channel logs for 2004 - 2010 are archived at http://tunes.org/~nef/logs/old/ ·· can't be searched
#osdev2 = #osdev @ Libera from 23may2021 to present
#osdev @ OPN/FreeNode from 3apr2001 to 23may2021
all other channels are on OPN/FreeNode from 2004 to present
http://bespin.org/~qz/search/?view=1&c=osdev&y=18&m=5&d=9
00:00:00 --- log: started osdev/18.05.09
00:00:12 --- join: grouse (~grouse@83-233-9-2.cust.bredband2.com) joined #osdev
00:05:49 --- join: elevated (~elevated@unaffiliated/elevated) joined #osdev
00:07:55 --- quit: sixand (Ping timeout: 264 seconds)
00:10:54 --- join: zeus1 (~zeus@197.239.1.54) joined #osdev
00:18:30 --- join: godel (~gonzalo@190.192.95.122) joined #osdev
00:21:31 --- join: sixand (~Thunderbi@175.44.160.12) joined #osdev
00:23:05 --- quit: jakogut_ (Quit: jakogut_)
00:24:12 --- quit: dbittman (Ping timeout: 246 seconds)
00:44:09 --- quit: vdamewood (Ping timeout: 246 seconds)
00:45:45 --- join: SGautam (~GautamS@unaffiliated/gautams) joined #osdev
00:47:21 --- join: kimundi (~Kimundi@dynip-129-217-079-084.wifi.tu-dortmund.de) joined #osdev
00:54:50 --- quit: quc (Remote host closed the connection)
00:55:02 --- join: quc (~quc@87.116.229.85) joined #osdev
00:57:35 --- quit: bemeurer (Quit: WeeChat 2.1)
00:57:47 --- quit: sixand (Quit: sixand)
00:58:16 --- quit: zeus1 (Ping timeout: 256 seconds)
01:02:28 --- quit: behalebabo (Excess Flood)
01:02:52 --- join: behalebabo (~behalebab@unaffiliated/behalebabo) joined #osdev
01:05:46 --- quit: godel (Ping timeout: 250 seconds)
01:16:19 --- quit: kimundi (Ping timeout: 240 seconds)
01:16:45 --- join: kimundi (~Kimundi@dynip-129-217-079-084.wifi.tu-dortmund.de) joined #osdev
01:16:58 --- quit: Pseudonym73 (Remote host closed the connection)
01:24:17 --- quit: oaken-source (Quit: Lost terminal)
01:24:45 <klange> let the adventure begin https://gist.github.com/klange/7701981001ef5f974c9cfe84d9aa46b4
01:24:47 <bslsk05> gist.github.com: things marked undeclared when building python for toaru-nih · GitHub
01:25:08 --- join: oaken-source (~oaken-sou@141.89.226.146) joined #osdev
01:26:00 <Mutabah> Not a lot
01:26:15 <klange> mostly math, a bit of wide characters, and some locale
01:26:22 <Mutabah> and snprintf?
01:27:04 <klange> yeah, seems I never did get around to adding the bounded version of that
01:28:21 --- quit: doug16k (Remote host closed the connection)
01:29:13 --- join: spare (~user@unaffiliated/spareproject) joined #osdev
01:30:54 --- quit: MarchHare (Ping timeout: 250 seconds)
01:32:54 --- join: doug16k (~dougx@198-91-135-200.cpe.distributel.net) joined #osdev
01:34:11 --- quit: ACE_Recliner (Remote host closed the connection)
03:40:16 --- quit: clog (^C)
03:40:16 --- log: stopped osdev/18.05.09
03:40:27 --- log: started osdev/18.05.09
03:40:27 --- join: clog (~nef@bespin.org) joined #osdev
03:40:27 --- topic: 'Operating System Development || Don't ask to ask---just ask! || For 3+ LoC, use a pastebin (for example https://gist.github.com/) || Channel logs: http://osdev-logs.qzx.com || Today's log: http://osdev-now.qzx.com || Visit http://wiki.osdev.org and http://forum.osdev.org || Books: http://wiki.osdev.org/Books'
03:40:27 --- topic: set by Mutabah!~tpg@pdpc/supporter/student/thepowersgang on [Sat Sep 16 19:44:15 2017]
03:40:27 --- names: list (clog lachlan_s `Guest00000 noldor doug16k fujisan light2yellow Kimundi_ NaNkeen Tazmain Kazinsal ^Inuyasha^ vdamewood behalebabo quc SGautam elevated grouse aalm Arcaelyx bork andrei-n_ Guest21615 RudyValencia Lowl3v3l xenos1984 variable plonk nj0rd t3hn3rd bluezinc dengke eremitah smeso awang__ mavhq abra0 Vercas kuldeep merry sham1 lucebac tj Guest34373 Defcronyke sprocklem umpc frolv Darksecond_ port8443 diginet X-Scale hussein absurdistani Belxjander jpo)
03:40:27 --- names: list (radens Jari-- zaquest tsurai booyah Burgundy zhiayang jack_rabbit bcos_ magnific1ab multi_io ZeDestructor rafaeldelucena funnel kasumi-owari dormito kanzure mrush_ pie_ rorx chedr tux3 ornitorrincos trousermonkey[m] dude1231- Peetz0r AntiSpamMeta FreeFull Naabed-_ pisculichi Youmu hollande Guest99251 dmh cja Brain SlyFawkes ptx bauen1 rain1 D4rk mlen Dreg `Ivan mawk Fatalnix butterthebuddha Neo shirk jimi1_ zentrum_ Simon70 oshogbo_ jjuran sssnap Patater)
03:40:27 --- names: list (hunterlabs Celelibi air josuedhg_ atk jordyd epony CWiz trn Kydd Gaudasse eam htafdwes bpye awordnot _sfiguser nightmared LambdaComplex empy kleinweby klys aiwakura ZipCPU pointfree R3x_ zgrepc ovf paulbarker swetland dmj` tktech xiphias MDude Mutabah gamozo gattuso grumble brynet Oxyd76 sleepyspider byte512 FireFly M_D_K Affliction hiei mpetch ephemer0l_ ecraven puckipedia bslsk05 Amaan Khaotic Nach0z nsh exezin lxpz kbeckmann reda oldtopman geordi)
03:40:27 --- names: list (CompanionCube sae2 ybden h4nnibal froggey lava pixelherodev SirCmpwn hcd[m] sha-2 snsapiens Meow-J___ divine_ pg12 peterbjornx EvilJStoker aejsmith davxy nshp opios lkurusa pikhq Xark matviy izabera beaky allight tyler569 Lucretia clever stee wgrant Shockk GomJabbar milesrout SM0TVI mniip Griwes transistor cc0 brad_d eax64 MindlessDrone mrvn robert_ jeaye dustinm` klange klange-tsun phr3ak lecx fristonio[m] Stary[m] am2on Wallbraker equalunique[m] llmm_)
03:40:27 --- names: list (htafdwes[m] jsgrant[m] caseypme[m] Drakonis[m] hl Ringil[m] VOIDShana[m] kataria[m] rw-r-r-0644[m] azonenberg latentprion gtucker huaw Sjors_ hohum renopt samirhali dograt apetresc alexlegg grange_c r3n_ Toldierone ohnx XgF alphawarrior rpcope chrisf Meisaka bipul mischief shakalaka drsn0w Boreeas Guest92125 JodaZ_ shikhin @ChanServ Benjojo sferrini Nokurn Compy_ danielinux olsner celadon tuxillo c0riolis barbershopper Colin_M_ stux shakeel Staphylo newnix)
03:40:27 --- names: list (sahibatko Olgierd mwk zerd md_5 SlashLife thinkpol catern promach shymega valerius qoxncyha kutio mowi malbx- hydraz Tobba cruxeternus nikivi BronzeBeard anonnumberanon graphitemaster Ameisen andrewrk acidx tkln j00ru Gynvael GreaseMonkey mrcivvy fusta chaignc sigtau glenda Stary farb heddwch tomaw grapte kbw Tanjoodo colona_ dwn emdub_ geist uplime krzywix ensign metaglog Dotti marky _mjg_ amygara m712 bitch amj Helio- heinrich5991 corecode z0ttel PyroLagus)
03:40:27 --- names: list (CcxWrk mlugg blueglass Wild talin goncalo Mikaku earlz SN4T14 dostoyevsky ikonia ampotos szpght_ dom96 [com]buster _ether_ _Hyper_ Brnocrist newsham jfu systmkor jp HeTo stux- pharaun zopsi eren`` Aerath)
03:41:26 --- join: Humble (~hchiramm@106.208.155.53) joined #osdev
03:58:27 --- join: t3h_n3rd (~Logic_Bom@cpc91224-cmbg18-2-0-cust219.5-4.cable.virginm.net) joined #osdev
03:59:23 --- join: andrei-n__ (~andrei@51.214-65-87.adsl-dyn.isp.belgacom.be) joined #osdev
03:59:49 --- quit: absurdistani (Ping timeout: 240 seconds)
04:01:25 --- join: nortega (~nortega@gateway/tor-sasl/deathsbreed) joined #osdev
04:02:04 --- join: oaken-source (~oaken-sou@p5DDB4811.dip0.t-ipconnect.de) joined #osdev
04:02:21 --- quit: andrei-n_ (Ping timeout: 260 seconds)
04:02:21 --- quit: t3hn3rd (Ping timeout: 260 seconds)
04:07:55 --- join: w41 (~w41@unaffiliated/w41) joined #osdev
04:08:26 --- quit: elevated (Quit: bye)
04:08:39 --- quit: Humble (Ping timeout: 256 seconds)
04:09:30 --- quit: NaNkeen (Ping timeout: 250 seconds)
04:11:21 --- join: elevated (~elevated@unaffiliated/elevated) joined #osdev
04:13:18 --- join: Humble (~hchiramm@106.208.155.53) joined #osdev
04:16:07 --- join: NaNkeen (~nankeen@115.164.80.206) joined #osdev
04:17:43 --- join: spare (~user@unaffiliated/spareproject) joined #osdev
04:19:19 --- quit: oaken-source (Ping timeout: 264 seconds)
04:21:33 --- quit: Humble (Ping timeout: 255 seconds)
04:22:55 --- join: Humble (~hchiramm@106.208.197.174) joined #osdev
04:30:47 --- join: mra90 (~Martin@host-85-202-159-241.sta.tvknaszapraca.pl) joined #osdev
04:34:30 --- quit: elevated (Quit: bye)
04:35:30 --- quit: Humble (Ping timeout: 250 seconds)
04:39:27 --- join: elevated (~elevated@unaffiliated/elevated) joined #osdev
04:43:55 --- join: CheckDavid (uid14990@gateway/web/irccloud.com/x-aclvnfznhlwqwjkn) joined #osdev
04:47:43 --- join: Humble (~hchiramm@106.208.197.174) joined #osdev
04:48:47 --- join: dennis95 (~dennis@i577BCE46.versanet.de) joined #osdev
04:57:37 --- quit: epony (Remote host closed the connection)
04:58:19 --- quit: `Guest00000 (Ping timeout: 240 seconds)
04:59:25 --- join: epony (~nym@79-100-134-61.ip.btc-net.bg) joined #osdev
05:02:03 --- quit: Kimundi_ (Ping timeout: 255 seconds)
05:05:51 --- quit: nortega (Quit: Vivu lante, vivu feliĉe!)
