#archlinux32 | Logs for 2019-03-07

Back
[00:09:52] <buildmaster> i686/lsd is broken (says rechenknecht).
[00:27:09] <buildmaster> i686/sqlitebrowser is broken (says rechenknecht).
[00:31:41] -!- guys has quit [Ping timeout: 255 seconds]
[00:34:07] -!- oaken-source has quit [Ping timeout: 240 seconds]
[00:45:58] -!- guys has joined #archlinux32
[00:59:52] -!- jason0597 has joined #archlinux32
[01:00:45] <jason0597> are things here usually this quiet?
[01:10:20] <buildmaster> i686/deepin-dock is broken (says buildknecht2).
[01:11:53] <buildmaster> i686/deepin-daemon is broken (says rechenknecht).
[01:12:29] -!- MrBIOS has quit [Quit: MrBIOS]
[01:16:06] <buildmaster> i686/deepin-launcher is broken (says buildknecht2).
[01:16:16] <buildmaster> i686/deepin-session-ui is broken (says rechenknecht).
[01:18:33] <buildmaster> i686/startdde is broken (says eurobuild3).
[01:30:20] <buildmaster> i686/golang-deepin-gir is broken (says buildknecht2).
[01:30:52] <buildmaster> i686/foomatic-db is broken (says eurobuild3).
[02:11:50] <buildmaster> i686/vulkan-headers are broken (says buildknecht2).
[02:13:01] -!- City-busz has quit [Remote host closed the connection]
[02:13:01] -!- elibrokeit has quit [Read error: Connection reset by peer]
[02:25:03] <buildmaster> i686/python-aws-xray-sdk is broken (says rechenknecht).
[02:28:14] <buildmaster> i686/reflector is broken (says buildknecht2).
[02:34:31] <bill-auger> jason0597: ive never heard and sounds eminating from this channel
[02:36:07] <bill-auger> nothing but text tex blah more text - mostly about development and maintenence - there is a forum for user support
[02:36:22] <buildmaster> i686/golang-deepin-lib is broken (says buildknecht2).
[02:36:33] <bill-auger> what are you trying to do exactly?
[02:37:57] -!- jason0597 has quit [Ping timeout: 245 seconds]
[03:22:02] <buildmaster> i686/nvidia-390xx-lts are broken (says rechenknecht).
[03:40:28] -!- T`aZ has quit [Ping timeout: 245 seconds]
[03:42:19] -!- T`aZ has joined #archlinux32
[03:50:20] <buildmaster> i686/josm is broken (says rechenknecht).
[03:57:45] -!- guys has quit [Ping timeout: 252 seconds]
[05:09:29] <buildmaster> i686/grafana is broken (says eurobuild3).
[05:19:33] -!- DepositePirate_ has quit [Ping timeout: 256 seconds]
[05:19:57] -!- DepositePirate_ has joined #archlinux32
[05:26:32] -!- ofara_ has quit [*.net *.split]
[05:26:32] -!- rcf has quit [*.net *.split]
[05:26:32] -!- KitsuWhooa has quit [*.net *.split]
[05:53:38] -!- ofara_ has joined #archlinux32
[05:53:38] -!- rcf has joined #archlinux32
[05:53:38] -!- KitsuWhooa has joined #archlinux32
[06:16:15] -!- City-busz has joined #archlinux32
[06:16:20] -!- elibrokeit has joined #archlinux32
[08:00:25] <buildmaster> i686/thunderbird is broken (says buildknecht).
[08:03:00] -!- oaken-source has joined #archlinux32
[08:14:17] -!- deep42thought has joined #archlinux32
[08:14:17] <buildmaster> Hi deep42thought!
[08:14:17] <buildmaster> !rq deep42thought
[08:14:18] <phrik> buildmaster: <deep42thought> so you're a garbage collector?
[08:18:05] <buildmaster> i686/d-stdx-allocator is broken (says buildknecht).
[08:23:05] <buildmaster> i686/python-apsw is broken (says buildknecht).
[08:26:32] <deep42thought> jason0597: registering to the bugtracker alone is not sufficient to report bugs - your account needs to be promoted manually by one of the admins - which I just did
[08:26:41] -!- jason0597 has joined #archlinux32
[08:48:26] <KitsuWhooa> I'll ask again since there seems to be more activity now, if it doesn't get annoying. Is it known that librsvg is trying to use SSE2, and is there a blocker for fixing it?
[08:48:53] <deep42thought> it's not annoying
[08:49:12] <deep42thought> there are some packages, where sse2 are creeping in
[08:49:24] <KitsuWhooa> rustc is definitely one of them
[08:49:25] <deep42thought> unfortunately, I'm no expert on that terrain (abaumann is)
[08:49:40] <KitsuWhooa> I guess I'll poke them when they get online
[08:49:45] <deep42thought> there are plans for a 'pentium3' architecture where sse2 would be allowed
[08:50:02] <deep42thought> but otoh, the main problem is to check for unallowed instructions in other architectures
[08:50:07] <KitsuWhooa> I was asking because I saw the pkgbuild for librsvg and it has flags that should disable SSE2
[08:50:19] <deep42thought> oh, let me have a look
[08:50:28] <KitsuWhooa> I know it was broken at some point, then it was fixed, and then it broke again
[08:50:41] <KitsuWhooa> it basically means I can't run GTK applications
[08:51:00] <deep42thought> and librsvg is also not available on i486 iirc
[08:51:04] <deep42thought> (due to rust)
[08:51:51] <KitsuWhooa> Hm, that might explain a lot
[08:51:54] <KitsuWhooa> I'm not on i486
[08:52:08] <deep42thought> yes, i486 should have no sse, sse2 and no cmov
[08:52:22] <KitsuWhooa> I'm on a Pentium 3 Coppermine though
[08:52:24] <deep42thought> but i686 having sse2 is becoming a problem for more and more users
[08:52:34] <KitsuWhooa> yes, definitely
[08:52:40] <KitsuWhooa> I'm getting SIGILL on movsd :p
[08:52:44] <deep42thought> hence the idea for another architecture which may have sse2
[08:54:14] <KitsuWhooa> I was referring to this earlier https://git.archlinux32.org
[08:54:52] <deep42thought> I'm looking at it right now - it's probably just the regex failing or something ...
