Wednesday, 2020-09-23

armpitsakoman, I would fire it off. RP is sleeping00:03
*** goliath <goliath!> has quit IRC00:24
*** sakoman <sakoman!~steve@> has quit IRC00:47
*** sakoman <sakoman!~steve@> has joined #yocto00:50
*** leon-anavi <leon-anavi!~Leon@> has quit IRC00:59
*** kiwi_29 <kiwi_29!> has joined #yocto01:03
*** kiwi_29 <kiwi_29!> has quit IRC01:10
*** vmeson <vmeson!> has quit IRC01:20
*** vmeson <vmeson!> has joined #yocto01:21
*** camus1 <camus1!~Instantbi@> has joined #yocto01:28
*** kaspter <kaspter!~Instantbi@> has quit IRC01:29
*** camus1 is now known as kaspter01:29
*** sw_ <sw_!> has joined #yocto01:58
*** sxiii <sxiii!> has quit IRC01:59
*** kiwi_29 <kiwi_29!> has joined #yocto02:01
*** kiwi_29 <kiwi_29!> has quit IRC02:03
*** kiwi_29 <kiwi_29!> has joined #yocto02:18
*** Crofton|cloud_ <Crofton|cloud_!sid401373@gateway/web/> has joined #yocto02:19
*** zbooth- <zbooth-!~zbooth@2600:1f16:181:f300:d350:982:b6f5:216c> has joined #yocto02:19
*** clementp[m]1 <clementp[m]1!cperonmatr@gateway/shell/> has joined #yocto02:20
*** rewitt_ <rewitt_!~rewitt@unaffiliated/rewitt> has joined #yocto02:21
*** rewitt_ <rewitt_!~rewitt@unaffiliated/rewitt> has quit IRC02:23
*** rubdos_ <rubdos_!~rubdos@2a02:578:859d:700:8b44:5716:382d:a7da> has quit IRC02:27
*** apsychogirl[m] <apsychogirl[m]!apsychogir@gateway/shell/> has quit IRC02:27
*** clementp[m] <clementp[m]!cperonmatr@gateway/shell/> has quit IRC02:27
*** Crofton|cloud <Crofton|cloud!sid401373@gateway/web/> has quit IRC02:27
*** zbooth <zbooth!~zbooth@2600:1f16:181:f300:d350:982:b6f5:216c> has quit IRC02:27
*** Crofton|cloud_ is now known as Crofton|cloud02:27
*** kiwi_29 <kiwi_29!> has quit IRC02:31
*** kiwi_29 <kiwi_29!> has joined #yocto02:32
*** rubdos_ <rubdos_!~rubdos@2a02:578:859d:700:8b44:5716:382d:a7da> has joined #yocto02:33
*** apsychogirl[m] <apsychogirl[m]!apsychogir@gateway/shell/> has joined #yocto02:33
*** kiwi_29 <kiwi_29!> has quit IRC02:36
*** camus1 <camus1!~Instantbi@> has joined #yocto02:40
*** kaspter <kaspter!~Instantbi@> has quit IRC02:40
*** camus1 is now known as kaspter02:40
*** hpsy1 <hpsy1!~hpsy@> has joined #yocto02:46
*** hpsy <hpsy!~hpsy@> has quit IRC02:47
*** sakoman <sakoman!~steve@> has quit IRC03:05
*** maudat <maudat!> has quit IRC03:06
*** georgem <georgem!~georgem@> has quit IRC03:21
*** georgem <georgem!~georgem@> has joined #yocto03:21
*** sakoman <sakoman!~steve@> has joined #yocto03:25
*** georgem_home <georgem_home!uid210681@gateway/web/> has quit IRC03:43
*** kiwi_29 <kiwi_29!> has joined #yocto04:08
*** kiwi_29 <kiwi_29!> has joined #yocto04:09
*** kiwi_29 <kiwi_29!> has quit IRC04:12
*** feddischson <feddischson!> has joined #yocto04:16
*** Bunio_FH <Bunio_FH!> has quit IRC04:19
*** yizhao <yizhao!~zhaoyi@> has joined #yocto04:48
*** jobroe <jobroe!> has joined #yocto04:51
*** kiwi_29 <kiwi_29!> has joined #yocto04:52
*** rsalveti <rsalveti!uid117878@gateway/web/> has quit IRC04:58
*** sh00p <sh00p!> has quit IRC05:13
*** kiwi_29_ <kiwi_29_!> has joined #yocto05:15
*** kiwi_29 <kiwi_29!> has quit IRC05:15
*** kiwi_29 <kiwi_29!> has joined #yocto05:21
*** kiwi_29_ <kiwi_29_!> has quit IRC05:21
*** kiwi_29 <kiwi_29!> has quit IRC05:29
*** sno <sno!> has quit IRC05:34
*** cp- <cp-!> has quit IRC05:37
*** cp- <cp-!> has joined #yocto05:42
*** AndersD <AndersD!> has joined #yocto05:43
*** jobroe <jobroe!> has quit IRC05:43
*** jobroe <jobroe!> has joined #yocto05:43
*** agust <agust!> has joined #yocto05:44
*** sh00p <sh00p!> has joined #yocto05:46
*** pohly <pohly!> has joined #yocto05:52
*** NiksDev <NiksDev!~NiksDev@> has quit IRC05:59
*** NiksDev <NiksDev!~NiksDev@> has joined #yocto05:59
*** sno <sno!~sno@> has joined #yocto06:02
*** vmeson <vmeson!> has quit IRC06:05
*** kaspter <kaspter!~Instantbi@> has quit IRC06:06
*** kaspter <kaspter!~Instantbi@> has joined #yocto06:06
*** rob_w <rob_w!~rob@unaffiliated/rob-w/x-1112029> has joined #yocto06:06
*** beneth <beneth!> has joined #yocto06:08
*** chris_ber <chris_ber!~quassel@> has joined #yocto06:14
*** gonkulator <gonkulator!~brandon@> has quit IRC06:17
*** gonkulator <gonkulator!~brandon@> has joined #yocto06:19
*** AndersD <AndersD!> has quit IRC06:22
*** cp- <cp-!> has quit IRC06:23
*** cp- <cp-!> has joined #yocto06:29
*** tgamblin <tgamblin!> has quit IRC06:33
*** sstiller <sstiller!> has joined #yocto06:43
*** mckoan|away is now known as mckoan06:52
*** fl0v0 <fl0v0!~fvo@> has joined #yocto06:53
*** tgamblin <tgamblin!> has joined #yocto06:53
*** sh00p <sh00p!> has quit IRC06:55
*** alejandrohs <alejandrohs!~alejandro@2605:6000:1306:9479::a59> has joined #yocto06:58
*** leon-anavi <leon-anavi!~Leon@> has joined #yocto07:01
*** aurelien <aurelien!~user@fsf/member/aurelien> has quit IRC07:02
*** gsalazar <gsalazar!5e3dbd6b@gateway/web/cgi-irc/> has joined #yocto07:02
*** aurelien <aurelien!~user@fsf/member/aurelien> has joined #yocto07:20
*** Bunio_FH <Bunio_FH!> has joined #yocto07:21
yannwhat should be released as 3.1.3 is around 44ab5a8 rather than the current dunfell tip, right ?07:55
*** itseris <itseris!> has quit IRC08:01
*** itseris <itseris!> has joined #yocto08:01
ptsnevesWhen is Gatesgarth expected to be released?08:05
paulbarkerptsneves: Usually the release is late October, sometimes early November08:06
ptsnevespaulbarker thanks for the info. Can you tell what is the process for the branch? Is it branched from master only on release or before the release there is a gatesgarth -next?08:09
paulbarkerptsneves: Typically the branch is created shortly before release, at the point where master may diverge while the new release goes through final QA and release process08:11
