According to IT8718F_V0 3_(for C version).pdf, it looks like 0xf0 should be 0x08.
Signed-off-by: Josh Profitt zorn169@gmail.com
On Thu, Oct 09, 2008 at 08:26:09AM -0400, Josh wrote:
According to IT8718F_V0 3_(for C version).pdf, it looks like 0xf0 should be 0x08.
Signed-off-by: Josh Profitt zorn169@gmail.com
Index: util/superiotool/ite.c
--- util/superiotool/ite.c (revision 3642) +++ util/superiotool/ite.c (working copy) @@ -295,7 +295,7 @@ 0x00,NANA,NANA,EOT}}, {0x5, "Keyboard", {0x30,0x60,0x61,0x62,0x63,0x70,0x71,0xf0,EOT},
{0x01,0x00,0x60,0x00,0x64,0x01,0x02,0x00,EOT}},
{0x01,0x00,0x60,0x00,0x64,0x01,0x02,0x08,EOT}},
Maybe, maybe not. Page 56 says 0x08, but on page 32 it says 0x00. The datasheet contradicts itself here, and we don't know which one is actually the correct default. We should probably document this in the code, though.
The difference is that KBC clock is either 8 or 12 MHz, which is board-dependent, unclear what could be the default for the chip.
Uwe.
On Thu, Oct 9, 2008 at 10:59 AM, Uwe Hermann uwe@hermann-uwe.de wrote:
On Thu, Oct 09, 2008 at 08:26:09AM -0400, Josh wrote:
According to IT8718F_V0 3_(for C version).pdf, it looks like 0xf0 should be 0x08.
Signed-off-by: Josh Profitt zorn169@gmail.com
Index: util/superiotool/ite.c
--- util/superiotool/ite.c (revision 3642) +++ util/superiotool/ite.c (working copy) @@ -295,7 +295,7 @@ 0x00,NANA,NANA,EOT}}, {0x5, "Keyboard", {0x30,0x60,0x61,0x62,0x63,0x70,0x71,0xf0,EOT},
{0x01,0x00,0x60,0x00,0x64,0x01,0x02,0x00,EOT}},
{0x01,0x00,0x60,0x00,0x64,0x01,0x02,0x08,EOT}},Maybe, maybe not. Page 56 says 0x08, but on page 32 it says 0x00. The datasheet contradicts itself here, and we don't know which one is actually the correct default. We should probably document this in the code, though.
The difference is that KBC clock is either 8 or 12 MHz, which is board-dependent, unclear what could be the default for the chip.
So shouldn't we make this configurable via mainboard Config.lb, and use the value dumped by superiotool?
-Corey
On 09.10.2008 20:08, Corey Osgood wrote:
On Thu, Oct 9, 2008 at 10:59 AM, Uwe Hermann uwe@hermann-uwe.de wrote:
On Thu, Oct 09, 2008 at 08:26:09AM -0400, Josh wrote:
According to IT8718F_V0 3_(for C version).pdf, it looks like 0xf0 should be 0x08.
Signed-off-by: Josh Profitt zorn169@gmail.com
Index: util/superiotool/ite.c
--- util/superiotool/ite.c (revision 3642) +++ util/superiotool/ite.c (working copy) @@ -295,7 +295,7 @@ 0x00,NANA,NANA,EOT}}, {0x5, "Keyboard", {0x30,0x60,0x61,0x62,0x63,0x70,0x71,0xf0,EOT},
{0x01,0x00,0x60,0x00,0x64,0x01,0x02,0x00,EOT}},
{0x01,0x00,0x60,0x00,0x64,0x01,0x02,0x08,EOT}},Maybe, maybe not. Page 56 says 0x08, but on page 32 it says 0x00. The datasheet contradicts itself here, and we don't know which one is actually the correct default. We should probably document this in the code, though.
The difference is that KBC clock is either 8 or 12 MHz, which is board-dependent, unclear what could be the default for the chip.
So shouldn't we make this configurable via mainboard Config.lb, and use the value dumped by superiotool?
Why not simply boot a board with that superio under coreboot where the value is left uninitialized? Or is there no such coreboot-supported board?
Regards, Carl-Daniel
On Thu, Oct 09, 2008 at 02:08:32PM -0400, Corey Osgood wrote:
Index: util/superiotool/ite.c
--- util/superiotool/ite.c (revision 3642) +++ util/superiotool/ite.c (working copy) @@ -295,7 +295,7 @@ 0x00,NANA,NANA,EOT}}, {0x5, "Keyboard", {0x30,0x60,0x61,0x62,0x63,0x70,0x71,0xf0,EOT},
{0x01,0x00,0x60,0x00,0x64,0x01,0x02,0x00,EOT}},
{0x01,0x00,0x60,0x00,0x64,0x01,0x02,0x08,EOT}},Maybe, maybe not. Page 56 says 0x08, but on page 32 it says 0x00. The datasheet contradicts itself here, and we don't know which one is actually the correct default. We should probably document this in the code, though.
The difference is that KBC clock is either 8 or 12 MHz, which is board-dependent, unclear what could be the default for the chip.
So shouldn't we make this configurable via mainboard Config.lb, and use the value dumped by superiotool?
If the keyboard doesn't work for some board, this could be one reason, yes. Keyboard does work for most boards out of the box though. And yes, I think you can already set this via Config.lb. However, I don't think we actually have a board with IT8718F so far, so this is not yet relevant :)
Uwe.
On 10.10.2008 00:41, Uwe Hermann wrote:
On Thu, Oct 09, 2008 at 02:08:32PM -0400, Corey Osgood wrote:
Index: util/superiotool/ite.c
--- util/superiotool/ite.c (revision 3642) +++ util/superiotool/ite.c (working copy) @@ -295,7 +295,7 @@ 0x00,NANA,NANA,EOT}}, {0x5, "Keyboard", {0x30,0x60,0x61,0x62,0x63,0x70,0x71,0xf0,EOT},
{0x01,0x00,0x60,0x00,0x64,0x01,0x02,0x00,EOT}},
{0x01,0x00,0x60,0x00,0x64,0x01,0x02,0x08,EOT}},Maybe, maybe not. Page 56 says 0x08, but on page 32 it says 0x00. The datasheet contradicts itself here, and we don't know which one is actually the correct default. We should probably document this in the code, though.
The difference is that KBC clock is either 8 or 12 MHz, which is board-dependent, unclear what could be the default for the chip.
So shouldn't we make this configurable via mainboard Config.lb, and use the value dumped by superiotool?
If the keyboard doesn't work for some board, this could be one reason, yes. Keyboard does work for most boards out of the box though. And yes, I think you can already set this via Config.lb. However, I don't think we actually have a board with IT8718F so far, so this is not yet relevant :)
In theory, asking ITE could work as well, but all my communication attempts have been met with deafening silence...
Regards, Carl-Daniel
On Thu, 09 Oct 2008, Uwe Hermann wrote:
Maybe, maybe not. Page 56 says 0x08, but on page 32 it says 0x00. The datasheet contradicts itself here, and we don't know which one is actually the correct default. We should probably document this in the code, though.
The difference is that KBC clock is either 8 or 12 MHz, which is board-dependent, unclear what could be the default for the chip.
Ok, I see. There was at least one other discrepancy like that in the datasheets, so I'll at least document them in my it8726f superiotool patch.
Josh
-
Carl-Daniel Hailfinger -
Corey Osgood -
Josh -
Uwe Hermann