Tartarus changed the topic of #u-boot to: SOURCE MOVED TO https://source.denx.de/u-boot/u-boot.git / U-Boot v2024.10, v2025.01-rc4 are OUT / Merge Window is CLOSED, next branch is OPEN / Release v2025.01 is scheduled for 06 January 2025 / Channel archives at https://libera.irclog.whitequark.org/u-boot
vagrantc has quit [Quit: leaving]
mmu_man has quit [Ping timeout: 265 seconds]
mmu_man has joined #u-boot
zibolo_ has quit [Ping timeout: 245 seconds]
zibolo has joined #u-boot
MyNetAz has quit [Remote host closed the connection]
MyNetAz has joined #u-boot
mmu_man has quit [Ping timeout: 264 seconds]
prabhakalad has quit [Ping timeout: 245 seconds]
prabhakalad has joined #u-boot
rvalue- has joined #u-boot
rvalue has quit [Ping timeout: 248 seconds]
rvalue- is now known as rvalue
warpme has joined #u-boot
dsimic has quit [Ping timeout: 272 seconds]
dsimic has joined #u-boot
ikarso has quit [Quit: Connection closed for inactivity]
<sjg1> Tartarus: That was the UPL series. I've been trying to get it figured out and finally made some progress. I sent a short follow-up today. I was hoping you might apply both. If you don't want to, let me know
<sjg1> naoki: the u-boot-any thing just means it cannot find u-boot.bin or similar. The list it shows you is the images it found. But for your application I suspect you don't need those values
<sjg1> So just try to update SPL to not access the binman symbols
<naoki> sjg1: thanks, Kwiboo told me to add no-write-symbols
<sjg1> Oh yes that will stop it too
<naoki> Btw i2c_eeprom.c doesn't support "atmel,24c16"...
<naoki> Is there any difference between 24c16 and 24c16a?
<naoki> Oh CMD_EEPROM is deprecated...
<naoki> What is the right way to access eeprom from command prompt?
warpme has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
goliath has joined #u-boot
vardhan has joined #u-boot
warpme has joined #u-boot
alpernebbi has joined #u-boot
ikarso has joined #u-boot
monstr has joined #u-boot
warpme has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
warpme has joined #u-boot
mmu_man has joined #u-boot
zsoltiv_ has quit [Ping timeout: 252 seconds]
<Tartarus> sjg1: A 60 part series isn't reviewable
___nick___ has joined #u-boot
xroumegue has joined #u-boot
___nick___ has quit [Client Quit]
warpme has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
___nick___ has joined #u-boot
naoki has quit [Quit: naoki]
<marex> Tartarus: sjg1: try spoonfeeding patches ? Send them in smaller chunks so they can get in easily ?
vardhan has quit [Ping timeout: 252 seconds]
<marex> that seems to work
monstr has quit [Remote host closed the connection]
totkeks has joined #u-boot
ikarso has quit [Quit: Connection closed for inactivity]
<SlimeyOTG> Much ❤️
pgreco_ has joined #u-boot
pgreco has quit [Ping timeout: 264 seconds]
<totkeks> what is this? emojis in 8bit chat? :D
<marex> I think unicode , some terminals now interpret it , systemd uses it a bit
<SlimeyOTG> Taking it back to 79
<quinq> UTF-8 only exists since ~30 years
<totkeks> same with this weird ipv6. but just like the internet, that will never gain acceptance
<SlimeyOTG> https://imgur.com/a/XeZmUbI CPU from the first Terminator
<quinq> Didn't know terminators were running on qoriq
SlimeyOTG has quit [Ping timeout: 276 seconds]
SlimeyOTG has joined #u-boot
<shadow> this anomalous LwIP dhcp behavior is not something I have mental bandwidth for :| I'm adding loads of printf debug statements to at least be able to provide more information.
<marex> shadow: did you talk to Jerome ?
<shadow> not beyond the mailing list reply
<marex> Tartarus: is there a way to have commit board/vendor/common/Makefile and include that in build without having to include ../common/stuff.o in each board/vendor/board/Makefile ?
<marex> HAVE_VENDOR_COMMON_LIB maybe ?
