Logs on 2021-10-12 (liberachat/#xmonad)
| 00:06:23 | × | banc quits (banc@gateway/vpn/airvpn/banc) (Ping timeout: 268 seconds) |
| 01:07:26 | × | catman quits (~catman@user/catman) (Ping timeout: 268 seconds) |
| 01:09:49 | → | catman joins (~catman@user/catman) |
| 01:34:21 | × | haskl[error] quits (~haskl@user/haskl) (Read error: Connection reset by peer) |
| 01:34:34 | → | haskl joins (~haskl@user/haskl) |
| 01:34:35 | × | vanvik quits (~vanvik@82.194.204.210) (Quit: Ping timeout (120 seconds)) |
| 01:34:55 | → | vanvik joins (~vanvik@82.194.204.210) |
| 01:35:20 | × | Rockj\ quits (~rockj@chromie.geekrevolution.net) (Ping timeout: 260 seconds) |
| 01:35:57 | → | Rockj\ joins (~rockj@chromie.geekrevolution.net) |
| 02:14:07 | <jakefromstatefar> | I'm working on the tree rewrite. I've started constructing data types, so far it's going great... |
| 02:36:19 | → | nomadxx3 joins (~lanomadx@208.91.67.219) |
| 02:37:10 | × | nomadxxxx3 quits (~lanomadx@69.167.36.238) (Ping timeout: 252 seconds) |
| 02:45:56 | × | humky quits (~humky@user/humky) (Ping timeout: 245 seconds) |
| 02:46:24 | → | humky joins (~humky@user/humky) |
| 02:50:55 | × | td_ quits (~td@94.134.91.135) (Ping timeout: 252 seconds) |
| 02:52:27 | → | td_ joins (~td@muedsl-82-207-238-248.citykom.de) |
| 02:56:25 | × | catman quits (~catman@user/catman) (Ping timeout: 252 seconds) |
| 04:02:44 | → | catman joins (~catman@user/catman) |
| 06:14:30 | → | wonko joins (~wjc@62.115.229.50) |
| 06:27:06 | × | wonko quits (~wjc@62.115.229.50) (Ping timeout: 260 seconds) |
| 06:34:39 | × | lambdabot quits (~lambdabot@haskell/bot/lambdabot) (*.net *.split) |
| 06:34:39 | × | slep quits (~slep@cpc150002-brnt4-2-0-cust437.4-2.cable.virginm.net) (*.net *.split) |
| 06:34:39 | × | eyenx quits (~m-5paiux@eyenx.ch) (*.net *.split) |
| 06:34:39 | × | rieper quits (~riepernet@sxbeta1.geo.uni-leipzig.de) (*.net *.split) |
| 06:34:39 | × | deebo_ quits (~globe@stonebay32.com) (*.net *.split) |
| 06:34:39 | × | piele quits (~piele@tbonesteak.creativeserver.net) (*.net *.split) |
| 06:34:39 | × | Buliarou1 quits (~gypsydang@185.207.166.57) (*.net *.split) |
| 06:34:39 | × | zawaken quits (~zawaken@user/zawaken) (*.net *.split) |
| 06:38:31 | → | mc47 joins (~mc47@xmonad/TheMC47) |
| 06:56:19 | → | lambdabot joins (~lambdabot@haskell/bot/lambdabot) |
| 06:56:19 | → | slep joins (~slep@cpc150002-brnt4-2-0-cust437.4-2.cable.virginm.net) |
| 06:56:19 | → | zawaken joins (~zawaken@user/zawaken) |
| 06:56:19 | → | eyenx joins (~m-5paiux@eyenx.ch) |
| 06:56:19 | → | rieper joins (~riepernet@sxbeta1.geo.uni-leipzig.de) |
| 06:56:19 | → | deebo_ joins (~globe@stonebay32.com) |
| 06:56:19 | → | piele joins (~piele@tbonesteak.creativeserver.net) |
| 06:56:19 | → | Buliarou1 joins (~gypsydang@185.207.166.57) |
| 07:05:43 | × | haskl quits (~haskl@user/haskl) (Ping timeout: 268 seconds) |
| 07:07:44 | → | haskl joins (~haskl@user/haskl) |
| 07:30:29 | × | mc47 quits (~mc47@xmonad/TheMC47) (Quit: Leaving) |
| 07:37:16 | × | Solitary quits (~Solitary@user/solitary) (Ping timeout: 265 seconds) |
| 07:59:31 | × | werneta quits (~werneta@70-142-214-115.lightspeed.irvnca.sbcglobal.net) (Ping timeout: 252 seconds) |
| 08:02:01 | → | cfricke joins (~cfricke@user/cfricke) |
| 08:04:35 | × | geekosaur quits (~geekosaur@xmonad/geekosaur) (Killed (NickServ (GHOST command used by allbery_b))) |
| 08:04:35 | → | allbery_b joins (~geekosaur@xmonad/geekosaur) |
| 08:04:38 | allbery_b | is now known as geekosaur |
| 08:17:37 | × | geekosaur quits (~geekosaur@xmonad/geekosaur) (Killed (NickServ (GHOST command used by allbery_b))) |
