nightseas Posted August 3, 2017 Posted August 3, 2017 I'm starting to add RK3399 to Armbian, in my spare time. Here's the board I got, RK3399 official EVB, which spent me 500CNY (~63EURO) on Taobao, quite cheap comparing to Firefly. https://item.taobao.com/item.htm?spm=a1z09.2.0.0.1f30a53flpk7IA&id=554454257549&_u=31t0eea9a1c Spec: - Rockchip RK3399 (2xA72 + 4xA53, Mali T860 MP4) - 4GBytes Dual channel LPDDR3 (64-bit in total) - 8GBytes eMMC, Micro SD Slot - USB 3.0 Type-A, USB Type-C (DP over Type-C), USB2.0, HDMI - 1G Ethernet - Raspberry Pi compatible 2x20 pin header - Other interfaces: PCIe 2.0 x4, MIPI CSI & DSI, eDP, LVDS.... You can download Cadence OrCAD format schematic and Allegro format PCB files for free. It's open sourced and you'll also find BOM, gerber files ready for production. http://opensource.rock-chips.com/wiki_Excavator_sapphire_board I was able to build Linux 4.4 + Debian 9 with build scripts from Rockchip. And it looks like the OpenGLES and OpenCL are ready? (not fully tested). VPU still has some problem. Will continue to update in this post... 1
hipboi Posted August 3, 2017 Posted August 3, 2017 Looks promising Update: I just found another website called vamrs sell the Sapphire board. https://store.vamrs.com/products/rockchip-rk3399-sapphire-board
Myy Posted August 3, 2017 Posted August 3, 2017 Concerning OpenGL ES tests : You might need to download one of these drivers in a folder, create the various libGLESv1_CM.so{,.1} libGLESv2.so{,.2} libOpenCL.so{,.1} libEGL.so{,.1} links to the libmali you downloaded and then set LD_LIBRARY_PATH when testing OpenGL ES little programs. I might be missing a few file links... So, if you downloaded libmali-midgard-4th-r13p0-gbm.so in /tmp/test-mali for example, this should do the trick : cd /tmp/test-mali for link in libGLESv1_CM.so{,.1} libGLESv2.so{,.2} libOpenCL.so{,.1} libEGL.so{,.1}; do ln -s libmali-midgard-4th-r13p0.so $link; done LD_LIBRARY_PATH=/tmp/test-mali ./es2gears Change es2gears by the name of that the OpenGL ES program binary you want to test of course. Note that if /dev/mali0 is not read/writeable by your user, the OpenGL ES program might fail with a Mali Driver error message telling you that you do not have the right GPU. It's a permission issue in that case. If you don't have a /dev/mali0, it's a driver integration issue however. No Vulkan support for the moment, however. The ARM guys do not seem to have released any driver for the Firefly RK3399. (´・ω・` )
nightseas Posted August 4, 2017 Author Posted August 4, 2017 18 hours ago, Myy said: Concerning OpenGL ES tests : You might need to download one of these drivers in a folder, create the various libGLESv1_CM.so{,.1} libGLESv2.so{,.2} libOpenCL.so{,.1} libEGL.so{,.1} links to the libmali you downloaded and then set LD_LIBRARY_PATH when testing OpenGL ES little programs. I might be missing a few file links... So, if you downloaded libmali-midgard-4th-r13p0-gbm.so in /tmp/test-mali for example, this should do the trick : cd /tmp/test-mali for link in libGLESv1_CM.so{,.1} libGLESv2.so{,.2} libOpenCL.so{,.1} libEGL.so{,.1}; do ln -s libmali-midgard-4th-r13p0.so $link; done LD_LIBRARY_PATH=/tmp/test-mali ./es2gears Change es2gears by the name of that the OpenGL ES program binary you want to test of course. Note that if /dev/mali0 is not read/writeable by your user, the OpenGL ES program might fail with a Mali Driver error message telling you that you do not have the right GPU. It's a permission issue in that case. If you don't have a /dev/mali0, it's a driver integration issue however. No Vulkan support for the moment, however. The ARM guys do not seem to have released any driver for the Firefly RK3399. (´・ω・` ) I just did some simple test and didn't know if it was in a correct way. As I was using the rk-rootfs-build scripts, I assumed that the packs were working. Unfortunately, no time for further test. Take a look at the deb packs here: https://github.com/rockchip-linux/rk-rootfs-build/tree/master/packages/arm64/libmali And actually ARM guys won't do anything for Firefly RK3399. Rockchip will, but only in there official repo. Screenshot of es2gears test.
