I went all the way back to kernel 2.6.12 and that masking was never needed. I suppose it was there in anticipation of software PEC support, but that was never added to the driver (and never will be, as this is made obsolete by hardware PEC).

Quoted 9 lines

I'm also removing initialization to 0, which is not needed either, and would prevent the compiler from reporting an actual usage of uninitialized variables.

---
  drivers/i2c/busses/i2c-i801.c |    4 ++--
  1 file changed, 2 insertions(+), 2 deletions(-)