ptsnevespaulbarker that is very useful thanks.08:15
*** creich <creich!> has quit IRC08:17
*** Ninic0c0 <Ninic0c0!> has joined #yocto08:19
Ninic0c0Hello humans ! I didn't find many informations about wich release process devs use inside Yocto. I mean which version of application add to the image, extract release note ... Maybe Yocto is not develop in this way. So any clue ? Thx :)08:21
yannI'm a building an ipk of a prebuilt library (libcef) and I'm surprised by do_package_qa not complaining about NEEDED libs not in RDEPENDS, what am I missing ?  The lib is otherwise properly handled as other ELF objects, with debug symbols stripped into -dbg package08:22
*** creich <creich!> has joined #yocto08:24
yannNinic0c0: you may be interested by the buildhistory feature, to track all package versions included in your image08:25
*** mbulut <mbulut!> has joined #yocto08:25
Ninic0c0yann Thx for repplying, the question is more how to manage dev/prod librairies/application inside Yocto, in order to build for example an image with all apps from develop/RC1 branches ...08:26
ptsnevesyann if you want to fix some version of a recipe then you can have PREFERRED_VERSION_<pn> = <version>08:28
qschulzNinic0c0: leverage PREFERRED_VERSION_<package> in local.conf or create different distros one for prod one for dev?08:28
ptsnevesotherwise keep in mind yocto defaults to the highest semantic versioned recipe version08:28
Ninic0c0qschulz different distros ?! Sounds really good advice. I will double check thx :)08:29
yannNinic0c0: buildhistory may still help you for the "extract release note" part08:30
Ninic0c0If combien both souble be the trick. Yhx a lot guys :)08:30
*** opennandra <opennandra!> has joined #yocto08:35
*** NiksDev <NiksDev!~NiksDev@> has quit IRC08:41
*** NiksDev <NiksDev!~NiksDev@> has joined #yocto08:42
*** yizhao <yizhao!~zhaoyi@> has quit IRC08:45
*** tgamblin_ <tgamblin_!> has joined #yocto08:45
*** florian_kc <florian_kc!~florian_k@Maemo/community/contributor/florian> has joined #yocto08:46
*** tgamblin <tgamblin!> has quit IRC08:46
*** yizhao <yizhao!~zhaoyi@> has joined #yocto08:50
*** tgamblin <tgamblin!> has joined #yocto08:50
*** tgamblin_ <tgamblin_!> has quit IRC08:51
*** florian_kc <florian_kc!~florian_k@Maemo/community/contributor/florian> has quit IRC08:53
*** florian_kc <florian_kc!~florian_k@Maemo/community/contributor/florian> has joined #yocto08:53
*** florian_kc is now known as florian08:53
yannhm, I'm mistaken thinking it is a package_qa issue, my problem is package_do_shlibs not being run (and bitbake claiming it is not defined for the package, despite the run.package_do_shlibs.* being generated08:56
*** yizhao <yizhao!~zhaoyi@> has quit IRC08:57
*** sno <sno!~sno@> has quit IRC08:58
yann... indeed it is, and log.do_package does show it does not find those dependencies, but only as NOTE!08:59
*** sno <sno!~sno@> has joined #yocto09:00
yannanyone knows why is "Couldn't find shared library provider" a just note and not a warning ?09:00
*** RobertBerger <RobertBerger!~rber@2a02:587:3b0d:ecec:617c:e634:e2e2:11b9> has joined #yocto09:05
*** kaspter <kaspter!~Instantbi@> has quit IRC09:12
*** kpo_ <kpo_!> has quit IRC09:13
*** kpo_ <kpo_!> has joined #yocto09:13
*** d32 <d32!2ef3c527@> has quit IRC09:18
*** NiksDev <NiksDev!~NiksDev@> has quit IRC09:19
*** pev <pev!> has joined #yocto09:24
ptsnevesthat is a good point, and i believe i have been bitten by that09:33
ptsnevesand then on the rootfs install issues happen09:33
*** kaspter <kaspter!~Instantbi@> has joined #yocto09:34
RPI suspect its the issue of validating that there aren't "normal" use cases which would then start warning09:34
ptsnevescan you think of an example? Would the warnings not be mitigated with insane skips?09:42
RPptsneves: I haven't given it a lot of thought. I have spent quite a few years changing defaults and chasing down all the corner cases which get exposed though, I'm a bit jaded by it.09:55
* RP has spent the last couple of days chasing down the impact of a well intended cleanup that breaks things :(09:55
ptsneveswell then thanks. Might i know what was the cleanup that broke things. Perhaps i have some insight or some help if you have a task09:57
*** vmeson <vmeson!> has joined #yocto09:58
RPptsneves: and the addition of OECORE_NATIVE_SYSROOT to buildtools environment setup scripts09:59
RPptsneves: ultimately that results in this
RPptsneves: I wrote the original patch above after a lot of discussion about "broken behaviour" in buildtools on the mailing list10:00
ptsneveswill have a look. Regarding buildtools we have a completely isolated environment with buildtools, down to the loader. The only quirk we had was that sometimes the buildtools paths get into the state cache, due to them not being in the cleanup list and destroyed the state cache sometimes. We fixed it and we are very happy10:02
ptsneveswhen i mean cleanup list i meant the FIXME list of the state cache10:03
RPptsneves: I think I have a patch in master-next which unsets that variable and some fixes in autobuilder-helper which will hopefully sort the problems10:04
RPptsneves: the buildtools paths shouldn't be making it into sstate, if they are we should look into and fix that10:04
ptsnevesi think that maybe due to the fact that we have a much bigger buildtools list than the default in vanilla yocto. The paths get there through $(which <prog>) that happens in autotools for example, although i am sure there are other build systems that do that as well.10:07