[08:56:11] <KitsuWhooa> It'd be great if it's just the regex failing
[08:56:44] <deep42thought> yup, that sed does nothing :-)
[08:56:47] <KitsuWhooa> might be worth making it fatal if it fails, if possible, to prevent this from happening in the future again, assuming that is indeed issue
[09:00:52] -!- oaken-source has quit [Ping timeout: 250 seconds]
[09:01:13] -!- abaumann has joined #archlinux32
[09:01:13] <buildmaster> Hi abaumann!
[09:01:13] <buildmaster> !rq abaumann
[09:01:14] <phrik> buildmaster: <abaumann> wow: my first wayland experience of the blury kind (in libvirt).. there is a small black line moving from the top to the bottom of the screen.. an homage to old CRT monitors, I suppose. ;-)
[09:01:17] <deep42thought> Hi abaumann!
[09:01:24] <abaumann> hi deep42though
[09:01:36] <abaumann> target-cpu=pentium3
[09:01:51] <abaumann> isn't that excplicitely drawing in SSE2?
[09:02:15] <abaumann> *abaumann checks his notebook with very confisuing Intel technology drawings..
[09:02:15] <deep42thought> the regex did not apply
[09:02:19] <abaumann> ah
[09:02:26] <deep42thought> 'cargo build' was renamed to '$(CARGO) build'
[09:02:37] <abaumann> ah, this explains it.
[09:02:51] <abaumann> rust on the other side is still the old version, the current version doesn't build (again)
[09:03:01] <deep42thought> abaumann: is this right, what I said here? https://bbs.archlinux32.org
[09:03:02] <phrik> Title: What constitutes an i686? AMD K6-2+ is not one? / Installation / Arch Linux 32 Forums (at bbs.archlinux32.org)
[09:03:07] <abaumann> I also have a libsecret __float128 blocker currently in core..
[09:04:00] <abaumann> cmov for sure.
[09:04:31] <abaumann> every decently new i686 should have SSE2 actually.
[09:04:57] <deep42thought> but we can't really rely on being "decently new", can we? ;-p
[09:05:14] <abaumann> of course not. ;-)
[09:05:50] <deep42thought> KitsuWhooa: I see no easy way to make a failing regex fatal
[09:05:52] <abaumann> it's rather confusing: AMDs may lack SSE2 and SSE along with CMOV.
[09:06:03] <deep42thought> sed does not exit non-zero if it didn't chang the file ...
[09:06:07] <abaumann> every i686 has CMOV
[09:06:16] <KitsuWhooa> deep42thought: ah, fair enough...
[09:06:25] <abaumann> SSE should be there almost everywhere, SSE2 on recent pentiums
[09:06:38] <deep42thought> we could "chmod -w" the target and fail if sed succeeds ...
[09:06:40] <deep42thought> hmm
[09:06:50] <abaumann> mmh. I'll hunt a drawing or something. You're right, we should put that somewhere prominent on the web page.
[09:07:34] <abaumann> SSE is since pentium3
[09:07:38] <deep42thought> errrrr sed is strange
[09:07:57] <deep42thought> I can "sed -i"-edit a file which is not writeable O.o
[09:08:16] <abaumann> huh? that's cool. so it just does nothing?
[09:08:29] <deep42thought> [erich@nlopc43 ~]$ echo x > /tmp/bla
[09:08:29] <deep42thought> [erich@nlopc43 ~]$ chmod -w /tmp/bla
[09:08:29] <deep42thought> [erich@nlopc43 ~]$ sed -i 's/x/y/' /tmp/bla
[09:08:29] <deep42thought> [erich@nlopc43 ~]$ echo $?
[09:08:29] <deep42thought> 0
[09:08:29] <deep42thought> [erich@nlopc43 ~]$ cat /tmp/bla
[09:08:29] <deep42thought> y
[09:08:51] <KitsuWhooa> maybe you can use awk instead, although I'm not too familiar with it
[09:08:54] <abaumann> -i being a gnu extension..
[09:09:45] <deep42thought> regarding architecture branches: I think, we should simply put up a list of cpu flags required for running i486, i686 (and the comming-soon pentium3)
[09:10:09] <KitsuWhooa> I still don't think "pentium3" would be an accurate representation of SSE2 supported CPUs
[09:10:10] <deep42thought> then everyone can check against their flags in /proc/cpuinfo
[09:10:12] <KitsuWhooa> if that's what you meant
[09:10:35] <deep42thought> KitsuWhooa: yes, that's what I mean - what other name is more fitting?
[09:10:37] <abaumann> hi KitsuWhooa.
[09:10:42] <KitsuWhooa> hello!
[09:10:44] <abaumann> yeah, the naming is not ideal..
[09:10:59] <deep42thought> i(6.5)86
[09:11:03] <abaumann> but i686_with_mmx_sse_sse2_and_probably_more is also not a good name ;-)
[09:11:05] <KitsuWhooa> deep42thought: I said it because I have a P3 and it doesn't do SSE2 :p
[09:11:11] <KitsuWhooa> so it can be very confusing
[09:11:23] <KitsuWhooa> https://tasossah.com
[09:11:24] <phrik> Title: P III cpuinfo (at tasossah.com)
[09:12:07] <deep42thought> and if we just call it 'coppermine'?
[09:12:12] <deep42thought> is that too much honor for intel?
[09:12:25] <abaumann> nah. there is too much confusion about Intel brand names, I think.
[09:12:31] <KitsuWhooa> definitely
[09:12:38] <abaumann> Pentium for instance was reused in 2009 IIRC
[09:12:44] <abaumann> for not the i585 thingy..
[09:13:01] <abaumann> i486: no CMOV, no MMX, no SSE, no SSE2
[09:13:11] <abaumann> i686: CMOV, MMX, SSE
[09:13:22] <deep42thought> no SSE2
[09:13:22] <abaumann> pentium3: CMOV, MMX, SSE, SSE2
[09:13:30] <abaumann> yep.
[09:13:45] <deep42thought> well, but "pentium3" might be confusing
[09:13:50] <abaumann> I think, that was the naming which we had in mind.
[09:13:56] <deep42thought> as KitsuWhooa's pentium3 does not have sse3
[09:13:58] <abaumann> i686_with_sse2?
[09:14:00] <deep42thought> sse2
[09:14:01] <KitsuWhooa> yeah
[09:14:07] <abaumann> maybe not a bad idea.
[09:14:08] <deep42thought> i786
[09:14:23] <KitsuWhooa> i686_sse2 could also work
[09:14:26] <deep42thought> oh, you mean 'sse2' as architecture name?
[09:14:30] <abaumann> technically Pentium M or Pentium 4 is also an option
[09:14:51] <abaumann> IIRC SSE2 was introduced with Pentium M cores
[09:15:00] <abaumann> pentiumm looks a little bit silly :-)
[09:15:04] <abaumann> pentium4?
[09:15:16] <abaumann> which is much more informative than pentium3, but more correct.
[09:15:40] <abaumann> yeah. I also have a PIII with SSE2 (Celeron, EEEPC) and an older PIII without..
[09:15:49] <deep42thought> "... die Intel mit dem Intel Pentium 4 einf├╝hrte."
[09:16:14] <abaumann> so. pentium3 -> pentium4?
[09:16:17] <deep42thought> yes
[09:16:23] <deep42thought> I'm in favour of that
[09:16:34] <KitsuWhooa> The P III Katmai I have also doesn't support sse2
[09:16:36] <KitsuWhooa> thought I'd mention it
[09:16:37] <abaumann> ok. me too
[09:16:48] <abaumann> yes. you did.
[09:17:04] <abaumann> I was just checking on my park of old machines.. :-)
[09:17:09] <abaumann> flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov clflush dts acpi mmx fxsr sse sse2 ss tm pbe nx bts cpuid
[09:17:12] <abaumann> eeepc
[09:17:36] <KitsuWhooa> I was saying that both Coppermine and Katmai don't have SSE2, so it's not only one codename
[09:18:22] <KitsuWhooa> So, next question I guess. Will librsvg be rebuilt, or do I need to wait for a new upstream release?
[09:18:50] <deep42thought> I rescheduled it
[09:19:03] <KitsuWhooa> Ah, okay. Thank you
[09:19:07] <KitsuWhooa> much appreciated
[09:19:18] <deep42thought> https://archlinux32.org
[09:19:19] <phrik> Title: Arch Linux 32 - List of Package Builds (at archlinux32.org)
[09:19:29] <KitsuWhooa> was looking for that page, actually. Thanks
[09:19:30] <deep42thought> ah, damn
[09:19:42] <deep42thought> that's still trying to compile the old commit ...
[09:20:16] <abaumann> There are other Intel features setting CPUs apart like control-flow-enforcement. I had some CPUs not being able to translate this into multi-byte NOPs.
[09:20:19] <deep42thought> it's even linked, now :) : archlinux32.org -> "Build Master" -> "Build List"
[09:20:28] <abaumann> Somewhere in the Pentium II area of CPUs
[09:20:41] <KitsuWhooa> deep42thought: indeed. Apologies, didn't see it
[09:20:46] <deep42thought> np
[09:20:48] <abaumann> Pentium III (Coppermine), 701.559 MHz, flags : fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca cmov pse36 mmx fxsr sse cpuid
[09:20:51] <abaumann> yep. no sse2
[09:23:16] <abaumann> Pentium S 75, no cmov, sse, sse2, mmx
[09:23:35] <deep42thought> looks like we need 2^4=16 architectures ;-)
[09:24:14] <abaumann> sweet ;-)
[09:24:38] <deep42thought> still, the question remains, how we identify the bad binaries
[09:24:42] <deep42thought> :-/
[09:25:07] <KitsuWhooa> I guess one way would be to disassemble and check for instructions
[09:26:03] <abaumann> yes. that's not the problem. Some packages decide NOT to have this configurable, some rely on uname to report the right architecture (and setarch is not doing the proper job, I'm looking at you glibc).
[09:26:25] <KitsuWhooa> Oh, I see...
[09:26:29] <abaumann> Some (gcc/binutils) just but it security-relevant code in the middle of the generated code, no flags involved.
[09:26:35] <abaumann> *put
[09:27:12] <abaumann> yes, the art of cross-compilation and multi-architectes goes slowly away.. we get into a 64-bit Intel/Intel mono-culture, that's my personal fear.
[09:27:19] <abaumann> *multi-architecture
[09:27:20] <abaumann> *multi-architectures
[09:27:36] <KitsuWhooa> yeeeeah
[09:27:48] <abaumann> *abaumann considers to change his old keyboard with one with less springy and sticky keys..
[09:28:41] <abaumann> basically, the assumption is: 64-bit Intel made SSE2 a minimal requirement anyway, so most people assume, that 32-bit Intel processors should all support SSE2 by now..
[09:28:42] -!- oaken-source has joined #archlinux32
[09:28:48] <abaumann> .. which is not the worst assumption..
[09:28:56] <abaumann> ..just not for legacy systems.. :-)
[09:29:25] <KitsuWhooa> chances are, those people don't have an interest in supporting those systems either
[09:30:31] <KitsuWhooa> in all fairness, I maintain software that assumes endianness, so I'm not the one to talk :p
[09:30:32] <abaumann> first, you need to have such old hardware around for testing and you need time to fix hard-core bugs.
[09:30:38] <KitsuWhooa> exactly
[09:30:47] <abaumann> endianess. wow! what's that. ;-)
[09:31:00] <abaumann> I have some old sparc machines..
[09:31:09] <abaumann> ..Linux supported them better in the past.
[09:31:16] <abaumann> all but sparc64 is more or leass dead now.
[09:31:20] <deep42thought> endianess is the question wether you use serialize() or ezilaires()
[09:31:20] <abaumann> *less
[09:31:26] <KitsuWhooa> linux has a hard time accessing the floppy drive on my desktop :p
[09:31:28] <abaumann> :-)
[09:31:35] <abaumann> LOL
[09:31:58] <abaumann> 3 1/2 inch I hope. :-)
[09:32:23] <KitsuWhooa> Yeah
[09:32:30] <abaumann> could be worse.
[09:32:46] <abaumann> I have 5 1/4 ones, and no working drive anymore to read them..
[09:33:07] <KitsuWhooa> well, it gets annoying if you run blkid and it hangs trying to access the floppy
[09:33:10] <abaumann> 8-inch is a thing from my childhood, I think, I saw one drive ones in a computer room.
[09:33:10] <KitsuWhooa> and then you need to reboot the entire system
[09:33:40] <abaumann> *abaumann tries blkid /dev/fd0 on his vintage machine..
[09:34:15] <abaumann> /dev/fd0: PTTYPE="atari"
[09:34:30] <abaumann> mmh. not exactly the result I was expecting.. but nothing hangs..
[09:34:35] <KitsuWhooa> [88198.892749] print_req_error: I/O error, dev fd0, sector 0 flags 0
[09:34:37] <KitsuWhooa> [88198.892757] floppy: error 10 while reading block 0
[09:34:38] <KitsuWhooa> I got that :p
[09:34:58] <abaumann> that's a real floppy drive and not one of those cheap USB floppy emulators..
[09:35:06] <abaumann> .. those never worked for me.
[09:35:06] <KitsuWhooa> yeah
[09:35:13] <KitsuWhooa> mine's a real one
[09:35:40] <abaumann> [ 1016.508929] print_req_error: I/O error, dev fd0, sector 0
[09:35:49] <abaumann> that could also be an empty drive or a failing drive
[09:36:10] <KitsuWhooa> Somehow an AM3+ board came with a floppy controller and a PATA one. I did not miss the chance to buy it. It's been my primary desktop since :p
[09:36:33] <KitsuWhooa> I am sure the drive and the cable work fine, so there's that
[09:36:47] <KitsuWhooa> sometimes it throws that error, other times it just hangs and throws a trace
[09:36:59] <KitsuWhooa> and then another trace for the blocked userspace application trying to access it
[09:37:37] <abaumann> that's exacly showing the problem. Is the kernel driver broken for some floppy drives out there, or is it the hardware?
[09:37:59] <KitsuWhooa> most likely the kernel drive
[09:38:03] <KitsuWhooa> worked fine on windows last I checked
[09:38:09] <KitsuWhooa> and also didn't hardlock if the floppy itself had bad sectors
[09:38:17] <abaumann> if it works with older kernels or other operating systems, then for sure
[09:38:19] <KitsuWhooa> *driver
[09:39:31] <KitsuWhooa> I haven't had much of a need for it recently, as all the old computers and floppies are at home, and the laptop with the coppermine has wifi
[09:39:35] <KitsuWhooa> so I just haven't bothered
[09:42:53] <KitsuWhooa> Also, can you reschedule librsvg with the right commit please? Last build is done
[09:43:08] <deep42thought> the buildmaster is working on it
[09:43:19] <KitsuWhooa> ah
[09:43:23] * KitsuWhooa just didn't see it on the list
[09:43:34] <deep42thought> it's still putting it there
[09:43:50] <KitsuWhooa> Yeah, I know nothing about any of this :p
[09:43:50] <deep42thought> (thunderbird-i18n takes some time to be inserted)
[09:43:52] <KitsuWhooa> Thanks
[09:49:55] -!- thePiGrepper has joined #archlinux32
[10:01:52] <buildmaster> i686/golang-golang-x-image is broken (says buildknecht2).
[10:09:32] <deep42thought> abaumann: https://archlinux32.org
[10:09:34] <phrik> Title: Arch Linux 32 - Architecture Overview (at archlinux32.org)
[10:21:41] <abaumann> looking good.
[10:21:52] <deep42thought> I'm adding a small description, though
[10:22:02] <deep42thought> so it doesn't look tooooo minimalistic ;-P
[10:22:14] <abaumann> good idea
[10:22:31] <abaumann> libsecret didn't build, now it does? *puzzle*
[10:34:03] <abaumann> ah. multi-byte NOPS are part of SSE.
[10:34:32] <buildmaster> i686/libsecret is broken (says buildknecht).
[10:34:39] <abaumann> aha. now again.
[10:34:53] <deep42thought> :-)
[10:35:11] <deep42thought> I'll be afk, checking for nlopc46 ...
[10:35:54] <abaumann> there is also a FNOP..
[10:36:00] <abaumann> cool :-)
[10:36:27] <abaumann> origin 8087 co-processor.
[10:36:46] <abaumann> anf FMOV.. yes..
[10:37:05] <abaumann> "Instruction set by marketing"
[10:37:25] <abaumann> that's NOT what MIPS stands for. ;-)
[10:44:26] -!- thePiGrepper has quit [Ping timeout: 250 seconds]
[10:49:52] -!- titus_livius has joined #archlinux32
[10:50:26] <abaumann> "dbus-run-session meson test".. urgh.
[10:51:15] <abaumann> I wonder why tests have to be started with meson with a new D-bus session..
[10:52:04] -!- oaken-source has quit [Ping timeout: 268 seconds]
[10:59:56] <deep42thought> re
[11:01:51] <deep42thought> what can possibly be special about NOP to prepend an F?
[11:05:36] <buildmaster> i686/ffmpegthumbs are broken (says nlopc46).
[11:06:11] <buildmaster> i686/mplayer is broken (says buildknecht).
[11:06:36] -!- oaken-source has joined #archlinux32
[11:10:29] -!- thePiGrepper has joined #archlinux32
[11:12:03] <buildmaster> i686/kfilemetadata is broken (says buildknecht).
[11:14:06] <buildmaster> i686/aubio is broken (says nlopc46).
[11:21:49] -!- Fear has quit [Ping timeout: 244 seconds]
[11:36:13] -!- Fear has joined #archlinux32
[12:00:03] <buildmaster> i686/opencv is broken (says rechenknecht).
[12:04:05] <buildmaster> i686/kamoso is broken (says nlopc46).
[12:12:50] <KitsuWhooa> https://archlinux32.org seems to have died
[12:12:55] <KitsuWhooa> it's returning HTTP/2 500
[12:13:06] <deep42thought> *grrr*
[12:13:33] <buildmaster> i686/libringclient is broken (says rechenknecht).
[12:14:21] <deep42thought> unmatched ()
[12:14:22] <deep42thought> ...
[12:14:25] <deep42thought> sry
[12:14:40] <deep42thought> I thought, I had tested that ...
[12:14:47] <buildmaster> i686/mgba is broken (says nlopc46).
[12:14:48] <KitsuWhooa> it's fine
[12:24:59] <buildmaster> i686/carla is broken (says nlopc46).
[12:33:03] <buildmaster> i686/obs-studio is broken (says nlopc46).
[12:35:27] -!- oaken-source has quit [Ping timeout: 240 seconds]
[12:36:45] <KitsuWhooa> Something slightly unrelated
[12:37:06] <KitsuWhooa> git.archlinux32.org has an AAAA record but doesn't respond to requests on port 80 over IPv6
[12:37:17] <deep42thought> tyzoid: ^
[12:37:19] <KitsuWhooa> or 443 for that matter
[12:43:44] <abaumann> deep42thought: well, as the 8087 was a CPU on it's own, it had it's own opcode for doing nothing FNOP.. I suppose..
[12:44:02] <deep42thought> still strange
[12:44:15] <deep42thought> I always thought $ffffffff.. just did nothing
[12:44:35] <abaumann> git.archlinux32.org is broken for IPv6 for quite some time, I think.
[12:44:36] <deep42thought> and there is no need for another opcode that does nothing
[12:44:49] <deep42thought> it's some port forwarding thing on tyzoid's server
[12:44:58] <deep42thought> but unfortunately, also my ssh tunnel is broken :-(
[12:44:58] <abaumann> /etc/hosts on the buildmaster still has an IPv4 entry..
[12:45:21] <deep42thought> should I simply remove the ipv6 address?
[12:45:26] <abaumann> on contraire. you want the copro to do nothing while doing something on the main cpu..
[12:45:30] <abaumann> ..so it makes sense.
[12:45:34] <deep42thought> ah, it's a cname ...
[12:46:00] <deep42thought> abaumann: well, that's the "ignore unknown opcodes" feature ;-P
[12:46:11] <abaumann> :-)
[12:46:19] <deep42thought> yeah, ok, I see, that it's useful ...
[12:46:56] <deep42thought> abaumann: I was thinking about https://www.archlinux32.org
[12:46:58] <phrik> Title: Arch Linux 32 - Package Differences Reports (at www.archlinux32.org)
[12:47:05] <deep42thought> currently it only shows diffs between i486 and i686
[12:47:22] <deep42thought> and my idea was to also show diffs between i[46]86 and x86_64
[12:47:47] <deep42thought> otoh, to implement this cleanly, we would need to track _upstream_ packages in the database, too
[12:47:49] <abaumann> yeah. that would be extremly handy in order to see what packages lag behind..
[12:47:51] <deep42thought> what do you think?
[12:48:04] <abaumann> you could grab the RSS feed from upstream.
[12:48:16] <abaumann> sort of phantom packages for x86_64
[12:48:22] <deep42thought> grabbing the packages is no problem (all relevant machines run on arch ;-P)
[12:48:30] <abaumann> ah. true that. :-)
[12:48:43] <deep42thought> my problem is with inserting/tracking them in the db
[12:49:10] <deep42thought> currently I have a large array in php of x86_64 packages and a large mysql table of i[46]86 packages ...
[12:49:10] <abaumann> it should not trigger builds and should appear as already in stable?
[12:49:34] <deep42thought> it could be a completely separate table
[12:49:39] <abaumann> ..ah. a huge diff in php? :-)
[12:49:57] <deep42thought> yes, but it would massively increase the size of the output of the mysql query
[12:50:03] <abaumann> a separate table is a good idea.
[12:50:09] <deep42thought> which currently only is the diff between i486 and i686
[12:50:19] <deep42thought> yeah, separate table it will be
[12:50:42] <abaumann> worth a try: apply brute force till proven wrong. :-)
[12:50:52] <deep42thought> !grab abaumann
[12:50:52] <phrik> deep42thought: Tada!
[12:51:26] <deep42thought> I just wanted a second opinion before I put another 10k packages into the database :-D
[12:51:51] <abaumann> well.. it's a mysql.. almost any amount of data is too much for it too handle.. so it shouldn't matter too much..
[12:51:59] <deep42thought> !grab abaumann
[12:51:59] <phrik> deep42thought: Tada!
[12:52:01] <abaumann> *it to handle*
[12:52:11] <abaumann> damn, grabbed with a spelling mistake ;-)
[12:52:19] <deep42thought> gotcha
[13:06:19] -!- thePiGrepper has quit [Ping timeout: 255 seconds]
[13:20:33] -!- thePiGrepper has joined #archlinux32
[13:33:37] -!- DepositePirate_ has quit [Quit: Connection reset by beer]
[13:39:08] -!- DepositePirate has joined #archlinux32
[13:44:35] <deep42thought> abaumann: you think, having reanimate-pacman is really a good idea?
[13:47:12] <buildmaster> i686/aubio is broken (says rechenknecht).
[13:47:18] <deep42thought> >.<
[14:08:09] -!- oaken-source has joined #archlinux32
[14:23:32] <abaumann> huh?
[14:23:42] <deep42thought> on the buildmaster
[14:23:52] <deep42thought> this script looks like it's semi-fixing a serious problem
[14:24:03] <abaumann> I didn't put it there.
[14:24:11] <deep42thought> O.o
[14:24:18] <abaumann> I put a pacman-static there because I broke pacman with a symlink :-)
[14:29:15] <deep42thought> abaumann: the differences report is now ready https://archlinux32.org
[14:29:32] <abaumann> oh. cool. :-)
[14:29:57] <deep42thought> dark-red means, there is a newer package already built, light red means it's the newest built package (but still older than some other package)
[14:30:01] <abaumann> firefox 1.66 *sigh*
[14:30:19] <abaumann> this is comparing our stable against upstream?
[14:30:33] <deep42thought> our stable and testing
[14:30:35] <deep42thought> against upstream
[14:30:37] <buildmaster> i686/pulseeffects are broken (says rechenknecht).
[14:30:47] <deep42thought> but for the "is there a newer package" also staging is considered
[14:30:52] <abaumann> ah. and if something is queued, it has a differnt colour.
[14:31:02] <deep42thought> ah, no, staging is even in the listing :-D
[14:31:11] <deep42thought> no
[14:31:23] <deep42thought> being in the build list has no influence on the color
[14:31:28] <deep42thought> only built packages are considered
[14:31:32] <abaumann> ah. ok.
[14:31:49] <deep42thought> (and there is no guarantee, that the newer, build package is not outdated, too)
[14:31:51] <abaumann> very handy, indeed.
[14:33:09] <deep42thought> updates are somewhat asynchronous (pacman updates every 10 minutes, the mysql database every 30 minutes, get-package-update even with another interval)
[14:33:11] <abaumann> *abaumann has to fix a bicycle now, flat tire..
[14:33:17] <deep42thought> have fun!
[14:33:28] <abaumann> yeah, I will.. and some dirty hands probably.. :-)
[14:33:40] <deep42thought> fitting to your keyboard? ;-)
[14:33:49] <abaumann> sort of, yeah. ;-)
[14:33:56] <abaumann> the update period is fine.
[14:34:13] <abaumann> the builds on the slaves are not that fast, so this delay should not matter..
[14:34:27] <deep42thought> well, it depends
[14:34:36] <deep42thought> when they're idle, they can be pretty fast ...
[14:35:05] <abaumann> true
[14:37:07] -!- oaken-source has quit [Ping timeout: 240 seconds]
[14:50:13] <buildmaster> i686/aubio is broken (says rechenknecht).
[15:21:23] <buildmaster> i686/deepin-qt5integration is broken (says rechenknecht).
[15:29:37] <abaumann> on the buildmaster, settting CacheDir doens't always work (not for the chroot building scripts), they still fiddle with /var/cache/pacman/pkg, so I set up the bind mount point again..
[15:29:59] <abaumann> *abaumann had a look at his bike, recognized the fact he has to buy some tubes and tires first..
[15:31:00] <deep42thought> I think, the devtools build scripts might take the cache dir from the config inside /usr/share/devtools/pacman*.conf
[15:31:05] <deep42thought> ... or /etc/pacman.conf
[15:31:19] <deep42thought> I'm always confused, what get's actually used in the end
[15:31:36] <abaumann> adding a bind mountpoint is much simpler :-)
[15:31:47] <deep42thought> yes, indeed
[15:36:45] <buildmaster> i686/thunderbird is broken (says buildknecht2).
[15:38:19] <abaumann> yippie. the Mozilla break of the day.. :-)
[15:38:58] <deep42thought> buildmaster: tell us more!
[15:38:58] <buildmaster> deep42thought: knoidgests
[15:39:07] <deep42thought> \o/ still works :-D
[15:39:28] <abaumann> :-)
[15:56:20] <buildmaster> i686/chromium is broken (says nlopc46).
[15:56:47] -!- davor has quit [Ping timeout: 240 seconds]
[15:57:01] <abaumann> buildmaster: tell us more!
[15:57:02] <buildmaster> abaumann: XOrialunchicappids
[15:57:20] <abaumann> LOL
[15:57:24] <deep42thought> it's an amazing(ly useless) feature, isn't it?
[15:57:35] <abaumann> indeed :-)
[15:57:49] -!- davor has joined #archlinux32
[15:57:55] <deep42thought> markov-generated random words (this channel is the input for the markov generator)
[15:59:15] <abaumann> reminds me of a scientific paper generator a collegue at work found some years ago. :-)
[15:59:41] <deep42thought> snarxive vs. arxive?
[15:59:47] <abaumann> dunno.
[16:00:22] <abaumann> it was papers in the realm of computer science.. one even got accepted to a conference :->
[16:00:23] <deep42thought> there you have to decide wether a paper title is auto-generated or actually published on arxiv.org
[16:00:30] <deep42thought> lol
[16:02:27] <deep42thought> ok, I have to leave
[16:02:37] <deep42thought> cu tomorrow (my internet at home is still broken ...)
[16:02:41] <abaumann> ok, I have fun with glib2 and libsecret..
[16:02:59] -!- deep42thought has quit [Quit: Leaving.]
[16:06:34] <buildmaster> i686/vlc is broken (says nlopc46).
[16:09:03] <buildmaster> i686/virtualbox is broken (says nlopc43).
[16:22:44] -!- oaken-source has joined #archlinux32
[16:46:28] <buildmaster> i686/analitza is broken (says nlopc46).
[16:46:31] <buildmaster> i686/geary is broken (says buildknecht2).
[16:48:37] <buildmaster> i686/libkcddb is broken (says rechenknecht).
[16:53:03] <buildmaster> i686/libkcompactdisc is broken (says nlopc46).
[16:55:12] <buildmaster> i686/libkmahjongg is broken (says rechenknecht).
[17:10:54] <buildmaster> i686/kimap is broken (says buildknecht2).
[17:11:01] <buildmaster> i686/kpimtextedit is broken (says nlopc46).
[17:11:03] <buildmaster> i686/kldap is broken (says rechenknecht).
[17:21:16] <buildmaster> i686/kmbox is broken (says buildknecht2).
[17:29:08] <buildmaster> i686/akonadi-notes are broken (says nlopc46).
[17:32:48] <buildmaster> i686/kitinerary is broken (says buildknecht2).
[17:33:44] <abaumann> libsecret fails also on 64-bit, I bet it has to do with all this dbus-run-session stuff..
[17:33:55] <abaumann> running the mocked tests in a dbus environemnt..
[17:34:26] <abaumann> and glib2 fails inside its own logger with a G_BREAKPOINT, the logging code in there looks WAY TOO COMPLICATED..
[17:35:15] <abaumann> "Message recipient disconnected from message bus without replying" is the actually message in mock_service_stop.
[17:35:21] <abaumann> *actual
[17:36:02] <abaumann> so, it seems, this is just some sort of a race while testing dbus thingies - at least, I hope.
[17:36:19] <abaumann> what I don't get: why do those things pass upstream?
[17:36:31] <abaumann> or should I say, passed.
[17:36:33] <buildmaster> i686/kmousetool is broken (says nlopc46).
[17:36:53] <abaumann> mmh. something also seems to be lurking in KDE..
[17:37:29] <buildmaster> i686/kaccounts-integration is broken (says rechenknecht).
[17:37:29] -!- abaumann has quit [Quit: leaving]
[17:44:56] <buildmaster> i686/ksystemlog is broken (says rechenknecht).
[17:49:47] -!- oaken-source has quit [Ping timeout: 240 seconds]
[17:52:16] <buildmaster> i686/keditbookmarks are broken (says rechenknecht).
[17:54:06] -!- oaken-source has joined #archlinux32
[17:54:40] <buildmaster> i686/khelpcenter is broken (says nlopc46).
[18:05:30] <buildmaster> i686/kross-interpreters are broken (says rechenknecht).
[18:05:49] <buildmaster> i686/kate is broken (says eurobuild3).
[18:05:50] <buildmaster> i686/blinken is broken (says nlopc46).
[18:13:38] <buildmaster> i686/kalzium is broken (says nlopc46).
[18:15:53] <buildmaster> i686/kgeography is broken (says rechenknecht).
[18:20:35] <buildmaster> i686/kiten is broken (says nlopc46).
[18:22:04] <buildmaster> i686/kbruch is broken (says eurobuild3).
[18:23:52] <buildmaster> i686/klettres are broken (says rechenknecht).
[18:25:52] -!- buildmaster has quit [Remote host closed the connection]
[18:26:05] -!- buildmaster has joined #archlinux32
[18:27:22] <buildmaster> i686/kmplot is broken (says nlopc46).
[18:32:40] <buildmaster> i686/kturtle is broken (says rechenknecht).
[18:34:31] <buildmaster> i686/minuet is broken (says nlopc46).
[18:36:31] -!- buildmaster has quit [Remote host closed the connection]
[18:36:42] -!- buildmaster has joined #archlinux32
[18:37:40] <buildmaster> i686/ktouch is broken (says eurobuild3).
[18:40:47] <buildmaster> i686/rocs are broken (says rechenknecht).
[18:42:05] <buildmaster> i686/step is broken (says nlopc46).
[18:48:41] <buildmaster> i686/granatier is broken (says rechenknecht).
[18:50:58] <buildmaster> i686/bomber is broken (says eurobuild3).
[18:50:59] <buildmaster> i686/kapman is broken (says nlopc46).
[18:54:54] -!- ofara has quit [Quit: ofara]
[18:56:07] <buildmaster> i686/kblocks are broken (says rechenknecht).
[18:59:15] <buildmaster> i686/kdiamond is broken (says nlopc46).
[19:00:30] -!- AndrevS has joined #archlinux32
[19:03:27] <buildmaster> i686/kgoldrunner is broken (says rechenknecht).
[19:06:00] <buildmaster> i686/kbreakout is broken (says eurobuild3).
[19:07:05] <buildmaster> i686/killbots are broken (says nlopc46).
[19:10:40] <buildmaster> i686/kjumpingcube is broken (says rechenknecht).
[19:14:12] <buildmaster> i686/kmines are broken (says nlopc46).
[19:20:35] <buildmaster> i686/knetwalk is broken (says rechenknecht).
[19:22:14] <buildmaster> i686/klines are broken (says eurobuild3).
[19:22:15] <buildmaster> i686/kolf is broken (says nlopc46).
[19:29:45] <buildmaster> i686/kpatience is broken (says nlopc46).
[19:30:52] <buildmaster> i686/kreversi is broken (says rechenknecht).
[19:44:44] <buildmaster> i686/ksirk is broken (says eurobuild3).
[19:46:04] <buildmaster> i686/ksudoku is broken (says rechenknecht).
[19:47:06] <buildmaster> i686/ksquares are broken (says nlopc46).
[19:49:07] -!- buildmaster has quit [Remote host closed the connection]
[19:49:19] -!- buildmaster has joined #archlinux32
[19:53:41] <buildmaster> i686/lskat is broken (says rechenknecht).
[19:55:29] <buildmaster> i686/palapeli is broken (says nlopc46).
[20:03:23] <buildmaster> i686/kcolorchooser is broken (says rechenknecht).
[20:05:44] <buildmaster> i686/kolourpaint is broken (says nlopc46).
[20:05:51] <buildmaster> i686/kubrick is broken (says eurobuild3).
[20:12:40] <buildmaster> i686/spectacle is broken (says rechenknecht).
[20:16:31] <buildmaster> i686/dragon is broken (says nlopc46).
[20:20:54] <buildmaster> i686/ffmpegthumbs are broken (says rechenknecht).
[20:51:22] <buildmaster> i686/kamoso is broken (says rechenknecht).
[20:56:21] <buildmaster> i686/juk is broken (says nlopc46).
[21:00:02] <buildmaster> i686/kmix is broken (says rechenknecht).
[21:04:24] <buildmaster> i686/kwave is broken (says nlopc46).
[21:04:37] -!- Dimtree has quit [Ping timeout: 245 seconds]
[21:07:41] <buildmaster> i686/kdenetwork-filesharing is broken (says rechenknecht).
[21:11:30] <buildmaster> i686/krdc is broken (says nlopc46).
[21:15:35] <buildmaster> i686/zeroconf-ioslave is broken (says rechenknecht).
[21:16:34] <buildmaster> i686/kio-extras are broken (says buildknecht2).
[21:21:52] <buildmaster> i686/cervisia is broken (says nlopc46).
[21:23:10] <buildmaster> i686/kde-dev-utils are broken (says rechenknecht).
[21:29:24] <buildmaster> i686/kirigami-gallery is broken (says rechenknecht).
[21:31:27] <buildmaster> i686/kompare is broken (says nlopc46).
[21:35:43] <buildmaster> i686/poxml is broken (says rechenknecht).
[21:39:43] <buildmaster> i686/ark is broken (says nlopc46).
[21:42:53] <buildmaster> i686/kbackup is broken (says rechenknecht).
[21:46:13] <buildmaster> i686/kcharselect is broken (says nlopc46).
[21:50:07] <buildmaster> i686/kdf is broken (says rechenknecht).
[21:52:42] <buildmaster> i686/kfloppy is broken (says nlopc46).
[21:57:45] <buildmaster> i686/ktimer is broken (says rechenknecht).
[21:59:35] <buildmaster> i686/print-manager is broken (says nlopc46).
[22:09:34] <buildmaster> i686/dune is broken (says rechenknecht).
[22:10:40] <buildmaster> i686/tracker-miners are broken (says buildknecht2).
[22:12:40] -!- buildmaster has quit [Remote host closed the connection]
[22:12:52] -!- buildmaster has joined #archlinux32
[22:14:50] <buildmaster> i686/xf86-video-amdgpu is broken (says buildknecht2).
[22:20:47] -!- yans has joined #archlinux32
[23:05:24] -!- bill-auger has quit [Remote host closed the connection]
[23:07:50] -!- bill-auger has joined #archlinux32
[23:29:02] <buildmaster> i686/gaupol is broken (says nlopc46).
[23:57:03] <buildmaster> i686/khangman is broken (says nlopc46).
[23:59:40] -!- oaken-source has quit [Ping timeout: 246 seconds]