2023-05-02 19:33:23 <ikke> nmeum: fyi https://go.dev/doc/devel/release#go1.20.4
2023-05-02 19:33:42 <nmeum> yea, i already opened an mr :)
2023-05-02 19:33:49 <nmeum> !46650
2023-05-02 19:34:07 <ikke> (y)
2023-05-02 19:34:24 <psykose> send it
2023-05-02 19:34:24 <psykose> :p
2023-05-02 19:47:51 <nmeum> fought succesfully(?) with google spreadsheets i now have time to press the merge button
2023-05-02 19:48:21 <nmeum> will do rebuilds in a sec
2023-05-02 19:49:30 <nmeum> ohh 1.19.9 fails tests on arm
2023-05-02 19:51:39 <psykose> nothing in commits looks relevant
2023-05-02 19:57:42 <nmeum> hmhmhmhmh, not sure
2023-05-03 00:09:39 <omni> !46671 !46672
2023-05-10 12:24:12 <ptrc> ncopa: despite the patch backported, TIOCSTI is still enabled by default on linux-lts, shouldn't that be set to =n instead?
2023-05-10 12:26:33 <ptrc> $ apk fetch --stdout linux-lts | tar xOz boot/config-lts | grep TIOCSTI
2023-05-10 12:26:34 <ptrc> CONFIG_LEGACY_TIOCSTI=y
2023-05-10 13:09:14 <ncopa> which patch?
2023-05-10 13:09:32 <ncopa> usually this happens due to something else pull in it as a dep
2023-05-10 13:27:39 <ptrc> the 0002-tty-Allow-TIOCSTI-to-be-disabled.patch 
2023-05-10 13:29:14 <ptrc> see line 45, it defaults to =y
2023-05-10 13:29:23 <ptrc> ( which means it's still enabled )
2023-05-10 13:29:33 <ptrc> https://gitlab.alpinelinux.org/alpine/aports/-/blob/6d54100d/main/linux-lts/0002-tty-Allow-TIOCSTI-to-be-disabled.patch#L45
2023-05-12 08:05:24 <psykose> ncopa: you still have to CONFIG_LEGACY_TIOCSTI=n
2023-05-12 08:10:43 <ncopa> hum
2023-05-12 08:14:59 <ncopa> config is generated with savedefconfig, so it only a diff against the defconfig
2023-05-12 08:16:36 <ncopa> ok, i'll try fix it
2023-05-12 08:17:44 <psykose> yeah you have to put that in the diff
2023-05-12 08:17:47 <psykose> because the default is =y
2023-05-12 08:20:00 <ncopa> i thought it woudl just work to use the generated savedefconfig
2023-05-12 08:20:21 <ncopa> use # CONFIG_LEGACY_TIOCSTI is not set
2023-05-12 08:20:41 <psykose> if something is =y then it will always be =y unless you put =n
2023-05-12 08:21:14 <psykose> pretty sure a #Comment would not ever set it to a non-default value
2023-05-12 08:22:14 <ncopa> i think it does
2023-05-12 08:22:26 <ncopa> kernel config is weird
2023-05-12 08:23:33 <psykose> if it did it would not be =y
2023-05-12 08:25:10 <ncopa> seems like ` # CONFIG_FEATURE is not set` is the proper way to disable a feature. not `CONFIG_FEATURE=n`
2023-05-12 08:29:09 <ncopa> well, CONFIG_FEATURE=n seems to be ok nowadays: https://lore.kernel.org/all/20220226123755.85213-1-masahiroy@kernel.org/
2023-05-12 08:32:41 <ncopa> nope, that commit was never merged. proper way to disable a feature is still # CONFIG_FEATURE is not set
2023-05-12 08:33:18 <ncopa> but I still have to add  `# CONFIG_LEGACY_TIOCSTI is not set`
2023-05-17 14:32:02 <psykose> https://www.interruptlabs.co.uk//articles/linux-ipv6-route-of-death
2023-05-17 14:48:16 <minimal> good thing it appears to be disabled by default (from looking at sysctl output) ;-)
2023-05-30 15:39:48 <ikke> New openssl advisories
2023-05-30 15:40:22 <ikke> openssl 3.0.9, 3.1.1, 1.1.1u released
2023-05-30 15:40:56 <ikke> https://www.openwall.com/lists/oss-security/2023/05/30/1