05:06:45 --- quit: vdamewood (Quit: Textual IRC Client: www.textualapp.com)
05:10:21 --- quit: latentprion (Quit: Lost terminal)
05:11:07 --- join: vdamewood (~vdamewood@unaffiliated/vdamewood) joined #osdev
05:15:58 --- join: glauxosdever (~alex@ppp-94-64-152-249.home.otenet.gr) joined #osdev
05:16:39 --- join: Kimundi_ (~Kimundi@dynip-129-217-079-084.wifi.tu-dortmund.de) joined #osdev
05:20:00 --- quit: vdamewood (Quit: Textual IRC Client: www.textualapp.com)
05:34:55 --- join: CrystalMath (~coderain@reactos/developer/theflash) joined #osdev
05:35:14 --- join: vmlinuz (~vmlinuz@187.101.157.248) joined #osdev
05:35:14 --- quit: vmlinuz (Changing host)
05:35:14 --- join: vmlinuz (~vmlinuz@unaffiliated/vmlinuz) joined #osdev
05:44:10 --- quit: light2yellow (Quit: light2yellow)
05:48:07 --- join: zlsyx (~zlsyx@193.2.224.100) joined #osdev
05:50:32 --- join: `Guest00000 (~user@37.113.160.19) joined #osdev
05:50:52 --- join: Shikadi (~Shikadi@cpe-98-10-34-205.rochester.res.rr.com) joined #osdev
05:53:40 <izabera> do i need to save the entire avx state every time i change process?
05:54:53 --- quit: lachlan_s (Quit: Connection closed for inactivity)
06:00:08 --- quit: Humble (Read error: Connection reset by peer)
06:00:56 <Lowl3v3l> izabera: if you want every process to have its own state, yes. and don't forget the SSE stuff.
06:02:08 --- quit: doug16k (Remote host closed the connection)
06:08:21 --- join: doug16k (~dougx@198-91-135-200.cpe.distributel.net) joined #osdev
06:13:01 --- quit: Kimundi_ (Quit: quit)
06:13:42 --- join: kimundi (~Kimundi@dynip-129-217-079-084.wifi.tu-dortmund.de) joined #osdev
06:14:41 --- join: absurdistani (~bradford@50-200-254-170-static.hfc.comcastbusiness.net) joined #osdev
06:17:43 --- quit: mrvn (Ping timeout: 248 seconds)
06:18:07 --- quit: elevated (Quit: bye)
06:31:39 --- join: JusticeEX (~justiceex@pool-98-113-143-43.nycmny.fios.verizon.net) joined #osdev
06:42:49 --- quit: Lowl3v3l (Quit: Leaving.)
06:51:22 <izabera> is there a way to detect if the state changed since this process started?
06:51:36 <izabera> well, not started
06:51:44 <izabera> i meant the last time that i executed this process
06:52:01 <izabera> like a cpu flag that tells you if the registers are dirty
06:52:01 <Mutabah> Depends...
06:52:11 <izabera> at least for stuff like avx
06:52:20 <lkurusa> you could disable those instructions and monitor the illegal opcode exception
06:52:23 <Mutabah> So, I think the newest version of the FPU save instruction has a validity bitmap
06:52:47 <Mutabah> But, as lkurusa said, you can just disable the FPU and re-enable it on the exception
06:53:02 <Mutabah> (I think it's a "coprocessor not present" exception?)
06:53:27 <izabera> is the fpu involved in any way for stuff like mov %xmm0, %xmm1 ?
06:53:40 --- quit: CheckDavid (Quit: Connection closed for inactivity)
06:53:58 --- join: Humble (~hchiramm@117.97.216.248) joined #osdev
06:54:00 <lkurusa> i don’t think so
06:54:05 <Mutabah> Weell... yes
06:54:23 <Mutabah> Kinda
06:54:39 <Mutabah> There's a flag that can disable all of those registers such that an excepiton is raised when they're used
06:54:59 <lkurusa> SSE seems to generate #UD if it’s not enabled
06:55:04 <izabera> is this flag settable from non prigileged rings?
06:55:13 --- quit: brynet (Ping timeout: 240 seconds)
06:55:20 --- join: brynet (~brynet@brynet6.biz.tm) joined #osdev
06:55:53 <Mutabah> Aah, it's not outright disabling the SSE
06:56:09 <Mutabah> you set the `TS` bit in FLAGS
06:56:21 <Mutabah> and any attempt to use the FPU/SSE will generate a #NM
06:56:28 <izabera> so i should be able to do this from userspace, right?
06:57:01 <izabera> wait ts is just the trap flag
06:57:17 <Mutabah> Nah, that's TF
06:57:23 <Mutabah> TS is 'task switched'
06:57:34 <Mutabah> (I think?)
06:58:32 <lkurusa> TS is task switched
06:58:43 <lkurusa> is that bit even working in long mode?
06:58:50 <Mutabah> Think so
06:58:58 <xenos1984> yes, TS can be used for lazy FPU / SSE / AVX state saving restoring
06:59:03 <xenos1984> yes it does
06:59:08 <lkurusa> but I think TS is in CR0?
06:59:18 <xenos1984> I'm using that in my kernel
06:59:26 <xenos1984> let me see...
06:59:28 <Mutabah> Whup! yep, it is
06:59:32 <Mutabah> Misread my code
06:59:43 <lkurusa> that’s cool
07:01:11 <xenos1984> it raises a "device not available" when this flag is set and one tries to use any such instruction (FPU / SSE / MMX / AVX)
07:01:29 --- quit: kimundi (Ping timeout: 256 seconds)
07:01:32 <izabera> it would be nice if this page https://wiki.osdev.org/CPU_Registers_x86 went a bit more in depth about this
07:01:35 <bslsk05> wiki.osdev.org: CPU Registers x86 - OSDev Wiki
07:02:03 --- join: oaken-source (~oaken-sou@p5DDB4811.dip0.t-ipconnect.de) joined #osdev
07:02:26 <xenos1984> so when you switch tasks you set TS, and when you get a "device not available", you save the registers to the task who last used them, and load the ones of the task that now wants to use them
07:02:40 <xenos1984> (and clear the flag)
07:03:09 --- join: Asu (~sdelang@AMarseille-658-1-120-96.w86-219.abo.wanadoo.fr) joined #osdev
07:04:21 <izabera> this is all very valuable information, thank you
07:10:29 --- join: nokdoot (~nokdoot@123.212.109.70) joined #osdev
07:10:46 <nokdoot> paging is too hard to me
07:10:48 <nokdoot> T.T
07:11:57 --- quit: brynet (Ping timeout: 246 seconds)
07:12:02 --- join: kimundi (~Kimundi@dynip-129-217-079-084.wifi.tu-dortmund.de) joined #osdev
07:15:36 --- join: baschdel (~baschdel@2a01:5c0:10:3d11:bca2:7797:3876:4668) joined #osdev
07:17:17 --- quit: variable (Quit: Found 1 in /dev/zero)
07:19:06 --- quit: zopsi (Ping timeout: 255 seconds)
07:27:06 --- quit: baschdel (Ping timeout: 250 seconds)
07:28:02 --- quit: xenos1984 (Quit: Leaving.)
07:30:51 --- join: Kimundi_ (~Kimundi@dynip-129-217-124-219.wifi.tu-dortmund.de) joined #osdev
07:32:36 --- quit: kimundi (Ping timeout: 246 seconds)
07:32:55 --- join: hmmmm (~sdfgsf@pool-72-79-167-102.sctnpa.east.verizon.net) joined #osdev
07:34:53 --- join: elevated (~elevated@unaffiliated/elevated) joined #osdev
07:37:09 --- quit: NaNkeen (Ping timeout: 246 seconds)
07:46:41 --- quit: Fatalnix (Changing host)
07:46:41 --- join: Fatalnix (~FatalNIX@unaffiliated/fatalnix) joined #osdev
07:53:58 --- join: NaNkeen (~nankeen@115.164.80.206) joined #osdev
07:55:36 --- quit: SGautam (Quit: Leaving)
08:01:09 --- join: xenos1984 (~xenos1984@22-164-191-90.dyn.estpak.ee) joined #osdev
08:04:14 --- join: lachlan_s (uid265665@gateway/web/irccloud.com/x-mtlghomptaqcdzlf) joined #osdev
08:04:39 --- join: promach2 (~promach@bb219-74-174-136.singnet.com.sg) joined #osdev
08:06:12 --- join: freakazoid0223 (~IceChat9@pool-108-52-244-197.phlapa.fios.verizon.net) joined #osdev
08:07:22 --- quit: promach2 (Client Quit)
08:08:53 <lachlan_s> Anyone know of any good online irc clients?
08:09:03 <lachlan_s> I currently use irccloud, but it's a bit lacking in some ways.
08:10:29 --- quit: spare (Quit: leaving)
08:11:34 --- join: sixand (~Thunderbi@112.5.237.192) joined #osdev
08:12:42 <izabera> why is mounting a filesystem a privileged operation?
08:14:08 --- join: toddw (~toddw@unaffiliated/toddw) joined #osdev
08:16:00 --- quit: sixand (Ping timeout: 246 seconds)
08:16:40 --- join: sixand (~Thunderbi@112.5.238.205) joined #osdev
08:29:25 --- quit: Humble (Read error: Connection reset by peer)
08:33:31 --- quit: glauxosdever (Quit: leaving)
08:33:56 --- quit: grouse (Quit: Leaving)
08:35:42 <lkurusa> lachlan_s (IRC): riot.im/
08:36:49 <lachlan_s> @lkurusa: Isn't that just for matrix?