nightseas Posted August 4, 2017 Author Posted August 4, 2017 Now the bootloader and kernel built with Armbian scripts are running. But there's a strange issue, in which the ramdisk system can not mount rootfs partition on SD card (SanDisk Ultra 8GB Class10 Micro SD): Begin: Loading essential drivers ... done. Begin: Running /scripts/init-premount ... done. Begin: Mounting root file system ... Begin: Running /scripts/local-top ... done. Begin: Running /scripts/local-premount ... done. Begin: Will now check root file system ... fsck from util-linux 2.27.1 [/sbin/fsck.ext4 (1) -- /dev/mmcblk0p1] fsck.ext4 -a -C0 /dev/mmcblk0p1 rootfs: recovering journal rootfs: clean, 145627/219024 files, 636583/875520 blocks done. mount: mounting /dev/mmcblk0p1 on /root failed: No such device done. Begin: Running /scripts/local-bottom ... done. Begin: Running /scripts/init-bottom ... mount: mounting /dev on /root/dev failed: No such file or directory done. mount: mounting /run on /root/run failed: No such file or directory run-init: current directory on the same filesystem as the root: error 0 Target filesystem doesn't have requested /sbin/init. run-init: current directory on the same filesystem as the root: error 0 run-init: current directory on the same filesystem as the root: error 0 run-init: current directory on the same filesystem as the root: error 0 run-init: current directory on the same filesystem as the root: error 0 run-init: current directory on the same filesystem as the root: error 0 No init found. Try passing init= bootarg. BusyBox v1.22.1 (Ubuntu 1:1.22.0-15ubuntu1) built-in shell (ash) Enter 'help' for a list of built-in commands. (initramfs) But the fsck.ext4 successfully operated the partition, and in the ramdisk system I was able to mount the system manually. (initramfs) ls /dev/ ttyFIQ0 tty24 fuse network_latency rfkill tty25 vpu_service network_throughput mem tty26 rkvdec memory_bandwidth null tty27 ptmx mali0 port tty28 ttyS1 mmcblk1 zero tty29 ttyS3 mmcblk1rpmb full tty30 ttyS0 mmcblk1boot1 random tty31 ttyS2 mmcblk1boot0 urandom tty32 hwrng usbmon3 kmsg tty33 ram0 usbmon4 tty tty34 loop-control usbmon5 console tty35 loop0 usbmon6 tty0 tty36 loop1 input vcs tty37 loop2 i2c-9 vcsa tty38 loop3 hdmi_hdcp1x vcs1 tty39 loop4 fb0 vcsa1 tty40 loop5 vendor_storage tty1 tty41 loop6 dri tty2 tty42 loop7 pts tty3 tty43 zram0 fd tty4 tty44 net stdin tty5 tty45 ppp stdout tty6 tty46 usbmon0 stderr tty7 tty47 usbmon1 char tty8 tty48 bus block tty9 tty49 usbmon2 disk tty10 tty50 uinput rtc tty11 tty51 i2c-0 vcs2 tty12 tty52 rtc0 vcsa2 tty13 tty53 i2c-1 vcs3 tty14 tty54 i2c-4 vcsa3 tty15 tty55 mapper vcs4 tty16 tty56 vhci vcsa4 tty17 tty57 uhid vcs5 tty18 tty58 ashmem vcsa5 tty19 tty59 iio:device0 vcs6 tty20 tty60 snd vcsa6 tty21 tty61 mmcblk0 tty22 tty62 mmcblk0p1 tty23 tty63 cpu_dma_latency (initramfs) [ 27.588426] random: nonblocking pool is initialized (initramfs) (initramfs) (initramfs) mount -t ext4 /dev/mmcblk0p1 /root [ 43.477832] EXT4-fs (mmcblk0p1): mounted filesystem with writeback data mode. Opts: (null) (initramfs) ls /root/ sys tmp usr lost+found media opt boot home proc selinux run var lib dev mnt bin root etc srv sbin The boot scripts and full log are attached. Just ignore the eDP and SDIO WiFi errors, because they are not connected. I guess I'll need to check the kernel config (also attached) then.
Igor Posted August 4, 2017 Posted August 4, 2017 Which patches do you use when building? Make sure you include at least those two: https://github.com/armbian/build/blob/master/patch/kernel/rockchip-default/packaging-4.x-with-postinstall-scripts.patch https://github.com/armbian/build/blob/master/patch/kernel/rockchip-default/bash_to_afterinstall.patch Bug in boot script: setenv bootargs "earlyprintk root=${rootdev} rw console=${console} rootfstype=${rootfstype} loglevel=${verbosity} rootwait ${extraargs}" Change this in config: CONFIG_LOCALVERSION="" # CONFIG_LOCALVERSION_AUTO is not set
Myy Posted August 4, 2017 Posted August 4, 2017 5 hours ago, nightseas said: I just did some simple test and didn't know if it was in a correct way. As I was using the rk-rootfs-build scripts, I assumed that the packs were working. Unfortunately, no time for further test. Take a look at the deb packs here: https://github.com/rockchip-linux/rk-rootfs-build/tree/master/packages/arm64/libmali And actually ARM guys won't do anything for Firefly RK3399. Rockchip will, but only in there official repo. Screenshot of es2gears test. Indeed, wzyy2 tend to prepackage the mali drivers in the rk-rootfs-build repository, but the one in the libmali repository seems to be updated more regularly. That said, if the drivers work, better not update them. Too bad for ARM, since their DDK was more up to date (obviously) and incorporated all the techs possible. That said, they often say they won't provide the drivers for any board and then release them anyway a year later. Now, it looks like the screenshot was eaten by a grue. (´・ω・` ) You might be interested by the new Rockchip wiki here. It has various information on how to compile or install certain Rockchip specific components on common distributions : http://opensource.rock-chips.com/ This might help you if you're stuck with few things.
fossxplorer Posted August 5, 2017 Posted August 5, 2017 Too bad this board isn't available anymore. I also get " is sorry that the baby is temporarily off the shelves, you can stroll around the other baby" like CNX-Soft reported.
nightseas Posted August 6, 2017 Author Posted August 6, 2017 On 8/4/2017 at 5:05 PM, Igor said: Which patches do you use when building? Make sure you include at least those two: https://github.com/armbian/build/blob/master/patch/kernel/rockchip-default/packaging-4.x-with-postinstall-scripts.patch https://github.com/armbian/build/blob/master/patch/kernel/rockchip-default/bash_to_afterinstall.patch Bug in boot script: setenv bootargs "earlyprintk root=${rootdev} rw console=${console} rootfstype=${rootfstype} loglevel=${verbosity} rootwait ${extraargs}" Change this in config: CONFIG_LOCALVERSION="" # CONFIG_LOCALVERSION_AUTO is not set Thanks a lot for the hints. I chose the Marvell 64-bit patch instead of RK3288 to replace compressed zImage with the binary Image. https://github.com/armbian/build/blob/master/patch/kernel/mvebu64-default/packaging-4.x-with-postinstall-scripts.patch Now Armbian 5.32 (Ubuntu 16.04 Desktop) is successfully running on my board. I forked the repo and make my own commit to add the support for RK3399. You can find the changes to original repo here: https://github.com/nightseas/armbian-build/ Know issues: 1. No GPU or VPU pack installed for now; 2. There's a binary tool from RK used to convert u-boot, which can't be packed into the u-boot deb because it's for x86_64. Means you can't update bootloader in a running Armbian. Next step will be GPU and VPU userspace lib porting. Screenshot of desktop, with CPU, temperature, and frequency: 2
