**** BEGIN LOGGING AT Sat Dec 22 03:00:02 2018 Dec 22 07:03:58 uboot-omap isn't building for me Dec 22 07:05:10 /usr/include/libfdt_env.h:71:30: error: conflicting types for 'fdt64_t' Dec 22 07:05:11 typedef uint64_t FDT_BITWISE fdt64_t; Dec 22 07:05:55 vs: Dec 22 07:05:57 typedef __be64 fdt64_t; Dec 22 09:01:51 Borromini: pong Dec 22 09:08:34 jow: hi. Dec 22 09:08:51 i have a 18.06 mbedtls 2.14.1 bump ready based on the master commit Dec 22 09:09:22 (all i did was fix up the patches so they'd apply cleanly) Dec 22 09:09:30 want me to send it in? Dec 22 09:13:57 i have compile-tested it but on my mt7621 i'm using openssl so haven't run-tested it yet. Dec 22 09:49:12 Borromini: that'd be nice, yes. But I can't promise to look into it before after christmas Dec 22 09:52:15 jow: all i wanted was to avoid you'd do it by yourself while i had already done it. there's no rush. Dec 22 09:52:22 can save you a bit Dec 22 09:52:33 * just wanted to help out a bit Dec 22 10:57:29 I used to have files in "files/boot/*", seems like those are deleted now. Anyone happen to know which commit made that impossible? (I'll have to bisect to find the commit otherwise) Dec 22 10:58:20 I agree the folder name I chose might not have been the best Dec 22 11:08:07 barhom: https://git.lede-project.org/31075313bf69fff3803b7c1da1e0ed7e5773d999 Dec 22 11:09:39 KanjiMonster: Awesome Dec 22 11:57:29 danitool: if you have time, can you check if https://git.openwrt.org/?p=openwrt/staging/jogo.git;a=shortlog;h=refs/heads/brcm63xx-next if it fixes the 6348 pinmux and the gpiohog issues for you? Dec 22 11:58:09 danitool: couldn't add https://patchwork.ozlabs.org/patch/997455/ though as the patch is broken (got cut off it seems) Dec 22 11:58:44 KanjiMonster: ok I'll test it Dec 22 15:41:12 KanjiMonster: the bcm6348 pinmux worked ok Dec 22 15:41:29 KanjiMonster: the gpio hog fix also worked ok Dec 22 15:42:29 however the VH4032 board still has the pci and ethernet leds pinctrl missed Dec 22 16:25:42 KanjiMonster: the BCM6348 pincontrol still has some issues, some groups own the wrong GPIOs. This causes a failure in a button (GPIO36) on my board. Dec 22 16:26:00 KanjiMonster: https://pastebin.com/eJRhCc6k Dec 22 16:27:05 KanjiMonster: if this helps, I made a table with the relevant groups https://openwrt.org/docs/techref/hardware/soc/soc.broadcom.bcm63xx/pinmux Dec 22 16:28:15 the EXT_MII only has the GPIO32 on the group0 for sure **** ENDING LOGGING AT Sat Dec 22 18:13:37 2018 **** BEGIN LOGGING AT Sat Dec 22 18:16:09 2018 Dec 22 18:55:57 Hauke: know why ath79 and ramips define cpu_has_rixi 0 and CONFIG_CPU_HAS_RIXI=y ? Dec 22 18:59:42 mangix: I do not know, it could be that this is an optional ase extension and it is not activated for the CPU in these SoCs Dec 22 19:04:57 hrm either the kernel or the config is wrong then Dec 22 19:05:05 *openwrt config Dec 22 19:35:29 mangix: CONFIG_CPU_HAS_RIXI gets selected when you select MIPS32R2 in the kernel configuration Dec 22 19:53:37 build #1187 of brcm47xx/generic is complete: Success [build successful] Build details are at http://phase1.builds.lede-project.org/builders/brcm47xx%2Fgeneric/builds/1187 Dec 22 19:55:36 hrm upstream uclibc++ has not made a release in 7 years while adding features/fixing bugs Dec 22 20:51:32 it's curious that target/linux/ath79/patches-4.14/0014-MIPS-ath79-finetune-cpu-overrides.patch overrides rixi. Dec 22 21:42:09 ldir: rixi is probable feature, so instead of forcing it to 1 you could remove the line and check the runtime value with a printk Dec 22 21:42:31 KanjiMonster: I was just suggesting the same ;-) Dec 22 21:42:52 this is documented on page 235 here: https://s3-eu-west-1.amazonaws.com/downloads-mips/I7200/I7200+product+launch/MIPS_nanoMIPS32_PRA_06_09_MD01251.pdf Dec 22 21:43:32 ldir: when you are at it, can you please print out the config3 register please Dec 22 21:43:59 I also think that CPU_HAS_RIXI actually only means CPU_MAY_HAVE_RIXI, not CPU_WILL_HAVE_RIXI Dec 22 21:44:30 KanjiMonster: yes me too Dec 22 21:49:01 Hauke: btw, I noticed that /proc/cpuinfo would not show any rixi related info; IMO it would be useful to show wether its supported by the cpu ( https://elixir.bootlin.com/linux/latest/source/arch/mips/kernel/proc.c#L36 ) Dec 22 21:49:31 KanjiMonster: yes I think so too Dec 22 21:49:48 KanjiMonster: some years ago I hacked more of the cpu features into cpuinfo Dec 22 21:49:58 but I think I do not have the patch any more Dec 22 21:50:41 I will have a look at this and probably send a patch **** ENDING LOGGING AT Sun Dec 23 03:00:00 2018