ospModelViewer + dragon.xml = no pixels about ospray HOT 6 CLOSED

Hey, Tim,

Can you tell us what branch (or git hash) you have been using?
Also, please run with "--osp:debug", and let us know the output, as well
as what architecture you have been using.

I just updated to current sources (in devel tree), and rebuilt, and it
works fine on my side ... and the output it makes are actually exactly
the right ones, so I'm a bit confused ... !?

Ingo

On 5/18/15 3:33 PM, timsandstrom wrote:

Look like data is loading correctly, but nothing being drawn. No CPU
activity at all.
This was pulled/built from sources today.

(tao:~/local/ospray-data/dragon) >
/u/wk/tsandstr/local/ospray/build/ospModelViewer dragon.xml
#ospModelViewer: starting to process cmdline arguments
#ospModelViewer: done parsing. found model with

• num meshes : 1 [922849]
• num instances: 1 [922849]
• num unique triangles : 922849
• num instanced triangles: 922849 #ospModelViewer: number of
  instances matches number of meshes, creating single model that
  contains all meshes #ospModelViewer: adding parsed geometries to
  ospray model #ospModelViewer: committing model #osp/trimesh:
  Verifying index buffer ... #osp/trimesh: Verifying vertex buffer
  ... #ospModelViewer: done creating ospray model. #ospModelViewer:
  Adding a hard coded directional light as the sun. #ospModelViewer:
  done creating window. Press 'Q' to quit. #ospModelViewer: set
  world bounds [(-6.83587, -0.431891, -6.93969); (7.11329, 7.60655,
  5.10773)], motion speed 0.0201081

—
Reply to this email directly or view it on GitHub
#17.

from ospray.

Arg, new to git, not sure how to get that info.

Lemme try... well, doesn't look like v0.8.1 does it? [i.e. I probably messed up] :(

If yours work then it's probably user-error on my part.

(tao:~/local/ospray) > git rev-parse --verify HEAD
8723ce6

(tao:~/local/ospray) > git log -n 1
commit 8723ce6
Merge: e7cb749 f639788
Author: Gregory P. Johnson [email protected]
Date: Fri May 8 11:56:13 2015 -0500

 Merge branch 'release-0.8'

ospModelViewer --osp:debug dragon.xml

Embree Ray Tracing Kernels 2.3.3 (May 18 2015)
Compiler : GCC 4.9.2
Platform : Linux (64bit)
CPU : SSE SSE2 SSE3 SSSE3 SSE41 SSE42 POPCNT AVX F16C RDRAND
Features : intersection_filter bufferstride
general:
build threads = 1
verbosity = 2
triangles:
accel = default
builder = default
traverser = default
replications = 2
motion blur triangles:
accel = default
builder = default
traverser = default
hair:
accel = default
builder = default
traverser = default
replications = 3
subdivision surfaces:
accel = default
#ospModelViewer: starting to process cmdline arguments
#ospModelViewer: done parsing. found model with

• num meshes : 1 [922849]
• num instances: 1 [922849]
• num unique triangles : 922849
• num instanced triangles: 922849
  #ospModelViewer: number of instances matches number of meshes, creating single model that contains all meshes
  #ospModelViewer: adding parsed geometries to ospray model
  #ospModelViewer: committing model
  #osp/trimesh: Verifying index buffer ...
  #osp/trimesh: Verifying vertex buffer ...
  building BVH4 with sse2::BVH4Builder() ... [DONE]
  dt = 7068.84ms, perf = 0.130552 Mprim/s
  primitives = 922849, vertices = 0, hash= 26839253013198374
  sah = 1.0000, depth = 18
  used = 68.3 MB, perPrimitive = 74.0 B
  alignedNodes = 79799 (89.0% filled) (10.2 MB) (15.0% of total)
  leaves = 204397 (58.1 MB) (85.0% of total)
  vertices = 0 (0.0 MB) (0.0% of total) (75.0% used)
  created scene intersector
  accels[0]
  intersector1 = sse41::BVH4Triangle4Intersector1Moeller
  intersector4 = sse42::BVH4Triangle4Intersector4HybridMoellerNoFilter
  selected scene intersector
  intersector1 = sse41::BVH4Triangle4Intersector1Moeller
  intersector4 = sse42::BVH4Triangle4Intersector4HybridMoellerNoFilter
  #ospModelViewer: done creating ospray model.
  #ospModelViewer: Adding a hard coded directional light as the sun.
  #ospModelViewer: done creating window. Press 'Q' to quit.
  #ospModelViewer: set world bounds [(-6.83587, -0.431891, -6.93969); (7.11329, 7.60655, 5.10773)], motion speed 0.0201081
  Embree: Invalid operation, (rtcIntersect8 not supported)
  Embree: Invalid operation, (rtcIntersect8 not supported)
  ...

On 05/18/2015 06:25 PM, Ingo Wald wrote:

Hey, Tim,

Can you tell us what branch (or git hash) you have been using?
Also, please run with "--osp:debug", and let us know the output, as well
as what architecture you have been using.

I just updated to current sources (in devel tree), and rebuilt, and it
works fine on my side ... and the output it makes are actually exactly
the right ones, so I'm a bit confused ... !?

Ingo

On 5/18/15 3:33 PM, timsandstrom wrote:

Look like data is loading correctly, but nothing being drawn. No CPU
activity at all.
This was pulled/built from sources today.

(tao:~/local/ospray-data/dragon) >
/u/wk/tsandstr/local/ospray/build/ospModelViewer dragon.xml
#ospModelViewer: starting to process cmdline arguments
#ospModelViewer: done parsing. found model with

• num meshes : 1 [922849]
• num instances: 1 [922849]
• num unique triangles : 922849
• num instanced triangles: 922849 #ospModelViewer: number of
  instances matches number of meshes, creating single model that
  contains all meshes #ospModelViewer: adding parsed geometries to
  ospray model #ospModelViewer: committing model #osp/trimesh:
  Verifying index buffer ... #osp/trimesh: Verifying vertex buffer
  ... #ospModelViewer: done creating ospray model. #ospModelViewer:
  Adding a hard coded directional light as the sun. #ospModelViewer:
  done creating window. Press 'Q' to quit. #ospModelViewer: set
  world bounds [(-6.83587, -0.431891, -6.93969); (7.11329, 7.60655,
  5.10773)], motion speed 0.0201081

—
Reply to this email directly or view it on GitHub
#17.

—
Reply to this email directly or view it on GitHub #17 (comment).

from ospray.

upon further inspection, my ispc is out of date [and llvm as well...] rebuilding... apologies :(

from ospray.

Ah; old ISPC version might explain it - the current version uses
multi-ISA build that may not be fully supported by an older ISPC. (We do
actually check the ISPC version when you first run cmake to check for
that case; but that check doesn't work if you update in an existing
repo, which is what I suppose happened).

Please update your ISPC, and ideally remove your build directory and
re-ccmake from scratch in a new build directory (just to be sure).

Let us know if that fixes it; until verified that it works I'll leave
this open.

Cheers,

Ingo

On 5/18/15 8:52 PM, timsandstrom wrote:

Arg, new to git, not sure how to get that info.
Lemme try... well, doesn't look like v0.8.1 does it.

If yours work then it's probably user-error on my part.

(tao:~/local/ospray) > git rev-parse --verify HEAD
8723ce6

(tao:~/local/ospray) > git log -n 1
commit 8723ce6
Merge: e7cb749 f639788
Author: Gregory P. Johnson [email protected]
Date: Fri May 8 11:56:13 2015 -0500

Merge branch 'release-0.8'

---

ospModelViewer --osp:debug dragon.xml

Embree Ray Tracing Kernels 2.3.3 (May 18 2015)
Compiler : GCC 4.9.2
Platform : Linux (64bit)
CPU : SSE SSE2 SSE3 SSSE3 SSE41 SSE42 POPCNT AVX F16C RDRAND
Features : intersection_filter bufferstride
general:
build threads = 1
verbosity = 2
triangles:
accel = default
builder = default
traverser = default
replications = 2
motion blur triangles:
accel = default
builder = default
traverser = default
hair:
accel = default
builder = default
traverser = default
replications = 3
subdivision surfaces:
accel = default
#ospModelViewer: starting to process cmdline arguments
#ospModelViewer: done parsing. found model with

• num meshes : 1 [922849]
• num instances: 1 [922849]
• num unique triangles : 922849
• num instanced triangles: 922849
  #ospModelViewer: number of instances matches number of meshes,
  creating single model that contains all meshes
  #ospModelViewer: adding parsed geometries to ospray model
  #ospModelViewer: committing model
  #osp/trimesh: Verifying index buffer ...
  #osp/trimesh: Verifying vertex buffer ...
  building BVH4 with sse2::BVH4Builder() ... [DONE]
  dt = 7068.84ms, perf = 0.130552 Mprim/s
  primitives = 922849, vertices = 0, hash= 26839253013198374
  sah = 1.0000, depth = 18
  used = 68.3 MB, perPrimitive = 74.0 B
  alignedNodes = 79799 (89.0% filled) (10.2 MB) (15.0% of total)
  leaves = 204397 (58.1 MB) (85.0% of total)
  vertices = 0 (0.0 MB) (0.0% of total) (75.0% used)
  created scene intersector
  accels[0]
  intersector1 = sse41::BVH4Triangle4Intersector1Moeller
  intersector4 = sse42::BVH4Triangle4Intersector4HybridMoellerNoFilter
  selected scene intersector
  intersector1 = sse41::BVH4Triangle4Intersector1Moeller
  intersector4 = sse42::BVH4Triangle4Intersector4HybridMoellerNoFilter
  #ospModelViewer: done creating ospray model.
  #ospModelViewer: Adding a hard coded directional light as the sun.
  #ospModelViewer: done creating window. Press 'Q' to quit.
  #ospModelViewer: set world bounds [(-6.83587, -0.431891, -6.93969);
  (7.11329, 7.60655, 5.10773)], motion speed 0.0201081
  Embree: Invalid operation, (rtcIntersect8 not supported)
  Embree: Invalid operation, (rtcIntersect8 not supported)
  ...

---

On 05/18/2015 06:25 PM, Ingo Wald wrote:

Hey, Tim,

Can you tell us what branch (or git hash) you have been using?
Also, please run with "--osp:debug", and let us know the output, as well
as what architecture you have been using.

I just updated to current sources (in devel tree), and rebuilt, and it
works fine on my side ... and the output it makes are actually exactly
the right ones, so I'm a bit confused ... !?

Ingo

On 5/18/15 3:33 PM, timsandstrom wrote:

Look like data is loading correctly, but nothing being drawn. No CPU
activity at all.
This was pulled/built from sources today.

(tao:~/local/ospray-data/dragon) >
/u/wk/tsandstr/local/ospray/build/ospModelViewer dragon.xml
#ospModelViewer: starting to process cmdline arguments
#ospModelViewer: done parsing. found model with

• num meshes : 1 [922849]
• num instances: 1 [922849]
• num unique triangles : 922849
• num instanced triangles: 922849 #ospModelViewer: number of
  instances matches number of meshes, creating single model that
  contains all meshes #ospModelViewer: adding parsed geometries to
  ospray model #ospModelViewer: committing model #osp/trimesh:
  Verifying index buffer ... #osp/trimesh: Verifying vertex buffer
  ... #ospModelViewer: done creating ospray model. #ospModelViewer:
  Adding a hard coded directional light as the sun. #ospModelViewer:
  done creating window. Press 'Q' to quit. #ospModelViewer: set
  world bounds [(-6.83587, -0.431891, -6.93969); (7.11329, 7.60655,
  5.10773)], motion speed 0.0201081

—
Reply to this email directly or view it on GitHub
#17.

—
Reply to this email directly or view it on GitHub
#17 (comment).

—
Reply to this email directly or view it on GitHub
#17 (comment).

from ospray.

OK, further updates...

so, cmake/gcc.make was not parsing the gcc --version output correctly
thus setting OSPRAY_COMPILER_SUPPORTS_AVX and OSPRAY_COMPILER_SUPPORTS_AVX2 to FALSE.

fixing that though, resulted in a compiler fail:

cd /u/wk/tsandstr/local/ospray/release/builddir/ospray/intel64 && g++ -DEMBREE_DISABLE_HAIR=1 -D__BUFFER_STRIDE__ -D__EXPORT_ALL_SYMBOLS__ -D__FIX_RAYS__ -D__INTERSECTION_FILTER__ -D__TARGET_AVX2__ -D__TARGET_AVX__ -D__TARGET_SSE41__ -D__TARGET_SSE42__ -fPIC -fno-strict-aliasing -DDEBUG -g -Wstrict-aliasing=1 -I/u/wk/tsandstr/local/ospray/release -I/u/wk/tsandstr/local/glut-3.7/include -I/u/wk/tsandstr/local/ospray -I/u/wk/tsandstr/local/ospray/release/builddir/ospray/intel64 -I/u/wk/tsandstr/local/ospray/ospray/embree -I/u/wk/tsandstr/local/ospray/ospray/embree/include -I/u/wk/tsandstr/local/ospray/ospray/embree/common -I/u/wk/tsandstr/local/ospray/ospray/embree/kernels -I/u/wk/tsandstr/local/ospray/ospray/embree/kernels/xeon -mf16c -mavx2 -mfma -mlzcnt -mbmi -mbmi2 -o CMakeFiles/ospray_embree_avx2.dir/embree/kernels/xeon/geometry/instance_intersector8.cpp.o -c /u/wk/tsandstr/local/ospray/ospray/embree/kernels/xeon/geometry/instance_intersector8.cpp
/tmp/ccLi6Kxi.s: Assembler messages:
/tmp/ccLi6Kxi.s:3365: Error: suffix or operands invalid for `vpcmpeqd'

manually setting OSPRAY_COMPILER_SUPPORTS_AVX2 to FALSE in cmake/gcc.make
results in a successful compile and rendering...

So, thanks so much for your attention and help,

-Tim

CPU: Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz
OS: Linux version 2.6.32-504.16.2.el6.x86_64 ([email protected])
(gcc version 4.4.7 20120313 (Red Hat 4.4.7-9) (GCC) ) #1 SMP Tue Mar 10 17:01:00 EDT 2015
gcc: 4.9.2
ispc:Intel(r) SPMD Program Compiler (ispc), 1.8.2dev (build commit ebe165dff87213ce @ 20150519, LLVM 3.5)

(tao:~/local) > ./bin/llvm-as -version
LLVM (http://llvm.org/):
LLVM version 3.5.1
Optimized build with assertions.
Built May 18 2015 (21:49:13).
Default target: x86_64-unknown-linux-gnu
Host CPU: core-avx-i

from ospray.

The new release of OSPRay (and the master branch) now correctly detects the version of GCC.

from ospray.