nightseas Posted August 6, 2017 Author Posted August 6, 2017 On 8/4/2017 at 5:42 PM, Myy said: Indeed, wzyy2 tend to prepackage the mali drivers in the rk-rootfs-build repository, but the one in the libmali repository seems to be updated more regularly. That said, if the drivers work, better not update them. Too bad for ARM, since their DDK was more up to date (obviously) and incorporated all the techs possible. That said, they often say they won't provide the drivers for any board and then release them anyway a year later. Now, it looks like the screenshot was eaten by a grue. (´・ω・` ) You might be interested by the new Rockchip wiki here. It has various information on how to compile or install certain Rockchip specific components on common distributions : http://opensource.rock-chips.com/ This might help you if you're stuck with few things. Well, the biggest problem is that the mali userspace lib from RK is not compatible to the xserver and desktop display infrastructure installed in ARMBIAN. You have to install RK's xserver packs and also new desktop from some unkonwn source (maybe that's the Debian source under developing or source from RK). I have faced it once when I was customizing a RK3288 board and I'm sure that it'll be still standing in my way. So Let me do some experiments and make as many functions work as I can.
nightseas Posted August 6, 2017 Author Posted August 6, 2017 14 hours ago, fossxplorer said: Too bad this board isn't available anymore. I also get " is sorry that the baby is temporarily off the shelves, you can stroll around the other baby" like CNX-Soft reported. I asked the owner about the stock. There are less than 100 boards in stock and I see the correct quantity in system now. I'm not sure if I can help on ordering boards because of possible issues on China Customs (tax stuff), production quality or transportation damage. Just don't have time to test all the boards.
Myy Posted August 7, 2017 Posted August 7, 2017 On 2017/8/6 at 11:24 AM, nightseas said: Well, the biggest problem is that the mali userspace lib from RK is not compatible to the xserver and desktop display infrastructure installed in ARMBIAN. Indeed. That's the main reason why I'm using DRM/KMS OpenGL ES code with the latest Rockchip libmali (with GBM support) to test the 3D drivers. Now that glmark2 has incorporated my patches, it should be easy to compile and use on any board with working DRM and OpenGL drivers. You might want to give it a try. Compile it like this : ./waf configure --with-flavors=drm-glesv2 ./waf ./waf install # Not mandatory, you can execute the binary from its build folder Then LD_LIBRARY_PATH=/path/to/mali/drivers glmark2-es2-drm You'll certainly be limited to VSync, however the score on the crystal bunny and the (ugly) terrain should provide some good statistics on the GPU.
YelloApple Posted August 11, 2017 Posted August 11, 2017 I have the same board...But it seems the sd card can't work too....Do you have any idea?
Myy Posted August 11, 2017 Posted August 11, 2017 Are you sure that's not a naming issue ? Like using mmcblk0 instead of mmcblk1 or something ? That's a common issue I have when switching between 4.4 and mainline kernels on my RK3288 boards.
YelloApple Posted August 11, 2017 Posted August 11, 2017 41 minutes ago, Myy said: Are you sure that's not a naming issue ? Like using mmcblk0 instead of mmcblk1 or something ? That's a common issue I have when switching between 4.4 and mainline kernels on my RK3288 boards. It seems to be a driver problem...I use "dmesg" command to look for the mmc driver kernel message but find nothing about it...
