Switch to DuckDuckGo Search
   September 15, 2019  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | >

Toggle Join/Part | bottom
[00:00:33] <Reinhilde> you're probably --disable'ing everything that i'm not --disable'ing
[00:01:51] <despair86> >silly broken bits
[00:01:54] <despair86> wut
[00:03:27] <despair86> as far as i can tell, sol2.11.4 and illumos are roughly source-compat, less kstat v2 and xpg7 *shrugs*
[00:04:00] <richlowe> despair86: there's a bunch of changes to the gdb in solaris that don't actually do anything.
[00:04:28] <despair86> ohh
[00:07:46] <Reinhilde> sys_uuidsys
[00:17:19] <andyf> jlevon - https://github.com/omniosorg/illumos-omnios/pull/537 is what I am proposing to do about maintaining /etc/versions/build on OmniOS
[00:17:37] <andyf> Not an SMF service in the end but tied into `bootadm update-archive`
[00:19:05] <andyf> I don't know if it is something that you'd be interested in seeing go upstream
[00:20:49] *** Reinhilde is now known as AmyMalik
[00:46:04] *** kujira <kujira!~null@mobile-166-176-186-244.mycingular.net> has quit IRC (Ping timeout: 265 seconds)
[00:54:34] *** jimklimov <jimklimov!~jimklimov@ip-86-49-254-64.net.upcbroadband.cz> has quit IRC (Quit: Leaving.)
[01:29:29] <jlevon> andyf: I don't see why it couldn't go into illumos-gate, sure. we'd not do that, but that's something for us not illumos-gate.
[02:13:53] <jbk> hrm..
[02:14:49] <jbk> so rust allows identifiers to use unicode codepoints... so when demangling, the resulting name could contain non-ASCII codepoints
[02:15:00] <jbk> I'm wondering how best to handle that in the demangling library
[02:15:56] <jbk> it encoded them using punycode (yay!), so in the process of decoding, I think I can end up with a wchar_t * (I've not really done much here)
[02:17:17] <jbk> and could then just wcstombs or such
[02:18:21] <jbk> the existing demanglers always return a UTF-8 encoded string regardless of locale
[02:18:44] <jbk> which seems wrong
[02:19:16] <despair86> yikes
[02:24:17] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Ping timeout: 276 seconds)
[03:15:38] *** Teknix <Teknix!~pds@172.58.43.210> has quit IRC (Ping timeout: 240 seconds)
[03:19:07] *** Teknix <Teknix!~pds@172.58.43.210> has joined #illumos
[03:20:00] *** ed209 <ed209!~ed209@165.225.128.67> has quit IRC (Remote host closed the connection)
[03:20:08] *** ed209 <ed209!~ed209@165.225.128.67> has joined #illumos
[03:41:39] *** patdk-lap <patdk-lap!~patrickdk@208.94.189.182> has quit IRC (Ping timeout: 258 seconds)
[03:43:38] *** XV8 <XV8!~XV8@2601:151:4403:9c0:2d8d:3ca8:d3d0:e24e> has quit IRC (Ping timeout: 245 seconds)
[03:48:17] *** patdk-lap <patdk-lap!~patrickdk@208.94.189.182> has joined #illumos
[04:01:19] *** mecrobio <mecrobio!~mecrobio@199.red-88-4-59.dynamicip.rima-tde.net> has joined #illumos
[04:25:01] <LeftWing> jbk: I would want an explicitly UTF-8 symbol to get at an explicitly UTF-8 version
[04:25:35] <LeftWing> Which could be distinct from a wchar/other locale sensitive version
[04:29:59] <richlowe> I would want the library to give me utf that we then tried to encode
[04:33:17] <richlowe> and to be as far away form anything like wchar_t as possible, for as long as possible
[04:33:18] <richlowe> if that's not clear
[04:38:48] <jbk> well when you do the punycode decoding, you're basically left with a uint_t[] of codepoints -- so the question is what to do with that
[04:39:23] <jbk> the other implementations take each codepoint and translate into a UTF-8 byte sequence
[04:39:37] <jbk> i suspect we probably want to respect the current locale in some manner
[04:39:45] <richlowe> the client would need to encode it, presumably.
[04:39:55] <jbk> what client?
[04:39:56] <richlowe> if it's going to display it, or the like
[04:40:03] <richlowe> whoever is using your demangling library
[04:40:14] <richlowe> if it's going to display the string, it needs to be encoded as the locale
[04:40:22] <richlowe> if it's doing something else with it, perhaps it does not, and in fact should not
[04:40:25] <richlowe> so you can't really decide for it
[04:42:27] <jbk> so just return the utf-8 encoded string, and let whatever do any conversions it wants?
[04:43:20] *** mecrobio <mecrobio!~mecrobio@199.red-88-4-59.dynamicip.rima-tde.net> has quit IRC (Remote host closed the connection)
[04:43:25] <richlowe> That's what I think
[04:43:36] <richlowe> there's likely better people to take advice frm
[04:51:08] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[04:56:52] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Remote host closed the connection)
[04:57:37] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[05:06:52] *** XV8 <XV8!~XV8@2601:151:4403:9c0:959a:2ddf:7325:17a9> has joined #illumos
[05:57:12] <LeftWing> Right, I would at a minimum want a way to get a (char *) of the UTF-8 encoding of the string
[05:57:27] <LeftWing> So that I could do whatever I then need to do with it
[05:57:52] <LeftWing> But I can see how some programs (e.g., clients that will display it in a terminal or whatever) would want a locale-sensitive one as well
[05:58:26] <LeftWing> But I wouldn't want to conflate those two, and if I had to have one or the other I'd have the UTF-8-explicitly version
[06:07:48] *** XV8 <XV8!~XV8@2601:151:4403:9c0:959a:2ddf:7325:17a9> has quit IRC (Ping timeout: 245 seconds)
[06:39:59] *** wonko <wonko!~quassel@75.52.174.33> has quit IRC (Quit: https://quassel-irc.org - Chat comfortably. Anywhere.)
[06:45:47] *** wonko <wonko!~quassel@2600:1700:5ee7:1c80:5ce0:c3ff:fe1e:104e> has joined #illumos
[06:48:44] *** psydroid <psydroid!psydroidma@gateway/shell/matrix.org/x-vutofcaznmqnfmmg> has quit IRC (*.net *.split)
[06:48:44] *** SpicyShibe <SpicyShibe!~DB@2601:187:8400:5::83c> has quit IRC (*.net *.split)
[06:48:44] *** igitoor <igitoor!igitur@unaffiliated/contempt> has quit IRC (*.net *.split)
[06:48:45] *** gjnoonan <gjnoonan!sid95422@gateway/web/irccloud.com/x-ysjhvgjeurzqymqh> has quit IRC (*.net *.split)
[06:48:45] *** Woodstock <Woodstock!~grumpf@beiss.dessnourst.de> has quit IRC (*.net *.split)
[06:48:45] *** ballew <ballew!sid244342@gateway/web/irccloud.com/x-ndgxneqmzexscfxq> has quit IRC (*.net *.split)
[06:48:45] *** ryaeng <ryaeng!sid364071@gateway/web/irccloud.com/x-ddjkglquiwoqkcwp> has quit IRC (*.net *.split)
[06:48:45] *** c3bolla <c3bolla!c3bolla@gateway/shell/panicbnc/x-tqzrzbukmjtjvkbi> has quit IRC (*.net *.split)
[06:48:45] *** sensille <sensille!~arne@dsnsi.rzone.de> has quit IRC (*.net *.split)
[06:48:57] *** Qatz <Qatz!~DB@2601:187:8400:5::83c> has joined #illumos
[06:49:00] *** c3bolla <c3bolla!c3bolla@gateway/shell/panicbnc/x-uoikgxlcooehkjga> has joined #illumos
[06:49:04] *** ryaeng <ryaeng!sid364071@gateway/web/irccloud.com/x-ulbnhunqlczaqlxw> has joined #illumos
[06:49:09] *** ballew <ballew!sid244342@gateway/web/irccloud.com/x-kmwllnzoglmokyuq> has joined #illumos
