| minute | screenshot from my newly upgraded reform https://mastodon.social/@mntmn/110703398144982527 | 00:11 |
|---|---|---|
| minute | now (not now now, but later now) would be a good time to explore compressed memory | 00:22 |
| josch | did you... try suspend? ;) | 00:24 |
| minute | i did not! and i will not!!!! | 00:28 |
| minute | ahaha | 00:28 |
| minute | no, i will look at it at some point in the next weeks... it should be possible in theory | 00:28 |
| josch | no worries, i'm not having my hopes up :) | 00:29 |
| minute | that's always good when facing suspend challenges | 00:31 |
| josch | and vagrant gave me another interesting idea earlier... once u-boot support is upstreamed, we can use u-boot-menu to choose the correct dtb to load before booting linux | 00:31 |
| josch | then you can literally just swap out the module and re-use your system without changes | 00:32 |
| minute | ah, i have a bit broken system atm | 00:33 |
| minute | > linux-image-arm64 : Depends: linux-image-6.4.0-0-reform2-arm64 (= 6.4.1-1~exp1+reform20230710T195054Z1) but 6.4.1-1~exp1+reform20230712T185438Z1 is to be installed | 00:33 |
| minute | because i didn't install all the .debs when i was still at work | 00:33 |
| minute | i wonder if i can silence this until tomorrow and install blueman anyway somehow | 00:33 |
| josch | you can remove linux-image-arm64 | 00:33 |
| josch | just don't forget to re-install it later | 00:33 |
| josch | the actual data is in linux-image-6.4.0-0-reform2-arm64 | 00:34 |
| josch | run this: | 00:34 |
| josch | dpkg -L linux-image-arm64 | 00:34 |
| josch | then you see that the package doesn't contain anything | 00:34 |
| josch | it's just a package that always depends on the latest kernel | 00:34 |
| josch | the package exists so that one can just depend on linux-image-arm64 and get the latest kernel without hardcoding version numbers | 00:34 |
| jfred | being able to just swap out modules without changing anything on-disk is the dream :) | 00:43 |
| jfred | (well, really if you could do that with upstream distro images too... I'm sure that's a lot more work haha) | 00:44 |
| josch | jfred: mainly missing for that is all the patches being upstreamed | 00:47 |
| minute | josch: thank you very much, that worked | 01:19 |
| violet | minute: if you're getting into zram my recommendation generally is to use zstd as the compression algorithm and set the zram device size to 3/4 your physical memory size. this usually works most of the time. you can sometimes go higher depending on the workload (I used to run 6/4) but i found that accidentally loading some .tar.gzs into tmpfs while already under heavy memory pressure froze my system on | 01:22 |
| violet | more than one occasion so i dropped back down. | 01:22 |
| + kuno_ (sid299442@id-299442.helmsley.irccloud.com) | 01:25 | |
| + skalk_ (~skalk@vond.sysret.de) | 01:25 | |
| + sterni_ (~lukas@user/sterni) | 01:26 | |
| + rah_ (rah@verain.settrans.net) | 01:27 | |
| + violet_ (~vi@user/violet) | 01:27 | |
| + nocko_ (~nock@user/nocko) | 01:27 | |
| violet_ | idk if any of that went through so ill copy paste into a log | 01:27 |
| - S0rin (QUIT: Ping timeout: 240 seconds) (~S0rin@user/s0rin) | 01:28 | |
| violet_ | minute: https://dpaste.com/BD6ZPL9FF zram recommendations (ive been using it for a long time) | 01:28 |
| violet_ | in case it's interesting | 01:29 |
| + S0rin (~S0rin@user/s0rin) | 01:30 | |
| - kuno (QUIT: *.net *.split) (sid299442@id-299442.helmsley.irccloud.com) | 01:32 | |
| - sevan (QUIT: *.net *.split) (~sevan@user/venture37) | 01:32 | |
| - sterni (QUIT: *.net *.split) (~lukas@user/sterni) | 01:32 | |
| - rah (QUIT: *.net *.split) (rah@verain.settrans.net) | 01:32 | |
| - skalk (QUIT: *.net *.split) (~skalk@vond.sysret.de) | 01:32 | |
| - leonardo (QUIT: *.net *.split) (leonardo@user/leonardo) | 01:32 | |
