ChanServ changed the topic of #haiku to: Open-source operating system that specifically targets personal computing. | https://haiku-os.org | Nightlies: https://download.haiku-os.org | Bugtracker: https://dev.haiku-os.org | SCM: https://git.haiku-os.org/ | Logs: https://oftc.irclog.whitequark.org/haiku | Matrix: #haiku:matrix.org | XMPP: #haiku%irc.oftc.net@irc.jabberfr.org
Coldfirex has joined #haiku
dpirate has joined #haiku
Asterisk has joined #haiku
<nekobot> [haiku/haiku] waddlesplash pushed 1 commit to master [hrev57963] - https://git.haiku-os.org/haiku/log/?qt=range&q=a26c33bd9764+%5E72be912bae8c
<nekobot> [haiku/haiku] a26c33bd9764 - DefaultBuildProfiles: We now have Python 3.10.
grobe0ba has quit [Quit: ZNC 1.8.2 - https://znc.in]
grobe0ba has joined #haiku
dalme_ has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
dalme_ has joined #haiku
mmu_man has quit [Ping timeout: 480 seconds]
itaipu has quit [Ping timeout: 480 seconds]
<botifico> [haiku/website] waddlesplash pushed 1 commit to master [+2/-1/±0] https://github.com/haiku/website/compare/badcfdd387e1...82b6c66402c5
<botifico> [haiku/website] trungnt2910 82b6c66 - blog/trungnt2910: Debugging progress 3 (#700)
itaipu has joined #haiku
coolcoder613_ has joined #haiku
coolcoder613 has quit [Read error: Connection reset by peer]
dalme_ has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
coolcoder613_ has quit [Ping timeout: 480 seconds]
SLema has quit [Quit: Vision[]: i've been blurred!]
coolcoder613 has joined #haiku
Skipp_OSX has joined #haiku
Skipp_OSX_ has joined #haiku
Skipp_OSX_ has quit []
Skipp_OSX_ has joined #haiku
Skipp_OSX has quit [Ping timeout: 480 seconds]
vdamewood has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
<botifico> [haiku/website] waddlesplash pushed 1 commit to master [+1/-0/±0] https://github.com/haiku/website/compare/82b6c66402c5...6f8f90fa5696
<botifico> [haiku/website] waddlesplash 6f8f90f - Activity Report, July.
linuxmaster has joined #haiku
vdamewood has joined #haiku
AlienSoldier has joined #haiku
diver1 has joined #haiku
diver has quit [Ping timeout: 480 seconds]
coolcoder613_ has joined #haiku
coolcoder613 has quit [Read error: Connection reset by peer]
coolcoder613 has joined #haiku
coolcoder613_ has quit [Read error: Connection reset by peer]
AlienSoldier has quit [Quit: Vision[]: i've been blurred!]
Chai-T-Rex has quit [Remote host closed the connection]
ChaiTRex has joined #haiku
HaikuUser has joined #haiku
HaikuUser has quit []
TMM has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
TMM has joined #haiku
tuaris has quit [Quit: Leaving.]
Begasus has joined #haiku
<Begasus> g'morning peeps
<Begasus> k, jmairboeck thanks :) fixed install part on this end for fluidsynth2 now also :)
AD_Haiku_onPC_ has joined #haiku
<Begasus> eeps: 0% tests passed, 26 tests failed out of 26 :P
AD_Anime2 has quit [Ping timeout: 480 seconds]
<Begasus> better :D 100% tests passed, 0 tests failed out of 26
zayd has quit [Quit: ZNC 1.8.2+deb2+deb11u1 - https://znc.in]
coolcoder613_ has joined #haiku
coolcoder613 has quit [Read error: Connection reset by peer]
<botifico> [haikuports/haikuports] Begasus pushed 1 commit to master [+0/-0/±1] https://github.com/haikuports/haikuports/compare/675a99fe8204...bb9579c207f3
<botifico> [haikuports/haikuports] Begasus bb9579c - libinstpatch, fix fixPkgconfig (#10819)
yann64 has joined #haiku
<botifico> [haikuports/haikuports] Begasus pushed 1 commit to master [+0/-14/±0] https://github.com/haikuports/haikuports/compare/bb9579c207f3...d87d1854cd07
<botifico> [haikuports/haikuports] OscarL d87d185 - qemu: drop older recipes. (#10816)
HaikuUser has joined #haiku
HaikuUser has quit []
<botifico> [haikuports/haikuports] Begasus pushed 1 commit to master [+0/-4/±0] https://github.com/haikuports/haikuports/compare/d87d1854cd07...1d9f09c05cd1
<botifico> [haikuports/haikuports] OscarL 1d9f09c - nodejs: retire older recipes. (#10802)
<botifico> [haikuports/haikuports] Begasus pushed 1 commit to master [+2/-2/±0] https://github.com/haikuports/haikuports/compare/1d9f09c05cd1...59197ef5c949
<botifico> [haikuports/haikuports] MaximeChretien 59197ef - minipro: update to version 0.7 (#10791)
yann64 has left #haiku [#haiku]
yann64 has joined #haiku
yann64 has left #haiku [#haiku]
<botifico> [haikuports/haikuports] Begasus pushed 1 commit to master [+0/-0/±1] https://github.com/haikuports/haikuports/compare/59197ef5c949...6781412355e8
<botifico> [haikuports/haikuports] OscarL 6781412 - crawl: move to Python 3.10. (#10800)
itaipu has quit [Ping timeout: 480 seconds]
Nephele has joined #haiku
<Begasus> Hi nephele
<Begasus> switched 32bit to nightly :)
<Begasus> biab, morning chores :D
Begasus_32 has joined #haiku
ChaiTRex has quit [Remote host closed the connection]
<Begasus> re, for a bit :)
ChaiTRex has joined #haiku
jmairboeck has joined #haiku
<jmairboeck> Hi Begasus
<jmairboeck> about fluidsynth: I thought the easiest solution would be to move the whole function to the .c file and thus compile it as C instead of C++. I think that the compiler would probably not inline a 45 line function anyway :) (at least by default)
mmu_man has joined #haiku
DKnoto has quit [Ping timeout: 480 seconds]
DKnoto has joined #haiku
<Begasus> hi jmairboeck
<Begasus> well, seems it wasn't really needed jmairboeck :) but if nothing else it's a start for when it "does" need a newer version :)
<Begasus> ps, fluidsynth fails to build with gcc2 here (checked with both flac arch versions)
nipos has left #haiku [Error from remote client]
nipos has joined #haiku
<Begasus> should push the changes for older flac to have a devel package again
<phschafft> Good morning. or something similar.
