All of lore.kernel.org
 help / color / mirror / Atom feed
From: Samuel Pitoiset <samuel.pitoiset-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
To: Ilia Mirkin <imirkin-FrUbXkNCsVf2fBVCVOL8/A@public.gmane.org>
Cc: "mesa-dev-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org"
	<mesa-dev-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org>,
	"11.0"
	<mesa-stable-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org>,
	"nouveau-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org"
	<nouveau-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org>
Subject: Re: [Mesa-dev] [PATCH] nv50: avoid using inline vertex data submit when gl_VertexID is used
Date: Mon, 24 Aug 2015 22:12:41 +0200	[thread overview]
Message-ID: <55DB7AB9.5080600@gmail.com> (raw)
In-Reply-To: <CAKb7Uvi-29ZZxo57Y1iWxf3A0U98EN4EroGvQ55L4PBqRXPF8w-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>



On 08/24/2015 10:02 PM, Ilia Mirkin wrote:
> Edge flag stuff is annoying. Pretty sure only blender uses it. shade
> model = flat should get fixed on nv50 before edge flags, since blender
> uses that too, and it's produces much worse visual artifacts.

No rush for this one though.

>
> I'm having second thoughts about this patch. I think I'm going to go
> back to my previous approach of just calling
> nv50_vertex_arrays_validate when vbo_fifo && vertexid. I suspect that
> vertexid usage with small draws from client buffers is next to
> inexistent, no need to re-emit this stuff so often.

Good, I'd be happy to have a look at this second approach.

