From mboxrd@z Thu Jan 1 00:00:00 1970 From: bugzilla-daemon-CC+yJ3UmIYqDUpFQwHEjaQ@public.gmane.org Subject: [Bug 96355] New: Performance: extra&costly SSBO validation even when SSBO aren't used Date: Fri, 03 Jun 2016 07:53:01 +0000 Message-ID: Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============1902649715==" Return-path: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: nouveau-bounces-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org Sender: "Nouveau" To: nouveau-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org List-Id: nouveau.vger.kernel.org --===============1902649715== Content-Type: multipart/alternative; boundary="14649403820.dBE7c.7797"; charset="UTF-8" --14649403820.dBE7c.7797 Date: Fri, 3 Jun 2016 07:53:02 +0000 MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: http://bugs.freedesktop.org/ Auto-Submitted: auto-generated https://bugs.freedesktop.org/show_bug.cgi?id=3D96355 Bug ID: 96355 Summary: Performance: extra&costly SSBO validation even when SSBO aren't used Product: Mesa Version: git Hardware: Other OS: All Status: NEW Severity: normal Priority: medium Component: Drivers/DRI/nouveau Assignee: nouveau-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org Reporter: gregory.hainaut-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org QA Contact: nouveau-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org Hello, I'm currently trying to profile my application (PCSX2) with Mesa. I don't k= now if my GPU (GTX760) is properly reclocked but my app is often CPU limited. It could just be the IO operation that are very slow. Anyway, Perf-Event shows that nvc0_validate_buffers is (too) often called. + 8.78% 7.98% pcsx2_GSReplayL nouveau_dri.so nvc0_validate_buffers My understanding of the code is that every time we switch a shader program,= a full SSBO bind/validation is called. nvc0_set_shader_buffers will dirty buf= fer state (NVC0_NEW_3D_BUFFERS). The trick is that my application doesn't use S= SBO (only UBO). Is it expected to call SSBO validation code when the shader pro= gram doesn't use them? If not, a validation shortcut will be nice. If it can help, here the backtrace from nvc0_set_shader_buffers=20 #0 nvc0_set_shader_buffers (pipe=3D0x87c51e0, shader=3D0, start=3D16, nr= =3D16, buffers=3D0x0) at nvc0/nvc0_state.c:1331 #1 0xf464acc4 in st_bind_ssbos (shader=3D0x8b106bc, shader_type=3D0, st=3D= 0x877ca38, st=3D0x877ca38) at state_tracker/st_atom_storagebuf.c:86 #2 0xf464ad0d in bind_vs_ssbos (st=3D0x877ca38) at state_tracker/st_atom_storagebuf.c:101 #3 0xf4647411 in st_validate_state (st=3D0x877ca38, pipeline=3DST_PIPELINE= _RENDER) at state_tracker/st_atom.c:289 #4 0xf46638ef in st_draw_vbo (ctx=3D0x8801f60, prims=3D0xffffa990, nr_prim= s=3D1, ib=3D0xffffa980, index_bounds_valid=3D0 '\000', min_index=3D4294967295, max_index=3D4294967295, tfb_vertcount=3D0x0,=20 stream=3D0, indirect=3D0x0) at state_tracker/st_draw.c:176 #5 0xf46270f9 in vbo_validated_drawrangeelements (ctx=3Dctx@entry=3D0x8801= f60, mode=3Dmode@entry=3D4, index_bounds_valid=3D0 '\000', start=3D4294967295, end=3D4294967295, count=3D6, type=3D5125,=20 indices=3D0x25258, basevertex=3D19047, numInstances=3D1, baseInstance= =3D0) at vbo/vbo_exec_array.c:849 #6 0xf46274bc in vbo_exec_DrawElementsBaseVertex (mode=3D4, count=3D6, typ= e=3D5125, indices=3D0x25258, basevertex=3D19047) at vbo/vbo_exec_array.c:1007 #7 0xf6ddf422 in shared_dispatch_stub_702 (mode=3D4, count=3D6, type=3D512= 5, indices=3D0x25258, basevertex=3D19047) at shared-glapi/glapi_mapi_tmp.h:212= 35 #8 0xf6362e0a in Draw (this=3D, this=3D, basevertex=3D, mode=3D) Feel free to ask trace/debug info. Best regards --=20 You are receiving this mail because: You are the QA Contact for the bug. You are the assignee for the bug.= --14649403820.dBE7c.7797 Date: Fri, 3 Jun 2016 07:53:02 +0000 MIME-Version: 1.0 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: http://bugs.freedesktop.org/ Auto-Submitted: auto-generated
Bug ID 96355
Summary Performance: extra&costly SSBO validation even when SSBO = aren't used
Product Mesa
Version git
Hardware Other
OS All
Status NEW
Severity normal
Priority medium
Component Drivers/DRI/nouveau
Assignee nouveau@lists.freedesktop.org
Reporter gregory.hainaut@gmail.com
QA Contact nouveau@lists.freedesktop.org 