<Begasus> Moin phschafft :)
<Begasus> not sure what happend to jpeg_turbo x512[m], from what I can see in the source references it points to a DESTINATION path that shouldn't interfere with a build setup?
<x512[m]> It seems it want absolute paths, not relative.
<Begasus> last change on that recipe was 2 years ago, it worked then, so why doesn't it now?
<Begasus> something must have changed ?
<x512[m]> Who knows. Did you tested it?
<Begasus> I always test a build before I do any PR's :)
<x512[m]> I mean test right now.
<Begasus> k, flac13 builds fine for both arch's
<Begasus> yeah, getting the same error you mentioned in the issue
<x512[m]> Things that originally worked may break after.
<Begasus> hence my confusion
<x512[m]> This patch fix issue for me.
Nephele has quit [Quit: Vision[]: i've been blurred!]
<Begasus> had been thinking there, but $cmakeDirArgs was already in there also (from before last change)
<Begasus> also using $prefix/develop/headers etall will brake things on 32bit
Anarchos has joined #haiku
<x512[m]> I am leaving a possibility that I am doing something wrong, so it is better to test it by someone else first.
<x512[m]> It reproduce both for x86_64 and riscv64.
<x512[m]> Tried updating HaikuPorter, no effect.
<Begasus> x512[m], try adding (in the original recipe) "-DCMAKE_INSTALL_INCLUDEDIR=$includeDir \"
<Begasus> that worked here
<Begasus> weirdly enough
<jmairboeck> doesn't $cmakeDirArgs use the relative variants?
<jmairboeck> I think that might be the problem here
<Begasus> not sure what is happening there jmairboeck, $cmakeDirArgs should deal with that
<x512[m]> It seems not only CMAKE_INSTALL_INCLUDEDIR is wrong.
<Begasus> k, fluidsynth2 builds fine on _x86 with older and new flac version
<jmairboeck> if $cmakeDirArgs uses a relative path but the CMakeLists.txt expects an absolute one, that won't work
<x512[m]> libjpeg_turbo do some obscure things in its CmakeLists.txt.
<Begasus> jikes
<Begasus> then something has changed in cmake jmairboeck, as it worked fine 2 years ago?
<jmairboeck> maybe the $cmakeDirArgs definition was changed in haikuporter?
<jmairboeck> I think CMake GNUInstallDirs expects relative paths, so as a default it makes sense I think
<Begasus> that was added after the last change to jpeg-turbo
<jmairboeck> yes, so it looks like that recipe needs to be fixed
<Begasus> it should be an easy fix, but even then, won't this bite us with other $cmakeDirArgs later on?
<jmairboeck> if there are any other weird CMake scripts that don't use default stuff
<Begasus> I don't see anything wrong really with this statement: INCLUDES DESTINATION ${CMAKE_INSTALL_INCLUDEDIR}
<Begasus> ah: include(cmakescripts/GNUInstallDirs.cmake)
<Begasus> they are indeed using something weird in there
<Begasus> x512[m], try this https://bpa.st/62XA
_-Caleb-_ has left #haiku [#haiku]
_-Caleb-_ has joined #haiku
<Anarchos> jmairboeck: do you know if ghostscript has a device for haiku ? With gs 8 , there was a device bealpha4 for beos.
<jmairboeck> Anarchos: I don't know
AD_Anime2 has joined #haiku
AD_Haiku_onPC_ has quit [Ping timeout: 480 seconds]
mmu_man has quit [Ping timeout: 480 seconds]
<Anarchos> jmairboeck: it is the last stopper for BeTeX : having a n invocation on ghostscript to preview .gs files. Or should i add a requirement on BGhostView ?
mmu_man has joined #haiku
<botifico> [haikuports/haikuports] Begasus pushed 1 commit to master [+1/-1/±0] https://github.com/haikuports/haikuports/compare/6781412355e8...4bc458ac9b1d
<botifico> [haikuports/haikuports] phw 4bc458a - picard: bump version to 2.12.1 (#10828)
<botifico> [haikuports/haikuports] Begasus pushed 1 commit to master [+2/-2/±0] https://github.com/haikuports/haikuports/compare/4bc458ac9b1d...e9c1a70b3a60
<botifico> [haikuports/haikuports] OscarL e9c1a70 - links: update to version 2.30. (#10823)
<Begasus> jmairboeck, if guile3 is fine I guess you could merge that one too?
mmu_man has quit [Ping timeout: 480 seconds]
<jmairboeck> There was the question whether it should become the default instead of 2.2. What do you think?
<Begasus> I guess it couldn't hurt to move forward
<Begasus> as long as there isn't anything breaking current deps on the older one :)
<Begasus> haven't looked deeply enough, does it have a conflict on the devel package?
<jmairboeck> also, I didn't do any "real-world" testing yet, and some of the test cases that come with it crash, although I haven't really investigated yet
<jmairboeck> no conflicts, that's why I have used --program-suffix=-3.0 to avoid that
<Begasus> hmm in that case maybe keep the older one as default
<jmairboeck> that was the other question: should they conflict instead?
<Begasus> can lilypond use the new cmd suffixed?
<jmairboeck> it doesn't require the commands, only the library
<Begasus> as said, haven't looked enough, does it provide a devel package with a *.so link?
<jmairboeck> the libraries have the version in their name anyway, the headers also in the path
mmu_man has joined #haiku
<jmairboeck> it's only the tools and the config script that would conflict
<Begasus> devel:libguile_3.0$secondaryArchSuffix should be no conflict on that
<jmairboeck> exactly
<Begasus> guile-config conflicts though
<jmairboeck> yes
<jmairboeck> when not using --program-suffix
<Begasus> why does the devel package require the tools one (don't see that in the other recipes)?
<Begasus> if not stricktly needed I would try to avoid it there, if something requires one of the cmd's in tools it should require that cmd then?
<jmairboeck> I added that because guile-config is written in scheme and requires gule to run
<jmairboeck> I could replace that with cmd:guile_3.0 if you like that better
HaikuUser has joined #haiku
<Begasus> k, so "required" for devel then :)
<HaikuUser> Hi all.
<Begasus> nah it's ok, as long as there are no "unneeded" cycling deps around :)
<Begasus> Hi HaikuUser
<HaikuUser> Have to change name
<Begasus> ;)
rann has quit [Quit: Connection closed for inactivity]
<Begasus> had my share with gettext jmairboeck :)
<Begasus> and kio/kcmutils ...
Pitmaster has joined #haiku
<Pitmaster> This must be better
HaikuUser has quit []
<Begasus> ah, hey buurman :)
<Pitmaster> Buurman? Ik?
* Begasus guesses :)
<Begasus> ja, noorderbuur :)
<Pitmaster> Oke, dan woon jij zuiderlijk van mij. Breda?
<Begasus> B Limburg :)
<Pitmaster> Oke, maar waar ken ik je van? Of jij mij?
<Begasus> nergens (denk ik), zag bij je login *.nl staan :)
<Pitmaster> Ah, zijn er zo weinig NL-ers hier?
<Begasus> niet zoveel neen, of niet zichtbaar via nl provider*
<Begasus> back to English * :)
<Pitmaster> Er is een NL groep, ik ga daar even kennis maken
<Begasus> there is still #haiku-nl (calm one though) :)
unimatrix has joined #haiku
yeti has quit [Ping timeout: 480 seconds]
unimatrix is now known as yeti
Pitmaster has quit [Quit: Vision[]: i've been blurred!]
_-Caleb-_ has left #haiku [#haiku]
_-Caleb-_ has joined #haiku
<Begasus> jmairboeck, could you use the command-suffix on the older guile versions too? (haven't checked if the tools are used by anything, but I doubt that, don't have any of the tools packages for it installed) :)
<botifico> [haikuports/haikuports] Begasus pushed 1 commit to master [+0/-0/±1] https://github.com/haikuports/haikuports/compare/e9c1a70b3a60...57619f93c455
<botifico> [haikuports/haikuports] Begasus 57619f9 - jpeg_turbo, revbump, fixes build (after changes in haikuporter $cmakeDirArgs handling) (#10829)
_-Caleb-_ has left #haiku [#haiku]
tuaris has joined #haiku
<jmairboeck> Begasus: yes, I could switch guile 2.2 to use that and remove it from 3.0 instead. I'll leave guile 1 as is, as that is just a legacy version.
_-Caleb-_ has joined #haiku
dalme_ has joined #haiku
<Begasus> /Opslag/haikuports/app-office/texmacs/texmacs-1.99.12.recipe:64: cmd:guile == 1.8.8
<Begasus> one found :)
<Begasus> k, that should still be fine I guess
zard has joined #haiku
zard has quit []
zard has joined #haiku
OrangeBomb has quit [Quit: Slacking off]
Coldfirex has quit [Remote host closed the connection]
OrangeBomb has joined #haiku
<botifico> [haikuports/haikuports] jmairboeck pushed 1 commit to master [+2/-2/±0] https://github.com/haikuports/haikuports/compare/57619f93c455...4b5b43edcff9
<botifico> [haikuports/haikuports] jmairboeck 4b5b43e - lilypond: bump version (#10787)
<botifico> [haikuports] jmairboeck deleted branch lilypond - https://github.com/haikuports/haikuports
<botifico> [haikuports/haikuports] jmairboeck deleted branch lilypond
itaipu has joined #haiku
itaipu has quit [Remote host closed the connection]
Coldfirex has joined #haiku
itaipu has joined #haiku
<jmairboeck> I think we should change the compat versions of the commands of texinfo 7 so that we can force the usage of it on 32 bit.
<jmairboeck> lilypond doesn't build with texinfo-6.1 (as expected) but that was the version that got picked by the x86_gcc2 build
<Anarchos> oh i see there is a new Activity & Contract report from waddlesplash !! Coool, will read it
<Begasus> you could use what PulkoMandy mentioned on the issue jmairboeck
<jmairboeck> which issue?
<jmairboeck> that's not the problem here, this isn't about trying to build something with gcc2, the problem is that texinfo 7.1 says "compat >= 4" mostly, so it is the same as the gcc2 version.
<Begasus> eeps
<jmairboeck> but that is obviously wrong because if they were fully compatible it wouldn't matter which version would be used for building
<Begasus> right, so fix the compat version? ;)
<Begasus> nextclout-client seems broken on 32bit nightly?
<Begasus> even with a rebuild I'm getting "TLS initialization failed"
<Begasus> and still no changes ... tsss :P
<jmairboeck> that's how I found it, through "blame" :)
Begasus_32 has quit [Quit: Vision[]: Gone to the dogs!]
<Begasus> https://repology.org/ down?
<Anarchos> Begasus: korli seems to be on vacation :) no new review on my BeTeX patch :)
<Begasus> don't know there Anarchos, he's been in the channel last week though :)
dqk_ has joined #haiku
<Begasus> jmairboeck, maybe an idea to add portVersionCompat there?
dqk has quit [Ping timeout: 480 seconds]
<Begasus> bbl
<botifico> [haiku/website] pulkomandy pushed 1 commit to master [+1/-0/±0] https://github.com/haiku/website/compare/6f8f90fa5696...0c6202990f53
<botifico> [haiku/website] Zardshard 0c62029 - Post GSOC status update
<botifico> [haikuports/haikuports] pulkomandy pushed 1 commit to master [+0/-0/±2] https://github.com/haikuports/haikuports/compare/4b5b43edcff9...666710323db7
<botifico> [haikuports/haikuports] jmairboeck 6667103 - texinfo: fix compat versions
zayd has joined #haiku
linuxmaster has quit [Ping timeout: 480 seconds]
BrunoSpr has joined #haiku
BrunoSpr has quit [Quit: Vision[]: i've been shreederd!]
<botifico> [haikuports/haikuports] jmairboeck pushed 1 commit to master [+0/-0/±1] https://github.com/haikuports/haikuports/compare/666710323db7...afb096c20ba2
<botifico> [haikuports/haikuports] jmairboeck afb096c - lilypond: require texinfo 7 specifically
TMM has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
TMM has joined #haiku
yann65 has joined #haiku
andymccall has joined #haiku
<andymccall> Hello!
<andymccall> There's a load of comments in the code from very early days of Haiku which has got an old/invalid email address of mine in. Even though it's only commends, can I submit an MR to update them with my new valid email address?
<zard> Probably? Don't see any reason why not
<Anarchos> andymccall: sure
Anarchos has quit [Quit: Page closed]
mmu_man has quit [Ping timeout: 480 seconds]
yann65 has quit [Remote host closed the connection]
mmu_man has joined #haiku
HaikuUser has joined #haiku
HaikuUser has quit []
nosycat has joined #haiku
<waddlesplash> andymccall: sure, of course. or you can just remove it, whatever you want :)
FreeFull has quit [Quit: rebooting]
FreeFull has joined #haiku
mmu_man has quit [Ping timeout: 480 seconds]
mmu_man has joined #haiku
erysdren has joined #haiku
HaikuUser has joined #haiku
<erysdren> morning
<nosycat> Hello!
<Skipp_OSX_> major breakthrough this morning, bi-directional color pickers support is working!
Skipp_OSX_ is now known as Skipp_OSX
HaikuUser has quit [Quit: Vision[]: i've been blurred!]
<zdykstra> that's awesome!
<PulkoMandy> let's see if I can build webkit_gtk today (thanks to everyone involved for fixing the problems in libevent and other things I reported :) )
figment has joined #haiku
dby has joined #haiku
Anarchos has joined #haiku
<botifico> [haikuports/haikuports] pulkomandy pushed 1 commit to master [+0/-0/±1] https://github.com/haikuports/haikuports/compare/afb096c20ba2...0ed615725783
<botifico> [haikuports/haikuports] pulkomandy 0ed6157 - haikuwebkit: remove now unneeded -fpermissive option
<Anarchos> PulkoMandy nice move !
tqh has joined #haiku
nosycat has quit [Quit: Leaving]
mmu_man has quit [Ping timeout: 480 seconds]
<zard> Nice, I see you're working on IOM's color picker :-)
* zard remembers you mentioning you were working on it last year
<Skipp_OSX> yeah I've been working on it a while plugging away... I've ripped out the Icon-O-Matic color picker and implemented my own so you can have it everywhere.
<zard> Little by little... +1
gouchi has joined #haiku
<Skipp_OSX> the bi-directional support was a major headache but we gotta have it in Icon-O-Matic at least.
<Skipp_OSX> The picker framework I've developed can be made to work with other kinds of pickers too, for example font pickers.
BrunoSpr has joined #haiku
mmu_man has joined #haiku
BrunoSpr has quit [Quit: Vision[]: Ich wurde eingeweicht!]
BrunoSpr has joined #haiku
BrunoSpr has quit [Quit: Vision[]: Ich wurde eingeweicht!]
<Begasus> closing down here
<Begasus> nice work Skipp_OSX! +1 :)
<Begasus> cu peeps!
Begasus has quit [Quit: Vision[]: i've been blurred!]
<phschafft> hm.
* phschafft wonders what the status of librhash is on Haiku.
bbjimmy has quit [Quit: Vision[]: i've been blurred!]
<phschafft> 1.4.x hm.
<phschafft> seems recent.
zard has quit [Quit: leaving]
<botifico> [haikuports/haikuports] pulkomandy pushed 1 commit to master [+0/-0/±2] https://github.com/haikuports/haikuports/compare/0ed615725783...cc5e1ccc05ec
<botifico> [haikuports/haikuports] pulkomandy cc5e1cc - webkit_gtk: update build to llvm18, openssl3, icu74
mmu_man has quit [Ping timeout: 480 seconds]
mmu_man has joined #haiku
<botifico> [haikuports/haikuports] pulkomandy pushed 5 commits to master [+5/-6/±1] https://github.com/haikuports/haikuports/compare/cc5e1ccc05ec...7b113b293a4c
<botifico> [haikuports/haikuports] pulkomandy 863626f - sdllopan: update source URI
<botifico> [haikuports/haikuports] pulkomandy d5eefec - Exomizer: update to latest version, older downloads were removed
<botifico> [haikuports/haikuports] pulkomandy ad103cb - Fatsort: update to current version
<botifico> [haikuports/haikuports] ... and 2 more commits.
kallisti5[m] has quit []
tulili[m] has quit [Quit: Bridge terminating on SIGTERM]
trungnt2910[m] has quit [Quit: Bridge terminating on SIGTERM]
Begasus[m] has quit [Quit: Bridge terminating on SIGTERM]
LjL[m] has quit [Quit: Bridge terminating on SIGTERM]
x512[m] has quit []
win8linux[m] has quit [Quit: Bridge terminating on SIGTERM]
Rocket[m] has quit [Quit: Bridge terminating on SIGTERM]
scops has quit [Quit: Bridge terminating on SIGTERM]
EnriqueMG[m] has quit [Quit: Bridge terminating on SIGTERM]
jimaz[m] has quit [Read error: Connection reset by peer]
andreasdr[m] has quit [Remote host closed the connection]
ErenOld[m] has quit [Quit: Bridge terminating on SIGTERM]
skepti[m] has quit [Remote host closed the connection]
funbreaker[m] has quit [Remote host closed the connection]
ilzu[m] has quit [Remote host closed the connection]
shzef[m] has quit [Remote host closed the connection]
c0m47053[m] has quit [Remote host closed the connection]
QuARCitits[m] has quit [Remote host closed the connection]
<nipos> Are there any known issues with USB storage being slow as hell? My server where the RAID isn't supported is really really slow and it's not fun to work with it. I thought with 24 CPU cores and about 100GB RAM it would be perfect for heavy compile tasks, but due to the I/O bottleneck it feels even slower than my old computer with only 3 cores, but a SATA hard disk :(
<nipos> cargo needed 5 minutes to download 39 crates (3.1 MB). And network speed isn't the issue here...
gouchi has quit [Remote host closed the connection]
<waddlesplash> is this beta4 or nightly?
<nipos> Nightly, updated to the latest version
<waddlesplash> USB3 storage on XHCI should be fast, I have gotten >100MB/s read
<waddlesplash> if you are not on USB3 storage or XHCI then it can be slow, yes
<nipos> The USB stick supports USB 3, but the server only has USB 2 ports unfortunately. But normally USB 2 still isn't *that* slow
<waddlesplash> yes, that's true
<waddlesplash> check syslog for errors I suppose
SLema has joined #haiku
<nipos> Also booting (where it only needs to read a few big package files) is rather fast here, but accessing random small files takes ages
<dovsienko> nipos: would it make more sense to rebuild the server on an ordinary SSD(s)?
<nipos> Haiku doesn't support the RAID controller in the server, so that's not an option. It just doesn't see the internal drives at all. I have 3 unused HDDs in it that I could use otherwise.
jmairboeck has quit [Quit: Konversation terminated!]
dqk_ has quit [Ping timeout: 480 seconds]
ChaiTRex has quit [Remote host closed the connection]
<dovsienko> nipos: one of the simplest ways out of this problem space would be to plug a PCIe SATA card that is supported, or replace the box with the motherboard
ChaiTRex has joined #haiku
<nipos> The syslog doesn't show anything suspicious. The message "KERN: slab memory manager: created area 0xffffff... (number)" appears rather often, but I don't think that has anyting to do with storage.
<waddlesplash> that's just the kernel heap allocating new areas
<dovsienko> you can benchmark it using "dd bs=1M ...", but you must be careful not to overwrite a production disk
<nipos> dovsienko: I don't plan buying more additional hardware. And I don't want to make big changes to the server hardware as I use the same machine for building OpenIndiana packages and don't want to break anything there.
<dovsienko> well, then you could do RAID in OpenIndiana and run Haiku in a VM with a virtual disk
<nipos> That could work, actually
<dovsienko> based on various experience, my first move to solve a hardware compatibility problem is to replace the hardware. it saves so much time.
<nipos> This is my replacement
<nipos> I have 2 computers I tried, both crash very often due to the AMD pre-Ryzen stability issues. And are rather slow. I thought a 2 x Intel Xeon machine would solve all problems, now it's the storage
<nipos> Oh and I have another server that I also tried, on that one the Haiku bootloader can't even find the boot partition (which is on the same USB stick), so that's not a solution either
<dovsienko> well, then the second most common move is to run the host OS that supports the hardware best, and to run everything else as VMs
<dovsienko> in my case that would be Linux (and actual hardware that works well with it)
<nipos> OpenIndiana works fine with the hardware, but I much prefer running on real hardware Instead of VMs. Especially with the built-in remote desktop and Serial over LAN features that the machine has, it's really nice to work with it.
<botifico> [haikuports/haikuports] pulkomandy pushed 2 commits to master [+0/-0/±2] https://github.com/haikuports/haikuports/compare/7b113b293a4c...c561dabc7cbc
<botifico> [haikuports/haikuports] pulkomandy 05d5d5a - roadfighter: checksum has changed
<botifico> [haikuports/haikuports] pulkomandy c561dab - qelectrotech: use github, tuxfamily hosting was removed
<botifico> [haikuports/haikuports] pulkomandy pushed 1 commit to master [+0/-0/±1] https://github.com/haikuports/haikuports/compare/c561dabc7cbc...67fc3b21c69a
<botifico> [haikuports/haikuports] pulkomandy 67fc3b2 - goonies: archive checksum has changed
<phschafft> erysdren: around?
<erysdren> hi
<erysdren> yes
<phschafft> :)
<erysdren> whats up
* phschafft pulls erysdren into a secret query, discussing the plan for world domiation or something.
DKnoto has quit [Ping timeout: 480 seconds]
tqh has quit [Quit: Leaving]
<waddlesplash> can you take a look at this please? it looks like it's due to your BListView drag changes
<waddlesplash> we shouldn't break BeOS compatibility, we need to add more checks to the BListView code here
DKnoto has joined #haiku
Pitmaster has joined #haiku
<Pitmaster> Guys, i have my beta 4 installed on VirtualBox. Gave him an USB stick 16 gig and used installation-program to install Haiku on my USB stick. It worked but on boot hanged. What can i do to let it work?
HaikuUser has joined #haiku
HaikuUser has quit []
dby has quit [Quit: Vision[]: i've been blurred!]
<phschafft> I'm thinking about writing a little library that allows you to query/set a few basic properties of files in a portable way.
<erysdren> sounds cool. what kind of properties?
<phschafft> with portable here meaning both cross-os but also cross-fs.
Pitmaster has quit [Quit: Vision[]: i've been blurred!]
<erysdren> unfortunately, that limits you to a handful of properites :P
<erysdren> but still sounds cool
<phschafft> specifically those relevant to my general work.
* phschafft looks at his list:
Anarchos has quit [Quit: Vision[]: i've been blurred!]
<phschafft> write mode, final state, media type, size, encoding, ISE, hashes.
Coldfirex has quit [Remote host closed the connection]
<phschafft> plus some more virtual properties/operations: parent object, hosting filesystem, allocation hints, ...
dqk has joined #haiku
LjL[m] has joined #haiku
dalme_ has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
<phschafft> erysdren: question answered?
<Skipp_OSX> ...arg
<Skipp_OSX> I'll take another look but I already fixed WonderBrush... I guess Stippi didn't want to accept my PR
<waddlesplash> yes, because we should not need a PR to fix crashes
<waddlesplash> clearly it worked on BeOS, if it does not on Haiku this is a bug
<Skipp_OSX> can you verify this worked on BeOS because BeOS implemented auto-scroll?
<Skipp_OSX> I meant, can you verify this worked on BeOS? Because BeOS implemented auto-scroll.
<Skipp_OSX> My guess is this is a post-BeOS change to Wonderbrush on Haiku before auto-scroll was reimplemented.
<nekobot> [haiku/haiku] waddlesplash pushed 1 commit to master [hrev57964] - https://git.haiku-os.org/haiku/log/?qt=range&q=9921f444625e+%5Ea26c33bd9764
<nekobot> [haiku/haiku] 9921f444625e - kernel/x86: Clear %edx in syscall handler if the return value isn't 64-bit.
<erysdren> phschafft: yeh, cool
<erysdren> but still sounds cool
<waddlesplash> Skipp_OSX: nope.
<waddlesplash> just downloaded Wonderbrush 1.6.0 which appears to have been built for R5
<waddlesplash> same crash
<waddlesplash> so, again, it's a regression from BeOS, we need to handle this properly
<Skipp_OSX> But I did fix it, I added a PR with my fix
<waddlesplash> Skipp_OSX: yes, I confirm WonderBrush 1.6.0 runs on BeOS R5.
<waddlesplash> no, you did not you fixed it in WonderBrush, but the fact is that this code works fine on BeOS R5
<waddlesplash> so your changes in Haiku are an API compatibility break with R5
<waddlesplash> so we need to somehow fix that
<Skipp_OSX> but you made me do that
<waddlesplash> ??
<Skipp_OSX> you reverted my fix
<Skipp_OSX> because you didn't want it to work on mouse up instead of down like BeOS did
<Skipp_OSX> remember?
<waddlesplash> I don't see any reverts from me in there
<waddlesplash> in fact the only changes that are me, are merges
<waddlesplash> and the one addition of B_SCROLL_VIEW_AWARE
<Skipp_OSX> well yeah you merged the reverse change, ok true not a revert
<waddlesplash> well the fact is that whatever is going on here we clearly have an API regression from BeOS R5 and we need to fix that
<Skipp_OSX> right because you made me regress from Be API
<Skipp_OSX> it should work on Mouse Up and then no crash
<waddlesplash> it did not crash before your changes...
<waddlesplash> if it worked on Mouse Up with BeOS, why does BeOS not crash either?
<Skipp_OSX> right
<waddlesplash> also the crash is a *stack overflow*
<Skipp_OSX> right because it worked on Mouse Up on BeOS
<waddlesplash> it's an infinite recursion
<waddlesplash> clearly we can avoid that
<Skipp_OSX> yeah I'm not sure exactly what's going on with that.
<Skipp_OSX> But I fixed the bug in Wonderbrush... I don't understand what the problem is.
<waddlesplash> well, did you try debugging it?
<waddlesplash> with the old code
<Skipp_OSX> yeah for a long time
<Skipp_OSX> I got the old code running right here
<Skipp_OSX> it's a NULL pointer access segfault IIRC, not a stack overflow...
<waddlesplash> for me the fault is on a "call" instruction with a really long stack
<waddlesplash> sure looks like a stack overflow
<waddlesplash> BListView::MouseMoved -> BListView::ScrollTo -> DragSortableListView::ScrollTo -> DragSortableListView::MouseMoved -> BListView::MouseMoved -> ...
<waddlesplash> this is about drag and drop, what does mouse-down selection have to do with anything?
mmu_man has quit [Ping timeout: 480 seconds]
<Skipp_OSX> yes now I remember... it is a stack overflow, GetMouse() calls ConvertFromScreen over and over again until it stack overflows and crashes
<waddlesplash> no?
<waddlesplash> the overflow is before that
<Skipp_OSX> you might be right... but not in the bug report I'm looking at
<waddlesplash> where?
<waddlesplash> the _ConvertFromScreen at the top isn't an overflow
<waddlesplash> that's expected
<waddlesplash> it's all the stuff below that which is the problem
<Skipp_OSX> yeah where is the _ConvertFromScreen crash report you're seeing, I'm just looking on my laptop in Debugger
<waddlesplash> _ConvertFromScreen is at the top but that's fine
<waddlesplash> the overflow is the hundreds of calls prior to it
<Skipp_OSX> what I'm seeing it is 100 _ConvertFromScree before it prior
<waddlesplash> huh, I don't
<waddlesplash> this looks like the break commit
<waddlesplash> we now call ScrollTo in MouseMoved
<Skipp_OSX> yeah probably the commit that broke it
_-Caleb-_ has left #haiku [#haiku]
mmu_man has joined #haiku
<waddlesplash> Skipp_OSX: if we add if (fTrack->is_dragging) to guard the scrollto that may fix it
<Skipp_OSX> https://pastebin.com/bft8Msnd <= bug report I'm looking at
<waddlesplash> yes, same trace here
<Skipp_OSX> ok cool
<waddlesplash> Skipp_OSX: this seems to fix it
<Skipp_OSX> hmmmmmm... and it doesn't break drag and drop?
<waddlesplash> nope
<waddlesplash> autoscroll and d&d still works in Locale preferences
<Skipp_OSX> oh but in WonderBrush?
<Skipp_OSX> ok if D&D still works in Locale that's a good sign
<waddlesplash> autoscroll and drag-to-reorder still work in WonderBrush too
<waddlesplash> not the same exact behavior though, but that makes sense because wonderbrush is managing it
<Skipp_OSX> it doesn't get trigged every time so just make sure to test a little bit more
<waddlesplash> there is no question this fixes it
<waddlesplash> it did get triggered basically every time
<Skipp_OSX> no
<Skipp_OSX> but 1 out of every 5-6 times maybe so if you draw a bunch of lines in WonderBrush and then drag them around and you can't make it crash then I'll be happy.
<waddlesplash> for me it was 100%
<Skipp_OSX> nah I can do it here without crashing easy
<waddlesplash> well so far I can get it to crash 100% without this patch and 0% with
<Skipp_OSX> ok but it does crash without too much difficulty
<waddlesplash> feel free to test the patch yourself
<Skipp_OSX> ok I guess that's it yeah uhhhhh ok
<Skipp_OSX> hey, on a related note, can we up the disk image size a bit, because there is very close to no room now
<waddlesplash> what do you mean?
<Skipp_OSX> (since packages got updated the other day the bar turned red)
<waddlesplash> you can increase it yourself...
<Skipp_OSX> just on the nightly image, make it a bit bigger
<waddlesplash> in your UserBuildConfig
<Skipp_OSX> yeah but for everybody else before R1B5 release I mean
<waddlesplash> R1B5 uses Release profile, different sizing there
<waddlesplash> and the base install image should be for installation not use anyway
<Skipp_OSX> oh ok well then for nightly images after that then
<Skipp_OSX> right, just 100 mb or so ought to do
<waddlesplash> Skipp_OSX: hmm. this change breaks click+drag in non-drag&drop to autoscrol
<waddlesplash> e.g. Locale "Formatting" list view window
<Skipp_OSX> I got it up here, what does it break?
<waddlesplash> Locale preferences "Formatting" tab list view
<Skipp_OSX> yeah I'm on it
<waddlesplash> if you click on an item + drag mouse outside list view, it scrolls
<waddlesplash> with the patch, it doesn't
<Skipp_OSX> oh I see it doesn't autoscroll right
<Skipp_OSX> right bc you have to be dragging, makes sense
<waddlesplash> ok, patch version 2
<Skipp_OSX> I'm already testing mouse is down...
<waddlesplash> this still has WonderBrush work but doesn't break click+drag-to-scroll
<waddlesplash> (revert first patch before applying)
<Skipp_OSX> so same thing but calling BView?
<waddlesplash> forcing base class call
<waddlesplash> instead of virtual to superclas
<Skipp_OSX> yeah
flag_ has joined #haiku
<Skipp_OSX> oh... am I going in a circle because I'm not calling the base class perhaps?
<Skipp_OSX> like I'm calling my own method recursively...
<waddlesplash> no, it's calling the superclass method
<waddlesplash> which calls us again
<waddlesplash> that's the recursion
<Skipp_OSX> the superclass method is BView though, right?
<Skipp_OSX> BView::ScrollTo I mean
<waddlesplash> sorry, I meant subclass
<phschafft> erysdren: basically I want to skip reimplementing everything all the time.
<phschafft> erysdren: and reading/writing the media type for example is just super handy.
<Skipp_OSX> ok
<Skipp_OSX> is WonderBrush no longer a package?
<waddlesplash> no, it is?
* coolcoder613_ waves
coolcoder613_ is now known as coolcoder613
<Skipp_OSX> what?
<Skipp_OSX> BListView or BView?
flag has quit [Ping timeout: 480 seconds]
<waddlesplash> it's the same either way
<waddlesplash> but I figured I'd be consistent
<Skipp_OSX> I'm confused now... how is it the same either way?
<waddlesplash> because BListView::ScrollTo just calls BView::ScrollTo
<Skipp_OSX> oh I see
<Skipp_OSX> ScrollTo without an index or point parameter
<Skipp_OSX> errr it has a point
<Skipp_OSX> ok sorry, ScrollTo with a point parameter instead of index
<Skipp_OSX> yes we're very explicit about that for some reason, so yes that's true
<Skipp_OSX> well other than the fact that I can't seem to find the Wonderbrush package
<Skipp_OSX> seems to work here
<waddlesplash> feel free to +2 then :P
<Skipp_OSX> well I haven't even tested it!
<waddlesplash> oh
<waddlesplash> pkgman install wonderbrush?
<Skipp_OSX> well not in Wonderbrush
<Skipp_OSX> yeah no good
<Skipp_OSX> it's probably gotta sync the data across first
<Skipp_OSX> is the update server down right now though?
<waddlesplash> no
<Skipp_OSX> ok it's just taking a bit then...
_-Caleb-_ has joined #haiku
AlienSoldier has joined #haiku
nipos has left #haiku [Error from remote client]
nipos has joined #haiku
<Skipp_OSX> ok, yeah it seems to work in Wonderbrush, let me just double check that it crashes in nightly
vdamewood has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
<Skipp_OSX> ok so my it's not crashing for me on WonderBrush here on hrev57943...
<waddlesplash> idk, crashed well and easily for me
<waddlesplash> just drag and reorder some items
<Skipp_OSX> under Object Filter right"
<Skipp_OSX> ?
<waddlesplash> yeah
<waddlesplash> I drew a bunch of random stuff then dragged it around
<Skipp_OSX> it's working fine here...
<Skipp_OSX> yeah me too
<waddlesplash> did you drag to the list edges?
<waddlesplash> if you drag above/below the list that's when it crashed easily
<Skipp_OSX> yeah O
<Skipp_OSX> I'm auto-scrolling and everything...
<waddlesplash> weird
<waddlesplash> crashes all the time for me...
<waddlesplash> well as long as the patch is OK we can merge it
<coolcoder613> Hi phschafft, erysdren
* phschafft waves to coolcoder613.
<nekobot> [haiku/haiku] waddlesplash pushed 1 commit to master [hrev57965] - https://git.haiku-os.org/haiku/log/?qt=range&q=cb92249c8839+%5E9921f444625e
<nekobot> [haiku/haiku] cb92249c8839 - kernel/timer: Use lockers in add_timer().
<Skipp_OSX> are you on 32-bit or 64-bit?
<erysdren> hiya
<waddlesplash> Skipp_OSX: 32-bit, for WonderBrush
<waddlesplash> I use 64-bit most of the time but don't have WonderBrush there
<Skipp_OSX> that's gotta be it
<coolcoder613> I have been working on my OS recently: https://github.com/coolcoder613eb/blog_os
<coolcoder613> I've mostly been working on line editing
<erysdren> cool!!
<phschafft> coolcoder613: was in fact about to ask on the status. hoped maybe we can have a chat about it the next days.
<Skipp_OSX> I'm on hrev57963 32-bit... no crash in Wonderbrush
<coolcoder613> There are some weird bugs that i'm pretending don't exist until I feel like tackling them
<phschafft> haha
<Skipp_OSX> either I'm going crazy or there is no crash happening
* phschafft secretly thinks coolcoder613 should fix GDT code in sirtx and could add line editing in lci ;)
<coolcoder613> Who, me?
<coolcoder613> If it's written in Rust...
<phschafft> more like C ;)
<coolcoder613> I don't program in C
<coolcoder613> Besides, have a look at my line editing code, then tell me if you want that in lci
<phschafft> I think we would both not get happy if you would try to work on lci ;)
<coolcoder613> And... Now I'm having trouble getting the bug to happen...
<phschafft> oh how I know that.
<phschafft> I'm still thinking about making a graphic with words and how much software developers like them.
<phschafft> e.g. 'always' and 'never' are most enjoyed. 'sometimes' clearly good bit begind. but 'rarely' is in the hate area.
<phschafft> ;)
<coolcoder613> Ok, maybe the bug only happend with the version that Claude made more DRY...
<coolcoder613> Having reverted that, I can't reproduce it, or another bug I just discovered
<coolcoder613> So, bakc to having an ~100 line read_line function
<coolcoder613> *back
<coolcoder613> Yay.
<coolcoder613> https://bpa.st/5XKA
<coolcoder613> I'm rolling my own escape codes, not using ansi...
<erysdren> dangerous!
<erysdren> :P
<arraybolt3> phschafft: I think 'works here' may rank even worse than 'rarely' :)
<Skipp_OSX> ok my laptop here I'm on hrev57483 gcc2 (yes I know it's old) and WonderBrush crashes like a mofo
<Skipp_OSX> I downloaded latest gcc2 nightly image, hrev57963 and no crash in vmware, or in 64-bit hrev 57963...
trungnt2910[m] has joined #haiku
x512[m] has joined #haiku
<Skipp_OSX> anybody running 32-bit Haiku on real hardware here that can help me out testing on WonderBrush?
xylobol[m] has joined #haiku
<coolcoder613> so.. Esc+i is start and end editable, Esc+c is clear screen, Esc+< and Esc+> is move the cursor left or right if editable...
<Skipp_OSX> I mean your patch does nothing waddlesplash, it calls the same method a different way, I don't see how that could fix anything.
<Skipp_OSX> if it makes you happy, it makes me happy, but I can't seem to reproduce the crash so I'm perplexed over here.
<coolcoder613> Skipp_OSX: I do run Haiku 32-bit on real hardware...
<phschafft> arraybolt3: over here I'm way too often saying 'depends on OpenSSL version'.
<coolcoder613> But it takes like a minute to boot up...
<Skipp_OSX> coolcoder613 can you do me a favor, install wonderbrush, then draw a bunch of lines, then drag and drop the lines in the list view under the Objects Filter and see if you can get it to crash? Should only take a 5-6 attempts at most.
<Skipp_OSX> and let me know what hrev you're running in AboutSystem or give me the crash report which will have that info
coolcoder613_32 has joined #haiku
<coolcoder613_32> Here I am...
<phschafft> coolcoder613: see? for SIRTX that is all control calls on the handle. with the defaults there is no way to control the cursor or alter screen attributes and stuff via writing to the terminal device.
<waddlesplash> Skipp_OSX: ?
dalme_ has joined #haiku
<waddlesplash> Skipp_OSX: adding BListView:: means the virtual table isn't used
<waddlesplash> so we don't call the subclass even potentially
<waddlesplash> so we don't recurse
<waddlesplash> that's the fix
<coolcoder613_32> What was I supposed to do in Wonderbrush?
<erysdren> "then draw a bunch of lines, then drag and drop the lines in the list view under the Objects Filter and see if you can get it to crash? Should only take a 5-6 attempts at most."
<coolcoder613_32> Ok, it crashed...
<coolcoder613_32> Skipp_OSX: Should I save a report?