Re: [Fwd: stack overflow "cause" found]

[Rik van Riel <riel@redhat.com>]

[-- Attachment #1: Type: text/plain, Size: 856 bytes --]

Ian Pratt wrote:
>>> Xen really should not be doing this.  We could either uninline some of
>>> these functions, or try gcc -O2 or gcc -Os.
>> OK, with -Os it doesn't quite build due to the built-ins not being
>> inlined.  With -O2 the stack usage changes dramatically...
> 
> It would be nice if -O2 worked. Quite a scary change give the amount of inline asm  though...

Not only does -O2 seem to work, but without it we get stack
overflows and things suddenly stop working.

I've been beating up my x86-64 system with a few VT and paravirt
guests for over an hour now, and it no longer crashes.  Without
-O2, I can make it crash within minutes...

-- 
"Debugging is twice as hard as writing the code in the first place.
Therefore, if you write the code as cleverly as possible, you are,
by definition, not smart enough to debug it." - Brian W. Kernighan

[-- Attachment #2: O2 --]
[-- Type: text/plain, Size: 461 bytes --]

--- xen/Rules.mk.noop	2006-07-26 17:44:02.000000000 -0400
+++ xen/Rules.mk	2006-07-26 17:44:16.000000000 -0400
@@ -52,7 +52,7 @@
 ALL_OBJS-$(ACM_SECURITY) += $(BASEDIR)/acm/built_in.o
 ALL_OBJS-y               += $(BASEDIR)/arch/$(TARGET_ARCH)/built_in.o
 
-CFLAGS-y               += -g -D__XEN__
+CFLAGS-y               += -g -D__XEN__ -O2
 CFLAGS-$(ACM_SECURITY) += -DACM_SECURITY
 CFLAGS-$(verbose)      += -DVERBOSE
 CFLAGS-$(crash_debug)  += -DCRASH_DEBUG

[-- Attachment #3: Type: text/plain, Size: 138 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel