dcsouthwick
commented
2 years ago Still digging in hopes of some functionality with anything other than 8_5_2. I find another user's report of the same issue, so following his hack, I can get one step further, but it doesn't seem clear where this requirement is coming from yet...
editing runme.sh I add a dirty hack to force the missing file to exist & add it to the image:
cp $ROOTDIR/images/tmp/boot/fsl-lx2160a-clearfog-cx.dtb $ROOTDIR/images/tmp/fsl-layerscape-lx2160a.dtb
... add to e2cp lineCool. That gets one step closer... but...
Scanning nvme 0:1...
Found /extlinux/extlinux.conf
Retrieving file: /extlinux/extlinux.conf
331 bytes read in 0 ms
1: primary kernel
Retrieving file: /boot/Image
41951744 bytes read in 27 ms (1.4 GiB/s)
append: console=ttyAMA0,115200 earlycon=pl011,mmio32,0x21c0000 default_hugepagesz=1024m hugepagesz=1024m hugepages=2 pci=pcie_bus_perf root=PARTUUID=30303030-01 rw rootwait
Retrieving file: /boot/fsl-layerscape-lx2160a.dtb
42467 bytes read in 1 ms (40.5 MiB/s)
Moving Image from 0x81100000 to 0x81200000, end=83b10000
## Flattened Device Tree blob at 81000000
Booting using the fdt blob at 0x81000000
Loading Device Tree to 000000009fff2000, end 000000009ffff5e2 ... OK
Releasing fan controller full speed gpio
fsl-mc: Deploying data path layout ... WARNING: Firmware returned an error (GSR: 0x3f)
Starting kernel ...early signs that several parts of the build script do not support serdes other than 8_5_2
[ 4.853491] ptp_qoriq: device tree node missing required elements, try automatic configuration
....
[ 5.129897] libphy: SFP I2C Bus: probed
[ 5.133781] sfp sfp-0: Host maximum power 2.0W
[ 5.138349] sfp sfp-0: No tx_disable pin: SFP modules will always be emitting.
[ 5.146109] libphy: SFP I2C Bus: probed
[ 5.149990] sfp sfp-1: Host maximum power 2.0W
[ 5.154541] sfp sfp-1: No tx_disable pin: SFP modules will always be emitting.
[ 5.162283] libphy: SFP I2C Bus: probed
[ 5.166156] sfp sfp-2: Host maximum power 2.0W
[ 5.170708] sfp sfp-2: No tx_disable pin: SFP modules will always be emitting.
[ 5.178441] libphy: SFP I2C Bus: probed
[ 5.182312] sfp sfp-3: Host maximum power 2.0W
[ 5.186872] sfp sfp-3: No tx_disable pin: SFP modules will always be emitting.
[ 5.429078] fsl_dpaa2_eth dpni.0 (unnamed net_device) (uninitialized): PHY [0x0000000008b96000:01] driver [Qualcomm Atheros AR8035] (irq=POLL)
[ 5.442941] fsl_dpaa2_eth dpni.0: Probed interface eth0
[ 5.450015] input: gpio-keys as /devices/platform/gpio-keys/input/input0
[ 5.457876] ALSA device list:
[ 5.460874] No soundcards found.
[ 5.482146] EXT4-fs (nvme0n1p1): mounted filesystem with ordered data mode. Opts: (null)
[ 5.490275] VFS: Mounted root (ext4 filesystem) on device 259:1.
[ 5.496744] devtmpfs: mounted
[ 5.501244] Freeing unused kernel memory: 6720K
[ 5.505848] Run /sbin/init as init process
[ 5.510281] Run /etc/init as init process
[ 5.514428] Run /bin/init as init process
[ 5.518597] Run /bin/sh as init process
/bin/sh: 0: can't access tty; job control turned off
#so, 14_5_2 build still thinks there are 4x sfp ports to bring up (despite SD1 config for a single QSFP)... as well as apparently missing initial device nodes.
Does anyone else have this issue? Thoughts?
Trying the last few commits of this repo to build images that support the clearfog variant of your board, but none are bootable. So far, I've tested SD1= 13/14/17/21 and all fail with the same error:
(boot logs from 14_5_2 build, which you seem to mention is tested in other tickets)
could you publish a known working image that demonstrates QSFP support? If the above issue is related to my build env, it would be helpful to have a working starting point...