* GPU tests on master next - core-image-x11 - sfp @ 2013-08-28 16:43 Daiane Angolini 2013-08-28 17:20 ` Daiane Angolini 2013-08-28 21:53 ` Christian Betz 0 siblings, 2 replies; 15+ messages in thread From: Daiane Angolini @ 2013-08-28 16:43 UTC (permalink / raw) To: meta-freescale I've been running some tests on master-next. I used the following commands. Do you know any other I could run? RESULTS: $ DISPLAY=:0 glxgears 108 frames in 5.0 seconds = 21.508 FPS $ DISPLAY=:0 es2gears_x11 1494 frames in 5.0 seconds = 298.740 FPS $ DISPLAY=:0 glmark2-es2 glmark2 Score: 212 $ DISPLAY=:0 glmark2-es2 --fullscreen glmark2 Score: 229 For the complete log: http://pastebin.com/KG8VEmAZ -- Daiane ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: GPU tests on master next - core-image-x11 - sfp 2013-08-28 16:43 GPU tests on master next - core-image-x11 - sfp Daiane Angolini @ 2013-08-28 17:20 ` Daiane Angolini 2013-08-28 18:00 ` Eric Bénard 2013-08-28 21:53 ` Christian Betz 1 sibling, 1 reply; 15+ messages in thread From: Daiane Angolini @ 2013-08-28 17:20 UTC (permalink / raw) To: meta-freescale On 08/28/2013 01:43 PM, Daiane Angolini wrote: > I've been running some tests on master-next. > > I used the following commands. Do you know any other I could run? > > The past email, was for SOFT fp > RESULTS: > $ DISPLAY=:0 glxgears > 108 frames in 5.0 seconds = 21.508 FPS > > $ DISPLAY=:0 es2gears_x11 > 1494 frames in 5.0 seconds = 298.740 FPS > > $ DISPLAY=:0 glmark2-es2 > glmark2 Score: 212 > > $ DISPLAY=:0 glmark2-es2 --fullscreen > glmark2 Score: 229 The results for HARD fp: $ DISPLAY=:0 glxgears 107 frames in 5.0 seconds = 21.395 FPS $ DISPLAY=:0 es2gears_x11 1451 frames in 5.0 seconds = 290.200 FPS $ DISPLAY=:0 glmark2-es2 glmark2 Score: 213 $ DISPLAY=:0 glmark2-es2 --fullscreen glmark2 Score: 228 > > > > For the complete log: > http://pastebin.com/KG8VEmAZ For the complete log for HARD fp: http://pastebin.com/q6sx6Y1h > > -- Daiane ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: GPU tests on master next - core-image-x11 - sfp 2013-08-28 17:20 ` Daiane Angolini @ 2013-08-28 18:00 ` Eric Bénard 2013-08-28 18:29 ` Daiane Angolini 0 siblings, 1 reply; 15+ messages in thread From: Eric Bénard @ 2013-08-28 18:00 UTC (permalink / raw) To: Daiane Angolini; +Cc: meta-freescale Le Wed, 28 Aug 2013 14:20:25 -0300, Daiane Angolini <daiane.angolini@freescale.com> a écrit : > On 08/28/2013 01:43 PM, Daiane Angolini wrote: > > I've been running some tests on master-next. > > > > I used the following commands. Do you know any other I could run? > > > > > > The past email, was for SOFT fp > > RESULTS: > > $ DISPLAY=:0 glxgears > > 108 frames in 5.0 seconds = 21.508 FPS > > > > $ DISPLAY=:0 es2gears_x11 > > 1494 frames in 5.0 seconds = 298.740 FPS > > > > $ DISPLAY=:0 glmark2-es2 > > glmark2 Score: 212 > > > > $ DISPLAY=:0 glmark2-es2 --fullscreen > > glmark2 Score: 229 > > The results for HARD fp: > > > $ DISPLAY=:0 glxgears > 107 frames in 5.0 seconds = 21.395 FPS > > $ DISPLAY=:0 es2gears_x11 > 1451 frames in 5.0 seconds = 290.200 FPS > > $ DISPLAY=:0 glmark2-es2 > glmark2 Score: 213 > > $ DISPLAY=:0 glmark2-es2 --fullscreen > glmark2 Score: 228 > so hardfp brings lower results than softfp ... interesting ;-) Eric ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: GPU tests on master next - core-image-x11 - sfp 2013-08-28 18:00 ` Eric Bénard @ 2013-08-28 18:29 ` Daiane Angolini 2013-08-28 20:27 ` Eric Bénard 0 siblings, 1 reply; 15+ messages in thread From: Daiane Angolini @ 2013-08-28 18:29 UTC (permalink / raw) To: Eric Bénard; +Cc: meta-freescale On 08/28/2013 03:00 PM, Eric Bénard wrote: > Le Wed, 28 Aug 2013 14:20:25 -0300, > Daiane Angolini <daiane.angolini@freescale.com> a écrit : > >> On 08/28/2013 01:43 PM, Daiane Angolini wrote: >>> I've been running some tests on master-next. >>> >>> I used the following commands. Do you know any other I could run? >>> >>> >> >> The past email, was for SOFT fp >>> RESULTS: >>> $ DISPLAY=:0 glxgears >>> 108 frames in 5.0 seconds = 21.508 FPS >>> >>> $ DISPLAY=:0 es2gears_x11 >>> 1494 frames in 5.0 seconds = 298.740 FPS >>> >>> $ DISPLAY=:0 glmark2-es2 >>> glmark2 Score: 212 >>> >>> $ DISPLAY=:0 glmark2-es2 --fullscreen >>> glmark2 Score: 229 >> >> The results for HARD fp: >> >> >> $ DISPLAY=:0 glxgears >> 107 frames in 5.0 seconds = 21.395 FPS >> >> $ DISPLAY=:0 es2gears_x11 >> 1451 frames in 5.0 seconds = 290.200 FPS >> >> $ DISPLAY=:0 glmark2-es2 >> glmark2 Score: 213 >> >> $ DISPLAY=:0 glmark2-es2 --fullscreen >> glmark2 Score: 228 >> > so hardfp brings lower results than softfp ... interesting ;-) I took only measure. One time. I haven't calculated the standard deviation or error from measures. I would say the results is just "equal". And, I'm not sure how much hfp and sfp impacts the result of a task executed mainly by hardware (gpu). Do you think, by the results, that there is still any bug? > > Eric > -- Daiane ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: GPU tests on master next - core-image-x11 - sfp 2013-08-28 18:29 ` Daiane Angolini @ 2013-08-28 20:27 ` Eric Bénard 2013-08-29 8:01 ` Alexandre Belloni 2013-08-29 11:49 ` Daiane Angolini 0 siblings, 2 replies; 15+ messages in thread From: Eric Bénard @ 2013-08-28 20:27 UTC (permalink / raw) To: Daiane Angolini; +Cc: meta-freescale Le Wed, 28 Aug 2013 15:29:29 -0300, Daiane Angolini <daiane.angolini@freescale.com> a écrit : > On 08/28/2013 03:00 PM, Eric Bénard wrote: > > Le Wed, 28 Aug 2013 14:20:25 -0300, > > Daiane Angolini <daiane.angolini@freescale.com> a écrit : > > > >> On 08/28/2013 01:43 PM, Daiane Angolini wrote: > >>> I've been running some tests on master-next. > >>> > >>> I used the following commands. Do you know any other I could run? > >>> > >>> > >> > >> The past email, was for SOFT fp > >>> RESULTS: > >>> $ DISPLAY=:0 glxgears > >>> 108 frames in 5.0 seconds = 21.508 FPS > >>> > >>> $ DISPLAY=:0 es2gears_x11 > >>> 1494 frames in 5.0 seconds = 298.740 FPS > >>> > >>> $ DISPLAY=:0 glmark2-es2 > >>> glmark2 Score: 212 > >>> > >>> $ DISPLAY=:0 glmark2-es2 --fullscreen > >>> glmark2 Score: 229 > >> > >> The results for HARD fp: > >> > >> > >> $ DISPLAY=:0 glxgears > >> 107 frames in 5.0 seconds = 21.395 FPS > >> > >> $ DISPLAY=:0 es2gears_x11 > >> 1451 frames in 5.0 seconds = 290.200 FPS > >> > >> $ DISPLAY=:0 glmark2-es2 > >> glmark2 Score: 213 > >> > >> $ DISPLAY=:0 glmark2-es2 --fullscreen > >> glmark2 Score: 228 > >> > > so hardfp brings lower results than softfp ... interesting ;-) > > I took only measure. One time. I haven't calculated the standard > deviation or error from measures. > > I would say the results is just "equal". > > And, I'm not sure how much hfp and sfp impacts the result of a task > executed mainly by hardware (gpu). > > > Do you think, by the results, that there is still any bug? > on the GPU point of view no. I find interesting that the hardfp gives lower results than the softfp : maybe that would be interesting to have true real life benchmarks on this side (not GPU centric) before defaulting hardfp in meta-fsl-arm for i.MX6 platforms. Eric ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: GPU tests on