[PATCH 1/2] drm/radeon: implement ACPI VFCT vbios fetch (v2)

2012-08-17 Thread Matthew Garrett
On Fri, Aug 17, 2012 at 12:56:31PM -0400, Alex Deucher wrote:
> > I guess we could leave it as is for now for -fixes and then switch it
> > use use the new exported symbol for -next?  Is it ok to export a new
> > symbol for -fixes?

I don't see why not, providing the ACPI people are happy with it.

> >> So, er, I had no clue how to clean up the return value of acpi_get_table
> >> - does this actually need to be cleaned up?  Or do you just get a
> >> pointer straight to the "real" ACPI table?
> >
> > Not sure on that.  Anyone know more about the acpi code?

Not sure what you're asking here - acpi_get_table returns acpi_status, 
not a pointer.

-- 
Matthew Garrett | mjg59 at srcf.ucam.org


[PATCH 1/2] drm/radeon: implement ACPI VFCT vbios fetch (v2)

2012-08-17 Thread Alex Deucher
Adding Matthew in case he as any ideas.

On Thu, Aug 16, 2012 at 3:51 PM, Alex Deucher  wrote:
> On Thu, Aug 16, 2012 at 3:40 PM, David Lamparter  
> wrote:
>> On Thu, Aug 16, 2012 at 03:13:46PM -0400, alexdeucher at gmail.com wrote:
>>> From: David L 
>> From: David Lamparter 
>>
>> There are still two rough edges left in here, I didn't get around to
>> clean it up, other stuff came up -- sorry...
>>
>>> This is required for pure UEFI systems.  The vbios is stored
>>> in ACPI rather than at the legacy vga location.
>>>
>>> Fixes:
>>> https://bugs.freedesktop.org/show_bug.cgi?id=26891
>>>
>>> V2: fix #ifdefs as per Greg's comments
>>>
>>> Signed-off-by: Alex Deucher 
>>> Cc: stable at vger.kernel.org
>>> ---
>> [...]
>>> + struct acpi_table_header *hdr;
>>> + /* acpi_get_table_with_size is not exported :( */
>>> + acpi_size tbl_size = 0x7fff;
>>
>> I was using acpi_get_table_with_size, but that needs an export, with
>> 0x7fff all the tests below are kinda useless because they always
>> succeed.  I think it'd be useful to keep the length checks in case some
>> vendor breaks their ACPI tables, so this needs an EXPORT_SYMBOL.
>
> I guess we could leave it as is for now for -fixes and then switch it
> use use the new exported symbol for -next?  Is it ok to export a new
> symbol for -fixes?
>
>>
>>> + UEFI_ACPI_VFCT *vfct;
>>> + GOP_VBIOS_CONTENT *vbios;
>>> + VFCT_IMAGE_HEADER *vhdr;
>>> +
>>> + if (!ACPI_SUCCESS(acpi_get_table("VFCT", 1, &hdr)))
>>> + return false;
>>> + if (tbl_size < sizeof(UEFI_ACPI_VFCT)) {
>>> + DRM_ERROR("ACPI VFCT table present but broken (too short 
>>> #1)\n");
>>> + goto out_unmap;
>>> + }
>>> +
>>> + vfct = (UEFI_ACPI_VFCT *)hdr;
>>> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) > tbl_size) {
>>> + DRM_ERROR("ACPI VFCT table present but broken (too short 
>>> #2)\n");
>>> + goto out_unmap;
>>> + }
>>> +
>>> + vbios = (GOP_VBIOS_CONTENT *)((char *)hdr + vfct->VBIOSImageOffset);
>>> + vhdr = &vbios->VbiosHeader;
>>> + DRM_INFO("ACPI VFCT contains a BIOS for %02x:%02x.%d %04x:%04x, size 
>>> %d\n",
>>> + vhdr->PCIBus, vhdr->PCIDevice, vhdr->PCIFunction,
>>> + vhdr->VendorID, vhdr->DeviceID, vhdr->ImageLength);
>>> +
>>> + if (vhdr->PCIBus != rdev->pdev->bus->number ||
>>> + vhdr->PCIDevice != PCI_SLOT(rdev->pdev->devfn) ||
>>> + vhdr->PCIFunction != PCI_FUNC(rdev->pdev->devfn) ||
>>> + vhdr->VendorID != rdev->pdev->vendor ||
>>> + vhdr->DeviceID != rdev->pdev->device) {
>>> + DRM_INFO("ACPI VFCT table is not for this card\n");
>>> + goto out_unmap;
>>> + };
>>> +
>>> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) + 
>>> vhdr->ImageLength > tbl_size) {
>>> + DRM_ERROR("ACPI VFCT image truncated\n");
>>> + goto out_unmap;
>>> + }
>>> +
>>> + rdev->bios = kmemdup(&vbios->VbiosContent, vhdr->ImageLength, 
>>> GFP_KERNEL);
>>> + ret = !!rdev->bios;
>>> +
>>> +out_unmap:
>>> + /* uh, no idea what to do here... */
>>
>> So, er, I had no clue how to clean up the return value of acpi_get_table
>> - does this actually need to be cleaned up?  Or do you just get a
>> pointer straight to the "real" ACPI table?
>
> Not sure on that.  Anyone know more about the acpi code?
>
> Alex


Re: [PATCH 1/2] drm/radeon: implement ACPI VFCT vbios fetch (v2)

2012-08-17 Thread Matthew Garrett
On Fri, Aug 17, 2012 at 12:56:31PM -0400, Alex Deucher wrote:
> > I guess we could leave it as is for now for -fixes and then switch it
> > use use the new exported symbol for -next?  Is it ok to export a new
> > symbol for -fixes?

I don't see why not, providing the ACPI people are happy with it.

> >> So, er, I had no clue how to clean up the return value of acpi_get_table
> >> - does this actually need to be cleaned up?  Or do you just get a
> >> pointer straight to the "real" ACPI table?
> >
> > Not sure on that.  Anyone know more about the acpi code?

Not sure what you're asking here - acpi_get_table returns acpi_status, 
not a pointer.

-- 
Matthew Garrett | mj...@srcf.ucam.org
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel


Re: [PATCH 1/2] drm/radeon: implement ACPI VFCT vbios fetch (v2)

2012-08-17 Thread Alex Deucher
Adding Matthew in case he as any ideas.

On Thu, Aug 16, 2012 at 3:51 PM, Alex Deucher  wrote:
> On Thu, Aug 16, 2012 at 3:40 PM, David Lamparter  wrote:
>> On Thu, Aug 16, 2012 at 03:13:46PM -0400, alexdeuc...@gmail.com wrote:
>>> From: David L 
>> From: David Lamparter 
>>
>> There are still two rough edges left in here, I didn't get around to
>> clean it up, other stuff came up -- sorry...
>>
>>> This is required for pure UEFI systems.  The vbios is stored
>>> in ACPI rather than at the legacy vga location.
>>>
>>> Fixes:
>>> https://bugs.freedesktop.org/show_bug.cgi?id=26891
>>>
>>> V2: fix #ifdefs as per Greg's comments
>>>
>>> Signed-off-by: Alex Deucher 
>>> Cc: sta...@vger.kernel.org
>>> ---
>> [...]
>>> + struct acpi_table_header *hdr;
>>> + /* acpi_get_table_with_size is not exported :( */
>>> + acpi_size tbl_size = 0x7fff;
>>
>> I was using acpi_get_table_with_size, but that needs an export, with
>> 0x7fff all the tests below are kinda useless because they always
>> succeed.  I think it'd be useful to keep the length checks in case some
>> vendor breaks their ACPI tables, so this needs an EXPORT_SYMBOL.
>
> I guess we could leave it as is for now for -fixes and then switch it
> use use the new exported symbol for -next?  Is it ok to export a new
> symbol for -fixes?
>
>>
>>> + UEFI_ACPI_VFCT *vfct;
>>> + GOP_VBIOS_CONTENT *vbios;
>>> + VFCT_IMAGE_HEADER *vhdr;
>>> +
>>> + if (!ACPI_SUCCESS(acpi_get_table("VFCT", 1, &hdr)))
>>> + return false;
>>> + if (tbl_size < sizeof(UEFI_ACPI_VFCT)) {
>>> + DRM_ERROR("ACPI VFCT table present but broken (too short 
>>> #1)\n");
>>> + goto out_unmap;
>>> + }
>>> +
>>> + vfct = (UEFI_ACPI_VFCT *)hdr;
>>> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) > tbl_size) {
>>> + DRM_ERROR("ACPI VFCT table present but broken (too short 
>>> #2)\n");
>>> + goto out_unmap;
>>> + }
>>> +
>>> + vbios = (GOP_VBIOS_CONTENT *)((char *)hdr + vfct->VBIOSImageOffset);
>>> + vhdr = &vbios->VbiosHeader;
>>> + DRM_INFO("ACPI VFCT contains a BIOS for %02x:%02x.%d %04x:%04x, size 
>>> %d\n",
>>> + vhdr->PCIBus, vhdr->PCIDevice, vhdr->PCIFunction,
>>> + vhdr->VendorID, vhdr->DeviceID, vhdr->ImageLength);
>>> +
>>> + if (vhdr->PCIBus != rdev->pdev->bus->number ||
>>> + vhdr->PCIDevice != PCI_SLOT(rdev->pdev->devfn) ||
>>> + vhdr->PCIFunction != PCI_FUNC(rdev->pdev->devfn) ||
>>> + vhdr->VendorID != rdev->pdev->vendor ||
>>> + vhdr->DeviceID != rdev->pdev->device) {
>>> + DRM_INFO("ACPI VFCT table is not for this card\n");
>>> + goto out_unmap;
>>> + };
>>> +
>>> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) + 
>>> vhdr->ImageLength > tbl_size) {
>>> + DRM_ERROR("ACPI VFCT image truncated\n");
>>> + goto out_unmap;
>>> + }
>>> +
>>> + rdev->bios = kmemdup(&vbios->VbiosContent, vhdr->ImageLength, 
>>> GFP_KERNEL);
>>> + ret = !!rdev->bios;
>>> +
>>> +out_unmap:
>>> + /* uh, no idea what to do here... */
>>
>> So, er, I had no clue how to clean up the return value of acpi_get_table
>> - does this actually need to be cleaned up?  Or do you just get a
>> pointer straight to the "real" ACPI table?
>
> Not sure on that.  Anyone know more about the acpi code?
>
> Alex
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel


Re: [PATCH 1/2] drm/radeon: implement ACPI VFCT vbios fetch (v2)

2012-08-17 Thread Greg KH
On Thu, Aug 16, 2012 at 03:13:46PM -0400, alexdeuc...@gmail.com wrote:
> From: David L 
> 
> This is required for pure UEFI systems.  The vbios is stored
> in ACPI rather than at the legacy vga location.
> 
> Fixes:
> https://bugs.freedesktop.org/show_bug.cgi?id=26891
> 
> V2: fix #ifdefs as per Greg's comments
> 
> Signed-off-by: Alex Deucher 
> Cc: sta...@vger.kernel.org
> ---
>  drivers/gpu/drm/radeon/radeon_bios.c |   62 
> ++
>  1 files changed, 62 insertions(+), 0 deletions(-)
> 
> diff --git a/drivers/gpu/drm/radeon/radeon_bios.c 
> b/drivers/gpu/drm/radeon/radeon_bios.c
> index 501f488..bf21f65 100644
> --- a/drivers/gpu/drm/radeon/radeon_bios.c
> +++ b/drivers/gpu/drm/radeon/radeon_bios.c
> @@ -32,6 +32,9 @@
>  
>  #include 
>  #include 
> +#ifdef CONFIG_ACPI
> +#include 
> +#endif

You forgot to remove this one :(

>  /*
>   * BIOS.
>   */
> @@ -476,6 +479,63 @@ static bool radeon_read_disabled_bios(struct 
> radeon_device *rdev)
>   return legacy_read_disabled_bios(rdev);
>  }
>  
> +#ifdef CONFIG_ACPI
> +static bool radeon_acpi_vfct_bios(struct radeon_device *rdev)
> +{
> + bool ret = false;
> + struct acpi_table_header *hdr;
> + /* acpi_get_table_with_size is not exported :( */
> + acpi_size tbl_size = 0x7fff;
> + UEFI_ACPI_VFCT *vfct;
> + GOP_VBIOS_CONTENT *vbios;
> + VFCT_IMAGE_HEADER *vhdr;
> +
> + if (!ACPI_SUCCESS(acpi_get_table("VFCT", 1, &hdr)))
> + return false;
> + if (tbl_size < sizeof(UEFI_ACPI_VFCT)) {
> + DRM_ERROR("ACPI VFCT table present but broken (too short 
> #1)\n");
> + goto out_unmap;
> + }
> +
> + vfct = (UEFI_ACPI_VFCT *)hdr;
> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) > tbl_size) {
> + DRM_ERROR("ACPI VFCT table present but broken (too short 
> #2)\n");
> + goto out_unmap;
> + }
> +
> + vbios = (GOP_VBIOS_CONTENT *)((char *)hdr + vfct->VBIOSImageOffset);
> + vhdr = &vbios->VbiosHeader;
> + DRM_INFO("ACPI VFCT contains a BIOS for %02x:%02x.%d %04x:%04x, size 
> %d\n",
> + vhdr->PCIBus, vhdr->PCIDevice, vhdr->PCIFunction,
> + vhdr->VendorID, vhdr->DeviceID, vhdr->ImageLength);
> +
> + if (vhdr->PCIBus != rdev->pdev->bus->number ||
> + vhdr->PCIDevice != PCI_SLOT(rdev->pdev->devfn) ||
> + vhdr->PCIFunction != PCI_FUNC(rdev->pdev->devfn) ||
> + vhdr->VendorID != rdev->pdev->vendor ||
> + vhdr->DeviceID != rdev->pdev->device) {
> + DRM_INFO("ACPI VFCT table is not for this card\n");
> + goto out_unmap;
> + };
> +
> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) + 
> vhdr->ImageLength > tbl_size) {
> + DRM_ERROR("ACPI VFCT image truncated\n");
> + goto out_unmap;
> + }
> +
> + rdev->bios = kmemdup(&vbios->VbiosContent, vhdr->ImageLength, 
> GFP_KERNEL);
> + ret = !!rdev->bios;
> +
> +out_unmap:
> + /* uh, no idea what to do here... */
> + return ret;
> +}
> +#else
> +static bool radeon_acpi_vfct_bios(struct radeon_device *rdev)