08:44:12 --- join: Humble (~hchiramm@106.208.227.163) joined #osdev
08:45:16 --- join: promach__ (~promach@bb219-74-174-136.singnet.com.sg) joined #osdev
08:48:37 <Drakonis[m]> izabera: i think that's changing soon
08:52:51 --- nick: promach__ -> promach2
08:53:42 --- quit: jimi1_ (Quit: Reconnecting)
08:54:04 --- join: jimi1 (~jimi@unaffiliated/jimi1) joined #osdev
08:54:07 --- quit: NaNkeen (Ping timeout: 264 seconds)
08:55:19 --- quit: Kimundi_ (Ping timeout: 240 seconds)
08:56:43 --- quit: sixand (Quit: sixand)
08:59:11 --- quit: Tazmain (Quit: Leaving)
09:09:33 --- quit: Asu (Ping timeout: 255 seconds)
09:09:38 --- join: Asu` (~sdelang@147.15.136.77.rev.sfr.net) joined #osdev
09:12:14 --- join: banisterfiend (~banister@ruby/staff/banisterfiend) joined #osdev
09:21:42 --- join: NaNkeen (~nankeen@115.164.80.206) joined #osdev
09:22:51 --- join: godel (~gonzalo@190.192.95.122) joined #osdev
09:26:42 --- quit: NaNkeen (Ping timeout: 250 seconds)
09:28:26 --- quit: w41 (Ping timeout: 260 seconds)
09:30:34 --- join: zeus1 (~zeus@h1b42.n1.ips.mtn.co.ug) joined #osdev
09:32:50 --- join: brynet (~brynet@brynet6.biz.tm) joined #osdev
09:33:39 --- join: Kimundi_ (~Kimundi@i577A9E3D.versanet.de) joined #osdev
09:40:11 <lachlan_s> Ayyy, installed ubuntu 18.04 on my surface pro
09:42:41 <lkurusa> lachlan_s (IRC): great
09:42:48 <lkurusa> it’s not just for Matrix, I’m using the Riot client to connect to IRC right now :-)
09:42:51 <lkurusa> added benefit is free bouncer
09:42:57 --- join: lldd_ (~atrapado@unaffiliated/atrapado) joined #osdev
09:43:41 --- join: zopsi (~zopsi@dir.ac) joined #osdev
09:43:48 <lachlan_s> I'll check it out
09:44:00 <lachlan_s> How fast is the site?
09:46:14 --- quit: zeus1 (Read error: Connection reset by peer)
09:47:01 <lkurusa> It’s not the fastest, but it’s all right
09:47:10 <lkurusa> Mobile clients are quite okay too
09:47:24 --- join: zeus1 (~zeus@41.210.155.66) joined #osdev
09:53:17 --- join: baschdel (~baschdel@2a01:5c0:10:3d11:bca2:7797:3876:4668) joined #osdev
09:54:46 --- join: tacco| (~tacco@i59F5256A.versanet.de) joined #osdev
09:56:29 --- join: w41 (~w41@unaffiliated/w41) joined #osdev
10:02:07 --- quit: _sfiguser (Quit: Leaving)
10:02:49 --- quit: sham1 (Ping timeout: 240 seconds)
10:05:07 --- join: sham1 (~sham1@212-146-44-37.aboa.net) joined #osdev
10:06:23 --- quit: t3h_n3rd (Read error: Connection reset by peer)
10:16:03 --- join: spare (~user@unaffiliated/spareproject) joined #osdev
10:19:20 --- join: mrvn (~mrvn@HSI-KBW-109-193-047-129.hsi7.kabel-badenwuerttemberg.de) joined #osdev
10:23:54 --- quit: promach2 (Ping timeout: 250 seconds)
10:26:23 --- join: fp7 (~fp7@unaffiliated/fp7) joined #osdev
10:27:20 --- quit: fujisan (Quit: Connection closed for inactivity)
10:27:27 --- quit: banisterfiend (Quit: My MacBook has gone to sleep. ZZZzzz…)
10:29:26 <lachlan_s> The only issue with ubuntu on this is that there is no battery indicator
10:32:55 <geist> batteries are for the birds mang
10:33:25 * lachlan_s waves to geist
10:35:10 --- quit: ptx (Read error: Connection reset by peer)
10:35:31 --- join: banisterfiend (~banister@ruby/staff/banisterfiend) joined #osdev
10:37:57 --- join: ptx (~ptx@563400ad.rev.stofanet.dk) joined #osdev
10:38:07 <lachlan_s> geist, do you have a pixelbook?
10:40:05 --- quit: Arcaelyx (Quit: My MacBook has gone to sleep. ZZZzzz…)
10:40:51 --- quit: banisterfiend (Quit: My MacBook has gone to sleep. ZZZzzz…)
10:42:03 <geist> yeah
10:43:18 <lachlan_s> I'm thinking of getting one since they can now run linux apps easily
10:46:18 --- join: glauxosdever (~alex@ppp-94-64-152-249.home.otenet.gr) joined #osdev
10:48:17 --- join: banisterfiend (~banister@ruby/staff/banisterfiend) joined #osdev
10:52:35 --- join: Asu (~sdelang@147.15.136.77.rev.sfr.net) joined #osdev
10:52:35 --- quit: Asu` (Quit: Konversation terminated!)
10:53:47 --- join: Tazmain (~Tazmain@unaffiliated/tazmain) joined #osdev
10:56:58 --- join: Asu` (~sdelang@AMarseille-658-1-120-96.w86-219.abo.wanadoo.fr) joined #osdev
10:57:58 --- quit: Asu (Ping timeout: 268 seconds)
10:58:56 --- join: Arcaelyx (~Arcaelyx@2604:2000:f0ca:7500:405:b9fe:538f:bb71) joined #osdev
10:59:01 --- quit: Arcaelyx (Client Quit)
10:59:22 --- join: ven0m (4dac8cd3@gateway/web/freenode/ip.77.172.140.211) joined #osdev
10:59:41 <ven0m> hi
11:00:03 <ven0m> x86 debug registers are used for ring0 (kernel) debugging?
11:00:18 <ven0m> or they have some use from user space?
11:00:32 <geist> dont think so, not sure any of them are visible from user mode
11:01:00 <ven0m> from the doc, they aren't
11:01:19 <ven0m> I am trying to find a real example on how they are used
11:01:23 <ven0m> do you know some?
11:08:43 --- quit: banisterfiend (Quit: My MacBook has gone to sleep. ZZZzzz…)
11:09:50 --- join: fujisan (uid4207@gateway/web/irccloud.com/x-cqquowboimhdxubb) joined #osdev
11:11:00 --- join: banisterfiend (~banister@ruby/staff/banisterfiend) joined #osdev
11:15:17 --- join: NaNkeen (~nankeen@115.164.80.206) joined #osdev
11:16:30 --- join: Asu (~sdelang@147.15.136.77.rev.sfr.net) joined #osdev
11:16:39 --- quit: toddw (Quit: Textual IRC Client: www.textualapp.com)
11:17:05 --- quit: Asu` (Ping timeout: 268 seconds)
11:17:26 --- quit: nokdoot (Remote host closed the connection)
11:17:39 --- quit: elevated (Quit: bye)
11:20:03 --- quit: NaNkeen (Ping timeout: 255 seconds)
11:27:19 <lkurusa> you can’t use them from userspace
11:27:21 <lkurusa> on Linux you can write to them from userspace using ptrace
11:27:45 <lkurusa> they are used for setting up watchpoints (i.e., “send me a #BP when someone reads from memory location X”
11:28:17 <lkurusa> actually, #DB
11:29:03 --- quit: JusticeEX (Ping timeout: 255 seconds)
11:32:59 --- join: sortie (~sortie@static-5-186-55-44.ip.fibianet.dk) joined #osdev
11:36:16 --- join: Arcaelyx (~Arcaelyx@2604:2000:f0ca:7500:405:b9fe:538f:bb71) joined #osdev
11:38:20 --- join: k4m1 (~k4m1@37-219-245-154.nat.bb.dnainternet.fi) joined #osdev
11:41:05 --- quit: nj0rd (Quit: WeeChat 2.1)
11:45:25 <doug16k> ven0m, my kernel has a gdb stub that uses them
11:46:05 <doug16k> https://github.com/doug65536/dgos/blob/master/kernel/arch/x86_64/cpu/control_regs.h#L218
11:46:07 <bslsk05> github.com: dgos/control_regs.h at master · doug65536/dgos · GitHub
11:46:38 <doug16k> https://github.com/doug65536/dgos/blob/master/kernel/arch/x86_64/gdbstub.cc
11:46:43 <bslsk05> github.com: dgos/gdbstub.cc at master · doug65536/dgos · GitHub
11:46:57 --- quit: banisterfiend (Quit: My MacBook has gone to sleep. ZZZzzz…)
11:47:44 --- join: ASPO (ad102383@gateway/web/freenode/ip.173.16.35.131) joined #osdev
11:47:54 <ASPO> Hey guys
11:48:08 <lachlan_s> Hey
11:48:22 <ASPO> I am learning to trap with MIPS
11:53:06 --- quit: Jari-- (Ping timeout: 260 seconds)
11:55:58 --- join: Asu` (~sdelang@243.82.136.77.rev.sfr.net) joined #osdev
11:59:33 --- quit: Asu (Ping timeout: 256 seconds)
12:03:05 --- quit: Tazmain (Quit: Leaving)
12:03:09 --- join: nj0rd (~nj0rd@i577BC0FD.versanet.de) joined #osdev
12:05:50 --- join: user10032 (~Thirteen@2a02:c7d:314e:b300:18d1:cd55:e1f0:b559) joined #osdev
12:07:47 --- quit: k4m1 (Quit: bbl)
12:15:05 --- join: banisterfiend (~banister@ruby/staff/banisterfiend) joined #osdev
12:16:00 --- join: m_t (~m_t@p5DDA2AC1.dip0.t-ipconnect.de) joined #osdev
12:26:28 --- quit: brynet (Quit: leaving)
12:45:39 --- join: bemeurer (~bemeurer@185.232.22.59) joined #osdev
12:45:41 --- join: JusticeEX (~justiceex@pool-98-113-143-43.nycmny.fios.verizon.net) joined #osdev
12:46:12 --- quit: bemeurer (Client Quit)
12:46:38 --- join: bemeurer (~bemeurer@2600:8802:5300:bb90:863a:4bff:fe06:c8b2) joined #osdev
12:47:51 --- join: brynet (~brynet@brynet6.biz.tm) joined #osdev
12:48:31 --- quit: Arcaelyx (Quit: My MacBook has gone to sleep. ZZZzzz…)
12:48:42 --- join: hmmmmm (~sdfgsf@pool-72-79-165-25.sctnpa.east.verizon.net) joined #osdev
12:50:24 --- quit: hmmmm (Ping timeout: 246 seconds)
12:52:27 --- quit: spare (Quit: leaving)
12:53:14 --- join: Arcaelyx (~Arcaelyx@2604:2000:f0ca:7500:405:b9fe:538f:bb71) joined #osdev
12:53:50 --- quit: baschdel (Ping timeout: 250 seconds)
12:54:51 --- quit: fp7 (Quit: .)
12:55:12 --- join: fp7 (~fp7@unaffiliated/fp7) joined #osdev
13:07:14 --- quit: awordnot (Read error: Connection reset by peer)
13:08:04 --- join: awordnot (~awordnot@67.184.137.130) joined #osdev
13:11:03 --- quit: `Guest00000 (Ping timeout: 246 seconds)
13:13:43 --- join: Asu (~sdelang@150.42.136.77.rev.sfr.net) joined #osdev
13:16:02 <Vercas> Windows updated some shit.
13:16:09 <Vercas> Now I cannot RDP into my server.