master next - core-image-x11 - sfp 2013-08-28 20:27 ` Eric Bénard @ 2013-08-29 8:01 ` Alexandre Belloni 2013-08-29 11:47 ` Daiane Angolini 2013-08-29 11:49 ` Daiane Angolini 1 sibling, 1 reply; 15+ messages in thread From: Alexandre Belloni @ 2013-08-29 8:01 UTC (permalink / raw) To: Eric Bénard; +Cc: meta-freescale Hi, On 28/08/2013 22:27, Eric Bénard wrote: > > on the GPU point of view no. > > I find interesting that the hardfp gives lower results than the > softfp : maybe that would be interesting to have true real life > benchmarks on this side (not GPU centric) before defaulting hardfp in > meta-fsl-arm for i.MX6 platforms. I'm not sure we can actually find real life benchmarks, it always depends on how intensive are the floating point operation compared to the rest of the system. Also, I remember seeing that the difference between softfp and hardfp is not as big on cortex-a9 as on cortex-a8. Actually, softfp is already using the VFP but using the soft ABI. Then, it has to copy values from integer registers to float registers and that is the overhead. This overhead is lower on cortex-a9 because the vfp is pipelined, this was not the case on cortex -a8. You can find interesting benchs here: https://wiki.linaro.org/OfficeofCTO/HardFloat/Benchmarks https://wiki.linaro.org/OfficeofCTO/HardFloat/Benchmarks201205 -- Alexandre Belloni, Free Electrons Embedded Linux, Kernel and Android engineering http://free-electrons.com ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: GPU tests on master next - core-image-x11 - sfp 2013-08-29 8:01 ` Alexandre Belloni @ 2013-08-29 11:47 ` Daiane Angolini 0 siblings, 0 replies; 15+ messages in thread From: Daiane Angolini @ 2013-08-29 11:47 UTC (permalink / raw) To: Alexandre Belloni; +Cc: meta-freescale On 08/29/2013 05:01 AM, Alexandre Belloni wrote: > Hi, > > On 28/08/2013 22:27, Eric Bénard wrote: >> >> on the GPU point of view no. >> >> I find interesting that the hardfp gives lower results than the >> softfp : maybe that would be interesting to have true real life >> benchmarks on this side (not GPU centric) before defaulting hardfp in >> meta-fsl-arm for i.MX6 platforms. > > I'm not sure we can actually find real life benchmarks, it always > depends on how intensive are the floating point operation compared to > the rest of the system. > > Also, I remember seeing that the difference between softfp and hardfp is > not as big on cortex-a9 as on cortex-a8. Actually, softfp is already > using the VFP but using the soft ABI. Then, it has to copy values from > integer registers to float registers and that is the overhead. This > overhead is lower on cortex-a9 because the vfp is pipelined, this was > not the case on cortex -a8. > > You can find interesting benchs here: > > https://wiki.linaro.org/OfficeofCTO/HardFloat/Benchmarks > https://wiki.linaro.org/OfficeofCTO/HardFloat/Benchmarks201205 sweet > > -- Daiane ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: GPU tests on master next - core-image-x11 - sfp 2013-08-28 20:27 ` Eric Bénard 2013-08-29 8:01 ` Alexandre Belloni @ 2013-08-29 11:49 ` Daiane Angolini 1 sibling, 0 replies; 15+ messages in thread From: Daiane Angolini @ 2013-08-29 11:49 UTC (permalink / raw) To: Eric Bénard; +Cc: meta-freescale On 08/28/2013 05:27 PM, Eric Bénard wrote: > Le Wed, 28 Aug 2013 15:29:29 -0300, > Daiane Angolini <daiane.angolini@freescale.com> a écrit : > >> On 08/28/2013 03:00 PM, Eric Bénard wrote: >>> Le Wed, 28 Aug 2013 14:20:25 -0300, >>> Daiane Angolini <daiane.angolini@freescale.com> a écrit : >>> >>>> On 08/28/2013 01:43 PM, Daiane Angolini wrote: >>>>> I've been running some tests on master-next. >>>>> >>>>> I used the following commands. Do you know any other I could run? >>>>> >>>>> >>>> >>>> The past email, was for SOFT fp >>>>> RESULTS: >>>>> $ DISPLAY=:0 glxgears >>>>> 108 frames in 5.0 seconds = 21.508 FPS >>>>> >>>>> $ DISPLAY=:0 es2gears_x11 >>>>> 1494 frames in 5.0 seconds = 298.740 FPS >>>>> >>>>> $ DISPLAY=:0 glmark2-es2 >>>>> glmark2 Score: 212 >>>>> >>>>> $ DISPLAY=:0 glmark2-es2 --fullscreen >>>>> glmark2 Score: 229 >>>> >>>> The results for HARD fp: >>>> >>>> >>>> $ DISPLAY=:0 glxgears >>>> 107 frames in 5.0 seconds = 21.395 FPS >>>> >>>> $ DISPLAY=:0 es2gears_x11 >>>> 1451 frames in 5.0 seconds = 290.200 FPS >>>> >>>> $ DISPLAY=:0 glmark2-es2 >>>> glmark2 Score: 213 >>>> >>>> $ DISPLAY=:0 glmark2-es2 --fullscreen >>>> glmark2 Score: 228 >>>> >>> so hardfp brings lower results than softfp ... interesting ;-) >> >> I took only measure. One time. I haven't calculated the standard >> deviation or error from measures. >> >> I would say the results is just "equal". >> >> And, I'm not sure how much hfp and sfp impacts the result of a task >> executed mainly by hardware (gpu). >> >> >> Do you think, by the results, that there is still any bug? >> > on the GPU point of view no. Great, the most important question I need to answer is "is that working". > > I find interesting that the hardfp gives lower results than the > softfp : maybe that would be interesting to have true real life > benchmarks on this side (not GPU centric) before defaulting hardfp in > meta-fsl-arm for i.MX6 platforms. With some time/test we can decide "which is the best" Nice ;-) > > Eric > -- Daiane ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: GPU tests on master next - core-image-x11 - sfp 2013-08-28 16:43 GPU tests on master next - core-image-x11 - sfp Daiane Angolini 2013-08-28 17:20 ` Daiane Angolini @ 2013-08-28 21:53 ` Christian Betz 2013-08-29 11:44 ` Daiane Angolini 1 sibling, 1 reply; 15+ messages in thread From: Christian Betz @ 2013-08-28 21:53 UTC (permalink / raw) Cc: meta-freescale On Wed, Aug 28, 2013 at 12:43 PM, Daiane Angolini <daiane.angolini@freescale.com> wrote: > I've been running some tests on master-next. > > I used the following commands. Do you know any other I could run? > > > RESULTS: > $ DISPLAY=:0 glxgears > 108 frames in 5.0 seconds = 21.508 FPS > > $ DISPLAY=:0 es2gears_x11 > 1494 frames in 5.0 seconds = 298.740 FPS > > $ DISPLAY=:0 glmark2-es2 > glmark2 Score: 212 > > $ DISPLAY=:0 glmark2-es2 --fullscreen > glmark2 Score: 229 this is great news. thank you for sharing. does this imply that windowed mode acceleration is working now? (i seem to recall it was broken) ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: GPU tests on master next - core-image-x11 - sfp 2013-08-28 21:53 ` Christian Betz @ 2013-08-29 11:44 ` Daiane Angolini 2013-08-29 17:33 ` Daiane Angolini 0 siblings, 1 reply; 15+ messages in thread From: Daiane Angolini @ 2013-08-29 11:44 UTC (permalink / raw) To: meta-freescale, christian.betz On 08/28/2013 06:53 PM, Christian Betz wrote: > On Wed, Aug 28, 2013 at 12:43 PM, Daiane Angolini > <daiane.angolini@freescale.com> wrote: >> I've been running some tests on master-next. >> >> I used the following commands. Do you know any other I could run? >> >> >> RESULTS: >> $ DISPLAY=:0 glxgears >> 108 frames in 5.0 seconds = 21.508 FPS >> >> $ DISPLAY=:0 es2gears_x11 >> 1494 frames in 5.0 seconds = 298.740 FPS >> >> $ DISPLAY=:0 glmark2-es2 >> glmark2 Score: 212 >> >> $ DISPLAY=:0 glmark2-es2 --fullscreen >> glmark2 Score: 229 > > this is great news. thank you for sharing. > > does this imply that windowed mode acceleration is working now? (i > seem to recall it was broken) I think so, but how can I test it (for sure)? I see the xterm on my screen > _______________________________________________ > meta-freescale mailing list > meta-freescale@yoctoproject.org > https://lists.yoctoproject.org/listinfo/meta-freescale > -- Daiane ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: GPU tests on master next - core-image-x11 - sfp 2013-08-29 11:44 ` Daiane Angolini @ 2013-08-29 17:33 ` Daiane Angolini 2013-08-30 15:53 ` Diego Rondini 0 siblings, 1 reply; 15+ messages in thread From: Daiane Angolini @ 2013-08-29 17:33 UTC (permalink / raw) To: meta-freescale, christian.betz On 08/29/2013 08:44 AM, Daiane Angolini wrote: > On 08/28/2013 06:53 PM, Christian Betz wrote: >> On Wed, Aug 28, 2013 at 12:43 PM, Daiane Angolini >> <daiane.angolini@freescale.com> wrote: >>> I've been running some tests on master-next. >>> >>> I used the following commands. Do you know any other I could run? >>> >>> >>> RESULTS: >>> $ DISPLAY=:0 glxgears >>> 108 frames in 5.0 seconds = 21.508 FPS >>> >>> $ DISPLAY=:0 es2gears_x11 >>> 1494 frames in 5.0 seconds = 298.740 FPS >>> >>> $ DISPLAY=:0 glmark2-es2 >>> glmark2 Score: 212 >>> >>> $ DISPLAY=:0 glmark2-es2 --fullscreen >>> glmark2 Score: 229 >> >> this is great news. thank you for sharing. >> >> does this imply that windowed mode acceleration is working now? (i >> seem to recall it was broken) > > I think so, but how can I test it (for sure)? I was told: If you're looking to verify the alpha.2 windowed mode optimizations, compare frame rates to the alpha.1 binaries for es2 applications in windowed mode: es2gears_x11, glmark2-es2, /opt/viv_samples/vdk/tutorial*_es20 They're roughly 10x higher, and comparable to fullscreen performance. You can see that glx hasn't been optimized because windowed mode for glxgears is still ~22 fps, where the fullscreen is ~300. -- Daiane ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: GPU tests on master next - core-image-x11 - sfp 2013-08-29 17:33 ` Daiane Angolini @ 2013-08-30 15:53 ` Diego Rondini 2013-08-30 16:53 ` Daiane Angolini 0 siblings, 1 reply; 15+ messages in thread From: Diego Rondini @ 2013-08-30 15:53 UTC (permalink / raw) To: meta-freescale > You can see that glx hasn't been optimized because windowed mode for > glxgears is still ~22 fps, where the fullscreen is ~300. Daiane, do you know if that means that now there's support for pure OpenGL 2.1 functionalities in the build? And by OpenGL 2.1 I really mean OpenGL, not OpenGL ES x.y. I can partially run glmark2 (not glmark2-es) with some errors, but driver is reporting: # glmark2 --fullscreen ** GLX does not support GLX_EXT_swap_control or GLX_MESA_swap_control! ** Failed to set swap interval. Results may be bounded above by refresh rate. ======================================================= glmark2 2012.12 ======================================================= OpenGL Information GL_VENDOR: Vivante Corporation GL_RENDERER: GC2000 Graphics Engine GL_VERSION: 2.1 2.0.1 ======================================================= ** GLX does not support GLX_EXT_swap_control or GLX_MESA_swap_control! ** Failed to set swap interval. Results may be bounded above by refresh rate. [build] use-vbo=false: FPS: 101 FrameTime: 9.901 ms ** GLX does not support GLX_EXT_swap_control or GLX_MESA_swap_control! ** Failed to set swap interval. Results may be bounded above by refresh rate. [build] use-vbo=true: FPS: 108 FrameTime: 9.259 ms So