Hi!
On 2019-11-11T18:12:44+0000, "Maciej W. Rozycki" <
ma...@wdc.com> wrote:
> Fix a problem with the libgo testsuite using a method to determine the
> compiler to use resulting in the tool being different from one the
> library has been built with, and causing a catastrophic failure from the
> lack of a suitable `--sysroot=' option where the `--with-build-sysroot='
> configuration option has been used to build the compiler resulting in
> the inability to link executables.
>
> Address this problem by providing a DejaGNU configuration file defining
> the compiler to use, via the GOC_UNDER_TEST TCL variable, set from $GOC
> by autoconf, which will have all the required options set for the target
> compiler to build executables in the environment configured
As we've found, this is conceptually problematic, as discussed in
<
https://inbox.sourceware.org/875y868...@euler.schwinge.homeip.net>
"Consider '--with-build-sysroot=[...]' for target libraries' build-tree testing (instead of build-time 'CC' etc.)
[PR109951]".
I therefore suggest to apply to libgo the conceptually same changes
as I've just pushed for libgomp:
<
https://inbox.sourceware.org/874jjzz...@euler.schwinge.homeip.net>
"libgomp: Consider '--with-build-sysroot=[...]' for target libraries' build-tree testing (instead of build-time 'CC'
etc.) [PR91884, PR109951]".
OK to push (via Ian/Go upstream) the attached
"libgo: Consider '--with-build-sysroot=[...]' for target libraries' build-tree testing (instead of build-time 'CC' etc.) [PR109951]"?
By the way, I've tested this one via hard-coding
'libgo/configure.ac:USE_DEJAGNU' to 'yes', and observing that my
"quick hack to replicate the original requirement"
('internal_error ("MISSING SYSROOT");') no longer triggers.
Grüße
Thomas
-----------------
Siemens Electronic Design Automation GmbH; Anschrift: Arnulfstraße 201, 80634 München; Gesellschaft mit beschränkter Haftung; Geschäftsführer: Thomas Heurung, Frank Thürauf; Sitz der Gesellschaft: München; Registergericht München, HRB 106955