Re: Gitlab runner timeout on evb-ast2600 ?

["Cédric Le Goater" <clg@kaod.org>]

On 4/11/25 17:01, Tom Rini wrote:
> On Fri, Apr 11, 2025 at 09:23:59AM +0200, Cédric Le Goater wrote:
>> Hi,
>>
>> On 4/11/25 01:14, Tom Rini wrote:
>>> On Fri, Apr 11, 2025 at 08:07:15AM +0930, Joel Stanley wrote:
>>>> On Mon, 7 Apr 2025 at 23:25, Eugen Hristev <eugen.hristev@linaro.org> wrote:
>>>>>
>>>>> Hi everyone,
>>>>>
>>>>> I keep getting an error on running the CI loop on -next on my tree :
>>>>>
>>>>> https://source.denx.de/u-boot/custodians/u-boot-at91/-/jobs/1090517
>>>>
>>>> Thanks for the report.
>>>>
>>>>>
>>>>> ! _pytest.outcomes.Exit: Lab timeout: Marking connection bad - no other
>>>>> tests will run !
>>>>>
>>>>> Any clue ? It's been like this for few days. It doesn't look like a
>>>>> faulty patch.
>>>>
>>>> I can't see that link you sent me, it 403s.
>>>>
>>>> Do you have any other traces or information to share?
>>>
>>> It's been intermittent recently. I was able to put qemu + gdb together
>>> and see it's one of the early initcalls that fails and hang()s but
>>> didn't see why. You should be able to use commit 1f0281294d8b ("Merge
>>> patch series "Annotate switch/case fallthrough cases"") as a starting
>>> point of failure.
>>>
>>
>> Do you have logs ?
>>
>> Things look ok from the QEMU side (latest 10.0.0-rc and 9.1) :
>>      uboot_build=/path/to/u-boot/
>>      flashsize=65536
>>      flash=./flash-ast2600-evb-uboot
>>      dd of=$flash bs=1K if=/dev/zero count=$flashsize
>>      dd of=$flash conv=notrunc bs=1K if=$uboot_build/spl/u-boot-spl.bin
>>      dd of=$flash conv=notrunc bs=1K if=$uboot_build/u-boot.img seek=64
>>      $ qemu-system-arm -M ast2600-evb -net nic,netdev=net0 -netdev user,id=net0 -nographic -snapshot -serial mon:stdio   -drive file=flash-ast2600-evb-uboot,format=raw,if=mtd
>>      qemu-system-arm: warning: nic ftgmac100.1 has no peer
>>      qemu-system-arm: warning: nic ftgmac100.2 has no peer
>>      qemu-system-arm: warning: nic ftgmac100.3 has no peer
>>      U-Boot SPL 2025.04-00921-gcb7555e93075 (Apr 11 2025 - 09:08:40 +0200)
>>      Trying to boot from RAM
>>      ## Checking hash(es) for config conf-1 ... OK
>>      ## Checking hash(es) for Image firmware-1 ... crc32+ OK
>>      ## Checking hash(es) for Image fdt-1 ... crc32+ OK
>>      U-Boot 2025.04-00921-gcb7555e93075 (Apr 11 2025 - 09:08:40 +0200)
> 
> Please note that it's inconsistent about what commit fails. I would
> expect 1f0281294d8b in U-Boot to fail but I don't know about
> cb7555e93075.
> 
> Also can you please check QEMU v8.2.0? That's what we're on, and I can
> try and test 9.1.3 or 10.0.0-rc and see if it's a QEMU problem instead.
> Thanks!

Same result :

qemu-system-arm -M ast2600-evb -net nic,netdev=net0 -netdev user,id=net0 -nographic -snapshot -serial mon:stdio   -drive file=flash-ast2600-evb-uboot,format=raw,if=mtd
qemu-system-arm: warning: nic ftgmac100.1 has no peer
qemu-system-arm: warning: nic ftgmac100.2 has no peer
qemu-system-arm: warning: nic ftgmac100.3 has no peer

U-Boot SPL 2025.04-00881-g1f0281294d8b (Apr 11 2025 - 17:23:25 +0200)
Trying to boot from RAM
## Checking hash(es) for config conf-1 ... OK
## Checking hash(es) for Image firmware-1 ... crc32+ OK
## Checking hash(es) for Image fdt-1 ... crc32+ OK


U-Boot 2025.04-00881-g1f0281294d8b (Apr 11 2025 - 17:23:25 +0200)

Model: Aspeed BMC
DRAM:  already initialized, 1008 MiB (capacity:1024 MiB, VGA:64 MiB), ECC off
Core:  266 devices, 24 uclasses, devicetree: separate
WDT:   Started watchdog@1e785000 with servicing every 1000ms (60s timeout)
MMC:   sdhci@1e750100: 0
Loading Environment from SPIFlash... SF: Detected mx25u51245g with page size 256 Bytes, erase size 64 KiB, total 64 MiB
*** Warning - bad CRC, using default environment

In:    serial@1e784000
Out:   serial@1e784000
Err:   serial@1e784000
Net:   eth0: ftgmac@1e660000, eth1: ftgmac@1e680000, eth2: ftgmac@1e670000, eth3: ftgmac@1e690000
Hit any key to stop autoboot:  0
libfdt fdt_check_header(): FDT_ERR_BADMAGIC
Error loading kernel FIT image
=> QEMU 8.2.10 monitor - type 'help' for more information
(qemu) q


What about the compiler ?


Thanks,

C.