ptsnevesso that detail may not be applicable to vanilla yocto. The point is that there is a way for paths of the buildtools to get into the state cache10:07
RPptsneves: we have fixed some areas that happens by setting it to <prog> without a path explicitly10:09
ptsnevesat least in older versions of poky. I normally work on quite an old version and am now preparing a huge build system upgrade to use the latest one because we need gcc10 which unfortunately does not exist in any release poky10:09
RPptsneves: if you have cases of that in OE-Core I'd be interested to see10:09
ptsneves@RP i will be more active when we upgrade our build system and then our reports will be confirmed to still be relevant to the community.10:10
RPptsneves: cool, as I said, we shouldn't be leaking host paths like that and we would actively fix it10:10
RPptsneves: how old is "older" for you atm?10:11
ptsnevespre per recipe sysroot. Krogoth if i am not mistken. It is so heavily patched and backported10:12
RPptsneves: ouch, yes. Lots of fixes since then!10:12
ptsneveswe have just so many arches and serve so many people that it is very hard to make a big upgrade10:13
ptsneveswe had a ppc that amazingly produced uimages of 8MB  with yocto.10:13
RPptsneves: its why we now have such a heavy test infrastructure for the core10:14
RPptsneves: that is nice :)10:14
RPptsneves: can I ask who "we" is? Just curious, I know people sometimes can't say10:14
ptsnevesi would rather not, as otherwise my engagement with the community would be limted10:16
*** kaspter <kaspter!~Instantbi@> has quit IRC10:18
*** sno <sno!~sno@> has quit IRC10:20
*** kaspter <kaspter!~Instantbi@> has joined #yocto10:21
*** kaspter <kaspter!~Instantbi@> has joined #yocto10:22
RPptsneves: no problem! As I said, just curious10:23
RPI'd much prefer you can engage!10:23
RPptsneves: I'm actually quite pleased buildtools is proving useful to people, that is helpful to know10:24
ptsnevesRP like i said without buildtools there would be no yocto as docker is not an option :)10:25
*** manuel1985 <manuel1985!> has joined #yocto10:31
*** camus1 <camus1!~Instantbi@> has joined #yocto10:32
*** kaspter <kaspter!~Instantbi@> has quit IRC10:34
*** camus1 is now known as kaspter10:34
yannRP: in this case I'd argue that the current behaviour can hide real problems - maybe we could just have a CI run to reveal any issues in OE ?  I guess that wouldn't be for gatesgarth, anyway :)10:43
RPyann: we can run a test through automated testing, yes. Its better than it used to be10:44
RPyann:  in theory I thought there were other QA tests which should pick up issues like that10:44
yannI thought so, but package_do_shlibs seems the natural place to check, since it's the one doing the lib lookup10:45
*** sno <sno!> has joined #yocto10:46
RPyann: you end up needing ways to hide the warning for specific packages/files and that infrastructure is in package_qa10:48
RPyann:  you want to send a patch we can try it and see how bad the issues are I guess10:49
*** goliath <goliath!> has joined #yocto10:50
*** rubdos <rubdos!~rubdos@2a02:578:859d:700:8b44:5716:382d:a7da> has joined #yocto11:08
*** rubdos_ <rubdos_!~rubdos@2a02:578:859d:700:8b44:5716:382d:a7da> has quit IRC11:09
*** jmiehe <jmiehe!> has joined #yocto11:10
yannk will do11:11
*** georgem_home <georgem_home!uid210681@gateway/web/> has joined #yocto11:27
*** berton <berton!~berton@> has joined #yocto11:37
*** Konsgnx <Konsgnx!> has joined #yocto12:10
*** Ox861726f6c64 <Ox861726f6c64!~Ox861726f@gateway/tor-sasl/ox861726f6c64> has joined #yocto12:17
*** opennandra <opennandra!> has quit IRC12:24
*** T_UNIX <T_UNIX!uid218288@gateway/web/> has joined #yocto12:26
*** ptsneves <ptsneves!b0dd7824@> has quit IRC12:28
T_UNIXam I mistaken or is bitbake incapable of handling backslashes in `file://` SRC_URI like `dev-disk-by\x2dlabel-foo.device`? It either fails while locating the file or while actually copying it.12:34
*** kpo_ <kpo_!> has quit IRC12:36
*** kpo_ <kpo_!> has joined #yocto12:36
*** bantu <bantu!~bantu@unaffiliated/bantu> has quit IRC12:39
*** bantu <bantu!~bantu@unaffiliated/bantu> has joined #yocto12:39
T_UNIXNinic0c0: thanks for your answer. I'm sorry, I do not see the point. What I see is that if I specify the filename as such, the fetcher parser will accept it. But the subsequent `cp` will eventually fail. If I specify the name using `'` or escape the backslashes in the SRC_URI path, the fetcher complains about a non-existend file :-/12:44
Ninic0c0T_UNIX FILESPATH is correctly set ?12:45
Ninic0c0share recipe12:45
T_UNIXNinic0c0: yeah, everything is fine. As a workaround, I renamed the file to not contain backslashes and it works.12:45
Ninic0c0ok :)12:46
*** ptsneves <ptsneves!b0dd7824@> has joined #yocto12:46
berton@T_UNIX: There is a open bug that I think is related to your issue12:46
*** clopez <clopez!> has quit IRC12:48
T_UNIXI am not sure why does not put the paths inside quotes `"cp -fpPRH '%s' '%s' % (file, destdir)". Or use `shutils` instead.12:50
T_UNIX@berton: spot on! :)12:50
*** clopez <clopez!> has joined #yocto12:50
T_UNIXI'm sorry I didn't see it right away. I've searched for `backslash` instead of `"\"` :-D12:50
ptsnevesT_UNIX maybe you can submit a patch?12:52
*** otavio_ <otavio_!> has quit IRC13:08
T_UNIXptsneves: is accepted?13:09
*** otavio_ <otavio_!> has joined #yocto13:09
*** leon-anavi <leon-anavi!~Leon@> has quit IRC13:09
T_UNIXor is just a public mirror?13:09
ptsnevesT_UNIX i do not think so. You need to submit a patch to the mailing list.
