[PATCH 1/2] x86: queensbay: Optimize disable_igd() a little

[PATCH 1/2] x86: queensbay: Optimize disable_igd() a little

[Bin Meng]

Initialize igd and sdvo to NULL so that we don't need to test the
return value of dm_pci_bus_find_bdf() later.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

 arch/x86/cpu/queensbay/tnc.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/arch/x86/cpu/queensbay/tnc.c b/arch/x86/cpu/queensbay/tnc.c
index 782ed863fe..e38c0198e4 100644
--- a/arch/x86/cpu/queensbay/tnc.c
+++ b/arch/x86/cpu/queensbay/tnc.c
@@ -18,18 +18,15 @@
 
 static int __maybe_unused disable_igd(void)
 {
-	struct udevice *igd, *sdvo;
+	struct udevice *igd = NULL;
+	struct udevice *sdvo = NULL;
 	int ret;
 
 	ret = dm_pci_bus_find_bdf(TNC_IGD, &igd);
-	if (ret)
-		return ret;
 	if (!igd)
 		return 0;
 
 	ret = dm_pci_bus_find_bdf(TNC_SDVO, &sdvo);
-	if (ret)
-		return ret;
 	if (!sdvo)
 		return 0;
 
-- 
2.25.1

[PATCH 2/2] x86: crownbay: Use external graphics card by default

[Bin Meng]

The board routes the Integrated Graphics Device (IGD) to an LVDS
panel, which is less popular than a PCIe based graphics card.

Disable the IGD so that it does not show up in the PCI configuration
space as a VGA display controller, so we can use an external PCIe
graphics card with whatever cable we have.

Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
---

 configs/crownbay_defconfig | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/configs/crownbay_defconfig b/configs/crownbay_defconfig
index 1208aad42f..d7ee0fe45e 100644
--- a/configs/crownbay_defconfig
+++ b/configs/crownbay_defconfig
@@ -8,9 +8,8 @@ CONFIG_MAX_CPUS=2
 CONFIG_DEFAULT_DEVICE_TREE="crownbay"
 CONFIG_VENDOR_INTEL=y
 CONFIG_TARGET_CROWNBAY=y
+CONFIG_DISABLE_IGD=y
 CONFIG_SMP=y
-CONFIG_HAVE_VGA_BIOS=y
-CONFIG_VGA_BIOS_ADDR=0xfffa0000
 CONFIG_GENERATE_PIRQ_TABLE=y
 CONFIG_GENERATE_MP_TABLE=y
 CONFIG_FIT=y
-- 
2.25.1

Re: [PATCH 2/2] x86: crownbay: Use external graphics card by default

[Simon Glass]

On Sat, 31 Jul 2021 at 07:17, Bin Meng <bmeng.cn@gmail.com> wrote:
>
> The board routes the Integrated Graphics Device (IGD) to an LVDS
> panel, which is less popular than a PCIe based graphics card.
>
> Disable the IGD so that it does not show up in the PCI configuration
> space as a VGA display controller, so we can use an external PCIe
> graphics card with whatever cable we have.
>
> diff --git a/configs/crownbay_defconfig b/configs/crownbay_defconfig
> index 1208aad42f..d7ee0fe45e 100644
> --- a/configs/crownbay_defconfig
> +++ b/configs/crownbay_defconfig
> @@ -8,9 +8,8 @@ CONFIG_MAX_CPUS=2
>  CONFIG_DEFAULT_DEVICE_TREE="crownbay"
>  CONFIG_VENDOR_INTEL=y
>  CONFIG_TARGET_CROWNBAY=y
> +CONFIG_DISABLE_IGD=y
>  CONFIG_SMP=y
> -CONFIG_HAVE_VGA_BIOS=y
> -CONFIG_VGA_BIOS_ADDR=0xfffa0000
>  CONFIG_GENERATE_PIRQ_TABLE=y
>  CONFIG_GENERATE_MP_TABLE=y
>  CONFIG_FIT=y
> --
> 2.25.1
>
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
>  configs/crownbay_defconfig | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)

Reviewed-by: Simon Glass <sjg@chromium.org>

Re: [PATCH 1/2] x86: queensbay: Optimize disable_igd() a little

[Simon Glass]

Hi Bin,

On Sat, 31 Jul 2021 at 07:17, Bin Meng <bmeng.cn@gmail.com> wrote:
>
> Initialize igd and sdvo to NULL so that we don't need to test the
> return value of dm_pci_bus_find_bdf() later.
>
> Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> ---
>
>  arch/x86/cpu/queensbay/tnc.c | 7 ++-----
>  1 file changed, 2 insertions(+), 5 deletions(-)
>
> diff --git a/arch/x86/cpu/queensbay/tnc.c b/arch/x86/cpu/queensbay/tnc.c
> index 782ed863fe..e38c0198e4 100644
> --- a/arch/x86/cpu/queensbay/tnc.c
> +++ b/arch/x86/cpu/queensbay/tnc.c
> @@ -18,18 +18,15 @@
>
>  static int __maybe_unused disable_igd(void)
>  {
> -       struct udevice *igd, *sdvo;
> +       struct udevice *igd = NULL;
> +       struct udevice *sdvo = NULL;
>         int ret;
>
>         ret = dm_pci_bus_find_bdf(TNC_IGD, &igd);
> -       if (ret)
> -               return ret;
>         if (!igd)
>                 return 0;

This is backwards. If ret returns 0 then we know igd is non-NULL. So
check the return value. That is what it is for. You can drop the check
of igd. Same below.

>
>         ret = dm_pci_bus_find_bdf(TNC_SDVO, &sdvo);
> -       if (ret)
> -               return ret;
>         if (!sdvo)
>                 return 0;
>
> --
> 2.25.1
>

Regards,
Simon

Re: [PATCH 1/2] x86: queensbay: Optimize disable_igd() a little

[Bin Meng]

Hi Simon,

On Mon, Aug 2, 2021 at 3:19 AM Simon Glass <sjg@chromium.org> wrote:
>
> Hi Bin,
>
> On Sat, 31 Jul 2021 at 07:17, Bin Meng <bmeng.cn@gmail.com> wrote:
> >
> > Initialize igd and sdvo to NULL so that we don't need to test the
> > return value of dm_pci_bus_find_bdf() later.
> >
> > Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
> > ---
> >
> >  arch/x86/cpu/queensbay/tnc.c | 7 ++-----
> >  1 file changed, 2 insertions(+), 5 deletions(-)
> >
> > diff --git a/arch/x86/cpu/queensbay/tnc.c b/arch/x86/cpu/queensbay/tnc.c
> > index 782ed863fe..e38c0198e4 100644
> > --- a/arch/x86/cpu/queensbay/tnc.c
> > +++ b/arch/x86/cpu/queensbay/tnc.c
> > @@ -18,18 +18,15 @@
> >
> >  static int __maybe_unused disable_igd(void)
> >  {
> > -       struct udevice *igd, *sdvo;
> > +       struct udevice *igd = NULL;
> > +       struct udevice *sdvo = NULL;
> >         int ret;
> >
> >         ret = dm_pci_bus_find_bdf(TNC_IGD, &igd);
> > -       if (ret)
> > -               return ret;
> >         if (!igd)
> >                 return 0;
>
> This is backwards. If ret returns 0 then we know igd is non-NULL. So
> check the return value. That is what it is for. You can drop the check
> of igd. Same below.
>

Ah, yes. I remembered why the original codes were written like this.
Actually the codes tried to be compatible that if the device is
already disabled it should not return error. I will reword the commit
message and add some comments here.

Regards,
Bin