<Tartarus> It should be automatic
<Tartarus> The cases where board/vendor/fooboard/Makefile grab ../common/bar.o is because there's not a common/Makefile, for maybe good maybe didn't know about the mechanism reasons
mmu_man has quit [Ping timeout: 244 seconds]
<marex> Tartarus: yeah, found it
<marex> Tartarus: thanks
mmu_man has joined #u-boot
SlimeyOTG has quit [Read error: Connection reset by peer]
SlimeyOTG has joined #u-boot
<shadow> marex: is Jerome active on IRC?
<marex> shadow: I dont know, sorry
<marex> Tartarus: ^ ?
<Tartarus> Yeah, not sure either
<shadow> thanks though. what I've got (for debug print statement output) is https://paste.debian.net/1342323/ where new_netif() branches to eth_init(), it tries first unplugged ethernet and times out, then again w/ second (plugged in) port successfully but somehow that doesn't stick around
SlimeyOTG has quit [Read error: Connection reset by peer]
<marex> shadow: what does not stick around ?
<marex> shadow: did you try ethact/ethprime env vars ?
<marex> maybe they have some impact
<shadow> marex: Tom did suggest and yes if I set ethact to the plugged-in second interface it works as expected
<marex> Tartarus: uh ... I have board Makefiles which look like ... ifndef XPL obj += stuff.o endif ... and the build fails for non-SPL because there is no object files for non-SPL generated, built-in,o is not build
<marex> *not built
<marex> shadow: so you want to make the stack test all network interface one after the other ?
<shadow> it already tests all network interface that it needs it just doesn't remember that it did this for whatever reason
<shadow> look at the pastebin
<Tartarus> marex: The board dir, or in the common dir?
<shadow> ethernet@16030000 is unplugged, ethernet@16040000 is plugged in and it sees this, but then it goes back to trying ethernet@16030000 and gives up (?)
<marex> Tartarus: board dir
<marex> Tartarus: I guess I have to move a bit of ifdeffery into C files
<Tartarus> Hmmm
<Tartarus> We should be OK with an empty board dir
SlimeyOTG has joined #u-boot
<marex> Tartarus: what about empty Makefile in board dir ?
<Tartarus> I am wondering and looking at an older commit of mine
SlimeyOTG has quit [Read error: Connection reset by peer]
SlimeyOTG has joined #u-boot
<marex> shadow: use some printf("%s[%d]\n", __func__, __LINE__); to get prints with actual function names and line numbers, so you can trace the code execution
<marex> shadow: also, #define DEBUG at the beginning of net-lwip.c to enable the debug() statement prints
<marex> that might give you some idea
<marex> shadow: also, print %s dev->name to find out which udevice * is passed around
<Tartarus> marex: It works for drivers, etc, so yes, we should
<marex> shadow: that might give you an idea where the udevice * changes back to the wrong one
<Tartarus> (I just did a quick local hack to confirm that an otherwise nothing build drivers/ directory gets built-in.o made)
SlimeyOTG has quit [Ping timeout: 252 seconds]
<shadow> thanks marex
<marex> Tartarus: drivers/ also contains subdirectories, right ?
<Tartarus> marex: Yes
SlimeyOTG has joined #u-boot
<marex> Tartarus: I'll keep digging
<Tartarus> But I just made things build in drivers/qe/ and not have anything to build in there (I quickly hacked out fdt.o) and it still went in and made built-in.o
la_mettrie has left #u-boot [#u-boot]
<marex> Tartarus: echo > disk/Makefile ; make -j
<marex> Tartarus: try that
<Tartarus> Failure: /usr/bin/ld: cannot find disk/built-in.o: No such file or directory
<marex> ;-)
<shadow> is there more than CONFIG_LOG=y needed to get debug to show up on console?
<marex> shadow: I think you need to configure log level
<marex> make menuconfig -> /CONFIG_LOG -> that's the menu you want
<shadow> thanks
<shadow> oh that's weird... backspace does not work in 'menuconfig' search /
<shadow> something with my qterminal and backspace in menuconfig not working, but is fine in gnome-terminal... anyways
<shadow> sync kconfig with upstream is suggested
SlimeyOTG has quit [Read error: Connection reset by peer]
SlimeyOTG has joined #u-boot
SlimeyOTG has quit [Ping timeout: 276 seconds]
SlimeyOTG has joined #u-boot
SlimeyOTG has quit [Read error: Connection reset by peer]
___nick___ has quit [Ping timeout: 265 seconds]
SlimeyOTG has joined #u-boot
<marex> shadow: does it work in linux-next menuconfig ?
