Hi Matt,
Starting with commit d877c643b ("glsl: Use M_PI_* macros."), I'm
seeing a few piglit failures on nvcf (GF116) for
generated_tests/spec/glsl-1.10/execution/built-in-functions/fs-atan-vec2-vec2.shader_test
and its vs/gs counterparts. Curiously it works fine on nv50-family
chips. I checked the generated shaders, and the only differences are
the constants[1]. Does the generated test need to be adjusted to
expect more (or less) precise values perhaps? Or is it likely that
there's some bug in the shader that was being masked by the less
correct constants?
I haven't actually looked at the atan formula -- could it be that
those constants were "off" on purpose? Any other ideas?
Thanks,
-ilia
[1] : shader diff. scalar, intel-style syntax.
--- passing 2014-04-19 23:14:25.000000000 -0400
+++ failing 2014-04-19 23:01:10.000000000 -0400
@@ -21,7 +21,7 @@
000000a0: bc00dde2 1af3f8c6 mov b32 $r3 0xbcfe31af
000000a8: 94011de2 18f6c527 mov b32 $r4 0x3db149e5
000000b0: 0c20dc00 30080000 fma rn f32 $r3 $r2 $r3 $r4
-000000b8: 54011de2 1af96f02 mov b32 $r4 0xbe5bc095
+000000b8: 50011de2 1af96f02 mov b32 $r4 0xbe5bc094
000000c0: 0c20dc00 30080000 fma rn f32 $r3 $r2 $r3 $r4
000000c8: 6c011de2 18ff243f mov b32 $r4 0x3fc90fdb
000000d0: 0c209c00 30080000 fma rn f32 $r2 $r2 $r3 $r4
@@ -32,15 +32,15 @@
000000f8: 43f1dc00 220e4000 set $p0 0x1 gt f32 0x0 c0[0x10]
00000100: 000021f4 40000000 join (not $p0) nop
00000108: 03f1dc00 218e4000 set $p0 0x1 le f32 0x0 c0[0x0]
-00000110: 70000002 2901243f $p0 add f32 $r0 $r0 0x40490fdc
-00000118: 70002002 2b01243f (not $p0) add f32 $r0 $r0 0xc0490fdc
+00000110: 6c000002 2901243f $p0 add f32 $r0 $r0 0x40490fdb
+00000118: 6c002002 2b01243f (not $p0) add f32 $r0 $r0 0xc0490fdb
00000120: 00001df4 40000000 join nop
00000128: 00001df4 40000000 B join nop
00000130: 00001de4 28004000 B mov b32 $r0 c0[0x0]
00000138: fc005c20 120e0000 set f32 $r1 gt f32 $r0 0x0
00000140: fc001c20 108e0000 set f32 $r0 lt f32 $r0 0x0
00000148: 00101d00 50000000 add rn f32 $r0 $r1 neg $r0
-00000150: 70001c12 30ff243f join mul f32 $r0 $r0 0x3fc90fdc
+00000150: 6c001c12 30ff243f join mul f32 $r0 $r0 0x3fc90fdb
00000158: 11205c44 10004000 B cvt f32 $r1 abs f32 c0[0x4]
00000160: dc105c02 30c8af31 mul f32 $r1 $r1 0x322bcc77
00000168: 00000007 60000005 joinat 0x2b0
@@ -64,7 +64,7 @@
000001f8: bc011de2 1af3f8c6 mov b32 $r4 0xbcfe31af
00000200: 94015de2 18f6c527 mov b32 $r5 0x3db149e5
00000208: 10311c00 300a0000 fma rn f32 $r4 $r3 $r4 $r5
-00000210: 54015de2 1af96f02 mov b32 $r5 0xbe5bc095
+00000210: 50015de2 1af96f02 mov b32 $r5 0xbe5bc094
00000218: 10311c00 300a0000 fma rn f32 $r4 $r3 $r4 $r5
00000220: 6c015de2 18ff243f mov b32 $r5 0x3fc90fdb
00000228: 1030dc00 300a0000 fma rn f32 $r3 $r3 $r4 $r5
@@ -75,15 +75,15 @@
00000250: 53f1dc00 220e4000 set $p0 0x1 gt f32 0x0 c0[0x14]
00000258: 000021f4 40000000 join (not $p0) nop
00000260: 13f1dc00 218e4000 set $p0 0x1 le f32 0x0 c0[0x4]
-00000268: 70104002 2901243f $p0 add f32 $r1 $r1 0x40490fdc
-00000270: 70106002 2b01243f (not $p0) add f32 $r1 $r1 0xc0490fdc
+00000268: 6c104002 2901243f $p0 add f32 $r1 $r1 0x40490fdb
+00000270: 6c106002 2b01243f (not $p0) add f32 $r1 $r1 0xc0490fdb
00000278: 00001df4 40000000 join nop
00000280: 00001df4 40000000 B join nop
00000288: 10005de4 28004000 B mov b32 $r1 c0[0x4]
00000290: fc109c20 120e0000 set f32 $r2 gt f32 $r1 0x0
00000298: fc105c20 108e0000 set f32 $r1 lt f32 $r1 0x0
000002a0: 04205d00 50000000 add rn f32 $r1 $r2 neg $r1
-000002a8: 70105c12 30ff243f join mul f32 $r1 $r1 0x3fc90fdc
+000002a8: 6c105c12 30ff243f join mul f32 $r1 $r1 0x3fc90fdb
000002b0: c0001d00 50004000 B add rn f32 $r0 $r0 neg c0[0x30]
000002b8: d0105d00 50004000 add rn f32 $r1 $r1 neg c0[0x34]
000002c0: 00001c00 58000000 mul rn f32 $r0 $r0 $r0
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