Make it inline, and then the compiler is smart enough to just delete the
whole if () check you make when you call it.

Third time's a charm?

greg k-h
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel


[PATCH 1/2] drm/radeon: implement ACPI VFCT vbios fetch (v2)

2012-08-16 Thread David Lamparter
On Thu, Aug 16, 2012 at 03:13:46PM -0400, alexdeucher at gmail.com wrote:
> From: David L 
From: David Lamparter 

There are still two rough edges left in here, I didn't get around to
clean it up, other stuff came up -- sorry...

> This is required for pure UEFI systems.  The vbios is stored
> in ACPI rather than at the legacy vga location.
> 
> Fixes:
> https://bugs.freedesktop.org/show_bug.cgi?id=26891
> 
> V2: fix #ifdefs as per Greg's comments
> 
> Signed-off-by: Alex Deucher 
> Cc: stable at vger.kernel.org
> ---
[...]
> + struct acpi_table_header *hdr;
> + /* acpi_get_table_with_size is not exported :( */
> + acpi_size tbl_size = 0x7fff;

I was using acpi_get_table_with_size, but that needs an export, with
0x7fff all the tests below are kinda useless because they always
succeed.  I think it'd be useful to keep the length checks in case some
vendor breaks their ACPI tables, so this needs an EXPORT_SYMBOL.

> + UEFI_ACPI_VFCT *vfct;
> + GOP_VBIOS_CONTENT *vbios;
> + VFCT_IMAGE_HEADER *vhdr;
> +
> + if (!ACPI_SUCCESS(acpi_get_table("VFCT", 1, &hdr)))
> + return false;
> + if (tbl_size < sizeof(UEFI_ACPI_VFCT)) {
> + DRM_ERROR("ACPI VFCT table present but broken (too short 
> #1)\n");
> + goto out_unmap;
> + }
> +
> + vfct = (UEFI_ACPI_VFCT *)hdr;
> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) > tbl_size) {
> + DRM_ERROR("ACPI VFCT table present but broken (too short 
> #2)\n");
> + goto out_unmap;
> + }
> +
> + vbios = (GOP_VBIOS_CONTENT *)((char *)hdr + vfct->VBIOSImageOffset);
> + vhdr = &vbios->VbiosHeader;
> + DRM_INFO("ACPI VFCT contains a BIOS for %02x:%02x.%d %04x:%04x, size 
> %d\n",
> + vhdr->PCIBus, vhdr->PCIDevice, vhdr->PCIFunction,
> + vhdr->VendorID, vhdr->DeviceID, vhdr->ImageLength);
> +
> + if (vhdr->PCIBus != rdev->pdev->bus->number ||
> + vhdr->PCIDevice != PCI_SLOT(rdev->pdev->devfn) ||
> + vhdr->PCIFunction != PCI_FUNC(rdev->pdev->devfn) ||
> + vhdr->VendorID != rdev->pdev->vendor ||
> + vhdr->DeviceID != rdev->pdev->device) {
> + DRM_INFO("ACPI VFCT table is not for this card\n");
> + goto out_unmap;
> + };
> +
> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) + 
> vhdr->ImageLength > tbl_size) {
> + DRM_ERROR("ACPI VFCT image truncated\n");
> + goto out_unmap;
> + }
> +
> + rdev->bios = kmemdup(&vbios->VbiosContent, vhdr->ImageLength, 
> GFP_KERNEL);
> + ret = !!rdev->bios;
> +
> +out_unmap:
> + /* uh, no idea what to do here... */

So, er, I had no clue how to clean up the return value of acpi_get_table
- does this actually need to be cleaned up?  Or do you just get a
pointer straight to the "real" ACPI table?

> + return ret;
> +}
> +#else
> +static bool radeon_acpi_vfct_bios(struct radeon_device *rdev)
> +{
> + return false;
> +}
> +#endif
>  
>  bool radeon_get_bios(struct radeon_device *rdev)
>  {
> @@ -484,6 +544,8 @@ bool radeon_get_bios(struct radeon_device *rdev)
>  
>   r = radeon_atrm_get_bios(rdev);
>   if (r == false)
> + r = radeon_acpi_vfct_bios(rdev);
> + if (r == false)
>   r = igp_read_bios_from_vram(rdev);
>   if (r == false)
>   r = radeon_read_bios(rdev);
> -- 
> 1.7.7.5
> 

Cheers,

-David
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 230 bytes
Desc: Digital signature
URL: 



[PATCH 1/2] drm/radeon: implement ACPI VFCT vbios fetch (v2)

2012-08-16 Thread Alex Deucher
On Thu, Aug 16, 2012 at 3:40 PM, David Lamparter  wrote:
> On Thu, Aug 16, 2012 at 03:13:46PM -0400, alexdeucher at gmail.com wrote:
>> From: David L 
> From: David Lamparter 
>
> There are still two rough edges left in here, I didn't get around to
> clean it up, other stuff came up -- sorry...
>
>> This is required for pure UEFI systems.  The vbios is stored
>> in ACPI rather than at the legacy vga location.
>>
>> Fixes:
>> https://bugs.freedesktop.org/show_bug.cgi?id=26891
>>
>> V2: fix #ifdefs as per Greg's comments
>>
>> Signed-off-by: Alex Deucher 
>> Cc: stable at vger.kernel.org
>> ---
> [...]
>> + struct acpi_table_header *hdr;
>> + /* acpi_get_table_with_size is not exported :( */
>> + acpi_size tbl_size = 0x7fff;
>
> I was using acpi_get_table_with_size, but that needs an export, with
> 0x7fff all the tests below are kinda useless because they always
> succeed.  I think it'd be useful to keep the length checks in case some
> vendor breaks their ACPI tables, so this needs an EXPORT_SYMBOL.

