lthadeus has quit [Remote host closed the connection]
mvlad has joined #yocto
ptsneves has quit [Quit: ptsneves]
florian_kc has joined #yocto
florian_kc is now known as florian
Guest18 has quit [Quit: Client closed]
Average_Joe has joined #yocto
<Average_Joe>
Hi all ... I'm new to yocto project and I'm trying to build my distro for a RISC-V
<Average_Joe>
I've finished an image with 'MACHINE="qemuriscv32"'
<Average_Joe>
the image is successfully created in tmp/deploy/images/qemuriscv32
<Average_Joe>
but, when I'm inspecting the instructions they don't look like RISC-V instructions ...
<Average_Joe>
am I missing something ?
mbulut has joined #yocto
<mcfrisk>
a layer with an empty __init__.py in meta-layer/lib/oeqa/runtime/cases broke all of parselogs.py open_text() resource file loading, only that path got searched. took way too long to find that out...
<Average_Joe>
oh, ... ignore my previous remark ... endianness got me (again)
<landgraf>
Average_Joe: been there, done that :(
<Average_Joe>
:-)
florian has quit [Ping timeout: 240 seconds]
jmd has joined #yocto
florian has joined #yocto
tnovotny has joined #yocto
jmiehe has joined #yocto
amitk has quit [Ping timeout: 255 seconds]
xmn has joined #yocto
leon-anavi has joined #yocto
Chaser has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
Chaser has joined #yocto
alessioigor has quit [Quit: alessioigor]
alessioigor has joined #yocto
kpo_ has quit [Ping timeout: 252 seconds]
alessioigor has quit [Quit: alessioigor]
alessioigor has joined #yocto
jmd has quit [Remote host closed the connection]
prabhakarlad has joined #yocto
prabhakar has joined #yocto
Guest43 has joined #yocto
<Guest43>
hello. the board i use has both an A core and a M7 one. i want to add a .bin file containing the firmware for the M core to bootfs but none of the thing i tried worked. what is the common way to do this?
<Guest43>
it's been weeks already that i search the internet without finding a valid answer
<rburton>
just make a recipe that ships that file
<Guest43>
well i tried that first
<Guest43>
but it doesn't appear in the filesystem uboot has access to
<rburton>
put it in that filesystem then. this is all something you're constructing...
<Guest43>
how can i put that in the uboot filesystem?
<Guest43>
i tried adding it to IMAGE_BOOT_FILES too, but it doesn't work either
<rburton>
what file system is uboot mounting, and how are you creating it?
<Guest43>
well i don't know actually
<rburton>
that's a good thing to understand then
<Guest43>
seems like it's done in what torizon provide
<rburton>
i'd be asking them for support then
<Guest43>
will do
<Guest43>
thanks for the advices
<rburton>
fwiw meta-arm has a BSP which writes a big blob of flash that contains uboot and all the firmware pieces
<Guest43>
ohhh lemme check that
jmiehe has quit [Quit: jmiehe]
<LetoThe2nd>
multiconfig FTW here :-)
<rburton>
Guest43: its corstone1000, it currently abuses a wic file write a flash blob with various pieces at explicit offsets so its easy to tell uboot where to read from.
<Guest43>
so if i find the relevant .wks file for my build i could amend it to get that file in bootfs?
<rburton>
well no i doubt it
<rburton>
you'd need to change your uboot to have the same behaviour
<Guest43>
hum that makes sense i guess
<rburton>
this is deeply hardware specific
<Guest43>
writing to toradex now
<rburton>
_how_ does the hardware boot, _how_ is the bootloader configured, etc
<Guest43>
certainly they'll have the right pointers
rob_w has quit [Remote host closed the connection]
jmiehe has joined #yocto
jmiehe has quit [Client Quit]
prabhakarlad has quit [Quit: Client closed]
zpfvo has quit [Ping timeout: 255 seconds]
zpfvo has joined #yocto
vladest has quit [Remote host closed the connection]
vladest has joined #yocto
zpfvo has quit [Ping timeout: 255 seconds]
Guest18 has joined #yocto
Daanct12 has quit [Quit: WeeChat 4.1.2]
zpfvo has joined #yocto
prabhakar has quit [Ping timeout: 276 seconds]
BobPungartnik has joined #yocto
GNUmoon2 has quit [Ping timeout: 240 seconds]
frieder has quit [Ping timeout: 268 seconds]
jmiehe has joined #yocto
alessioigor has quit [Quit: alessioigor]
alessioigor has joined #yocto
Dr_Who has joined #yocto
BobPungartnik has quit [Quit: Leaving]
Chaser_ has joined #yocto
Chaser has quit [Ping timeout: 256 seconds]
linfax has quit [Ping timeout: 240 seconds]
amitk has joined #yocto
GNUmoon2 has joined #yocto
alessioigor has quit [Quit: alessioigor]
alessioigor has joined #yocto
likewise has joined #yocto
frieder has joined #yocto
florian_kc has joined #yocto
Chaser_ has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
|Xagen has quit [Ping timeout: 268 seconds]
mason has joined #yocto
<mason>
Hi all! Has IMAGE_INSTALL_remove been deprecated? Got an error ending with "contains an operation using the old override syntax. Please convert this layer/metadata before attempting to use with a newer bitbake.
<rburton>
yes
<mason>
thank you
<rburton>
the override operator is : not _
<mason>
Thank you. Will give that a try.
<rburton>
there's a tool in scripts to migrate syntax if you've just upgraded
<mason>
Hanging my cluelessness out there, this is the first time we've tried removing something this way, so I suspect we're using the current syntax, but what I just added is obsolete.
<LetoThe2nd>
mason: for the record: its always more sensible to not add something undesired in the first place than to remove it again later.
<mason>
LetoThe2nd: Sure. In this case, I want to remove nfs-utils, which we're not explicitly requesting that I can see.
<LetoThe2nd>
mason: then the correct way is to find out what adds it :-)
mbulut has quit [Remote host closed the connection]
Guest18 has quit [Quit: Client closed]
<JPEW>
RP: jpew/hashserver-gc in poky-contrib
<paulg>
rburton, I was looking at your 64k patch and was wondering if you checked... when you build and boot a 64k page system, login and run "free" immediately after boot, does it have a considerably larger footprint than a default 4k build?
<paulg>
I'm sure it would have 20y ago but everything is so bloated now, that I'm guessing the level of fragmentation that 4k pages delivers is just needless.
<rburton>
i've not checked myself but my understanding is there's a larger footprint but its not _considerable_
<rburton>
and obviously clever allocators batch allocations etc etc
<rburton>
the performance win is quite notable
<paulg>
I dont doubt it one bit.
<rburton>
note that the apple m1 is 16kb pages only
<paulg>
at the risk of stating the obvious, 16 is still 4x better than the current default.
<LetoThe2nd>
simone: at FOSDEM usually the OpenEmbedded side takes precedence. We (as OE) are sharing a table with AGL, plus some presentations are related to the topic area.
<LetoThe2nd>
simone: and last but not least on Monday after FOSDEM, there is an OE Workshop
khazakar has quit [Quit: Connection closed for inactivity]
Chaser has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
alessioigor has quit [Quit: alessioigor]
<rburton>
vvn: you can always share
<rburton>
vvn: the autobuilder builds dunfell through to master and has a single sstate
<vvn>
perfect thank you
<yudjinn>
hello, this might not be the best place to ask, but I'm trying to add ntfs support to my build; specifically, r/w. I added the ntfs-3g driver, and on-target I can `mount -t ntfs3 /a /b`, but I cant get it to automatically use ntfs3. I added a udev rule to set the FS_ID to `ntfs3` if it was `ntfs` which seems to work (`udevadm info /a` shows "ntfs3"), but still having issues; thoughts?
simone has quit [Quit: Konversation terminated!]
<vvn>
rburton: the autobuilder uses different branches?
<rburton>
vvn: one sstate directory
<vvn>
yep but I meant to build dunfell and next versions, are they per-branch versions as usual?
<rburton>
not sure what you mean. when it builds dunfell it checks out dunfell.
<vvn>
ok
<vmeson>
vvn: one GIANT sstate directory! ;-) The cache is hashed so there are no collisions b/w dunfell, kirkstone, etc or if nothing has changed you can nust re-use the sstate object.
<vmeson>
*just re-use
<vvn>
you just need to make sure uninative is enabled in your distro, right?
<rburton>
no
<rburton>
that just means you can reuse native between different hosts
<vvn>
I understood this was recommended for sstate sharing
<rburton>
if you don't use uninative (uni = one, one-native) then native recipes need to be rebuilt
<rburton>
it improves sharing, but doesn't break sstate
<vvn>
do you guys ever use `. oe-init-build-env .` (i.e. using `.` as the BDIR)? :-)
<vmeson>
vvn: sure, at times but only if I've previously created the '.' directory using the oe-init-build-env script. As long as bitbake finds what it needs you should be fine. Why do you ask.
<cambrian_invader>
is it possible to get bitbake to ignore dependency issues long enough to print out the environment via -e?
<cambrian_invader>
I'm trying to debug a strange error with my kernel (Missing or unbuildable dependency chain was: ['my-kernel', "${@oe.utils.conditional('KERNEL_IMAGETYPE',"])
Vonter has quit [Ping timeout: 260 seconds]
Vonter has joined #yocto
<vvn>
vmeson: because in order to share a build system, it is convenient to provide a slightly biased repository layout containing a versioned conf/ directory with the site.conf and multiconfig files, alongside our local meta-* directories and submodules
<vvn>
(I see a build directory almost as a unlisted layer)
prabhakarlad has quit [Quit: Client closed]
tnovotny has quit [Quit: Leaving]
<vvn>
(also it's simpler to have paths relatives to ${TOPDIR} subdirectories rather than looking into parent directories)
vladest has quit [Remote host closed the connection]
<rburton>
vvn: if you want to provide local.conf then use templates like poky itself does, don't play games using . as build dir
Xagen has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
rob_w has quit [Read error: Connection reset by peer]