>
> On Mon, Aug 24, 2015 at 4:07 PM, Samuel Pitoiset
> <samuel.pitoiset@gmail.com> wrote:
>> Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
>>
>> This fix is simpler than I was expected. What about the edge flag stuff now?
>> :)
>>
>>
>> On 08/24/2015 05:51 PM, Ilia Mirkin wrote:
>>> The hardware only generates vertexid when vertices come from a VBO. This
>>> fixes:
>>>
>>>     vertexid-drawelements
>>>     vertexid-drawarrays
>>>
>>> Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
>>> Cc: "11.0" <mesa-stable@lists.freedesktop.org>
>>> ---
>>>    src/gallium/drivers/nouveau/nv50/nv50_program.c        | 1 +
>>>    src/gallium/drivers/nouveau/nv50/nv50_program.h        | 1 +
>>>    src/gallium/drivers/nouveau/nv50/nv50_state_validate.c | 3 ++-
>>>    src/gallium/drivers/nouveau/nv50/nv50_vbo.c            | 8 ++++++++
>>>    4 files changed, 12 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/src/gallium/drivers/nouveau/nv50/nv50_program.c
>>> b/src/gallium/drivers/nouveau/nv50/nv50_program.c
>>> index 02dc367..eff4477 100644
>>> --- a/src/gallium/drivers/nouveau/nv50/nv50_program.c
>>> +++ b/src/gallium/drivers/nouveau/nv50/nv50_program.c
>>> @@ -66,6 +66,7 @@ nv50_vertprog_assign_slots(struct nv50_ir_prog_info
>>> *info)
>>>          case TGSI_SEMANTIC_VERTEXID:
>>>             prog->vp.attrs[2] |= NV50_3D_VP_GP_BUILTIN_ATTR_EN_VERTEX_ID;
>>>             prog->vp.attrs[2] |=
>>> NV50_3D_VP_GP_BUILTIN_ATTR_EN_VERTEX_ID_DRAW_ARRAYS_ADD_START;
>>> +         prog->vp.vertexid = 1;
>>>             continue;
>>>          default:
>>>             break;
>>> diff --git a/src/gallium/drivers/nouveau/nv50/nv50_program.h
>>> b/src/gallium/drivers/nouveau/nv50/nv50_program.h
>>> index 5d3ff56..f4e8e94 100644
>>> --- a/src/gallium/drivers/nouveau/nv50/nv50_program.h
>>> +++ b/src/gallium/drivers/nouveau/nv50/nv50_program.h
>>> @@ -76,6 +76,7 @@ struct nv50_program {
>>>          ubyte psiz;        /* output slot of point size */
>>>          ubyte bfc[2];      /* indices into varying for FFC (FP) or BFC
>>> (VP) */
>>>          ubyte edgeflag;
>>> +      ubyte vertexid;
>>>          ubyte clpd[2];     /* output slot of clip distance[i]'s 1st
>>> component */
>>>          ubyte clpd_nr;
>>>       } vp;
>>> diff --git a/src/gallium/drivers/nouveau/nv50/nv50_state_validate.c
>>> b/src/gallium/drivers/nouveau/nv50/nv50_state_validate.c
>>> index b304a17..66dcf43 100644
>>> --- a/src/gallium/drivers/nouveau/nv50/nv50_state_validate.c
>>> +++ b/src/gallium/drivers/nouveau/nv50/nv50_state_validate.c
>>> @@ -503,7 +503,8 @@ static struct state_validate {
>>>        { nv50_validate_samplers,      NV50_NEW_SAMPLERS },
>>>        { nv50_stream_output_validate, NV50_NEW_STRMOUT |
>>>                                       NV50_NEW_VERTPROG | NV50_NEW_GMTYPROG
>>> },
>>> -    { nv50_vertex_arrays_validate, NV50_NEW_VERTEX | NV50_NEW_ARRAYS },
>>> +    { nv50_vertex_arrays_validate, NV50_NEW_VERTEX | NV50_NEW_ARRAYS |
>>> +                                   NV50_NEW_VERTPROG },
>>>        { nv50_validate_min_samples,   NV50_NEW_MIN_SAMPLES },
>>>    };
>>>    #define validate_list_len (sizeof(validate_list) /
>>> sizeof(validate_list[0]))
>>> diff --git a/src/gallium/drivers/nouveau/nv50/nv50_vbo.c
>>> b/src/gallium/drivers/nouveau/nv50/nv50_vbo.c
>>> index 600b973..fb4305f 100644
>>> --- a/src/gallium/drivers/nouveau/nv50/nv50_vbo.c
>>> +++ b/src/gallium/drivers/nouveau/nv50/nv50_vbo.c
>>> @@ -301,6 +301,14 @@ nv50_vertex_arrays_validate(struct nv50_context
>>> *nv50)
>>>       unsigned i;
>>>       const unsigned n = MAX2(vertex->num_elements,
>>> nv50->state.num_vtxelts);
>>>    +   /* A vertexid is not generated for inline data uploads. Have to use
>>> a
>>> +    * VBO. This check must come after the vertprog has been validated,
>>> +    * otherwise vertexid may be unset.
>>> +    */
>>> +   assert(nv50->vertprog->translated);
>>> +   if (nv50->vertprog->vp.vertexid)
>>> +      nv50->vbo_push_hint = 0;
>>> +
>>>       if (unlikely(vertex->need_conversion))
>>>          nv50->vbo_fifo = ~0;
>>>       else
>>

_______________________________________________
Nouveau mailing list
Nouveau@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/nouveau

      parent reply	other threads:[~2015-08-24 20:12 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-08-24 15:51 [PATCH] nv50: avoid using inline vertex data submit when gl_VertexID is used Ilia Mirkin
     [not found] ` <1440431513-29568-1-git-send-email-imirkin-FrUbXkNCsVf2fBVCVOL8/A@public.gmane.org>
2015-08-24 15:57   ` Tobias Klausmann
2015-08-24 16:19     ` [Nouveau] " Ilia Mirkin
2015-08-24 20:07   ` [Mesa-dev] " Samuel Pitoiset
     [not found]     ` <55DB7974.5090403-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2015-08-24 20:02       ` Ilia Mirkin
     [not found]         ` <CAKb7Uvi-29ZZxo57Y1iWxf3A0U98EN4EroGvQ55L4PBqRXPF8w-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2015-08-24 20:12           ` Samuel Pitoiset [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=55DB7AB9.5080600@gmail.com \
    --to=samuel.pitoiset-re5jqeeqqe8avxtiumwx3w@public.gmane.org \
    --cc=imirkin-FrUbXkNCsVf2fBVCVOL8/A@public.gmane.org \
    --cc=mesa-dev-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org \
    --cc=mesa-stable-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org \
    --cc=nouveau-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.