On Tue, Nov 24, 2020 at 02:01:58PM +0200, Ivan Serdyuk wrote:
> Richard,
> could you tell why gold linker is not available on RISC-V port of Fedora 33?

You're right that it's not built (as a subpackage of binutils):

  http://fedora.riscv.rocks/koji/buildinfo?buildID=186524

That happens because of this clause in the binutils spec file:

  # Note - in the future the gold linker may become deprecated.
  %ifnarch riscv64
  %bcond_without gold
  %else
  # RISC-V does not have ld.gold thus disable by default.
  %bcond_with gold
  %endif

That was added by David Abdurachmanov and Nick Clifton last year, so
I'm CC-ing both of them.  It's unclear exactly what the
"file installation issues" were.

  
https://src.fedoraproject.org/rpms/binutils/c/3edfd25eae3f2c6b2bbdf7d16320fd70f9456c99?branch=master

I've kicked off a local build without this to see if it would work now.

Rich.

> 
> On Sat, Nov 21, 2020 at 12:10 PM Ivan Serdyuk <[email protected]>
> wrote:
> 
>     Wei, Carlos: adding you to this discussion.
> 
>     Carlos, please ask your questions. What was not understood, regarding my
>     attempt?
> 
>     Hence that QEMU's host OS could be, say, Fedora 32+ x86_64.
> 
>     Ivan
> 
>     On Sat, Nov 21, 2020 at 5:36 AM Ivan Serdyuk 
> <[email protected]>
>     wrote:
> 
>         Hello.
> 
>         It is just a brief survey - but I am currently dealing with a messy
>         setup, for qemu.
> 
>         I have upgraded, after bootstrapping( the manual ) and I am able to
>         boot
> 
>         > Fedora (5.8.0-1.0.riscv64.fc33.riscv64) 33 (Rawhide)
> 
>         kernel, under qemu.
> 
>         I see a very abstract
> 
>         > $ cat /proc/cpuinfo
>         > processor : 0
>         > hart : 0
>         > isa : rv64imafdcsu
>         > mmu : sv48
>         .
> 
>         Doesn't really help.
>         I will suggest trying other emulators - but would have to compile 
> those
>         myself.
> 
>         I have some concerns about gmp, mpfr and mpc (since RISC-V assumes 
> some
>         extensions, to cover such demands) - I feel that it might have more
>         sense to get rid of overlaying, for the end-user use cases (which
>         gollvm project itself achieves).
> 
> 
>         Ivan
> 
>         --
>         You received this message because you are subscribed to the Google
>         Groups "golang-nuts" group.
>         To unsubscribe from this group and stop receiving emails from it, send
>         an email to [email protected].
>         To view this discussion on the web visit https://groups.google.com/d/
>         msgid/golang-nuts/
>         2c9788e3-5349-4f5e-bbfc-53f300cf439bn%40googlegroups.com.
> 

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-top is 'top' for virtual machines.  Tiny program with many
powerful monitoring features, net stats, disk stats, logging, etc.
http://people.redhat.com/~rjones/virt-top

-- 
You received this message because you are subscribed to the Google Groups 
"golang-nuts" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/golang-nuts/20201124121243.GC30079%40redhat.com.

Reply via email to