08:04 <PaulFertser> borneoa_: fyi the bug with gertty crash was fixed upstream, current master branch works out of the box

08:05 <borneoa_> PaulFertser: very fast! Cool

08:44 <PaulFertser> borneoa_: the reason is that by default PolyGerrit creates "patchset comments" now, it's basically like code comments to specific file but using a special reserved "/PATCHSET_LEVEL" filename. You can spot the difference in PolyGerrit too, they're marked "Patchset" and foldable.

08:44 <PaulFertser> I did some minimal investigation, contacted the devs via their IRC and they fixed it :)

10:43 <PaulFertser> tarekb: hey :) I'm testing reply via e-mail, it almost worked.

10:50 <Haohmaru> hm, v 0.11.0 says "CMSIS-DAP v2 (USB bulk based) adapter driver" <- would that automagically make things "faster" on an existing DAP-based debugger, or does it need a new firmware too?

10:51 <PaulFertser> Haohmaru: depends on whether your debugger supported non-HID mode or not.

10:52 <Haohmaru> no idea about that

10:52 <PaulFertser> Haohmaru: lsusb -vvv would tell

10:52 <Haohmaru> i'll check when i'm @home

10:52 <PaulFertser> Are you limited by the adapter, do you feel slowness?

10:53 <Haohmaru> i don't know how to "feel", iirc i did record a video of it and i must have shown it here

10:54 <PaulFertser> tarekb: sorry about the nitpicking but that let me see how CSS wasn't allowed for viewing from Jenkins and to test some other parts of the workflow.

10:54 <Haohmaru> you know my experience is equal to a very very small positive number, just a hair above zero

10:57 <Haohmaru> i can dig out the video too

10:57 <Haohmaru> but i don't remember if anyone really gave a comment on it

11:04 <Haohmaru> i don't know how much of the "slowness" comes from the IDE side (but i do have experience with using the normal gdb on normal computer programs, that's not slow certainly), how much comes from the gdb -> OS -> USB -> debugger link, and how much comes from the debugger's firmware

11:08 <Haohmaru> oh, did the website get pimped up or what

11:09 <PaulFertser> Haohmaru: pelican static instead of stupid php wordpress

11:09 <Haohmaru> kewl, i was eyeballing this pelican too for my own website, but i ended up going for Hugo coz my friend told me so

11:15 <Haohmaru> the docs look the same tho

11:17 <PaulFertser> Why would they be different?

11:17 <Haohmaru> i don't know

11:17 <PaulFertser> User doc is generated with texinfo, developer's guide with doxygen.

11:17 <Haohmaru> i have a user manual on my website too and that got repimped up now together with the rest of the site

11:18 <Haohmaru> (it's not online yet tho)

11:18 <Haohmaru> so texinfo.. it looks the same as the gcc docs

11:18 <Haohmaru> i rewrote my sh*t with asciidoc

18:19 <antto> PaulFertser, https://paste.debian.net/hidden/13886581/

18:20 <antto> that's the zl33prg debugger

18:22 <PaulFertser> antto: looks like hid only

18:22 <antto> :~(

21:42 <PaulFertser> borneoa_: probably we should enable submitWholeTopic https://gerrit-review.googlesource.com/Documentation/config-gerrit.html ?

21:43 <PaulFertser> Is that what you had in mind?

21:48 <borneoa_> PaulFertser: no. In a gerrit inside ST I can submit a whole series, even without setting the topic. It's named submit with parents or something similar. But in openocd we seldom have long series, and anyway every patch of the series has to be put +2, one by one. So not a real benefit.

21:50 <PaulFertser> borneoa_: guess you have different merge strategy there then.

21:51 <borneoa_> PaulFertser: possible. I should check

21:52 <PaulFertser> Gertty doesn't allow to add reviewers unfortunately

21:53 <PaulFertser> Not implemented because they do not use that

21:53 <PaulFertser> Coordinating on irc/matrix instead.

22:34 <borneoa_> PaulFertser: yep, the ST gerrit uses strategy "rebase always"

22:36 <PaulFertser> borneoa_: if you think that's appropriate I can switch.

22:37 <PaulFertser> borneoa_: and in general I'd be happy to make your life easier, just need to know how :)

22:57 <karlp> only reason git defaults to merge is because linux people have all these public branches

22:58 <karlp> ~100% of everyone else is better off with rebase always

23:10 <borneoa_> PaulFertser: no reason to change. I'm ok with the current setup

23:11 <borneoa_> karlp: today we have strategy cherry pick.

23:13 <borneoa_> karlp: so far it's working fine. I don't think "merge" to be suitable for ooenocd, we only have one branch.

23:13 <karlp> works the same, just makes it explicitly one by one

23:13 <karlp> yeah, skip merge policyt :)

23:15 <borneoa_> karlp: so with cherry pick we are free to merge in a different order, or skip one patch in the series. Sounds more flexible. Clearly it opens the door to make a big mess, but should be fine.

23:16 <karlp> :+1: