This is a simple patch which allows payloads to be placed in memory in the range of 0xf0000-0x100000, where the Coreboot tables live in v2. As long as the payload doesn't need the tables, it seems harmless, so why not just print a warning?
This allows v2 to load "legacybios" without having to have a separate loader.
Signed-off-by: Myles Watson mylesgw@gmail.com
Thanks, Myles
Myles Watson wrote:
This is a simple patch which allows payloads to be placed in memory in the range of 0xf0000-0x100000, where the Coreboot tables live in v2. As long as the payload doesn't need the tables, it seems harmless, so why not just print a warning?
It will definitely break your interrupt routing ... While I agree this is OK for testing, we have to find a better solution for this.
-----Original Message----- From: Stefan Reinauer [mailto:stepan@coresystems.de] Sent: Wednesday, May 21, 2008 3:57 PM To: Myles Watson Cc: Coreboot Subject: Re: [coreboot] Allow payloads to write over Coreboot tables, add a warning
Myles Watson wrote:
This is a simple patch which allows payloads to be placed in memory in the range of 0xf0000-0x100000, where the Coreboot tables live in v2. As long as the payload doesn't need the tables, it seems harmless, so why not just print a warning?
It will definitely break your interrupt routing ... While I agree this is OK for testing, we have to find a better solution for this.
Yes.
As soon as we decide how we want to pass information to legacybios, it should be clear what the better solution is.
One solution is a libpayload-based loader that copies the information and passes the relevant parts to legacybios. Until legacybios uses any of that information, though, it might as well be copied over.
Thanks, Myles
Myles Watson wrote:
-----Original Message----- From: Stefan Reinauer [mailto:stepan@coresystems.de] Sent: Wednesday, May 21, 2008 3:57 PM To: Myles Watson Cc: Coreboot Subject: Re: [coreboot] Allow payloads to write over Coreboot tables, add a warning
Myles Watson wrote:
This is a simple patch which allows payloads to be placed in memory in the range of 0xf0000-0x100000, where the Coreboot tables live in v2. As long as the payload doesn't need the tables, it seems harmless, so why not just print a warning?
It will definitely break your interrupt routing ... While I agree this is OK for testing, we have to find a better solution for this.
Yes.
As soon as we decide how we want to pass information to legacybios, it should be clear what the better solution is.
One solution is a libpayload-based loader that copies the information and passes the relevant parts to legacybios. Until legacybios uses any of that information, though, it might as well be copied over.
It'll be fine for testing and doesn't really break anything that did work before...
Acked-by: Stefan Reinauer stepan@coresystems.de
-----Original Message----- From: Stefan Reinauer [mailto:stepan@coresystems.de] Sent: Wednesday, May 21, 2008 4:03 PM To: Myles Watson Cc: 'Coreboot' Subject: Re: [coreboot] Allow payloads to write over Coreboot tables, add a warning
Myles Watson wrote:
-----Original Message----- From: Stefan Reinauer [mailto:stepan@coresystems.de] Sent: Wednesday, May 21, 2008 3:57 PM To: Myles Watson Cc: Coreboot Subject: Re: [coreboot] Allow payloads to write over Coreboot tables,
add
a warning
Myles Watson wrote:
This is a simple patch which allows payloads to be placed in memory in the range of 0xf0000-0x100000, where the Coreboot tables live in v2. As long as the payload doesn't need the tables, it seems harmless, so why not just print a warning?
It will definitely break your interrupt routing ... While I agree this is OK for testing, we have to find a better solution for this.
Yes.
As soon as we decide how we want to pass information to legacybios, it should be clear what the better solution is.
One solution is a libpayload-based loader that copies the information
and
passes the relevant parts to legacybios. Until legacybios uses any of
that
information, though, it might as well be copied over.
It'll be fine for testing and doesn't really break anything that did work before...
Acked-by: Stefan Reinauer stepan@coresystems.de
Rev 3343. Thanks, Myles
On Wed, May 21, 2008 at 04:01:40PM -0600, Myles Watson wrote:
One solution is a libpayload-based loader that copies the information and passes the relevant parts to legacybios.
I like this.
//Peter
-
Myles Watson -
Peter Stuge -
Stefan Reinauer