I guess we could leave it as is for now for -fixes and then switch it
use use the new exported symbol for -next?  Is it ok to export a new
symbol for -fixes?

>
>> + UEFI_ACPI_VFCT *vfct;
>> + GOP_VBIOS_CONTENT *vbios;
>> + VFCT_IMAGE_HEADER *vhdr;
>> +
>> + if (!ACPI_SUCCESS(acpi_get_table("VFCT", 1, &hdr)))
>> + return false;
>> + if (tbl_size < sizeof(UEFI_ACPI_VFCT)) {
>> + DRM_ERROR("ACPI VFCT table present but broken (too short 
>> #1)\n");
>> + goto out_unmap;
>> + }
>> +
>> + vfct = (UEFI_ACPI_VFCT *)hdr;
>> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) > tbl_size) {
>> + DRM_ERROR("ACPI VFCT table present but broken (too short 
>> #2)\n");
>> + goto out_unmap;
>> + }
>> +
>> + vbios = (GOP_VBIOS_CONTENT *)((char *)hdr + vfct->VBIOSImageOffset);
>> + vhdr = &vbios->VbiosHeader;
>> + DRM_INFO("ACPI VFCT contains a BIOS for %02x:%02x.%d %04x:%04x, size 
>> %d\n",
>> + vhdr->PCIBus, vhdr->PCIDevice, vhdr->PCIFunction,
>> + vhdr->VendorID, vhdr->DeviceID, vhdr->ImageLength);
>> +
>> + if (vhdr->PCIBus != rdev->pdev->bus->number ||
>> + vhdr->PCIDevice != PCI_SLOT(rdev->pdev->devfn) ||
>> + vhdr->PCIFunction != PCI_FUNC(rdev->pdev->devfn) ||
>> + vhdr->VendorID != rdev->pdev->vendor ||
>> + vhdr->DeviceID != rdev->pdev->device) {
>> + DRM_INFO("ACPI VFCT table is not for this card\n");
>> + goto out_unmap;
>> + };
>> +
>> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) + 
>> vhdr->ImageLength > tbl_size) {
>> + DRM_ERROR("ACPI VFCT image truncated\n");
>> + goto out_unmap;
>> + }
>> +
>> + rdev->bios = kmemdup(&vbios->VbiosContent, vhdr->ImageLength, 
>> GFP_KERNEL);
>> + ret = !!rdev->bios;
>> +
>> +out_unmap:
>> + /* uh, no idea what to do here... */
>
> So, er, I had no clue how to clean up the return value of acpi_get_table
> - does this actually need to be cleaned up?  Or do you just get a
> pointer straight to the "real" ACPI table?

Not sure on that.  Anyone know more about the acpi code?

Alex


[PATCH 1/2] drm/radeon: implement ACPI VFCT vbios fetch (v2)

2012-08-16 Thread Alex Deucher
On Thu, Aug 16, 2012 at 3:25 PM, Greg KH  wrote:
> On Thu, Aug 16, 2012 at 03:13:46PM -0400, alexdeucher at gmail.com wrote:
>> From: David L 
>>
>> This is required for pure UEFI systems.  The vbios is stored
>> in ACPI rather than at the legacy vga location.
>>
>> Fixes:
>> https://bugs.freedesktop.org/show_bug.cgi?id=26891
>>
>> V2: fix #ifdefs as per Greg's comments
>>
>> Signed-off-by: Alex Deucher 
>> Cc: stable at vger.kernel.org
>> ---
>>  drivers/gpu/drm/radeon/radeon_bios.c |   62 
>> ++
>>  1 files changed, 62 insertions(+), 0 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/radeon/radeon_bios.c 
>> b/drivers/gpu/drm/radeon/radeon_bios.c
>> index 501f488..bf21f65 100644
>> --- a/drivers/gpu/drm/radeon/radeon_bios.c
>> +++ b/drivers/gpu/drm/radeon/radeon_bios.c
>> @@ -32,6 +32,9 @@
>>
>>  #include 
>>  #include 
>> +#ifdef CONFIG_ACPI
>> +#include 
>> +#endif
>
> You forgot to remove this one :(

Argh!  I squashed that into the wrong patch when I rebased my fixes.

>
>>  /*
>>   * BIOS.
>>   */
>> @@ -476,6 +479,63 @@ static bool radeon_read_disabled_bios(struct 
>> radeon_device *rdev)
>>   return legacy_read_disabled_bios(rdev);
>>  }
>>
>> +#ifdef CONFIG_ACPI
>> +static bool radeon_acpi_vfct_bios(struct radeon_device *rdev)
>> +{
>> + bool ret = false;
>> + struct acpi_table_header *hdr;
>> + /* acpi_get_table_with_size is not exported :( */
>> + acpi_size tbl_size = 0x7fff;
>> + UEFI_ACPI_VFCT *vfct;
>> + GOP_VBIOS_CONTENT *vbios;
>> + VFCT_IMAGE_HEADER *vhdr;
>> +
>> + if (!ACPI_SUCCESS(acpi_get_table("VFCT", 1, &hdr)))
>> + return false;
>> + if (tbl_size < sizeof(UEFI_ACPI_VFCT)) {
>> + DRM_ERROR("ACPI VFCT table present but broken (too short 
>> #1)\n");
>> + goto out_unmap;
>> + }
>> +
>> + vfct = (UEFI_ACPI_VFCT *)hdr;
>> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) > tbl_size) {
>> + DRM_ERROR("ACPI VFCT table present but broken (too short 
>> #2)\n");
>> + goto out_unmap;
>> + }
>> +
>> + vbios = (GOP_VBIOS_CONTENT *)((char *)hdr + vfct->VBIOSImageOffset);
>> + vhdr = &vbios->VbiosHeader;
>> + DRM_INFO("ACPI VFCT contains a BIOS for %02x:%02x.%d %04x:%04x, size 
>> %d\n",
>> + vhdr->PCIBus, vhdr->PCIDevice, vhdr->PCIFunction,
>> + vhdr->VendorID, vhdr->DeviceID, vhdr->ImageLength);
>> +
>> + if (vhdr->PCIBus != rdev->pdev->bus->number ||
>> + vhdr->PCIDevice != PCI_SLOT(rdev->pdev->devfn) ||
>> + vhdr->PCIFunction != PCI_FUNC(rdev->pdev->devfn) ||
>> + vhdr->VendorID != rdev->pdev->vendor ||
>> + vhdr->DeviceID != rdev->pdev->device) {
>> + DRM_INFO("ACPI VFCT table is not for this card\n");
>> + goto out_unmap;
>> + };
>> +
>> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) + 
>> vhdr->ImageLength > tbl_size) {
>> + DRM_ERROR("ACPI VFCT image truncated\n");
>> + goto out_unmap;
>> + }
>> +
>> + rdev->bios = kmemdup(&vbios->VbiosContent, vhdr->ImageLength, 
>> GFP_KERNEL);
>> + ret = !!rdev->bios;
>> +
>> +out_unmap:
>> + /* uh, no idea what to do here... */
>> + return ret;
>> +}
>> +#else
>> +static bool radeon_acpi_vfct_bios(struct radeon_device *rdev)
>
> Make it inline, and then the compiler is smart enough to just delete the
> whole if () check you make when you call it.
>
> Third time's a charm?

