mzb

Please don't misunderstand my question: I'm not asking for you to "hurry up and deliver the finished product so I can use it". I do know how this game works, and I have seen evidence over the years of the colossal effort you put into this project. You've done a great job, as usual, producing an image *THAT WORKS* (so well) in such a short time ... I'm very grateful. It's because of efforts from you, and others in the Armbian project, that I get to use Debian on devices that probably wouldn't be supported otherwise. I'm simply asking what is preventing us/me/you/everyone from using Bullseye with the kernel you've used for the Sid image? Is there something in particular that's in the debian-sid (or armbian-sid?) userland that is required? There are a number of things I'd like to try out on this board (headless), and running Debian-stable makes that a lot easier (eg: PHP) without having to containerise everything.

I chose a 1TB Samsung MZ-V7S1T0BW 970 EVO Plus and one of the Radxa NVME heatsinks with 2x clear bands holding it on. No paste, no sticky pad, didn't remove the sticker (sue me ). It idles along at 35°C in the acrylic case with the Radxa fan (which is on the other side of the board, obviously!), while the CPU is also around 35°C (ambient is currently 25°C). During the fio tests below (after running several times) the drive's internal temperature peaked at 69°C, and the external temp at 59°C. It seemed like good value (and speed) for money at AU$179 ( ~ US$115? ) on Amazon(.au). root@rock5:~# hdparm -tT --direct /dev/nvme0n1 /dev/nvme0n1: Timing O_DIRECT cached reads: 3514 MB in 2.00 seconds = 1758.52 MB/sec Timing O_DIRECT disk reads: 5472 MB in 3.00 seconds = 1823.70 MB/sec root@rock5:~# fio --filename=/mnt/nvme/test.bin --direct=1 --rw=write --ioengine=libaio --bs=2m --iodepth=64 --size=10G --numjobs=1 --runtime=60 --time_base=1 --group_reporting --name=test-seq-write test-seq-write: (g=0): rw=write, bs=(R) 2048KiB-2048KiB, (W) 2048KiB-2048KiB, (T) 2048KiB-2048KiB, ioengine=libaio, iodepth=64 fio-3.32 Starting 1 process test-seq-write: Laying out IO file (1 file / 10240MiB) Jobs: 1 (f=1): [W(1)][100.0%][w=3064MiB/s][w=1532 IOPS][eta 00m:00s] test-seq-write: (groupid=0, jobs=1): err= 0: pid=17830: Fri Nov 11 12:22:56 2022 write: IOPS=1527, BW=3054MiB/s (3203MB/s)(179GiB/60041msec); 0 zone resets slat (usec): min=196, max=4245, avg=463.28, stdev=87.74 clat (usec): min=32206, max=80119, avg=41428.38, stdev=2070.37 lat (usec): min=32694, max=80591, avg=41891.66, stdev=2063.49 clat percentiles (usec): | 1.00th=[39584], 5.00th=[40109], 10.00th=[40109], 20.00th=[40109], | 30.00th=[40109], 40.00th=[40633], 50.00th=[40633], 60.00th=[40633], | 70.00th=[40633], 80.00th=[43779], 90.00th=[45351], 95.00th=[45351], | 99.00th=[45876], 99.50th=[46400], 99.90th=[49021], 99.95th=[51643], | 99.99th=[73925] bw ( MiB/s): min= 2992, max= 3096, per=100.00%, avg=3057.25, stdev=18.13, samples=119 iops : min= 1496, max= 1548, avg=1528.59, stdev= 9.04, samples=119 lat (msec) : 50=99.93%, 100=0.07% cpu : usr=14.26%, sys=59.82%, ctx=65623, majf=0, minf=11 IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=0.1%, >=64=99.9% submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0% complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.1%, >=64=0.0% issued rwts: total=0,91694,0,0 short=0,0,0,0 dropped=0,0,0,0 latency : target=0, window=0, percentile=100.00%, depth=64 Run status group 0 (all jobs): WRITE: bw=3054MiB/s (3203MB/s), 3054MiB/s-3054MiB/s (3203MB/s-3203MB/s), io=179GiB (192GB), run=60041-60041msec Disk stats (read/write): nvme0n1: ios=0/374636, merge=0/0, ticks=0/15378262, in_queue=15378262, util=99.85% root@rock5:~# fio --filename=/mnt/nvme/test.bin --direct=1 --rw=read --ioengine=libaio --bs=2m --iodepth=64 --size=10G --numjobs=1 --runtime=60 --time_base=1 --group_reporting --name=test-seq-read test-seq-read: (g=0): rw=read, bs=(R) 2048KiB-2048KiB, (W) 2048KiB-2048KiB, (T) 2048KiB-2048KiB, ioengine=libaio, iodepth=64 fio-3.32 Starting 1 process Jobs: 1 (f=1): [R(1)][100.0%][r=3145MiB/s][r=1572 IOPS][eta 00m:00s] test-seq-read: (groupid=0, jobs=1): err= 0: pid=17996: Fri Nov 11 12:24:25 2022 read: IOPS=1568, BW=3137MiB/s (3289MB/s)(184GiB/60040msec) slat (usec): min=130, max=6520, avg=402.38, stdev=133.63 clat (msec): min=18, max=159, avg=40.33, stdev= 2.01 lat (msec): min=18, max=163, avg=40.73, stdev= 2.04 clat percentiles (msec): | 1.00th=[ 37], 5.00th=[ 40], 10.00th=[ 40], 20.00th=[ 40], | 30.00th=[ 41], 40.00th=[ 41], 50.00th=[ 41], 60.00th=[ 41], | 70.00th=[ 41], 80.00th=[ 41], 90.00th=[ 42], 95.00th=[ 42], | 99.00th=[ 44], 99.50th=[ 46], 99.90th=[ 52], 99.95th=[ 72], | 99.99th=[ 138] bw ( MiB/s): min= 2327, max= 3164, per=100.00%, avg=3140.59, stdev=75.34, samples=119 iops : min= 1163, max= 1582, avg=1570.24, stdev=37.73, samples=119 lat (msec) : 20=0.05%, 50=99.83%, 100=0.09%, 250=0.02% cpu : usr=1.91%, sys=67.30%, ctx=54414, majf=0, minf=32787 IO depths : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=0.1%, >=64=99.9% submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0% complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.1%, >=64=0.0% issued rwts: total=94176,0,0,0 short=0,0,0,0 dropped=0,0,0,0 latency : target=0, window=0, percentile=100.00%, depth=64 Run status group 0 (all jobs): READ: bw=3137MiB/s (3289MB/s), 3137MiB/s-3137MiB/s (3289MB/s-3289MB/s), io=184GiB (198GB), run=60040-60040msec Disk stats (read/write): nvme0n1: ios=384613/0, merge=11752/0, ticks=15331918/0, in_queue=15331918, util=100.00%

I'm using a Baseus 65W GaN3 Pro CCGAN65-1ACC bought off eBay root@rock5:~# uname -a; cat /etc/issue; sensors | grep -A3 ^tcpm Linux rock5 5.10.72-rockchip-rk3588 #trunk.0080 SMP Sun Oct 23 18:45:28 UTC 2022 aarch64 GNU/Linux Armbian 22.11.0-trunk Sid \l tcpm_source_psy_4_0022-i2c-4-22 Adapter: rk3x-i2c in0: 20.00 V (min = +20.00 V, max = +20.00 V) curr1: 2.25 A (max = +2.25 A) Do this as root: update-alternatives --set iptables /usr/sbin/iptables-legacy update-alternatives --set ip6tables /usr/sbin/ip6tables-legacy Thanks @Igor, nice job. I notice that linux-image-legacy-rockchip-rk3588 is also listed in Packages of armbian-bullseye. What needs fixing to get rock5 to run -stable, or something less explosive than sid?