[06:49:16] *** gjnoonan <gjnoonan!sid95422@gateway/web/irccloud.com/x-itimzmcqqysgusiv> has joined #illumos
[06:51:01] *** psydroid <psydroid!psydroidma@gateway/shell/matrix.org/x-cuhimwpyrteuhtjr> has joined #illumos
[06:54:19] *** igitoor <igitoor!igitur@unaffiliated/contempt> has joined #illumos
[06:54:19] *** Woodstock <Woodstock!~grumpf@beiss.dessnourst.de> has joined #illumos
[06:54:19] *** sensille <sensille!~arne@dsnsi.rzone.de> has joined #illumos
[07:02:19] <AmyMalik> does anyone know if non-illumos gdb on elf platforms can debug cores made by an elf binary on an illumos system
[07:08:49] <LeftWing> AmyMalik: At the very least I don't know that they'd be able to interpret all of the ELF notes we might add to a core file
[07:38:13] *** despair86 <despair86!~despair@185.10.68.109> has quit IRC (Remote host closed the connection)
[07:38:39] *** despair86 <despair86!~despair@185.10.68.109> has joined #illumos
[08:38:12] *** despair86 <despair86!~despair@185.10.68.109> has quit IRC (Remote host closed the connection)
[08:38:33] *** despair86 <despair86!~despair@185.10.68.109> has joined #illumos
[08:55:44] *** alanc <alanc!~alanc@129.157.69.38> has quit IRC (Remote host closed the connection)
[08:56:12] *** alanc <alanc!~alanc@129.157.69.38> has joined #illumos
[10:00:42] *** despair86 <despair86!~despair@185.10.68.109> has quit IRC (Remote host closed the connection)
[10:01:02] *** despair86 <despair86!~despair@exit.i2p.rocks> has joined #illumos
[10:05:10] *** andy_js <andy_js!~andy@94.12.192.123> has joined #illumos
[10:08:33] <AmyMalik> is there any developer interest in making illumos run on a raspberry pi? apologies if we're way past that juncture
[10:08:57] <AmyMalik> but i haven't heard of anything except linux and riscos
[10:57:58] <andyf> jlevon ok thanks - I'll mark it for upstreaming at some point.. things are a bit hectic as we are in the last two weeks before freezing OmniOS ready for the next stable release
[11:12:54] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[11:35:59] *** jimklimov <jimklimov!~jimklimov@ip-86-49-254-64.net.upcbroadband.cz> has joined #illumos
[11:36:45] *** spectral` <spectral`!~julian@os.k/spectral> has quit IRC (Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net)
[11:36:56] *** spectral` <spectral`!~julian@spectral.powered.by.lunarbnc.net> has joined #illumos
[11:36:56] *** spectral` <spectral`!~julian@spectral.powered.by.lunarbnc.net> has quit IRC (Changing host)
[11:36:56] *** spectral` <spectral`!~julian@os.k/spectral> has joined #illumos
[11:49:11] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[11:52:05] *** baojg <baojg!~baojg@162.243.44.213> has joined #illumos
[11:53:01] *** merzo <merzo!~merzo@22-2-132-95.pool.ukrtel.net> has joined #illumos
[12:20:00] *** ed209 <ed209!~ed209@165.225.128.67> has quit IRC (Remote host closed the connection)
[12:20:08] *** ed209 <ed209!~ed209@165.225.128.67> has joined #illumos
[12:22:34] *** jellydonut <jellydonut!~quassel@s91904422.blix.com> has quit IRC (Quit: jellydonut)
[12:25:56] *** jellydonut <jellydonut!~quassel@s91904424.blix.com> has joined #illumos
[12:28:50] *** Exagone313 <Exagone313!exa@3131313.xyz> has quit IRC (Quit: see ya!)
