02:06 <m4t> is the fsl caam rng supposed to work on older qoriq ppc (p2041)? looking at recent changes e.g. https://lore.kernel.org/all/20211115070014.17586-1-gaurav.jain@nxp.com/ . i tried a couple things like explicitly calling sec_init() in arch/powerpc/cpu/mpc85xx/cpu_init.c and removing the conditional at https://github.com/u-boot/u-boot/blob/master/drivers/crypto/fsl/jr.c#L788-L792 but get 'caam_jr: caam

02:06 <m4t> not found' and 'SEC0: RNG4 SH0 instantiation failed with error 0xfffffffe'

02:06 <m4t> i'm not clear if the u-boot driver only supports newer chips. rng-caam seems to work fine in linux

02:08 <m4t> i'm mostly interested in getting it working for seeding kaslr

02:49 <hanetzer> sjg1: for the record: u-boot HEAD, 4f2c559b9a2ad86e03dffeef720257ea680707d0 as I write this, actually does work with (mostly) stock chromebook_speedy_defconfig. Problem is in the rk_vop.c driver, not finding the ahb reset line, so the display never inits and it 'appears' dead. I may have damaged my yoshi flex cable however, bummer there.

02:50 <hanetzer> however, having to work around this did let me get a bit of useful code in for u-boot :)

04:13 <Forty-Bot> hanetzer: yes, git send-email is fine

04:13 <Forty-Bot> patman is nice too

04:14 <hanetzer> ohyeh, forgot about patman

06:09 <hanetzer> `f43312c974eaeb7301cb2638aa4ab05ed7ca4c44 ARM: Prevent the compiler from using NEON registers' << makes chromebook_kevin_defconfig + CONFIG_CONSOLE_TRUETYPE fail build for me

17:49 <fmiz> Hello, I just received the email for the rc5. I've prepared a small patch to improve the documentation of fdtoverlays keyword in sysboot (extlinux) files, and added another keyword for it to follow the bootloader specification. Is it better if I wait until 2022.10 is released to send it?

17:50 <cambrian_invader> fmiz: you can send patches at any time; if it isn't a bug fix, base it on u-boot/next instead of /master

17:53 <fmiz> cambrian_invader: no it definetely does not fix a bug. If it gets accepted it will end up in the release following 2022.10?

17:53 <cambrian_invader> it should

17:54 <fmiz> Thanks. As soon as I figure out the right git commands, I'll send it :)

17:57 <vagrantc> i'd just send a patch ... if it's harmless enough it wouldn't hurt to have better documentation in the 2022.10 release ... you might have to resubmit later or something but ...

17:58 <vagrantc> also depends how large the diff is

17:59 <fmiz> vagrantc: look, this is the summary just below the email body that patman generated (based on master), boot/pxe_utils.c | 2 ++ doc/README.pxe | 4 ++++ doc/develop/distro.rst | 26 ++++++++++++++++++++++++++ 3 files changed, 32 insertions(+)

18:00 <fmiz> ...ugly.. very ugly message

18:02 <fmiz> vagrantc: it 2 lines added in pxe_utils (one is a comment, the other adds devicetree-overlays as an alias to fdtoverlays), everything else is documentation (split between README.pxe and doc/develop/distro.rst)

18:11 <vagrantc> maybe it's bad advice, but i'd try to get it into 2022.10 :)

18:13 <fmiz> vagrantc: I've rebased it to master, but now patman it not collaborating, it's throwing an error... UnicodeDecodeError: 'utf-8' codec can't decode byte 0xfc in position 4893390: invalid start byte

18:19 <vagrantc> eesh.

18:20 <vagrantc> probably some mixed encoding somewhere in the file?

18:32 <fmiz> vagrantc: I managed to get it to work, I'm not sure how

18:51 <fmiz> it's done, I sent the patch. I've done a mess with my local branches, I've double checked it, I hope I've sent the right thing