GL_VERSION is "2.1 2.0.1", not "OpenGL ES 2.0". Same goes for: # glxinfo name of display: :0 display: :0 screen: 0 direct rendering: Yes server glx vendor string: SGI server glx version string: 1.4 server glx extensions: GLX_ARB_multisample, GLX_EXT_import_context, GLX_EXT_texture_from_pixmap, GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_MESA_copy_sub_buffer, GLX_OML_swap_method, GLX_SGIS_multisample, GLX_SGIX_fbconfig, GLX_SGIX_pbuffer, GLX_SGI_make_current_read client glx vendor string: Vivante Corp client glx version string: 1.4 client glx extensions: GLX_ARB_get_proc_address, GLX_SGIX_fbconfig, GLX_SGIX_pbuffer, GLX_SGI_make_current_read GLX version: 1.4 GLX extensions: GLX_SGIX_fbconfig, GLX_SGIX_pbuffer, GLX_SGI_make_current_read OpenGL vendor string: Vivante Corporation OpenGL renderer string: GC2000 Graphics Engine OpenGL version string: 2.1 2.0.1 OpenGL shading language version string: 1.20 OpenGL extensions: GL_ARB_depth_texture, GL_ARB_fragment_program_shadow, GL_ARB_fragment_shader, GL_ARB_imaging, GL_ARB_multisample, GL_ARB_multitexture, GL_ARB_pixel_buffer_object, GL_ARB_shader_objects, GL_ARB_texture_compression, GL_ARB_texture_env_add, GL_ARB_texture_env_combine, GL_ARB_texture_non_power_of_two, GL_ARB_texture_rectangle, GL_ARB_vertex_buffer_object, GL_ARB_vertex_shader, GL_ARB_window_pos, GL_ATI_draw_buffers, GL_ATI_element_array, GL_ATI_separate_stencil, GL_ATI_texture_env_combine3, GL_ATI_vertex_array_object, GL_EXT_abgr, GL_EXT_bgra, GL_EXT_blend_color, GL_EXT_blend_equation_separate, GL_EXT_blend_func_separate, GL_EXT_color_table, GL_EXT_convolution, GL_EXT_draw_buffers2, GL_EXT_draw_instanced, GL_EXT_draw_range_elements, GL_EXT_framebuffer_blit, GL_EXT_framebuffer_multisample, GL_EXT_framebuffer_object, GL_EXT_histogram, GL_EXT_pixel_buffer_object, GL_EXT_polygon_offset, GL_EXT_rescale_normal, GL_EXT_secondary_color, GL_EXT_separate_specular_color, GL_EXT_stencil_two_side, GL_EXT_subtexture, GL_EXT_texture, GL_EXT_texture3D, GL_EXT_texture_array, GL_EXT_texture_buffer_object, GL_EXT_texture_compression_s3tc, GL_EXT_texture_cube_map, GL_EXT_texture_env_add, GL_EXT_texture_env_combine, GL_EXT_texture_filter_anisotropic, GL_EXT_texture_lod_bias, GL_EXT_texture_object, GL_EXT_texture_rectangle, GL_EXT_timer_query, GL_EXT_vertex_array, GL_NV_texture_rectangle, GL_SGIS_generate_mipmap, GL_SGIS_texture_lod, GL_SGI_color_matrix, WGL_ARB_extensions_string, WGL_EXT_extensions_string, WGL_EXT_swap_control So will be OpenGL be possible? Thanks, Diego ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: GPU tests on master next - core-image-x11 - sfp 2013-08-30 15:53 ` Diego Rondini @ 2013-08-30 16:53 ` Daiane Angolini 2013-09-02 8:15 ` Diego 0 siblings, 1 reply; 15+ messages in thread From: Daiane Angolini @ 2013-08-30 16:53 UTC (permalink / raw) To: Diego Rondini; +Cc: meta-freescale On 08/30/2013 12:53 PM, Diego Rondini wrote: >> You can see that glx hasn't been optimized because windowed mode for >> glxgears is still ~22 fps, where the fullscreen is ~300. > > Daiane, do you know if that means that now there's support for pure OpenGL 2.1 > functionalities in the build? > And by OpenGL 2.1 I really mean OpenGL, not OpenGL ES x.y. I know this is not the first GPU release that support OpenGL (I cannot ensure the version number because I really don't remember). So, this GPU versio DOES support OpenGL. During my testes I faced libGL.so AND libEGL.so. So, it does support both. > > I can partially run glmark2 (not glmark2-es) with some errors, but driver is > reporting: > # glmark2 --fullscreen > ** GLX does