Raspberry Pi Forums

blackshard83

Of course that's an approach, but it will work well only if your "background" does never change.

blackshard83

It is really not supported? It should be supported as long as the broadcom proprietary EGL/OpenGLES/OpenVG libraries are still supported. Look that the proprietary libraries don't work in X, they are supposed to work exclusively on a dispmanx layer. If you're using the opensource OpenGL driver (thus...

blackshard83

ARM is another story: no improvements to ISA, no more general purpose registers, no legacy compatibility to get rid of; What ??? Sorry, that is nonsense. 1) A64 is a completely different ISA to A32. It is greatly improved and streamlined for modern CPU's. 2) The number of registers has doubled. And...

blackshard83

Read this : ...cut... I will be very surprised that the gain is only 5% on ARM. On a 4-lane highway you put at most 2 times more cars than on a 2-lane highway, it's applied physics... In computer science there are also other problems of optimization of dependencies, etc, which potentially reduces t...

blackshard83

I hope it will be better when we have an OS and 64-bit applications available (theoretical performance gain of 30 to 70%). It is OT, but where you saw such gain of performance going 64-bits? It's not x86, performance gains are in the order of 5% or less on ARM, but memory usage may increase a lot d...

blackshard83

Khronos has nothing to do with implementation of Vulkan, they just define the standard and compliance.
It's not their intent or purpose to create the drivers for the devices.

blackshard83

Is the firmware blob still there?

blackshard83

Thanks a lot for looking into.
I tried the other approach (disable the output port, then do OMX_FreeBuffers and finally do the transition to Idle) and it works and it looks tidier. I updated the pull request with the new solution.

blackshard83

I don't have an answer, but OOI, do you *need* to do a cleanup? Can you not just exit, and let the system sort itself out? Well, it's not my use case. It should fit into a long running process, so it's not a viable option because it leaks resources, in particular it leaks the ppBufferHdr buffer all...

blackshard83

Hello, I made some changes to stock omxplayer to use egl_render to render into an EGLImageKHR resource using the egl_render component instead of usual video_render. It worked, but during the cleanup it freezed because the component hangs when it is going to OMX_StateLoaded state. I checked hello_vid...

blackshard83

using f2fs for the partition where chromium writes mounting the partition with commit=x argument (x is the number of seconds the cache has to be flushed after, look for documentation for full explanation) mount the partition with noatime argument, so there is the access time is not changed every tim...

blackshard83

I revive this old thread with an edge case. Here there is some code (and precompiled binary) that creates a pbuffer and uses vgGaussianBlur() to make some image manipulation using VC4. It does not render anything on screen, but running this while some audio is playing causes massive buffer underruns...

blackshard83

As an indirectly related matter which goes beyond dispmanx, EGL provides some facilities (EGLWaitNative, EGL_KHR_lock_surface extension , ...) to share native buffers between ARM an GPU. I never used it, so I don't know what is the impact in performance and other caveats, but in theory you can use G...

blackshard83

Just pass -march=armv6 to gcc compilation arguments. I'm doing this all the time (compiling on a Pi2, running on a Pi1) If you want to be sure, maybe you can add -mfpu=vfp and -mfloat-abi=hard too. You can patch CFLAGS variable in the Makefile or something like this, depending on the project. This i...

blackshard83

It depends on the quality of your converter. HDMI and USB are both digital representation of data, so there is no loss during transport of the signal: the quality of their DAC and the board circuitry is crucial for a good output. An onboard hat with a DAC is exactly the same, it just uses the intern...

blackshard83

The last time I took a deep dive into SDCARD performance, and specifically the more recent Sandisk cards. Pointed to several reasons most people anecdotally recall F2FS doesn't really perform much better than ext4: 1. Primarily F2FS was intended for raw memory chips without the built-in wear-leveli...

blackshard83

Is that regardless of whether you 'sync' beforehand. Or is data safe as long as applications fsync appropriately? IIRC, the default behaviour of linux is to flush data that has been cached for more than 30 seconds, using too much RAM and block IO if cached data is taking up more than a certain perc...

blackshard83

Ext4 is working well. F2fs hasn't been proven to be a worthwhile. When people have tried it for Android, they found that performance degrades over time, requiring them to reformat. Then they went back to Ext4. If you look on XDA, the general view of it isn't positive. To consider switching, we'd ne...

blackshard83

Just to add some other details, I'm noticing that calling eglSwapBuffers() starves all other threads of the same process which try to call eglMakeCurrent() and also other threads of the same process which try to invoke OMX functions to load videos. Juggling with priorities (lowering rendering thread...

blackshard83

Hello, I have a pure OpenVG application with two threads: the first thread is the graphics rendering loop, and the other thread does the resource upstreaming. The first thread does its work and then issues a call to eglSwapBuffers() call. eglSwapBuffes waits for vsync, blits the things and then does...

blackshard83

PeterO wrote: ↑
Sat Feb 24, 2018 8:39 am
BUT ISTR that might depend on the sound device having to support signed samples (and not all of them do that). I can't remember if the alsa code will do the conversion on the fly for you..

AFAIK signed 16 bit integers are the standard for 16 bit samples

blackshard83

Hello, I'm working on a font rendering engine that uses vector paths and renders them via OpenVG. Everything works fine at the moment, but using paths for font glyphs causes a lot of free blocks inside video memory. I provide exact segment and coords hints for the paths, create them using vgCreatePa...

blackshard83

You try to do some testing with hub-ctrl you can find somewhere here in the forums. Some time ago I tried it but I felt it was not really cutting power to the USB ports. I needed to really cut power to reset some stubborn USB sticks, but was not able to. It also works "reliably" only on rpi2 and upp...

blackshard83

Got it, thanks for the detailed and interesting explanation!

blackshard83

Hello, I detatched the framebuffer using this command: echo "soc:fb" > /sys/bus/platform/drivers/bcm2708_fb/unbind this detatches the framebuffer completely, so /dev/fb0 disappears. In this state launching: vcdbg reloc causes a lot of error rows like this: debug_sym: vc_mem_copy: Unable to open '/de...

Search found 92 matches

Re: Best way to handle sprites

Re: Recommendation for an alternative to OpenVG for RP4

Re: Raspberry Pi 4 temperature

Re: Raspberry Pi 4 temperature

Re: Raspberry Pi 4 temperature

Re: Do we get Vulkan on Pi 4?

Re: Raspberry Pi 4 Thread - general discussion

Re: egl_render, hangs at cleanup

Re: egl_render, hangs at cleanup

egl_render, hangs at cleanup

Re: Seeking the illusive zero write

Re: Analogue audio redux

Re: Getting direct access to a dispmanx resource

Re: Compile programs for the Raspberry Pi 1 and Zero's ARMv6 SoC

Re: On-Board HDMI Audio vs USB Audio vs Add-on DAC Card

Re: f2fs : why not distribution with root fs as f2fs

Re: f2fs : why not distribution with root fs as f2fs

Re: f2fs : why not distribution with root fs as f2fs

Re: EGL and concurrent resource upload

EGL and concurrent resource upload

Re: ALSA Programming

Reduce VRAM fragmentation

Re: Is there a way to turn off the Vdd on the USB connectors

Re: vcdbg: unable to open /dev/fb0 (unbound framebuffer)

vcdbg: unable to open /dev/fb0 (unbound framebuffer)