ptsnevesCan anybody tell me what is the advantage of having a do_install_append in the task append in base recipes disable the ability of other recipes to override behavior of the task13:13
*** pev <pev!> has quit IRC13:15
*** maudat <maudat!> has joined #yocto13:17
*** florian <florian!~florian_k@Maemo/community/contributor/florian> has quit IRC13:23
*** pev <pev!> has joined #yocto13:23
*** C-o-r-E <C-o-r-E!> has quit IRC13:31
*** C-o-r-E <C-o-r-E!> has joined #yocto13:31
T_UNIXptsneves: said and done:,,,20,0,0,0::recentpostdate%2Fsticky,,,20,2,0,7703495913:34
ptsnevesT_UNIX thats great. I will chime in on the patch. Thanks for fixing both placeholders13:36
ptsneveshave you verified it fixes it? I guess the next step would be to create a test case, which to be honest i have never done :(13:37
ptsnevesor better i tried and failed13:37
T_UNIXI tried it locally, yes. Unfortunatelly only with `zeus`. But the paragraph didn't change inbetween. Actually one would likely want to use shutils there, but that's kind of broken in combination with selinux:
ptsnevesgood to know :)13:40
ptsnevesi will test it in latest master in a bit13:41
T_UNIX90% of the time to create the patch was spent on setting up the ML stuff :-/ Maybe there'll be some kind of ML<-> GitHub/GitLab bridge one day.13:42
ptsnevesyeah. i also really dislike it and it is also one of the reasons i do not contribute more. I would be surprised if this is not a very negative factor for submissions.13:43
*** ericch <ericch!> has joined #yocto13:44
*** kaspter <kaspter!~Instantbi@> has quit IRC13:47
*** kaspter <kaspter!~Instantbi@> has joined #yocto13:47
*** RobertBerger <RobertBerger!~rber@2a02:587:3b0d:ecec:617c:e634:e2e2:11b9> has quit IRC13:50
*** rizzitello <rizzitello!~quassel@> has joined #yocto13:53
*** rizzitello <rizzitello!~quassel@> has quit IRC13:53
*** sakoman <sakoman!~steve@> has joined #yocto13:55
*** sstiller <sstiller!> has quit IRC13:58
*** vmeson <vmeson!> has quit IRC13:59
sakomanRP: looks like the issue with buildperf is still there :-(14:03
RPsakoman: yes, I was waiting for you to talk about this14:04
*** RobertBerger <RobertBerger!> has joined #yocto14:05
RPsakoman: I've been digging on your behalf and a) the issue wasn't fixed by michaels changes which means we need buildtools b) buildtools has some issues c) master buildtools doesn't work with dunfell14:05
sakomanJust sipping my first cup of coffee, so I'm almost awake ;-)14:05
RPsakoman: I put the jinja2 changes onto dunfell-next and there is a buildtools baking from that14:06
sakomanWell that's not good news14:06
RPsakoman: those changes probably can't be backported :(14:06
RPI thought I'd create a hacked up buildtools to unblock things for now14:07
RPsakoman: might unbreak master buildtools on dunfell14:07
sakomanThat sounds reasonable14:07
RPsakoman: this is all a bit of a mess :(14:07
RPsakoman: I did add a single change to dunfell to try and untangle the mess too14:07
RPa simple one14:07
sakomanLooks like a reasonable change. Too bad it didn't work14:10
RPsakoman: dunfell buildtools build and added to helper, running a new perf test build now14:10
sakomanFingers crossed14:11
RPsakoman: well, it partially works, we need that *and* a buildtools with jinja214:11
RPsakoman: if this works we then get to decide what to do with the changes in dunfell-next14:11
RP(post 3.1.3)14:11
RPsakoman: I would probably leave that bit for you ;-)14:11
sakomanNo worries, I 'll take care of that.  Post 3.1.3 we also need to chat about all of those dangerous bitbake changes you'd like to see in dunfell ;-)14:13
sakomanI'm going to need help sorting out the minimum set14:13
*** mous16 <mous16!> has joined #yocto14:16
mous16Good morning, everybody14:17
RPsakoman: yes, bitbake is diverging a bit now :)14:26
sgwMorning all14:38
sgwRP: I started looking at tracking networking failures in the oetest/testimage area, it's been a long time since I used that functionality.  Who knows that code well these days it seems untouched for a few years now (which is a good thing in some ways).14:41
ptsnevesT_UNIX you can find a good model for the test of your patch here,,,20,0,0,0::recentpostdate%2Fsticky,,,20,2,0,7689719314:41
RPsgw: I'm probably our main expert, for better or worse :/14:44
ptsneves@sgw i do not think that code is good. It is just that most people do not use it. There are controllers which i think are not working with the current code14:44
*** jobroe <jobroe!> has quit IRC14:45
sgwptsneves: I guess stable for some version of stable for the Autobuilders and Qemu are working.14:46
ptsnevesthe qemu was always working well but there was quite a divergence in code for qemu and other controllers.14:47
ptsnevesmeaning qemu had special code and did not use controllers, at least at the time. I put some patches some time ago to that code14:47
ptsneveslet me check if it is still the case14:48
sgwRP: I was trying to trace the basic setUp/tearDown code in meta/lib/oeqa/ and maybe my bitbake/yocto fu is just rusty, but I was not get bb.debug() output, should that be using more generic python (which I thought bb.debug was)14:49
RPptsneves: I think things do still need tweaks for non-qemu14:49
RPsgw: there is no "bb" inside oe-selftest so you need to use logger14:50
RPsgw: in some cases like testimage there is but not all14:50
sgwRP: ok I will putter around more and try to regain my mojo!14:51
*** william-yocto <william-yocto!> has joined #yocto14:51
*** vmeson <vmeson!> has joined #yocto14:51
RPsgw: I'd probably look at adding some kind of debug fall back when the ssh commands fail, i.e. hook it on ssh failure rather than setup/teardown14:51
sgwRP: just know I am digging around to try and catch networking failures, well mostly ping and ssh failures and validate if qemu is still alive via serial as you suggested.  Is there a bug I should take ownership of?14:52
ptsnevessgw not sure if it works for the tearup/down but i normally put the error messages in the asserts14:52
RPsgw: there is an ltp one which basically needs this to move forward14:52
RPptsneves: that gives you a one shot message but not a trace14:53
sgwsure I can look that direction, right now, I am just digging around and refreshing my memory.14:53
RPsgw: "no route to host" search in the bugzilla would probably find the bugs14:54
*** william-yocto <william-yocto!> has quit IRC14:57
sgwfound this one:
sgwbug #1400214:57
sgwI thought we had some irc magic that gave bug summaries14:58
RPsgw: yes, that is one14:58
RPsgw: yocti should do that14:58
sgwOk I will take ownership of that one from you.14:58
RPsgw: may be because you've already mentioned the url14:58
RPsgw: currently with Bruce, not me. Not sure why14:59
sgwzeddii: can I steal 14002 away from you?14:59
ptsnevesby the way i have a test framework with nfs boot for any kernel which is configured correctly. The nfs share is served from a wic image that is reloaded with every nfs connection15:00
zeddiisgw: sure. I won't be debugging ssh connection failures any time soon :D15:10
sgwzeddii: figured you might be a bugholder!  I will take it on.15:13
T_UNIXptsneves, : thanks for the feedback15:20
*** chris_ber <chris_ber!~quassel@> has quit IRC15:34
*** gsalazar <gsalazar!5e3dbd6b@gateway/web/cgi-irc/> has quit IRC15:39
*** gsalazar <gsalazar!5e3dbd6b@gateway/web/cgi-irc/> has joined #yocto15:41
*** kiwi_29 <kiwi_29!> has joined #yocto15:46
*** jmiehe <jmiehe!> has quit IRC15:51
*** kiwi_29 <kiwi_29!> has quit IRC15:53
*** kiwi_29 <kiwi_29!> has joined #yocto15:56
*** kiwi_29 <kiwi_29!> has quit IRC15:59
*** kiwi_29 <kiwi_29!> has joined #yocto16:04
*** kpo_ <kpo_!> has quit IRC16:08
*** kpo_ <kpo_!> has joined #yocto16:09
*** xtron <xtron!~xtron@> has joined #yocto16:16
*** fl0v0 <fl0v0!~fvo@> has quit IRC16:16
*** stacktrust <stacktrust!> has quit IRC16:21
*** stacktrust <stacktrust!> has joined #yocto16:23
*** stacktrust <stacktrust!> has quit IRC16:27
*** stacktrust <stacktrust!> has joined #yocto16:30
*** Sandrita83 <Sandrita83!d0586e2e@gateway/web/cgi-irc/> has quit IRC16:35
*** mckoan is now known as mckoan|away16:42
*** stacktrust <stacktrust!> has quit IRC16:47
kiwi_29Hello. My autotools based project for which I wrote yocto recipe fails with this message in autotools_do_configure . bison : command not found . How should I change recipe so that bison is available during configure time.16:48
*** Ox861726f6c64 <Ox861726f6c64!~Ox861726f@gateway/tor-sasl/ox861726f6c64> has quit IRC16:54
RPkiwi_29: add bison-native to DEPENDS?16:56
kiwi_29RP .. got it ..thanks !16:59
*** Bunio_FH <Bunio_FH!> has quit IRC16:59
*** roussinm <roussinm!> has joined #yocto17:01
*** Bunio_FH <Bunio_FH!> has joined #yocto17:01
*** pev <pev!> has quit IRC17:05
*** rizzitello <rizzitello!~quassel@> has joined #yocto17:07
*** rizzitello <rizzitello!~quassel@> has quit IRC17:08
*** pohly <pohly!> has quit IRC17:19
*** moto-timo <moto-timo!~ttorling@fsf/member/moto-timo> has quit IRC17:22
*** sakoman <sakoman!~steve@> has quit IRC17:25
*** kiwi_29 <kiwi_29!> has quit IRC17:26
*** sakoman <sakoman!> has joined #yocto17:27
*** kiwi_29 <kiwi_29!> has joined #yocto17:32
*** camus1 <camus1!~Instantbi@> has joined #yocto17:40
*** kaspter <kaspter!~Instantbi@> has quit IRC17:42
*** camus1 is now known as kaspter17:42
*** gallegoberlines <gallegoberlines!> has joined #yocto17:45
*** xtron <xtron!~xtron@> has quit IRC17:45
*** gallegoberlines <gallegoberlines!> has quit IRC17:46
manuel1985Does Yocto build on ARM?17:48
zeddiiarm-arm or cross arm ?17:49
zeddiiyes. It should. There's something in the autobuilder, and jonmason would know more.17:50
manuel1985Alright, thank you.17:50
sakomanRP: looks like your buildtools fixes worked!17:50
jonmasonarm64-arm64, not tried arm-arm17:50
ptsnevesarm arm must be mighty slow :)17:51
jonmasonyes, you need a new, beefy arm64 server to build in any reasonable timeframe17:52
manuel1985jonmason: Duly noted. As you seem to know a bit about arm: I never really understood the differences between the terms Arm, Arm64, aarch64, ARMv7, Armv8, and whatnot. Do you happen to know more about this?17:52
jonmasonwe're building qemuarm and qemuarmv5 on arm64 without issues17:53
manuel1985Oh yeah there's ARMv5 as well17:53
jonmasonmanuel1985: I know a fair amount17:53
manuel1985and armhf and whatnot17:53
jonmasonthe names suck17:53
jonmasonarm64 is easy, thats ARMv8.  aarch64 is just another name for arm6417:54
manuel1985I figure some of these terms specify the ABI, others are just marketing, others specify the ISA, others are product names....17:54
*** lucaceresoli <lucaceresoli!> has joined #yocto17:54
ptsnevesi think you figure right :)17:55
jonmasonAnything that's cortex5X or higher is ARMv817:55
jonmasoncortex-a5/7/9/15/etc is ARMv717:56
jonmasonthe "fun" part about arm is that you don't necessarily have to include some things17:56
manuel1985jonmason: Alright thanks. What's armhf? Hardfloat, I know, but... it's just a GCC compiler prefix, isn't it? I figure starting from some arm versions there is no chip anymore which lacks hardware-fp17:57
jonmasonso, the floating point is something that some vendors chose to do in software for ...reasons17:57
jonmasonso, the toolchain needed to know if you wanted hw or sw float17:57
jonmasonARMv8 requires HW float17:58
manuel1985Requires? Good to know. Thought you can /always/ just not use the hardware-floatingpoint-silicon and do things in software.17:58
jonmasonsimilarly, CRC was optional for ARMv8 but required for ARMv8.217:58
manuel1985Ah okay got it so it must be on the chip so it can call itself ARMv8.217:59
jonmasonobviously HW float is faster, but SW should work for everyone17:59
*** lucaceresoli_ <lucaceresoli_!~lucaceres@> has joined #yocto17:59
jonmasonsince companies are licensing the designs, its a choice.  Like when buying a car, do you get the seat warmers added to the LE or it is standard for the LX18:00
manuel1985So that's totally independent from which instruction set we use, is it? I know about Thumb, what are the official names of the other two? So I won't confuse it with other terms. They all sound so similar.18:00
*** kiwi_29 <kiwi_29!> has quit IRC18:01
manuel1985Hmmm I see. Good comparison.18:01
*** lucaceresoli <lucaceresoli!> has quit IRC18:01
jonmasonthere are NEON, VFP, Thumb, and other instructions that will work if the HW is there and break if not.  So the compiler must know (thumb should always work, but you get the idea about compiler aware)18:02
jonmasonThumb is a reduced sized instruction, so you can pack more instructions into a small amount of memory18:03
jonmasongreat for embedded, but doesn't make a ton of sense for larger things18:03
manuel1985Ui, didn't know about NEON and VFP. Thought the other two I was referring two were called Arm32 and Arm64.18:03
*** lucaceresoli_ <lucaceresoli_!~lucaceres@> has quit IRC18:03
jonmasonNEON is the SIMD stuff18:03
jonmasonVFP is vector floating point, which is the hf stuff18:04
jonmasonI've been living this page for about 2 weeks now :)18:04
jonmasonI'll eventually get the reworked arm tunings for OE-core18:05
manuel1985Thanks for the link. I skimmed it. So it seems all these names armv5, armv6, armv7 and so on are *architectures*, and Thumb for example is an *execution state*. One architecture can support several *execution states*.18:08
jonmasonunless you have old HW, you have armv7 or armv818:09
manuel1985When it comes to the little details, wording is significant. That's why I'm about serious about it :) Misconecptions can happen easily18:09
*** kiwi_29 <kiwi_29!> has joined #yocto18:10
RPsakoman: good that it works. We can hopefully get on and build 3.1.3 now then18:19
sakomanYes, I think we are good to go on 3.1.318:20
RPsakoman: I probably need to slow 3.2 M2 rc2 in first18:20
RPer, slot18:20
*** T_UNIX <T_UNIX!uid218288@gateway/web/> has quit IRC18:26
*** kiwi_29 <kiwi_29!> has quit IRC18:31
*** kiwi_29 <kiwi_29!> has joined #yocto18:37
*** sgw <sgw!> has left #yocto18:38
*** sgw <sgw!> has joined #yocto18:42
*** hpsy1 <hpsy1!~hpsy@> has quit IRC18:42
*** sgw1 <sgw1!> has joined #yocto18:42
*** hpsy <hpsy!~hpsy@> has joined #yocto18:43
*** kaspter <kaspter!~Instantbi@> has quit IRC18:45
*** camus1 <camus1!~Instantbi@> has joined #yocto18:45
*** camus1 is now known as kaspter18:47
*** kiwi_29 <kiwi_29!> has quit IRC18:50
*** lxc6 <lxc6!> has quit IRC18:54
*** rob_w <rob_w!~rob@unaffiliated/rob-w/x-1112029> has quit IRC18:56
*** opennandra <opennandra!> has joined #yocto18:57
*** sgw1 <sgw1!> has quit IRC19:03
*** sgw1 <sgw1!> has joined #yocto19:04
*** Ninic0c0 <Ninic0c0!> has quit IRC19:11
*** sgw <sgw!> has quit IRC19:12
*** kiwi_29 <kiwi_29!> has joined #yocto19:15
*** kiwi_29 <kiwi_29!> has quit IRC19:18
*** sgw <sgw!> has joined #yocto19:22
kergothHas anyone done any work on getting the nativesdk-ncurses recipe building for mingw yet?19:27
frayI thought I saw someone had.. but I might be confused19:28
*** lucaceresoli_ <lucaceresoli_!~lucaceres@> has joined #yocto19:29
*** sgw <sgw!> has quit IRC19:31
*** sgw <sgw!> has joined #yocto19:31
*** moto-timo <moto-timo!> has joined #yocto19:33
*** moto-timo <moto-timo!~ttorling@fsf/member/moto-timo> has joined #yocto19:33
kergothseems like it's supposed to be supported as of ncurses 5.8+, but the build fails still, guessing we need to configure it differently or something19:33
kergothwonder if that SDK_VENDOR fix was required for it?19:34
JPEWkergoth: If that fixes it we can backport it19:34
kergothi'll give it a try19:34
kergothI noticed that includes the -w64, so it's a possibility19:35
*** kiwi_29 <kiwi_29!> has joined #yocto19:37
JPEWkergoth: It would be nice if that works. I keep pushing back requests to get the TUI working in GDB because there's no ncurses implementation19:43
JPEWI had it working with pdcurses years ago, but GDB started using API it didn't support19:43
*** lucaceresoli_ <lucaceresoli_!~lucaceres@> has quit IRC19:44
*** kiwi_29_ <kiwi_29_!> has joined #yocto19:49
*** kiwi_29 <kiwi_29!> has quit IRC19:49
*** xtron <xtron!~xtron@> has joined #yocto19:50
*** xtron <xtron!~xtron@> has quit IRC19:51
*** xtron <xtron!~xtron@> has joined #yocto19:51
*** xtron1 <xtron1!~xtron@> has joined #yocto19:55
*** xtron <xtron!~xtron@> has quit IRC19:58
*** kiwi_29_ <kiwi_29_!> has quit IRC20:00
*** pohly <pohly!> has joined #yocto20:01
*** kiwi_29 <kiwi_29!> has joined #yocto20:04
*** sgw <sgw!> has left #yocto20:05
*** sgw1 <sgw1!> has quit IRC20:06
*** kiwi_29 <kiwi_29!> has quit IRC20:09
*** sgw <sgw!> has joined #yocto20:10
*** mous16 <mous16!> has quit IRC20:21
*** feddischson <feddischson!> has quit IRC20:39
*** lucaceresoli_ <lucaceresoli_!~lucaceres@> has joined #yocto20:39
*** opennandra <opennandra!> has quit IRC20:43
kergothJPEW: got it. if you kill the cf_cv_working_poll=yes which is done unconditionally via the environment (really?) in oe-core, and switch from --with-termlib=tinfo to --enable-term-driver --enable-sp-funcs, nativesdk-ncurses will build for mingw, regardless of SDK_VENDOR20:47
