IRC channel logs

<Pellescours>damo22: I’ll send you the patches for rumpdisk, you can test them and apply them if you find them fine

<damo22>you can create a dummy opt_ahcisata.h

<damo22>it just needs to #include it

<Pellescours>yes that’s what I did, I added the options being the old behavior

<damo22>cool

<Pellescours>I’m recompiling rumpdisk to test it before sending you the patches

<damo22>how did you update the sources

<damo22>did you use prune.sh and only commit those?

<Pellescours>rm the src folder, git clone netbsd, ther prune.sh

<Pellescours>yes

<damo22>cool

<damo22>netbsd wants us to upstream our patches

<Pellescours>o/

<damo22>once we have it working we just mail them into tech-kern

<damo22>they will review

<Pellescours>when trying to build hurd (not the debian package) I have undefined ref to « rumpns_bus_dmatag_subregion »

<Pellescours>I can see that they added this _bus_dmatag_subregion call to some pci stuff (seems related to ethernet devices)

<Pellescours>or at least something used in rumpusbdisk

<Pellescours>i worked, I run my VM with updated netbsd sources and rumpdisk

<Pellescours>piixide still loose interupts, so nothing changed on that part

<Pellescours>but can a patch of 223M be send by email? I’m not sure that emails providers accept it

<youpi>they won't :)

<sneek>youpi, you have 1 message!

<sneek>youpi, gfleury says: i just sent on glibc ML move of __pthread_sigstate, __pthread__sigstate and pthread_sigmask into libc.

<youpi>you can push a branch somewhere to pull from

<youpi>(yes, seen gfleury's mails)

<Pellescours>damo22: youpi: https://github.com/etienne02/rumpdisk-debian it’s the 2 commits after master

<Pellescours>To have a fully functionnal rump we need to define `rumpns_bus_dmatag_subregion` for rumpusbdisk

<Pellescours>but that’s a base to work on

<ZhaoM>youpi: I found some files ($(hurd)/exec/execmutations.h for example) don't have the copyright header, is it worth to add the copyright header?

<youpi>when they are quite trivial, it's not worth it

<youpi>(trivial source is non-copyrightable)

<ZhaoM>OK

<gnucode>sneek later tell azert I totally agree! The comment section on phoronix can be fairly mean. That's why I #neverReadTheComments But we are all mostly in agreement that announcing Hurd progress update are a good thing right?

<sneek>Got it.

<gfleury>Hi

<gfleury>youpi: oops i forgot signof. Should i send v2

<gfleury>signed-off

<azert>Hello all, in regards to the discussion we had about kms/drm, there is this interesting article https://blog.gtk.org/2023/11/15/introducing-graphics-offload/ that explains how things are organized from an user perspective

<sneek>azert, you have 1 message!

<sneek>azert, gnucode says: I totally agree! The comment section on phoronix can be fairly mean. That's why I #neverReadTheComments But we are all mostly in agreement that announcing Hurd progress update are a good thing right?

<azert>would be interesting to know what is the difference between dma-buf in Linux and memory objects in gnumach

<azert>and how all of this is far superior to mmap /dev/mem

IRC channel logs

2024-12-13.log