On the way.

Alex


[PATCH 1/2] drm/radeon: implement ACPI VFCT vbios fetch (v2)

2012-08-16 Thread alexdeuc...@gmail.com
From: David L 

This is required for pure UEFI systems.  The vbios is stored
in ACPI rather than at the legacy vga location.

Fixes:
https://bugs.freedesktop.org/show_bug.cgi?id=26891

V2: fix #ifdefs as per Greg's comments

Signed-off-by: Alex Deucher 
Cc: stable at vger.kernel.org
---
 drivers/gpu/drm/radeon/radeon_bios.c |   62 ++
 1 files changed, 62 insertions(+), 0 deletions(-)

diff --git a/drivers/gpu/drm/radeon/radeon_bios.c 
b/drivers/gpu/drm/radeon/radeon_bios.c
index 501f488..bf21f65 100644
--- a/drivers/gpu/drm/radeon/radeon_bios.c
+++ b/drivers/gpu/drm/radeon/radeon_bios.c
@@ -32,6 +32,9 @@

 #include 
 #include 
+#ifdef CONFIG_ACPI
+#include 
+#endif
 /*
  * BIOS.
  */
@@ -476,6 +479,63 @@ static bool radeon_read_disabled_bios(struct radeon_device 
*rdev)
return legacy_read_disabled_bios(rdev);
 }

+#ifdef CONFIG_ACPI
+static bool radeon_acpi_vfct_bios(struct radeon_device *rdev)
+{
+   bool ret = false;
+   struct acpi_table_header *hdr;
+   /* acpi_get_table_with_size is not exported :( */
+   acpi_size tbl_size = 0x7fff;
+   UEFI_ACPI_VFCT *vfct;
+   GOP_VBIOS_CONTENT *vbios;
+   VFCT_IMAGE_HEADER *vhdr;
+
+   if (!ACPI_SUCCESS(acpi_get_table("VFCT", 1, &hdr)))
+   return false;
+   if (tbl_size < sizeof(UEFI_ACPI_VFCT)) {
+   DRM_ERROR("ACPI VFCT table present but broken (too short 
#1)\n");
+   goto out_unmap;
+   }
+
+   vfct = (UEFI_ACPI_VFCT *)hdr;
+   if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) > tbl_size) {
+   DRM_ERROR("ACPI VFCT table present but broken (too short 
#2)\n");
+   goto out_unmap;
+   }
+
+   vbios = (GOP_VBIOS_CONTENT *)((char *)hdr + vfct->VBIOSImageOffset);
+   vhdr = &vbios->VbiosHeader;
+   DRM_INFO("ACPI VFCT contains a BIOS for %02x:%02x.%d %04x:%04x, size 
%d\n",
+   vhdr->PCIBus, vhdr->PCIDevice, vhdr->PCIFunction,
+   vhdr->VendorID, vhdr->DeviceID, vhdr->ImageLength);
+
+   if (vhdr->PCIBus != rdev->pdev->bus->number ||
+   vhdr->PCIDevice != PCI_SLOT(rdev->pdev->devfn) ||
+   vhdr->PCIFunction != PCI_FUNC(rdev->pdev->devfn) ||
+   vhdr->VendorID != rdev->pdev->vendor ||
+   vhdr->DeviceID != rdev->pdev->device) {
+   DRM_INFO("ACPI VFCT table is not for this card\n");
+   goto out_unmap;
+   };
+
+   if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) + 
vhdr->ImageLength > tbl_size) {
+   DRM_ERROR("ACPI VFCT image truncated\n");
+   goto out_unmap;
+   }
+
+   rdev->bios = kmemdup(&vbios->VbiosContent, vhdr->ImageLength, 
GFP_KERNEL);
+   ret = !!rdev->bios;
+
+out_unmap:
+   /* uh, no idea what to do here... */
+   return ret;
+}
+#else
+static bool radeon_acpi_vfct_bios(struct radeon_device *rdev)
+{
+   return false;
+}
+#endif

 bool radeon_get_bios(struct radeon_device *rdev)
 {
@@ -484,6 +544,8 @@ bool radeon_get_bios(struct radeon_device *rdev)

r = radeon_atrm_get_bios(rdev);
if (r == false)
+   r = radeon_acpi_vfct_bios(rdev);
+   if (r == false)
r = igp_read_bios_from_vram(rdev);
if (r == false)
r = radeon_read_bios(rdev);
-- 
1.7.7.5



Re: [PATCH 1/2] drm/radeon: implement ACPI VFCT vbios fetch (v2)

2012-08-16 Thread David Lamparter
On Thu, Aug 16, 2012 at 03:13:46PM -0400, alexdeuc...@gmail.com wrote:
> From: David L 
From: David Lamparter 

There are still two rough edges left in here, I didn't get around to
clean it up, other stuff came up -- sorry...

> This is required for pure UEFI systems.  The vbios is stored
> in ACPI rather than at the legacy vga location.
> 
> Fixes:
> https://bugs.freedesktop.org/show_bug.cgi?id=26891
> 
> V2: fix #ifdefs as per Greg's comments
> 
> Signed-off-by: Alex Deucher 
> Cc: sta...@vger.kernel.org
> ---
[...]
> + struct acpi_table_header *hdr;
> + /* acpi_get_table_with_size is not exported :( */
> + acpi_size tbl_size = 0x7fff;

I was using acpi_get_table_with_size, but that needs an export, with
0x7fff all the tests below are kinda useless because they always
succeed.  I think it'd be useful to keep the length checks in case some
vendor breaks their ACPI tables, so this needs an EXPORT_SYMBOL.

