#archlinux32 | Logs for 2022-02-17
Back
[01:35:25] -!- petris has quit [Ping timeout: 240 seconds]
[01:36:57] -!- petris has joined #archlinux32
[01:47:48] -!- finsternis has quit [Ping timeout: 240 seconds]
[02:11:41] -!- jirido has quit [Ping timeout: 256 seconds]
[02:13:23] -!- jirido has joined #archlinux32
[02:15:20] -!- KillerWasp has quit [Remote host closed the connection]
[02:16:32] -!- KillerWasp has joined #archlinux32
[02:28:50] -!- KillerWasp has parted #archlinux32
[03:25:18] -!- jonathon has quit [Remote host closed the connection]
[03:25:29] -!- jonathon has joined #archlinux32
[05:07:28] -!- volki57 has quit [Remote host closed the connection]
[05:10:12] -!- volki57 has joined #archlinux32
[07:04:04] <girls> KillerWasp: I'm using `trust extract-compat` - see https://git.eckner.net
[07:04:05] <phrik> Title: local-certificates - Erich/archlinuxewe - Unnamed repository; edit this file 'description' to name the repository. (at git.eckner.net)
[07:04:19] <girls> therefor, I put the certificates into /usr/share/ca-certificates/trust-source/anchors/
[09:43:18] -!- finsternis has joined #archlinux32
[11:47:55] -!- GNUtoo has quit [Ping timeout: 240 seconds]
[12:20:27] -!- GNUtoo has joined #archlinux32
[12:51:56] -!- abaumann has joined #archlinux32
[12:51:57] <buildmaster> Hi abaumann!
[12:51:57] <buildmaster> !rq abaumann
[12:51:57] <phrik> buildmaster: <abaumann> * abaumann looks at his hands.. counts them.. shakes his head and sighs about the fact he cannot grow some more..
[12:52:21] <abaumann> so, my build machines are switched off for now, I need them to test glibc 2.35 issues..
[12:52:55] <abaumann> If anyone knows how test cases in glibc-2.35/time/Makefile suddently get tests-time64 := -time64, I'm clad.
[12:52:58] <abaumann> *glad
[12:53:18] <abaumann> I cannot imagine something would rewrite those hand-written makefiles..
[12:53:46] <abaumann> and before I start to build 20 minutes in strace just to find out what is changing this file.. :->
[12:56:24] <abaumann> skip_test tst-adjtime time/Makefile
[12:56:24] <abaumann> skip_test tst-clock2 time/Makefile
[12:56:26] <abaumann> ahem. :-)
[12:58:46] <abaumann> I generally don't like the approach with sed and patching stuff.. patches are really hard to read compared to a simple plain diff.
[13:00:51] <abaumann> why is this not a problem upstream, I wonder?
[13:23:11] -!- abaumann has quit [Quit: leaving]
[13:27:24] -!- lithiumpt has quit [Ping timeout: 240 seconds]
[14:57:07] -!- lithiumpt has joined #archlinux32
[17:47:24] -!- sunshavi has quit [Ping timeout: 272 seconds]
[17:56:10] -!- sunshavi has joined #archlinux32
[19:58:17] <girls> but a diff is less maintainable
[20:23:51] -!- lithium_pt has joined #archlinux32
[20:23:55] -!- lithiumpt has quit [Ping timeout: 256 seconds]
[20:46:26] -!- abaumann has joined #archlinux32
[20:46:27] <buildmaster> Hi abaumann!
[20:46:27] <buildmaster> !rq abaumann
[20:46:27] <phrik> buildmaster: <abaumann> well, we found out that those disks _ARE_ actually slow disks then :-)
[20:47:07] <abaumann> girls: yes, maybe. But you need more brain power to understand it, especially sed brain power :-)
[20:48:21] <abaumann> glibc 2.35 is more or less ok (still some failing tests), binutils has one outstanding failing test "FAIL: Build ifunc-1a with PIE -z ibtplt", I don't even understand yet, what this is testing exacltly..
[20:49:09] <abaumann> "Generate Intel Indirect Branch Tracking (IBT) enabled PLT entries" mmg
[20:49:58] <abaumann> I hope we don't build double-indirected PLTs in libraries. Well, I'll yollo that test (aka ignore it)
[20:53:05] <T`aZ> last time I ignored a failing glibc test, was 20y ago (lfs system), make install anyway, no software could be launched anymore, "fun" night ahead that was
[20:57:52] <abaumann> yes. I know. :-)
[20:58:37] <abaumann> that's why I write it here. Uptream can also not test all stuff in glibc due to the way the build containers work. It's a tradeoff..
[20:59:26] <abaumann> Last time I yolloed a glibc, we run into the nice ISA 686 issue.. that was a mess to clean up.. *sigh*
[21:01:55] <abaumann> The binutils ibtplt we find out fast, when it goes wrong
[21:02:17] <abaumann> and the untested glibc tests are mainly around time64. And I don't expect to maintain Archlinux32 in 2038.. ;-)
[21:09:05] <T`aZ> isnt that more about using the correct version of struct time_t(64bits or not) , than witnessing the actual 32bits overflow in 2038 ?
[21:09:23] <T`aZ> it's only 16 years away
[21:12:41] <abaumann> 16 years, then I'm retired :-)
[21:12:55] <abaumann> I was there for the Y2K, that was fun enough..
[21:22:37] -!- bill-auger has quit [Ping timeout: 240 seconds]
[21:30:38] <sunshavi> Hi. @2038 I could still be using my compaq n600c :)
[22:21:59] <abaumann> binutils: /usr/include/libiberty/ansidecl.h exists in filesystem (owned by gcc)
[22:22:17] <abaumann> grr, some file hopping between gcc and binutils
[22:23:01] <abaumann> I need gcc to build a new gcc with a new binutils, so this demands maybe for a straw 'overwrite='*'' (aka force)
[22:23:51] <abaumann> or I make another run with binutils, where I delete /usr/include/libiberty, then I rebuild gcc, then I rebuild binutils with libiberty installed
[22:31:32] -!- abaumann has quit [Quit: leaving]
[22:37:36] -!- drathir_tor has quit [Ping timeout: 240 seconds]
[22:50:57] -!- bill-auger has joined #archlinux32
[23:33:24] -!- petris has quit [Ping timeout: 240 seconds]
[23:35:36] -!- petris has joined #archlinux32
[23:44:26] -!- drathir_tor has joined #archlinux32