kergoththe ncurses recipe is weird. it shoves a ton of arguments into a wrapper around oe_runconf (ncurses_configure) instead of just using EXTRA_OECONF20:48
*** mirzak <mirzak!sid303002@gateway/web/> has quit IRC20:49
*** dl9pf <dl9pf!sid395223@opensuse/member/dl9pf> has quit IRC20:49
*** awafaa <awafaa!sid716@gateway/web/> has quit IRC20:49
yannI have troubles with a cmake module: I have put FindCEF.cmake in ${datadir}/cmake/Modules/ and it does get into the recipe-sysroot of dependent packages, but then find_package(CEF) complains it cannot find the file, even though it is present in the sysroot directory added to CMAKE_MODULE_PATH in toolchain.cmake.  Any idea what could be wrong ?20:50
*** mdp <mdp!sid49840@gateway/web/> has quit IRC20:50
*** ndec <ndec!sid219321@linaro/ndec> has quit IRC20:50
*** ric96 <ric96!sid234506@gateway/web/> has quit IRC20:50
*** rhadye <rhadye!sid217449@gateway/web/> has quit IRC20:50
*** paulbarker <paulbarker!sid269702@gateway/web/> has quit IRC20:50
*** Crofton|cloud <Crofton|cloud!sid401373@gateway/web/> has quit IRC20:50
*** jonmason <jonmason!sid36602@gateway/web/> has quit IRC20:50
*** georgem_home <georgem_home!uid210681@gateway/web/> has quit IRC20:51
*** thierryE <thierryE!sid286446@gateway/web/> has quit IRC20:51
*** lh__ <lh__!sid77898@gateway/web/> has quit IRC20:51
*** ribalda <ribalda!sid306640@gateway/web/> has quit IRC20:51
*** awafaa <awafaa!sid716@gateway/web/> has joined #yocto20:51
*** dl9pf <dl9pf!sid395223@opensuse/member/dl9pf> has joined #yocto20:51
*** diamondman <diamondman!sid306859@gateway/web/> has quit IRC20:51
kergothJPEW: still failsi to install due to assumptions about .so and linker scripts, but that should be easy enough to resolve. i'll prep a couple pull reqs when i finish up20:51
*** mirzak <mirzak!sid303002@gateway/web/> has joined #yocto20:51
*** madisox <madisox!sid453692@gateway/web/> has quit IRC20:51
*** bradfa <bradfa!sid297668@gateway/web/> has quit IRC20:51
*** nohit <nohit!sid334887@gateway/web/> has quit IRC20:51
*** jonmason <jonmason!sid36602@gateway/web/> has joined #yocto20:52
*** ndec <ndec!sid219321@linaro/ndec> has joined #yocto20:52
*** nohit <nohit!sid334887@gateway/web/> has joined #yocto20:52
*** Crofton|cloud <Crofton|cloud!sid401373@gateway/web/> has joined #yocto20:52
*** georgem_home <georgem_home!uid210681@gateway/web/> has joined #yocto20:52
*** ribalda <ribalda!sid306640@gateway/web/> has joined #yocto20:52
*** paulbarker <paulbarker!sid269702@gateway/web/> has joined #yocto20:53
*** thierryE <thierryE!sid286446@gateway/web/> has joined #yocto20:53
*** bradfa <bradfa!sid297668@gateway/web/> has joined #yocto20:53
*** lh__ <lh__!sid77898@gateway/web/> has joined #yocto20:53
*** rhadye <rhadye!sid217449@gateway/web/> has joined #yocto20:53
*** diamondman <diamondman!sid306859@gateway/web/> has joined #yocto20:53
*** ric96 <ric96!sid234506@gateway/web/> has joined #yocto20:53
*** mdp <mdp!sid49840@gateway/web/> has joined #yocto20:53
*** madisox <madisox!sid453692@gateway/web/> has joined #yocto20:54
yannd'oh, ordering problem20:56
*** kiwi_29 <kiwi_29!> has joined #yocto20:58
JPEWkergoth: Cool. Is there any easy to test it?20:59
kergothmy goal is the same one you mentioned, interactive gdb-cross-canadian on windows. ideally also python, but that might require a recipe that grabs an upstream build..21:01
JPEWHeh, ya. Python in MinGW would be *really* nice, but I don't think it will be easy21:01
*** kiwi_29 <kiwi_29!> has quit IRC21:02
*** Konsgnx <Konsgnx!> has quit IRC21:03
*** kiwi_29 <kiwi_29!> has joined #yocto21:03
*** abelloni <abelloni!> has joined #yocto21:03
kergothJPEW: hmm, most mingw dlls end up in bindir next to the executables, as is needed due to dll search paths, but base_libdir and libdir are still lib, so some additional installation logic ends up breaking. I'm assuming we don't want to set those to bin due to extranous non-dll files ending up there, so is your process to just explicitly skip suchl ogic for mingw specifically?21:08
*** ptsneves <ptsneves!b0dd7824@> has quit IRC21:09
*** pohly <pohly!> has quit IRC21:10
*** lucaceresoli_ <lucaceresoli_!~lucaceres@> has quit IRC21:11
JPEWIt's possible it's just never really come up before21:11
JPEWkergoth: AKA I'm not sure :)21:11
moto-timoI see corner cases everywhere21:12
RPkergoth: I know some of the original decisions I made in there were simply due to lack of experience and knowledge of windows :(21:12
moto-timoRP: the correct decision21:13
RPmoto-timo: you need circular rooms ;-)21:13
moto-timo(at least at the time)21:13
moto-timoRP: yes. yes I do.21:13
moto-timowhat was the circular face watch denix had? motorola something?21:13
moto-timono corners21:13
moto-timogrr. what to do when kernel panics because it isn't finding rootfs? Is it my initramfs? is it my wic? dm-verity hates me21:15
JPEWmoto-timo: It's too hard to put Windows in a curved wall, hence all the corners21:15
moto-timoJPEW: round curved windows21:15
moto-timobetter yet, spherical like a deep diving submarine21:15
*** berton <berton!~berton@> has quit IRC21:17
* RP watches pseudo hang at his changes. I think this means I don't remember C :/21:18
moto-timoRP rewrites it in Rust21:19
kergothAny idea what a '.dll.a' file is? :) admittedly i know nothing about windows development21:20
JPEWkergoth: static library maybe?21:21
kergothugh, do_split_packages for .so* against libdir do not do what you'd hope on windows since the files don't live there.21:21
moto-timokergoth: statically linked dynamicly linked library21:21
moto-timothat would follow with old MS code style... as long as it had 5000 #ifdef clauses in it21:21
moto-timoyou can't make this up!