| 08:17:37 | → | allbery_b joins (~geekosaur@xmonad/geekosaur) |
| 08:17:40 | allbery_b | is now known as geekosaur |
| 09:21:56 | × | electr0n quits (~electr0n@about/security/founder/electr0n) (Remote host closed the connection) |
| 09:56:53 | → | Solitary joins (~Solitary@user/solitary) |
| 10:02:34 | → | banc joins (banc@gateway/vpn/airvpn/banc) |
| 10:07:29 | → | qbt joins (~qbt@user/edun) |
| 10:07:49 | ← | qbt parts (~qbt@user/edun) () |
| 11:42:52 | → | dschrempf joins (~dominik@070-207.dynamic.dsl.fonira.net) |
| 11:50:41 | → | azg256 joins (~user@78-56-98-5.static.zebra.lt) |
| 12:05:17 | × | geekosaur quits (~geekosaur@xmonad/geekosaur) (Remote host closed the connection) |
| 12:05:36 | → | geekosaur joins (~geekosaur@xmonad/geekosaur) |
| 13:35:56 | × | azg256 quits (~user@78-56-98-5.static.zebra.lt) (Remote host closed the connection) |
| 13:45:10 | → | azg256 joins (~user@78-56-98-5.static.zebra.lt) |
| 13:55:11 | × | azg256 quits (~user@78-56-98-5.static.zebra.lt) (Remote host closed the connection) |
| 14:03:29 | → | azg256 joins (~user@78-56-98-5.static.zebra.lt) |
| 14:08:19 | × | azg256 quits (~user@78-56-98-5.static.zebra.lt) (Quit: ERC (IRC client for Emacs 27.1)) |
| 14:25:17 | × | cfricke quits (~cfricke@user/cfricke) (Quit: WeeChat 3.3) |
| 14:25:47 | × | geekosaur quits (~geekosaur@xmonad/geekosaur) (Remote host closed the connection) |
| 14:27:58 | → | geekosaur joins (~geekosaur@xmonad/geekosaur) |
| 14:33:49 | → | defjam joins (~eb0t@90.213.43.87) |
| 14:50:44 | × | Eoco quits (~ian@x-160-94-179-157.acm.umn.edu) (Ping timeout: 245 seconds) |
| 14:51:43 | → | greyhash joins (~greyhash@198.62.219.7) |
| 14:52:30 | × | dschrempf quits (~dominik@070-207.dynamic.dsl.fonira.net) (Quit: WeeChat 3.3) |
| 14:53:22 | → | Eoco joins (~ian@x-160-94-179-157.acm.umn.edu) |
| 15:17:19 | × | greyhash quits (~greyhash@198.62.219.7) (Ping timeout: 252 seconds) |
| 15:36:42 | → | mc47 joins (~mc47@xmonad/TheMC47) |
| 15:41:44 | → | seschwar joins (~seschwar@user/seschwar) |
| 15:42:11 | → | werneta joins (~werneta@70-142-214-115.lightspeed.irvnca.sbcglobal.net) |
| 15:46:25 | × | defjam quits (~eb0t@90.213.43.87) (Ping timeout: 252 seconds) |
| 15:48:06 | → | defjam joins (~eb0t@90.213.43.87) |
| 16:04:21 | → | greyhash joins (~greyhash@198.62.219.18) |
| 16:09:48 | → | alternateved joins (~user@staticline-31-183-147-179.toya.net.pl) |
| 16:21:20 | <jakefromstatefar> | Thinking of changing stacksets list items to something from Data.Vector. Running operations with a large stack becomes slow... |
| 16:22:36 | <geekosaur> | oy, does it |
| 16:23:26 | geekosaur | left his desktop session running overnight once, about a decade ago, and forgot to configure something not to spam him. ended up nuking the session from outside |
| 16:23:48 | <geekosaur> | couple thousand windows overnight |
| 16:29:58 | × | greyhash quits (~greyhash@198.62.219.18) (Ping timeout: 268 seconds) |
| 16:44:04 | <Solid> | tbh, at the sizes we're normally dealing with here lists are completely fine (the use of strings is, imo, much more egregious) |
| 16:49:50 | × | mc47 quits (~mc47@xmonad/TheMC47) (Remote host closed the connection) |
| 16:53:59 | skewerr | is now known as spoonm |
| 17:02:55 | → | electr0n joins (~electr0n@about/security/founder/electr0n) |
| 17:18:19 | <jakefromstatefar> | Usually, it's not that big of a deal. But, I think it should be prevented against... Especially if it's not too hard to do. |
| 17:19:47 | <liskin> | one of the undocumented goals of xmonad is to have an easy to understand codebase with few dependencies |