Hello,

I'm currently trying to profile my application (PCSX2) with Mesa. I don't k=
now
if my GPU (GTX760) is properly reclocked but my app is often CPU limited. It
could just be the IO operation that are very slow.

Anyway, Perf-Event shows that nvc0_validate_buffers is (too) often called.
+    8.78%     7.98%  pcsx2_GSReplayL  nouveau_dri.so  nvc0_validate_buffers

My understanding of the code is that every time we switch a shader program,=
 a
full SSBO bind/validation is called. nvc0_set_shader_buffers will dirty buf=
fer
state (NVC0_NEW_3D_BUFFERS). The trick is that my application doesn't use S=
SBO
(only UBO). Is it expected to call SSBO validation code when the shader pro=
gram
doesn't use them? If not, a validation shortcut will be nice.

If it can help, here the backtrace from nvc0_set_shader_buffers=20

#0  nvc0_set_shader_buffers (pipe=3D0x87c51e0, shader=3D0, start=3D16, nr=
=3D16,
buffers=3D0x0) at nvc0/nvc0_state.c:1331
#1  0xf464acc4 in st_bind_ssbos (shader=3D0x8b106bc, shader_type=3D0, st=3D=
0x877ca38,
st=3D0x877ca38) at state_tracker/st_atom_storagebuf.c:86
#2  0xf464ad0d in bind_vs_ssbos (st=3D0x877ca38) at
state_tracker/st_atom_storagebuf.c:101
#3  0xf4647411 in st_validate_state (st=3D0x877ca38, pipeline=3DST_PIPELINE=
_RENDER)
at state_tracker/st_atom.c:289
#4  0xf46638ef in st_draw_vbo (ctx=3D0x8801f60, prims=3D0xffffa990, nr_prim=
s=3D1,
ib=3D0xffffa980, index_bounds_valid=3D0 '\000', min_index=3D4294967295,
max_index=3D4294967295, tfb_vertcount=3D0x0,=20
    stream=3D0, indirect=3D0x0) at state_tracker/st_draw.c:176
#5  0xf46270f9 in vbo_validated_drawrangeelements (ctx=3Dctx@entry=3D0x=
8801f60,
mode=3Dmode@entry=3D4, index_bounds_valid=3D0 '\000', start=3D429496729=
5,
end=3D4294967295, count=3D6, type=3D5125,=20
    indices=3D0x25258, basevertex=3D19047, numInstances=3D1, baseInstance=
=3D0) at
vbo/vbo_exec_array.c:849
#6  0xf46274bc in vbo_exec_DrawElementsBaseVertex (mode=3D4, count=3D6, typ=
e=3D5125,
indices=3D0x25258, basevertex=3D19047) at vbo/vbo_exec_array.c:1007
#7  0xf6ddf422 in shared_dispatch_stub_702 (mode=3D4, count=3D6, type=3D512=
5,
indices=3D0x25258, basevertex=3D19047) at shared-glapi/glapi_mapi_tmp.h:212=
35
#8  0xf6362e0a in Draw (this=3D<optimized out>, this=3D<optimized =
out>,
basevertex=3D<optimized out>, mode=3D<optimized out>)

Feel free to ask trace/debug info.

Best regards

You are receiving this mail because:
  • You are the QA Contact for the bug.
  • You are the assignee for the bug.
= --14649403820.dBE7c.7797-- --===============1902649715== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: base64 Content-Disposition: inline X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KTm91dmVhdSBt YWlsaW5nIGxpc3QKTm91dmVhdUBsaXN0cy5mcmVlZGVza3RvcC5vcmcKaHR0cHM6Ly9saXN0cy5m cmVlZGVza3RvcC5vcmcvbWFpbG1hbi9saXN0aW5mby9ub3V2ZWF1Cg== --===============1902649715==--