[PATCH] Add ACPI generic resume support

[Rudolf Marek]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello,

Following patch implements ACPI resume support for coreboot. The hardware main
hook will come in separate patch perhaps.

Signed-off-by: Rudolf Marek

Rudolf
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkngvEEACgkQ3J9wPJqZRNWQJgCeLV/QW++cZ+O4Jm/gX/F3O8TA
b2gAn2WDZjRLy9FsmuqRaXHgDjfhE4MT
=BbSl
-----END PGP SIGNATURE-----

[Peter Stuge]

Rudolf Marek wrote:
> Following patch implements ACPI resume support for coreboot. The hardware main
> hook will come in separate patch perhaps.
>
> Signed-off-by: Rudolf Marek

Acked-by: Peter Stuge

[Stefan Reinauer]

Wonderful!

On 11.04.2009 17:50 Uhr, Rudolf Marek wrote:
> Index: src/arch/i386/boot/Config.lb
>
> --- src/arch/i386/boot/Config.lb	(revision 4093)
> +++ src/arch/i386/boot/Config.lb	(working copy)
> @@ -14,4 +14,5 @@
>  if HAVE_ACPI_TABLES
>  object acpi.o
>  object acpigen.o
> +object wakeup.S
>  end

Can you please guard the code with CONFIG_RESUME_SUPPORT? Not everyone
wants the code included.
> Index: src/arch/i386/boot/acpi.c
>
> --- src/arch/i386/boot/acpi.c	(revision 4093)
> +++ src/arch/i386/boot/acpi.c	(working copy)
> @@ -27,6 +27,9 @@
> +int acpi_is_wakeup(void) {
> return (acpi_slp_type == 3);
>
> +}

Please, for functions, put the { on a new line.
> +static acpi_rsdp_t *valid_rsdp(acpi_rsdp_t *rsdp) {
> if (strncmp((char *)rsdp, RSDP_SIG, sizeof(RSDP_SIG) - 1) != 0)
> return NULL;

Maybe you can compare a 32bit number here instead of doing a strncmp?
Not sure if this is any timing critical, but it looks like a lot of byte
wise memory access can be spared here.
> +extern char *lowmem_backup;
> +extern char *lowmem_backup_ptr;
> +extern int lowmem_backup_size;

Please also put stuff like that in CONFIG_RESUME_SUPPORT

[Rudolf Marek]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello,
> Can you please guard the code with CONFIG_RESUME_SUPPORT? Not everyone
> wants the code included.

Yes I can but same as for the rest file I was just avoiding to have #ifdef
forests. I recently did some work in u-boot and it was a #ifdef hell.

Do you think I should put to ifdef all added functions (as a block)?

Btw someone can please have a look to rest of the patches?

Thanks,
Rudolf
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkniLmsACgkQ3J9wPJqZRNVYxwCg1jst+ZO7sUX9KYh4JO80Y205
MrEAoMo95SdzszfBT78a7qd/mx8B7LA/
=KlDE
-----END PGP SIGNATURE-----

[ron minnich]

On Sun, Apr 12, 2009 at 11:09 AM, Rudolf Marek  wrote:
> Do you think I should put to ifdef all added functions (as a block)?

Either that, or is seperate compilation an option?
> Btw someone can please have a look to rest of the patches?

Which ones? I have not been able to follow this closely.

ron

[Rudolf Marek]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
> Which ones? I have not been able to follow this closely.

The remaining one from me ;)

[coreboot] [PATCH] AMD revF resume support
[coreboot] [PATCH] ACPI general support for resume
[coreboot] [PATCH] VT8237/K8M890 resume support

R.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkniM0oACgkQ3J9wPJqZRNUK4ACgzUOPBz9FtTpI/rWMMIJX4Yo1
i9gAn1whTiekhdd6fpXFVKkMOIk7UVw9
=Y8od
-----END PGP SIGNATURE-----

[Rudolf Marek]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi all,

I fixed that as Stephan suggested. Committed revision 4101.

Rudolf
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknjf2wACgkQ3J9wPJqZRNWP7wCeO5ejiePFeJyzW5UvN1KMtfoE
YvMAn2EH/IludoCUUXAfP5tzYGbMtftw
=FD8x
-----END PGP SIGNATURE-----