nightseas Posted August 12, 2017 Author Posted August 12, 2017 On 2017/8/11 at 8:36 AM, YelloApple said: I have the same board...But it seems the sd card can't work too....Do you have any idea? Which kernel are you using? Kernel 4.4 from Rockchip official repo: https://github.com/rockchip-linux/kernel If it's mainline Kernel from kernel.org then which version is it? Other 3rd-party repo? There's a bug in the RK official u-boot, which is caused by missing dwmmc driver and leads to SD card boot failure. https://github.com/nightseas/rockchip-uboot/commit/80b82a7ba084b36c4f4ffbe7b67ffe98ff349f21
YelloApple Posted August 12, 2017 Posted August 12, 2017 (edited) On 12.8.2017 at 2:51 AM, nightseas said: Which kernel are you using? Kernel 4.4 from Rockchip official repo: https://github.com/rockchip-linux/kernel If it's mainline Kernel from kernel.org then which version is it? Other 3rd-party repo? There's a bug in the RK official u-boot, which is caused by missing dwmmc driver and leads to SD card boot failure. https://github.com/nightseas/rockchip-uboot/commit/80b82a7ba084b36c4f4ffbe7b67ffe98ff349f21 I use Rockchip official repo... THX for your uboot patch~ I have used your uboot from https://github.com/nightseas/rockchip-uboot/,but I can't find any SD card driver as well... In uboot ,I have tried the command "mmcinfo" but no any SD card be found. In linux, I have tried the command "ls /dev" but no any SD card node be found. “dmesg” command does not have any information about the SD card node, I have used your uboot source to compile, whether the need to patch the kernel? Update: Spoiler DDR Version 1.08 20170320 In Channel 0: LPDDR3, 800MHz CS = 0 MR0=0x58 MR1=0x58 MR2=0x58 MR3=0x58 MR4=0x2 MR5=0x1 MR6=0x5 MR7=0x0 MR8=0x1F MR9=0x1F MR10=0x1F MR11=0x1F MR12=0x1F MR13=0x1F MR14=0x1F MR15=0x1F MR16=0x1F CS = 1 MR0=0x58 MR1=0x58 MR2=0x58 MR3=0x58 MR4=0x2 MR5=0x1 MR6=0x5 MR7=0x0 MR8=0x1F MR9=0x1F MR10=0x1F MR11=0x1F MR12=0x1F MR13=0x1F MR14=0x1F MR15=0x1F MR16=0x1F Bus Width=32 Col=10 Bank=8 Row=15/15 CS=2 Die Bus-Width=32 Size=2048MB Channel 1: LPDDR3, 800MHz CS = 0 MR0=0x58 MR1=0x58 MR2=0x58 MR3=0x58 MR4=0x2 MR5=0x1 MR6=0x5 MR7=0x0 MR8=0x1F MR9=0x1F MR10=0x1F MR11=0x1F MR12=0x1F MR13=0x1F MR14=0x1F MR15=0x1F MR16=0x1F CS = 1 MR0=0x58 MR1=0x58 MR2=0x58 MR3=0x58 MR4=0x2 MR5=0x1 MR6=0x5 MR7=0x0 MR8=0x1F MR9=0x1F MR10=0x1F MR11=0x1F MR12=0x1F MR13=0x1F MR14=0x1F MR15=0x1F MR16=0x1F Bus Width=32 Col=10 Bank=8 Row=15/15 CS=2 Die Bus-Width=32 Size=2048MB 256B stride ch 0 ddrconfig = 0x101, ddrsize = 0x2020 ch 1 ddrconfig = 0x101, ddrsize = 0x2020 pmugrf_os_reg[2] = 0x3AA0DAA0, stride = 0xD OUT Boot1: 2017-04-07, version: 1.06 CPUId = 0x0 ChipType = 0x10, 1829 SdmmcInit=2 0 BootCapSize=100000 UserCapSize=7456MB FwPartOffset=2000 , 100000 SdmmcInit=0 20 StorageInit ok = 66413 LoadTrustBL No find bl30.bin No find bl32.bin RunBL31 0x10000 NOTICE: BL31: v1.3(debug):55aa0b4 NOTICE: BL31: Built : 16:55:03, Mar 31 2017 INFO: GICv3 with legacy support detected. ARM GICV3 driver initialized in EL3 INFO: boot cpu mask: 0 INFO: plat_rockchip_pmu_init(1089): pd status 3e INFO: BL31: Initializing runtime services INFO: BL31: Preparing for EL3 exit to normal world INFO: Entry point address = 0x200000 INFO: SPSR = 0x3c9 U-Boot 2017.07-g80b82a7 (Aug 12 2017 - 20:48:48 +0800) Model: Rockchip RK3399 Evaluation Board DRAM: 3.9 GiB MMC: dwmmc@fe320000: 1, sdhci@fe330000: 0 *** Warning - bad CRC, using default environment In: serial Out: serial Err: serial Model: Rockchip RK3399 Evaluation Board Net: Error: ethernet@fe300000 address not set. No