kergothugh, a lot of FILES lack ${HOST_EXEEXT} suffixes on binary names too, of course, cause who remembers to use that?21:22
kergothmoto-timo: haha21:23
JPEWOh, right21:23
JPEWLinking is *really* strange in MinGW21:23
moto-timoat least it's not cygwin21:24
*** xtron1 <xtron1!~xtron@> has quit IRC21:24
moto-timoI'll take msys2 any day of the week21:24
JPEWkergoth: I think they are "import libraries":
JPEWBasically little static libraries that help the linker link against a DLL21:25
kergothhuh. so they're static, but really they belong in -dev, not -staticdev21:26
kergothbecause of course.21:26
kergothi think i may be done for today.. :)21:26
JPEWkergoth: I believe so21:26
* RP was happy not to maintain meta-mingw21:26
* RP wonders if this revised patch will lock up bitbake21:27
moto-timoJPEW: that sounds better than the SO link I found21:28
moto-timoERROR: btrfs-tools-5.7-r0 do_package_qa: QA Issue: /usr/bin/btrfs-convert contained in package btrfs-tools requires, but no providers found in RDEPENDS_btrfs-tools? [file-rdeps]21:33
moto-timoERROR: btrfs-tools-5.7-r0 do_package_qa: QA Issue: /usr/bin/btrfs-convert contained in package btrfs-tools requires, but no providers found in RDEPENDS_btrfs-tools? [file-rdeps]21:33
moto-timowhat new hell is this? nothing should have changed that!21:33
*** beneth <beneth!> has left #yocto21:34
moto-timobtrfs-tools DEPENDS on e2fsprogs21:34
moto-timo"it built fine yesterday"21:34
*** ericch <ericch!> has quit IRC21:53
*** ericch <ericch!> has joined #yocto21:55
*** mirzak <mirzak!sid303002@gateway/web/> has quit IRC21:55
*** mirzak <mirzak!sid303002@gateway/web/> has joined #yocto21:57
*** manuel1985 <manuel1985!> has quit IRC22:04
*** manuel1985 <manuel1985!> has joined #yocto22:18
*** risca <risca!~quassel@> has quit IRC22:26
*** kiwi_29 <kiwi_29!> has quit IRC22:33
*** kiwi_29 <kiwi_29!> has joined #yocto22:33
*** manuel1985 <manuel1985!> has quit IRC22:34
*** kaspter <kaspter!~Instantbi@> has quit IRC22:36
*** camus1 <camus1!~Instantbi@> has joined #yocto22:36
*** maudat <maudat!> has quit IRC22:38
*** camus1 is now known as kaspter22:38
*** kiwi_29 <kiwi_29!> has quit IRC22:48
*** ernstp <ernstp!sid168075@gateway/web/> has quit IRC22:50
*** mithro <mithro!sid24875@gateway/web/> has quit IRC22:50
*** Crofton|cloud <Crofton|cloud!sid401373@gateway/web/> has quit IRC22:50
*** ribalda <ribalda!sid306640@gateway/web/> has quit IRC22:50
*** ric96 <ric96!sid234506@gateway/web/> has quit IRC22:50
*** diamondman <diamondman!sid306859@gateway/web/> has quit IRC22:50
*** rburton <rburton!sid1738@gateway/web/> has quit IRC22:50
*** ndec <ndec!sid219321@linaro/ndec> has quit IRC22:50
*** bradfa <bradfa!sid297668@gateway/web/> has quit IRC22:50
*** dl9pf <dl9pf!sid395223@opensuse/member/dl9pf> has quit IRC22:50
*** marex-cloud_ <marex-cloud_!sid137234@gateway/web/> has quit IRC22:50
*** nohit <nohit!sid334887@gateway/web/> has quit IRC22:51
*** madisox <madisox!sid453692@gateway/web/> has quit IRC22:51
*** thierryE <thierryE!sid286446@gateway/web/> has quit IRC22:51
*** georgem_home <georgem_home!uid210681@gateway/web/> has quit IRC22:51
*** jonmason <jonmason!sid36602@gateway/web/> has quit IRC22:51
*** awafaa <awafaa!sid716@gateway/web/> has quit IRC22:51
*** lh__ <lh__!sid77898@gateway/web/> has quit IRC22:51
*** paulbarker <paulbarker!sid269702@gateway/web/> has quit IRC22:51
*** Crofton|cloud <Crofton|cloud!sid401373@gateway/web/> has joined #yocto22:52
*** ndec <ndec!sid219321@linaro/ndec> has joined #yocto22:52
*** dagmcr <dagmcr!sid323878@gateway/web/> has quit IRC22:52
*** dagmcr <dagmcr!sid323878@gateway/web/> has joined #yocto22:52
*** nohit <nohit!sid334887@gateway/web/> has joined #yocto22:52
*** rhadye <rhadye!sid217449@gateway/web/> has quit IRC22:52
*** marex-cloud_ <marex-cloud_!sid137234@gateway/web/> has joined #yocto22:53
*** bradfa <bradfa!sid297668@gateway/web/> has joined #yocto22:53
*** georgem_home <georgem_home!uid210681@gateway/web/> has joined #yocto22:53
*** paulbarker <paulbarker!sid269702@gateway/web/> has joined #yocto22:54
*** thierryE <thierryE!sid286446@gateway/web/> has joined #yocto22:54
*** diamondman <diamondman!sid306859@gateway/web/> has joined #yocto22:54
*** ribalda <ribalda!sid306640@gateway/web/> has joined #yocto22:54
*** mithro <mithro!sid24875@gateway/web/> has joined #yocto22:54
*** madisox <madisox!sid453692@gateway/web/> has joined #yocto22:55
*** ric96 <ric96!sid234506@gateway/web/> has joined #yocto22:55
*** ernstp <ernstp!sid168075@gateway/web/> has joined #yocto22:55
*** rburton <rburton!sid1738@gateway/web/> has joined #yocto22:55
*** lh__ <lh__!sid77898@gateway/web/> has joined #yocto22:55
*** dl9pf <dl9pf!sid395223@opensuse/member/dl9pf> has joined #yocto22:55
*** rhadye <rhadye!sid217449@gateway/web/> has joined #yocto22:55
*** awafaa <awafaa!sid716@gateway/web/> has joined #yocto22:56
*** kiwi_29 <kiwi_29!> has joined #yocto22:56
*** jonmason <jonmason!sid36602@gateway/web/> has joined #yocto22:57
*** kiwi_29 <kiwi_29!> has quit IRC22:59
RPfray, seebs: I implemented a quick hacky client side path filter. For a sample recipe, it took the pseudo database entries from 45,000 entries down to 15622:59
RPIt also exposed some glitches I need to understand :/23:00
*** kiwi_29 <kiwi_29!> has joined #yocto23:00
RP - I know one issues it the early return is missing from fd dup code23:03
RP is a hacked up single recipe test of it23:03
* RP sleeps on it23:05
RP(the early return means its missing some fd dup code - looks like I really do need to sleep)23:05
*** kiwi_29 <kiwi_29!> has quit IRC23:07
*** kiwi_29 <kiwi_29!> has joined #yocto23:08
*** kiwi_29 <kiwi_29!> has quit IRC23:15
*** ericch <ericch!> has quit IRC23:43
*** kpo_ <kpo_!> has quit IRC23:48
*** kpo_ <kpo_!> has joined #yocto23:49

Generated by 2.17.2 by Marius Gedminas - find it at!