[Th] rpm at sparc64

[Jakub Bogusz]

On Mon, Jun 18, 2007 at 02:40:27PM -0400, Jeff Johnson wrote:
> On Jun 18, 2007, at 2:10 PM, Tomasz Mateja wrote:
> >> removed and it's built right now, btw binutils are current.
> >>
> > not working:
> >
> > [root at moon RPMS]# gdb rpm
> > GNU gdb 6.6
> > Copyright (C) 2006 Free Software Foundation, Inc.
> > GDB is free software, covered by the GNU General Public License,  
> > and you are
> > welcome to change it and/or distribute copies of it under certain
> > conditions.
> > Type "show copying" to see the conditions.
> > There is absolutely no warranty for GDB.  Type "show warranty" for  
> > details.
> > This GDB was configured as "sparc64-pld-linux"...
> > Using host libthread_db library "/lib64/libthread_db.so.1".
> > (gdb) run --rebuilddb
> > Starting program: /bin/rpm --rebuilddb
> >
> > Program received signal SIGBUS, Bus error.
> > 0xfffff8010000fbec in intGetEntry (h=0x26b4e0, tag=1184,
> > type=0x7feffef9518, p=0x7feffef951c, c=0x7feffef950c, minMem=1) at
> > header.c:1785
> > 1785            if (p) *p = NULL;
> 
> Likely alignment on sparc64. Pointers aligned on 64bit boundary on  
> sparc64?
> 
> One hack-a-round is:
>      memset(p, 0, sizeof(*p));
> instead of
>      *p = NULL;
> but you may die in caller instead.

Something odd.

In rpmdbAdd() (rpmdb.c) hcolor is declared locally as uint32_t (so it's
OK for it to be just 4-byte aligned).
Then (in rpmdb.c:3125) its address is casted to void** and passed as *p
through hge() finally to intGetEntry().
So intGetEntry() tries to set hcolor as (64-bit) pointer... and on
sparc64 it finally crashes on alignment, but on x86_64 it seems that it
overwrites some other value on stack (signalMask?). Am I wrong?

> FWIW, tag=1184 is RPMTAG_PACKAGECOLOR added on
> multilib systems. That's likely not PLD/sparc64.

That sparc64 experiment uses multilib.


-- 
Jakub Bogusz    http://qboosh.pl/