not support GLX_EXT_swap_control or GLX_MESA_swap_control! > ** Failed to set swap interval. Results may be bounded above by refresh rate. > ======================================================= > glmark2 2012.12 > ======================================================= > OpenGL Information > GL_VENDOR: Vivante Corporation > GL_RENDERER: GC2000 Graphics Engine > GL_VERSION: 2.1 2.0.1 > ======================================================= > ** GLX does not support GLX_EXT_swap_control or GLX_MESA_swap_control! > ** Failed to set swap interval. Results may be bounded above by refresh rate. > [build] use-vbo=false: FPS: 101 FrameTime: 9.901 ms > ** GLX does not support GLX_EXT_swap_control or GLX_MESA_swap_control! > ** Failed to set swap interval. Results may be bounded above by refresh rate. > [build] use-vbo=true: FPS: 108 FrameTime: 9.259 ms > > > So GL_VERSION is "2.1 2.0.1", not "OpenGL ES 2.0". > > Same goes for: > # glxinfo > name of display: :0 > display: :0 screen: 0 > direct rendering: Yes > server glx vendor string: SGI > server glx version string: 1.4 > server glx extensions: > GLX_ARB_multisample, GLX_EXT_import_context, GLX_EXT_texture_from_pixmap, > GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_MESA_copy_sub_buffer, > GLX_OML_swap_method, GLX_SGIS_multisample, GLX_SGIX_fbconfig, > GLX_SGIX_pbuffer, GLX_SGI_make_current_read > client glx vendor string: Vivante Corp > client glx version string: 1.4 > client glx extensions: > GLX_ARB_get_proc_address, GLX_SGIX_fbconfig, GLX_SGIX_pbuffer, > GLX_SGI_make_current_read > GLX version: 1.4 > GLX extensions: > GLX_SGIX_fbconfig, GLX_SGIX_pbuffer, GLX_SGI_make_current_read > OpenGL vendor string: Vivante Corporation > OpenGL renderer string: GC2000 Graphics Engine > OpenGL version string: 2.1 2.0.1 > OpenGL shading language version string: 1.20 > OpenGL extensions: > GL_ARB_depth_texture, GL_ARB_fragment_program_shadow, > GL_ARB_fragment_shader, GL_ARB_imaging, GL_ARB_multisample, > GL_ARB_multitexture, GL_ARB_pixel_buffer_object, GL_ARB_shader_objects, > GL_ARB_texture_compression, GL_ARB_texture_env_add, > GL_ARB_texture_env_combine, GL_ARB_texture_non_power_of_two, > GL_ARB_texture_rectangle, GL_ARB_vertex_buffer_object, > GL_ARB_vertex_shader, GL_ARB_window_pos, GL_ATI_draw_buffers, > GL_ATI_element_array, GL_ATI_separate_stencil, > GL_ATI_texture_env_combine3, GL_ATI_vertex_array_object, GL_EXT_abgr, > GL_EXT_bgra, GL_EXT_blend_color, GL_EXT_blend_equation_separate, > GL_EXT_blend_func_separate, GL_EXT_color_table, GL_EXT_convolution, > GL_EXT_draw_buffers2, GL_EXT_draw_instanced, GL_EXT_draw_range_elements, > GL_EXT_framebuffer_blit, GL_EXT_framebuffer_multisample, > GL_EXT_framebuffer_object, GL_EXT_histogram, GL_EXT_pixel_buffer_object, > GL_EXT_polygon_offset, GL_EXT_rescale_normal, GL_EXT_secondary_color, > GL_EXT_separate_specular_color, GL_EXT_stencil_two_side, > GL_EXT_subtexture, GL_EXT_texture, GL_EXT_texture3D, > GL_EXT_texture_array, GL_EXT_texture_buffer_object, > GL_EXT_texture_compression_s3tc, GL_EXT_texture_cube_map, > GL_EXT_texture_env_add, GL_EXT_texture_env_combine, > GL_EXT_texture_filter_anisotropic, GL_EXT_texture_lod_bias, > GL_EXT_texture_object, GL_EXT_texture_rectangle, GL_EXT_timer_query, > GL_EXT_vertex_array, GL_NV_texture_rectangle, GL_SGIS_generate_mipmap, > GL_SGIS_texture_lod, GL_SGI_color_matrix, WGL_ARB_extensions_string, > WGL_EXT_extensions_string, WGL_EXT_swap_control > > So will be OpenGL be possible? yes ^.