00:02 <sjg1> Tartarus: For me that doesn't give the same results, and takes over 3 minutes!

00:03 <sjg1> lvrp16: Yes that looks like a bug. I think the meaning of bootmeth_common_read_file() changed and the callers were not updated

00:04 <lvrp16> sjg1: thanks for the confirmation

00:05 <sjg1> lvrp16: Also get_relfile() has the same problem (size not set)

00:07 <lvrp16> sjg1: thanks for the heads up.

00:08 <lvrp16> bootstd makes my env so clean 😍

00:08 <Tartarus> sjg1: Yes, it takes about 4 minutes but it gives the correct results

00:08 <Tartarus> And means we don't have to try and duplicate the logic of parsing MAINTAINERS files in buildman

00:09 <Tartarus> I don't object to a new tool that parses those files and can be queried

00:09 <Tartarus> But since handling "N" format broke what we have in CI, I just want to move on

00:57 <sjg1> Are you sure it is correct? It complains about uniphier_ld4_sld8_defconfig, for example

01:09 <sjg1> lvrp16: Yes and you can use a txt environment too

01:10 <lvrp16> sjg1: yep I use it with ini and it is like magic. All the hacks I use to carry on top of master can all be dumped.

01:29 <sjg1> lvrp16: Good to hear!

01:55 <Tartarus> sjg1: Yes, it's correct, uniphier is all orphaned

01:56 <Tartarus> I've emailed the kernel folks to see if they want to or know someone who wants to pick it up

15:46 <Tartarus> sjg1: local networking issues or just weka networking issues?

16:33 <Tartarus> sjg1: And as a tangent / related to config fragments, what I'd kinda like to do long term is stop making boards.cfg and instead have buildman make a db of the Kconfig stuff, like moveconfig.py can be told to do, and use _that_ for figuring out what to build from what. In CI we'd just pass that along as an artifact to avoid generating it N times, which would be really terrible in Azure

16:33 <Tartarus> And in general it should be fine / fine enough with some dependency rules

16:56 <marex> day 0 ... white hat detect vulnerability, reports it to them

16:56 <marex> day 1 ... critical vuln fixed

16:56 <marex> day 2 ... remaining issues fixed

16:56 <marex> let's see about NXP ... day 0 ... vuln reported

16:56 <marex> day 14 ... query about any news

16:56 <marex> right now is like day 45 and still no CVE

16:57 <Tartarus> Not Sisko? ;)

16:58 <Tartarus> https://memory-alpha.fandom.com/wiki/In_the_Pale_Moonlight_(episode)

16:59 <marex> Tartarus: who's the romulans and who's the dominion in this setup ? :)

17:01 <opalfruit> Hi, does anyone know if the AC200 ethernet phy is supported by u-boot, i'm curious whether boards such as tanix tx6, which use the AC200 in the same package as the Allwinner chip, would be able to pxe boot from u-boot? I've not seen any defconfig's for such boards that seem to enable it

17:03 <marex> opalfruit: nope, but the driver looks simple enough to port

17:04 <opalfruit> thanks, port from where sorry, from a linux version or..?

17:04 <marex> opalfruit: linux, yeah

17:05 <marex> patch seems to have been posted at least

17:08 <opalfruit> gotcha, will have a look at that then, thanks, there doesn't happen to be a guide on writing / porting drivers to u-boot?

17:13 <marex> opalfruit: drivers/net/phy/intel_xway.c

17:13 <marex> opalfruit: fork this, plug in the few register writes, and you should be all set

17:15 <opalfruit> thanks!

17:21 <opalfruit> would i be right in thinking i'd also need to do some config over i2c too, looking at - https://lore.kernel.org/lkml/20200416185758.1388148-3-jernej.skrabec@siol.net/T/

17:24 <marex> oh, there is more registers to write

17:25 <marex> so yes, you migth need to implement MFD driver for this, see drivers/mfd (same as linux to no surprise)

19:42 <sjg1> Tartarus: this is where the code came from a8a0141bc2c buildman: Incorporate the genboardscfg.py tool

19:42 <Tartarus> Yeah

19:42 <Tartarus> And genboardscfg.py replaced some shell script or somehing, for MAKEALL

19:43 <Tartarus> Or maybe even a manually updated text file with instructions on how to re-sort it

20:48 <sjg1> Tartarus: Will send a v4. I can move the code to a new tool once we come up with a name :-)

20:48 <sjg1> https://www.irccloud.com/pastebin/J00IcNGW/

20:48 <sjg1> Tartarus: 6 minutes down to under 2 second

20:50 <apteryx> the unzip u-boot command is not documented, right

20:50 <apteryx> ?

21:30 <apteryx> it has online documentation (at the u-boot prompt), but I was expecting to see them listed in the manual