13:16:12 --- join: NaNkeen (~nankeen@115.164.80.206) joined #osdev
13:17:11 --- quit: Asu` (Ping timeout: 256 seconds)
13:19:02 --- part: mrvn left #osdev
13:19:18 --- join: _sfiguser (~sfigguser@130.108.214.77) joined #osdev
13:19:50 --- quit: w41 (Ping timeout: 250 seconds)
13:20:12 <ASPO> havent you said thanks microsoft enough to not use it?
13:21:09 <Vercas> :(
13:21:11 --- quit: NaNkeen (Ping timeout: 260 seconds)
13:25:00 --- join: w41 (~w41@unaffiliated/w41) joined #osdev
13:26:40 --- join: elevated (~elevated@unaffiliated/elevated) joined #osdev
13:27:54 --- quit: banisterfiend (Quit: My MacBook has gone to sleep. ZZZzzz…)
13:33:30 --- quit: zlsyx (Quit: Leaving...)
13:38:19 --- quit: zaquest (Ping timeout: 240 seconds)
13:38:38 <glauxosdever> One of my professors complained about Windows publicly during a class. He said he wanted to do something on a Windows 10 laptop that hadn't been booted since 6 months or so but it was sluggish because it was downloading updates, then it took 2 hours to shutdown because of installing those updates. He didn't say whether it broke something, but it's already "fun"
13:41:58 --- join: Asu` (~sdelang@150.42.136.77.rev.sfr.net) joined #osdev
13:42:36 --- quit: lldd_ (Quit: Leaving)
13:42:37 --- join: zaquest (~notzaques@5.128.210.30) joined #osdev
13:45:31 --- quit: Asu (Ping timeout: 256 seconds)
13:45:56 <qoxncyha> can someone explain how OpenGL calls get sent to the video card, step by step?
13:46:11 <sortie> They don't
13:46:38 <sortie> OpenGL calls are compiled into a sequence of special shader commands and small programs sent to the video card, which varies a lot from card to card
13:46:57 <qoxncyha> sortie: what channel is used to 'put' the commands into the video card?
13:47:15 <qoxncyha> is it mmap'ed?
13:47:28 <sortie> The kernel driver for the video card locates the PCI device for the video card and maps its registers
13:47:36 <sortie> The registers are probably memory mapped
13:47:50 <sortie> There's probably a DMA engine involved for efficient transfer to and from system memory
13:48:19 --- quit: w41 (Ping timeout: 240 seconds)
13:48:37 <sortie> qoxncyha: I recommend you learn more basic OpenGl stuff first <https://open.gl/>. Once you know how complicated shaders can be, you start to grasp how complicated the driver has to be.
13:48:38 <bslsk05> open.gl: OpenGL - Introduction
13:48:51 <qoxncyha> that's a bit condescending
13:49:25 <sortie> Basically the OpenGL implementation is a compiler from the shader language, and other basic OpenGL calls, to a native instruction set for the GPU
13:50:02 <sortie> qoxncyha: I'm being helpful. It seems to me that you don't understand some basics, so I'm recommending more topics to learn.
13:50:06 --- join: drakonis (~drakonis@unaffiliated/drakonis) joined #osdev
13:50:21 <qoxncyha> the code is compiled in the CPU, correct?
13:50:28 <sortie> Yes
13:50:50 <qoxncyha> how does the compiled code get to the video card? sorry if i'm repeating myself
13:51:00 <sortie> I have not implemented a graphics driver myself, it's very hard (you need to know a lot) and it's alot of work
13:51:43 <sortie> qoxncyha: The core of the video driver is a DMA engine that transfers memory to and from the GPU. There's probably some command packets that can be sent and received this way to asyncronously make the GPU do stuff.
13:52:00 <sortie> It varies from card to card.
13:52:26 <qoxncyha> in that case, is RAM always a middle layer?
13:52:42 <sortie> I don't know what that means
13:53:08 <qoxncyha> to get the compiled kernel into the GPU, does it always traverse RAM to get there?
13:53:10 <sortie> You can send commands that asks the GPU to copy memory from one location on the GPU to another location on the GPU
13:53:28 <sortie> The kernel is not compiled to the GPU
13:54:22 <sortie> Programs like "the pixel's red color is its X coordinate ” are compiled to GPU instructions and sent to the GPU for execution as part of a more complex program that draws a scene
13:55:00 <qoxncyha> sortie: i don't mean the linux kernel, lol. i mean to ask how compiled native instructions for the gpu get to the gpu
13:55:38 <sortie> I answered that
13:55:45 <sortie> qoxncyha: I don't know details.
13:55:50 <chrisf> this isnt entirely accurate.
13:55:50 <sortie> It varies from card to card.
13:55:57 <qoxncyha> sortie: no worries, i'm not asking for exact knowledge, just a high-level understanding
13:56:04 <qoxncyha> sortie: just to confirm, you said it is DMA?
13:56:17 <sortie> chrisf: Please correct me if I'm wrong, I don't know any specifics. I'm just answering him to the best of my knowledge :)
13:56:20 <chrisf> there /may/ be a DMA engine that can copy things without handholding.
13:56:47 <ASPO> CPU copies are dumb
13:56:50 <ASPO> in other words
13:57:24 <chrisf> you also may not need to copy anything -- some chunk of memory is typically accessible to both the CPU and GPU.
13:57:56 <qoxncyha> chrisf: makes sense
13:59:47 <qoxncyha> how is the GPU signalled to run the compiled native instructions? does it repeatedly run the chunk of memory on its own until changed?
14:00:24 --- quit: Arcaelyx (Quit: My MacBook has gone to sleep. ZZZzzz…)
14:01:40 <chrisf> typically there's a ringbuffer owned by the host kernel which the GPU frontend feeds from
14:02:20 --- join: banisterfiend (~banister@ruby/staff/banisterfiend) joined #osdev
14:03:27 <chrisf> the userspace driver will fill out fairly large command buffers with all the hardware commands for a bunch of draws, and then ask the kernel to make the buffer visible to the GPU and then insert an "execute this command buffer" command in the ring
14:04:39 <qoxncyha> chrisf: how does the kernel physically speak to the gpu?
14:05:22 <chrisf> memory mapped IO
14:06:57 <qoxncyha> that makes sense
14:07:14 <qoxncyha> are there any other ways to communicate to peripherals, in general, other than memory mapping?
14:12:49 --- quit: andrei-n__ (Ping timeout: 240 seconds)
14:13:03 --- join: `Guest00000 (~user@37.113.160.19) joined #osdev
14:13:45 <qoxncyha> i would be really interested in an answer to this question
14:14:16 --- quit: oaken-source (Ping timeout: 260 seconds)
14:14:49 --- join: immibis (~chatzilla@222-155-160-32-fibre.bb.spark.co.nz) joined #osdev
14:18:08 --- join: Arcaelyx (~Arcaelyx@2604:2000:f14a:2500:69dd:9d84:25f3:6f26) joined #osdev
14:20:11 <ven0m> with PTI in kernel memory space all physical memory is still mapped, right? So also userspace page tables?
14:21:25 --- quit: glauxosdever (Quit: leaving)
14:26:56 --- join: w41 (~w41@unaffiliated/w41) joined #osdev
14:30:24 <ASPO> Some cpus get extended with registers
14:31:07 <ASPO> but thats more embedded chips
14:31:45 <qoxncyha> how does data get to the serial port? also memory-mapped?
14:31:59 <ASPO> It is read through a register
14:32:06 <ASPO> and wrote thru
14:32:22 <ASPO> Go grab yourself an arduino
14:32:42 <ASPO> I think you should look up some DUI projects with hardware too starting with usb port
14:32:50 <ASPO> Especially for operating system-less designs
14:34:50 <chrisf> qoxncyha: almost all hw interfaces come down to some combination of registers mapped into a space you can bang on, and shared memory.
14:35:00 --- join: OZ__ (~AndrewBC@ip70-187-57-150.pn.at.cox.net) joined #osdev
14:36:27 --- join: isd (~isd@pool-71-174-32-198.bstnma.east.verizon.net) joined #osdev
14:37:28 <chrisf> that space might be a dedicated IO space (with 'in' / 'out' type instructions to access it) or mapped into the memory space somewhere (where you can use regular moves etc)
14:38:59 --- quit: banisterfiend (Quit: My MacBook has gone to sleep. ZZZzzz…)
14:39:05 <ven0m> hmm
14:40:25 <chrisf> ven0m: re userspace still being mapped in kernel mode, afaik yes, even with KPTI.
14:40:31 <izabera> ASPO: i read that as for operating (system)-less) designs instead of for (operating system)-less designs
14:40:49 <chrisf> ven0m: /not/ mapping the userspace makes copy_to_user / copy_from_user suck.
14:42:07 <ven0m> yeah, so kernel page tables map also processes memory
14:43:10 --- quit: user10032 (Quit: Leaving)
14:43:28 <chrisf> ven0m: you could design a system that doesnt map any process's userspace while the kernel is running
14:44:09 <ven0m> is it doable without any performance impact?
14:44:53 <chrisf> as before, it makes marshaling data between kernel and user more ugly
14:45:18 <aalm> no free lunches
14:45:53 <chrisf> and you murder the TLB even more than usual
14:45:54 --- join: banisterfiend (~banister@ruby/staff/banisterfiend) joined #osdev
14:46:05 <ven0m> right
14:46:11 --- join: SGautam (~GautamS@unaffiliated/gautams) joined #osdev
14:46:16 <ven0m> so it is not doable
14:46:18 <lachlan_s> Murdering things is bad
14:46:56 --- quit: `Guest00000 (Ping timeout: 250 seconds)
14:49:14 <lachlan_s> Is there a particular reason why it's necessary to turn on virtual memory in long mode?
14:49:46 <lachlan_s> Aside from the cpu shouting at you
14:50:11 <ASPO> The GBA had a hardware accelerated GPU per se. Except you used hardware registers and DMA to transfer data to its RAM which would write back to the devices VRAM which draws to the screen.
14:50:38 <ASPO> But yes that is still MMU shit
14:51:55 --- quit: drakonis (Remote host closed the connection)
14:55:06 --- quit: immibis (Ping timeout: 260 seconds)
14:56:40 <chrisf> lachlan_s: "because it's defined that way" / shouts at you if you dont
14:57:09 --- quit: banisterfiend (Quit: My MacBook has gone to sleep. ZZZzzz…)
14:57:22 <chrisf> is not much of a practical issue -- you want paging enabled anyway
14:57:43 --- quit: elevated (Quit: bye)
14:58:44 <lachlan_s> I mean, on a "managed" os, paging isn't necessary
14:59:12 <lachlan_s> You can make segmented heaps and stacks to get around it
14:59:28 <lachlan_s> Plus, you don't get the overhead of the tlb cache filling
14:59:31 --- quit: aalm (Ping timeout: 264 seconds)
15:00:09 --- join: aalm (~aalm@37-219-119-194.nat.bb.dnainternet.fi) joined #osdev
15:00:10 --- join: drakonis (~drakonis@unaffiliated/drakonis) joined #osdev
15:04:19 <chrisf> but you lose the ability to do memory mapped files, page attributes, etc.
15:04:58 --- quit: vmlinuz (Quit: Leaving)
15:05:10 <geist> lachlan_s: the reason is because that's what AMD decided
15:05:38 <geist> my guess is it's because since the 4 ring levels are functionally dead in 64bit, and segmentation doesn't allow you to otherwise keep user vs supervisor code separate, you need paging for that
15:05:46 <lachlan_s> I suppose that's a good enough reason as any
15:06:09 <chrisf> enormous amounts of area are burned making the paging machinery fast
15:06:27 <chrisf> you dont get a refund by not using it
15:06:43 <geist> plus fairly early on AMD added 1GB pages, though i dont think it was present in K8
15:06:57 <geist> but that generally effectively disables the overhead of the page tables
15:08:05 <chrisf> a certain popular OS only really supports 4K pages, and throws up all over itself if you use the bolt-on 2M page mechanism
15:08:12 <lachlan_s> How often are large pages like that used?
15:08:38 <lachlan_s> Seems like it'd totally mess up memory organization
15:08:53 <chrisf> lachlan_s: for a client platform, never. 2M is useful if it's transparent. 1G isnt.
15:09:07 --- quit: w41 (Ping timeout: 264 seconds)
15:09:45 <chrisf> 1G pages are useful if you actually have blocks of data around that size in your working set.
15:09:54 <lachlan_s> And the memory under those needs to be contiguous?
15:09:59 <chrisf> yes
15:10:01 <chrisf> HPC!
15:10:36 --- join: ALowther (~alowther@68.200.236.134) joined #osdev
15:11:33 <lachlan_s> Is there any downside to having more memory then necessary mapped in (aside from constricting its use)?
15:12:35 <geist> was talking more about from the point of view of effectively disabling paging, if that's what you want
15:12:45 <geist> you can map all of phys8ical memory using only 1GB pages and that's pretty good
15:12:48 --- quit: SGautam (Quit: Rekt)
15:14:31 <lachlan_s> Yep
15:14:56 --- quit: m_t (Quit: Leaving)
15:15:00 <chrisf> geist: except for the first GB being a mess, sure.
15:15:35 <geist> even if you go across PCI space, its still there
15:16:08 <lachlan_s> It just makes everything messy
15:16:37 <geist> well, look i was just giving you a suggestion, based on the previous notion of 'why was paging required'
15:16:40 <geist> take it or leave it
15:16:43 <geist> sheesh
15:16:47 --- join: NaNkeen (~nankeen@115.164.80.206) joined #osdev
15:17:38 <lachlan_s> Hey, hey, don't be like that. Thanks for explaining it to me!
15:17:42 <chrisf> geist: do you leave the firmware's PCI config largely intact, or remap everything out of the way?
15:17:43 <lachlan_s> I was more thinking outloud
15:20:13 <geist> generally it's a good idea to stick with what they already did
15:20:26 <geist> there's not a lot of value to re-laying out PCI space
15:20:38 <geist> especially if you want to maintain some sort of continuity across the PCI init
15:21:11 --- join: drakonis_ (~drakonis@unaffiliated/drakonis) joined #osdev
15:21:42 --- quit: NaNkeen (Ping timeout: 255 seconds)
15:22:58 --- quit: lachlan_s (Quit: Updating details, brb)
15:23:11 --- join: lachlan_s (uid265665@gateway/web/irccloud.com/x-lymdbifvdpivvzym) joined #osdev
15:24:12 --- quit: drakonis (Ping timeout: 250 seconds)
15:24:12 --- quit: Darksecond_ (Read error: Connection reset by peer)
15:25:18 --- quit: quc (Ping timeout: 255 seconds)
15:26:24 --- quit: ven0m (Quit: Page closed)
15:26:52 --- quit: Asu` (Remote host closed the connection)
15:27:31 --- join: Darksecond (~darksecon@a82-94-53-70.adsl.xs4all.nl) joined #osdev
15:28:03 <lachlan_s> What do you think of the mill cpus?
15:28:49 --- join: `Guest00000 (~user@37.113.160.19) joined #osdev
15:29:39 <chrisf> patents and slideware
15:29:55 <lachlan_s> That's fair
15:30:01 <lachlan_s> We haven't actually seen any hardware
15:30:08 <geist> yeah cpu architectures are a dime a dozen. real implementations are where it matters
15:30:15 <chrisf> i enjoy the presentations and there are some interesting ideas in there
15:31:03 <chrisf> but yes... silicon on my desk!
15:32:43 <lachlan_s> Do you think we'll be stuck with x86, arm, and riscv for a while then?
15:32:55 <lachlan_s> *or rather amd64
15:33:34 <geist> yes
15:33:58 <geist> as with most things, there's a very strong bias towards status quo
15:34:07 <lachlan_s> I haven't done anything with riscv yet. Is it nicer then the other two?
15:34:13 <geist> not particularly
15:34:22 <lachlan_s> :/
15:34:44 <geist> it's primary advantage as far as i can tell is that its royalty free
15:35:34 <lachlan_s> Well, that's nothing to scoff at
15:38:01 <lachlan_s> Didn't ibm do something with a new arch recently?
15:41:07 --- join: zaquest_ (~notzaques@5.128.210.30) joined #osdev
15:42:57 --- quit: zaquest (Ping timeout: 246 seconds)
15:43:47 <geist> hmm, dunno
15:44:04 <geist> kind of doubt it, they're pretty fixed on power and zmachine now
15:44:36 <lachlan_s> Must have been power that I was thinking of
15:48:38 --- quit: fujisan (Quit: Connection closed for inactivity)
15:50:19 --- quit: Kimundi_ (Ping timeout: 240 seconds)
15:55:17 --- quit: epony (Remote host closed the connection)
15:55:36 --- join: epony (~nym@79-100-134-61.ip.btc-net.bg) joined #osdev
15:58:32 <ASPO> https://blogs.windows.com/buildingapps/2018/05/08/visual-studio-support-for-windows-10-on-arm-development/
15:58:33 <bslsk05> blogs.windows.com: Early preview of Visual Studio support for Windows 10 on ARM development - Windows Developer BlogWindows Developer Blog
15:59:22 <ASPO> "These Always Connected PCs are thin, light, fast, and designed with instant-on 4G LTE connectivity and unprecedented battery life – now measured in days and weeks, not hours. "
16:01:28 --- quit: drakonis_ (Ping timeout: 250 seconds)
16:02:19 --- join: drakonis (~drakonis@unaffiliated/drakonis) joined #osdev
16:02:25 <lachlan_s> Those are pretty cool, but they don't seem to have much advantage over thin and lights these days
16:02:40 <geist> but they're unprecedented!
16:03:46 <ASPO> Hey I bought a nvidia gtx 1080 sc
16:03:57 <lachlan_s> My surface gets just as good battery life
16:04:03 <geist> ASPO: that's unprecented!
16:04:09 <lachlan_s> Well, it did before I installed ubuntu
16:04:27 <ASPO> I tell you what after spending a lot of my time with a zotac gt 730 (4gb ddr3) GPU I don't think I really care much about slim aha
16:04:41 <lachlan_s> Unprecedented!
16:04:50 <geist> unprecedented!
16:05:01 <ASPO> Precedented!
16:06:18 <lachlan_s> It's a unprecidented! idea, but I just don't see much use for it
16:07:50 <lachlan_s> I want one of these babies: https://www.windowscentral.com/sites/wpcentral.com/files/styles/xlarge/public/field/image/2017/10/A%20westworld-ui-reel5_0.jpg?itok=RNzeEesO
16:10:04 <lachlan_s> Or one of these: http://blogs-images.forbes.com/kevinmurnane/files/2017/03/Hand-terminal-eason-1_Syfy.jpg
16:10:08 <bslsk05> blogs-images.forbes.com <no title>
16:10:47 <ASPO> Tbh we should really start making portable screens that work off wifi so that could be possible
16:11:06 <ASPO> I don't need to lug my pc and monitor around if I wanted to be portable
16:11:25 <lachlan_s> Pretty much a "cloud" based thin client
16:11:53 <ASPO> If I hear cloud one more time from microsoft I am going to contact ISIS
16:12:03 <ASPO> What are their goals?
16:12:09 <ASPO> 'cloud computing'
16:12:10 <ASPO> more like
16:12:15 <ASPO> 'hey lets steal users data'
16:12:19 <lachlan_s> They're trying to help fix the drought in cali
16:12:26 <ASPO> oh tru
16:13:05 --- quit: mra90 (Read error: Connection reset by peer)
16:13:07 <ASPO> pretty sure cali is a drought
16:13:09 <ASPO> but yea
16:13:53 <ASPO> la $t0, callback_sceKernelCreateCallback_001
16:13:57 <ASPO> lw $a0, ($t0)
16:14:09 <geist> looks like mips
16:14:18 <geist> or maybe sparc
16:14:46 <lachlan_s> To be honest, I wouldn't be opposed to hat basically having your phone basically just run in a container that could easily be moved around between datacenters.
16:14:53 <ASPO> So I load the address of that variable into $t0, and load a word from it to $a0, would this be redundant? Yes its MIPS
16:15:14 <lachlan_s> So you could have localized datacenters virtualize your phone and stream it to your "hand-terminal"
16:15:27 <lachlan_s> And you'd get ridiculous battery life
16:15:33 <ASPO> Thats exactly what I said, stream it to your hand terminal
16:15:38 --- quit: sortie (Quit: Leaving)
16:15:44 <lachlan_s> Yep, what you said
16:16:16 <ASPO> DMA it from the video card to your memory to your hand terminal
16:16:25 <ASPO> Using black magic register work
16:16:42 <lachlan_s> I wonder what kind of latency you'd need for it to feel the same as modern smartphones
16:17:01 <ASPO> You only need to draw once every frame, lachlan
16:17:18 <ASPO> So as long as you have less than or equal to 60 ping u'd be full resolution each time
16:17:39 <lachlan_s> Right, but what's the latency between touching the screen and it reacting however way
16:17:47 <lachlan_s> 100 ms?
16:17:53 <lachlan_s> That's double over 5g
16:17:54 <ASPO> No more like 2 ms
16:18:06 <ASPO> Ever lan party before?
16:18:06 <lachlan_s> Hmm, not so doable
16:18:09 <lachlan_s> Yes
16:18:16 <ASPO> Ever look at the amazing ping?
16:18:25 <lachlan_s> I haven't actually
16:18:30 <lachlan_s> less then 10 or so, right?