ethernet found. Hit any key to stop autoboot: 0 => => mmc list dwmmc@fe320000: 1 sdhci@fe330000: 0 (eMMC) => mmc info Device: sdhci@fe330000 Manufacturer ID: 15 OEM: 100 Name: 8GND3 Timing Interface: High Speed Tran Speed: 52000000 Rd Block Len: 512 MMC version 5.0 High Capacity: Yes Capacity: 7.3 GiB Bus Width: 8-bit Erase Group Size: 512 KiB HC WP Group Size: 8 MiB User Capacity: 7.3 GiB WRREL Boot Capacity: 4 MiB ENH RPMB Capacity: 512 KiB ENH => mmc dev 1 Card did not respond to voltage select! mmc_init: -95, time 12 Edited August 14, 2017 by Tido added spoiler
nightseas Posted August 13, 2017 Author Posted August 13, 2017 19 hours ago, YelloApple said: I use Rockchip official repo... THX for your uboot patch~ I have used your uboot from https://github.com/nightseas/rockchip-uboot/,but I can't find any SD card driver as well... In uboot ,I have tried the command "mmcinfo" but no any SD card be found. In linux, I have tried the command "ls /dev" but no any SD card node be found. “dmesg” command does not have any information about the SD card node, I have used your uboot source to compile, whether the need to patch the kernel? Update: ... The driver works fine so that is not a like a software issue. I have tested the SanDisk Ultra 8GB/16GB C10 cards and seen no such error. If the U-Boot shows no mmc1 device then it means the dwmmc controller doesn't detect anything, and I'm afraid there's something wrong with your hardware or SD card.
YelloApple Posted August 13, 2017 Posted August 13, 2017 1 hour ago, nightseas said: The driver works fine so that is not a like a software issue. I have tested the SanDisk Ultra 8GB/16GB C10 cards and seen no such error. If the U-Boot shows no mmc1 device then it means the dwmmc controller doesn't detect anything, and I'm afraid there's something wrong with your hardware or SD card. I have tested the SanDisk Ultra 16GB C10 cards but also have the same error...So sad....
nightseas Posted August 14, 2017 Author Posted August 14, 2017 On 8/8/2017 at 2:02 AM, Myy said: Indeed. That's the main reason why I'm using DRM/KMS OpenGL ES code with the latest Rockchip libmali (with GBM support) to test the 3D drivers. Now that glmark2 has incorporated my patches, it should be easy to compile and use on any board with working DRM and OpenGL drivers. You might want to give it a try. Compile it like this : ./waf configure --with-flavors=drm-glesv2 ./waf ./waf install # Not mandatory, you can execute the binary from its build folder Then LD_LIBRARY_PATH=/path/to/mali/drivers glmark2-es2-drm You'll certainly be limited to VSync, however the score on the crystal bunny and the (ugly) terrain should provide some good statistics on the GPU. I tried to build libmali deb pack for RK3399 but failed. There's no instruction or guide, so I may need to ask Rockchip for help again. Then I manually linked the lib in /usr/lib and destroyed the the desktop GUI. After some experiment the desktop was back but still could not test gles2 as you suggested. The good news is, OpenCL works well. I'll do some OpenCV + OpenCL perf test and put the result here. 1
yasko Posted August 16, 2017 Posted August 16, 2017 @nightseas, This is a good news about OpenCL and RK3399. It's interesting to see your results about OpenCV + OpenCL on RK3399. I have found that ARM released their own Compute library . It supports both NEON and OpenCL and they claim that it has better performance than OpenCV. I am also playing with RK3399 (Yundoo Y8) and was able to compile ARM compute library and I'm planning to do some perf tests.
Andro Posted August 17, 2017 Posted August 17, 2017 Reading the above thread, is there any support for the Videostrong 6 core RK3399 board? Can Armbian run on that system?
nightseas Posted August 17, 2017 Author Posted August 17, 2017 1 hour ago, Andro said: Reading the above thread, is there any support for the Videostrong 6 core RK3399 board? Can Armbian run on that system? I have made the initial support for RK3399. Currently working on GPU stuff. https://github.com/nightseas/armbian-build
Myy Posted August 17, 2017 Posted August 17, 2017 The command to build most of the Rockchip repositories with a debian folder seems to be : dpkg-buildpackage -b -rfakeroot -us -uc That said, I would still highly recommend taking the .so library from their repository, put it in a folder, regenerate the appropriate links quickly and test GLES programs by invoking them like this : LD_LIBRARY_PATH=/path/to/the/library gles-program Installing the new OpenGL ES libraries in /usr/lib could generate serious issues. Testing DRM/KMS OpenGL ES programs, like glmark2, from a real console (CTRL+ALT+F1), or from weston (Wayland), will generally be easier than trying from X11.
nightseas Posted August 18, 2017 Author Posted August 18, 2017 On 8/16/2017 at 2:02 PM, yasko said: @nightseas, This is a good news about OpenCL and RK3399. It's interesting to see your results about OpenCV + OpenCL on RK3399. I have found that ARM released their own Compute library . It supports both NEON and OpenCL and they claim that it has better performance than OpenCV. I am also playing with RK3399 (Yundoo Y8) and was able to compile ARM compute library and I'm planning to do some perf tests. Attached is the full OpenCV 3.3.0 + OpenCL perf test report except video test part, which failed probably because of corrupted GLES. Just for your reference. x86_opencv_perf_reports_20170817.html 1
nightseas Posted August 18, 2017 Author Posted August 18, 2017 14 hours ago, Myy said: The command to build most of the Rockchip repositories with a debian folder seems to be : dpkg-buildpackage -b -rfakeroot -us -uc That said, I would still highly recommend taking the .so library from their repository, put it in a folder, regenerate the appropriate links quickly and test GLES programs by invoking them like this : LD_LIBRARY_PATH=/path/to/the/library gles-program Installing the new OpenGL ES libraries in /usr/lib could generate serious issues. Testing DRM/KMS OpenGL ES programs, like glmark2, from a real console (CTRL+ALT+F1), or from weston (Wayland), will generally be easier than trying from X11. Thanks for the hints. Do I also need to install Rockchip DRM packages while testing glmark2?
Myy Posted August 18, 2017 Posted August 18, 2017 It should not be necessary. The Rockchip DRM package main purpose is to exploit DRI capabilities and the RGA 2D acceleration on Rockchip systems, but the RGA 2D acceleration was deemed inefficient for X11 purposes, and the DRI is not useful when testing DRM capabilities directly from a console.
odlg Posted August 28, 2017 Posted August 28, 2017 Hello I have a Firefly RK3399 board and would like to test your build. Do you have downloadable images? Or do I need to setup a build env. Do you have flashing instructions? BR Ole
nightseas Posted August 30, 2017 Author Posted August 30, 2017 On 2017/8/28 at 8:38 PM, odlg said: Hello I have a Firefly RK3399 board and would like to test your build. Do you have downloadable images? Or do I need to setup a build env. Do you have flashing instructions? BR Ole You should better build your own image, with several configurations changed from rk3399-evb to firefly, such as dtb, uboot config...
nobe Posted August 31, 2017 Posted August 31, 2017 if some of you guys are still interested, the saphirre board is back in stock in the taobao website, but more expensive now (580CNY). 1
Recommended Posts