**** BEGIN LOGGING AT Thu Dec 31 02:59:57 2020
Dec 31 04:26:25 <aparcar[m]> mangix: any ideas about https://github.com/openwrt/openwrt/pull/3739 ?
Dec 31 05:20:39 >KGB-1< https://tests.reproducible-builds.org/openwrt/openwrt_tegra.html has been updated. (0% images and 97.0% packages reproducible in our current test framework.)
Dec 31 08:31:38 <mangix> I do not
Dec 31 09:12:50 <Borromini> pkgadd: i found my way to a few local marketplaces and getting all kinds of second hand stuff :P
Dec 31 09:13:00 <Borromini> pity there are no tri-band mt76 routers afaict
Dec 31 09:25:50 <rsalvaterra> mangix: https://github.com/openwrt/openwrt/pull/3738#issuecomment-752891453
Dec 31 09:26:10 <rsalvaterra> I have urngd running on MT7621 without issues, I believe… :/
Dec 31 09:26:25 <rsalvaterra> Am I missing something?
Dec 31 09:32:16 <mangix> yeah. it doesn't do anything
Dec 31 09:34:02 <mangix> i forgot the details, but it requires some CPU feature
Dec 31 09:34:15 <rsalvaterra> Hmm… It's not what I see when I cat /dev/random…
Dec 31 09:38:52 <rsalvaterra> The MIPS 1004Kc is basically a 24Kc with an extra pipeline stage (thread selection, for SMT), and according to the jitterentropy-rng paper (on which urngd is based, I believe), the 24Kc is a usable CPU: https://www.chronox.de/jent/doc/CPU-Jitter-NPTRNG.html#toc-Appendix-C
Dec 31 11:06:01 <reiffert> dedeckeh: FS#3539 was just closed but I wanted to keep working on it. Things have started to be not so solved/fixed/clear after all
Dec 31 11:16:22 <dedeckeh> reiffert:I will re-open FS#3539
Dec 31 11:17:37 <reiffert> Thank you
Dec 31 11:30:17 <deivid> hello! i'm trying to build a single package following the wiki guide for this..but regardless of what i do during config time, when i run `make tools/install` i get recursive deps on PACKAGE_kmod*, even though i unselected them unselected them from .config
Dec 31 11:30:39 <deivid> ofcourse as per the docs i did try `make world` and that one works (even though it shows all the same errors)
Dec 31 11:43:00 <russell--> deivid: when you refer to a wiki guide, it is helpful to provide the link to that guide
Dec 31 11:43:27 <deivid> whoops - i followed https://openwrt.org/docs/guide-developer/single.package
Dec 31 12:12:10 <karlp> also, share what you've got so far
**** ENDING LOGGING AT Thu Dec 31 12:40:58 2020

**** BEGIN LOGGING AT Thu Dec 31 12:44:19 2020
Dec 31 13:23:14 <russell--> deivid: what is the single package?
Dec 31 15:12:42 <rr123> https://madaidans-insecurities.github.io/guides/linux-hardening.html
Dec 31 15:15:11 <cp-> Happy New Year 2021 to everyone
Dec 31 18:13:00 >KGB-0< https://tests.reproducible-builds.org/openwrt/openwrt_mediatek.html has been updated. (100.0% images and 97.3% packages reproducible in our current test framework.)
Dec 31 19:34:33 <aparcar[m]> mangix: ping
Dec 31 19:35:39 <rsalvaterra> Hauke: ping
Dec 31 19:41:17 <Hauke> rsalvaterra: pong
Dec 31 19:41:53 <rsalvaterra> Hauke: Ah, I mailed you already. :) You didn't remove the makefile escaping from the awk script when testing in the terminal. :P
Dec 31 19:42:53 <rsalvaterra> That $$$$3 will translate to $3. :)
Dec 31 19:46:12 <aparcar[m]> Hauke: you're ssl suggestion sounds good
Dec 31 19:46:17 <aparcar[m]> *your
Dec 31 19:46:41 <aparcar[m]> do you know if baptiste will work on that?
Dec 31 19:46:42 <Hauke> rsalvaterra: with only one $ it works
Dec 31 19:47:46 <rsalvaterra> Hauke: I had the opposite problem when writing the patch, figuring out how much escaping I needed. :D
Dec 31 19:48:22 <rsalvaterra> Anyway, thanks for testing on 32-bit, I hadn't tested, but I was confident. ;)
Dec 31 19:48:57 <Hauke> on the server side we just activated the session cache and it was used: https://git.openwrt.org/f2c8f6dc3249b506b915741d12905402dfffe162
Dec 31 19:50:33 <Hauke> and this one: https://git.openwrt.org/e8a1469
Dec 31 19:58:19 <aparcar[m]> Hauke: great so we only have to upgrade apache on d.o.o?
Dec 31 20:08:42 <Hauke> aparcar[m]: we should test it before
Dec 31 20:09:08 <Hauke> if it really brings an improvment, I only tetsed this with mbedtls and in server mode
Dec 31 20:10:00 <Hauke> openssl is highly optimized for performance on most architectures, mbedtls is not
Dec 31 20:10:08 <Hauke> I do not know the state of wolfssl on mips
Dec 31 20:15:57 <aparcar[m]> so how to test? setup a server with only ecc enabled?
Dec 31 20:17:24 <Hauke> aparcar[m]: yes you can test this by just using wget
Dec 31 20:17:50 <Hauke> ecc vs. rsa is only to make handshakes faster
Dec 31 20:18:04 <Hauke> the other aproache is to reduce the number of handshakes needed
Dec 31 20:18:21 <Hauke> I think we should try both and if both are good try to get both implemented
Dec 31 20:20:56 <aparcar[m]> dangole: ping
Dec 31 20:22:12 <aparcar[m]> so what would be the ideal solution? ecc for quick handshakes and mbedtls/wolfssl with session caches?
Dec 31 20:26:56 <Hauke> both together
Dec 31 20:27:10 <Hauke> if it really makes a difference
Dec 31 23:02:57 <mangix> aparcar[m]: pong
Dec 31 23:03:43 <mangix> Hauke: wolfssl is a pain with transmission
Dec 31 23:21:30 <Hauke> mangix: ;-)
Dec 31 23:22:42 <mangix> Hauke: pain meaning instant crash when adding a torrent
Dec 31 23:22:46 <mangix> I can't figure it out
Dec 31 23:53:41 <mangix> https://www.phoronix.com/scan.php?page=news_item&px=Linux-MIPS-Vulnerabilities
Dec 31 23:53:53 <mangix> unfortunate
Jan 01 00:12:22 <mrkiko> An excellent new year to all of you guys!!
Jan 01 00:53:37 <Hauke> angix: I think someone said that most MIPS CPU cores used in routers are probably not affected by Meltdown and spectre, because they are in order and have a static memory boundary between kernel and user space.
Jan 01 01:04:36 <mangix> Hauke: right. but upstream development of MIPS is pretty much dead. Except for Longsoon.
**** ENDING LOGGING AT Fri Jan 01 03:00:13 2021