> + UEFI_ACPI_VFCT *vfct;
> + GOP_VBIOS_CONTENT *vbios;
> + VFCT_IMAGE_HEADER *vhdr;
> +
> + if (!ACPI_SUCCESS(acpi_get_table("VFCT", 1, &hdr)))
> + return false;
> + if (tbl_size < sizeof(UEFI_ACPI_VFCT)) {
> + DRM_ERROR("ACPI VFCT table present but broken (too short 
> #1)\n");
> + goto out_unmap;
> + }
> +
> + vfct = (UEFI_ACPI_VFCT *)hdr;
> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) > tbl_size) {
> + DRM_ERROR("ACPI VFCT table present but broken (too short 
> #2)\n");
> + goto out_unmap;
> + }
> +
> + vbios = (GOP_VBIOS_CONTENT *)((char *)hdr + vfct->VBIOSImageOffset);
> + vhdr = &vbios->VbiosHeader;
> + DRM_INFO("ACPI VFCT contains a BIOS for %02x:%02x.%d %04x:%04x, size 
> %d\n",
> + vhdr->PCIBus, vhdr->PCIDevice, vhdr->PCIFunction,
> + vhdr->VendorID, vhdr->DeviceID, vhdr->ImageLength);
> +
> + if (vhdr->PCIBus != rdev->pdev->bus->number ||
> + vhdr->PCIDevice != PCI_SLOT(rdev->pdev->devfn) ||
> + vhdr->PCIFunction != PCI_FUNC(rdev->pdev->devfn) ||
> + vhdr->VendorID != rdev->pdev->vendor ||
> + vhdr->DeviceID != rdev->pdev->device) {
> + DRM_INFO("ACPI VFCT table is not for this card\n");
> + goto out_unmap;
> + };
> +
> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) + 
> vhdr->ImageLength > tbl_size) {
> + DRM_ERROR("ACPI VFCT image truncated\n");
> + goto out_unmap;
> + }
> +
> + rdev->bios = kmemdup(&vbios->VbiosContent, vhdr->ImageLength, 
> GFP_KERNEL);
> + ret = !!rdev->bios;
> +
> +out_unmap:
> + /* uh, no idea what to do here... */

So, er, I had no clue how to clean up the return value of acpi_get_table
- does this actually need to be cleaned up?  Or do you just get a
pointer straight to the "real" ACPI table?

> + return ret;
> +}
> +#else
> +static bool radeon_acpi_vfct_bios(struct radeon_device *rdev)
> +{
> + return false;
> +}
> +#endif
>  
>  bool radeon_get_bios(struct radeon_device *rdev)
>  {
> @@ -484,6 +544,8 @@ bool radeon_get_bios(struct radeon_device *rdev)
>  
>   r = radeon_atrm_get_bios(rdev);
>   if (r == false)
> + r = radeon_acpi_vfct_bios(rdev);
> + if (r == false)
>   r = igp_read_bios_from_vram(rdev);
>   if (r == false)
>   r = radeon_read_bios(rdev);
> -- 
> 1.7.7.5
> 

Cheers,

-David


signature.asc
Description: Digital signature
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel


Re: [PATCH 1/2] drm/radeon: implement ACPI VFCT vbios fetch (v2)

2012-08-16 Thread Alex Deucher
On Thu, Aug 16, 2012 at 3:40 PM, David Lamparter  wrote:
> On Thu, Aug 16, 2012 at 03:13:46PM -0400, alexdeuc...@gmail.com wrote:
>> From: David L 
> From: David Lamparter 
>
> There are still two rough edges left in here, I didn't get around to
> clean it up, other stuff came up -- sorry...
>
>> This is required for pure UEFI systems.  The vbios is stored
>> in ACPI rather than at the legacy vga location.
>>
>> Fixes:
>> https://bugs.freedesktop.org/show_bug.cgi?id=26891
>>
>> V2: fix #ifdefs as per Greg's comments
>>
>> Signed-off-by: Alex Deucher 
>> Cc: sta...@vger.kernel.org
>> ---
> [...]
>> + struct acpi_table_header *hdr;
>> + /* acpi_get_table_with_size is not exported :( */
>> + acpi_size tbl_size = 0x7fff;
>
> I was using acpi_get_table_with_size, but that needs an export, with
> 0x7fff all the tests below are kinda useless because they always
> succeed.  I think it'd be useful to keep the length checks in case some
> vendor breaks their ACPI tables, so this needs an EXPORT_SYMBOL.

I guess we could leave it as is for now for -fixes and then switch it
use use the new exported symbol for -next?  Is it ok to export a new
symbol for -fixes?

>
>> + UEFI_ACPI_VFCT *vfct;
>> + GOP_VBIOS_CONTENT *vbios;
>> + VFCT_IMAGE_HEADER *vhdr;
>> +
>> + if (!ACPI_SUCCESS(acpi_get_table("VFCT", 1, &hdr)))
>> + return false;
>> + if (tbl_size < sizeof(UEFI_ACPI_VFCT)) {
>> + DRM_ERROR("ACPI VFCT table present but broken (too short 
>> #1)\n");
>> + goto out_unmap;
>> + }
>> +
>> + vfct = (UEFI_ACPI_VFCT *)hdr;
>> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) > tbl_size) {
>> + DRM_ERROR("ACPI VFCT table present but broken (too short 
>> #2)\n");
>> + goto out_unmap;
>> + }
>> +
>> + vbios = (GOP_VBIOS_CONTENT *)((char *)hdr + vfct->VBIOSImageOffset);
>> + vhdr = &vbios->VbiosHeader;
>> + DRM_INFO("ACPI VFCT contains a BIOS for %02x:%02x.%d %04x:%04x, size 
>> %d\n",
>> + vhdr->PCIBus, vhdr->PCIDevice, vhdr->PCIFunction,
>> + vhdr->VendorID, vhdr->DeviceID, vhdr->ImageLength);
>> +
>> + if (vhdr->PCIBus != rdev->pdev->bus->number ||
>> + vhdr->PCIDevice != PCI_SLOT(rdev->pdev->devfn) ||
>> + vhdr->PCIFunction != PCI_FUNC(rdev->pdev->devfn) ||
>> + vhdr->VendorID != rdev->pdev->vendor ||
>> + vhdr->DeviceID != rdev->pdev->device) {
>> + DRM_INFO("ACPI VFCT table is not for this card\n");
>> + goto out_unmap;
>> + };
>> +
>> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) + 
>> vhdr->ImageLength > tbl_size) {
>> + DRM_ERROR("ACPI VFCT image truncated\n");
>> + goto out_unmap;
>> + }
>> +
>> + rdev->bios = kmemdup(&vbios->VbiosContent, vhdr->ImageLength, 
>> GFP_KERNEL);
>> + ret = !!rdev->bios;
>> +
>> +out_unmap:
>> + /* uh, no idea what to do here... */
>
> So, er, I had no clue how to clean up the return value of acpi_get_table
> - does this actually need to be cleaned up?  Or do you just get a
> pointer straight to the "real" ACPI table?