| - nocko (QUIT: *.net *.split) (~nock@user/nocko) | 01:32 | |
| - violet (QUIT: *.net *.split) (~vi@user/violet) | 01:32 | |
| * violet_ -> violet | 01:32 | |
| * kuno_ -> kuno | 01:32 | |
| + leonardo (leonardo@user/leonardo) | 01:34 | |
| - mjw (QUIT: Ping timeout: 245 seconds) (~mjw@gnu.wildebeest.org) | 01:34 | |
| minute | violet: very good info, thanks | 01:39 |
| + sevan (~sevan@2001:470:1f1d:1d6:5a55:caff:fe24:ed4) | 01:43 | |
| - mtm- (QUIT: Ping timeout: 246 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 02:02 | |
| + bkeys (~Thunderbi@static-198-54-130-101.cust.tzulo.com) | 02:14 | |
| * sterni_ -> sterni | 02:22 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 02:38 | |
| - yewscion (QUIT: Remote host closed the connection) (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 02:38 | |
| - ajr (QUIT: Quit: Connection closed for inactivity) (uid609314@user/ajr) | 03:27 | |
| jfred | how significant are the performance gains from the laird antenna? | 03:39 |
| jfred | think I'm finally going to actually order the a311d module and if the new antenna makes a significant difference I might throw it onto that order too | 03:40 |
| josch | i didn't measure it but for me the difference is between "no reception in the bedroom" and "same speed in the bedroom as everywhere else in the house" | 03:40 |
| jfred | seems pretty helpful then | 03:40 |
| josch | same in my office at work: with the old antenna i didn't have reception in my office and with the laird antenna i had the same flaky connection that i had with my intel notebook before | 03:40 |
| jfred | might as well tack it on, it doesn't add that much more to the cost | 03:41 |
| jfred | though would feel like a waste to peel the old antenna off its holder and junk it... maybe I'll also laser cut a new holder for it at the hackerspace :P | 03:42 |
| + sigrid_ (~sigrid@ftrv.se) | 03:57 | |
| - sigrid (QUIT: Quit: Reconnecting) (~sigrid@ftrv.se) | 03:58 | |
| - mlarkin (QUIT: Quit: leaving) (~mlarkin@047-036-115-056.res.spectrum.com) | 04:03 | |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 04:08 | |
| violet | usb2 is honestly kinda more fine that a lot of ppl give it credit for a think | 04:09 |
| violet | a lot of usb2's bad rep is really really bad usb drives that already ran at way below usb2's data rate | 04:09 |
| jfred | yeah | 04:17 |
| jfred | and also thinking about it I honestly don't often need fast USB drives anyway | 04:18 |
| jfred | slow USB transfer speeds are a bit painful if I'm writing a live USB or something, but I don't actually do that very often | 04:18 |
| + mlarkin (~mlarkin@047-036-115-056.res.spectrum.com) | 04:47 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 05:00 | |
| - S0rin (QUIT: Ping timeout: 246 seconds) (~S0rin@user/s0rin) | 05:02 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 05:03 | |
| + S0rin (~S0rin@user/s0rin) | 05:04 | |
| violet | and for everytihng else you basically dont benefit from usb3 | 05:27 |
| violet | like even a wifi dongle, you're not gonna saturate the 480mbit/s of usb2 unless you have a particularly optimized wifi setup. a gigabit ethernet adapter yeah you will | 05:27 |
| + yewscion (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 05:38 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 05:51 | |
| - jjbliss (QUIT: Ping timeout: 246 seconds) (~jjbliss@1464766-static.elnsmiaa.metronetinc.net) | 06:10 | |
| + jjbliss (~jjbliss@1464766-static.elnsmiaa.metronetinc.net) | 06:10 | |
| - sevan (QUIT: Ping timeout: 246 seconds) (~sevan@2001:470:1f1d:1d6:5a55:caff:fe24:ed4) | 06:12 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 07:07 | |
| - klardotsh (QUIT: Ping timeout: 245 seconds) (~klardotsh@98.97.113.255) | 07:13 | |
| + klardotsh (~klardotsh@98.97.113.255) | 07:18 | |
| - S0rin (QUIT: Ping timeout: 245 seconds) (~S0rin@user/s0rin) | 07:24 | |
| + S0rin (~S0rin@user/s0rin) | 07:26 | |
| - klardotsh (QUIT: Quit: zzz) (~klardotsh@98.97.113.255) | 08:30 | |
| violet | hdmi has been stable all afternoon/evening for me on my 1080p display, ty again for that | 09:19 |
| - lifning (QUIT: Ping timeout: 260 seconds) (~lifning@user/lifning) | 09:20 | |
| + lifning (~lifning@user/lifning) | 10:03 | |
| - lifning (QUIT: Ping timeout: 245 seconds) (~lifning@user/lifning) | 10:07 | |
| amospalla[m] | I've always used small zram volumes (say, 1/4 of my main memory) on my hosts, should I raise that? | 10:10 |
| amospalla[m] | Also, I use zram for /tmp volume (ext2 is enough for it, no need for a file system log). That is still ram fast, avoids solid state disks writes and saves memory by enabling zstd compression. | 10:12 |
| amospalla[m] | At the end, zram is just a block device. | 10:13 |
| - Gooberpatrol66 (QUIT: Ping timeout: 245 seconds) (~Gooberpat@user/gooberpatrol66) | 10:22 | |
| violet | amospalla[m]: i would just use tmpfs instead of zram on /tmp personally | 10:22 |
| violet | because tmpfs gets moved to swap as-needed already | 10:22 |
| violet | so you dont really gain anything by making it a zram device, and in fact you lose things because of the filesystem overhead and the fact that space gets permanently allocated and not zeroed when files get deleted | 10:23 |
| violet | so tmpfs on /tmp + zram swap device is (in my opinion) a better option | 10:24 |
| amospalla[m] | violet: ahh didn't know that, thank you. | 10:26 |
| amospalla[m] | Then yes, on a device with 4 or 8 gib of memory zswap makes more sense. | 10:27 |
| violet | yeah i did not know that either for a long time but ended up investigating when i was working on a system with 2 gigs of ram and was hyperoptimizing things | 10:27 |
| violet | i.e. https://artemis.sh/2022/01/12/life-at-800mhz.html | 10:27 |
| violet | 2 gigs ram, this one has a 1.5 gig zram i think | 10:27 |
| violet | and the entire browser userdata gets loaded into tmpfs to keep the browser IndexDB/sqlite/localstorage queries fast to dramatically speed up webapps | 10:28 |
| violet | (internal storage is dreadfully slow) | 10:28 |
| violet | and that ends up having a really really good compression ratio of like 2:1 or 2.5:1 | 10:29 |
| + lifning (~lifning@user/lifning) | 10:29 | |
| amospalla[m] | Oh, I already read that post of yours some time ago :) | 10:30 |
| violet | practically speaking what happens is all the stuff that isnt actively in use makes its way into zram and all the stuff that is actually in currently in use by the currently open web-app stays uncompressed | 10:30 |
| violet | oh neat! | 10:30 |
| amospalla[m] | And some others. | 10:30 |
| amospalla[m] | I've used firefox profile on zram, but lately I stopped, just because my current ssd feels fast enough and I like simplicity. | 10:31 |
| violet | on reform of course you have a real nvme drive so zswap would be quite nice, but i just dont really like wearing my drives with swap so i stick to zram even when i have decent storage | 10:31 |
| amospalla[m] | My script was a mess, maybe some day I rewrite it. | 10:32 |
| violet | which. i probably dont need to these days, im not likely to actually wear out the drives with swap. but *shrug* | 10:32 |
| amospalla[m] | My current gpd pocket 2 internal emmc feels quite good for anything I do, even firefox profile. | 10:32 |
| violet | whats the fun in computing without a little personal superstition | 10:32 |
| amospalla[m] | Aren't modern emmc fast enough? | 10:33 |
| violet | it really depends on the emmc | 10:33 |
| amospalla[m] | I will try the internal Pocket Reform emmc first, but at the end I suspect I'll move to nvme, just because my fear to burn the emmc. | 10:33 |
| violet | yeah those small emmc are not so great for write cycles | 10:34 |
| violet | we burned one pretty early into owning a chromebook with one. but part of that was using btrfs which apparently at the time had like 11x write amplification for small writes (yikes) | 10:34 |
| violet | and browsers do a *lot* of small writes | 10:35 |
| amospalla[m] | Yes, they do, I guess I'll buy some nmve before the pocket arrives. | 10:36 |
| violet | nowadays if i cant replace the flash storage i try to be suuuper gentle with it and only use it for booting | 10:37 |
| violet | and do everything else on something replaceable | 10:37 |
| amospalla[m] | Yes, that's the reason. | 10:37 |
| amospalla[m] | For a couple of years I've used the internal eMMC of the gpd pocket 2 for root partition, and anything else on a raid1 between a usb pendrive and a micro-sd card (that thing doesn't have any disk slot). | 10:39 |
| amospalla[m] | Everything writeable was there, including /var/intensive-writeable-folders, etc | 10:39 |
| amospalla[m] | F2FS is a great file system for that, my pendrive and sdcard survived to months of daily usage for anything I needed. | 10:40 |
| violet | oh nice | 10:41 |
| + mjw (~mjw@gnu.wildebeest.org) | 11:53 | |
| josch | violet: i browsed your blog. You also rebased the ffmpeg v4l2-requests patchset? I did the same for the reform but unfortunately there are quite some artifacts with hantro still... | 11:59 |
| josch | https://source.mnt.re/reform/reform-debian-packages/-/blob/main/patches/ffmpeg | 11:59 |
| minute | as a test i just downloaded the game "Dex" from gog and ran it on a311d reform (it's weirdly a 80386 linux executable), works perfectly | 12:14 |
| - buckket (QUIT: Quit: buckket) (~buckket@vps.buckket.org) | 12:17 | |
| minute | i do have /dev/video0 btw but i don't know what it can do yet (vdec) | 12:17 |
| + buckket (~buckket@vps.buckket.org) | 12:18 | |
| - mjw (QUIT: Ping timeout: 252 seconds) (~mjw@gnu.wildebeest.org) | 13:40 | |
| * mark_ -> mjw | 13:48 | |
| - mtm (QUIT: Ping timeout: 252 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 14:03 | |
| Boostisbetter | great to hear. | 14:09 |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 16:09 | |
| josch | i'll be afk until monday -- hope nothing serious pops up until then :) | 16:32 |
| minute | josch: enjoy! | 16:33 |
| - ec0 (QUIT: Ping timeout: 250 seconds) (~ec0@vps-446f4f39.vps.ovh.ca) | 16:35 | |
| + ec0 (~ec0@vps-446f4f39.vps.ovh.ca) | 16:39 | |
| minute | ordered the 36x LS1028AXE7PQA chips for part 2 of the production from farnell. with 18 weeks lead time | 16:40 |
| + sevan (~sevan@2001:470:1f1d:1d6:5a55:caff:fe24:ed4) | 16:43 | |
| - sevan (QUIT: Changing host) (~sevan@2001:470:1f1d:1d6:5a55:caff:fe24:ed4) | 16:45 | |
| + sevan (~sevan@user/venture37) | 16:45 | |
| - bkeys (QUIT: Remote host closed the connection) (~Thunderbi@static-198-54-130-101.cust.tzulo.com) | 16:59 | |
| minute | fallout new vegas just works https://mastodon.social/@mntmn/110707499124420208 | 17:34 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 18:21 | |
| Boostisbetter | YES!!! Best Fallout to date! | 18:44 |
| + mark_ (~mjw@gnu.wildebeest.org) | 18:56 | |
| + ajr (uid609314@user/ajr) | 18:58 | |
| Boostisbetter | any tips for getting it to run on the Reform with the imx8 in it? | 18:58 |
| minute | Boostisbetter: that won't work | 19:00 |
| minute | i mean, it won't be playable i think | 19:00 |
| - mjw (QUIT: Killed (NickServ (GHOST command used by mark_!~mjw@gnu.wildebeest.org))) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 19:05 | |
| * mark_ -> mjw | 19:05 | |
| + mark_ (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 19:05 | |
| Boostisbetter | that is a bummer. Which soc are you using then? | 19:08 |
| - lifning (QUIT: Ping timeout: 245 seconds) (~lifning@user/lifning) | 19:34 | |
| + klardotsh (~klardotsh@98.97.114.65) | 20:08 | |
| minute | Boostisbetter: i upgraded to A311D | 20:09 |
| Boostisbetter | how much of a performance delta is the default SoC with 8gb RAM in the Pocket between the A311D? | 20:10 |
| minute | Boostisbetter: CPU wise the performance is maybe double? and GPU wise a lot more | 20:12 |
| minute | 7z benchmark total score is around 1.7x of imx8mp with 2ghz | 20:13 |
| Boostisbetter | is it possible to get teh A311D in the Pocket? | 20:13 |
| minute | i hope so | 20:13 |
| minute | it should be, i just need to test the display etc | 20:13 |
| Boostisbetter | that is exciting stuff. Anything that can play new vegas is more than powerful enough fro me. | 20:13 |
| vkoskiv | Veeery excited about the A311D when that starts shipping 8) | 20:14 |
| minute | i just ordered the 100x new heatsinks today | 20:21 |
| minute | cm4 had in tests only a tiny bug with sd card level shifter, will fix it and then order a batch tomorrow | 20:22 |
| + tretinha_ (3a571d9f43@2604:bf00:561:2000::1151) | 20:34 | |
| + noam_ (81879d1ffa@2604:bf00:561:2000::dfc) | 20:34 | |
| + henesy_ (d7619ffbc2@2604:bf00:561:2000::143) | 20:35 | |
| + Zaba_ (80b9b4b35e@2604:bf00:561:2000::116) | 20:35 | |
| + cmahns_ (8fe824803c@2604:bf00:561:2000::10cd) | 20:35 | |
| + laumann_ (~quassel@2a0a-e5c0-2-2-0-c8ff-fe68-bef1.loves.ipv6.at.ungleich.ch) | 20:36 | |
| + plomlomp0m (~plom@play.plomlompom.com) | 20:37 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 20:38 | |
| - qbit (QUIT: *.net *.split) (~qbit@h.suah.dev) | 20:42 | |
| - minute (QUIT: *.net *.split) (~mntirc@softboy.mntmn.com) | 20:42 | |
| - plomlompom (QUIT: *.net *.split) (~plom@play.plomlompom.com) | 20:42 | |
| - laumann (QUIT: *.net *.split) (~quassel@user/laumann) | 20:42 | |
| - cmahns (QUIT: *.net *.split) (8fe824803c@user/cmahns) | 20:42 | |
| - Zaba (QUIT: *.net *.split) (80b9b4b35e@2604:bf00:561:2000::116) | 20:42 | |
| - henesy (QUIT: *.net *.split) (d7619ffbc2@2604:bf00:561:2000::143) | 20:42 | |
| - tretinha (QUIT: *.net *.split) (3a571d9f43@2604:bf00:561:2000::1151) | 20:42 | |
| - noam (QUIT: *.net *.split) (81879d1ffa@2604:bf00:561:2000::dfc) | 20:42 | |
| - ethulhu (QUIT: *.net *.split) (~ethulhu@nora.ethulhu.co.uk) | 20:42 | |
| * cmahns_ -> cmahns | 20:42 | |
| * henesy_ -> henesy | 20:42 | |
| * tretinha_ -> tretinha | 20:42 | |
| * noam_ -> noam | 20:42 | |
| * Zaba_ -> Zaba | 20:42 | |
| + minute (~mntirc@softboy.mntmn.com) | 20:48 | |
| + qbit (~qbit@2602:ff16:3:0:1:3a0:0:1) | 20:50 | |
| + klardotsh_ (~klardotsh@98.97.114.65) | 21:59 | |
| - laumann_ (QUIT: Ping timeout: 240 seconds) (~quassel@2a0a-e5c0-2-2-0-c8ff-fe68-bef1.loves.ipv6.at.ungleich.ch) | 21:59 | |
| - klardotsh (QUIT: Read error: Connection reset by peer) (~klardotsh@98.97.114.65) | 21:59 | |
| + laumann (~quassel@2a0a-e5c0-2-2-0-c8ff-fe68-bef1.loves.ipv6.at.ungleich.ch) | 22:00 | |
| Boostisbetter | saw that a couple of days ago and i have to say that is a pretty big bummer. | 22:29 |
| - jacobk (QUIT: Ping timeout: 264 seconds) (~quassel@47-186-122-163.dlls.tx.frontiernet.net) | 22:33 | |
| minute | Boostisbetter: saw what? | 22:33 |
| - laumann (QUIT: Changing host) (~quassel@2a0a-e5c0-2-2-0-c8ff-fe68-bef1.loves.ipv6.at.ungleich.ch) | 22:47 | |
| + laumann (~quassel@user/laumann) | 22:47 | |
| + laumann_ (~quassel@2a0a-e5c0-2-2-0-c8ff-fe68-bef1.loves.ipv6.at.ungleich.ch) | 23:01 | |
| - laumann (QUIT: Ping timeout: 260 seconds) (~quassel@user/laumann) | 23:01 | |
| Boostisbetter | minute: I saw the news that Nucs were getting discontinued. It is a bummer for sure. | 23:05 |
| Boostisbetter | minute: any idea when the a311d is going to see up in the shop? | 23:07 |
| + robin (~robin@user/terpri) | 23:23 | |
| - robin (QUIT: Read error: Connection reset by peer) (~robin@user/terpri) | 23:31 | |
| + robin (~robin@user/terpri) | 23:32 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!