16:18:31 <ASPO> I could get less than 10ms ping with wifi on 2 old vista laptops
16:18:50 <ASPO> I think 8 but idk
16:19:28 <ASPO> I have two PC's to try it
16:19:44 <ASPO> But that involves taking out java for doing it in a reasonable time
16:19:47 <OZ__> Wouldn't you rather graphics be local but instead talk about filesystem access in the scheme of it all?
16:19:50 <Belxjander> LAN-wise pings?
16:20:30 <ASPO> Yes Belxjander
16:20:32 <ASPO> OZ__: ?
16:20:53 --- quit: dennis95 (Quit: Leaving)
16:21:04 * Belxjander has multiple RPi machines along with a Laptop and desktop here and they all "ping <IPaddr>" respond with a maximum lag ~5ms at the worst and that is regardless of wired/wireless (I have both wired and wireless machines on the same LAN here)
16:21:05 <ASPO> Theres a program that does this already but does it very badly
16:21:12 --- join: banisterfiend (~banister@ruby/staff/banisterfiend) joined #osdev
16:21:26 --- quit: _sfiguser (Ping timeout: 260 seconds)
16:22:34 <Belxjander> ASPO: is MIPS a RISC style Instruction Setup or a CISC style Instruction setup ?
16:22:37 --- quit: ephemer0l_ (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.)
16:23:03 <ASPO> yes
16:23:45 <Belxjander> ASPO: if it is a RISC style then you need to load any r[N} with an address and use it for target reading into another r[N] as memory is exclusively "load" and "store" operation accessed
16:23:56 --- join: ephemer0l_ (~ephemer0l@pentoo/user/ephemer0l) joined #osdev
16:24:07 <ASPO> Remember that ping is a measurement of round trip
16:24:12 <Belxjander> I have a PowerPC desktop here and it's also a RISC arch Instruction Setup
16:24:15 <ASPO> So lag wouldn't be horrible at all with touching
16:24:40 --- quit: xenos1984 (Quit: Leaving.)
16:25:01 <Belxjander> ASPO: actually it would more depend on the amount of processing at each end more than the network transfer (Locally at least)
16:25:35 <ASPO> Belxjander: We talking wifi here hombre
16:26:01 <ASPO> Belxjander: I am not sure what you are telling me tehre.
16:27:30 <Belxjander> in reference to touch are you talking about two devices operating a wifi while the cases are in contact ?
16:27:39 <Belxjander> NFC range?
16:28:09 <Belxjander> or touch-screen usage for an Application that is not on the Touchscreen device ?
16:28:59 --- quit: fp7 (Read error: Connection reset by peer)
16:31:11 <lachlan_s> The latter
16:31:32 <lachlan_s> I mean, so you had a datacenter in each city that could run these containers and stream them to your device
16:31:43 <lachlan_s> Using 5g, the latency could be very low
16:32:20 <Belxjander> ugh.... I'd only ever do that over Wifi on a LAN setup... not over Cellular Wireless
16:33:01 * Belxjander has seen Cellullar wireless that is claimed to be good turn to total crap when too many handsets were in close proximity
16:34:03 <lachlan_s> What if you used mesh networks so the more devices in proximity, the higher the bandwidth
16:34:16 <lachlan_s> You could break up the data and multiplex it
16:34:22 --- join: _sfiguser (~sfigguser@95.141.40.109) joined #osdev
16:35:43 <lachlan_s> Figure out the maximum bandwidth that you can stream to the mass of devices and distribute it amongst all of them
16:36:31 --- quit: banisterfiend (Quit: My MacBook has gone to sleep. ZZZzzz…)
16:40:38 --- join: manwar (~manwar@41.37.214.159) joined #osdev
16:43:08 --- quit: manwar (Client Quit)
16:51:22 <geist> lachlan_s: like a big raid0 floppy drive array?
16:54:19 <lachlan_s> geist: exactly lol
16:55:16 <lachlan_s> I mean, r?emote desktops seem to work alright right
16:55:24 --- quit: `Guest00000 (Ping timeout: 246 seconds)
16:58:27 <lachlan_s> Might as well make them better by adding buzzwords
16:59:57 --- quit: JusticeEX (Ping timeout: 240 seconds)
17:01:07 <Ameisen> MIPS is pretty explicitly RISC
17:02:58 <klange> Mm, sounds like riscy business.
17:03:27 --- quit: Brain (Read error: Connection reset by peer)
17:03:49 <Ameisen> calm down tom cruise
17:03:54 <lachlan_s> Seems like localized 5g can be between 5-6 ms
17:04:40 --- join: Brain (~brain@brainwave.brainbox.cc) joined #osdev
17:07:52 --- nick: jp -> picle
17:07:56 --- nick: picle -> pickle
17:08:42 <lachlan_s> How long it takes light to actually move across that distance is not a significant part of that time, so it should be possible to expand that quite a bit without increasing latency
17:08:56 --- nick: pickle -> jp
17:09:14 --- join: Jari-- (~vai@85-76-105-31-nat.elisa-mobile.fi) joined #osdev
17:09:17 <lachlan_s> So, say you can get a constant 10ms ping
17:09:28 <lachlan_s> Would that be low enough for a good user experience
17:10:22 --- quit: drakonis (Ping timeout: 250 seconds)
17:10:46 --- quit: isd (Quit: Leaving.)
17:10:53 <lachlan_s> The touch latency on an s8 is about 90ms lol
17:12:31 --- join: `Guest00000 (~user@37.113.160.19) joined #osdev
17:15:08 --- quit: bemeurer (Ping timeout: 250 seconds)
17:17:39 --- join: NaNkeen (~nankeen@115.164.80.206) joined #osdev
17:17:40 --- quit: ALowther (Remote host closed the connection)
17:19:25 --- join: MarchHare (~Starwulf@mo-184-5-204-232.dhcp.embarqhsd.net) joined #osdev
17:19:42 --- join: banisterfiend (~banister@ruby/staff/banisterfiend) joined #osdev
17:22:33 --- quit: NaNkeen (Ping timeout: 256 seconds)
17:24:04 --- join: promach2 (~promach@bb219-74-174-136.singnet.com.sg) joined #osdev
17:25:46 --- join: dbittman (~dbittman@2601:647:ca00:1651:b26e:bfff:fe31:5ba2) joined #osdev
17:32:13 --- join: JusticeEX (~justiceex@pool-98-113-143-43.nycmny.fios.verizon.net) joined #osdev
17:33:59 --- join: jakogut (~jakogut@162.251.69.147) joined #osdev
17:34:24 --- quit: gamozo (Remote host closed the connection)
17:37:40 --- join: drakonis (~drakonis@unaffiliated/drakonis) joined #osdev
17:37:43 --- join: jakogut_ (~jakogut@68.116.23.122) joined #osdev
17:39:09 --- quit: jakogut (Ping timeout: 246 seconds)
17:39:09 --- nick: jakogut_ -> jakogut
17:43:47 <ASPO> 90ms (touch) + 2-8ms (wifi) is < 100ms, which isn't doable per se
17:44:04 <ASPO> but the real important thing is they wouldn't notice
17:44:29 <lachlan_s> It'd be pretty cool
17:44:54 <ASPO> Ofc different phones have different ms delays, for example, some have 28ms
17:45:03 <lachlan_s> Yeah, true
17:45:14 <ASPO> Which is definitely doable because less than 60
17:45:26 <ASPO> It would be truely impossible to tell a lag
17:45:31 <ASPO> Because less than fps
17:46:08 <ASPO> but < 100 ms is definitely playable, especially when it comes to video games
17:46:17 --- join: ACE_Recliner (~ACE_Recli@c-73-18-225-48.hsd1.mi.comcast.net) joined #osdev
17:46:34 <ASPO> However many reports from a game called league of the difference between 24 and 100 ms ping is pretty phenominal, as they can actually tell
17:46:53 <ASPO> I think we are already geared to face 100ms lag tho
17:46:56 <ASPO> Yunno
17:47:04 <lachlan_s> Ordinary user won't notice especially because it's just tapping and whatnot
17:47:10 <lachlan_s> Not gaming mostly
17:47:11 --- join: drakonis_ (~drakonis@unaffiliated/drakonis) joined #osdev
17:47:35 <ASPO> I guarantee that the delay of shipping 1080p jpg or raw won't be noticable
17:47:50 <lachlan_s> More like 4k
17:47:54 <lachlan_s> But still
17:48:14 <ASPO> 4k the same as 1080p
17:48:22 <ASPO> You just can't go too far
17:48:30 --- quit: drakonis (Ping timeout: 250 seconds)
17:48:52 <ASPO> As soon as your ms hits over fps then you are lagging behind
17:49:22 <ASPO> because it takes 1.x frames to get the picture from the current drawn frame
17:49:27 <lachlan_s> Well, you'd need ~16 ms then
17:49:46 <ASPO> no you'd need anything < 60ms
17:49:59 <ASPO> Which is 120 ping
17:50:48 <lachlan_s> Why 60ms?
17:50:56 <lachlan_s> It's 60 fps, which is 16.6 ms between frames
17:50:58 <ASPO> Ping is to + from
17:51:32 --- quit: drakonis_ (Ping timeout: 250 seconds)
17:52:18 <ASPO> Well I guess
17:52:31 --- join: drakonis (~drakonis@unaffiliated/drakonis) joined #osdev
17:52:49 <ASPO> Ehhck I kinda messed up what I was saying
17:52:59 <lachlan_s> Alright, I think I got ya
17:53:44 --- join: vaibhav (~vnagare@125.16.97.123) joined #osdev
17:54:04 <ASPO> 120 ping is drawing 2 frames, 60 is the pipe to the display. 60ms is doable because its 30 pipe to the display, which is less than 2 times a frame, but still will be super lag (frameskip to the rescue)
17:54:28 <lachlan_s> Alright, I see
17:54:29 <lachlan_s> Yeah
17:54:34 --- quit: Belxjander (Ping timeout: 250 seconds)
17:55:46 <ASPO> 120 pipe to the display is 8, 60 is 4, 30 is 2, 15 is <1
17:56:24 <ASPO> 16.666666666666666666666666666667 is our goal to match perfect ping for 60fps.
17:56:40 --- join: ALowther (~alowther@68.200.236.134) joined #osdev
17:56:44 --- quit: drakonis (Ping timeout: 250 seconds)
17:56:51 <ASPO> so as long as you have less than 33 or 34 ping you'd be a-okay
17:57:18 --- join: Belxjander (~Belxjande@sourcemage/Mage/Abh-Elementalist) joined #osdev
17:57:27 <ASPO> Which is DEFINITELY doable, because I get like 8-10 with a shitty acer aspire
17:57:48 <ASPO> 8-10 ping, which is 4-5 pipe to
17:57:50 <lachlan_s> It should be doable to a city-wide datacenter as well
17:58:06 <ASPO> no
17:58:32 <ASPO> Too much risk
17:58:32 <lachlan_s> Yes
17:58:44 <lachlan_s> why
17:58:46 <ASPO> you'd then involve cryptographic functions of sha
17:58:52 <ASPO> similar to one time password
17:58:59 <ASPO> Which is a truncated hmac of a sha
17:59:12 <lachlan_s> Encrypted streams of data aren't a new thing
17:59:39 <lachlan_s> What's wrong with encrypting it?
