Re: Fixup INTEGER_CST

2012-10-08 Thread Jan Hubicka
> > 2) As we query the type_hash while we are rewritting the types, > > we run into instability of the hashtable. This manifests itself > > as an ICE when one adds sanity check that while merging function > > types their arg types are equivalent, too. > > This ICEs compiling i

Re: Fixup INTEGER_CST

2012-10-08 Thread Jan Hubicka
> > 2) As we query the type_hash while we are rewritting the types, > > we run into instability of the hashtable. This manifests itself > > as an ICE when one adds sanity check that while merging function > > types their arg types are equivalent, too. > > This ICEs compiling i

Re: Fixup INTEGER_CST

2012-10-08 Thread Richard Guenther
On Mon, Oct 8, 2012 at 11:18 AM, Jan Hubicka wrote: >> On Sun, Oct 7, 2012 at 7:22 PM, Jan Hubicka wrote: >> >> On Sun, Oct 7, 2012 at 5:15 PM, Jan Hubicka wrote: >> >> > Hi, >> >> > I added a santy check that after fixup all types that lost in the >> >> > merging are >> >> > really dead. And

Re: Fixup INTEGER_CST

2012-10-08 Thread Jan Hubicka
> On Sun, Oct 7, 2012 at 7:22 PM, Jan Hubicka wrote: > >> On Sun, Oct 7, 2012 at 5:15 PM, Jan Hubicka wrote: > >> > Hi, > >> > I added a santy check that after fixup all types that lost in the > >> > merging are > >> > really dead. And it turns out we have some zombies around. > >> > > >> > INT

Re: Fixup INTEGER_CST

2012-10-08 Thread Richard Guenther
On Sun, Oct 7, 2012 at 7:22 PM, Jan Hubicka wrote: >> On Sun, Oct 7, 2012 at 5:15 PM, Jan Hubicka wrote: >> > Hi, >> > I added a santy check that after fixup all types that lost in the merging >> > are >> > really dead. And it turns out we have some zombies around. >> > >> > INTEGER_CST needs s

Re: Fixup INTEGER_CST

2012-10-07 Thread Jan Hubicka
NTEGER_CST with overflow is changed in place because it is allowed to do so). Honza > > Richard. > > > > > PR lto/54839 > > * lto/lto.c (remember_with_vars): Also fixup INTEGER_CST. > >

Re: Fixup INTEGER_CST

2012-10-07 Thread Richard Guenther
em where they end up used unfixed. Richard. > > PR lto/54839 > * lto/lto.c (remember_with_vars): Also fixup INTEGER_CST. > (fixup_integer_cst): New functoin. > (lto_ft_type): Fixup BASETYPE of methods and offsets. > Index: lto/lto.c > =

Fixup INTEGER_CST

2012-10-07 Thread Jan Hubicka
corrupt the hashtable used by tree.c's sharing code Bootstrapped/regtested x86_64-linux, OK? PR lto/54839 * lto/lto.c (remember_with_vars): Also fixup INTEGER_CST. (fixup_integer_cst): New functoin. (lto_ft_type): Fixup BASETYPE of methods and offsets. Index