[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Bug in mpn/m68k/lshift.asm
From: |
Kevin Ryde |
Subject: |
Re: Bug in mpn/m68k/lshift.asm |
Date: |
Sat, 03 Aug 2002 08:17:52 +1000 |
User-agent: |
Gnus/5.090007 (Oort Gnus v0.07) Emacs/21.1 (i386-debian-linux-gnu) |
[Bug reports to address@hidden please, not address@hidden
Camm Maguire <address@hidden> writes:
>
> #1 0x80079656 in __gmpz_mul_2exp (w=0xeffffa7c, u=0x8001177c, cnt=2147573446)
> at mul_2exp.c:52
That cnt parameter looks unlikely, but maybe it's just the debugger
playing up since the cnt and limb_cnt below seem ok.
> (gdb) p wp
> $14 = 0x8021aee0
> (gdb) p limb_cnt
> $16 = 0
> (gdb) p abs_usize
> $17 = 1
> (gdb) p cnt
> $18 = 11
> (gdb) p u->_mp_d
> $19 = (mp_limb_t *) 0x801fdfb8
Hmm. It ought to write a limb to 0x8021aee0 and leave 0x8021aedc
alone. You might have to step through the asm code to see if it goes
wrong somewhere.
In the code a1 is the destination pointer and the size in d6 will be
decremented and found to be zero so the only store should be the one
at Lend.
I don't have an m68k to test at the moment, but that code has been
exercised in the past.
- Re: Bug in mpn/m68k/lshift.asm,
Kevin Ryde <=