Bug#338148: [alpha] regression: ocamlopt.opt segfaults

2005-12-10 Thread Falk Hueffner
Falk Hueffner <[EMAIL PROTECTED]> writes: > Maybe like this: Actually, what I really want is: --- ocaml-3.09.0.orig/asmcomp/alpha/emit.mlp2005-10-14 18:41:06.0 +0 200 +++ ocaml-3.09.0/asmcomp/alpha/emit.mlp 2005-12-10 20:50:11.0 +0100 @@ -740,7 +740,7 @@ ` .fmask 0x0,

Bug#338148: [alpha] regression: ocamlopt.opt segfaults

2005-12-10 Thread Falk Hueffner
Mike Furr <[EMAIL PROTECTED]> writes: > Falk Hueffner wrote: >> AFAICS, the caml_apply2 posted to >> http://caml.inria.fr/mantis/view.php?id=3912 does not indicate an >> actual problem. ".prologue 0" means the function will not rely on pv >> ($27) being set correctly, and it will not clobber gp ($

Bug#338148: [alpha] regression: ocamlopt.opt segfaults

2005-12-10 Thread Mike Furr
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Falk Hueffner wrote: > AFAICS, the caml_apply2 posted to > http://caml.inria.fr/mantis/view.php?id=3912 does not indicate an > actual problem. ".prologue 0" means the function will not rely on pv > ($27) being set correctly, and it will not clobber gp

Bug#338148: [alpha] regression: ocamlopt.opt segfaults

2005-12-10 Thread Falk Hueffner
Hi, AFAICS, the caml_apply2 posted to http://caml.inria.fr/mantis/view.php?id=3912 does not indicate an actual problem. ".prologue 0" means the function will not rely on pv ($27) being set correctly, and it will not clobber gp ($29). Clobbering pv is OK. So the problem is probably elsewhere. --