<shadow> marex: appears to be negatory on that.
<shadow> or ... sorry I tried u-boot -next
<shadow> moment please will try the kernel
<shadow> marex: linux-next/master menuconfig is fine, works with qterminal
SlimeyOTG has quit [Read error: Connection reset by peer]
SlimeyOTG has joined #u-boot
SlimeyOTG has quit [Read error: Connection reset by peer]
SlimeyOTG has joined #u-boot
SlimeyOTG has quit [Read error: Connection reset by peer]
SlimeyOTG has joined #u-boot
<marex> shadow: huh
<marex> shadow: and u-boot/master does not ?
SlimeyOTG has quit [Ping timeout: 276 seconds]
<marex> Tartarus: when did the kconfig got synced last ?
SlimeyOTG has joined #u-boot
<marex> shadow: seems Yamada-san has been very active
<shadow> marex: correct. applying the changes from the pull request I linked makes it work in qterminal (and the comment there is that those are already in upstream Linux kconfig)
<Tartarus> marex: It's on v4.20 still
<Tartarus> I've made a few passes at moving to v5.0 but hit problems
<Tartarus> Some parts like the parsing could likely be synced separate from other logic
<Tartarus> (ie scripts/kconfig/ separate from scripts/Makefile.*)
<marex> shadow: do you want to submit the fixes ?
<shadow> I think it will work itself out if y'all want to sync with upstream
ikarso has joined #u-boot
mmu_man has quit [Ping timeout: 252 seconds]
SlimeyOTG has quit [Read error: Connection reset by peer]
SlimeyOTG has joined #u-boot
naoki has joined #u-boot
<shadow> marex: we just patch it in from linux-next? for fun they've changed the spelling from 'heigth' => 'height' in defines so I can't just copypaste those three files over.
mmu_man has joined #u-boot
<shadow> up to y'all if you want to carry some patches on menuconfig in u-boot; I'm not that bothered by it since if/when it gets synced to upstream the fix exists there and a workaround is to use a different terminal
SlimeyOTG has quit [Read error: Connection reset by peer]
SlimeyOTG has joined #u-boot
SlimeyOTG has quit [Read error: Connection reset by peer]
SlimeyOTG has joined #u-boot
RoganDawes has joined #u-boot
<RoganDawes> Hi folks. I'm looking for some help with an old iMX28 target (Wink Hub v1). Fabio Estevam helped with an initial config, and I was able to push u-boot to my board using SDP over USB. However, since then I have managed to destroy my console UART by putting 12V through it. While I wait for a new board to ship to me, I'm trying to find another way to
<RoganDawes> get a console on it. I have confirmed that the CPU still works, as I can still see it enumerating over USB as a "Freescale Inc ROM Recovery" device when I ground the flash's data pins while applying power. I do have JTAG access as well, via a FTDI2232H adapter.
<RoganDawes> My primary problem at the moment is actually getting a working u-boot that I can interact with. I see two possible responses currently when running mxsldr u-boot.sb. Either the device disconnects from USB, and re-enumerates again as the same "ROM Recovery" device. Or else nothing appears to happen at all, and the USB endpoint becomes unresponsive,
<RoganDawes> but does not disconnect (according to `dmesg -w`)
SlimeyOTG has quit [Read error: Connection reset by peer]
SlimeyOTG has joined #u-boot
<RoganDawes> I have tried loading u-boot-spl as suggested by https://docs.u-boot.org/en/latest/board/nxp/ls1046ardb.html#ls1046ardb-jtag, and executing `resume 0` since that is the start of OCRAM on the i.MX28. However, that appears to crash, and I can no longer execute `halt` in openocd, unless I first `reset`. Trying to set the PC `$pc=0`, and then single
<RoganDawes> stepping, I get to $pc=0x0054, before it jumps off into the ROM at 0xfffffafa. I'm unfamiliar with gdb, so am not really sure how to proceed from here.
