Re: [Xen-devel] Re: [xen-unstable test] 8803: regressions - FAIL

[Ian Campbell <Ian.Campbell@xxxxxxxxxx>]

On Thu, 2011-09-01 at 17:23 +0100, Ian Jackson wrote:
> ~xen.org writes ("[xen-unstable test] 8803: regressions - FAIL"):
> > flight 8803 xen-unstable real [real]
> > http://www.chiark.greenend.org.uk/~xensrcts/logs/8803/
> >  build-i386                    4 xen-build              fail REGR. vs. 8769
> 
> gcc -O1 -fno-omit-frame-pointer -m32 -march=i686 -g -fno-strict-aliasing 
> -std=gnu99 -Wall -Wstrict-prototypes -Wno-unused-value -Wdeclaration-after-
> statement  -fno-builtin -fno-common -Wredundant-decls -iwithprefix include 
> -Werror -Wno-pointer-arith -pipe -I/home/osstest/build.8803.build-i386/xe
> n-unstable/xen/include  
> -I/home/osstest/build.8803.build-i386/xen-unstable/xen/include/asm-x86/mach-generic
>  -I/home/osstest/build.8803.build-i386/xe
> n-unstable/xen/include/asm-x86/mach-default -msoft-float -fno-stack-protector 
> -fno-exceptions -Wnested-externs -fno-optimize-sibling-calls -nostdinc
>  -g -D__XEN__ -DVERBOSE -fno-omit-frame-pointer -DCONFIG_FRAME_POINTER -MMD 
> -MF .domain_page.o.d -c domain_page.c -o domain_page.o
> domain_page.c: In function 'map_domain_page_global':
> domain_page.c:211: error: negative width in bit-field '<anonymous>'

Which is:
    /* At least half the ioremap space should be available to us. */
    BUILD_BUG_ON(IOREMAP_VIRT_START + (IOREMAP_MBYTES << 19) >= FIXADDR_START);

Looking at 8769 vs this revision my bet is:
        
        changeset:   23802:bb9b81008733
        user:        Laszlo Ersek <lersek@xxxxxxxxxx>
        date:        Wed Aug 31 15:16:14 2011 +0100
        files:       xen/include/asm-x86/config.h
        description:
        x86: Increase the default NR_CPUS to 256
        
        Changeset 21012:ef845a385014 bumped the default to 128 about one and a
        half years ago. Increase it now to 256, as systems with eg. 160
        logical CPUs are becoming (have become) common.
        
        Signed-off-by: Laszlo Ersek <lersek@xxxxxxxxxx>
        
        
        diff -r d54cfae72cd1 -r bb9b81008733 xen/include/asm-x86/config.h
        --- a/xen/include/asm-x86/config.h      Wed Aug 31 15:15:41 2011 +0100
        +++ b/xen/include/asm-x86/config.h      Wed Aug 31 15:16:14 2011 +0100
        @@ -50,7 +50,7 @@
         #ifdef MAX_PHYS_CPUS
         #define NR_CPUS MAX_PHYS_CPUS
         #else
        -#define NR_CPUS 128
        +#define NR_CPUS 256
         #endif
         
         #ifdef __i386__

I think we need (untested):

# HG changeset patch
# User Ian Campbell <ian.campbell@xxxxxxxxxx>
# Date 1314894881 -3600
# Node ID 4309ff9535001bdca8db93a439edd86bb4c447cd
# Parent  bb97bd46df6c6d8562759a964ebf6c31b6361a7a
xen/x86: only support >128 CPUs on x86_64

32 bit cannot cope with 256 cpus and hits:

    /* At least half the ioremap space should be available to us. */
    BUILD_BUG_ON(IOREMAP_VIRT_START + (IOREMAP_MBYTES << 19) >= FIXADDR_START);

Signed-off-by: Ian Campbell <ian.campbell@xxxxxxxxxx>

diff -r bb97bd46df6c -r 4309ff953500 xen/include/asm-x86/config.h
--- a/xen/include/asm-x86/config.h      Thu Sep 01 16:03:21 2011 +0100
+++ b/xen/include/asm-x86/config.h      Thu Sep 01 17:34:41 2011 +0100
@@ -49,6 +49,8 @@
 
 #ifdef MAX_PHYS_CPUS
 #define NR_CPUS MAX_PHYS_CPUS
+#elif defined __i386__
+#define NR_CPUS 128
 #else
 #define NR_CPUS 256
 #endif



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