--- Log for 24.10.122 Server: tungsten.libera.chat Channel: #rockbox --- Nick: rb-logbot Version: Dancer V4.16 Started: 2 days and 23 hours ago 00.01.01 Quit chris_s (Quit: Connection closed) 00.18.28 Quit massiveH (Read error: Connection reset by peer) 00.18.54 # <_bilgus> well we could always up the limit but there is like 65 -80 bytes per entry 00.19.24 # <_bilgus> based on my back of envelope calcs 00.23.29 *** Saving seen data "./dancer.seen" 00.37.10 Quit m01 (Quit: Konversation terminated.) 00.39.17 Join m01 [0] (~quassel@vps-b172b88b.vps.ovh.net) 01.19.23 Quit amiconn (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.) 01.19.23 Quit pixelma (Quit: .) 01.21.47 Join amiconn [0] (~jens@p200300ea874e1f000e5ba1111f349910.dip0.t-ipconnect.de) 01.21.50 Join pixelma [0] (marianne@p200300ea874e1f00305e95fffec66ff3.dip0.t-ipconnect.de) 02.04.39 Join chris_s [0] (~chris_s@ip-095-223-073-182.um35.pools.vodafone-ip.de) 02.06.22 # I was a bit surprised that the fds would even stay open, but looking at font_load() that appears to be by design 02.06.37 # "cached font: pf->fd stays open until the font is unloaded " 02.20.58 # 7 fonts sounds like a lot for a theme 02.23.26 # <_bilgus> agreed 02.23.32 *** Saving seen data "./dancer.seen" 02.23.33 # btw, it is not easy to find a theme by name or resolution unless i am searching wrong 02.24.26 # 'all themes' is the only way i found 02.25.16 # 1 typeface in 7 sizes/weights then 02.25.21 # <_bilgus> spork, at the bottom of https://themes.rockbox.org/ 'You can also directly search for specific themes here: '? 02.26.29 # sure, but like chris_s mentions a resolution that is not an index as such 02.26.49 # sorry, it is 02.26.57 # <_bilgus> font scaling would probably be a good add 02.27.07 # i would bump that search to the top of the page 02.28.18 # hmm, i have a 320x240 device, i must try that theme 02.28.31 # <_bilgus> eh most people aren't searching for a specific theme persay they probably scroll through the pics first 02.28.49 # most people are not most people 02.29.09 # especially with rockbox, everybody has their own usage 02.29.41 # <_bilgus> my point being just as many never need the search fuction as do 02.30.06 # of course 02.30.27 # but it is 2 lines and now it is 'buried' at the bottom 02.31.05 # i found it, i will be good bow 02.31.07 # now 02.31.51 # Maybe there should be a warning when loading a theme that could result in unexpected behavior due to all fds being used up 02.32.38 # <_bilgus> ick 02.33.22 # <_bilgus> msybe a splash about no file handles left 02.33.29 # <_bilgus> if there isn't one 02.33.37 # yeah 02.34.08 # maybe not when loading the theme 02.34.11 # make the number of fd's dependent on the available memory 02.34.20 # if memory usage is the concern 02.34.41 # that specific theme is going to run on the less constrained devices 02.35.31 # <_bilgus> its probably not being that the screen size is pretty big 02.35.37 # although if you're just showing a splash when/that there are no fd's left, the user may not realize it has anything to do with the theme used 02.36.39 # <_bilgus> no probably not but it seems a lot of code to do it just based on the theme font number rather than just having a splash that covers all cases 02.36.51 # true 02.37.14 # <_bilgus> but i'd go for increasing the fds for the bigger memory devices 02.37.18 # <_bilgus> per spork 02.37.40 # <_bilgus> although whats to stop the next theme from using most of those 02.38.15 # <_bilgus> better would probably be to force unloading some of the cache as well 02.39.13 # <_bilgus> make an event EV_FD_DEPLETED or something 02.39.40 # <_bilgus> have the cache register to listen and kick out the oldest 02.39.48 # ipod video has 32MB so a couple of extra 80 bytes would not hurt 02.46.18 Quit paulk (Ping timeout: 264 seconds) 02.46.40 Join paulk [0] (~paulk@vpn-0-22.aquilenet.fr) 02.47.10 Quit chris_s (Quit: Connection closed) 02.55.04 # <_bilgus> g#4738 02.55.07 # 3Gerrit review #4738 at https://gerrit.rockbox.org/r/c/rockbox/+/4738 : 3increase the number of file and directory handles for devices with more ram by William Wilgus 02.56.55 # <_bilgus> I *think* thats proper 02.56.59 # i would argue <32 and >=32 as on 32 it would be fine 02.58.34 # and there are no devices with 33 or so, the next step probably is 64 02.59.36 # <_bilgus> i upped all except the < 8 mb 03.00.49 # <_bilgus> i just put in a little less for 16mb or less 03.00.52 # yes, i get that 03.01.42 # just do not think the <=32 is needed and <32 would do too 03.02.07 # ipodvideo had 32 i think 03.02.42 # <_bilgus> i'd rather be a little conservative on devices I can't test 03.02.56 # fair enough 03.16.01 # <_bilgus> well the one peoblem I see with having a splash when you run out of descriptors is it'd probably make the device wholy unusable with the splashes 03.16.25 # <_bilgus> so I think maybe limiting the number of times it gets displayed prudent 03.25.02 Join chris_s [0] (~chris_s@ip-095-223-073-182.um35.pools.vodafone-ip.de) 03.31.14 # nice – one possible edge case is that grayscale devices may crash when the grayscale overlay (lib/grey.h) is running at the time that the splash is displayed. But that's probably a negligible concern that applies to other scenarios where an error splash is displayed as well. 03.32.37 # sorry, apps/plugins/lib/grey.h 03.33.53 # number of fonts used is probably not an issue there 03.42.03 # <_bilgus> mmm that is a good point I could see running out of handles in a plugin 04.23.35 *** Saving seen data "./dancer.seen" 05.05.24 Quit chris_s (Quit: Connection closed) 06.23.39 *** Saving seen data "./dancer.seen" 08.23.40 *** No seen item changed, no save performed. 10.23.44 *** No seen item changed, no save performed. 11.38.35 Join othello7 [0] (~Thunderbi@pool-100-36-133-223.washdc.fios.verizon.net) 12.23.49 *** No seen item changed, no save performed. 12.44.02 Quit paulcarroty (Quit: Bridge terminating on SIGTERM) 12.44.02 Quit MayeulC (Quit: Bridge terminating on SIGTERM) 12.44.03 Quit e-user[m] (Quit: Bridge terminating on SIGTERM) 12.44.03 Quit popcorn9499[m] (Quit: Bridge terminating on SIGTERM) 12.44.03 Quit blbro[m] (Quit: Bridge terminating on SIGTERM) 12.44.13 Quit underpantsgnome[ (Quit: Bridge terminating on SIGTERM) 12.48.26 Join MayeulC [0] (~mayeulc@2001:470:69fc:105::35e) 12.54.47 Join paulcarroty [0] (~paulcarro@2001:470:69fc:105::216) 12.54.48 Join underpantsgnome[ [0] (~tinybronc@2001:470:69fc:105::2:1af6) 12.54.54 Join popcorn9499[m] [0] (~popcorn94@2001:470:69fc:105::2:11cc) 12.55.00 Join e-user[m] [0] (~e-usermat@2001:470:69fc:105::1:2763) 12.55.21 Join blbro[m] [0] (~blbrostra@2001:470:69fc:105::8f7) 14.23.52 *** Saving seen data "./dancer.seen" 15.47.41 Join lebellium [0] (~lebellium@2a01cb040109a6003c06b1e9d9cadce8.ipv6.abo.wanadoo.fr) 16.19.30 Quit kirvesAxe (*.net *.split) 16.19.30 Quit gevaerts (*.net *.split) 16.19.30 Quit Xeha (*.net *.split) 16.19.31 Quit user890104 (*.net *.split) 16.19.40 Join user890104 [0] (~Venci@freemyipod/user890104) 16.20.06 Join kirvesAxe [0] (kirvesaxe@user/kirvesaxe) 16.22.07 Join Xeha [0] (~Xeha@user/Xeha) 16.23.54 *** Saving seen data "./dancer.seen" 16.24.48 Join gevaerts [0] (~fg@user/gevaerts) 17.11.01 Quit mixfix41 (Ping timeout: 272 seconds) 18.23.55 *** Saving seen data "./dancer.seen" 18.34.07 Quit lebellium (Quit: Leaving) 19.27.25 Quit rb-bluebot (Ping timeout: 252 seconds) 19.27.25 Quit bluebrother (Ping timeout: 252 seconds) 19.29.05 Join bluebrother [0] (~dom@user/bluebrother) 19.42.01 Join rb-bluebot [0] (~rb-bluebo@rockbox/bot/utility) 20.23.58 *** Saving seen data "./dancer.seen" 21.10.43 Quit yosafbridge (K-Lined) 21.25.27 Join yosafbridge [0] (~yosafbrid@static.38.6.217.95.clients.your-server.de) 21.37.12 Quit kakaszsz (Ping timeout: 260 seconds) 21.51.10 Join kakaszsz [0] (~koniu@cpc107003-dals23-2-0-cust91.20-2.cable.virginm.net) 22.00.36 Quit tomato (Quit: The Lounge - https://thelounge.chat) 22.01.22 Join tomato [0] (~tomato@user/tomato) 22.24.01 *** Saving seen data "./dancer.seen" 22.34.37 Quit othello7 (Quit: othello7) 22.47.25 Join othello7 [0] (~Thunderbi@pool-100-36-133-223.washdc.fios.verizon.net)