[12:28:56] <tsoome> AmyMalik: if anything then arm64 (most likely)….
[12:35:33] *** jellydonut <jellydonut!~quassel@s91904424.blix.com> has quit IRC (Quit: jellydonut)
[12:48:48] *** Exagone313 <Exagone313!exa@3131313.xyz> has joined #illumos
[12:49:14] *** Exagone313 <Exagone313!exa@3131313.xyz> has quit IRC (Remote host closed the connection)
[12:53:42] *** Exagone313 <Exagone313!exa@3131313.xyz> has joined #illumos
[12:57:37] *** Exagone313 <Exagone313!exa@3131313.xyz> has quit IRC (Client Quit)
[12:59:05] *** Exagone313 <Exagone313!exa@3131313.xyz> has joined #illumos
[12:59:58] *** Exagone313 <Exagone313!exa@3131313.xyz> has quit IRC (Remote host closed the connection)
[13:08:11] *** Exagone313 <Exagone313!exa@3131313.xyz> has joined #illumos
[13:33:32] *** jellydonut <jellydonut!~quassel@s91904421.blix.com> has joined #illumos
[14:15:12] *** despair86 <despair86!~despair@exit.i2p.rocks> has quit IRC (Remote host closed the connection)
[14:15:34] *** despair86 <despair86!~despair@exit.i2p.rocks> has joined #illumos
[14:30:55] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[14:34:34] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[14:40:17] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[14:41:53] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[14:50:44] *** despair86 <despair86!~despair@exit.i2p.rocks> has quit IRC (Remote host closed the connection)
[14:51:15] *** despair86 <despair86!~despair@exit.i2p.rocks> has joined #illumos
[15:01:21] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[15:01:51] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[15:08:24] *** jrick <jrick!~jrick@unaffiliated/jrick> has quit IRC (Ping timeout: 246 seconds)
[15:30:02] *** MilkmanDan <MilkmanDan!~dan@wilug/expat/MilkmanDan> has quit IRC (Ping timeout: 245 seconds)
[15:31:33] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Ping timeout: 245 seconds)
[15:32:00] *** MilkmanDan <MilkmanDan!~dan@wilug/expat/MilkmanDan> has joined #illumos
[15:34:23] *** Kurlon <Kurlon!~Kurlon@cpe-67-253-141-249.rochester.res.rr.com> has quit IRC (Remote host closed the connection)
[16:15:41] *** despair86 <despair86!~despair@exit.i2p.rocks> has quit IRC (Ping timeout: 258 seconds)
[16:31:02] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[16:36:04] *** XV8 <XV8!~XV8@2601:151:4403:9c0:a942:b7ba:af78:a46d> has joined #illumos
[16:36:08] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Ping timeout: 245 seconds)
[16:45:45] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[16:46:13] *** clapont <clapont!~clapont@unaffiliated/clapont> has quit IRC (Ping timeout: 246 seconds)
[16:47:31] *** kohju <kohju!~kohju@gw.justplayer.com> has quit IRC (Quit: Tiarra 0.1+svn-36726: SIGINT received; exit)
[16:48:55] *** kohju <kohju!~kohju@gw.justplayer.com> has joined #illumos
[16:49:27] <toasterson> AmyMalik (IRC): FreeBSD has a port and images for Raspberry Pi up until 3
[16:49:42] <toasterson> and Raspberry4 is arm64 iirc
[16:50:55] <toasterson> Raspberry has been arm64 since 3 and 2 v1.2
[16:53:30] *** clapont <clapont!~clapont@unaffiliated/clapont> has joined #illumos
[16:56:42] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[16:58:02] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[17:07:44] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[17:09:35] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[17:14:14] *** mnowak_ <mnowak_!~mnowak_@94.142.238.232> has quit IRC (Quit: Leaving)
[17:16:51] *** Teknix <Teknix!~pds@172.58.43.210> has quit IRC (Ping timeout: 240 seconds)
[17:18:22] *** clapont <clapont!~clapont@unaffiliated/clapont> has quit IRC (Ping timeout: 245 seconds)
[17:20:43] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Ping timeout: 245 seconds)
[17:22:14] *** mnowak_ <mnowak_!~mnowak_@94.142.238.232> has joined #illumos
[17:22:23] *** Teknix <Teknix!~pds@172.58.27.94> has joined #illumos
[17:25:17] *** clapont <clapont!~clapont@unaffiliated/clapont> has joined #illumos
[17:26:20] *** Teknix <Teknix!~pds@172.58.27.94> has quit IRC (Remote host closed the connection)
[17:28:17] *** Teknix <Teknix!~pds@172.58.27.94> has joined #illumos
[17:37:25] *** Teknix <Teknix!~pds@172.58.27.94> has quit IRC (Remote host closed the connection)
[17:38:48] *** Teknix <Teknix!~pds@172.58.27.94> has joined #illumos
[17:41:31] *** Teknix <Teknix!~pds@172.58.27.94> has quit IRC (Remote host closed the connection)
[17:47:18] *** Teknix <Teknix!~pds@172.58.44.69> has joined #illumos
[17:47:18] *** clapont <clapont!~clapont@unaffiliated/clapont> has quit IRC (Read error: Connection reset by peer)
[17:53:04] *** clapont <clapont!~clapont@unaffiliated/clapont> has joined #illumos
[18:21:36] *** clapont <clapont!~clapont@unaffiliated/clapont> has quit IRC (Read error: Connection reset by peer)
[18:41:25] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has quit IRC (Ping timeout: 265 seconds)
[19:13:33] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has joined #illumos
[19:13:45] *** Teknix <Teknix!~pds@172.58.44.69> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[20:18:58] *** despair86 <despair86!~despair@exit.i2p.rocks> has joined #illumos
[20:30:56] <AmyMalik> will illumos ever again run on 32 bit or is there just no interest
[20:32:09] <despair86> the bits are still there
[20:32:30] <despair86> but probably only for RISC archs with 64-bit atomics
[20:32:46] <jbk> well do you mean kernel or userland
[20:33:15] <jbk> i don't know if anyone wants to invest the time to keep both 32 and 64 bit kernels working
[20:33:48] <despair86> both, i heard someone got a UFS-based Solaris running on Acorn v7 a while back with GCC 5.x...
[20:33:51] <despair86> but yeah
[20:34:03] *** XV8 <XV8!~XV8@2601:151:4403:9c0:a942:b7ba:af78:a46d> has quit IRC (Ping timeout: 245 seconds)
[20:34:12] <despair86> interest is low now that Acorn et al have 64-bit archs now
[20:34:19] <despair86> ^H^H^H
[20:35:15] <AmyMalik> jbk: i mean kernel on i86pc
[20:35:53] <despair86> oh
[20:36:01] * despair86 shrugs
[20:38:33] <jbk> probably not..I mean no one's going to stop anyone from reviving the bits if they want, but I don't think anyone currently doing any kernel work has expressed a desire to revive it
[20:39:34] <despair86> ^
[20:40:00] <despair86> the only 32-bit variant of sol2.11 left is tribblix milestone 20
[20:47:04] *** despair86 <despair86!~despair@exit.i2p.rocks> has quit IRC (Quit: Leaving)
[20:50:02] *** despair86 <despair86!~despair@exit.i2p.rocks> has joined #illumos
[21:16:20] *** jimklimov <jimklimov!~jimklimov@ip-86-49-254-64.net.upcbroadband.cz> has quit IRC (Quit: Leaving.)
[21:16:32] *** jimklimov <jimklimov!~jimklimov@ip-86-49-254-64.net.upcbroadband.cz> has joined #illumos
[21:20:55] *** jimklimov <jimklimov!~jimklimov@ip-86-49-254-64.net.upcbroadband.cz> has quit IRC (Ping timeout: 265 seconds)
[21:57:12] *** despair86 <despair86!~despair@exit.i2p.rocks> has quit IRC (Remote host closed the connection)
[21:57:33] *** despair86 <despair86!~despair@exit.i2p.rocks> has joined #illumos
[21:59:36] <AmyMalik> this is probably not illumos' problem
[21:59:38] <AmyMalik> but
[21:59:48] <AmyMalik> procfs.c: In member function 'virtual void procfs_target::create_inferior(const char*, const string&, char**, int)':
[21:59:50] <AmyMalik> procfs.c:3062:12: error: 'char* std::strncpy(char*, const char*, std::size_t)' output truncated before terminating nul copying as many bytes from a string as its length [-Werror=stringop-truncation]
[21:59:52] <AmyMalik> strncpy (tryname, p, len);
[21:59:54] <AmyMalik> ~~~~~~~~^~~~~~~~~~~~~~~~~
[21:59:57] <AmyMalik> procfs.c:3061:19: note: length computed here
[21:59:58] <AmyMalik> len = strlen (p);
[22:00:00] <AmyMalik> ~~~~~~~^~~
[22:00:02] <AmyMalik> attempting to compile gdb head
[22:00:04] <AmyMalik> who do I report to?
[22:04:09] *** mnowak_ <mnowak_!~mnowak_@94.142.238.232> has quit IRC (Quit: Leaving)
[22:06:57] *** mnowak_ <mnowak_!~mnowak_@94.142.238.232> has joined #illumos
[22:07:02] <jbk> that sounds like a gdb problem if i'm reading it correctly
[22:07:20] <jbk> I _think_ it's maybe wanting to see 'strncpy(tryname, p, len - 1)'
[22:07:38] <jbk> or + 1
[22:08:36] <jbk> probably len + 1
[22:08:52] <jbk> i'd maybe try that and see if the error goes away
[22:09:21] <AmyMalik> jbk: are you telling me I need to edit gdb's source code? ;p
[22:09:37] <jbk> well i'm assuming that procfs.c is from the gdb source and not illumos
[22:10:07] <AmyMalik> yes
[22:10:09] <AmyMalik> it is
[22:10:19] <jbk> and the error doesn't appear to be something with our headers or such
[22:10:25] <jbk> but a logic error in that file
[22:10:34] <AmyMalik> procfs.c: In member function 'virtual void procfs_target::create_inferior(const char*, const string&, char**, int)':
[22:10:36] <AmyMalik> procfs.c:3062:12: error: 'char* std::strncpy(char*, const char*, std::size_t)' specified bound depends on the length of the source argument [-Werror=stringop-overflow=]
[22:10:38] <AmyMalik> strncpy (tryname, p, len + 1);
[22:10:40] <AmyMalik> ~~~~~~~~^~~~~~~~~~~~~~~~~~~~~
[22:10:43] <AmyMalik> and +1 introduced a logic error
[22:10:50] <AmyMalik> jbk: you're absolutely right
[22:11:02] <AmyMalik> i'm only here because i'm compiling on illumos and everyone hates sun, including me
[22:12:26] <AmyMalik> by "including me" I mean I hate how gnuconfigure scripts detect 'i386-pc-solaris2.11' when they should detect 'x86_64-pc-solaris2.11'
[22:13:55] <AmyMalik> i'll go bother ##binutils
[22:14:29] <jbk> yeah.. looking at the surrounding code.. it might need something like
[22:14:44] <jbk> size_t trylen = strlen(path) + strlen(shell_file) + 2;
[22:14:54] <jbk> tryname = (char *)alloca(trylen);
[22:14:55] <jbk> ...
[22:14:59] *** clapont <clapont!~clapont@unaffiliated/clapont> has joined #illumos
[22:15:00] <jbk> strncpy(tryname, p, trylen);
[22:15:04] <jbk> to make it happy
[22:15:13] <jbk> actually no
[22:15:28] <jbk> it's just a bad error
[22:15:43] <AmyMalik> a bad error?
[22:15:48] <jbk> a false alarm
[22:15:52] <jbk> it's not an actually error
[22:16:04] <AmyMalik> it is an error though because gdb uses -Werror=all
[22:16:26] <jbk> because sizeof (tryname) > strlen(path), p is a subset of path
[22:16:38] <jbk> and it wants to copy a portion of path into tryname
[22:17:08] <jbk> yeah, the compiler isn't smart enough to know that it's safe
[22:17:12] *** clapont <clapont!~clapont@unaffiliated/clapont> has quit IRC (Read error: Connection reset by peer)
[22:17:44] <AmyMalik> the whole stack is gnu from the compiler on up, how can they not know
[22:17:45] <jbk> (which would be somewhat challenging)
[22:18:09] <jbk> it has nothing to do with gnu or not
[22:18:22] <jbk> it's a matter of how sophisticated the analysis in the complier can be
[22:18:39] <AmyMalik> i know it's nothing to do with who
[22:18:43] <jbk> and they probably don't test every version of gdb with every version of gcc
[22:18:58] <AmyMalik> but you'd think the developers would be smart enough to, you know
[22:19:02] <AmyMalik> MAKE IT WORK ON THEIR OWN SOFTWARE?
[22:19:34] <AmyMalik> especially since gdb is the only debugger worth mentioning for Novices™ on most unix
[22:19:47] <AmyMalik> i know, that's like expecting microsoft to test windows against itself
[22:20:00] *** ed209 <ed209!~ed209@165.225.128.67> has quit IRC (Remote host closed the connection)
[22:20:03] <jbk> well you're more likely to get the gdb people to fix it either be explicitly muting the error there or someone reworking the code to make the compiler happy
[22:20:07] *** ed209 <ed209!~ed209@165.225.128.67> has joined #illumos
[22:20:38] <jbk> well when you look at the toolchain, the number of possible combinations tends to explode
[22:21:02] <AmyMalik> GNU have become what they hate, tbh
[22:21:10] <AmyMalik> and that's microsoft
[22:21:45] <jbk> i mean here, probably a simpler, and less likely approach would be something like
[22:22:09] <AmyMalik> this is how tesseract fixed it https://github.com/tesseract-ocr/tesseract/pull/1398/commits/caa4faff6580d630dee82cf01c0e29ffdd9e90cf
[22:22:23] <AmyMalik> to use memcpy instead of strncpy?
[22:22:47] <jbk> hmm no
[22:22:48] <AmyMalik> the program might crash though
[22:22:52] <jbk> maybe change the type of 'len' to size_t
[22:22:56] <jbk> at line 3048
[22:23:09] <jbk> well try that first
[22:23:11] <jbk> then memcpy
[22:23:33] <AmyMalik> still unhappy
[22:23:53] <richlowe> if you're trying to fix the procfs largefile things, it's noble, but annoying as hell.
[22:24:14] <richlowe> there's a reason I gave up :)
[22:24:57] <AmyMalik> richlowe: it's not that, i don't think
[22:25:27] <AmyMalik> jbk: there's this braindead way in which gdb calculates the length of char *p.
[22:25:35] <AmyMalik> 3058: if (p1 != NULL)
[22:25:37] <AmyMalik> 3059: len = p1 - p;
[22:25:39] <AmyMalik> 3060: else
[22:25:41] <AmyMalik> 3061: len = strlen (p);
[22:26:16] <jbk> well the simpler thing would be to make a copy of path, then use strtok or strsep
[22:27:23] <jbk> basically (using pseudocode) paths = split(path, ':'); for each p in path { tryname = p + shell_file; if (access(tryname) .. }
[22:27:58] <AmyMalik> this is daunting as fuck
[22:29:25] <richlowe> if it makes you feel better, git just auto-merged one of my makefiles with a random zfs test shell script.