Not sure on that.  Anyone know more about the acpi code?

Alex
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel


Re: [PATCH 1/2] drm/radeon: implement ACPI VFCT vbios fetch (v2)

2012-08-16 Thread Alex Deucher
On Thu, Aug 16, 2012 at 3:25 PM, Greg KH  wrote:
> On Thu, Aug 16, 2012 at 03:13:46PM -0400, alexdeuc...@gmail.com wrote:
>> From: David L 
>>
>> This is required for pure UEFI systems.  The vbios is stored
>> in ACPI rather than at the legacy vga location.
>>
>> Fixes:
>> https://bugs.freedesktop.org/show_bug.cgi?id=26891
>>
>> V2: fix #ifdefs as per Greg's comments
>>
>> Signed-off-by: Alex Deucher 
>> Cc: sta...@vger.kernel.org
>> ---
>>  drivers/gpu/drm/radeon/radeon_bios.c |   62 
>> ++
>>  1 files changed, 62 insertions(+), 0 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/radeon/radeon_bios.c 
>> b/drivers/gpu/drm/radeon/radeon_bios.c
>> index 501f488..bf21f65 100644
>> --- a/drivers/gpu/drm/radeon/radeon_bios.c
>> +++ b/drivers/gpu/drm/radeon/radeon_bios.c
>> @@ -32,6 +32,9 @@
>>
>>  #include 
>>  #include 
>> +#ifdef CONFIG_ACPI
>> +#include 
>> +#endif
>
> You forgot to remove this one :(

Argh!  I squashed that into the wrong patch when I rebased my fixes.

>
>>  /*
>>   * BIOS.
>>   */
>> @@ -476,6 +479,63 @@ static bool radeon_read_disabled_bios(struct 
>> radeon_device *rdev)
>>   return legacy_read_disabled_bios(rdev);
>>  }
>>
>> +#ifdef CONFIG_ACPI
>> +static bool radeon_acpi_vfct_bios(struct radeon_device *rdev)
>> +{
>> + bool ret = false;
>> + struct acpi_table_header *hdr;
>> + /* acpi_get_table_with_size is not exported :( */
>> + acpi_size tbl_size = 0x7fff;
>> + UEFI_ACPI_VFCT *vfct;
>> + GOP_VBIOS_CONTENT *vbios;
>> + VFCT_IMAGE_HEADER *vhdr;
>> +
>> + if (!ACPI_SUCCESS(acpi_get_table("VFCT", 1, &hdr)))
>> + return false;
>> + if (tbl_size < sizeof(UEFI_ACPI_VFCT)) {
>> + DRM_ERROR("ACPI VFCT table present but broken (too short 
>> #1)\n");
>> + goto out_unmap;
>> + }
>> +
>> + vfct = (UEFI_ACPI_VFCT *)hdr;
>> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) > tbl_size) {
>> + DRM_ERROR("ACPI VFCT table present but broken (too short 
>> #2)\n");
>> + goto out_unmap;
>> + }
>> +
>> + vbios = (GOP_VBIOS_CONTENT *)((char *)hdr + vfct->VBIOSImageOffset);
>> + vhdr = &vbios->VbiosHeader;
>> + DRM_INFO("ACPI VFCT contains a BIOS for %02x:%02x.%d %04x:%04x, size 
>> %d\n",
>> + vhdr->PCIBus, vhdr->PCIDevice, vhdr->PCIFunction,
>> + vhdr->VendorID, vhdr->DeviceID, vhdr->ImageLength);
>> +
>> + if (vhdr->PCIBus != rdev->pdev->bus->number ||
>> + vhdr->PCIDevice != PCI_SLOT(rdev->pdev->devfn) ||
>> + vhdr->PCIFunction != PCI_FUNC(rdev->pdev->devfn) ||
>> + vhdr->VendorID != rdev->pdev->vendor ||
>> + vhdr->DeviceID != rdev->pdev->device) {
>> + DRM_INFO("ACPI VFCT table is not for this card\n");
>> + goto out_unmap;
>> + };
>> +
>> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) + 
>> vhdr->ImageLength > tbl_size) {
>> + DRM_ERROR("ACPI VFCT image truncated\n");
>> + goto out_unmap;
>> + }
>> +
>> + rdev->bios = kmemdup(&vbios->VbiosContent, vhdr->ImageLength, 
>> GFP_KERNEL);
>> + ret = !!rdev->bios;
>> +
>> +out_unmap:
>> + /* uh, no idea what to do here... */
>> + return ret;
>> +}
>> +#else
>> +static bool radeon_acpi_vfct_bios(struct radeon_device *rdev)
>
> Make it inline, and then the compiler is smart enough to just delete the
> whole if () check you make when you call it.
>
> Third time's a charm?

On the way.

Alex
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel


[PATCH 1/2] drm/radeon: implement ACPI VFCT vbios fetch (v2)

2012-08-16 Thread Greg KH
On Thu, Aug 16, 2012 at 03:13:46PM -0400, alexdeucher at gmail.com wrote:
> From: David L 
> 
> This is required for pure UEFI systems.  The vbios is stored
> in ACPI rather than at the legacy vga location.
> 
> Fixes:
> https://bugs.freedesktop.org/show_bug.cgi?id=26891
> 
> V2: fix #ifdefs as per Greg's comments
> 
> Signed-off-by: Alex Deucher 
> Cc: stable at vger.kernel.org
> ---
>  drivers/gpu/drm/radeon/radeon_bios.c |   62 
> ++
>  1 files changed, 62 insertions(+), 0 deletions(-)
> 
> diff --git a/drivers/gpu/drm/radeon/radeon_bios.c 
> b/drivers/gpu/drm/radeon/radeon_bios.c
> index 501f488..bf21f65 100644
> --- a/drivers/gpu/drm/radeon/radeon_bios.c
> +++ b/drivers/gpu/drm/radeon/radeon_bios.c
> @@ -32,6 +32,9 @@
>  
>  #include 
>  #include 
> +#ifdef CONFIG_ACPI
> +#include 
> +#endif

