See patch.
Uwe.
On 06.10.2009 16:34, Uwe Hermann wrote:
* Make the ROM write-protected.
Does that mean we have to unprotect the ROM in flashrom?
Regards, Carl-Daniel
On Tue, Oct 06, 2009 at 04:42:54PM +0200, Carl-Daniel Hailfinger wrote:
On 06.10.2009 16:34, Uwe Hermann wrote:
* Make the ROM write-protected.
Does that mean we have to unprotect the ROM in flashrom?
Regards, Carl-Daniel
I am happy with that if, and only if, the flashrom chipset enable takes care of that.
Luc Verhaegen.
On Tue, Oct 06, 2009 at 04:49:22PM +0200, Luc Verhaegen wrote:
* Make the ROM write-protected.
Does that mean we have to unprotect the ROM in flashrom?
Yes.
I think long-term we want a (runtime/CMOS) option for coreboot to enable/disable ROM protection. For now, I guess every chipset does a different thing and/or doesn't explicitly protect or unprotect the ROM at all.
As a safe default I think disabling the ROM write access is the best thing to do from a user's point of view.
I am happy with that if, and only if, the flashrom chipset enable takes care of that.
Sure, flashrom does that. And for other chipsets where we might write-protect the ROM later we also have flashrom chipset support. I cannot think of any chipset where we have coreboot support but no flashrom support (even for CK804 and MCP55 we have).
Uwe.
On Tue, Oct 6, 2009 at 8:49 AM, Uwe Hermann uwe@hermann-uwe.de wrote:
I think long-term we want a (runtime/CMOS) option for coreboot to enable/disable ROM protection. For now, I guess every chipset does a different thing and/or doesn't explicitly protect or unprotect the ROM at all.
Going back to the beginning, from the old days, we always left it write-enabled. That was amistake on our part and it is good to see it being fixed.
ron
Dear Uwe,
Am Dienstag, den 06.10.2009, 16:34 +0200 schrieb Uwe Hermann:
Index: src/southbridge/amd/cs5530/cs5530.h
--- src/southbridge/amd/cs5530/cs5530.h (Revision 4726) +++ src/southbridge/amd/cs5530/cs5530.h (Arbeitskopie) @@ -27,7 +27,13 @@
[…]
+#define LOWER_ROM_ADDRESS_RANGE (1 << 0) +#define ROM_WRITE_ENABLE (1 << 1) +#define UPPER_ROM_ADDRESS_RANGE (1 << 2) +#define BIOS_ROM_POSITIVE_DECODE (1 << 5)
Is this spacing intentional?
Thanks,
Paul
On Tue, Oct 06, 2009 at 08:16:17PM +0200, Paul Menzel wrote:
Dear Uwe,
Am Dienstag, den 06.10.2009, 16:34 +0200 schrieb Uwe Hermann:
Index: src/southbridge/amd/cs5530/cs5530.h
--- src/southbridge/amd/cs5530/cs5530.h (Revision 4726) +++ src/southbridge/amd/cs5530/cs5530.h (Arbeitskopie) @@ -27,7 +27,13 @@
[…]
+#define LOWER_ROM_ADDRESS_RANGE (1 << 0) +#define ROM_WRITE_ENABLE (1 << 1) +#define UPPER_ROM_ADDRESS_RANGE (1 << 2) +#define BIOS_ROM_POSITIVE_DECODE (1 << 5)
Is this spacing intentional?
Yes. It looks strange in the patch/email, but if you apply the patch and look at the file in a text editor it'll be nicely aligned.
Uwe.
On Tue, Oct 6, 2009 at 8:34 AM, Uwe Hermann uwe@hermann-uwe.de wrote:
See patch.
Uwe,
Can the call go into a cs5530_enable_rom() call go into a the southbridge init instead of every mainboard? Would it ever be mainboard specific?
Marc
Am Dienstag, den 06.10.2009, 16:34 +0200 schrieb Uwe Hermann:
See patch.
Acked-by: Patrick Georgi patrick.georgi@coresystems.de
The rom enable call can go into some common southbridge code once the code flow allows it (currently it does not)
On Wed, Oct 07, 2009 at 04:18:06PM +0200, Patrick Georgi wrote:
Am Dienstag, den 06.10.2009, 16:34 +0200 schrieb Uwe Hermann:
See patch.
Acked-by: Patrick Georgi patrick.georgi@coresystems.de
The rom enable call can go into some common southbridge code once the code flow allows it (currently it does not)
Thanks, r4731. Will take care of moving the calls later, as soon as the infrastructure is there.
Uwe.