<Kwiboo>
have had issue with similar when using full DEBUG build, have not sent any patch because I did not know if it was correct or may have other strange side effects
<Kwiboo>
have also seen issues with asserts in DM core because of vbe simple, however, think that issue may have been fixed already
<qschulz>
Kwiboo: with the suggested patch I reach the CLI
<qschulz>
s/patch/change/
frieder has quit [Remote host closed the connection]
goliath has quit [Quit: SIGSEGV]
ikarso has joined #u-boot
persmule has joined #u-boot
mmu_man has quit [Quit: reboot]
mmu_man has joined #u-boot
<sjg1>
qschulz: Enabling DEBUG should not cause a failure, unless there is actually an underlying issue to fix. Perhaps we need to enable DEBUG in CI in some places?
<qschulz>
sjg1: I guess we could have some sandbox test check an assert that is known to fail so that when we build with DEBUG it still passes, but with -DASSERT or however we want to call the symbol, it does fail?
<qschulz>
Kwiboo: select DM_WARN or increase the LOG_MAX_LEVEL value to 8+ to see the logs
ukky_ has quit [Ping timeout: 240 seconds]
ukky has quit [Ping timeout: 256 seconds]
ukky has joined #u-boot
ukky_ has joined #u-boot
Clamor has quit [Ping timeout: 252 seconds]
Clamor has joined #u-boot
enok has quit [Ping timeout: 240 seconds]
ikarso has quit [Quit: Connection closed for inactivity]
<pbrobinson>
hey Tartarus, got a q about upstream DT pieces, how does that work for tar file releases? Should we expect the tar files to be complete or do we need to pull DTs from elsewhere?
<Tartarus>
pbrobinson: Uh, good question, did I mess up the tarballs?
<marex>
there are tarball of U-Boot releases? I thought after the XZ f....p everyone started pulling stuff from git
<marex>
most certainly distros like centos or such
<pbrobinson>
Tartarus: I think I'm probably missing something TBH
ladis has quit [Quit: Leaving]
iprusov has quit [Quit: WeeChat 3.5]
<Tartarus>
pbrobinson: Well, tbh I did not think about if I need new flags to have subtrees included
<Tartarus>
pbrobinson: OK, they should be included, yes
<Tartarus>
I ran the command that's part of my release script and see "u-boot-HEAD/dts/upstream/src/arm/allwinner/sun8i-a23-q8-tablet.dts"
<pbrobinson>
Tartarus: I see them now I'm digging in ./dts/upstream/src/arm64/rockchip/rk3566-quartz64-b.dts, but I'm not seeing an output dtb when I build it
<pbrobinson>
but TBH I think the board has other config issues so I think I may go an experiment with a different one
<Tartarus>
OK
joeskb7 has quit [Ping timeout: 252 seconds]
Jones42 has joined #u-boot
goliath has joined #u-boot
joeskb7 has joined #u-boot
Jones42 has quit [Ping timeout: 268 seconds]
Clamor has quit [Read error: Connection reset by peer]
ikarso has joined #u-boot
enok has joined #u-boot
<marex>
pbrobinson: why not pull the sources from git ?
grs has quit [Remote host closed the connection]
grs has joined #u-boot
naoki has joined #u-boot
<Tartarus>
There's good reasons too for distributions to use archives not SCM, SCM is a big download sometimes and always getting worsee
<Tartarus>
I'm sure everyone groans when fetching the kernel via git in OE :)
<marex>
Tartarus: have you heard about DLDIR yet ?
<marex>
:)
<marex>
Tartarus: just do incremental fetch on every release, that's fast
<marex>
also git.k.o has many mirrors, so that's fast to fetch too
<quinq>
That doesn't mean very much without an actual context
<Tartarus>
linux-yocto is the bane of my builds every time I start poking something for a new customer and yes, I have a local mirror and archive
<Tartarus>
It's between that or one of the other big git pulls as to what will take the longest with --runall=fetch
<quinq>
At work some team started using yocto
<marex>
Tartarus: and you do that exactly ... once ...
<marex>
Tartarus:
<quinq>
They need like GB of storage space just for that
<marex>
yes, you need a lot of space to build the entire OS
<quinq>
And that's only for base repo, then you start building and the monster unleashes its full wrath
<quinq>
I don't understand why people cope with this kind of thing
<Tartarus>
marex: and it sucks so bad every time, compared with 95% of the packages which are fetched quickly
<marex>
Tartarus: why do you even keep fetching the stuff repeatedly ?
<marex>
Tartarus: also, are you sure the other stuff that is fetched quietly does match what's in the git repo (cfr XZ mess) ?
<Tartarus>
The cadence of when I need to start a new build for a new customer ends up such that yes, linux-yocto and glibc/binutils/etc end up being noticeable pain points compared with everything else, even when not cache hits
<Tartarus>
So I can see how if everything defaulted to git clones instead it would be even worse
<Tartarus>
And yes, the whole XZ mess shows a number of issues
<marex>
... solved by using git clones
<Tartarus>
No? That was also "re-generate your autoconf files" too
<marex>
Tartarus: and how can you be sure the whatever autoreeverything.sh script isn't malicious ?
<marex>
(or whatever is in the auto* files is not somehow problematic?)
<Tartarus>
I mean, OE would have been fine with xz as an archive even if it got the trojan one because autotools.bbclass ?
Hypfer has quit [Quit: Ping timeout (120 seconds)]
<marex>
Tartarus: the trojan wasnt in git in the first place, so pulling from git would avoid that problem altogether
<Tartarus>
And fedora/debian/etc hitting git repos even more so than file archives would be a bigger impact on self-hosted projects
Hypfer has joined #u-boot
<Tartarus>
Yes, a problem with xz was including things in the archives (generated files) that aren't in git. If they didn't do that, it wouldn't have been a vector
<marex>
Tartarus: surely they have local code mirror, like salsa. for debian ?
* cambrian_invader
would take a tarball over git any day
jclsn has quit [Quit: WeeChat 4.2.2]
mmu_man has joined #u-boot
Hypfer0 has joined #u-boot
Hypfer has quit [Ping timeout: 246 seconds]
Hypfer0 is now known as Hypfer
slobodan__ has quit [Ping timeout: 268 seconds]
mmu_man has quit [Ping timeout: 272 seconds]
mmu_man has joined #u-boot
Shiz has quit [Ping timeout: 272 seconds]
gsz has quit [Ping timeout: 240 seconds]
goliath has quit [Quit: SIGSEGV]
slobodan__ has joined #u-boot
enok has quit [Ping timeout: 240 seconds]
Danct12 has quit [Remote host closed the connection]