| 17:20:02 | <liskin> | this makes it easy for non-Haskellers to get stuff done in their configs |
| 17:20:26 | <liskin> | and it also makes xmonad a good starting point in learning Haskell |
| 17:21:35 | <liskin> | so using Data.Vector in particular would probably need to be justified by an existence of an actual performance problem that it solves |
| 17:22:22 | <jakefromstatefar> | Unless I'm mistaken, it is a performance problem, caused by the use of lists. Not an overly frequent one, but nevertheless. |
| 17:23:33 | <Solid> | I'd really like to see a reproducible example of that |
| 17:23:35 | <liskin> | (although I do have to admit that I've considered whether the zippers used in StackSet really are the best data structure: we walk them over and over so many times that the time saved by not having to reconstruct the whole list on the infrequent update is probably negligible) |
| 17:24:01 | <liskin> | jakefromstatefar: I think we have a serious misunderstanding of what "existence of an actual performance problem" means :-) |
| 17:25:05 | <liskin> | as far as I understand it, an actual performance problem exists if there are real people reporting a problem that they encounter during real usage of the software |
| 17:26:49 | <liskin> | so far we have one real geekosaur saying that his ancient configuration broke once with thousands of windows, and I'm somewhat certain that it's a scenario that's likely to break Xorg/picom/compton/xcompmgr far sooner than xmonad |
| 17:27:43 | <jakefromstatefar> | Solid: Open 20 windows, in a layout that resizes frequently, like accordion. Now, try to switch windows, notice the delay? Try to close windows quickly, count how many times you press your close keybind. Notice, that it misses input. |
| 17:27:51 | <geekosaur> | yeh, thousands of windows is not at all a normal use case for X11, much less window managers |
| 17:28:31 | <jakefromstatefar> | I'm not speaking of thousands, 15+ is when the problem I'm describing arises. Most noticeable at 20-25+. |
| 17:28:57 | <jakefromstatefar> | Hiding windows via limiting the windows in a layout doesn't fix this problem either. |
| 17:33:42 | <liskin> | That may actually be a bug in the specific layout. |
| 17:33:46 | <Solid> | there is absolutely no way that a linked list of length 25 is giving anyone performance problems (otherwise, we'd _really_ be in big trouble with our use of strings :)) |
| 17:34:34 | <liskin> | grid with 50+ windows works rather well here |
| 17:35:08 | <liskin> | although I'm fairly sure it would be a complete disaster if I upgraded picom, as it got bloated a lot in yshui's hands |
| 17:36:05 | <Solid> | (also, switching windows is O(1) amortised in our case anyways, so if you're having slowness issues with that it's definitely something else) |
| 17:36:09 | <liskin> | it can also be related to logging to xmobar: there is or was code that fetches window titles of all windows, even if they're not used in the PP |
| 17:37:05 | <liskin> | so yeah, that's definitely a bug worth playing with, but some of us would be real money on the fix not being anywhere near StackSet :-) |
| 17:37:58 | <Solid> | the window title stuff should live in some logger now and not in the dynamicLogString (?) impl anymore |
| 17:38:47 | <liskin> | Solid: oh, indeed! good. |
| 17:39:37 | <liskin> | my xmonad.hs does that anyway, though, and still handles those 60 windows fine (although fine is subjective here; different people have different latency perceptions) |
| 17:41:54 | <liskin> | (if fetching of window titles is proven by profiling to be culprit, that'd be an easy fix: cache it and listen for PropertyEvents) |
| 18:06:17 | × | geekosaur quits (~geekosaur@xmonad/geekosaur) (Remote host closed the connection) |
| 18:08:06 | → | geekosaur joins (~geekosaur@xmonad/geekosaur) |
| 18:13:02 | → | greyhash joins (~greyhash@198.62.219.18) |
| 18:14:16 | → | wonko joins (~wjc@62.115.229.50) |
| 18:18:14 | × | geekosaur quits (~geekosaur@xmonad/geekosaur) (Remote host closed the connection) |
| 18:18:36 | → | geekosaur joins (~geekosaur@xmonad/geekosaur) |
| 18:20:29 | × | geekosaur quits (~geekosaur@xmonad/geekosaur) (Remote host closed the connection) |
| 18:20:42 | → | geekosaur joins (~geekosaur@xmonad/geekosaur) |
| 18:23:28 | × | geekosaur quits (~geekosaur@xmonad/geekosaur) (Remote host closed the connection) |
| 18:23:56 | → | geekosaur joins (~geekosaur@xmonad/geekosaur) |
| 18:38:39 | × | geekosaur quits (~geekosaur@xmonad/geekosaur) (Killed (NickServ (GHOST command used by allbery_b))) |
| 18:38:39 | → | allbery_b joins (~geekosaur@xmonad/geekosaur) |
| 18:38:42 | allbery_b | is now known as geekosaur |
| 18:49:34 | × | geekosaur quits (~geekosaur@xmonad/geekosaur) (Ping timeout: 252 seconds) |
| 19:04:31 | → | geekosaur joins (~geekosaur@xmonad/geekosaur) |
| 19:08:24 | → | a6a45081-2b83 joins (~aditya@2601:249:4300:1296:88ec:cc73:84d4:1507) |
| 19:14:07 | × | geekosaur quits (~geekosaur@xmonad/geekosaur) (Killed (NickServ (GHOST command used by allbery_b))) |
| 19:14:07 | → | allbery_b joins (~geekosaur@xmonad/geekosaur) |
| 19:14:10 | allbery_b | is now known as geekosaur |
| 19:18:09 | × | geekosaur quits (~geekosaur@xmonad/geekosaur) (Killed (NickServ (GHOST command used by allbery_b))) |
| 19:18:12 | → | allbery_b joins (~geekosaur@xmonad/geekosaur) |
| 19:18:16 | allbery_b | is now known as geekosaur |
| 19:21:53 | × | abhixec quits (~abhixec@c-67-169-139-16.hsd1.ca.comcast.net) (Remote host closed the connection) |
| 19:37:55 | × | greyhash quits (~greyhash@198.62.219.18) (Ping timeout: 250 seconds) |
| 20:26:32 | <FOSSHuman[m]> | <liskin> "although I'm fairly sure it..." <- Do you use a fork of Picom?? Wondering bcs I use yshui's version of Picom and it slows down resizing of windows alot of windows are opened.. (If I kill Picom, window resizing is quick again even with more than 6 windows) |
| 20:26:47 | <FOSSHuman[m]> | * of windows when alot of |
| 20:27:19 | geekosaur | still uses compton becuase he doesn't like the things he keeps hearing about picom |
| 20:28:20 | <FOSSHuman[m]> | lol, just went with Picom bcs I guess I thought it was more "modern" and "faster" maybe?? |
| 20:33:23 | × | wonko quits (~wjc@62.115.229.50) (Ping timeout: 250 seconds) |
| 20:41:25 | <liskin> | FOSSHuman[m]: not a fork, just an old version |
| 20:41:45 | <liskin> | 7.4 is what I have here, that's still pretty fast |
| 20:42:09 | <liskin> | I don't need any blurs or shadows or anything, just basic compositing |
| 20:42:36 | <liskin> | I don't even remember why I updated from the original compton |
| 21:17:48 | × | a6a45081-2b83 quits (~aditya@2601:249:4300:1296:88ec:cc73:84d4:1507) (Read error: Connection reset by peer) |
| 21:50:32 | × | seschwar quits (~seschwar@user/seschwar) (Quit: :wq) |
| 23:03:26 | × | scardinal quits (~supreme@customer-212-237-101-39.ip4.gigabit.dk) (Ping timeout: 245 seconds) |
| 23:14:27 | × | alternateved quits (~user@staticline-31-183-147-179.toya.net.pl) (Ping timeout: 265 seconds) |
| 23:15:14 | → | scardinal joins (~supreme@customer-212-237-101-39.ip4.gigabit.dk) |
All times are in UTC on 2021-10-12.