17:59:47 <ASPO> Its pointless
18:00:26 <ASPO> Well I mean if you mean data center as if your entire computer is somewhere then...
18:00:45 <ASPO> Yeah perhaps a local data center is okay, but its still a shitty idea because security
18:01:04 --- quit: ALowther (Ping timeout: 250 seconds)
18:01:18 <ASPO> You'd still have to do the algorithm so people can't watch your porn with you
18:01:28 <lachlan_s> Yeah
18:02:37 <ASPO> So your 1080p has to arrive at your computer between the delay of the next frame pull request
18:02:48 --- quit: `Guest00000 (Ping timeout: 250 seconds)
18:02:56 <Belxjander> okay...challenge time...
18:02:58 <ASPO> and displayed at a uniform time
18:03:25 <Belxjander> I have a 10.1 inch Android Tablet where the "touch" function of the display is broken...
18:03:45 <lachlan_s> Plug in a mouse
18:03:50 <lachlan_s> I did that with my old nexus 7
18:03:52 <ASPO> What solution do you want other than to suicide?
18:04:11 <ASPO> '"touch" function of the display is broken...' unprecedented
18:04:21 --- quit: freakazoid0223 (Ping timeout: 246 seconds)
18:04:30 <Belxjander> would it be viable to sort out some means of replacing the android on it (the device hardware specs are open c/o Sony as it is an *original* Tablet Z)
18:04:33 <ASPO> I'd give up if my gtx 1080 I recently got just died
18:05:15 <Belxjander> lachlan_s: I already got a UB Hub that lets me use 3 normal USB devices from the tablet...
18:05:54 <Belxjander> I'm just considering what sort of hackable materials would be usable on the tablet if I were to use it as a barebones single function device like a wireless display for my laptop
18:05:59 --- join: chao-tic (~chao@203.97.21.86) joined #osdev
18:06:22 <ASPO> bel: use a potato keyboard
18:07:00 <ASPO> Kartofel ist sehr gut
18:08:01 <ASPO> Guys I think the biggest and most shittiest gap we would have to take to fight with this is
18:08:10 <ASPO> Learning to program android apps
18:08:13 <lachlan_s> oh no
18:08:34 <lachlan_s> I've strayed far away from writing apps for mobile devices
18:08:41 <ASPO> same
18:08:52 <ASPO> not as far away as html though
18:09:00 <ASPO> html needs replaced
18:09:09 <lachlan_s> I actually like it
18:09:12 * ASPO holds up torch
18:09:17 <lachlan_s> It has issues, but it's decently good at what it does
18:09:26 <ASPO> Not really
18:09:39 <ASPO> Its like python trying to be a programming language
18:09:52 <lachlan_s> Well, python *is* a programming language
18:10:02 <lachlan_s> And html is a markup language
18:10:08 <ASPO> You can find your way around a mass amount of fuckery to do what you want yes.
18:10:26 <ASPO> Python is a scripting language that isn't compiled to native machine language
18:10:35 <lachlan_s> uh huh
18:10:48 <ASPO> Python generates assembly huh
18:10:55 <ASPO> For my specific architecture?
18:11:15 <lachlan_s> I think pypy jits python
18:11:43 <ASPO> jits is better, yes, but not native machine language
18:12:13 <lachlan_s> Well, technically, it is yeah
18:12:13 <ASPO> are*
18:12:37 <ASPO> nah man they just optimize bytecode during runtime
18:12:46 <ASPO> bytecode != native machine language
18:13:13 <lachlan_s> Nah, pypy does actually jit
18:13:52 <ASPO> Its no
18:14:34 <lachlan_s> It's yes
18:14:41 <ASPO> Its not
18:14:52 <ASPO> Anyways
18:15:10 <ASPO> I might have to check out android development because I've been meaning to do a bit of it, and this would be a perfect intro
18:15:26 <lachlan_s> https://pypy.org
18:15:28 <bslsk05> pypy.org: PyPy - Welcome to PyPy
18:15:33 <lachlan_s> You should, yeah
18:15:35 --- join: MaryJaneInChain (~user@223.255.127.32) joined #osdev
18:15:48 <ASPO> k bye
18:15:51 --- quit: ASPO (Quit: Page closed)
18:17:32 <lachlan_s> Bye
18:17:40 <lachlan_s> Good talking to ya
18:23:05 --- join: ALowther (~alowther@68.200.236.134) joined #osdev
18:23:33 --- join: freakazoid0223 (~IceChat9@pool-108-52-244-197.phlapa.fios.verizon.net) joined #osdev
18:30:19 --- quit: Belxjander (Ping timeout: 240 seconds)
18:31:21 --- join: Belxjander (~Belxjande@sourcemage/Mage/Abh-Elementalist) joined #osdev
18:32:05 --- quit: _sfiguser (Quit: Leaving)
18:36:33 --- quit: promach2 (Ping timeout: 255 seconds)
18:37:19 --- quit: chao-tic (Ping timeout: 240 seconds)
18:41:19 --- join: MaryJane` (~user@223.255.127.32) joined #osdev
18:43:10 --- quit: MaryJaneInChain (Ping timeout: 250 seconds)
18:43:56 --- join: sdk93 (uid290141@gateway/web/irccloud.com/x-pevrrkhjxpaolztf) joined #osdev
18:44:32 --- join: `Guest00000 (~user@37.113.160.19) joined #osdev
18:47:40 <klange> Not sure if it's a vxworks thing or specific to this embedded system I use at work, but there's a shell that provides a very interesting bit of functionality: it just calls symbols with whatever arguments you give... I'm not sure if it understand argument sizes, but the concept is really amusing
18:48:40 <klange> like it implements some basic shell functions like `cd`... as a symbol for a function that takes a string argument, so you `cd "whatever"` and it calls cd("whatever")... you want to open and read a file? `fopen "file","r"` and it'll spit out the return value in decimal and hex and you can toss it at `fgetc`
18:48:42 --- quit: Belxjander (Ping timeout: 255 seconds)
18:50:29 <pixelherodev> Huh
18:50:37 <pixelherodev> Interesting
18:51:14 --- quit: tacco| ()
18:53:02 --- quit: ALowther (Remote host closed the connection)
18:54:00 --- join: Belxjander (~Belxjande@sourcemage/Mage/Abh-Elementalist) joined #osdev
19:04:12 --- quit: Belxjander (Ping timeout: 246 seconds)
19:04:21 --- join: drakonis (~drakonis@unaffiliated/drakonis) joined #osdev
19:05:04 --- join: SGautam (~GautamS@unaffiliated/gautams) joined #osdev
19:06:54 --- join: Belxjander (~Belxjande@sourcemage/Mage/Abh-Elementalist) joined #osdev
19:07:05 --- quit: Shikadi (Ping timeout: 256 seconds)
19:08:00 --- join: ALowther (~alowther@68.200.236.134) joined #osdev
19:11:56 --- join: NaNkeen (~nankeen@115.164.80.206) joined #osdev
19:13:30 --- join: promach_ (~phung@155.69.206.255) joined #osdev
19:13:50 <promach_> What do you guys think about the difference between perf and callgrind ?
19:15:42 --- join: Arcaelyx_ (~Arcaelyx@2604:2000:f14a:2500:6527:47c5:4de2:5f5d) joined #osdev
19:16:27 --- quit: NaNkeen (Ping timeout: 246 seconds)
19:18:12 --- quit: Arcaelyx (Ping timeout: 246 seconds)
19:26:56 --- quit: Belxjander (Ping timeout: 260 seconds)
19:27:57 <Mutabah> promach_: Iirc, perf is a sampling profiler, while callgrind is full instrumentation
19:28:10 <Mutabah> (actually, callgrind is simulation)
19:31:45 --- join: Belxjander (~Belxjande@sourcemage/Mage/Abh-Elementalist) joined #osdev
19:31:52 <promach_> simulation , huh ?
19:32:26 <Mutabah> well, pretty sure it does simulation to get its instruction-level counts
19:32:35 <Mutabah> might just be doing trickery with the trap flag though
19:32:48 <promach_> Mutabah: so, which one is more accurate and suitable ?
19:33:28 --- join: chao-tic (~chao@203.97.21.86) joined #osdev
19:33:28 <Mutabah> valgrind is accurate, but SLOOOOW
19:33:58 <Mutabah> I _think_ that perf can either do sampling, or use profiling hooks emitted by the compiler (but might want to check on that)
19:35:25 <SGautam> i think thats the point of valgrind
19:35:26 --- quit: Arcaelyx_ (Read error: Connection reset by peer)
19:35:27 --- quit: JusticeEX (Quit: Lost terminal)
19:37:15 --- quit: ALowther (Remote host closed the connection)
19:37:48 --- quit: chao-tic (Ping timeout: 246 seconds)
19:38:15 --- join: Arcaelyx (~Arcaelyx@2604:2000:f14a:2500:6527:47c5:4de2:5f5d) joined #osdev
19:42:10 <promach_> valgrind or callgrind ?
19:42:13 <promach_> I am confused now
19:44:49 --- quit: geordi (Remote host closed the connection)
19:46:56 <Mutabah> promach_: callgrind is a tool in the valrind suite
19:48:29 <geist> merp!
19:49:10 --- join: NaNkeen (~nankeen@203.188.234.181) joined #osdev
19:53:29 --- join: variable (~variable@freebsd/developer/variable) joined #osdev
20:00:54 --- join: geordi (~geordi@ns510081.ip-192-99-4.net) joined #osdev
20:01:24 <SGautam> herb!
20:07:36 --- quit: SGautam (Quit: Leaving)
20:10:22 --- join: chao-tic (~chao@203.97.21.86) joined #osdev
20:11:08 --- quit: trn (Ping timeout: 260 seconds)
20:13:05 --- join: spencerb (~quassel@c-73-106-187-191.hsd1.ga.comcast.net) joined #osdev
20:17:05 --- quit: nj0rd (Ping timeout: 256 seconds)
20:20:16 --- quit: variable (Quit: /dev/null is full)
20:20:30 --- quit: NaNkeen (Ping timeout: 246 seconds)
20:20:41 <geist> huh thats funny. a i386 ubuntu 17.10 server install i have has just suggsted and let me install an 18.04 upgrade
20:20:47 <geist> i thought 18.04 was no more on i386
20:20:55 <geist> it's doing the upgrade now
20:21:47 <klange> Hm?
20:22:33 <klange> 32-bit x86 builds of Ubuntu are still available, and the repos are still maintained.
20:22:42 <geist> there are just no more .isos
20:22:45 <klange> They just don't provide 32-bit desktop live CDs.
