Sui Jingfeng
2023-Sep-04 19:57 UTC
[Nouveau] [RFC, drm-misc-next v4 3/9] drm/radeon: Implement .be_primary() callback
From: Sui Jingfeng <suijingfeng at loongson.cn> On a machine with multiple GPUs, a Linux user has no control over which one is primary at boot time. This patch tries to solve the mentioned problem by implementing the .be_primary() callback. Pass radeon.modeset=10 on the kernel cmd line if you really want the device bound by radeon to be the primary video adapter, no matter what VGAARB say. Cc: Alex Deucher <alexander.deucher at amd.com> Cc: Christian Koenig <christian.koenig at amd.com> Signed-off-by: Sui Jingfeng <suijingfeng at loongson.cn> --- drivers/gpu/drm/radeon/radeon_device.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index 71f2ff39d6a1..b661cd3a8dc2 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -1263,6 +1263,14 @@ static const struct vga_switcheroo_client_ops radeon_switcheroo_ops = { .can_switch = radeon_switcheroo_can_switch, }; +static bool radeon_want_to_be_primary(struct pci_dev *pdev) +{ + if (radeon_modeset == 10) + return true; + + return false; +} + /** * radeon_device_init - initialize the driver * @@ -1425,7 +1433,7 @@ int radeon_device_init(struct radeon_device *rdev, /* if we have > 1 VGA cards, then disable the radeon VGA resources */ /* this will fail for cards that aren't VGA class devices, just * ignore it */ - vga_client_register(rdev->pdev, radeon_vga_set_decode, NULL); + vga_client_register(rdev->pdev, radeon_vga_set_decode, radeon_want_to_be_primary); if (rdev->flags & RADEON_IS_PX) runtime = true; -- 2.34.1
Christian König
2023-Sep-05 05:50 UTC
[Nouveau] [RFC, drm-misc-next v4 3/9] drm/radeon: Implement .be_primary() callback
Am 04.09.23 um 21:57 schrieb Sui Jingfeng:> From: Sui Jingfeng <suijingfeng at loongson.cn> > > On a machine with multiple GPUs, a Linux user has no control over which one > is primary at boot time.Question is why is that useful? Should we give users the ability to control that? I don't see an use case for this. Regards, Christian.> This patch tries to solve the mentioned problem by > implementing the .be_primary() callback. Pass radeon.modeset=10 on the > kernel cmd line if you really want the device bound by radeon to be the > primary video adapter, no matter what VGAARB say. > > Cc: Alex Deucher <alexander.deucher at amd.com> > Cc: Christian Koenig <christian.koenig at amd.com> > Signed-off-by: Sui Jingfeng <suijingfeng at loongson.cn> > --- > drivers/gpu/drm/radeon/radeon_device.c | 10 +++++++++- > 1 file changed, 9 insertions(+), 1 deletion(-) > > diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c > index 71f2ff39d6a1..b661cd3a8dc2 100644 > --- a/drivers/gpu/drm/radeon/radeon_device.c > +++ b/drivers/gpu/drm/radeon/radeon_device.c > @@ -1263,6 +1263,14 @@ static const struct vga_switcheroo_client_ops radeon_switcheroo_ops = { > .can_switch = radeon_switcheroo_can_switch, > }; > > +static bool radeon_want_to_be_primary(struct pci_dev *pdev) > +{ > + if (radeon_modeset == 10) > + return true; > + > + return false; > +} > + > /** > * radeon_device_init - initialize the driver > * > @@ -1425,7 +1433,7 @@ int radeon_device_init(struct radeon_device *rdev, > /* if we have > 1 VGA cards, then disable the radeon VGA resources */ > /* this will fail for cards that aren't VGA class devices, just > * ignore it */ > - vga_client_register(rdev->pdev, radeon_vga_set_decode, NULL); > + vga_client_register(rdev->pdev, radeon_vga_set_decode, radeon_want_to_be_primary); > > if (rdev->flags & RADEON_IS_PX) > runtime = true;