[00:05] RichieSams: Core 2 Duo E6750 doesn't sound like a toaster to me [00:05] Unrelated tp ScummVM, but we had a user complaining that saving data from our application was very slow. [00:06] After weeks of investigation it turns out it was his antvirus [00:06] because he was saving in a directory that was automatically scanned [00:07] ahh, that'll do it [00:07] LordHoto: i7 bleeding edge or go home! Ha ha ha. kidding [00:08] We'll see when he uploads a save. I think criezy may be right about built up state [00:09] since we never really reset the VM [00:10] RichieSams: Trust me, I have a Skylake i7... it's a PITA :-P [00:10] Because saves directly serialize/deserialize the lua state [00:10] RichieSams: he wrote "Yes, it still happens and right from the start of the game." [00:10] grrr [00:11] then maybe it's your AV theory [00:11] or some other file watcher program [00:11] that's a bit of a stretch, but yeah that was just to say not to discount an external factor [00:12] maybe if he's using all his memory. ie. it's being paged in and out of swap? [00:12] that could slow things down [00:13] or if he's using valgrind :P [00:13] also valid. :P [00:13] did that finally finish? [00:13] Yes it did. [00:14] But the use of uninitialized memory didn't speak much to me. [00:14] All in Common::GZipWriteStream::write(void const*, unsigned int) [00:15] Called from Sword25::PersistenceService::saveGame(unsigned int, Common::String const&) [00:15] criezy: yeah, disable compressed savefiles [00:15] criezy: otherwise it might not give you the exact write operation which causes that [00:15] brb driving home [00:15] RichieSams (628e303e@gateway/web/freenode/ip.98.142.48.62) left irc: Quit: Page closed [00:16] criezy: (you can pass "false" as additional parameter to openSaveForWriting) [00:16] criezy: *openForSaving [00:16] You answered my question before I even asked it :) [00:17] criezy: well, I had m_kiewitz ask me about a similar thing during his AGI work :-P [00:17] criezy: and this looks all too familiar :-P [00:23] Cheeseness (~cheesenes@ppp118-208-202-83.lns20.hba1.internode.on.net) joined #scummvm. [00:32] t0by (~t0by@host234-243-dynamic.54-82-r.retail.telecomitalia.it) left irc: Remote host closed the connection [00:39] mgerhardy_ (~mgerhardy@46.189.78.184) joined #scummvm. [00:41] mgerhardy (~mgerhardy@46.189.78.184) left irc: Ping timeout: 240 seconds [00:41] snover (~Adium@unaffiliated/snover) left irc: Quit: Leaving. [00:45] RichieSams (~RichieSam@host-66-235-81-67.warpspeed1.net) joined #scummvm. [00:46] erdic (~erdic@unaffiliated/motley) left irc: Ping timeout: 240 seconds [00:46] erdic (~erdic@unaffiliated/motley) joined #scummvm. [00:47] That's much better. [00:47] SylvainTV (~SylvainTV@ALille-653-1-470-201.w92-142.abo.wanadoo.fr) left irc: Read error: Connection reset by peer [00:47] Just wondering, are games like Eternam and Willy Beamish on the to-do list to run on ScummVM? [00:47] I don't have time to look at it now but I will record it in a ne w bug. [00:48] Eternam I have no idea. [00:48] There was some work on Willy Beamish I believe. [00:49] Ah right: http://wiki.scummvm.org/index.php/DGDS [00:49] ah interesting thx [01:14] sirlemonhead (bduncan22@86.45.247.211) left #scummvm. [01:38] Strangerke_ (~Strangerk@cable-85.28.84.13.coditel.net) joined #scummvm. [01:40] Strangerke (~Strangerk@cable-85.28.84.13.coditel.net) left irc: Ping timeout: 240 seconds [01:40] Nick change: Strangerke_ -> Strangerke [01:43] criezy (~criezy@host86-139-77-100.range86-139.btcentralplus.com) left irc: Quit: criezy [02:10] m_kiewitz (~m_kiewitz@x5f7109a0.dyn.telefonica.de) left irc: Quit: technology isn't intrinsically good or evil. It's how it's used. Like the Death Ray. [02:13] frankyboy_ (~franky@ppp109-252-17-141.pppoe.spdop.ru) left irc: Remote host closed the connection [02:26] Vampire0_ (~Vampire@jEdit/Vampire) joined #scummvm. [02:30] Vampire0 (~Vampire@jEdit/Vampire) left irc: Ping timeout: 276 seconds [02:38] snover (~Adium@unaffiliated/snover) joined #scummvm. [02:38] #scummvm: mode change '+o snover' by ChanServ!ChanServ@services. [03:16] Schnabeltierchen (~Schnabelt@xdsl-213-196-229-72.netcologne.de) left irc: Ping timeout: 245 seconds [03:17] Schnabeltierchen (~Schnabelt@xdsl-78-34-207-146.netcologne.de) joined #scummvm. [03:28] LordHoto (~loom@unaffiliated/lordhoto) left irc: Ping timeout: 240 seconds [03:30] LordHoto (~loom@unaffiliated/lordhoto) joined #scummvm. [03:30] #scummvm: mode change '+o LordHoto' by ChanServ!ChanServ@services. [04:11] dreammaster (~dreammast@c-73-167-118-204.hsd1.ma.comcast.net) left irc: [04:21] jammm (~jam@106.221.136.135) joined #scummvm. [04:21] jammmm (~jam@106.221.157.90) left irc: Read error: Connection reset by peer [04:28] Port build status changed with 331a9f49: Failure: master-amigaos4 [05:02] Poly-C (~Poly-C@gentoo/developer/Polynomial-C) joined #scummvm. [05:05] Polynomial-C (~Poly-C@gentoo/developer/Polynomial-C) left irc: Ping timeout: 272 seconds [05:43] RichieSams (~RichieSam@host-66-235-81-67.warpspeed1.net) left irc: Quit: Leaving [06:04] LordHoto (~loom@unaffiliated/lordhoto) left irc: Quit: ... [06:18] Lightkey (~jonas@p200300764F5A250722CF30FFFE083718.dip0.t-ipconnect.de) left irc: Ping timeout: 240 seconds [06:22] I gave up on the old linux laptop with the mysterious Maemo toolchain on it and decided to start anew on a Linux VM (that way I don't have to depend on a specific machine - we can just image it) [06:23] Took this opportunity to update the wiki with more precise info on how to build for maemo in 2016 (when Debian has archives all the needed files) [06:23] also bought a battery and charger for my maemo device .. and tested my new toolchain [06:23] I finally got it to work but: [06:25] we have a regression with the keymapper that happened sometime after v1.7.0 .. the latest 1.8.0git basically isn't behaving properly with the keymapper .. unsurprising because the keymapper code had 4 years of bitrot [06:27] I will bisect first and then see what we need to do. I just need it working for the 1.8.0 release and then I can give it the proper attention afterwards. It is a PITA to bisect because it only happens on the real device and isn't a nice compile time error (and the real device is slow as molasses) [06:28] (1.7.0 freshly built with the new toolchain has no bug so it is the "GOOD" and the latest git is the "BAD") [06:29] frankyboy_ (~franky@ppp109-252-17-141.pppoe.spdop.ru) joined #scummvm. [06:29] ny00123 (~ny00123@cbl217-132-129-134.bb.netvision.net.il) joined #scummvm. [06:31] Lightkey (~jonas@p200300764F5A250922CF30FFFE083718.dip0.t-ipconnect.de) joined #scummvm. [06:43] Harekiet (~harekiet@kittens.harekiet.com) left irc: Ping timeout: 245 seconds [06:59] 9b8b737 is what broke the keymapper - (probably because my config already has length 4 (REM/VIR/..) and now it is expecting length 5 (REMP/VIRT/...) .. this is sort of a backwards compatibility breakage but I am ok with it I guess [07:00] I actually have "REM" and "VIR" hardcoded in maemo.cpp *facepalm* [07:02] snover (~Adium@unaffiliated/snover) left irc: Quit: Leaving. [07:05] rootfather (~rootfathe@p200300422D781780E906E02F28F5A306.dip0.t-ipconnect.de) joined #scummvm. [07:15] GitHub170 (~GitHub170@192.30.252.41) joined #scummvm. [07:15] [scummvm] tsoliman pushed 1 new commit to master: https://git.io/vgnkW [07:15] scummvm/master f28ced2 Tarek Soliman: MAEMO: Fix default bindings... [07:15] GitHub170 (GitHub170@192.30.252.41) left #scummvm. [07:17] borosky (~broosky@088156132138.dynamic-ww-4.vectranet.pl) joined #scummvm. [07:17] broosky (~broosky@088156132138.dynamic-ww-4.vectranet.pl) left irc: Ping timeout: 256 seconds [07:19] frankyboy_ (~franky@ppp109-252-17-141.pppoe.spdop.ru) left irc: Ping timeout: 256 seconds [07:22] Port build status changed with f28ced2e: Success: master-amigaos4 [07:24] Harekiet (~harekiet@kittens.harekiet.com) joined #scummvm. [07:40] Harekiet (~harekiet@kittens.harekiet.com) left irc: Ping timeout: 250 seconds [07:47] Harekiet (~harekiet@kittens.harekiet.com) joined #scummvm. [08:13] jhns (~acidtest@5.9.234.86) left irc: Quit: leaving [08:14] jhns (~johannes@5.9.234.86) joined #scummvm. [08:15] jhns (johannes@5.9.234.86) left #scummvm. [08:18] LyTchet (~alexandre@host-95-182-242-222.dynamic.voo.be) joined #scummvm. [08:36] LordHoto (~loom@unaffiliated/lordhoto) joined #scummvm. [08:36] #scummvm: mode change '+o LordHoto' by ChanServ!ChanServ@services. [08:52] TMM (~hp@fsf/member/pdpc.professional.tmm) left irc: Quit: Ex-Chat [08:54] waltervn (~waltervn@82-171-142-149.ip.telfort.nl) joined #scummvm. [08:54] #scummvm: mode change '+o waltervn' by ChanServ!ChanServ@services. [08:55] morning [09:00] Henke37 (~Henrik@2001:2002:c2ec:d1b0:a51c:2f8c:a598:fa68) joined #scummvm. [09:03] t0by (~t0by@host208-20-dynamic.5-87-r.retail.telecomitalia.it) joined #scummvm. [09:03] #scummvm: mode change '+v t0by' by ChanServ!ChanServ@services. [09:09] ajax16384 (~User@109.60.138.138) joined #scummvm. [09:09] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services. [09:11] Strangerke_ (~Strangerk@cable-85.28.84.13.coditel.net) joined #scummvm. [09:14] Strangerke (~Strangerk@cable-85.28.84.13.coditel.net) left irc: Ping timeout: 240 seconds [09:14] Nick change: Strangerke_ -> Strangerke [09:18] t0by (~t0by@host208-20-dynamic.5-87-r.retail.telecomitalia.it) left irc: Ping timeout: 250 seconds [09:22] m_kiewitz (~m_kiewitz@x5f710502.dyn.telefonica.de) joined #scummvm. [09:22] #scummvm: mode change '+o m_kiewitz' by ChanServ!ChanServ@services. [09:40] t0by (~t0by@host208-20-dynamic.5-87-r.retail.telecomitalia.it) joined #scummvm. [09:40] #scummvm: mode change '+v t0by' by ChanServ!ChanServ@services. [09:41] TMM (~hp@fsf/member/pdpc.professional.tmm) joined #scummvm. [09:53] t0by (~t0by@host208-20-dynamic.5-87-r.retail.telecomitalia.it) left irc: Remote host closed the connection [09:54] t0by (~t0by@host208-20-dynamic.5-87-r.retail.telecomitalia.it) joined #scummvm. [09:54] #scummvm: mode change '+v t0by' by ChanServ!ChanServ@services. [10:12] frankyboy_ (~franky@ppp109-252-17-141.pppoe.spdop.ru) joined #scummvm. [10:27] _sev|work (~sev@proxy-gw-a.booking.com) joined #scummvm. [10:27] _sev|work (~sev@proxy-gw-a.booking.com) left irc: Changing host [10:27] _sev|work (~sev@scummvm/undead/sev) joined #scummvm. [10:27] #scummvm: mode change '+o _sev|work' by ChanServ!ChanServ@services. [11:09] _sev|work_ (~sev@proxy-gw-a.booking.com) joined #scummvm. [11:09] _sev|work_ (~sev@proxy-gw-a.booking.com) left irc: Changing host [11:09] _sev|work_ (~sev@scummvm/undead/sev) joined #scummvm. [11:09] #scummvm: mode change '+o _sev|work_' by ChanServ!ChanServ@services. [11:11] LoneStarr (LoneStarr@ip25053d66.dynamic.kabel-deutschland.de) joined #scummvm. [11:12] ChrisKeys (LoneStarr@ip25053d66.dynamic.kabel-deutschland.de) left irc: Ping timeout: 264 seconds [11:12] Paaris (paaris@soju.fi) left irc: Ping timeout: 264 seconds [11:12] _sev|work (~sev@scummvm/undead/sev) left irc: Ping timeout: 264 seconds [11:13] ced117 (~ced117@opensuse/member/ced117) left irc: Ping timeout: 264 seconds [11:13] heroux (sandroco@gateway/shell/insomnia247/x-gvqdcjqdqkubuhzz) left irc: Ping timeout: 264 seconds [11:14] heroux (sandroco@gateway/shell/insomnia247/x-spdbxfihwrxmatet) joined #scummvm. [11:15] LyTchet (~alexandre@host-95-182-242-222.dynamic.voo.be) left irc: [11:26] ced117 (~ced117@opensuse/member/ced117) joined #scummvm. [11:28] quiet in here this morning... [11:29] frankyboy_ (~franky@ppp109-252-17-141.pppoe.spdop.ru) left irc: Ping timeout: 240 seconds [11:30] Action: t0by screams [11:30] Action: t0by rattles pots [11:36] Action: _sev|work_ mumbles [11:36] Nick change: _sev|work_ -> _sev|work [11:39] Nick change: LoneStarr -> ChrisKeys [11:41] GitHub107 (~GitHub107@192.30.252.46) joined #scummvm. [11:41] [scummvm] lordhoto pushed 1 new commit to master: https://git.io/vgnDd [11:41] scummvm/master 29b37f4 Johannes Schickel: COMMON: Add render mode for Macintosh.... [11:41] GitHub107 (GitHub107@192.30.252.46) left #scummvm. [11:49] Paaris (paaris@soju.fi) joined #scummvm. [12:21] jammm (~jam@106.221.136.135) left irc: Read error: Connection reset by peer [12:21] jammmm (~jam@106.221.136.195) joined #scummvm. [12:54] LordHoto (~loom@unaffiliated/lordhoto) left irc: Quit: brb [12:56] LordHoto (~loom@unaffiliated/lordhoto) joined #scummvm. [12:56] #scummvm: mode change '+o LordHoto' by ChanServ!ChanServ@services. [13:06] borosky (~broosky@088156132138.dynamic-ww-4.vectranet.pl) left irc: Ping timeout: 240 seconds [13:14] GitHub96 (~GitHub96@192.30.252.34) joined #scummvm. [13:14] [scummvm] m-kiewitz pushed 1 new commit to master: https://git.io/vgnpo [13:14] scummvm/master efb6532 Martin Kiewitz: AGI: Rewrote stringWordWrap()... [13:14] GitHub96 (GitHub96@192.30.252.34) left #scummvm. [13:24] GitHub81 (~GitHub81@192.30.252.45) joined #scummvm. [13:24] [scummvm] m-kiewitz pushed 1 new commit to master: https://git.io/vgnjC [13:24] scummvm/master f954603 Martin Kiewitz: AGI: GfxMgr::drawBox now using kRenderMacintosh [13:24] GitHub81 (GitHub81@192.30.252.45) left #scummvm. [13:27] broosky (~broosky@088156132138.dynamic-ww-4.vectranet.pl) joined #scummvm. [14:15] frankyboy_ (~franky@ppp109-252-17-141.pppoe.spdop.ru) joined #scummvm. [14:33] waltervn (~waltervn@82-171-142-149.ip.telfort.nl) left irc: Ping timeout: 248 seconds [14:42] <_sev|work> rootfather: may I help somehow? [14:44] jammm (~jam@117.195.114.32) joined #scummvm. [14:44] jammmm (~jam@106.221.136.195) left irc: Read error: Connection reset by peer [14:51] jamm (~jam@unaffiliated/jamm) joined #scummvm. [14:54] jammm (~jam@117.195.114.32) left irc: Ping timeout: 276 seconds [15:07] ajax16384 (~User@109.60.138.138) left irc: Read error: Connection reset by peer [15:11] hi _sev|work, thanks for asking, but my question is quite specific to the German translation :) [15:12] btw, what's the deadline for changes to the translation of the NEWS and README files for 1.8.0? [15:14] heroux (sandroco@gateway/shell/insomnia247/x-spdbxfihwrxmatet) left irc: Ping timeout: 248 seconds [15:27] mgerhardy_ (~mgerhardy@46.189.78.184) left irc: Quit: Ex-Chat [15:32] RichieSams (628e303e@gateway/web/freenode/ip.98.142.48.62) joined #scummvm. [15:34] RichieSams (628e303e@gateway/web/freenode/ip.98.142.48.62) left irc: Client Quit [15:36] RichieSams (628e303e@gateway/web/freenode/ip.98.142.48.62) joined #scummvm. [15:37] <_sev|work> rootfather: are you subscribed to scummvm-devel? [15:41] GitHub190 (~GitHub190@192.30.252.46) joined #scummvm. [15:41] [scummvm] m-kiewitz pushed 1 new commit to master: https://git.io/vgcaT [15:41] scummvm/master f09a15f Martin Kiewitz: AGI: Render mode Macintosh added... [15:41] GitHub190 (GitHub190@192.30.252.46) left #scummvm. [15:46] WooShell (woo@ip1886aec7.dynamic.kabel-deutschland.de) joined #scummvm. [15:51] meow =^.^= [15:52] Nick change: RichieSams -> RichieSams|work [16:00] Port build status changed with f09a15f2: Failure: master-mingw-w32 [16:00] nooo [16:04] popey_ (~popey@ubuntu/member/popey) joined #scummvm. [16:04] GitHub39 (~GitHub39@192.30.252.42) joined #scummvm. [16:04] [scummvm] m-kiewitz pushed 1 new commit to master: https://git.io/vgc6n [16:04] scummvm/master a7ffda3 Martin Kiewitz: AGI: fix build powf() -> pow() [16:04] GitHub39 (GitHub39@192.30.252.42) left #scummvm. [16:12] borosky (~broosky@088156132138.dynamic-ww-4.vectranet.pl) joined #scummvm. [16:12] Vampire0 (~Vampire@jEdit/Vampire) joined #scummvm. [16:15] broosky (~broosky@088156132138.dynamic-ww-4.vectranet.pl) got netsplit. [16:15] Vampire0_ (~Vampire@jEdit/Vampire) got netsplit. [16:15] popey (~popey@ubuntu/member/popey) got netsplit. [16:15] |Cable| (~cable@238.147.165.83.dynamic.reverse-mundo-r.com) got netsplit. [16:15] waltervn (~waltervn@82-171-142-149.ip.telfort.nl) joined #scummvm. [16:15] #scummvm: mode change '+o waltervn' by ChanServ!ChanServ@services. [16:20] TMM (~hp@fsf/member/pdpc.professional.tmm) left irc: Quit: Ex-Chat [16:20] |Cable| (~cable@238.147.165.83.dynamic.reverse-mundo-r.com) returned to #scummvm. [16:24] Port build status changed with a7ffda3b: Success: master-mingw-w32 [16:26] popey (~popey@ubuntu/member/popey) got lost in the net-split. [16:26] Vampire0_ (~Vampire@jEdit/Vampire) got lost in the net-split. [16:26] broosky (~broosky@088156132138.dynamic-ww-4.vectranet.pl) got lost in the net-split. [16:27] syke (~Adium@142-254-29-170.dsl.dynamic.fusionbroadband.com) joined #scummvm. [16:31] Nick change: popey_ -> popey [16:39] I suppose mort.dat and access.dat should be packaged? [16:41] demonimin (~demonimin@unaffiliated/demonimin) left irc: Remote host closed the connection [16:46] wjp: I suppose so. This is probably a good point to look at dists/scummvm.rc.in again to assure that all data files are listed there too. [16:47] ajax16384 (~User@ip33.net130.n37.ru) joined #scummvm. [16:47] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services. [16:53] GitHub178 (~GitHub178@192.30.252.34) joined #scummvm. [16:53] [scummvm] wjp pushed 1 new commit to master: https://git.io/vgc7K [16:53] scummvm/master bdec938 Willem Jan Palenstijn: DISTS/REDHAT: Package mort.dat, access.dat [16:53] GitHub178 (GitHub178@192.30.252.34) left #scummvm. [16:54] syke (~Adium@142-254-29-170.dsl.dynamic.fusionbroadband.com) left irc: Quit: Leaving. [16:54] wysiwtf (arvenius@whatyouseeis.wtf) left irc: Ping timeout: 240 seconds [17:00] snover (~Adium@unaffiliated/snover) joined #scummvm. [17:00] #scummvm: mode change '+o snover' by ChanServ!ChanServ@services. [17:01] Strangerke (~Strangerk@cable-85.28.84.13.coditel.net) left irc: Ping timeout: 248 seconds [17:05] jamm (~jam@unaffiliated/jamm) left irc: Ping timeout: 252 seconds [17:05] jammm (~jam@106.221.153.206) joined #scummvm. [17:14] wysiwtf (arvenius@whatyouseeis.wtf) joined #scummvm. [17:24] geep (~JoeBelow@216.185.241.19) joined #scummvm. [17:27] geep (~JoeBelow@216.185.241.19) left irc: Client Quit [17:31] Littleboy (~littleboy@c-50-133-137-235.hsd1.ma.comcast.net) joined #scummvm. [17:31] #scummvm: mode change '+o Littleboy' by ChanServ!ChanServ@services. [17:41] GitHub85 (~GitHub85@192.30.252.42) joined #scummvm. [17:41] [scummvm] m-kiewitz pushed 1 new commit to master: https://git.io/vgcjb [17:41] scummvm/master c61c89d Martin Kiewitz: SCI: QfG1/2/3/4 character file detection changed... [17:41] GitHub85 (GitHub85@192.30.252.42) left #scummvm. [17:47] _sev|work (~sev@scummvm/undead/sev) left irc: Ping timeout: 272 seconds [17:48] geep (~JoeBelow@216.185.241.19) joined #scummvm. [17:52] geep (~JoeBelow@216.185.241.19) left irc: Client Quit [17:54] for god's sake sourceforge [17:54] The request could not be satisfied. [17:54] This distribution is not configured to allow the HTTP request method that was used for this request. The distribution supports only cachable requests. [17:54] The sourceforge.net website is temporarily in static offline mode. [17:54] Only a very limited set of project pages are available until the main website returns to service. [17:55] waltervn (~waltervn@82-171-142-149.ip.telfort.nl) left irc: Ping timeout: 240 seconds [17:57] waltervn (~waltervn@82-171-142-149.ip.telfort.nl) joined #scummvm. [17:57] #scummvm: mode change '+o waltervn' by ChanServ!ChanServ@services. [17:58] my daughter keeps pushing the power button, instead of escape, when she wants to skip dialogue :P [17:58] (in Putt-Putt) [18:00] syke (~Adium@142-254-29-170.dsl.dynamic.fusionbroadband.com) joined #scummvm. [18:13] sirlemonhead (~bduncan22@86-45-247-211-dynamic.agg2.ddm.bbh-prp.eircom.net) joined #scummvm. [18:30] can someone explain to me how VectorRendererSpec is seemingly successfully defining out-of-class template methods in a cpp file? [18:30] [md5] (~md5@unaffiliated/md5/x-729473) joined #scummvm. [18:30] #scummvm: mode change '+o [md5]' by ChanServ!ChanServ@services. [18:33] snover: which part is strange? [18:34] if you mean the instantiation, there is only one template argument with only 2 options (uint16, uint32), and those are instantiated by createRenderer() in the same file [18:35] ah, i see. ok. that makes more sense. [18:35] but if you were to use VectorRendererSpec with a different PixelType in a different file, it would lead to missing definitions [18:44] GitHub154 (~GitHub154@192.30.252.41) joined #scummvm. [18:44] [scummvm] m-kiewitz pushed 1 new commit to master: https://git.io/vgCYR [18:44] scummvm/master 04de727 Martin Kiewitz: SCI: Blocking ScummVM auto-save slot 0 for saving... [18:44] GitHub154 (GitHub154@192.30.252.41) left #scummvm. [18:47] <_sev> tonight is branching [18:49] m_kiewitz: did you test mother goose? It also has a custom savegame script [18:50] yes [18:50] that game doesn't use slot 0 right? IIRC [18:50] well i actually figured, that loading saved games from GMM directly results in some graphics being messed up in mother goose [18:50] jones uses slot 0 because of hard-coding [18:51] I mean MG [18:51] well mother goose basically says the first time "save this game" and that uses the first slot (we control that) [18:51] I seem to remember that it *not* using slot 0 was a problem at one pointer, because we renumbered everything to start at 0 [18:51] *point [18:51] we use virtual slot ids for 5 or 6 years now, because of all sorts of weird scripts [18:52] so when a game calls kSaveGame on slot 0 or 1 or whatever, we assume that the game wants to create a new save [18:52] that is not affected by my commit [18:53] rootfather (~rootfathe@p200300422D781780E906E02F28F5A306.dip0.t-ipconnect.de) left irc: [18:53] wjp: do you have any time to script collapse other selector accesses? there are a bunch in e.g. ctor_Plane and I would love to be able to quickly go through this constructor instead of&slowly :) [18:53] waltervn: please check jones if possible. it should still work [18:54] i wonder if a commit broke mixed up mother goose loading [18:54] well mixed up mother goose SCI1 [18:57] argh and mixed up mother goose SCI1.1 crashes in kDeviceInfo [18:57] within case K_DEVICE_INFO_GET_DEVICE: { [18:58] wjp: can you do a bisect on mother goose SCI1? [18:59] maybe its port related. back in 2010 it definitely worked properly, at least as far as i remember [19:00] mixed up mother goose sci0 works properly [19:03] Small dumb in-between question about SCI games (i.e. LSL1 or QFG1) on ScummVM Android. How would I bring up the virtual keyboard for commands? [19:08] lskovlun (~lars@194.19.137.125) joined #scummvm. [19:09] in mixed up mother goose SCI1.1 SegManager::getString() is called on raw data and I see only a valid "a" there, without any terminating NUL. Which then results in a crash in Common::String because it runs a strlen() on it [19:11] MUMG SCI1.1 worked in ScummVM 1.7.0 [19:17] MUGM SCI1 restoring also worked properly in 1.7.0, but only for restoring within the game. when i restore using GMM it seems to have never worked properly [19:18] well the latter works, but parts of the background are not drawn properly [19:35] jammm (~jam@106.221.153.206) left irc: Read error: Connection reset by peer [19:35] jamm (~jam@unaffiliated/jamm) joined #scummvm. [19:40] syke (~Adium@142-254-29-170.dsl.dynamic.fusionbroadband.com) left irc: Quit: Leaving. [19:47] SylvainTV (~SylvainTV@ALille-653-1-470-201.w92-142.abo.wanadoo.fr) joined #scummvm. [19:47] #scummvm: mode change '+o SylvainTV' by ChanServ!ChanServ@services. [19:54] m_kiewitz: what exactly do you want me to test in what? [19:55] wjp: mixed up mother goose SCI1 -> start game -> watch intro -> save -> restore save (not via GMM!) -> graphical issue [19:55] mixed up mother goose SCI1.1 -> start game -> watch intro -> try to save -> uninitialized memory access in kDeviceInfo [19:57] both of those didn't occur in 1.7.0, well I'm not 100% sure about the kDeviceInfo issue. Apple IIgs in AGI also didn't crash as much as my development build, but it still crashes [19:57] crashed [19:57] not seeing a crash in SCI1.1 [19:58] set a breakpoint on kDeviceInfo [19:58] sci debugger or an outside debugger? [19:58] game passes a raw reference to a 1 byte string [19:58] outside debugger [19:59] maybe it works for you by accident. valgrind should definitely complain at least [19:59] the 1 byte string is then returned (it's an "a" for me) and that crashes inside Common::String, because it tries to check the length of the string [20:01] it's doing a "K_DEVICE_INFO_GET_DEVICE" (sub-opcode 0) [20:03] did SegManager always use Common::Strings? for getString()? [20:04] looking at kq5 asm, that only ever looks at 2 bytes of that string [20:05] I apparently turned it from char* to Common::String in 2009 [20:06] well in that case it's really only 1 valid byte [20:06] in kq5 it has two different behaviours, depending on if path[1] == ':' or not [20:06] all the ones after that are the typical filler bytes made by AppVerifier, that i use [20:07] so the issue was there all the time? [20:07] can you check 1.3.1 or something and what's exactly at the location of the pointer? [20:08] i guess we really should improve data security checks, like checking for missing NULs and so on [20:08] although i even wonder where it actually points at. a, maybe A:? [20:09] my work in progress saved game code is doing parts of that well it figures out certain pointers and saves them in a generic way so that straight offsets are not really needed [20:10] it's basically turning a direct pointer into a segment into "this points to 3rd string in segment X, string offset X" [20:10] that way you can apply or remove most script patches and saved games will still work most of the time [20:11] girafe (~girafe@AGrenoble-651-1-477-112.w90-42.abo.wanadoo.fr) joined #scummvm. [20:13] and i also don't know what to do about that fan game issue. The game calls kSaveGame with slot 999 and we currently assume that it's some serious script issue and error() out [20:13] which fan game does that? [20:13] [vEX] (~vex@h95n22-spaa-a12.ias.bredband.telia.com) left irc: Ping timeout: 248 seconds [20:13] i don't like adding 999 as another "acceptable" slot for new slots [20:14] lskovlun: https://sourceforge.net/p/scummvm/bugs/7007/ [20:14] "Cascade Quest" [20:14] when we are calling with slot 0-99, we assume that the scripts want to create a new slot and do exactly that [20:15] and then we have the virtual slots, 100-199, which are our internal actual slots [20:15] so kq5's GetDevice basically returns the first two chars + nul if path[1] == ':', and otherwise it returns the first two chars + nul of GetCwd() [20:15] i had to change that from 1000-1999 to 100-199 because of mixed up mother goose, which requires a script patch [20:15] [vEX] (~vex@h95n22-spaa-a12.ias.bredband.telia.com) joined #scummvm. [20:15] wjp: so it tries to figure out the current drive letter and returns it?! [20:16] anyway i think i will probably add a script patch for that game, because i don't really like allowing random slot numbers. [20:16] if it would ask for slot 99, it would have worked already [20:17] didn't SCI have some internal limit on slots anyway? like 20 or 30 or so? [20:17] well although i guess saving to slot 999 all the time would create new saved games, which means at some point it would fail anyway [20:19] oh right and it would have worked as well before i lowered the virtual ids to the 100-199 range [20:26] t0by (~t0by@host208-20-dynamic.5-87-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds [20:27] criezy (~criezy@host86-139-77-100.range86-139.btcentralplus.com) joined #scummvm. [20:27] #scummvm: mode change '+o criezy' by ChanServ!ChanServ@services. [20:27] wjp: I believe we simply returned '/' in FreeSCI. It is platform specific anyway. [20:30] we do the same [20:30] but we do read the input string, only to then do nothing with it (except possible print it in a debug message) [20:33] #scummvm: mode change '+o [vEX]' by ChanServ!ChanServ@services. [20:35] maybe MUMG is doing something else first - i.e. relying on the return value of other things in DeviceInfo? [20:35] I don't own the game; I've never had the opportunity to look [20:36] it does call a few in a row [20:40] but the input doesn't come from there AFAICT [20:42] other than arg_0 cleaning stuff up on itself before copying members from the other object, is there any tell-tale sign to determine whether a subroutine is a copy constructor versus a copy assignment operator? default initialisations, maybe? [20:44] I'm fairly sure I mix those up all the time [20:44] the conclusion I came to is, does it matter? [20:45] heh. i suppose youre right. :) [20:46] snover: sorry, these other selector lookups are proving a bit tricky to automatically isolate so far [20:46] wjp: no worries, i figured they wouldnt be quite so simple [20:48] <[md5]> btw, regarding sciAudio: the games that use it don't use all of its functionality [20:48] if we assume they are the same in SCI3, I could tell you what they are [20:48] I have a nice little enumeration that works for all of SCI3 [20:48] (but apparently not earlier) [20:50] <[md5]> btw, lskovlun: there are some subops in SCI3 for ops 0x26 and 0x27 [20:51] <[md5]> there's a TODO for these [20:51] snover: I'll see if I can get the majority of them done, though [20:51] <[md5]> do these follow any pattern? i.e. the selectors of a specific object? [20:51] snover: hopefully the number of annoying exceptions is limited [20:52] wjp: thank you very much 🙏 [20:52] <[md5]> right now, we support 0x4c, 0x4d and 0x4e (with a possible TODO) [20:53] <[md5]> which are -info- (pop, push) and -super- (pop) [20:57] [md5] that is because they do not have proper selector numbers in SCI3 [20:57] <[md5]> hm? [20:57] <[md5]> what do you mean? [20:58] -info- for example does not have a selector number [20:58] so you can't retrieve it using the ordinary opcodes [20:58] <[md5]> so the scripts find it from that call? [20:58] instead, there is this opcode [20:58] <[md5]> right [20:58] similar for the others [20:59] <[md5]> and how do you know it's -info-? [21:00] because the system classes use it in an analogous way to older SCI? [21:00] and because the scripts in SCI3 basically work already [21:01] <[md5]> I think I added the superclass case [21:01] <[md5]> but it's been yeras [21:01] <[md5]> years * [21:01] <[md5]> hm [21:02] 0x4f pushes the superclass to stack (or at least the same value that 0x4e loads to acc) [21:02] it appears that nothing uses it, otherwise we would have seen by onw [21:03] <[md5]> yep [21:04] <[md5]> ah [21:04] <[md5]> SV supports these [21:05] <[md5]> it calls them _infoToa and _superToa [21:10] GitHub30 (~GitHub30@192.30.252.41) joined #scummvm. [21:10] [scummvm] m-kiewitz pushed 1 new commit to master: https://git.io/vgCKT [21:10] scummvm/master eaf0aeb Martin Kiewitz: AGI: Make it possible to also force CGA/EGA/VGA... [21:10] GitHub30 (GitHub30@192.30.252.41) left #scummvm. [21:12] ajax16384 (~User@ip33.net130.n37.ru) left irc: Read error: Connection reset by peer [21:13] syke (~Adium@142-254-29-170.dsl.dynamic.fusionbroadband.com) joined #scummvm. [21:13] I had to figure these out to get anything going at all; if you look at the algorithm in Object::initSelectorsSci3, that one came from a disassembler I wrote [21:13] there's nothing like it in the original interpreter [21:14] it just loads the CSC files into memory and operates on the memory directly [21:14] relying on the compiler to have eliminated any type errors [21:16] <[md5]> ouch [21:16] MUMG, Jones, and maybe one of the Doctor Brain games had some interesting savegame scripts [21:16] in AGI, I remember Manhunter 1 & 2 doing something unusual as well [21:16] <[md5]> lskovlun: I thought they would have learned at some point that operating on the memory directly can lead to all sorts of chaos... [21:17] syke (~Adium@142-254-29-170.dsl.dynamic.fusionbroadband.com) left irc: Client Quit [21:17] [md5] apparently not [21:18] <[md5]> :/ [21:18] curiously, the information is there to tell them apart [21:19] *properties and methods [21:19] <[md5]> "Oops, you did something we weren't expecting" should be more like "Oops, we messed up memory in ways we weren't expecting" [21:20] right, and that's essentially what the original interpreters said [21:20] <[md5]> yep :) [21:20] Action: lskovlun [md5] I managed to hide that message in the source for a product I built at IBM - in a comment, the users would never see it [21:21] Action: lskovlun can't figure out how to send private messages just now [21:22] lskovlun: msg [21:24] for some reason my fingers typed /me instead of /msg [21:30] dang fingers have a mind of their own! :) [21:32] aaaargh, it's calling FontMgr::BuildFBM with a huge set of selector lookups as arguments [21:32] EditText()? [21:32] CreateTextBitmap [21:32] ah [21:32] yes, that too [21:39] :-D [21:43] in order: result?, textView, textRect, text, fore, back, font, skip, borderColor, dimmed. (that's the short one) [21:44] textView is populated with view/loop/cel, textRect is populated with textLeft/textTop/etc [21:53] GitHub24 (~GitHub24@192.30.252.42) joined #scummvm. [21:53] [scummvm] bluegr pushed 2 new commits to master: https://git.io/vgCyc [21:53] scummvm/master 7ffc4f9 Filippos Karapetis: SCI: Silence false positive warning in MSVC [21:53] scummvm/master 88e2673 Filippos Karapetis: SCI: Clean up the op_infoToa and op_superToa SCI3 opcodes [21:53] GitHub24 (GitHub24@192.30.252.42) left #scummvm. [21:55] nice :) [21:56] ny00123 (~ny00123@cbl217-132-129-134.bb.netvision.net.il) left irc: Quit: Leaving [21:59] oh msvc. [21:59] you silly. [22:05] <[md5]> yeah, it does throw some warnings when variable assignment logic is not straightforward [22:11] t0by (~t0by@host208-20-dynamic.5-87-r.retail.telecomitalia.it) joined #scummvm. [22:11] #scummvm: mode change '+v t0by' by ChanServ!ChanServ@services. [22:14] GitHub3 (~GitHub3@192.30.252.42) joined #scummvm. [22:14] [scummvm] bluegr pushed 1 new commit to master: https://git.io/vgCQ1 [22:14] scummvm/master b843780 Filippos Karapetis: MADS: Change the engine description... [22:14] GitHub3 (GitHub3@192.30.252.42) left #scummvm. [22:14] "will support" [22:15] <[md5]> eventually [22:15] but is it appropriate as long as we only support that one? [22:15] <[md5]> well, there is tons of hardcoded logic for the other two games already [22:15] <[md5]> the original engine name was MADS anyway, not Rex [22:15] there should be some indication of what the underlying games are [22:16] <[md5]> there is, but not in the configure.engine file [22:16] snover: slowly going through them: http://www.usecode.org/scummvm/sq6/sq6_hide_getSelector.idc [22:16] <[md5]> wjp: impressive! :) [22:17] I wonder why there isn't some sort of more direct support for this kind of thing [22:17] wjp: excellent. did you find a reasonable signature for automation then? [22:17] <[md5]> what kind of thing? [22:18] "folding" of inlined code sequences [22:19] there's some research on it out there, but it's apparently quite hard [22:19] snover: all manual so far [22:19] snover: mainly in the hopes of starting to see patterns, which there kind of are [22:20] folding inlined code sequences is hard to do /reliably/ [22:20] snover: that file has 39 out of 147 of them [22:20] snover: oh, and maybe I should've said that I don't recommend _saving_ yet after applying that .idc, in case I screwed up... [22:20] (although they're of course very easy to undo by just deleting the hidden areas) [22:21] wjp: luckily folds are rea& yeah :) [22:21] but if you're not copying your idb occasionally then please do.. :p [22:21] come on guys live on the edge [22:23] there are selectors for pitch/roll/yaw in the enum that I have... maybe they were looking to write a flight sim in SCI32? [22:23] I'm quite glad they didn't [22:24] <[md5]> perhaps for games like Shivers 2? [22:24] wait, they didn't? [22:24] :) [22:24] <[md5]> it would be fun to search for items needed to land a plane [22:25] <[md5]> "bad guys want to shoot you down, and you have to find the key hidden inside the pilot's smelly socks to get out of this alive" [22:25] <[md5]> great game, that one [22:26] well it might have fit into e.g. RAMA? [22:27] <[md5]> perhaps, yeah [22:27] <[md5]> can't remember any smelly socks in that one :/ [22:27] <[md5]> :P [22:31] GitHub62 (~GitHub62@192.30.252.46) joined #scummvm. [22:31] [scummvm] sev- tagged desc/1.8.0pre at ba42d2a: https://git.io/vgCFi [22:31] GitHub62 (GitHub62@192.30.252.46) left #scummvm. [22:31] GitHub161 (~GitHub161@192.30.252.46) joined #scummvm. [22:31] [scummvm] sev- created branch-1-8 (+1 new commit): https://git.io/vgCFP [22:31] scummvm/branch-1-8 b7f1921 Eugene Sandulenko: RELEASE: This is 1.8.0pre [22:31] GitHub161 (GitHub161@192.30.252.46) left #scummvm. [22:33] GitHub182 (~GitHub182@192.30.252.46) joined #scummvm. [22:33] [scummvm] sev- pushed 1 new commit to master: https://git.io/vgCF5 [22:33] scummvm/master 0930018 Eugene Sandulenko: RELEASE: This is 1.9.0git [22:33] GitHub182 (GitHub182@192.30.252.46) left #scummvm. [22:33] already? [22:33] GitHub75 (~GitHub75@192.30.252.45) joined #scummvm. [22:33] [scummvm] sev- tagged desc/1.9.0git at 05ca788: https://git.io/vgCFd [22:33] GitHub75 (GitHub75@192.30.252.45) left #scummvm. [22:33] Topic changed on #scummvm by _sev!sev@scummvm/undead/sev: ScummVM 1.7.0 || 1.9.0 branched || http://www.scummvm.org/faq/ || No warez, no Game Requests, read the FAQ! || Channel logs: http://logs.scummvm.org [22:33] wasn't that scheduled for Sunday? [22:33] Topic changed on #scummvm by _sev!sev@scummvm/undead/sev: ScummVM 1.7.0 || 1.8.0 branched || http://www.scummvm.org/faq/ || No warez, no Game Requests, read the FAQ! || Channel logs: http://logs.scummvm.org [22:33] <_sev> oh [22:33] <_sev> right [22:33] <_sev> well [22:34] <_sev> too late :/ [22:34] <_sev> I mixed the tagging friday with this one [22:34] oh well, no matter [22:35] <_sev> yes, it's not tagging [22:38] <_sev> wjp: are you on Mac? [22:40] girafe (~girafe@AGrenoble-651-1-477-112.w90-42.abo.wanadoo.fr) left irc: Read error: Connection reset by peer [22:40] girafe (~girafe@AGrenoble-651-1-477-112.w90-42.abo.wanadoo.fr) joined #scummvm. [22:49] _sev: is the upload process different now that we're no longer hosting with SF? [22:49] <_sev> yes [22:49] <_sev> I did not figure it out yet [22:49] that's fine .. no rush [22:50] <_sev> so most probalbly I upload by myself [22:50] <_sev> or if somebody could find soome simple uploader with OAuth auntentication, so we could use github accounts... [22:54] There's always gist.github.com but I don't think it will like (large) binaries [22:54] GitHub108 (~GitHub108@192.30.252.41) joined #scummvm. [22:54] [scummvm] m-kiewitz pushed 1 new commit to master: https://git.io/vgCpn [22:54] scummvm/master 088c4d1 Martin Kiewitz: AGI: Adding Macintosh palettes for GR+SQ2... [22:54] GitHub108 (GitHub108@192.30.252.41) left #scummvm. [22:59] There seems to be a free 1GB of storage and 1GB of bandwidth for every user and org with Git LFS https://github.com/blog/1986-announcing-git-large-file-storage-lfs [23:02] I guess you can communicate with github's storage service using this API https://github.com/github/git-lfs/blob/master/docs/api/README.md [23:03] specifically the storage API section: https://github.com/github/git-lfs/blob/master/docs/api/README.md#storage-api [23:04] GitHub71 (~GitHub71@192.30.252.42) joined #scummvm. [23:04] [scummvm-tools] criezy pushed 1 new commit to master: https://git.io/vgCh7 [23:04] scummvm-tools/master 1984f6b Thierry Crozat: NEWS: Mention new tools for The Prince and the Coward and Tony Tough [23:04] GitHub71 (GitHub71@192.30.252.42) left #scummvm. [23:04] GitHub16 (~GitHub16@192.30.252.41) joined #scummvm. [23:04] [scummvm] m-kiewitz pushed 1 new commit to master: https://git.io/vgCh5 [23:04] scummvm/master bb9052b Martin Kiewitz: AGI: Switch to CLUT3 for Macintosh as default [23:04] GitHub16 (GitHub16@192.30.252.41) left #scummvm. [23:05] snover: ok, I did mess up two of these getSelector calls [23:06] if you have them applied, delete the hidden areas starting around 106B0 and 231DF [23:15] t0by (~t0by@host208-20-dynamic.5-87-r.retail.telecomitalia.it) left irc: Remote host closed the connection [23:17] Henke37 (~Henrik@2001:2002:c2ec:d1b0:a51c:2f8c:a598:fa68) left irc: Read error: Connection reset by peer [23:19] Strangerke (~Strangerk@cable-85.28.84.13.coditel.net) joined #scummvm. [23:19] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services. [23:19] hi guys [23:19] <[md5]> :O [23:19] <[md5]> hi Strangerke [23:21] What's up dude? :) [23:22] hi Strangerke [23:23] lskovlun: do you have any idea why setting some properties on objects sets bit 3 in the _info_ field? [23:25] is that inside AddScreenItem/AddPlane/friends? [23:25] as far as I can tell, all property setters [23:25] that sounds strange [23:26] (so this is interpreter code writing to script objects) [23:26] if it touches a property with index between 26 and 44, it ors _info_ with 8 [23:27] (26 and 44 inclusive) [23:28] dreammaster (~dreammast@c-73-167-118-204.hsd1.ma.comcast.net) joined #scummvm. [23:28] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services. [00:00] --- Sat Feb 6 2016