^ -- Daiane ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: GPU tests on master next - core-image-x11 - sfp 2013-08-30 16:53 ` Daiane Angolini @ 2013-09-02 8:15 ` Diego 2013-09-02 12:26 ` Daiane Angolini 0 siblings, 1 reply; 15+ messages in thread From: Diego @ 2013-09-02 8:15 UTC (permalink / raw) To: Daiane Angolini; +Cc: meta-freescale In data venerdì 30 agosto 2013 13:53:23, Daiane Angolini ha scritto: > On 08/30/2013 12:53 PM, Diego Rondini wrote: > >> You can see that glx hasn't been optimized because windowed mode for > >> glxgears is still ~22 fps, where the fullscreen is ~300. > > > > Daiane, do you know if that means that now there's support for pure OpenGL > > 2.1 functionalities in the build? > > And by OpenGL 2.1 I really mean OpenGL, not OpenGL ES x.y. > > I know this is not the first GPU release that support OpenGL (I cannot > ensure the version number because I really don't remember). So, this GPU > versio DOES support OpenGL. > > During my testes I faced libGL.so AND libEGL.so. So, it does support both. > I have those libs too, but it looks like gpu-viv-bin-mx6q-3.5.7-1.0.0-alpha.2- hfp.bin doesn't include any GL header, so while the library is there, there's no official way to interface to libGL.so. Do you have any information about that issue? Should I file a bug report? Thanks, Diego ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: GPU tests on master next - core-image-x11 - sfp 2013-09-02 8:15 ` Diego @ 2013-09-02 12:26 ` Daiane Angolini 0 siblings, 0 replies; 15+ messages in thread From: Daiane Angolini @ 2013-09-02 12:26 UTC (permalink / raw) To: Diego; +Cc: meta-freescale On 09/02/2013 05:15 AM, Diego wrote: > In data venerdì 30 agosto 2013 13:53:23, Daiane Angolini ha scritto: >> On 08/30/2013 12:53 PM, Diego Rondini wrote: >>>> You can see that glx hasn't been optimized because windowed mode for >>>> glxgears is still ~22 fps, where the fullscreen is ~300. >>> >>> Daiane, do you know if that means that now there's support for pure OpenGL >>> 2.1 functionalities in the build? >>> And by OpenGL 2.1 I really mean OpenGL, not OpenGL ES x.y. >> >> I know this is not the first GPU release that support OpenGL (I cannot >> ensure the version number because I really don't remember). So, this GPU >> versio DOES support OpenGL. >> >> During my testes I faced libGL.so AND libEGL.so. So, it does support both. >> > > I have those libs too, but it looks like gpu-viv-bin-mx6q-3.5.7-1.0.0-alpha.2- > hfp.bin doesn't include any GL header, so while the library is there, there's > no official way to interface to libGL.so. > > Do you have any information about that issue? Should I file a bug report? No, I don't have any other information. Please, submit a bug report ;-) This is only a "bug" from hfp package? Could you, please, test sfp as well and report a complete bug report? Thanks in advance. > > Thanks, > Diego > > -- Daiane ^ permalink raw reply [flat|nested] 15+ messages in thread
end of thread, other threads:[~2013-09-02 12:29 UTC | newest] Thread overview: 15+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2013-08-28 16:43 GPU tests on master next - core-image-x11 - sfp Daiane Angolini 2013-08-28 17:20 ` Daiane Angolini 2013-08-28 18:00 ` Eric Bénard 2013-08-28 18:29 ` Daiane Angolini 2013-08-28 20:27 ` Eric Bénard 2013-08-29 8:01 ` Alexandre Belloni 2013-08-29 11:47 ` Daiane Angolini 2013-08-29 11:49 ` Daiane Angolini 2013-08-28 21:53 ` Christian Betz 2013-08-29 11:44 ` Daiane Angolini 2013-08-29 17:33 ` Daiane Angolini 2013-08-30 15:53 ` Diego Rondini 2013-08-30 16:53 ` Daiane Angolini 2013-09-02 8:15 ` Diego 2013-09-02 12:26 ` Daiane Angolini
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.