camus has quit [Read error: Connection reset by peer]
camus1 is now known as camus
jpuhlman__ has joined #yocto
jpuhlman_ has quit [Ping timeout: 252 seconds]
gioyik_ has quit [Ping timeout: 244 seconds]
gioyik_ has joined #yocto
roussinm1 has quit [Quit: WeeChat 3.3-dev]
camus1 has joined #yocto
camus has quit [Read error: Connection reset by peer]
camus1 is now known as camus
Tokamak has quit [Ping timeout: 252 seconds]
Tokamak has joined #yocto
sakoman has quit [Quit: Leaving.]
gioyik_ is now known as gioyik
amitk has joined #yocto
Tokamak has quit [Ping timeout: 252 seconds]
camus1 has joined #yocto
camus has quit [Read error: Connection reset by peer]
camus1 is now known as camus
ant__ has quit [Quit: Leaving]
rber|res has joined #yocto
<rber|res>
Hi
<rber|res>
Can someone give me hint what could possibly modify a binary in a package when it's installed on the host while baking an image? In the package it's correct.
<rber|res>
@LetoThe2nd - Ned wirkli - OeSi training ;)
<LetoThe2nd>
rber|res != reliableembedded
<LetoThe2nd>
rber|res: sog eana, pscht, geheimnis, aber eigenlich waars vui bessa sie dadn windous ei-ou-tii nemma.
frieder has joined #yocto
rob_w has joined #yocto
mckoan|away is now known as mckoan
<mckoan>
good morning
<LetoThe2nd>
mckoan: howdy
OnkelUlla has quit [Ping timeout: 240 seconds]
zpfvo has joined #yocto
OnkelUlla has joined #yocto
<mcfrisk>
FYI: stress-ng in poky has diverged between dunfell and master. the point release update patches differ
florian has joined #yocto
WadeBerrier[m] has quit [Ping timeout: 246 seconds]
timbert[m] has quit [Ping timeout: 246 seconds]
thekappe has quit [Quit: WeeChat 1.9.1]
prabhakarlad has joined #yocto
<qschulz>
morning folks
hpsy has joined #yocto
zyga has joined #yocto
crawler has joined #yocto
Vonter has quit [Ping timeout: 258 seconds]
Vonter has joined #yocto
<perdmann_>
Hi, my board hangs at "Starting Kernel..." after upgrading to a newer kernel version. Is there an easy way to debug this? i sadly dont have a hardware debugger
<mcfrisk>
perdmann_: serial console, that's what one needs to debug bootloader and earl kernel things
<mcfrisk>
the HW support needs to be there and it needs to be configure for SW pieces
Schlumpf has joined #yocto
<RP>
JPEW: objcopy in package.bbclass
<perdmann_>
mcfrisk: is there an option for more outputs via serial console?
<qschulz>
perdmann_: earlyprintk support in the kernel probably yes, you need to enable it, configure it correctly and setup the bootloader to enable it (pass it to the kernel command line AFAIR)
<qschulz>
perdmann_: also, I had a similar issue with incompatible TF-A with a newer kernel. I updated TF-A too and then it worked better :) Don't know if it applies to your case, but just FYI
MysticMice has quit [Quit: Leaving]
<perdmann_>
qschulz: Trusted Firmware?
<qschulz>
perdmann_: yes
<qschulz>
4.14 kernel on i.mx8mm with 2.0 TF-A, 5.4 kernel with 2.2 TF-A (IIRC, don't remember exactly the version of the TF-A)
<qschulz>
4.14 with 2.2 TF-A and 5.4 with 2.0 TF-A didn't work
<qschulz>
kernel and TF-A coming from NXP repos
thekappe has joined #yocto
zeddii has quit [Ping timeout: 268 seconds]
zeddii has joined #yocto
florian_kc has joined #yocto
gioyik has quit [Quit: WeeChat 3.1]
zeddii has quit [Ping timeout: 252 seconds]
camus1 has joined #yocto
camus has quit [Read error: Connection reset by peer]
camus1 is now known as camus
bps has quit [Ping timeout: 252 seconds]
zeddii has joined #yocto
zeddii has quit [Ping timeout: 246 seconds]
bps has joined #yocto
bps has joined #yocto
paulg has joined #yocto
zeddii has joined #yocto
xmn has quit [Quit: ZZZzzz…]
WadeBerrier[m] has joined #yocto
leon-anavi has joined #yocto
zeddii has quit [Ping timeout: 252 seconds]
crawler has quit [Ping timeout: 252 seconds]
zeddii has joined #yocto
crawler has joined #yocto
crawler has quit [Excess Flood]
crawler has joined #yocto
crawler has quit [Remote host closed the connection]
xmn has joined #yocto
zeddii has quit [Ping timeout: 252 seconds]
camus has quit [Ping timeout: 252 seconds]
camus has joined #yocto
<Schlumpf>
Hi, is there any chance to tell bitbake to build only one specific target of a cmake project?
florian_kc has quit [Ping timeout: 272 seconds]
camus has quit [Remote host closed the connection]
<perdmann_>
Hi, i try to fetch some files from an SVN, but i always get this error: bb.data_smart.ExpansionError: Failure expanding variable SRCPV, expression was ${@bb.fetch2.get_srcrev(d)} which triggered exception AttributeError: 'NoneType' object has no attribute 'group'
<yates>
but the library seems like a great idea. 64 bit integers sound massive to me. i worked with 16-bit and 32-bit in TMS320C54x fixed-point DSP assembly back in the day.
zeddii is now known as zedd
Schlumpf has quit [Remote host closed the connection]
Schlumpf has joined #yocto
<rfs613>
yates: fixed point support got added to gcc, so the need for the library kind of went away. I vaguely seem to recall that Nico tweaked this library and that possibly that's what gcc uses.
<yates>
rfs613: interesting history.
yumasi has joined #yocto
<rfs613>
a quick grep of the gcc source tree suggests my memory may be incorrect...
<yates>
i'm trying to determine what would cause a QA error when PACKAGE_NO_GCONV = 0 but not when it is 1
<yates>
one thing is if there are other directories under ${libdir}/ than gconv
<rburton>
JPEW: do you think we still care about running reproducible selftest for more architectures?
<yates>
even if they were empty, would this QA error be triggered?
<JPEW>
rburton: Ya
<JPEW>
rburton: ARM would be the next one IMHO ;)
<rburton>
the current test just uses whatever the machine is
<yates>
s/QA error/QA issue/
<rburton>
JPEW: i guess the easy solution is to just hardcode qemux86-64 and qemuarm64
<yumasi>
Hi! I am trying to build a signed fitImage. Everything goes well, I have a signature in the final image, put u-boot fails on checking the sha256 of the configuration section of the image. Looking up with mkimage -l, I see that the "hash value" is indeed "undefined". Anyone already encountered this error before or know how to fix it?
<JPEW>
rburton: I'm confused, are you saying that it's already testing qemuarm64?
<rburton>
it tests whatever the current machine is
<rburton>
which defaults to qemux86-64
<JPEW>
rburton: Ah right. So does the AB only run the test on qemux86-64 then?
<rburton>
yes
<rburton>
We could punt this into the 'run selftest for arm64' bug
<rburton>
actually might be the best idea
<rburton>
I should retest that...
<JPEW>
rburton: Works for me :)
zedd has quit [Remote host closed the connection]
zeddii has joined #yocto
<perdmann_>
Is there any way to escape charecters in variables? \ is not working for me
bps has joined #yocto
bps has joined #yocto
timbert[m] has joined #yocto
<yates>
packaging question: will a QA issue be flagged if empty subdirectories in the source exist that were not created in the destination ($(D))?
<yates>
by inference i have determined the answer to this question is "yes" but I want to double-check
<smurray>
yates: packaging only looks in ${D}, so I'm pretty sure the answer is no
<yates>
smurray: it must be looking at source directories/files too in order to do this QA issue check, right?
<yates>
how could that possibly be true? what is it comparing to to determine there is an issue?
<qschulz>
yates: the check is done between what's in ${D} and what's in FILES_*
<smurray>
yates: that's telling you the directory exists in ${D}, but no FILES_*
<qschulz>
if after everything's been packaged, there are still files, you get a QA Issue
<smurray>
qschulz: sniped me ;)
<yates>
ah, ok.
<rburton>
the assumption is that everything that appeared in do_install should be in a package
<rburton>
so if something gets installed but not packaged, you get that warning
<qschulz>
and IIRC, empty directories still need to be packaged. So if your target always create some directory but not always put things in it, you might need to add /some/dir/ in addition to /some/dir/my-file in FILES_*
<qschulz>
your makefile/cmake/meson/whatever target*
<rburton>
correct, empty directories are still things that need to be packaged
<rburton>
either package it, or don't install it
<yates>
rburton: ok you seriously confused me with "if something gets installed but not packaged". I thought installing IS packaging. Or at least the first step, i.e., copying stuff to a staging area, which then subsequently gets packaged
<rburton>
yeah we went through this last week
<rburton>
do_install puts stuff into a staging area, do_package splits it up into packages
<rburton>
packaging is just turning the output of do_install into packages by doing what FILES_* says
yumasi has quit [Remote host closed the connection]
yumasi has joined #yocto
<yates>
ok let me regroup/rethink this then. thanks for the (re)clarification.
<rburton>
4.3.5.3 is do_install, 4.3.5.4 is do_package
<yates>
+1
zeddii has quit [Remote host closed the connection]
zeddii has joined #yocto
dev1990 has joined #yocto
zpfvo has quit [Remote host closed the connection]
florian has quit [Quit: Ex-Chat]
goliath has quit [Quit: SIGSEGV]
florian_kc has quit [Ping timeout: 246 seconds]
zeddii has quit [Ping timeout: 258 seconds]
yumasi has quit [Ping timeout: 246 seconds]
zeddii has joined #yocto
Schlumpf has quit [Quit: Konversation terminated!]
frieder has quit [Remote host closed the connection]
droman is now known as stkw0
<khem>
RP: regarding dwarfsrcfiles, -7 errorcode could happen if it read past the filename and got really a long pathname perhaps, is it a race or can it be reproduced with some particular package
camus has quit [Ping timeout: 246 seconds]
camus has joined #yocto
leo_sandoval has quit [Quit: Client closed]
Tokamak has joined #yocto
leonanavi has quit [Quit: Leaving]
zyga has quit [Ping timeout: 252 seconds]
camus has quit [Ping timeout: 258 seconds]
camus1 has joined #yocto
camus1 is now known as camus
ant__ has joined #yocto
Vineela has joined #yocto
leo_sandoval has joined #yocto
camus1 has joined #yocto
camus has quit [Ping timeout: 272 seconds]
camus1 is now known as camus
florian_kc has joined #yocto
prabhakarlad has quit [Quit: Client closed]
Guest3394 is now known as mithro
<RP>
khem: there is a reproducer in the bug now, I worked it out was a SIGBUS
Vineela has quit [Remote host closed the connection]
leo_sandoval has quit [Quit: Client closed]
tkoskine_ is now known as tkoskine
<yates>
what is the purpose of the gconv folder /usr/lib/gconv?
<yates>
the glib-locale recipe generates many files into it