[Intel-xe] [PATCH v2 05/11] drm/xe: Remove dependency on intel_gt_regs.h

[Vivi, Rodrigo]

On Fri, 2023-02-24 at 12:25 -0800, Lucas De Marchi wrote:
> On Fri, Feb 24, 2023 at 07:47:04PM +0000, Rodrigo Vivi wrote:
> > On Fri, 2023-02-24 at 10:34 -0800, Matt Roper wrote:
> > > On Fri, Feb 24, 2023 at 10:29:14AM -0800, Lucas De Marchi wrote:
> > > > On Fri, Feb 24, 2023 at 10:06:14AM -0800, Matt Roper wrote:
> > > > > On Thu, Feb 16, 2023 at 04:52:20PM -0800, Lucas De Marchi
> > > > > wrote:
> > > > > > Create regs/xe_gt_regs.h file with all the registers and
> > > > > > bit
> > > > > > definitions used by the xe driver. Eventually the registers
> > > > > > may
> > > > > > be
> > > > > > defined in a different way and since xe doesn't supported
> > > > > > below
> > > > > > gen12,
> > > > > > the number of registers touched is much smaller, so create
> > > > > > a
> > > > > > new header.
> > > > > > 
> > > > > > The definitions themselves are direct copy from the
> > > > > > gt/intel_gt_regs.h file, just sorting the registers by
> > > > > > address.
> > > > > > Cleaning those up and adhering to a common coding style is
> > > > > > left
> > > > > > for
> > > > > > later.
> > > > > > 
> > > > > > Signed-off-by: Lucas De Marchi <lucas.demarchi at intel.com>
> > > > > > ---
> > > > > >  drivers/gpu/drm/i915/i915_reg_defs.h  |   2 +
> > > > > >  drivers/gpu/drm/xe/regs/xe_gt_regs.h  | 287
> > > > > > ++++++++++++++++++++++++++
> > > > > >  drivers/gpu/drm/xe/xe_execlist.c      |   2 +-
> > > > > >  drivers/gpu/drm/xe/xe_force_wake.c    |   3 +-
> > > > > >  drivers/gpu/drm/xe/xe_ggtt.c          |   2 +-
> > > > > >  drivers/gpu/drm/xe/xe_gt.c            |   3 +-
> > > > > >  drivers/gpu/drm/xe/xe_gt_clock.c      |   2 +-
> > > > > >  drivers/gpu/drm/xe/xe_gt_mcr.c        |   3 +-
> > > > > >  drivers/gpu/drm/xe/xe_guc.c           |   2 +-
> > > > > >  drivers/gpu/drm/xe/xe_guc_ads.c       |   4 +-
> > > > > >  drivers/gpu/drm/xe/xe_guc_pc.c        |   2 +-
> > > > > >  drivers/gpu/drm/xe/xe_hw_engine.c     |   2 +-
> > > > > >  drivers/gpu/drm/xe/xe_irq.c           |   2 +-
> > > > > >  drivers/gpu/drm/xe/xe_lrc.c           |   2 +-
> > > > > >  drivers/gpu/drm/xe/xe_mmio.c          |   2 +-
> > > > > >  drivers/gpu/drm/xe/xe_mocs.c          |   3 +-
> > > > > >  drivers/gpu/drm/xe/xe_reg_sr.c        |   3 +-
> > > > > >  drivers/gpu/drm/xe/xe_reg_whitelist.c |   3 +-
> > > > > >  drivers/gpu/drm/xe/xe_ring_ops.c      |   2 +-
> > > > > >  drivers/gpu/drm/xe/xe_tuning.c        |   3 +-
> > > > > >  drivers/gpu/drm/xe/xe_wa.c            |   2 +-
> > > > > >  21 files changed, 308 insertions(+), 28 deletions(-)
> > > > > >  create mode 100644 drivers/gpu/drm/xe/regs/xe_gt_regs.h
> > > > > > 
> > > > > > diff --git a/drivers/gpu/drm/i915/i915_reg_defs.h
> > > > > > b/drivers/gpu/drm/i915/i915_reg_defs.h
> > > > > > index b5e06b6a9478..a6080c983a77 100644
> > > > > > --- a/drivers/gpu/drm/i915/i915_reg_defs.h
> > > > > > +++ b/drivers/gpu/drm/i915/i915_reg_defs.h
> > > > > > @@ -136,6 +136,8 @@ typedef struct {
> > > > > >         u32 reg;
> > > > > >  } i915_mcr_reg_t;
> > > > > > 
> > > > > > +#define MCR_REG(offset)        ((const i915_mcr_reg_t){
> > > > > > .reg =
> > > > > > (offset) })
> > > > > 
> > > > > Should there be a corresponding removal of this definition
> > > > > from
> > > > > drivers/gpu/drm/i915/gt/intel_gt_regs.h ?
> > > > > 
> > > > > Aside from that,
> > > > > 
> > > > > Reviewed-by: Matt Roper <matthew.d.roper at intel.com>
> > > > 
> > > > 
> > > > good catch. I was trying not to change the i915 headers, but
> > > > this
> > > > one went unoticed while fixing the build.
> > > > 
> > > > I wonder if this should be a separate patch to merge in
> > > > drm-intel-next by itself. It's harmless though and we could
> > > > keep it
> > > > in the xe branch, as long as also removing from
> > > > drivers/gpu/drm/i915/gt/intel_gt_regs.h
> > > > 
> > > > Rodrigo / Matt, are you ok with this approach?
> > 
> > yeap, anything touching outside xe needs a separated patch anyway.
> > and I believe it can already be sent do intel-gfx, right?!
> 
> the problem is to depend on that patch. If we are rebasing the
> drm-xe-next branch with a non-ff. Then we could simply have the patch
> in both places and handle the conflict. If we are not rebasing, then
> is it ok to have the patch in both places and deal with a possible
> conflict in a merge (or backmerge if we are not yet in drm-next)?

we will have to rebase, so let's add the patch in both places and
deal with the conflicts.

> 
> Lucas De Marchi