[PATCH libdrm v2 5/5] intel: get gen once for gen >= 9

Lucas De Marchi lucas.demarchi at intel.com
Wed Sep 5 18:01:47 UTC 2018


On Wed, Aug 29, 2018 at 03:31:11PM +0100, Chris Wilson wrote:
> Quoting Lucas De Marchi (2018-08-29 01:35:32)
> > We don't need to call IS_GEN() for each gen >= 9: we can rather use the
> > new intel_is_genx() helper to iterate the pciids array once.
> > 
> > Signed-off-by: Lucas De Marchi <lucas.demarchi at intel.com>
> > ---
> >  intel/intel_bufmgr_gem.c | 8 +-------
> >  intel/intel_decode.c     | 8 ++------
> >  2 files changed, 3 insertions(+), 13 deletions(-)
> > 
> > diff --git a/intel/intel_bufmgr_gem.c b/intel/intel_bufmgr_gem.c
> > index 8c3a4b20..d6587b76 100644
> > --- a/intel/intel_bufmgr_gem.c
> > +++ b/intel/intel_bufmgr_gem.c
> > @@ -3656,13 +3656,7 @@ drm_intel_bufmgr_gem_init(int fd, int batch_size)
> >                 bufmgr_gem->gen = 7;
> >         else if (IS_GEN8(bufmgr_gem->pci_device))
> >                 bufmgr_gem->gen = 8;
> > -       else if (IS_GEN9(bufmgr_gem->pci_device))
> > -               bufmgr_gem->gen = 9;
> > -       else if (IS_GEN10(bufmgr_gem->pci_device))
> > -               bufmgr_gem->gen = 10;
> > -       else if (IS_GEN11(bufmgr_gem->pci_device))
> > -               bufmgr_gem->gen = 11;
> > -       else {
> > +       else if (!intel_get_genx(bufmgr_gem->pci_device, &bufmgr_gem->gen)) {
> >                 free(bufmgr_gem);
> >                 bufmgr_gem = NULL;
> >                 goto exit;
> 
> And while you are here, don't exit for an unknown gen, just pretend it's
> a future one. i915_pciids.h should contain *all* ids, even reserved, for
> exactly this reason. (The same behaviour is relied on elsewhere so that
> we don't get caught out by some one retrospectively introducing a new
> chip. Don't even get started on the abuse of pci-id for sub-gen
> encoding...)

This can be done as a separate series, otherwise we will just start
shoving a lot of things on top of this series and never get it merged.

Lucas De Marchi


More information about the dri-devel mailing list