SlimeyOTG has quit [Read error: Connection reset by peer]
SlimeyOTG has joined #u-boot
RoganDawes has quit [Changing host]
RoganDawes has joined #u-boot
<shadow> hi RoganDawes
SlimeyOTG has quit [Ping timeout: 276 seconds]
SlimeyOTG has joined #u-boot
<RoganDawes> Hi Shadow
<marex> RoganDawes: is that the musb ?
<marex> RoganDawes: mx28 does not contain musb, it contains CI HDRC , chipidea IP , which is already enabled in your config
<marex> you can also try https://archive.fosdem.org/2023/schedule/event/semihosting_uboot/ , paging cambrian_invader
<marex> RoganDawes: if you want to start stuff on mxs via usb, try https://source.denx.de/denx/mxsldr
joeskb7 has quit [Quit: Lost terminal]
<marex> it has been forever since I tried mxs and jtag
joeskb7 has joined #u-boot
joeskb7 has quit [Client Quit]
joeskb7 has joined #u-boot
<RoganDawes> marex: thanks for the link to the semihosting talk. will definitely check that out.
<RoganDawes> I get the same results with `mxsldr u-boot.sb` and with `snagrecover -s imx28 -f wink.yaml`, where wink.yaml just points to u-boot.sb, according to the default template.
<RoganDawes> Will watch the video in the morning. 1am here :-p
lehmanju has quit [Quit: The Lounge - https://thelounge.chat]
<marex> RoganDawes: can you mux the UART on another set of pins btw ?
lehmanju has joined #u-boot
<marex> RoganDawes: ones which are not ... destroyed ?
<marex> or use AUART instead of DUART ?
mmu_man has quit [Ping timeout: 265 seconds]
joeskb7 has quit [Quit: Lost terminal]
joeskb7 has joined #u-boot
<RoganDawes> marex: I might be able to use an AUART instead of the destroyed DUART. Only thing is that they are all connected to various radio peripherals. But at least I can hopefully snoop the pins and see what is being output.
<RoganDawes> will have to try and trace the signals to the radios and see if any of the UART pins are brought out to a header pin somewhere.
<marex> RoganDawes: mux DUART to some other pins ? is that an option ?
mmu_man has joined #u-boot
<xypron> Tartarus : U-Boot's implementation of the YAFFS2 file-system is not integrated into the file-system functions. Upstream YAFFS2 developement stopped 2012. Linux eliminated it in 2010. - Do we really want to keep it?
<RoganDawes> marex: I'm not sure. Maybe I could redirect it to the status LED, that should be sufficiently exposed for me to be able to see it?
<RoganDawes> I wouldn't know how to even start doing that, though!
<Tartarus> xypron: It should probably just be removed, yes
<marex> RoganDawes: try it
<RoganDawes> how?
<marex> RoganDawes: check HW_PINCTRL_MUXSEL in MCIMX28RM.pdf
<marex> board/freescale/mx28evk/iomux.c: MX28_PAD_PWM0__DUART_RX,
<marex> seems like it is configurable in the board file
<shadow> U-Boot documentation says stuff a '#define LOG_DEBUG' at the top of a file and the message will be written regardless of CONFIG_LOG_DEFAULT_LEVEL but uh... nothing happens. Tried it with CONFIG_LOG=y CONFIG_LOG_CONSOLE=y and without. If I set the default log level with menuconfig then all debug messages everywhere show up which is not really so good.
<shadow> sure I've mis-read or otherwise missed some detail
<shadow> for '#define LOG_DEBUG' to work correctly I need to compile with CONFIG_LOG ? or does that note about 'Using DEBUG' in documentation infer that CONFIG_LOG is not necessary because such thing will be a printf() statement.
SlimeyOTG has quit [Read error: Connection reset by peer]
joeskb7 has quit [Quit: Lost terminal]
SlimeyOTG has joined #u-boot
joeskb7 has joined #u-boot
<SlimeyOTG> Hehe quinq
<SlimeyOTG> Skynet is a fan of yocto
<SlimeyOTG> Btw how well do you know the qorq target?
<quinq> It's a vast family
<quinq> I know a bit of T2080 and LX2