On 3/23/26 10:11, Bruno Haible wrote:
Dennis Clarke wrote:
(2) with the hacked in 60000 value
real 3150.73
user 2270.45
sys 426.53
OK, this disproves the "infinite loop" and "gcc bug" hypotheses.
You merely have a really slow program execution.
centauri$ echo $CFLAGS
-g -O0 -mcpu=21164 -mgas -mexplicit-relocs -fno-fast-math -fno-builtin
-fno-unsafe-math-optimizations -mno-soft-float -mfp-trap-mode=sui
-mfp-rounding-mode=n -mtrap-precision=i -mieee-with-inexact
-mieee-conformant
Much of the slowness is probably due to the "-O0 -fno-builtin" options.
"-O0" instead of "-O2" can be 3x to 10x slower.
Bruno
I don't care about the time. Why would anyone care about such a thing
with a machine from 1997! ha ha ha
I suspect that runtime to be bogus anyways. I would need to extract the
tarball and re-do everything to be sure and that just would be silly.
* * * Put in a longer timeframe on that ALARM please. * * *
Building a toolchain on the beast will require that I can debug stuff
later. At least the initial first pass toolchain. Of course a GCC
compiler bootstrap will NOT be doing anything like those CFLAGS.
Also why does the maillist not post anything via the web interface :
https://lists.gnu.org/archive/html/bug-m4/2026-03/
--
--
Dennis Clarke
RISC-V/SPARC/PPC/ARM/CISC
UNIX and Linux spoken