You forgot to remove this one :(

>  /*
>   * BIOS.
>   */
> @@ -476,6 +479,63 @@ static bool radeon_read_disabled_bios(struct 
> radeon_device *rdev)
>   return legacy_read_disabled_bios(rdev);
>  }
>  
> +#ifdef CONFIG_ACPI
> +static bool radeon_acpi_vfct_bios(struct radeon_device *rdev)
> +{
> + bool ret = false;
> + struct acpi_table_header *hdr;
> + /* acpi_get_table_with_size is not exported :( */
> + acpi_size tbl_size = 0x7fff;
> + UEFI_ACPI_VFCT *vfct;
> + GOP_VBIOS_CONTENT *vbios;
> + VFCT_IMAGE_HEADER *vhdr;
> +
> + if (!ACPI_SUCCESS(acpi_get_table("VFCT", 1, &hdr)))
> + return false;
> + if (tbl_size < sizeof(UEFI_ACPI_VFCT)) {
> + DRM_ERROR("ACPI VFCT table present but broken (too short 
> #1)\n");
> + goto out_unmap;
> + }
> +
> + vfct = (UEFI_ACPI_VFCT *)hdr;
> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) > tbl_size) {
> + DRM_ERROR("ACPI VFCT table present but broken (too short 
> #2)\n");
> + goto out_unmap;
> + }
> +
> + vbios = (GOP_VBIOS_CONTENT *)((char *)hdr + vfct->VBIOSImageOffset);
> + vhdr = &vbios->VbiosHeader;
> + DRM_INFO("ACPI VFCT contains a BIOS for %02x:%02x.%d %04x:%04x, size 
> %d\n",
> + vhdr->PCIBus, vhdr->PCIDevice, vhdr->PCIFunction,
> + vhdr->VendorID, vhdr->DeviceID, vhdr->ImageLength);
> +
> + if (vhdr->PCIBus != rdev->pdev->bus->number ||
> + vhdr->PCIDevice != PCI_SLOT(rdev->pdev->devfn) ||
> + vhdr->PCIFunction != PCI_FUNC(rdev->pdev->devfn) ||
> + vhdr->VendorID != rdev->pdev->vendor ||
> + vhdr->DeviceID != rdev->pdev->device) {
> + DRM_INFO("ACPI VFCT table is not for this card\n");
> + goto out_unmap;
> + };
> +
> + if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) + 
> vhdr->ImageLength > tbl_size) {
> + DRM_ERROR("ACPI VFCT image truncated\n");
> + goto out_unmap;
> + }
> +
> + rdev->bios = kmemdup(&vbios->VbiosContent, vhdr->ImageLength, 
> GFP_KERNEL);
> + ret = !!rdev->bios;
> +
> +out_unmap:
> + /* uh, no idea what to do here... */
> + return ret;
> +}
> +#else
> +static bool radeon_acpi_vfct_bios(struct radeon_device *rdev)

Make it inline, and then the compiler is smart enough to just delete the
whole if () check you make when you call it.

Third time's a charm?

greg k-h


[PATCH 1/2] drm/radeon: implement ACPI VFCT vbios fetch (v2)

2012-08-16 Thread alexdeucher
From: David L 

This is required for pure UEFI systems.  The vbios is stored
in ACPI rather than at the legacy vga location.

Fixes:
https://bugs.freedesktop.org/show_bug.cgi?id=26891

V2: fix #ifdefs as per Greg's comments

Signed-off-by: Alex Deucher 
Cc: sta...@vger.kernel.org
---
 drivers/gpu/drm/radeon/radeon_bios.c |   62 ++
 1 files changed, 62 insertions(+), 0 deletions(-)

diff --git a/drivers/gpu/drm/radeon/radeon_bios.c 
b/drivers/gpu/drm/radeon/radeon_bios.c
index 501f488..bf21f65 100644
--- a/drivers/gpu/drm/radeon/radeon_bios.c
+++ b/drivers/gpu/drm/radeon/radeon_bios.c
@@ -32,6 +32,9 @@
 
 #include 
 #include 
+#ifdef CONFIG_ACPI
+#include 
+#endif
 /*
  * BIOS.
  */
@@ -476,6 +479,63 @@ static bool radeon_read_disabled_bios(struct radeon_device 
*rdev)
return legacy_read_disabled_bios(rdev);
 }
 
+#ifdef CONFIG_ACPI
+static bool radeon_acpi_vfct_bios(struct radeon_device *rdev)
+{
+   bool ret = false;
+   struct acpi_table_header *hdr;
+   /* acpi_get_table_with_size is not exported :( */
+   acpi_size tbl_size = 0x7fff;
+   UEFI_ACPI_VFCT *vfct;
+   GOP_VBIOS_CONTENT *vbios;
+   VFCT_IMAGE_HEADER *vhdr;
+
+   if (!ACPI_SUCCESS(acpi_get_table("VFCT", 1, &hdr)))
+   return false;
+   if (tbl_size < sizeof(UEFI_ACPI_VFCT)) {
+   DRM_ERROR("ACPI VFCT table present but broken (too short 
#1)\n");
+   goto out_unmap;
+   }
+
+   vfct = (UEFI_ACPI_VFCT *)hdr;
+   if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) > tbl_size) {
+   DRM_ERROR("ACPI VFCT table present but broken (too short 
#2)\n");
+   goto out_unmap;
+   }
+
+   vbios = (GOP_VBIOS_CONTENT *)((char *)hdr + vfct->VBIOSImageOffset);
+   vhdr = &vbios->VbiosHeader;
+   DRM_INFO("ACPI VFCT contains a BIOS for %02x:%02x.%d %04x:%04x, size 
%d\n",
+   vhdr->PCIBus, vhdr->PCIDevice, vhdr->PCIFunction,
+   vhdr->VendorID, vhdr->DeviceID, vhdr->ImageLength);
+
+   if (vhdr->PCIBus != rdev->pdev->bus->number ||
+   vhdr->PCIDevice != PCI_SLOT(rdev->pdev->devfn) ||
+   vhdr->PCIFunction != PCI_FUNC(rdev->pdev->devfn) ||
+   vhdr->VendorID != rdev->pdev->vendor ||
+   vhdr->DeviceID != rdev->pdev->device) {
+   DRM_INFO("ACPI VFCT table is not for this card\n");
+   goto out_unmap;
+   };
+
+   if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) + 
vhdr->ImageLength > tbl_size) {
+   DRM_ERROR("ACPI VFCT image truncated\n");
+   goto out_unmap;
+   }
+
+   rdev->bios = kmemdup(&vbios->VbiosContent, vhdr->ImageLength, 
GFP_KERNEL);
+   ret = !!rdev->bios;
+
+out_unmap:
+   /* uh, no idea what to do here... */
+   return ret;
+}
+#else
+static bool radeon_acpi_vfct_bios(struct radeon_device *rdev)
+{
+   return false;
+}
+#endif
 
 bool radeon_get_bios(struct radeon_device *rdev)
 {
@@ -484,6 +544,8 @@ bool radeon_get_bios(struct radeon_device *rdev)
 
r = radeon_atrm_get_bios(rdev);
if (r == false)
+   r = radeon_acpi_vfct_bios(rdev);
+   if (r == false)
r = igp_read_bios_from_vram(rdev);
if (r == false)
r = radeon_read_bios(rdev);
-- 
1.7.7.5

___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel