August 3, 2011 
< | 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 | 31 | >
August 3, 2011 [00:00:23] *** Ingenu has left #haiku[00:00:36] <negusnyul> vooshy: Only the standard 1024x768, 800x600 etc. options are there :([00:01:46] <vooshy> odd certainly there on my ideapad-s10 and my hp laptop[00:02:04] <negusnyul> this is a dell XPS[00:02:21] *** Barrett has quit IRC[00:02:31] <scgtrp> negusnyul: if you decide to try the mode list hacking way (again, i have no idea how real hardware will react to this), add this to the table in src/add-ons/accelerants/common/create_display_modes.cpp:[00:02:35] <scgtrp> {{85500, 1360, 1440, 1552, 1792, 768, 771, 777, 795, B_POSITIVE_VSYNC}, B_CMAP8, 1366, 768, 0, 0, MODE_FLAGS}, /* Vesa_Monitor_@60Hz_(1366X768) */[00:03:01] *** brrant has quit IRC[00:03:11] <negusnyul> scgtrp: nice, will try that tomorrow[00:03:13] * scgtrp disappears[00:04:05] * vooshy grep's scgtrp[00:04:40] * augiedoggie doesn't get that one[00:04:40] <negusnyul> also, this laptop has nvidia optimus... guess that's make things more complicated... (hopefully Haiku just ignores it)[00:04:57] <saivert> Isn't that sketchy on Linux even?[00:05:05] <saivert> switching between two GPUs like that[00:05:15] <augiedoggie> I think mmu_man has trouble with that on his thinkpad[00:06:16] <negusnyul> saivert: I'm fine with Linux, using only the intel graphics (haven't tried bumblebee yet), it has enough performance for everything except games[00:06:42] *** PasNox has quit IRC[00:07:07] *** lefthandman has joined #haiku[00:10:33] *** LordLobo has quit IRC[00:11:37] *** negusnyul has quit IRC[00:13:20] *** HaikuUser2 has quit IRC[00:16:16] *** thies has joined #haiku[00:16:53] <CIA-73> Haiku: kallisti5 * r42545 /haiku/trunk/headers/private/graphics/radeon_hd/r600_reg.h http://dev.haiku-os.org/changeset/42545 :[00:16:53] <CIA-73> * Register additions[00:16:53] <CIA-73> * No functional change[00:19:02] <CIA-73> Haiku: kallisti5 * r42546 /haiku/trunk/src/add-ons/kernel/drivers/graphics/radeon_hd/radeon_hd.cpp http://dev.haiku-os.org/changeset/42546 : * Rename PCI rom shared area.. isn't AtomBios until we verify it is[00:22:43] *** f1assistance has joined #haiku[00:22:45] *** lefthandman has quit IRC[00:22:51] <CIA-73> Haiku: czeidler * r42547 /haiku/trunk/src/servers/app/decorator/DefaultDecorator.cpp http://dev.haiku-os.org/changeset/42547 : Check if the dirty region is valid. Part of #7896.[00:23:48] *** oxoocoffee has joined #haiku[00:24:57] *** oxoocoffee has quit IRC[00:26:09] <CIA-73> HaikuPorts: scottmc * r1754 /haikuports/trunk/ (3 files in 3 dirs) http://ports.haiku-files.org/changeset/1754 : Update patches for apr and subversion to make use of AC_SEARCH_LIBS to detect need for -lnetwork.[00:26:40] *** oxoocoffee has joined #haiku[00:27:00] *** Cr0wb4r has joined #haiku[00:27:23] *** Barrett has joined #haiku[00:28:04] *** thies has quit IRC[00:28:45] *** vooshy has quit IRC[00:29:52] <CIA-73> Haiku: czeidler * r42548 /haiku/trunk/src/servers/app/ (5 files in 2 dirs) http://dev.haiku-os.org/changeset/42548 :[00:29:53] <CIA-73> If a window is hidden remove it from the S&T group. This happens when[00:29:53] <CIA-73> MediaPlayer goes fullscreen. Maybe not optimal but at least consistent with[00:29:53] <CIA-73> terminal which also left the S&T group in fullscreen mode. This is because the[00:29:53] <CIA-73> terminal has no decorator in fullscreen mode and thus can't be stacked any more[00:29:53] <CIA-73> (maybe this should be solved in the future...). Fixes #7895, #7896.[00:31:50] *** oxoocoffee has quit IRC[00:34:45] *** OmniMancer has joined #haiku[00:43:24] *** AlienSoldier has joined #haiku[00:45:37] <CIA-73> Haiku: czeidler * r42549 /haiku/trunk/src/servers/app/Window.cpp http://dev.haiku-os.org/changeset/42549 : Set the top most window look when switching between windows in a stack.[00:51:17] *** lorenzo5930 has quit IRC[00:58:57] *** OmniMancer has quit IRC[01:04:13] *** lorenzo5930 has joined #haiku[01:06:28] *** Barrett has quit IRC[01:07:50] *** DaaT has quit IRC[01:11:22] *** _arjen_ has quit IRC[01:11:27] *** OmniMancer has joined #haiku[01:13:56] <CIA-73> Haiku: kallisti5 * r42550 /haiku/trunk/src/add-ons/accelerants/radeon_hd/ (accelerant.cpp accelerant.h bios.cpp bios.h) http://dev.haiku-os.org/changeset/42550 :[01:13:56] <CIA-73> * refactor accelerant debugging[01:13:56] <CIA-73> * clone VGA rom shared area in accelerant[01:13:56] <CIA-73> * enable access, and make a copy of the VGA bios[01:13:56] <CIA-73> * give malloc'ed VGA bios pointer to AtomBIOS parser[01:13:57] <CIA-73> * Still invalid BIOS magic[01:13:58] <CIA-73> * TODO : Move atomBIOS pointer and reorganize some stuff[01:16:25] <CIA-73> Haiku: kallisti5 * r42551 /haiku/trunk/src/add-ons/kernel/drivers/graphics/radeon_hd/radeon_hd.cpp http://dev.haiku-os.org/changeset/42551 : * var typo fix in driver rom shared area creation[01:23:36] *** MAX2 has joined #haiku[01:27:39] *** Sikosis has joined #haiku[01:27:39] *** ChanServ sets mode: +o Sikosis[01:30:09] *** mmu_man has quit IRC[01:37:02] *** Pinaraf has quit IRC[01:39:40] *** luroh has quit IRC[01:49:28] *** dak_ has quit IRC[01:55:05] *** krckor has joined #haiku[02:04:26] *** oco2 has quit IRC[02:04:35] *** LiquidInsect has quit IRC[02:16:22] *** Cr0wb4r has quit IRC[02:23:07] *** Skipp_OSXI has quit IRC[02:24:35] *** krckor has quit IRC[02:24:40] *** jmelesky has quit IRC[02:26:56] *** noz_mob has quit IRC[02:27:17] *** noz_mob has joined #haiku[02:54:20] *** oxoocoffee has joined #haiku[03:09:41] *** saivert has quit IRC[03:10:00] *** saivert has joined #haiku[03:10:11] *** Teknomancer has quit IRC[03:10:26] *** Teknomancer has joined #haiku[03:12:00] *** Hery has quit IRC[03:13:08] *** marshan has quit IRC[03:14:16] *** HaikuUser has joined #haiku[03:14:21] *** HaikuUser is now known as BinkySMC2[03:15:00] *** marshan has joined #haiku[03:22:43] <BinkySMC2> hello all[03:23:24] <Duggan> hey[03:23:28] <Duggan> hey Skipp_OSX you around?[03:30:10] *** oxoocoffee has quit IRC[03:30:34] *** dreamed has joined #haiku[03:33:09] *** OmniMancer has quit IRC[03:37:53] *** oxoocoffee has joined #haiku[03:39:34] <CIA-73> Haiku: czeidler * r42552 /haiku/trunk/src/servers/app/ (10 files in 3 dirs) http://dev.haiku-os.org/changeset/42552 : Move flags and look into the tab too. The flags are needed to determine e.g. whether or not the zoom button should be drawn.[03:54:25] *** freakazoid0223 has quit IRC[04:00:21] *** volci has quit IRC[04:01:03] <Skipp_OSX> Duggan: I am[04:01:34] <Duggan> are you committing changes to the deskbar?[04:01:36] <BinkySMC2> hi skippy ![04:02:47] <Duggan> I updated to r42538 last night and the deskbar is usually everywhere but where it should be :P hehe[04:03:03] <BinkySMC2> deskbar is moving around ?[04:04:02] <Duggan> yep[04:04:12] <BinkySMC2> weirdness[04:05:46] *** OmniMancer has joined #haiku[04:06:50] <Skipp_OSX> Duggan: nope, not my fault, blame czeidler[04:06:59] <Duggan> lol ok[04:07:12] <Duggan> thanks :)[04:07:28] <Duggan> how goes the bug smashing btw?[04:07:30] <BinkySMC2> how did stack and tile mess up the deskbar ?[04:08:22] <Duggan> SAT is messed up above and beyond the deskbar[04:08:29] <Duggan> both are messed up in this particular build[04:08:57] <Skipp_OSX> Duggan: not good, the applications are expanded, resize window is being called. Everything is right, but the height is wrong.[04:09:11] <Duggan> the deskbar autoraises too often (anytime it's over it's rect, not just when another window isn't over it) and it moves down the side of the screen and sometimes resizes to a 1x1 (or so) rect[04:09:22] <Skipp_OSX> Duggan: The only thing I can think to do is to manually calculate the size instead of relying on the size of the expando->Frame()[04:09:26] <Duggan> Skipp_OSX well, that's better than it was :)[04:09:44] <Skipp_OSX> Duggan: I saw a bug report on that...[04:09:51] <Duggan> it'll be based on the expando->Frame() but you just need to figure out what to add to it[04:10:18] <Skipp_OSX> Duggan: well, Expando Frame just doesn't include the size of the window items[04:10:24] <Duggan> SAT doesn't like appropriately redrawing window tabs and some other issues with window rendering etc[04:10:40] <Duggan> Skipp_OSX the apps too, right?[04:10:59] <Duggan> there's 4 things that you have to worry about, the leaf menu, the tray, the apps, and the windows[04:11:04] <Duggan> right?[04:11:38] <Skipp_OSX> well, the apps and windows only really, the left and tray are separate[04:13:01] <Skipp_OSX> Duggan: see but it calculates the height be fExpando->Frame().bottom[04:13:40] <Duggan> it shouldn't be.... should be fExpando->Frame().bottom - fExpando->Frame().top[04:13:42] <Skipp_OSX> Duggan: This works unless the Deskbar crashes. When it does, fExpando->Frame().bottom only includes the heights of the apps and not the window items.[04:14:33] <Duggan> what event is this in again?[04:14:43] <BinkySMC2> is the work on SAT exposoing maybe some older bugs that just never got noticed ?[04:14:50] <Skipp_OSX> Duggan: I don't think so because I am pretty sure that fExpando->Frame().bottom gets you the frame of the entire window... which is what you want.[04:15:26] *** MAX3 has joined #haiku[04:15:39] *** MAX2 has quit IRC[04:16:24] <Skipp_OSX> Duggan: but... for some reason isn't including the window list in it after a crash... I have no idea why. It makes 0 sense to me[04:16:51] <Skipp_OSX> Duggan: do you want to see my patch to test?[04:17:09] <Duggan> no, it gets you the bottom coordinate of the window... if the top of the expando isn't the top of the screen, you don't get an accurate result by just using bottom()[04:17:23] <Duggan> I can take a look over it, sure[04:17:35] <Duggan> I might test, I'll have to see[04:25:25] <Skipp_OSX> Duggan: http://pastebin.com/uufitnhE[04:25:45] <Duggan> thanks, taking a look...[04:32:53] <Skipp_OSX> Duggan: let me know if you see anything really agregious[04:38:19] <Duggan> looking hehe... had to step out for a sec but I'm back... going to have to take a look at the originals and compare to get an idea[04:40:02] <Skipp_OSX> Duggan: originals: http://haiku.it.su.se:8180/source/search?q=&defs=&refs=&path=src%2Fapps%2Fdeskbar&hist=[04:41:03] <Duggan> hehe thanks[04:41:58] <Skipp_OSX> Duggan: testing this is kind of a biatch... you have to check the Show Application Expander and Expand New Teams flags in the Deskbar settings. Then reboot to save. When it comes back up load a terminal, kill Deskbar and restart it.[04:42:36] <Duggan> that sucks[04:42:42] <Skipp_OSX> The biatch is that you have to restart. If you don't then it won't save your settings and killing Deskbar will wipe them out.[04:42:53] <augiedoggie> why don't you just ask deskbar to quit?[04:43:03] <Skipp_OSX> augiedoggie: ... how do you do that?[04:43:10] <augiedoggie> hey Deskbar QUIT[04:43:43] <Skipp_OSX> augiedoggie: worth a shot...[04:44:10] <Skipp_OSX> augiedoggie: can I just say, `hey Deskbar SaveSettings` ???[04:44:18] <scgtrp> wait, is that an actual command?[04:44:21] <augiedoggie> not like that[04:44:24] * scgtrp is amused[04:44:52] <augiedoggie> you might be able to dig through the sources and find a message to send that will work[04:45:38] <augiedoggie> processcontroller may be able to send a quit request to deskbar too[04:45:47] <augiedoggie> haven't tried that[04:45:59] <Duggan> scgtrp yes, "hey" is an actual command :P[04:46:36] <scgtrp> (i was equally amused upon realizing "cat > kitten" was a valid unix command)[04:46:47] <Duggan> hehe[04:47:43] <Skipp_OSX> augiedoggie: well I'll be darned that did work.[04:48:04] <Skipp_OSX> augiedoggie: okay, much easier way to test. Set the preferences, then `hey Deskbar QUIT` then restart Deskbar[04:48:21] <Skipp_OSX> err I mean Duggan sorry[04:49:03] *** oxoocoffee has quit IRC[04:49:22] <Duggan> http://www.csd.uwo.ca/staff/magi/personal/humour/Computer_Audience/Funny%20UNIX%20Shell%20Commands.html[04:49:44] <Duggan> thanks, Skipp_OSX, augiedoggie[04:51:17] <Skipp_OSX> or in one line `hey Deskbar QUIT && Deskbar`[04:51:48] <augiedoggie> hm, might need a sleep command in there if Deskbar doesn't quit quick enough[04:52:15] <augiedoggie> or it'll try and launch another one while it's still running[04:52:17] <Skipp_OSX> augiedoggie: worked for me (in VBox)[04:52:33] <Duggan> VBox = slow? :)[04:53:07] <Duggan> seems to work for me too[04:53:21] <Skipp_OSX> Duggan: you see the problem[04:53:23] <Skipp_OSX> ?[04:53:41] <Duggan> not yet[04:53:43] <Skipp_OSX> ok[04:53:55] <Duggan> I do notice you have alot of ifs but not many elses[04:55:32] <Skipp_OSX> Duggan: well, like I said before, the ifs are really just precautionary... BarView() is going to be valid unless something really bad happens, I guess I could try and construct a new BarView in the else case.[04:55:34] *** OmniMancer has quit IRC[04:55:54] <jrabbit> the python lib I thought worked is busted[04:55:54] <Skipp_OSX> but if something bad happened... then that probably won't work either.[04:56:00] <jrabbit> gotta rewrite it I think[04:56:13] <Duggan> Skipp_OSX depends on what bad thing happened[04:56:14] *** OmniMancer has joined #haiku[04:56:19] <Duggan> wb OmniMancer[04:57:14] <Duggan> Skipp_OSX I see a potential problem...[04:58:59] <Skipp_OSX> Duggan: okay... what is it?[04:59:00] <Duggan> the code you moved to functions, I don't like how those for loops look... it may be perfectly valid but make it readable, not impressive :)[04:59:13] <Duggan> and I know you didn't do one of them[04:59:55] <Skipp_OSX> Duggan: no I didn't, I just copied. For whatever reason the author decided to go bottom up instead of top down.[05:00:11] <Duggan> for (int32 teamIndex = fExpando->CountItems(); teamIndex-- > 0;) { is less readable (in my opinion) than for (int32 teamIndex = fExpando->CountItems(); teamIndex > 0; teamIndex-- ) {[05:00:30] <Skipp_OSX> Duggan: didn't want to be thrown off by expanding teams he said, whatever that means[05:00:48] <Duggan> anyway, that's not the big issue[05:00:55] <Duggan> I think the issue is the value of teamIndex[05:01:46] <Duggan> teamIndex is going from CountItems() down to 1 ( > 0) yet it's used in the fExpando->ItemAt(teamIndex)...[05:02:00] <Duggan> most of the time items are numbered 0 to count -1[05:03:06] <Duggan> try using ItemAt(teamIndex-1) ? ... I don't even know if that would be related, but I think it might be worth a shot to try? ...[05:03:28] <Duggan> I'm not as adept at this particular code as you are[05:03:48] <Skipp_OSX> Duggan: you mean int32 teamIndex = fExpando->CountItems() - 1; teamIndex >= 0; teamIndex-- ??[05:04:00] <Skipp_OSX> for (int32 teamIndex = fExpando->CountItems() - 1; teamIndex >= 0; teamIndex--)[05:04:07] *** phalax has joined #haiku[05:04:18] <Duggan> you can do it either that way or just subtract one from teamIndex when you use it[05:04:36] <Duggan> it's the same, just a question of when you subtract the 1[05:05:16] <Skipp_OSX> Duggan: no, that is different, it would go from CountItems() -1 to 0 instead of CountItems() to 1[05:05:42] <CIA-73> Haiku: kallisti5 * r42553 /haiku/trunk/ (5 files in 2 dirs) http://dev.haiku-os.org/changeset/42553 :[05:05:42] <CIA-73> * move bios functions into bios.cpp[05:05:42] <CIA-73> * implement various methods to pull AtomBIOS from card[05:05:42] <CIA-73> * add some missing registers to headers from linux drm driver[05:05:44] <Skipp_OSX> Duggan: I copied that code though so I assume that it must be right[05:06:01] <Duggan> the value passed to fExpando->ItemAt() would be the same though, regardless of where the 1 is subtracted[05:06:45] <Duggan> or I'll say, for ever instance where teamIndex is used inside the loop[05:08:11] <Duggan> for (int x = 0; x < 10; x++) cout<<x+1; is the same as for(int x = 1; x <= 10; x++) cout<<x;[05:08:20] <Skipp_OSX> Duggan: lets see what happens if I go from CountItems() - 1 to 0 instead of CountItems() to 1[05:08:43] <Skipp_OSX> Duggan: most likely outcome is a spectacular crash[05:08:44] <Duggan> the result is the same i mean[05:08:51] <Duggan> probably, but try :)[05:10:22] <Skipp_OSX> Duggan: no crash, but no change either... that is strange[05:10:39] <Duggan> try it 2 or 3 times[05:10:41] <Duggan> not just once[05:11:33] <BinkySMC2> go Kallisti5 ![05:11:55] <Duggan> also make sure you change both loops[05:12:18] <Skipp_OSX> Duggan: I opened a bunch of apps, moved Deskbar around like a madman, used the hey trick a couple times, no crash[05:12:23] <Duggan> SaveExpandedItems() and ExpandItems()[05:12:26] *** freakazoid0223 has joined #haiku[05:12:33] <Skipp_OSX> Duggan: same exact result. Yes, I changed both instances[05:12:41] <Duggan> you recompiled and replaced the executable? (just making sure)[05:13:10] <Duggan> I couldn't tell you how many times I got pissed because something I did should have fixed a bug but I forgot to replace the exec[05:13:22] <Duggan> or for whatever reason was executing the wrong one[05:13:52] <Skipp_OSX> Duggan: I am double checking because I was expecting a crash. They can't both be right[05:14:08] *** AlienSoldier has quit IRC[05:14:36] <Duggan> I wouldn't think that the way it was was correct, the value might be clamped inside the ItemAt() call or something[05:15:12] *** oxoocoffee has joined #haiku[05:15:30] <Duggan> try the sigIndex loop too[05:15:35] <Skipp_OSX> Duggan: I think the author just wrote it that way to make it fit into 80 chars[05:16:24] <Duggan> maybe but I still hate that syntax :/[05:16:35] <Duggan> I personally don't limit my code to 80 character lines[05:16:41] *** phalax has quit IRC[05:17:00] <Duggan> I try to keep it to a page, but even then I'm not anal[05:17:02] <Skipp_OSX> Duggan: well, I double checked. No change... I am perplexed[05:17:18] <Duggan> did you try changing the sigIndex loop in ExpandItems?[05:17:21] <Skipp_OSX> Duggan: well the coding guidelines say to stay under 80 chars[05:17:32] <Duggan> 0 to CountItems-1 instead of 1 to CountItems()?[05:17:50] <Duggan> I know, but I don't abide by their coding guidelines in my personal code :)[05:18:17] <Duggan> if I were to submit something I may modify it after i'm done, but not while I'm working on it[05:18:31] <Skipp_OSX> Duggan: no, let me try that[05:18:43] <Duggan> alright[05:20:11] <Skipp_OSX> Duggan: personally, I hate all this backwords list wallking. I would do it 0 to CountItems() - 1 unless there is a really good reason not to... which for the life of me I can't think of one.[05:24:33] *** OmniMancer has quit IRC[05:25:33] <Skipp_OSX> Duggan: well, I got Deskbar to hang once... but I don't know if it is related and I can't reproduce it.[05:28:27] <Skipp_OSX> yeah, I can't do it again, I have like 20 apps opened all expanded and I am flinging Deskbar around like a madman and it works.[05:28:31] <Duggan> I can't think of one either, change it if you like and try it[05:28:44] <Duggan> collapse some of them[05:29:38] <Skipp_OSX> Duggan: It is a little bit slow now but otherwise works.[05:30:11] <Skipp_OSX> Duggan: nm, I got it... Deskbar is hung[05:30:12] *** oxoocoffee has quit IRC[05:30:35] <Skipp_OSX> Duggan: not crashed mind you, just hung[05:31:26] <Skipp_OSX> Duggan: doesn't respond to any input, won't redraw, but no crash dialog, it is still running. Let me try an alpha3 and see if I can crash Deskbar there too[05:33:41] <Duggan> alright[05:35:35] <BinkySMC2> several QT apps crash deskbar, anything with a QT deskbar replicant seems to be enough actually[05:35:55] <Skipp_OSX> Duggan: so far no, I can't get Deskbar to hang in alpha3. Let me revert and see if I can get Deskbar to hang[05:37:21] <BinkySMC2> I'd be willing to put a $50 spot on whoever can find and fix that annoying ass bug BTW[05:40:28] <Duggan> alright[05:40:41] *** OmniMancer has joined #haiku[05:44:22] <Skipp_OSX> Duggan: well, I have good news and bad news and bad news[05:44:43] <Skipp_OSX> The good news is that we aren't crazy.[05:45:39] <Skipp_OSX> The bad news is that with around 20 apps open, half expanded half not, I can get Deskbar to crash right away in BString::ICompare() which is almost definitely my code[05:45:41] <Duggan> lol alright[05:45:59] <Duggan> hmm ok, what's the bad news?[05:46:30] <Skipp_OSX> That Deskbar also hangs when I fix bad news #1[05:46:46] <Skipp_OSX> so... what this means is... I probably need to add some locking.[05:46:56] <Duggan> alright, what's your fix for bad news #1?[05:47:18] <Skipp_OSX> Duggan: The fix for bad news #1 is to go from CountItems() - 1 to 0 instead of CountItems() to 1[05:48:56] * Duggan is looking for a call to BString::ICompare()...[05:49:03] *** Tapionp has quit IRC[05:49:35] <Duggan> the ExpandItems() function?[05:49:50] <Duggan> yep, got it..[05:50:56] <Skipp_OSX> Duggan: wait... does it call the middle part of the for loop before it runs the first time of after?[05:51:24] <Skipp_OSX> Duggan: no wait... it doesn't matter.[05:51:48] <Skipp_OSX> Duggan: so, this means that the current code most likely has an off-by-one error in it[05:52:04] <Duggan> if they only still included the BeBook with Haiku :/..........[05:52:36] <Skipp_OSX> Although you never see it because it doesn't manifest itself because the object is too long lived.[05:52:53] <Skipp_OSX> err too short lived I mean[05:53:06] <Duggan> ok, change it back to Compare() not ICompare)([05:53:10] <Duggan> there's no sense in using ICompare[05:53:34] *** nsuperbus has joined #haiku[05:53:36] <Duggan> unless you can cite some excuse[05:54:11] <augiedoggie> um, the bebook is one of the optional packages[05:54:14] <Skipp_OSX> Duggan: well, is the sig case senstive?[05:54:27] <Duggan> augiedoggie there (until recently) was a link to it by default on the desktop[05:54:46] <augiedoggie> that's a limitation of installoptionalpackage[05:54:51] <Duggan> Skipp_OSX Haiku is case sensitive[05:54:53] <Skipp_OSX> I mean, is application/vendor-application different from Application/VenDoR-AppLiCaTion ??[05:55:04] <Duggan> augiedoggie what do you mean? installoptionalpackage had nothing to do with it...[05:55:05] <augiedoggie> the link is created if you include bebook in the build profile[05:55:16] <Duggan> ah[05:55:19] <Duggan> well if they hadn't removed it...[05:55:20] <Duggan> :P[05:55:28] <Duggan> Skipp_OSX yes it is[05:55:31] <Skipp_OSX> I would think those 2 sigs should be treated as the same[05:55:36] <Duggan> nope[05:55:37] <Skipp_OSX> Duggan: so that is why I use ICompare[05:55:46] <Duggan> then don't :) because they're different[05:55:55] <Skipp_OSX> okay I can live with that[05:57:30] *** noz_mob has quit IRC[05:59:57] <Skipp_OSX> I would image that you having 2 apps with sigs only different by case would be a rare case that would cause problems anyway[06:00:27] <Duggan> some vendors may do that, you never know[06:00:46] <Duggan> especially if "vendor" is used to refer to some guy sitting in his house writing code... like me :P[06:00:50] <Duggan> (but I don't do that)[06:02:55] *** MAX3 has quit IRC[06:06:59] *** mmadia has quit IRC[06:07:37] <Skipp_OSX> Duggan: well, I can't get Deskbar to hang anymore so perhaps that bug is either fixed... or maybe it is in app_server or something[06:08:50] <Skipp_OSX> or maybe it was in ICompare... I don't know[06:09:20] <Skipp_OSX> but in any case we are down to only good news. We are not crazy and I fixed yet another subtle Deskbar bug[06:10:08] <Skipp_OSX> well, except for the bad news that started all this, a crashed Deskbar does not have the right height on startup.[06:12:08] <OmniMancer> or perhaps you only hid the bug :P[06:12:22] <Skipp_OSX> OmniMancer: no, I revealed it[06:12:30] <Skipp_OSX> well actually Duggan did...[06:13:46] <Skipp_OSX> OmniMancer: It is a very subtle off by 1 error that is hard to see[06:14:26] <Skipp_OSX> OmniMancer: http://haiku.it.su.se:8180/source/xref/src/apps/deskbar/BarView.cpp Line 488 Counts from CountItems() to 1, it should count from CountItems() - 1 to 0[06:16:31] <Duggan> back, sorry[06:16:57] <Duggan> I did?[06:17:21] <Skipp_OSX> Duggan: well, I think so[06:18:10] <Duggan> so what is the status of the existing bug? still off by 1 team/window height?[06:19:46] <Skipp_OSX> Duggan: Here is what I think is the case, now I could be wrong, but, it looks like the current (trunk) code has an off by 1 error in BarView.cpp line 488 (and 481)[06:21:06] <Duggan> ok, gotta let the dog out, when I get back I'll take a look[06:22:42] *** BinkySMC2 has quit IRC[06:29:54] <Duggan> alright, lets see...[06:30:45] <Duggan> take "if (signature == NULL) continue;" out[06:30:58] <Duggan> lines 483, 484[06:31:50] *** oxoocoffee has joined #haiku[06:32:01] <Duggan> that was put in to prevent a crash if ItemAt() returns an invalid signature..... aka: if an invalid sigIndex was passed to it..... like would happen if sigIndex = expandedItems->CountItems()[06:32:15] *** oxoocoffee has quit IRC[06:32:46] <Duggan> but we fixed that, so that check shouldn't be necessary anymore[06:32:49] <Duggan> Skipp_OSX that make sense?[06:34:10] <Duggan> and due to the similar fix in the other loop, you can take "item != NULL &&" out in line 491[06:35:26] <Duggan> all those things are doing is hiding other problems if they exist... so if it crashes after removing those things we might have a better idea as to why[06:35:39] <Duggan> Skipp_OSX anybody home? lol[06:36:26] <Skipp_OSX> Duggan: I am here.[06:36:43] <Skipp_OSX> Duggan: yes, I can remove those lines... I don't see how that would fix my problem but I can remove them[06:36:52] <Duggan> it wouldn't fix it but I think I just found it[06:37:02] <Duggan> it has to do with the logic of the 2nd loop[06:37:23] <Duggan> ok nope, didn't find it[06:37:35] <Duggan> there's a break; there, so that's a Good Thing[06:38:55] <Duggan> lines 499 - 501 shouldn't be necessary... not like that...[06:39:20] <Duggan> BList should self destruct when it goes out of scope at the end of the function and clean itself out.... if you want to do it manually I'm sure there's a single member function to do all that...[06:40:11] <Duggan> it's an object, not a pointer, so it should clean itself up... declared on line 456[06:41:27] <Duggan> oh ok I see... BList doesn't do it itself :/[06:42:12] <Duggan> I guess it can't... oh well[06:42:56] <Duggan> ok, forget what I said about 499-501 :P[06:43:07] <augiedoggie> that's why BObjectList is recommended[06:43:15] <Duggan> gotcha[06:43:29] <Duggan> thanks, augiedoggie[06:45:05] <Skipp_OSX> yeah BList is not very smart...[06:45:18] *** MatthewH12 has quit IRC[06:45:28] <Skipp_OSX> again, I only chose BList because the original author did[06:45:43] <Duggan> yep, BList is just a wrapper for a void*[]... or something like that hehe[06:45:46] <Skipp_OSX> Then again this mysterious original author seems to have made a few mistakes...[06:45:59] <Duggan> so it's index starts at 0[06:46:05] <augiedoggie> yeah, this code probably predates BObjectList[06:46:19] <Duggan> wow, it's an opentracker file...[06:46:27] <Duggan> talk about ancient...[06:47:37] *** MatthewH12 has joined #haiku[06:48:29] <Duggan> wait yeah, I think the error is in that 2nd loop...[06:49:05] *** OmniMancer has quit IRC[06:50:09] <Duggan> Skipp_OSX swap the logic of the 2nd loop......[06:50:19] <Duggan> replace line 491 with:[06:50:51] <Duggan> if (signature->Compare(item->Signature())==0) {[06:51:03] <Duggan> and change 492 to true[06:52:05] <Duggan> it probably wouldn't hurt to add an else to that if either...[06:52:39] <Duggan> else {item->ToggleExpandedState(false);} // (notice no "break;")[06:53:19] <Skipp_OSX> Duggan: I really don't know what line 491 means...[06:54:06] <Duggan> the outer loop iterates through the expandeditems, setting "signature" to the sig of each item in the expandeditems list[06:54:14] <Skipp_OSX> Duggan: you want me to take out the NULL checks?[06:54:30] <Duggan> the inner loop loops through all items and (currently) collapses items not in the list[06:54:38] <Duggan> one sec Skipp_OSX[06:54:58] <Skipp_OSX> Duggan: this is in ExpandItems() ???[06:56:02] <Skipp_OSX> Duggan: no, the inner loop expands items that are in the list[06:56:57] <Skipp_OSX> Duggan: all items are unexpanded because when you drag deskbar fExpando is deleted and recreated. That is why I need to save the expanded items in a list and re-expand them each time.[06:57:33] <Duggan> then what does ToggleExpandedState(false) mean?[06:58:13] <Duggan> sounds to me like it's collapsing, not expanding[06:58:57] <Duggan> the contents of the condition lend to that belief as well[07:01:31] <Skipp_OSX> Duggan: well, confusingly it means, Toggle the state (not expanded -> expanded) and the false means, don't resize the window[07:02:00] <Duggan> that's pretty confusing...[07:02:02] <Skipp_OSX> Duggan: I don't resize the window because I do it later down in the function. I don't want to resize the window for each item, only once at the end. Thus the false.[07:02:02] *** BDifferent has joined #haiku[07:02:15] <Duggan> ok, then the problem still remains[07:02:30] <Skipp_OSX> Duggan: which problem is this again?[07:02:49] <Duggan> what's the condition?[07:03:07] <Duggan> if (item != NULL && !signature->Compare(item->Signature())) {[07:03:12] <Duggan> taking the null check out....[07:03:21] <Duggan> if (!signature->Compare(item->Signature())) {[07:03:55] <BDifferent> Are the Network preferences supposed to work at the moment? I.e., is it sufficient to choose DHCP and "Apply" to set up my NIC or do I have to do additional things to set up networking?[07:04:27] <Duggan> BDifferent there are still a few problems with network prefs at the moment[07:04:31] <Skipp_OSX> Duggan: I have no NULL check there[07:04:48] <Duggan> ok, but BString::Compare() returns 0 of they're equal right?[07:04:51] <Skipp_OSX> BDifferent: but to answer your question it is sufficient[07:05:00] <Skipp_OSX> Duggan: yes, like strmpr[07:05:03] <Skipp_OSX> strcmp[07:05:50] <Duggan> so what's the def of !-1, !0, and !1?[07:06:20] <scgtrp> 0, 1, and 0[07:07:17] <Duggan> ok then there's no error there :/[07:07:58] * Duggan wonders what's wrong with "==0"...[07:08:25] <scgtrp> that's how i usually check the return value of strcmpish functions[07:08:35] <scgtrp> but !a and a==0 are the same[07:08:56] <Duggan> hd space is no longer at a premium... we can use ==0 these days without having to count how many bits it's going to take to store...[07:09:05] <Duggan> but whatever :P[07:09:39] <Duggan> Skipp_OSX is it given that in all cases the items are already collapsed?[07:09:42] <Skipp_OSX> Duggan: it is just a style choice a == 0 or !a whichever[07:09:46] <scgtrp> ^[07:09:54] <Duggan> Skipp_OSX question is still readability :/[07:10:07] <Duggan> hehe[07:10:17] <Skipp_OSX> Duggan: it is a given... okay, I'll admit that it is a bit confusing[07:10:24] <scgtrp> i use == 0 when i'm actually comparing for whether something equals 0, and ! when i'm testing boolean-ish things[07:10:52] <scgtrp> also, pointers, because 'not p' kinda sounds like 'p isn't valid'[07:10:58] <BDifferent> @Skipp_OSX: Okay it works now. Yesterday I tried but did not succeed, today I tried again but had forgotten to plug in the cable ... With the cable it works now :-)[07:11:03] <Skipp_OSX> when ChangeState() gets called, and that is a lot of the time, PlaceApplicationBar() destroys and rebuilds fExpando[07:11:11] <Duggan> BDifferent hehe glad you got it working[07:11:50] <Skipp_OSX> when fExpando is destroyed whether or not an item is expanded is lost.[07:12:15] <Skipp_OSX> in other words, all items are not expanded.[07:13:00] <Duggan> gotcha[07:13:04] <Skipp_OSX> In the trunk code it saves whether or not an item is expanded right in ChangeState. This works unless you switch modes.[07:13:30] <Duggan> which you've preserved[07:13:38] <Duggan> which is what this was all about to begin with, right?[07:13:54] <Skipp_OSX> When you switch modes, to say horizontal mode, it still saves whether or not the items are expanded, but, since in horizontal mode nothing can be expanded since fExpando doesn't exist[07:14:06] <Duggan> yep, gotcha[07:14:14] <Skipp_OSX> so when you switch back you lose it.[07:14:34] <Duggan> here's a dumb question...[07:14:56] <Skipp_OSX> What I did was to change it so that it stores in a list that is on the BarView that doesn't go away. So that way it continues to save the expanded items even when fExpando is deleted.[07:15:16] <Duggan> if "expand new applications" is checked, and you move the bar to horizontal, start a new app, then move it back to vertical, will the application be expanded?[07:15:16] <Skipp_OSX> But I still have to re-expand each time.[07:15:29] <Skipp_OSX> Duggan: in trunk, no, in my code, yes[07:15:41] <Duggan> what if expand new apps isn't checked?[07:15:50] <Skipp_OSX> Duggan: then no[07:15:50] <Duggan> it all works properly?[07:15:56] <Duggan> ok[07:15:59] <Duggan> just checking :)[07:16:27] <Skipp_OSX> Duggan: it's okay, it is a fair question, and I had to do a bit of work and testing in order to get it this far.[07:16:35] <Duggan> no doubt[07:17:38] <Skipp_OSX> Duggan: my call in BarApp.cpp AddTeam (see my patch) is what makes that work.[07:17:55] <Duggan> gotcha[07:18:24] <Duggan> have you tested it after removing those null checks?[07:18:31] <Skipp_OSX> when you add a team, it checks the expandNewTeams flag, if set, it adds the item to the expanded list in any mode. If the flag is not set then the item is not expanded on startup so it doesn't get added to the list[07:18:48] <Duggan> gotcha[07:19:10] <Skipp_OSX> Duggan: no, but I really don't think that I should remove them. They prevent bad things from happening if some NULL item gets put in the list somehow[07:19:23] <Duggan> it never will if you do the loops properly[07:19:24] <Skipp_OSX> Duggan: I agree that it is a little paranoid, but, it really doesn't hurt much[07:19:32] <Duggan> which you are now[07:19:35] <Duggan> here's the question....[07:19:41] <Duggan> how did you fix tracker not expanding?.....[07:19:51] <Duggan> because I bet these errors in the loops fixed that too[07:20:00] <Duggan> I bet that's where that bug originated[07:20:05] <BDifferent> hm, does not work; the Network Prefs replicant in the Deskbar says "Ready", however, installoptionalpackage/wget does not find haiku-files.org, and ping 192.168.100.1 (status page of the router) says "Network unreachable"[07:20:25] <BDifferent> ... been some time since I used Haiku/BeOS. What can I try next?[07:21:19] <Duggan> I can write programs that use the network, I know a little about networking, but I don't know enough to help much with problems with it these days[07:21:28] <augiedoggie> what does ifconfig say your address is?[07:21:44] <augiedoggie> (or the network replicant in the deskbar)[07:22:16] <Duggan> Skipp_OSX it never got to the 0th element.... which would have been Tracker.... so it never would have gotten expanded by default[07:22:49] <BDifferent> 169.254.0.228[07:22:55] <Skipp_OSX> Duggan: I agree... sort of... but it did get Tracker[07:23:09] <Duggan> how did you fix it though?[07:23:21] <augiedoggie> BDifferent: yeah, that's not valid, you'll have to look through your syslog to see why dhcp isn't working[07:23:50] <BDifferent> okay, it's in boot/var/ IIRC? :-)[07:23:56] <augiedoggie> /var/log[07:24:08] <Skipp_OSX> Duggan: I did, and yet it is still broken in the same way, Deskbar is too short by exactly the height required to fit the window list items[07:25:06] <Duggan> oh I thought you said you fixed the non-expanding-tracker problem[07:25:26] <Duggan> wait what....[07:25:35] <Duggan> I thought you said it was 1 item off.... not a whole list off[07:25:58] <Skipp_OSX> Duggan: well, I did, that prevents a crash[07:26:26] <Skipp_OSX> Duggan: fixing the off by 1 fixes Deskbar crashing...[07:27:20] <Skipp_OSX> Duggan: but it doesn't fix the if Deskbar crashes when you restart it Deskbar is too short problem.[07:28:04] <Duggan> and how "too short" is deskbar in that case?[07:28:14] <Duggan> n windows?[07:30:06] <Skipp_OSX> Duggan: like I said, it is short by the number of Window Items you have open... in other words it is just tall enough to fit the appliations sans expanded Window Items[07:30:17] *** ddavid123 has joined #haiku[07:31:07] <Skipp_OSX> Duggan: and if you unexpand a window and re-expand it, or do anything that calls ChangeState() it corrects the issue.[07:31:38] <Skipp_OSX> err unexpand and re-expand an app I mean[07:31:53] <BDifferent> syslog contains "Send DHCP_DISCOVER to 255.255.255.255:67" several times. Later "Timeout shift ... "[07:32:23] <BDifferent> could it be a firewall issue?[07:33:14] <augiedoggie> unlikely[07:33:15] <Skipp_OSX> Duggan: I just took out all of the NULL checks and tried again. I got the same behavior as expected[07:33:28] <ddavid123> I have a tiny insignificant just-for-my-information question to ask.[07:34:06] <ddavid123> Why is the /system directory not named /haiku like Be[07:34:23] <ddavid123> Like Be was /be[07:36:02] <ddavid123> I suppose it really doesn't matter[07:36:12] <augiedoggie> BDifferent: I would try switching to static and then back, or reboot[07:37:45] <BDifferent> @augiedoggie I will do that later, I have to go to work now (7:30 AM here :-) ) Thanks for your help.[07:37:51] *** BDifferent has quit IRC[07:37:58] <Duggan> Skipp_OSX no crash?[07:38:13] *** OmniMancer has joined #haiku[07:38:28] <Duggan> ddavid123 probably just to be more generic than Be was[07:39:32] <ddavid123> To accomodate the future distros of Haiku to solve future naming problems[07:39:39] <Duggan> most likely[07:39:40] <ddavid123> ?[07:40:34] <ddavid123> If the the directory was named /haiku, they would have to rename it and cause problems?[07:40:35] <Duggan> distros / forks / whatever[07:40:50] <ddavid123> have to recompile?[07:40:51] <Skipp_OSX> Duggan: no crash[07:40:55] <Duggan> Skipp_OSX good :)[07:41:04] <Skipp_OSX> Duggan: Like I said, paranoia[07:41:41] <Duggan> hehe... the only time you should ever get a NULL is if the index you pass it is out of bounds.... which is the case when the index = CountItems()[07:42:02] <Duggan> but index will never = CountItems() now, so it *should* be safe[07:42:25] <Skipp_OSX> Duggan: I agree in principle[07:42:57] <Duggan> really, if you want to keep the checks in there you can... but (in principle as you say) they would only take up cycles[07:43:10] <Duggan> hehe[07:43:33] <Duggan> yeah I know, I used the wrong phrase :P[07:45:46] <ddavid123> I am thinking about buying an old G5 imac from ebay and install Haiku on it! That should prove fun and interesting![07:46:17] <ddavid123> Will have to build the Haiku ppc image myself[07:47:05] <ddavid123> I think Haiku would just run great on it.[07:47:17] <Duggan> as great as a ppc build would run :)[07:47:34] <ddavid123> Or, better yet a used intel imac[07:48:00] <Duggan> that would probably do better... I'm not a ppc guy but I think there are still some problems with the ppc build[07:48:06] *** OmniMancer has quit IRC[07:48:28] <Skipp_OSX> ddavid123: You do realize that Haiku doesn't boot on PPC at this moment correct?[07:48:49] <ddavid123> I thought it has build options for ppc[07:48:54] <Duggan> it does[07:49:01] <Duggan> but the build still has problems[07:49:06] <ddavid123> oh[07:49:10] <ddavid123> I see[07:49:26] <ddavid123> not fully supports the ppc instruction set?[07:49:33] <Skipp_OSX> Duggan: I replaced all of the index's with i and j, now everything fits in 80 chars easily[07:49:47] <Duggan> I don't think that's the problem, I think it's just alot of bugs in bad places[07:50:02] <Duggan> Skipp_OSX but not very descriptive :)[07:50:05] <ddavid123> that is ok, I will wait for R1[07:50:09] <ddavid123> maybe[07:50:14] <Duggan> try intel :)[07:50:24] <ddavid123> lol * 1000000[07:50:24] <Duggan> you're running it already, aren't you?[07:50:35] <ddavid123> A Mac?[07:50:39] <Duggan> Haiku[07:50:46] <ddavid123> Yes[07:50:49] <Skipp_OSX> Duggan: sure, I don't really think you need much description for an index.[07:51:00] <Duggan> ddavid123 native?[07:51:02] <Duggan> or vm?[07:51:05] <ddavid123> Yes[07:51:10] <ddavid123> Native[07:51:11] <Duggan> ah ok[07:51:34] <ddavid123> And almost everything works![07:51:43] <Duggan> Skipp_OSX might want to talk to one of the devs to see what they say, they might prefer it not be like that.... I don't know so I won't say[07:51:44] <ddavid123> he he[07:51:52] <Duggan> ddavid123 hehe yep[07:52:51] <ddavid123> Duggan I am going to buy a system 76 desktop computer fairly soon, I should just add a second hard drive and dual-boot![07:52:57] <Skipp_OSX> anyway, so, I think that while I've made some progress here, I am no closer to fixing the original bug.[07:53:03] *** ericxjo has joined #haiku[07:54:39] <ddavid123> I wonder, are there any x86 emulators that you can install as the OS and then emulate the x86 platform to boot haiku from[07:55:00] <augiedoggie> xen[07:55:04] <ddavid123> like it is it's own kernel[07:55:13] <ddavid123> just the emulator[07:56:18] *** OmniMancer has joined #haiku[07:57:15] <ddavid123> I may do that[07:57:22] <ddavid123> brb[08:02:13] *** ddavid123 has quit IRC[08:04:31] *** scgtrp has quit IRC[08:05:59] *** scgtrp has joined #haiku[08:06:00] *** LiquidInsect has joined #haiku[08:10:48] <Skipp_OSX> Duggan: well, thanks for looking at the code, you have helped me find a few problems[08:11:49] *** kPb_in has joined #haiku[08:16:29] *** jjido has joined #haiku[08:21:19] *** jjido has quit IRC[08:23:04] *** jjido has joined #haiku[08:27:21] <Duggan> back, sorry[08:27:32] <Duggan> Skipp_OSX no problem, I'd like to help fix that final one[08:28:05] <Duggan> I don't think you answered my question though...[08:28:30] <Duggan> the size of the deskbar, is it off by 1 item or by some number of items?[08:36:38] *** Pinaraf has joined #haiku[08:36:55] <Skipp_OSX> Duggan: It is off by the number of Window Items[08:37:22] <Skipp_OSX> Duggan: Basically it is just tall enough to fit the number of apps you got as though there were no expanded windows[08:37:23] <Duggan> ok, I was under the impression previously that it was 1 window item, but I got the idea today that it was more[08:37:28] <Skipp_OSX> err expanded items[08:37:32] <Duggan> gotcha[08:37:41] <Duggan> and where does that expansion actually take place?[08:38:26] <Skipp_OSX> Duggan: teamItem->ToggleExpandState(false);[08:39:18] <Skipp_OSX> Duggan: and if you'd like to see that it is http://haiku.it.su.se:8180/source/xref/src/apps/deskbar/TeamMenuItem.cpp Line 496[08:39:25] *** jjido has quit IRC[08:41:04] <Duggan> thanks for the link, let me take a look[08:41:35] <Duggan> wtf.... 498?....[08:41:45] <Skipp_OSX> 496[08:42:02] <Duggan> oh duh... ok[08:42:24] <Duggan> this is the individual object's toggle state function... my bad.... of course the state will toggle if you're in a toggle state function lol[08:43:04] <Skipp_OSX> Duggan: well, I call that from the ExpandItems() function in BarView[08:43:21] <Duggan> have you made any changes to this function?[08:43:30] <Skipp_OSX> Duggan: no, no changes there[08:43:39] <Duggan> well theoretically it works then...[08:44:34] <Duggan> ok looking back at your diff....[08:44:52] <Duggan> line 214 (of the diff)...[08:45:37] <Duggan> hold on I need to find that file..[08:47:29] <Duggan> weird...[08:47:33] <Duggan> or wierd...[08:47:59] <Duggan> did you add the AddExpandedItem() function?[08:50:16] <Skipp_OSX> Duggan: yes I did, AddExpandedItem(), RemoveExpandedItems() SaveExpandedItems() are all new[08:50:46] <Duggan> ah ok.... just strange that it shows you removed an Invalidate() call .... which otherwise wouldn't have been inside a function at all?.......[08:51:31] <Skipp_OSX> Duggan: It is a little hard to follow but I didn't remove the Invalidate() call, I just moved it to after my function call.[08:52:03] <Duggan> btw, there's another ICompare() :)[08:52:19] <Skipp_OSX> Duggan: yeah, I got them all. Why don't I give you a new patch[08:52:27] <Duggan> ah ok thanks :)[08:55:03] <Duggan> oh another thing.... you added the fExpandedItems to TBarView... where was it originally again?[08:55:37] *** Animal-X has quit IRC[08:55:58] <Skipp_OSX> Duggan: http://pastebin.com/MLp018UP[08:56:19] *** Animal-X has joined #haiku[08:56:27] <Skipp_OSX> Duggan: it was originally a List created in ChangeState()[08:56:36] <Duggan> danke schön[08:57:04] <Duggan> and it's scope was only valid inside ChangeState()?[09:00:14] <Skipp_OSX> Duggan: yeah, and that was the problem[09:02:47] <Duggan> ok, do me a favour...[09:02:52] <Duggan> I know this is going to sound stupid or pointless...[09:03:22] <Duggan> line 183 of the diff... the RemoveExpandedItems() call in TBarView::ExpandItems().... comment it out and test[09:03:43] <Duggan> ok nevermind[09:03:52] <Duggan> I keep forgetting it's just BStrings :/[09:04:11] <Duggan> hrm......[09:04:14] <Skipp_OSX> Duggan: no I need that line[09:04:20] <Skipp_OSX> Duggan: I have a theory[09:04:24] <Duggan> eh....[09:04:27] <Duggan> thinking about it again...[09:04:33] <Duggan> see, that's the static list....[09:04:39] <Duggan> and you're emptying it...[09:04:54] <Duggan> something doesn't seem right about that[09:04:58] <Skipp_OSX> Duggan: yes but after I've expanded it. So I want to empty it[09:05:14] <Duggan> yeah but next time it redraws it recreates that list?[09:05:28] <Duggan> or changes, not so much redraws[09:05:39] <Skipp_OSX> Duggan: I have no idea what is going to happen to the items in the meantime, they may get unexpanded, so, I empty the list, then, when something happens I build it up again.[09:06:09] <Skipp_OSX> Duggan: if I don't remove the items from the list then they will stay expanded forever[09:06:20] <Duggan> ok gotcha[09:06:28] <Skipp_OSX> Duggan: anyway, I have a theorey[09:06:35] <Duggan> I just figure leave the list alone and only add or remove things as it's changed[09:06:39] <Duggan> so what's your theory?[09:07:27] <Skipp_OSX> In BarApp.cpp I call BarView()->UpdatePlacement() just before I call fBarWindow->Show() in the BarApp Constructor.[09:08:06] <Skipp_OSX> Now, because the window isn't shown yet, it is locked, it can't be resized. So even though I am telling it to resize it won't[09:08:21] <Duggan> possibly, swap them and try that[09:08:34] <Skipp_OSX> Duggan: well, I did that. Deskbar crashes on startup[09:08:39] <Duggan> it *shouldn't* matter, but it might...[09:08:47] <Duggan> where does it crash? in Show()?[09:10:02] <Skipp_OSX> Duggan: no, it crashes in BarView::PlaceBeMenu when it calls Bounds()[09:10:20] <Duggan> very interesting...[09:10:28] *** markos_ has joined #haiku[09:10:59] <Skipp_OSX> now, who knows why this is happening, I don't, but, I think that that is the reason that it isn't getting resized.[09:11:13] <Duggan> have you tried commenting out the UpdatePlacement() call?[09:12:16] <Skipp_OSX> Duggan: http://haiku.it.su.se:8180/source/xref/src/apps/deskbar/BarView.cpp Line 215 is where it crashes[09:12:21] <Skipp_OSX> I think[09:12:30] <Duggan> UpdatePlacement wasn't originally called until AttachedToWindow()....[09:12:41] <Skipp_OSX> Duggan: true[09:13:13] <Skipp_OSX> Duggan: that is actually earlier,[09:13:26] <Duggan> ahhhhh[09:13:28] <Duggan> put it back there :)[09:13:35] <Skipp_OSX> Duggan: I can't[09:13:37] <Duggan> why?[09:13:49] *** ericxjo has quit IRC[09:14:13] <Skipp_OSX> The line fBarWindow = new TBarWindow(); creates the BarWindow, which creates the BarView which calls AttachedToWindow() which calls UpdatePlacement()... or used to at least, but...[09:14:54] <Skipp_OSX> That happens before the the teams get added which means that any app that is open on startup will NOT get expanded if I call it there.[09:15:07] <Skipp_OSX> That is why I moved UpdatePlacement() down[09:16:24] <Duggan> I'm iffy about that...[09:17:02] <Skipp_OSX> Duggan: what I could do is move fBarWindow = new TBarWindow(); down instead and put UpdatePlacement() back in AttachedToWindow()[09:17:44] <Skipp_OSX> Duggan: but, because that is called before fBarWindow->Show() it still won't resize the window for me because the window is locked[09:18:18] <Skipp_OSX> I think...[09:18:32] <Duggan> it doesn't make sense that the window would be locked if it's not visible... it would make sense to update the window before you display it if necessary... but I'll take your word for it[09:20:37] <Duggan> it doesn't make sense that UpdatePlacement() would crash if called after Show()[09:21:33] <Skipp_OSX> Duggan: I am sorry, it is unlocked on show...[09:21:38] <Skipp_OSX> so here is the idea.[09:21:56] <Duggan> alright[09:22:08] *** Sikosis has quit IRC[09:22:21] <Skipp_OSX> Show the window which activates it and unlocks it, then immediately unlock it, call BarView()->UpdatePlacement(), then lock it again[09:22:36] <Skipp_OSX> no I said that wrong[09:22:50] <Skipp_OSX> Show which unlocks, lock, UpdatePlacement, unlock[09:23:13] <Skipp_OSX> That should fix the crash and fix the Resizing... I hope[09:23:14] <Duggan> try it?[09:23:51] <Skipp_OSX> Duggan: I really think this could work. I just have to compile which will take a few minutes[09:23:54] *** MatthewH12 has quit IRC[09:23:54] *** Sikosis has joined #haiku[09:23:54] *** ChanServ sets mode: +o Sikosis[09:23:57] <Duggan> you're the educated one here, I'm just trying to help as best I can :)[09:24:16] <Skipp_OSX> For some reason when I switch branches on git it takes a long time to compile, I don't know why[09:24:19] <Duggan> np, I'll be here... if not I'll be smoking so I'll be back soon enough hehe[09:24:35] <Duggan> so I might as well get it out of the way... brb hehe[09:24:52] <Skipp_OSX> Duggan: okay, I really think this could work provided that I am right about what is going on, otherwise, spectactular crash[09:25:06] <Skipp_OSX> okay, go smoke, it'll be a few minutes[09:33:08] <Duggan> hehe[09:33:09] <Duggan> back[09:33:20] *** Pinaraf has quit IRC[09:36:24] <Skipp_OSX> Duggan: yup that did it[09:36:34] <Duggan> fixed it?[09:38:07] <Duggan> ninjas, biker gangs, football players, and an apparent abundance of extra terrestrial lifeforms or at least an incredibly literal interpretation of the lyrics...[09:38:22] <Duggan> Jim Steinman was on acid when he did Total Eclipse of the Heart...[09:38:28] <Duggan> or whoever it was that directed the video...[09:38:57] <Skipp_OSX> Duggan: yes, that fixed it[09:39:15] <Duggan> well congratulations :)[09:39:24] <Duggan> clean it up, submit the patch :)[09:39:39] <Duggan> or maybe see what you need to do to merge it with the recent changes...[09:39:54] <Duggan> (and maybe fix some of those new bugs :'( )[09:39:58] <Skipp_OSX> It makes sense now that I understand why the window wasn't resizing, it wasn't shown tet.[09:40:05] <Skipp_OSX> yet[09:40:33] <Duggan> I guess I just don't understand the whole locking thing that well yet... sorry[09:40:59] <Skipp_OSX> Duggan: don't worry, niether do I, but I am learning[09:45:38] <Duggan> well I'm glad you finally figured it out :)[09:48:14] <Skipp_OSX> Duggan: me too![09:50:37] <Duggan> so what's next?[10:06:14] *** tqh has joined #haiku[10:06:18] <Skipp_OSX> Duggan: check out this video I made showing off another feature I am working on: http://www.youtube.com/user/jscipione#p/a/u/0/Yr-9Y5huFs0[10:08:36] <Duggan> can't :P[10:09:05] * Duggan is in Haiku which doesn't play well with youtube.[10:10:57] <Skipp_OSX> ah yes[10:11:31] <Skipp_OSX> Duggan: well, basically, it is a feature to increase the size of your Deskbar icons via a slider in the preferences.[10:11:32] <Duggan> otherwise I'd be happy to hehe[10:11:41] <Duggan> ah interesting[10:11:57] <Duggan> I'm trying to work on 3d hw accel[10:12:04] <Duggan> but I'm sure you know that already lol[10:12:07] <Skipp_OSX> Duggan: you can have any size from 16x16 to 64x64[10:12:20] <Duggan> I'll stick to 16x16 :D[10:12:21] <Duggan> hehe[10:12:28] <Skipp_OSX> ... but, it still needs some work[10:12:35] <Duggan> if you need any help, let me know[10:13:03] <Skipp_OSX> Duggan: okay, well, I really wanted to get that bug fixed, but now that it is, I will probably go work on the Haiku Book some more.[10:13:21] <Skipp_OSX> Enough Deskbar for now[10:13:45] <Duggan> hehe ok[10:13:53] <Duggan> I guess I'm working on DRI drivers tonight...[10:13:59] <Skipp_OSX> Duggan: actually, you can see the progress I've made on that here: http://74.74.222.96[10:14:00] <Duggan> now that I got a DRM driver to compile...[10:14:03] <Skipp_OSX> no flash required[10:14:28] <Duggan> nice[10:14:47] <Skipp_OSX> Duggan: well that sounds good too, 3d acceleration would be a great feature to have, especially for my Deskbar resizing feature.[10:14:58] <Duggan> I was thinking games, but yeah that too :P[10:15:26] <Skipp_OSX> I'd love to be able to smoothly animate between sizes[10:15:49] <Skipp_OSX> but for that you need a 3d acclerated app_server... and that is definitely an R2 feature.[10:16:03] *** adominguez has joined #haiku[10:16:15] <Skipp_OSX> but to get a 3d accelerated app_server, first you need 3d drivers[10:16:42] *** kPb_in has quit IRC[10:16:47] <Duggan> working on that... hehe[10:17:36] <Duggan> I don't think smoothly animating between sizes has much to do with 3d....[10:17:47] <Duggan> maybe only so far as scaling a 2d plane, but not much 3d[10:18:04] <Duggan> but hw accel anything would be better than what we've got[10:18:22] <Skipp_OSX> Duggan: I belive that you need 3d hardware to make it smooth. with 2d it is just too slow[10:19:09] *** OmniMancer has quit IRC[10:19:30] <Skipp_OSX> I guess that I could "cheat" just resize the bitmap until you let your mouse button up and then render from vector[10:19:57] <Duggan> that's a start[10:20:00] <Skipp_OSX> Right now I am re-rendering a bitmap at each step which is non-ideal[10:20:20] <Duggan> yep, I'd go with the previous idea[10:20:33] <Skipp_OSX> yeah, well, toss that in the todo pile[10:21:18] <Duggan> hehe I think my todo pile can only handle one thing right now hehe[10:21:46] <Skipp_OSX> yeah well, 3d acceleration is a big task[10:21:48] <tqh> you might want to profile what is slow. It should be doable imo.[10:22:12] <Skipp_OSX> tqh: I am pretty sure that rendering the bitmap from the HVIF is what is slow[10:22:38] *** iMe has quit IRC[10:22:57] <tqh> yes, but that is at a macro level. Why is that so slow...[10:23:20] <Skipp_OSX> tqh: although I am not sure. But I'd imagine that I could resize a bitmap much faster, it wouldn't look as good, but that is okay because I would re-render moments later[10:23:43] <tqh> I hate workarounds[10:23:55] *** _taos_ has quit IRC[10:24:07] <Skipp_OSX> tqh: It is slow because I have to grab the icon from disk and render it each time[10:24:18] <Duggan> you have to grab it from disk each time?[10:24:30] <tqh> first take a really hard look at the problem, and if nothing else do a workaraound and add a big fat comment about it being sub-optimal.[10:25:19] <Skipp_OSX> tqh: well, it isn't that slow, but it doesn't look great rerendering at 8px increments[10:25:34] <tqh> hmm, I'd expect it to be in cache after first time.[10:25:37] <Skipp_OSX> rerendering at 1px increments would look better, but would be much too slow[10:25:59] <Skipp_OSX> well, is much too slow[10:26:04] <tqh> Skipp_OSX, but the big question is, why is it slow and can we make it faster.[10:26:21] <Skipp_OSX> tqh: It is slow because rendering a bitmap from an HVIF is slow[10:26:48] <tqh> that's way to abstract to be an answer[10:26:53] <Skipp_OSX> slow if you are doing it 10 times per second anyway[10:27:45] <Skipp_OSX> tqh: well, it is probably also slow because Deskbar keeps redrawing itself each step of the way too.[10:27:51] <jrabbit> has anyoen written HVIF stuff for other OSes?[10:28:04] <tqh> jrabbit, not that I know.[10:28:09] <jrabbit> like a gimp export or inkscape option? :\[10:28:44] <Skipp_OSX> jrabbit: I don't think HVIF really makes sense on other OS's really. At least OS's that already have a preferred icon format[10:28:45] <tqh> Skipp_OSX, but since redrawing is slow, the next question would be: what in redrawing is slow?[10:29:18] <tqh> actually it makes a lot of sense. Especially on the web.[10:29:23] <Skipp_OSX> tqh: I don't know really[10:30:01] <Skipp_OSX> tqh: But I don't think that scaling a bitmap is that much of a workaround, it is pretty simple really[10:30:55] <Skipp_OSX> scaling a bitmap can be very very fast. The bilinear scaling that Haiku uses is pretty optimized. It doesn't always look great, but it is fast.[10:31:18] *** iMe has joined #haiku[10:31:47] <Skipp_OSX> bilinear scaling is used by Tracker to scale up BeOS icons in Tracker[10:32:00] <tqh> more code paths, more code, it all adds up. 1 workaround is pretty easy to keep track of, but since it still slow you will need it in the next scaling code and the next and the next. It is always better to try and fix things.[10:32:55] <Skipp_OSX> tqh: well, like I said, the real solution is to do the scaling in the graphics card using openGL. That would be fast and wouldn't need any workarounds.[10:33:51] <Skipp_OSX> tqh: but, since there is no 3d acceleration in Haiku right now, that isn't really an option right now.[10:33:57] <augiedoggie> opengl is still bitmap scaling[10:34:19] <tqh> I don't think it should be slow in software either. You think 3d will be available on ARM and other architectures?[10:35:00] <Skipp_OSX> tqh: no, but then you would just revert back to the 2d scaling in that case.[10:36:19] <Skipp_OSX> I guess one thing I could do would be to pre-render icons at every size between 16x16 and 64x64 and then just switch between them[10:36:40] <tqh> so you have three code paths for one simple thing: scaling vector, scaling with 3d and scaling with bitmap. It's not an elegant solution and it's getting bloated.[10:37:02] <Skipp_OSX> tqh: hmmmm, well, how would you approach the problem?[10:37:34] *** GeekShadow has joined #haiku[10:37:35] *** GeekShadow has joined #haiku[10:38:18] <Skipp_OSX> tqh: You could just have the scaling with 3d codepath and rely on fast software rendering for cases where 3d is not available.[10:38:19] <tqh> I would try and find out what exactly makes rendering so slow. An icon is a simplish object, it's very little data, the rendering shouldn't be that hard and it's quite a small destination bitmap[10:38:45] <tqh> something is suboptimal in rendering.[10:39:10] <Skipp_OSX> tqh: I agree[10:40:05] <Skipp_OSX> tqh: what I know is that OS's that do fast bitmap scaling do it with 3d hardware, so I assume that is the bottleneck.[10:40:18] <tqh> try to make a testprogram that only renders an icon, and then profile it with a profiler (there is one, I don't know how to use t) and see what takes time.[10:40:20] *** OmniMancer has joined #haiku[10:41:08] <Skipp_OSX> tqh: well, I guess I will have to do that eventually. Right now I have a bunch of problems that come first.[10:41:44] <Skipp_OSX> tqh: I want to rework the status tray to use the layout system[10:42:05] <tqh> ofc. But take a long look at the real problem before adding workarounds.[10:42:11] <Skipp_OSX> tqh: right now it is too inflexable for my needs.[10:42:40] <tqh> It looks nice.[10:43:24] <Skipp_OSX> tqh: It looks nice now, but, make the icons 64x64 in horizontal mode and then it looks terrible[10:44:01] <jrabbit> Skipp_OSX: I can't author HVIF in other programs though[10:44:49] *** kPb_in has joined #haiku[10:45:18] <jrabbit> thats the main issue[10:45:33] <Duggan> DRI sucks worse than I thought...[10:45:59] <jrabbit> god this trial is a cluster[10:46:06] <Skipp_OSX> tqh: well, and there are other problems too. I need to make the icons shrink in case you make them big and open a bunch of applications for instance.[10:46:42] <Skipp_OSX> tqh: Otherwise they will just be positioned below the screen (in vertical mode) and that really doesn't work.[10:47:24] <Skipp_OSX> I guess you can run into the same problem at 16x16 if you have enough apps open. If you use expando view it becomes a problem pretty quickly actually.[10:47:25] <Duggan> jrabbit trial?[10:47:47] <jrabbit> Mubarak's trial opening arraignment[10:47:58] <Skipp_OSX> But that is a problem for another day[10:48:17] <tqh> Skipp_OSX, there are always a lot of challanges. Btw sounds like that constraint programming could be used for that kind of problem. Might want to ask cziedler a bit about that.[10:48:19] <Duggan> interesting... I assume he's being tried en absentia?[10:48:28] <Skipp_OSX> good night all[10:48:32] <Duggan> or whatever those stupid latin phrases are...[10:48:34] <Duggan> l8r Skipp_OSX[10:48:36] <Duggan> congrats again[10:48:43] *** Skipp_OSX has quit IRC[10:48:52] <Duggan> btw jrabbit yes, I've been so busy coding I haven't even turned on the news in weeks...[10:49:04] <jrabbit> heh :)[10:49:33] <jrabbit> Duggan: you're working with 3D stuff right?[10:49:37] <Duggan> yep[10:49:42] <Duggan> well, trying to hehe[10:50:39] <jrabbit> it could be really beneficial, I just learned that iOS uses openGL heavily for its rendering via Apple's coreimage/*[10:51:08] <Duggan> it could be really beneficial for games :P[10:51:08] <Duggan> hehe[10:51:59] <jrabbit> games don't need no stinkin' 3D :P[10:52:11] <jrabbit> but yeah[10:52:25] <Duggan> bah, games are the future of Haiku :P[10:53:33] <Duggan> don't argue with me, I'm a visionary and I'll prove it![10:53:36] <Duggan> ... in 20 years...[10:55:59] * jrabbit goes to play some 3d games just to be a hypocrite[11:06:51] *** Ingenu has joined #haiku[11:10:39] <OmniMancer> openGL makes a pretty good cross platform GUI :P[11:19:59] <Teknomancer> so does SDL :P[11:20:02] <Teknomancer> Morning all[11:20:33] <OmniMancer> yea[11:20:49] <OmniMancer> but OpenGL is probably available in slightly more places[11:20:59] <OmniMancer> even if it requires different init methods in all of them[11:24:55] <Duggan> hey Teknomancer[11:25:09] <Teknomancer> hi Duggan[11:29:05] * Duggan sighs.[11:29:20] <Duggan> DRI is really going to suck...[11:31:03] <Duggan> so far I've got about 4000 lines worth of errors... meh, go figure[11:32:07] *** lorenzo5930 has quit IRC[11:39:47] *** ab5tract has joined #haiku[11:44:53] *** lorenzo5930 has joined #haiku[11:51:29] *** tqh has quit IRC[12:00:33] *** ericxjo has joined #haiku[12:01:01] *** tqh has joined #haiku[12:28:31] *** mmlr_work has joined #haiku[12:31:32] *** mmlr_work is now known as mmlr_away[12:31:52] *** LiquidInsect has quit IRC[12:35:53] *** negusnyul has joined #haiku[12:36:33] *** ericxjo has quit IRC[12:37:48] *** Nozy has joined #haiku[12:37:56] *** smop has quit IRC[12:44:38] *** smop has joined #haiku[12:57:42] *** Nozy has quit IRC[13:00:24] *** negusnyul has quit IRC[13:02:26] *** Animal-X has quit IRC[13:02:39] *** brianb_ has joined #haiku[13:05:13] *** brianb has quit IRC[13:07:57] *** ab5tract has quit IRC[13:09:28] *** zgaga has joined #haiku[13:11:46] *** negusnyul has joined #haiku[13:14:12] *** Animal-X has joined #haiku[13:14:51] *** tqh has quit IRC[13:17:58] <negusnyul> scgtrp: tried that screen resolution trick you told me yesterday, no success sadly :( actually, nothing happened, in the boot menu, I only see 800x600 and 1024x768 modes... maybe it's a bios problem[13:20:14] *** Barrett has joined #haiku[13:21:32] *** ab5tract has joined #haiku[13:21:39] <negusnyul> I tried booting on my desktop too, but it just restarts the computer (very early in the boot process, only the Loading... text appears then reboot), any ideas how can I debug/fix this?[13:30:47] *** tqh has joined #haiku[13:32:05] *** negusnyul has quit IRC[13:34:55] *** survery has joined #haiku[13:34:59] *** survery has left #haiku[13:37:01] *** negusnyul has joined #haiku[13:42:41] *** jjido has joined #haiku[13:43:20] *** jjido has quit IRC[13:51:50] *** swiftos has joined #haiku[13:54:07] *** Barrett has quit IRC[13:54:59] *** zgaga has quit IRC[14:06:05] *** Cr0wb4r has joined #haiku[14:06:31] *** volci has joined #haiku[14:12:14] *** Cr0wb4r has quit IRC[14:40:40] * Teknomancer goes to burn a Haiku live CD[14:43:40] <Teknomancer> btw, whats the way to check if haiku is using VESA drivers or using native radeon drivers?[14:43:44] <Teknomancer> anything like lsmod?[14:49:52] <OmniMancer> opening the screen preflet and hovering over the monitor should say[14:50:17] <OmniMancer> otherwise I think you can listimage the appserver and grep for accelerants[14:51:00] <Teknomancer> cool, thanks[14:51:27] <Ingenu> omg old people from the BeOS days talking ![14:51:39] <Ingenu> well ten years older at least ;p[14:51:50] <Teknomancer> :)[14:52:02] <Ingenu> not tried latest Haiku[14:52:07] <Teknomancer> me neither[14:52:10] <Teknomancer> i'm burning a live CD[14:52:19] <Teknomancer> i plan to hit the stores sometime to buy a new laptop[14:52:32] <Teknomancer> i want it to be able to run Haiku decently[14:52:44] <Ingenu> last time I thought: it's slick, it's fast, it's responsive... back to windows to code and play games... ^^[14:53:00] <Teknomancer> play games is true :)[14:53:20] <Ingenu> I also like Visual Studio, used to it since I've been using it for 12 years now[14:54:00] <Ingenu> although now I sometimes use the command line (heresy ! GUI for the win !)[14:54:01] <Ingenu> ;)[14:56:45] <OmniMancer> the commandline is useful[14:56:57] <OmniMancer> it is those who wish to use the command line for everything[14:57:09] <OmniMancer> or the GUI for everything that lose out[14:57:17] *** oxoocoffee has joined #haiku[14:58:06] *** oxoocoffee has quit IRC[15:04:11] *** brrant has joined #haiku[15:04:57] <Teknomancer> hmm what's the SVN revision for Haiku R1 alpha3[15:06:33] *** Nozy has joined #haiku[15:07:29] <Teknomancer> nvm[15:11:21] *** Nozy has quit IRC[15:11:22] <Teknomancer> cd burnt :)[15:12:53] *** cherrypai has quit IRC[15:13:27] *** adominguez has quit IRC[15:13:50] *** mmu_man has joined #haiku[15:14:31] *** cherrypie has joined #haiku[15:15:54] *** jjido has joined #haiku[15:18:50] *** Barrett has joined #haiku[15:33:44] *** Barrett has quit IRC[15:37:17] *** phoudoin has joined #haiku[15:37:17] *** ChanServ sets mode: +o phoudoin[15:37:23] <phoudoin> hi[15:40:05] *** zgaga has joined #haiku[16:00:13] *** jjido has quit IRC[16:00:43] *** OmniMancer has quit IRC[16:03:49] <mmu_man> plop[16:06:06] *** judgen has joined #haiku[16:08:33] *** MatthewH12 has joined #haiku[16:27:52] *** swiftos has quit IRC[16:28:06] *** kPb_in has quit IRC[16:33:10] *** The123king has joined #haiku[16:38:08] *** tqh2 has joined #haiku[16:45:35] *** dr_evil has joined #haiku[16:49:24] *** dr_evil_ has quit IRC[16:49:28] *** Tapionp has joined #haiku[16:59:46] *** papasmurf has joined #haiku[17:02:35] *** volci has quit IRC[17:18:37] *** The123king has quit IRC[17:21:24] *** XenoDuck has joined #haiku[17:22:50] *** MatthewH12 has quit IRC[17:23:59] *** tqh has quit IRC[17:24:00] *** tqh2 has quit IRC[17:32:51] *** Guest79521 has joined #haiku[17:33:57] *** Guest79521 is now known as Colin_Finck[17:34:03] *** Colin_Finck has joined #haiku[17:35:29] <binkySMC> hello[17:36:47] *** humdinger has joined #haiku[17:38:43] *** tqh2 has joined #haiku[17:40:41] *** _taos_ has joined #haiku[17:46:49] *** kPb_in has joined #haiku[17:49:05] *** Tapionp has quit IRC[17:58:34] *** kPb_in has quit IRC[18:05:01] *** jmelesky has joined #haiku[18:11:28] *** Malmis has quit IRC[18:13:03] <helf> anyone wanna buy a sprint epic 4g?[18:13:57] *** NickDANEjer has joined #haiku[18:13:58] <NickDANEjer> boink[18:18:01] *** HaikuUser has joined #haiku[18:18:16] *** HaikuUser has quit IRC[18:18:26] <binkySMC> how much for the phone ?[18:20:00] *** PasNox has joined #haiku[18:24:53] <helf> i dunno[18:24:55] <helf> $100?[18:24:59] <helf> its the phone, a charger[18:25:03] <helf> i think i might have the box it came in[18:25:09] <helf> its a little worn but not bad[18:25:31] <helf> has one of those god awful pentil samoled screens people like for some reason[18:26:17] *** humdinger has quit IRC[18:26:49] *** Barrett has joined #haiku[18:26:50] <binkySMC> link ?[18:27:11] <helf> to what? the actual phone?[18:27:15] <helf> I havent uploaded any photos yet[18:27:18] <helf> Ill do it tonight[18:27:44] <helf> my dad used it for awhile before switching phones. Its been sitting on my desk for ages.[18:30:25] <binkySMC> just a link to which model it is[18:35:24] *** kPb_in has joined #haiku[18:35:27] *** Skipp_OSX has joined #haiku[18:37:29] *** idefix_xifedi has joined #haiku[18:39:12] <Skipp_OSX> Duggan?[18:40:08] *** evildaemon has quit IRC[18:42:29] <zgaga> many links for software on bebits.com are dead[18:42:53] <helf> yeah, thats why I always hated sites that didnt locally host files[18:44:57] *** Barrett has quit IRC[18:54:53] *** Barrett has joined #haiku[18:57:07] *** Malmis has joined #haiku[18:57:24] *** Colin_Finck is now known as Colin_Finck|lap[18:59:05] *** PulkoMandy has joined #haiku[18:59:25] <binkySMC> most of the stuff from bebits is on haikuware, at least the stuff that was available was anyways[19:01:31] <PulkoMandy> bebits still has the data about unavailable stuff[19:01:45] <binkySMC> thats true[19:01:47] <PulkoMandy> author names, mail addresses and so on may be an invaluable resource to get some programs back[19:01:56] <PulkoMandy> (and yes, it already worked several times)[19:02:02] <binkySMC> question is, how many are worth bringing back ?[19:02:27] <PulkoMandy> well, porting an app from BeOS, even an R3 one, is much easier than writing from scratch[19:02:35] <PulkoMandy> that is, for the one that can be opensourced[19:02:54] <binkySMC> I wonder how many of the opportunitys are left at this time ?[19:03:00] <binkySMC> I agree, definately easier[19:03:07] <PulkoMandy> but the response is usually positive, if you manage to get in touch with the author[19:03:20] <PulkoMandy> well, recovered Sawteeth sourcecode just one year ago[19:03:29] <PulkoMandy> along with all the code from his original authors[19:04:00] <PulkoMandy> could work for other projects as well, just costs a day or so tracking the author, for month or years they spent writing the software[19:05:24] *** _taos_ has quit IRC[19:06:08] *** mmlr_away has quit IRC[19:07:13] *** aksr has joined #haiku[19:08:24] *** tqh has joined #haiku[19:10:21] * GeneralMaximus wishes someone could track the author of BeCommander[19:10:34] <GeneralMaximus> what a fine piece of work[19:11:32] *** DaaT has joined #haiku[19:15:24] <binkySMC> BeCommander ?[19:16:00] <GeneralMaximus> a midnight commander clone for BeOS[19:16:10] <GeneralMaximus> pretty neat.[19:16:36] *** brobostigon has quit IRC[19:16:47] *** brobostigon has joined #haiku[19:16:48] <binkySMC> what does it do ? file manager ?[19:16:54] <GeneralMaximus> aye.[19:17:01] <GeneralMaximus> it's still available on Haikuware.[19:17:08] <GeneralMaximus> but only the binaries[19:18:51] *** judgen has quit IRC[19:19:14] *** NickDANEjer has quit IRC[19:19:32] <Skipp_OSX> well, there is Seeker...[19:19:41] <zgaga> does haikuware.com have apps for downloads stored locally?[19:19:56] <binkySMC> haikuware keeps pretty much everything on their own server IIRC[19:20:13] <PulkoMandy> ... including binaries[19:20:21] <PulkoMandy> but their backup plan seems to work so far[19:21:48] *** aksr has quit IRC[19:22:21] <GeneralMaximus> google is being too clever for its own good these days. i looked up "seeker BeOS" and it returned results for "seeker bears". no "did you mean?" either.[19:22:44] <GeneralMaximus> oh no, there _is_ a did you mean[19:22:46] <GeneralMaximus> phew[19:23:31] <Skipp_OSX> GeneralMaximus: http://darkwyrm.beemulated.net/downloads.htm[19:23:33] <binkySMC> I m curious why someone would hack haikuware, was it malicous intent or just a do it to do it thing ?[19:23:57] <binkySMC> yeah, when is be-emulated comming back online ! anyone know ?[19:24:04] <PulkoMandy> GeneralMaximus: you need to put each word between quotes these days[19:24:38] <PulkoMandy> binkySMC: was a well known hacking-script AFAICT[19:24:45] <PulkoMandy> so just an easy to hack website[19:26:03] <GeneralMaximus> Seeker looks sweet.[19:26:11] <binkySMC> just curious if it was a boredom thing, or a malicous thing Ie " I hate BEOS so I'll hack haikuware" or " lets hack a site becuase I am bored and a spoiled manchild"[19:26:18] <GeneralMaximus> PulkoMandy: yes. especiall with programming queries.[19:26:23] *** phoudoin has quit IRC[19:26:42] <Skipp_OSX> binkySMC: well, the site being hacked angered you, so, mission accomplished for the hacker. I guess that is reason enough.[19:26:46] <GeneralMaximus> binkySMC: i'd say it was just script kiddies being script kiddies.[19:26:54] <GeneralMaximus> binkySMC: for the lulz.[19:27:22] <binkySMC> thats what I was wondering, well what angers me is someone disrespecting another persons hardwork and shtting on it, ya, it does anger me, it should anger everyone[19:28:46] <PulkoMandy> was just a nice occasion to test the backup plan :)[19:30:34] *** aksr has joined #haiku[19:31:15] *** johndrinkwater has joined #haiku[19:43:28] *** kPb_in has quit IRC[19:43:40] *** Colin_Finck|lap has quit IRC[19:48:06] *** zgaga has quit IRC[19:53:32] <binkySMC> well, the back up plan does work obviously 1[19:56:40] <Skipp_OSX> binkySMC: In the mind of the cracker they are actually helping the person by pointing out security flaws. Silly I know, but that is the mindset.[19:58:18] *** Barrett has quit IRC[20:00:10] *** AmineKhaldi has joined #haiku[20:03:04] *** AmineKhaldi_ has quit IRC[20:06:39] *** zgaga has joined #haiku[20:07:44] <binkySMC> no, it's no different then spray painting a wall.[20:08:53] <PulkoMandy> Skipp_OSX: well, depends on the people[20:09:19] <PulkoMandy> some people actually point the security flaws, when they get in, but these usually don't erase the website[20:09:46] <binkySMC> well mor elike, you send the admin a email like, hey you have a security hole here ---->[20:09:56] <binkySMC> this is just plain malicous vandalism[20:19:04] <CIA-73> Haiku: kallisti5 * r42554 /haiku/trunk/ (8 files in 3 dirs) http://dev.haiku-os.org/changeset/42554 :[20:19:04] <CIA-73> * move obtaining / copying the vga bios into the driver.[20:19:04] <CIA-73> * add missing r500 header[20:19:04] <CIA-73> * replace r600 headers with newer one from kernel[20:19:38] <DaaT> go go kallisti5!![20:21:43] <zgaga> what's wrong with KWord? When I click on any menu it doesn't do anything?[20:22:30] *** Pinaraf has joined #haiku[20:23:26] <zgaga> Menus are normally drawn, but clicking on them does nothing[20:27:56] *** luroh has joined #haiku[20:37:32] <dr_evil> I need your help to legally file in for a claim worth £37,850,000 where you will part with 30% of the total sum at the end of claim[20:37:39] <Duggan> hey Skipp_OSX what's up?[20:38:18] <dr_evil> and no one else to ask than a random stranger. those spammers suck[20:38:38] *** lorenzo5930 has quit IRC[20:39:27] <augiedoggie> and they all seem to think I suffer from erectile dysfunction :/[20:39:52] <Skipp_OSX> hey Duggan, I updated my patch a little bit more. I went to bed and I thought about what would happen if for some crazy reason UpdatePlacement() never got called. So, I put it back in the BarView constructor. This means that it will get called 2x on startup potentially. But, I think that is better than it potentially getting called 0 times.[20:39:57] * luroh remembers the good old days when they spammed via fax[20:40:00] <Duggan> augiedoggie you dont!?[20:40:26] <Duggan> Skipp_OSX under what condition would it not be called at all?[20:40:40] <Skipp_OSX> Duggan: if the window could not be locked for some crazy reason.[20:41:09] <Duggan> alright[20:41:19] <Skipp_OSX> Duggan: if UpdatePlacement() is not called Deskbar will be running but won't draw so you would be completely screwed. That to me is unacceptable and it really distressed me.[20:41:43] <Duggan> it shouldn't, I don't think there will be a problem locking the window[20:42:04] <Skipp_OSX> So, I decided that the tiny performance hit is much preferrable to a 0.0000001% chance that Deskbar wouldn't get drawn.[20:42:15] <PulkoMandy> the locking wait if the window is already locked[20:42:27] <PulkoMandy> it resumes either if the window is free, or it is closed[20:42:36] <PulkoMandy> in the first case execution will continue as expected[20:42:43] <PulkoMandy> in the other case you're screwed anyway[20:42:55] <Duggan> I personally also think recreating the expanded list every time it's updated is bad...keeping a running list would probably be better[20:43:29] <Duggan> but... your project :) I'm just glad you fixed that bug hehe[20:44:10] *** brechtm has joined #haiku[20:44:19] <Skipp_OSX> PulkoMandy: well, the thing is, if it doesn't lock for some reason. Deskbar doesn't get drawn if I didn't call UpdatePlacement() already which is fatal. But, if I call UpdatePlacement() the lock and call it again and it fails to lock, it just doesn't resize Deskbar which is much less of a problem.[20:44:58] <Skipp_OSX> Duggan: I agree that it is bad but that problem is much bigger than just this patch.[20:45:13] <PulkoMandy> well if you don't need the lock... do it before or after :)[20:45:25] <PulkoMandy> anyway, it will be an improvement, I guess[20:46:52] <Skipp_OSX> PulkoMandy: I don't think I am describing the problem very well.[20:47:36] <PulkoMandy> :)[20:48:00] <Skipp_OSX> PulkoMandy: Lets put it this way. I fixed all the real problems, now I am fretting over potential problems.[20:49:26] *** zgaga has quit IRC[20:50:17] <Skipp_OSX> if something really bad happens try my hardest to fail gracefully instead of failing spectacularly[20:54:05] <Skipp_OSX> I also replaced C style casts with static_cast so that on failure it doesn't fallback onto a reinterpret_cast which if the stars are aligned a certain way and your computer gets hit by a gamma ray at just thr right time might cause a serious problem.[20:56:06] <Duggan> well commit your stuff or submit a patch, I'm sure if somebody doesn't like the way you're doing something, it won't take long to get changed :) hehe[20:56:19] <Duggan> at least then, any problems won't be on you[20:57:51] *** lollo64it has joined #haiku[20:58:23] <Skipp_OSX> Duggan: I did submit a patch: http://dev.haiku-os.org/ticket/7051[20:58:35] <Duggan> ah ok :)[21:00:43] *** lollo64it has quit IRC[21:01:51] *** lollo64it has joined #haiku[21:06:23] *** bjw has quit IRC[21:06:49] *** bjw has joined #haiku[21:06:59] *** MatthewH12 has joined #haiku[21:12:45] <Skipp_OSX> what is the command to set a decorator?[21:13:31] <scgtrp> setdecor iirc[21:13:40] <scgtrp> i'm not sure if it's working atm[21:14:34] <Skipp_OSX> scgtrp: apparently all the decorators have been removed in trunk[21:25:34] <Ingenu> decorators, those are the things that make the OS looks like another OS right ?[21:25:41] <Ingenu> (well not only but where used for)[21:25:53] <Skipp_OSX> Ingenu: precisely[21:32:32] *** zgaga has joined #haiku[21:32:44] *** zgaga has quit IRC[21:33:17] *** hrga has joined #haiku[21:37:02] *** hrga has quit IRC[21:45:43] *** DaaT has quit IRC[21:48:56] <binkySMC> Nice patch Skippy_osx[21:49:09] <binkySMC> setdecor xxxxxxxxxxx[21:49:21] <binkySMC> also I thought a gui was put into decorators ?[21:49:59] <binkySMC> its under appaearance in prefrences as well[21:58:54] <Skipp_OSX> binkySMC: All decorators but default have been broken and removed in the latest builds so, no decorators for now.[21:59:21] <Skipp_OSX> binkySMC: Thanks for the praise. I hope that the devs agree.[22:00:23] *** Barrett has joined #haiku[22:02:16] *** ab5tract has quit IRC[22:02:23] <binkySMC> I know the decorators are broken, but the preflet is there to select them now, hopefully thye get fixed soon 1[22:02:24] <binkySMC> ![22:02:25] *** Hery has joined #haiku[22:05:12] <Skipp_OSX> binkySMC: I think someone is going to have to champion that cause before they will come back.[22:05:18] <saivert> Isn't stack and tile part of that?[22:05:45] <Skipp_OSX> saivert: S&T has been integrated into app_server. It is now included by default. And it is much improved as well.[22:06:53] *** idefix_xifedi has left #haiku[22:07:02] <Duggan> Skipp_OSX is that integration completed?[22:08:00] *** tqh2 has quit IRC[22:08:42] *** MAX2 has joined #haiku[22:09:37] *** NeonLicht has quit IRC[22:09:38] <binkySMC> what would be required to bring the decorators back ? SAT is a window manager feature, not a decorator ?[22:11:19] <saivert> binkySMC: yes but previously it was implemented as a decorator, but apparently it is now part of app_server instead. so it should work across decorators (hopefullY9[22:13:45] *** ddavid123 has joined #haiku[22:16:14] <Skipp_OSX> correct, S&T is now available to all decorators. Although getting that to work for other decorators would still take some work. Probably not a lot of work in the case of the BeDecorator, but for Mac and Win, something new would need to be done since those decorators are not tabs.[22:16:59] <Skipp_OSX> All in all it is not a big loss though. S&T is much cooler than looking like a Mac from 1999 IMHO[22:18:10] <ddavid123> perhaps new decorators can be developed and integrated into the app server to give Haiku a Compizesqe feel?[22:18:57] <ddavid123> But without the cube thing[22:19:02] <ddavid123> Compiz[22:19:17] <ddavid123> It is a joke[22:19:18] <Duggan> something they should have implemented to begin with is vertically stacking tabs :/[22:20:03] <ddavid123> perhaps some animations when switching virtual desktops?[22:20:26] <ddavid123> to smooth the transition[22:20:41] <Skipp_OSX> ddavid123: lots of animations would be nice to have later on[22:20:42] *** tqh2 has joined #haiku[22:20:52] <ddavid123> Yes, later on[22:20:58] <Duggan> much later on...[22:20:59] <ddavid123> Perhaps R2[22:21:12] <Skipp_OSX> ddavid123: There is some interest in a compositing app_server, some people have even claimed to have started working on it.[22:21:30] <ddavid123> I am interested![22:21:38] <binkySMC> whats the benefit of a compositing app server ???[22:22:01] <Duggan> a truly 3d desktop would be nice... but obviously not until 3d hw accel is implemented...[22:22:03] <ddavid123> in my opinion that is where it should be if there is to be one[22:23:16] <ddavid123> It would be great if the new open source NVidia and ATI drivers were ported to Haiku![22:23:31] <Duggan> nvidia released open source drivers?[22:23:32] <binkySMC> what drivers would those be ?[22:23:36] <Skipp_OSX> binkySMC: animations and shadows and the like[22:23:38] <binkySMC> gallium ?[22:23:42] <ddavid123> yes[22:23:50] <Duggan> you mean intel right?[22:24:04] <ddavid123> Nvidia and ATI[22:24:10] <binkySMC> its not really a Drive like your thinking, you still have to setup th ehardware using the system drivers.[22:24:34] <ddavid123> I don't remember what they are named, but there are people writing the drivers from scratch[22:24:54] <ddavid123> Nouveau[22:25:00] <ddavid123> perhaps[22:25:10] <ddavid123> sounds fammiliar[22:25:11] <Duggan> if people are writing the drivers from scratch then they weren't released by the vendor.... ATI and Intel released open source drivers, NVidia has not (that I'm aware of)[22:26:04] <ddavid123> NVidia has not released open source drivers, right, the only open source NVidia drivers are 3rd party[22:26:09] <ddavid123> community[22:26:38] <binkySMC> haiku can't support animations, shadows now ? strange[22:27:00] <Skipp_OSX> but of course the open source drivers are Linux and Xorg drivers so they do not port over to Haiku easily.[22:27:14] <ddavid123> true[22:27:23] <Duggan> Mesa is already here, DRM / DRI / Gallium are not[22:27:31] <ddavid123> but it could be done[22:27:44] <Duggan> only with a great deal of effort...[22:27:50] <Skipp_OSX> Duggan: well, as I understand even the Gallium3d drivers are very Linux-centric at this point[22:28:11] <ddavid123> Only if the people thought such things were worth the effort[22:28:28] <Duggan> ddavid123 I'm looking into DRM/DRI/Gallium at present[22:28:28] <Skipp_OSX> Give it a few years and you may see some nice 3d drivers in Haiku, but I wouldn't expect them anytime soon.[22:28:38] <ddavid123> ok[22:28:44] <ddavid123> Duggan[22:28:54] <Duggan> I don't know if I'm going to work on porting them, but I'm weighing porting them vs borrowing some code for our drivers / accelerants[22:28:57] <Skipp_OSX> Duggan: speaking of that, are you aware of the Gallium3d branch?[22:29:07] <Duggan> Skipp_OSX yes[22:29:12] <Skipp_OSX> Duggan: okay good.[22:29:24] <Duggan> I tried merging the current trunk with it.... it was hell... I gave up :P[22:29:36] <Duggan> it's not documented well either.... only the commit logs[22:29:42] <Skipp_OSX> Duggan: yeah, it is quite old at this point too[22:29:51] <Duggan> very much so[22:30:34] <Skipp_OSX> Duggan: I am sure it would be a lot of work to integrate, and then a bunch more work to incorporate the latest Gallium stuff, and a whole bunch more work to make everything work, and a whole bunch more work to integrate it properly[22:30:37] *** mmadia has joined #haiku[22:31:13] <Skipp_OSX> Duggan: in other words. It is a whole bunch of work :) But probably faster than starting from scratch[22:32:16] <Duggan> you'd be surprised :P[22:33:17] <binkySMC> I'm with duggan, gallium is a design for linux and its goofy graphics architecture.[22:33:22] *** Barrett has quit IRC[22:35:29] <Duggan> I'm not a big fan of the linux graphics stack... it's notoriously slow[22:35:38] *** jjido has joined #haiku[22:35:44] <Duggan> but if porting it is what gets Haiku 3d hw acceleration, I guess it's worth it....[22:35:50] <Duggan> ... we can always fork it and optimize later :P[22:36:11] <binkySMC> I think that'd be more work though[22:36:27] <Duggan> kind of like how we forked mesa... I don't see any of our port of mesa being pushed upstream anytime soon[22:36:56] <Duggan> binkySMC in the long run no, as we could still keep the compatibility with the drivers so we still get free driver upgrades in the future[22:37:47] <Duggan> unless something inherant in the architecture of the stack makes it inefficient, then we would simply have to fix it and keep track of how to keep fixing it in future versions (or just fork outright and forget about upgrading from their trunk)[22:37:58] <Duggan> but that would make it more difficult to get better drivers in the future[22:38:15] <binkySMC> ehh, I am no c++ or programming expert by any stretch, the best parts and the most likely to be optimized parts of gallium are the llvm jit and the state trackers. everything else has to be done with the native driver anyways. so why bother porting the linux cruft ?[22:38:47] <negusnyul> ahh 3d drivers... I think it would be better to have proper 2d drivers for newer cards first :)[22:39:20] <Duggan> negusnyul the only benefit of 2d hw accel is the blitter[22:39:34] <CIA-73> Haiku: kallisti5 * r42555 /haiku/trunk/headers/os/drivers/PCI.h http://dev.haiku-os.org/changeset/42555 : * define PCI add-on rom flags that are normally defined in PCI header files.[22:40:23] <negusnyul> I mean working modesetting for example[22:40:47] <negusnyul> to be able to use all supported resolutions[22:40:52] <Duggan> I don't guess I understand what you mean... I can change graphics modes fine[22:41:21] <binkySMC> gallium won't work without the 2d drivers to acess the cards hardware register, clock speed etc.[22:41:43] <binkySMC> I can link the email I got from one of the gallium devs a while back[22:41:52] <binkySMC> have to do it later when I get home though[22:42:06] <negusnyul> Duggan: for example, I cannot use widescreen resolutions on my laptop[22:42:15] <Duggan> ah[22:42:39] <Duggan> binkySMC that's nothing to do with 2d/3d, that's just transferring information to the card[22:43:29] <negusnyul> also, anybody has some tips how to debug/fix a very-early-in-the-boot-process crash?[22:43:59] <negusnyul> I only get the Loading... text then my computer reboots[22:44:13] <negusnyul> I cannot even access the boot menu by pressing space/shift[22:46:57] *** jjido has quit IRC[22:47:31] <binkySMC> if you can't set clockspeeds,powersettings,screenresolution etc and you can't command the card to read from the ring buffer 3d drivers are useless, you can compile the code but it won't get there and if the card isn't put into 3d mode it won't process the code.[22:49:09] *** Barrett has joined #haiku[22:49:22] <CIA-73> Haiku: kallisti5 * r42556 /haiku/trunk/src/add-ons/kernel/drivers/graphics/radeon_hd/radeon_hd.cpp http://dev.haiku-os.org/changeset/42556 :[22:49:22] <CIA-73> * use new PCI.h rom flags[22:49:22] <CIA-73> * clean up tracing[22:49:22] <CIA-73> * remove infinite loop :)[22:51:03] *** ericxjo has joined #haiku[22:51:17] <CIA-73> Haiku: korli * r42557 /haiku/trunk/src/kits/interface/Window.cpp http://dev.haiku-os.org/changeset/42557 : Patch from X512 (#7408): only send input method aware messages for active windows[22:52:50] *** NeonLicht has joined #haiku[23:00:12] *** ddavid123 has quit IRC[23:02:59] *** tqh2 has quit IRC[23:05:14] *** tqh2 has joined #haiku[23:06:20] *** tqh2 has quit IRC[23:07:28] *** brechtm has quit IRC[23:08:58] *** Qeos|2 has joined #haiku[23:09:21] *** freakazoid0223 has quit IRC[23:11:08] *** tqh2 has joined #haiku[23:11:28] *** Qeos has quit IRC[23:11:31] *** Barrett has quit IRC[23:11:50] *** freakazoid0223 has joined #haiku[23:12:22] *** NeonLicht has quit IRC[23:13:14] <Duggan> this is really going to suck :/[23:15:05] *** stef_ has joined #haiku[23:15:45] *** stef has quit IRC[23:19:13] *** volcimaster has joined #haiku[23:20:01] *** NeonLicht has joined #haiku[23:20:22] <binkySMC> heh ?[23:21:18] <Duggan> interesting, it would appear there are no gallium channels on freenode...[23:21:42] <binkySMC> oh,that would suck[23:22:18] <Duggan> according to irc.netsplit.de 's channel search, "gallium" returns 0 results[23:22:19] *** tqh2 has quit IRC[23:22:46] <binkySMC> I'll check hang on[23:22:47] *** papasmurf has quit IRC[23:23:03] <Duggan> (querying the channel list in Vision is insanely slow)[23:23:15] *** _arjen_ has joined #haiku[23:23:30] *** volcimaster has quit IRC[23:24:46] <binkySMC> I just ran through the channel lists, I didn;t see any gallium channels[23:25:03] <Duggan> #dri-devel covers everything BUT gallium[23:25:04] <binkySMC> I don't think that many people are actually working on it in reality[23:26:21] <Duggan> maybe I should give BSD another chance and help...[23:27:19] <binkySMC> ??????[23:27:28] <Skipp_OSX> check mesa[23:27:33] <binkySMC> doh ![23:27:40] *** tqh2 has joined #haiku[23:28:02] <Duggan> good point, it is part of mesa now, but I'd think they'd have some reference to it where the search would have returned something :/[23:28:18] <Duggan> gotta run for a bit... back soon[23:29:22] <binkySMC> ok[23:31:17] <binkySMC> #nouveau[23:31:24] <binkySMC> thats the gallium3d channel[23:34:07] <binkySMC> #radeon[23:34:22] <binkySMC> #radeonhd[23:36:59] *** nsuperbus has quit IRC[23:42:48] *** luroh has quit IRC[23:44:57] *** Barrett has joined #haiku[23:45:24] <CIA-73> Haiku: kallisti5 * r42558 /haiku/trunk/src/add-ons/kernel/drivers/graphics/radeon_hd/ (radeon_hd.cpp radeon_hd_private.h) http://dev.haiku-os.org/changeset/42558 :[23:45:24] <CIA-73> * don't trample PCI rom config[23:45:24] <CIA-73> * better error checking[23:45:24] <CIA-73> * the driver can now locate the AtomBIOS on real hw![23:47:21] *** OmniMancer has joined #haiku[23:48:29] <binkySMC> look at Kallisti5 go ![23:52:17] *** oxoocoffee has joined #haiku[23:54:57] *** oxoocoffee has quit IRC[23:55:34] *** NeonLicht has quit IRC[23:56:38] *** NeonLicht has joined #haiku[23:58:30] *** OmniMancer has quit IRC