[9front] Weird >>= operator behaviour in 5c/5l

[Jonas Amoson <jonas.amoson@gmail.com>]

Dear list,

It looks to me that the operation x >>= 0 doesn't generate the
correct result with 5c/5l (32-bit arm on Rpi 1b+). Example:

x = 2;
x >>= 0;

Gives x == 0 on arm, but on 386 it yields x == 2 as expected.
The non shorted form: x = x >> 0; Gives the correct result.
Is this a known issue, or I am missing something?

Looking at the assembly, the operation seems to be eliminated
if using the long form (giving the correct result) but using the
shorted form generates an instruction SRL $0, R1, R2, which
doesn't seem to work. Other shifts (e.g. x>>=1) do work (e.g.
generating SRL $1, R1, R2).

//Jonas Amoson

#include <u.h>
#include <libc.h>

void
main()
{
        unsigned int x = 2;
        print("%ud\n", x >>= 0);
        exits(nil);
}