20:22:46 --- join: bemeurer (~bemeurer@2600:8802:5300:bb90:863a:4bff:fe06:c8b2) joined #osdev
20:22:53 <geist> no more server isos either
20:23:06 <geist> except it seems to be letting me upgrade to 18.04, which implies they'll maintain for 5 years
20:24:11 --- join: wowww (~weechat@unaffiliated/icetooth) joined #osdev
20:25:39 --- quit: freakazoid0223 (Quit: Never put off till tomorrow, what you can do the day after tomorrow)
20:26:25 --- join: NaNkeen (~nankeen@122.0.31.67) joined #osdev
20:31:58 --- join: nj0rd (~nj0rd@i577BC09C.versanet.de) joined #osdev
20:34:01 --- quit: NaNkeen (Ping timeout: 260 seconds)
20:34:55 --- quit: Griwes (Ping timeout: 264 seconds)
20:36:36 --- quit: Belxjander (Ping timeout: 246 seconds)
20:37:55 --- join: Griwes (~griwes@znc.reaver-project.org) joined #osdev
20:38:31 --- quit: jakogut (Ping timeout: 264 seconds)
20:42:02 --- join: Belxjander (~Belxjande@sourcemage/Mage/Abh-Elementalist) joined #osdev
20:42:50 --- join: latentprion (~latentpri@sel4/latentprion) joined #osdev
20:47:48 --- quit: drakonis (Read error: Connection reset by peer)
20:51:15 --- nick: wowww -> tubular
20:51:59 --- join: NaNkeen (~nankeen@203.188.234.181) joined #osdev
20:54:53 --- quit: tux3 (Ping timeout: 265 seconds)
20:58:02 --- join: trn (jhj@prone.ws) joined #osdev
20:58:45 --- quit: NaNkeen (Ping timeout: 255 seconds)
21:00:22 --- join: tux3 (tux@cmdline.org) joined #osdev
21:04:42 --- join: pounce (~pounce@c-24-11-27-195.hsd1.ut.comcast.net) joined #osdev
21:06:27 --- join: bcos (~bcos@CPE-58-170-185-157.lns3.win.bigpond.net.au) joined #osdev
21:07:13 --- part: tubular left #osdev
21:09:52 --- quit: bcos_ (Ping timeout: 265 seconds)
21:17:20 <izabera> Mutabah: what slowdown would you consider acceptable for a profiler accurate as callgrind, and what's too slow?
21:18:13 --- quit: GomJabbar (Ping timeout: 240 seconds)
21:20:00 <doug16k> callgrind causes 20x to 50x slowdown in my experience
21:23:03 --- quit: tux3 (Ping timeout: 255 seconds)
21:23:14 <geist> yah valgrind i think jits and runs, so there's a fair amount of slowdown
21:25:37 --- join: tux3 (tux@cmdline.org) joined #osdev
21:32:24 <doug16k> geist, I upgraded to 18.04 today. worked okay. I had to add myself to the kvm group to run qemu with kvm enabled
21:33:31 <geist> yah i've upgraded other machines to 18.04, just didn' tknow i386 was still on the table
21:33:39 <doug16k> it failed to detect hot plugging my headset once since upgrading. had to reboot
21:34:46 <doug16k> they supposedly updated the k10temp module to read ryzen cpu temp. I get nonsense reading though (91C)
21:35:47 --- join: GomJabbar (GomJabbar@gateway/shell/panicbnc/x-hmczxvwwhvhgrham) joined #osdev
21:36:22 <doug16k> I guess AMD is never making a family 17h BKDG :(
21:37:00 <geist> :(
21:38:03 <geist> i should use my contacts at work to get ahold of one
21:38:14 <geist> surely we have a relationship with AMD
21:38:24 <doug16k> probably
21:41:00 --- quit: jp (Disconnected by services)
21:41:10 --- join: jp (ayy@youlosethega.me) joined #osdev
21:41:18 <pounce> I dare someone to write an OS without branches
21:41:32 <klange> branches are overrated anyway
21:41:49 <pounce> completely immune to speculative execution bugs!
21:41:53 <klange> for best results it needs to be a virtual machine so you can guarantee there's no branches in userspace
21:43:10 --- quit: bork (Disconnected by services)
21:43:29 --- join: bork (ayy@cpe-76-173-133-37.hawaii.res.rr.com) joined #osdev
21:43:40 --- quit: lachlan_s (Quit: Connection closed for inactivity)
21:53:32 --- quit: epony (Read error: Connection reset by peer)
21:55:06 --- join: xenos1984 (~xenos1984@2001:bb8:2002:200:6651:6ff:fe53:a120) joined #osdev
21:55:45 --- quit: pounce (Quit: WeeChat 2.1)
21:56:03 --- quit: CrystalMath (Quit: Support Free Software - https://www.fsf.org/)
21:56:45 --- quit: Belxjander (Ping timeout: 246 seconds)
21:57:16 --- quit: xenos1984 (Client Quit)
21:57:19 --- quit: eam (Ping timeout: 240 seconds)
21:57:27 --- join: eam (~eam@li155-76.members.linode.com) joined #osdev
21:57:36 --- join: xenos1984 (~xenos1984@2001:bb8:2002:200:6651:6ff:fe53:a120) joined #osdev
21:58:03 --- join: epony (~nym@79-100-134-61.ip.btc-net.bg) joined #osdev
22:01:47 --- quit: xenos1984 (Client Quit)
22:01:51 <doug16k> amd seems to think the PPR _is_ the BKDG: https://community.amd.com/thread/220452
22:01:53 <bslsk05> community.amd.com: Where I can download BKDG for 17h? | Community
22:02:16 --- join: Belxjander (~Belxjande@sourcemage/Mage/Abh-Elementalist) joined #osdev
22:03:11 --- join: xenos1984 (~xenos1984@2001:bb8:2002:200:6651:6ff:fe53:a120) joined #osdev
22:11:55 --- quit: chao-tic (Quit: WeeChat 1.0.1)
22:18:49 --- quit: doug16k (Remote host closed the connection)
22:23:43 --- join: doug16k (~dougx@198-91-135-200.cpe.distributel.net) joined #osdev
22:25:17 --- quit: doug16k (Remote host closed the connection)
22:26:53 --- join: doug16k (~dougx@198-91-135-200.cpe.distributel.net) joined #osdev
22:30:06 --- quit: tux3 (Ping timeout: 255 seconds)
22:30:42 --- quit: doug16k (Remote host closed the connection)
22:31:11 --- join: Pseudonym73 (~Pseudonym@e-nat-unimelb-128-250-0-48.uniaccess.unimelb.edu.au) joined #osdev
22:32:16 --- join: tux3 (tux@cmdline.org) joined #osdev
22:32:19 --- join: doug16k (~dougx@198-91-135-200.cpe.distributel.net) joined #osdev
22:35:03 --- quit: Belxjander (Ping timeout: 255 seconds)
22:36:03 --- quit: doug16k (Remote host closed the connection)
22:36:16 <Mutabah> Anyone know if using a union in C to convert between a float and an integer is defined?
22:37:07 --- join: Belxjander (~Belxjande@sourcemage/Mage/Abh-Elementalist) joined #osdev
22:38:04 --- join: doug16k (~dougx@198-91-135-200.cpe.distributel.net) joined #osdev
22:40:10 <klange> I believe so.
22:40:22 <klange> It's often the defacto example of using a union to do weird type things.
22:41:20 <klange> I think you might want to make sure that int is unsigned and of the appropriate length, eg float + uint32_t, double + uint64_t
22:44:34 <Mutabah> Oh, yeah, sure. Just had this vauge memory of it breaking strict aliasing
22:45:17 <doug16k> memcpy is the safest way to type pun
22:45:35 <Mutabah> Yeah, that's what I'm doing
22:46:08 <doug16k> with builtin memcpy enabled, the compiler should see what you mean and do it optimally
22:52:48 --- join: andrei-n__ (~andrei@51.214-65-87.adsl-dyn.isp.belgacom.be) joined #osdev
22:54:06 --- part: GomJabbar left #osdev
22:58:27 --- quit: rw-r-r-0644[m] (*.net *.split)
23:02:11 --- quit: shirk (Ping timeout: 260 seconds)
23:02:21 --- join: shirk (~shirk@bitspin.org) joined #osdev
23:02:27 --- join: m_t (~m_t@p5DDA17AD.dip0.t-ipconnect.de) joined #osdev
23:04:46 --- quit: cruxeternus (Remote host closed the connection)
23:06:43 --- quit: `Guest00000 (Ping timeout: 264 seconds)
23:08:08 --- join: zhiayang_ (~zhiayang@124.197.97.227) joined #osdev
23:09:08 --- join: m3nt4L (~asvos@2a02:587:a023:f000:3285:a9ff:fe8f:665d) joined #osdev
23:09:11 --- quit: zhiayang (Ping timeout: 260 seconds)
23:16:54 --- quit: banisterfiend (Quit: My MacBook has gone to sleep. ZZZzzz…)
23:23:40 <doug16k> funny how my 1060 GPU never goes below 55C, but my CPU is usually around 34C, lol
23:27:09 --- join: AndrewBC_ (~AndrewBC@ip70-187-57-150.pn.at.cox.net) joined #osdev
23:28:38 <promach_> Could anyone tell what kind of division algorithm is used in https://gitrepos.estec.esa.int/taste/panda/blob/master/examples/CHStone/dfdiv/softfloat.c#L241 ?
23:28:39 <bslsk05> gitrepos.estec.esa.int: examples/CHStone/dfdiv/softfloat.c · master · TASTE / PandA · GitLab
23:29:13 <latentprion> a programmatic one
23:29:47 <promach_> ??
23:30:24 --- quit: OZ__ (Ping timeout: 255 seconds)
23:30:38 <klange> looks like a bunch of edge case handling and I think the actual division is down in the while loop at the bottom
23:30:48 --- quit: zeus1 (Quit: WeeChat 2.1)
23:31:06 <promach_> add128() ?
23:35:57 <klange> there and in this function: https://gitrepos.estec.esa.int/taste/panda/blob/master/etc/libbambu/softfloat/softfloat-macros#L546
23:35:58 <bslsk05> gitrepos.estec.esa.int: etc/libbambu/softfloat/softfloat-macros · master · TASTE / PandA · GitLab
23:36:12 <klange> but yes, loop over addition (or subtraction) is a standard approach to basic division
23:37:38 <promach_> I do not understand
23:39:12 <Mutabah> It's long division (if you remember maths classes from primary school)
23:45:06 --- join: quc (~quc@87.116.229.85) joined #osdev
23:46:17 --- join: w41 (~w41@unaffiliated/w41) joined #osdev
23:53:04 --- join: `Guest00000 (~user@37.113.160.19) joined #osdev
23:54:42 --- quit: ACE_Recliner (Remote host closed the connection)
23:57:56 <promach_> Mutabah: but (sbits64) rem0) < 0 does no make sense
23:59:59 --- log: ended osdev/18.05.09