[22:29:31] <richlowe> so everything sucks, not just gdb.
[22:29:41] <AmyMalik> richlowe: what?
[22:29:48] <AmyMalik> jbk: compiles fine with memcpy
[22:30:01] <AmyMalik> possible bug introduced?
[22:30:27] <jbk> i don't think so -- it's NUL terminating the string
[22:31:00] <jbk> and the size of tryname should always be large enough to hold the results
[22:39:14] <richlowe> AmyMalik: is a working gdb, or is something else wrong too?
[22:40:08] <AmyMalik> JUST got to the CXXLD gdb phase, haven't tried it on the file i need to debug yet
[22:40:18] <richlowe> oh, yuck.
[22:40:34] <richlowe> How advanced a gdb user are you?
[22:41:09] <richlowe> Everything I remember how to do seemed to largely work, when I tried it. But that's really not very much.
[22:41:35] <richlowe> I think stopping on syscalls or signals is the thing I remember that did not (it's just not supported by the target).
[22:41:51] <richlowe> maybe just syscalls, it stopped on sigsegv obviously.
[22:42:51] <AmyMalik> richlowe: I'm just trying to post-mortem a program
[22:43:07] <AmyMalik> a program that seems to keep turning things into nul pointers
[22:43:21] <richlowe> anyway, I'm interested in reports of what doesn't work. People have been complaining for years, but not really able to explain why.
[22:44:44] <AmyMalik> richlowe: are you a representative of gnu binutils? :-)
[22:44:50] *** andy_js <andy_js!~andy@94.12.192.123> has quit IRC (Quit: andy_js)
[22:45:08] <richlowe> No, illumos for the moment at least.
[22:46:14] <richlowe> but I've been working on something gdb related in there.
[22:46:24] <AmyMalik> right let's actually try this mofo now that I have a compiled 'gdb' in $PATH
[22:46:45] <AmyMalik> warning: Unexpected size of section `.reg2/1' in core file.
[22:46:47] <AmyMalik> for what it's worth
[22:47:22] <AmyMalik> i seem to be mostly in business
[22:47:27] <AmyMalik> now I just need to recompile the app -O0
[22:48:17] <richlowe> the .reg2/1 warnings are allegedly normal.
[22:48:26] <AmyMalik> allegedly.
[22:48:29] <richlowe> Oracle "fixed" them by just removing the code that prints them
[22:48:41] <AmyMalik> SHAME, Oracle, SHAME
[22:48:53] <richlowe> they state it's something to do with the structure size being variable, and gdb not really being capable of coping.
[22:49:04] <richlowe> which given some of the code involved, could be true?
[22:52:43] <AmyMalik> right, I'm going to confront the sole surviving developer of this application (which is based on eggdrop 1.3.28...) with the backtrace
[23:06:17] *** jeffpc <jeffpc!~jeffpc@josefsipek.net> has quit IRC (Ping timeout: 245 seconds)
[23:07:04] *** jeffpc <jeffpc!~jeffpc@josefsipek.net> has joined #illumos
[23:08:50] *** XV8 <XV8!~XV8@2601:151:4403:9c0:49f0:b582:4cac:6f84> has joined #illumos
[23:29:43] *** despair86 <despair86!~despair@exit.i2p.rocks> has quit IRC (Remote host closed the connection)
[23:30:00] *** despair86 <despair86!~despair@exit.i2p.rocks> has joined #illumos
[23:40:04] *** merzo <merzo!~merzo@22-2-132-95.pool.ukrtel.net> has quit IRC (Read error: Connection reset by peer)
[23:40:14] *** merzo <merzo!~merzo@22-2-132-95.pool.ukrtel.net> has joined #illumos
top

   September 15, 2019  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | >