IRC channel logs

2024-09-25.log

back to list of logs

<gnucodeIrcIsAwes> https://mail.gnu.org/archive/html/bug-hurd/2023-01/msg00132.html that's an interesting email.
<gnucodeIrcIsAwes>It says...if you have a working hurd installation, then you can cross compile it via ./configure && make
<youpi>not cross-compile, just compile
<youpi>the original question was about just compiling
<gnucodeIrcIsAwes>youpi: so I am running Debian Hurd on my T43...if I wanted to use my T43 (X86_64) to build an AArch64 image, that would be cross compiling right?
<gnucodeIrcIsAwes>I guess that I am asking...what is the easiest way to run AArch64 Hurd in qemu?
<youpi>it would be cross-compiling, yes
<gnucodeIrcIsAwes>gotcha.
<youpi>azert's link provides the bits about booting with qemu
<gnucodeIrcIsAwes>thanks
<azert>Hi
<sneek>azert, you have 3 messages!
<sneek>azert, gnucodeIrcIsAwes says: thanks for the debian pastebin for how to use the AArch64 port!
<sneek>azert, gnucodeIrcIsAwes says: would you be willing to license the bit of code as GFDL 1.2+ ? That's what most of the wiki is licensed as. If it's licensed that way, it makes it easier to put that sort of thing in say the Hurd manual.
<sneek>azert, gnucodeIrcIsAwes says: please do share the patches for the uart
<azert>gnucode: ok
<azert>Here are three patches. https://paste.debian.net/1330354/
<azert> https://paste.debian.net/1330355/
<azert> https://paste.debian.net/1330356/
<azert>The first one is the serial driver for my board, it’s a very common chip since it’s the one found on x86
<azert>So I think it is generally very useful.
<azert>The second one it is some kind of fix up that didn’t deliver what I wanted, maybe solid_black can review it
<azert>The third one is solid_black patch to add tracing, in case it was lost
<azert>I think that the second patch is because I thought the timer was broken, while in fact it was just the memory error I mentioned yesterday
<azert>Dynamically allocated memory would allocate memory that didn’t exist
<azert>So I’m really not sure if the second patch is needed
<azert>Possibly it is an improvement
<azert>None of this code is ready to go, it is just useful for testing
<azert>The timer might still be broken..
<azert>In general, I’m not an arm programmer, I’m sure someone with experience with arm hardware could figure this out in minutes
<azert>Although, the memory allocation system might be daunting.. netbsd diverged quite a lot in that sector
<azert>solid_black: I’ve read in the mailing list that you did some cool gtk + c++ work, I am curious, what did you do ?