movl issue

movl issue

[Flavio Leitner]

Last one, now it is compiling.


----

Fixed to use 'mov' instead of 'movl' when dealing with segments.

Signed-off-by: Flavio B. Leitner <fbl@conectiva.com.br>

===== linux-2.6.11-xen-sparse/include/asm-xen/asm-i386/mmu_context.h 1.16 vs edited =====
--- 1.16/linux-2.6.11-xen-sparse/include/asm-xen/asm-i386/mmu_context.h	2005-06-01 12:37:20 -04:00
+++ edited/linux-2.6.11-xen-sparse/include/asm-xen/asm-i386/mmu_context.h	2005-06-29 18:03:08 -04:00
@@ -34,7 +34,7 @@
 	 * are always kernel segments while inside the kernel. Must
 	 * happen before reload of cr3/ldt (i.e., not in __switch_to).
 	 */
-	__asm__ __volatile__ ( "movl %%fs,%0 ; movl %%gs,%1"
+	__asm__ __volatile__ ( "mov %%fs,%0 ; mov %%gs,%1"
 		: "=m" (*(int *)&current->thread.fs),
 		  "=m" (*(int *)&current->thread.gs));
 	__asm__ __volatile__ ( "movl %0,%%fs ; movl %0,%%gs"
===== linux-2.6.11-xen-sparse/include/asm-xen/asm-i386/system.h 1.21 vs edited =====
--- 1.21/linux-2.6.11-xen-sparse/include/asm-xen/asm-i386/system.h	2005-06-16 07:43:56 -04:00
+++ edited/linux-2.6.11-xen-sparse/include/asm-xen/asm-i386/system.h	2005-06-29 17:37:46 -04:00
@@ -84,7 +84,7 @@
 #define loadsegment(seg,value)			\
 	asm volatile("\n"			\
 		"1:\t"				\
-		"movl %0,%%" #seg "\n"		\
+		"mov %0,%%" #seg "\n"		\
 		"2:\n"				\
 		".section .fixup,\"ax\"\n"	\
 		"3:\t"				\
@@ -102,7 +102,7 @@
  * Save a segment register away
  */
 #define savesegment(seg, value) \
-	asm volatile("movl %%" #seg ",%0":"=m" (*(int *)&(value)))
+	asm volatile("mov %%" #seg ",%0":"=m" (*(int *)&(value)))
 
 /*
  * Clear and set 'TS' bit respectively
----
-- 
Flávio Bruno Leitner <fbl@conectiva.com.br>
[0EA2 7F40 4CF4 1E63 4AF6  33C0 3E10 E205 F251 EDDA]

Re: movl issue

[Vincent Hanquez]

On Thu, Jun 30, 2005 at 09:29:39AM -0300, Flavio Leitner wrote:
> Last one, now it is compiling.
> Fixed to use 'mov' instead of 'movl' when dealing with segments.

thoses one will get fix when we will upgrade the sparse tree to 2.6.12,
which should be done soon.

(the upgrading patch is available for testing for anybody wanting it)

-- 
Vincent Hanquez

Re: movl issue

[Flavio Leitner]

On Thu, Jun 30, 2005 at 03:33:47PM +0200, Vincent Hanquez wrote:
> On Thu, Jun 30, 2005 at 09:29:39AM -0300, Flavio Leitner wrote:
> > Last one, now it is compiling.
> > Fixed to use 'mov' instead of 'movl' when dealing with segments.
> 
> thoses one will get fix when we will upgrade the sparse tree to 2.6.12,
> which should be done soon.

Ok, no problem. 

> (the upgrading patch is available for testing for anybody wanting it)

Where I can get it? 


-- 
Flávio Bruno Leitner <fbl@conectiva.com.br>
[0EA2 7F40 4CF4 1E63 4AF6  33C0 3E10 E205 F251 EDDA]

Re: movl issue

[Rik Van Riel]

On Thu, 30 Jun 2005, Vincent Hanquez wrote:

> (the upgrading patch is available for testing for anybody wanting it)

Where is it ? ;)

-- 
The Theory of Escalating Commitment: "The cost of continuing mistakes is
borne by others, while the cost of admitting mistakes is borne by yourself."
  -- Joseph Stiglitz, Nobel Laureate in Economics

Re: movl issue

[Vincent Hanquez]

Hi,

sorry for the time to answer,

I've put the patch here:

http://www.cl.cam.ac.uk/Research/SRG/netos/xen/downloads/temp/xen-upgrade-2612-rc1.diff.bz2

before applying the patch you must rename
linux-2.6.11-xen-sparse to linux-2.6.12-xen-sparse

then do:
mkdir patches/linux-2.6.12
cp patches/linux-2.6.11/smp-alts.patch patches/linux-2.6.12/
cp patches/linux-2.6.11/x86_64-linux.patch patches/linux-2.6.12/

2 last caveats:
  - serial console problem on i386 with SMP
  - compilation on x86_64 fail, but you can just remove the offending
    line in arch/xen/kernel/vmlinux.lds.S

      phys_startup_64 = startup_64 - LOAD_OFFSET;

    this is for the moment harmless to remove it, since it's
    related to kexec and it will works just fine without this.

we would like to hear if there's any others regressions to fix.
it would be nice before upgrading the real repository for everybody.

thanks,
-- 
Vincent Hanquez

Re: movl issue

[Chris Wright]

* Vincent Hanquez (vincent.hanquez@cl.cam.ac.uk) wrote:
> Hi,
> 
> sorry for the time to answer,
> 
> I've put the patch here:
> 
> http://www.cl.cam.ac.uk/Research/SRG/netos/xen/downloads/temp/xen-upgrade-2612-rc1.diff.bz2

Last patch and this one have been working fine for me.  This one spews
warnings such as:

include/asm/hw_irq.h:56:1: warning: "CALL_FUNCTION_VECTOR" redefined
In file included from include/asm-xen/asm/irq.h:16,
                 from include/linux/irq.h:21,
                 from include/asm/hardirq.h:6,
                 from include/linux/hardirq.h:6,
                 from include/linux/interrupt.h:11,
                 from include/linux/netdevice.h:514,
                 from include/linux/icmpv6.h:154,
                 from net/compat.c:19:
include/asm-xen/asm-x86_64/mach-xen/irq_vectors.h:85:1: warning: this is the location of the previous definition

include/asm/hw_irq.h:53:1: warning: "INVALIDATE_TLB_VECTOR" redefined
In file included from include/asm-xen/asm/irq.h:16,
                 from include/linux/irq.h:21,
                 from include/asm/hardirq.h:6,
                 from include/linux/hardirq.h:6,
                 from include/linux/interrupt.h:11,
                 from include/net/sock.h:48,
                 from lib/kobject_uevent.c:24:
include/asm-xen/asm-x86_64/mach-xen/irq_vectors.h:84:1: warning: this is the location of the previous definition


include/asm/hw_irq.h:54:1: warning: "RESCHEDULE_VECTOR" redefined
In file included from include/asm-xen/asm/irq.h:16,
                 from include/linux/irq.h:21,
                 from include/asm/hardirq.h:6,
                 from include/linux/hardirq.h:6,
                 from include/linux/interrupt.h:11,
                 from include/linux/netdevice.h:514,
                 from include/net/sock.h:48,
                 from lib/kobject_uevent.c:24:
include/asm-xen/asm-x86_64/mach-xen/irq_vectors.h:83:1: warning: this is the location of the previous definition


Also, below if rediffed portion of xen/i386/kernel/Makefile which no
longer applied due to recent changes. (untested)

--- a/linux-2.6.12-xen-sparse/arch/xen/i386/kernel/Makefile.orig	2005-07-01 14:14:31.000000000 -0700
+++ b/linux-2.6.12-xen-sparse/arch/xen/i386/kernel/Makefile	2005-07-01 14:38:16.000000000 -0700
@@ -33,6 +33,7 @@ obj-$(CONFIG_X86_MPPARSE)	+= mpparse.o
 obj-$(CONFIG_X86_LOCAL_APIC)	+= apic.o
 c-obj-$(CONFIG_X86_LOCAL_APIC)	+= nmi.o
 obj-$(CONFIG_X86_IO_APIC)	+= io_apic.o
+c-obj-$(CONFIG_X86_REBOOTFIXUPS)+= reboot_fixups.o
 c-obj-$(CONFIG_X86_NUMAQ)	+= numaq.o
 c-obj-$(CONFIG_X86_SUMMIT_NUMA)	+= summit.o
 c-obj-$(CONFIG_MODULES)		+= module.o
@@ -53,11 +54,11 @@ c-obj-$(CONFIG_SCx200)		+= scx200.o
 # Note: kbuild does not track this dependency due to usage of .incbin
 $(obj)/vsyscall.o: $(obj)/vsyscall-int80.so $(obj)/vsyscall-sysenter.so
 targets += $(foreach F,int80 sysenter,vsyscall-$F.o vsyscall-$F.so)
-targets += vsyscall.lds
+targets += vsyscall-note.o vsyscall.lds
 
 # The DSO images are built using a special linker script.
 quiet_cmd_syscall = SYSCALL $@
-      cmd_syscall = $(CC) -nostdlib -m32 $(SYSCFLAGS_$(@F)) \
+      cmd_syscall = $(CC) -m elf_i386 -nostdlib $(SYSCFLAGS_$(@F)) \
 		          -Wl,-T,$(filter-out FORCE,$^) -o $@
 
 export CPPFLAGS_vsyscall.lds += -P -C -U$(ARCH)
@@ -67,7 +68,8 @@ SYSCFLAGS_vsyscall-sysenter.so	= $(vsysc
 SYSCFLAGS_vsyscall-int80.so	= $(vsyscall-flags)
 
 $(obj)/vsyscall-int80.so $(obj)/vsyscall-sysenter.so: \
-$(obj)/vsyscall-%.so: $(src)/vsyscall.lds $(obj)/vsyscall-%.o FORCE
+$(obj)/vsyscall-%.so: $(src)/vsyscall.lds \
+		      $(obj)/vsyscall-%.o FORCE
 	$(call if_changed,syscall)
 
 # We also create a special relocatable object that should mirror the symbol
@@ -82,13 +84,15 @@ $(obj)/vsyscall-syms.o: $(src)/vsyscall.
 	$(call if_changed,syscall)
 
 c-link	:= init_task.o
-s-link	:= vsyscall-int80.o vsyscall-sysenter.o vsyscall-sigreturn.o vsyscall.lds.o
+s-link	:= vsyscall-int80.o vsyscall-sysenter.o vsyscall-sigreturn.o vsyscall.lds.o syscall_table.o
 
 $(patsubst %.o,$(obj)/%.c,$(c-obj-y) $(c-obj-m) $(c-link)) $(patsubst %.o,$(obj)/%.S,$(s-obj-y) $(s-link)):
 	@ln -fsn $(srctree)/arch/i386/kernel/$(notdir $@) $@
 
 $(obj)/vsyscall-int80.S: $(obj)/vsyscall-sigreturn.S
 
+$(obj)/entry.o: $(src)/entry.S $(src)/syscall_table.S
+
 obj-y	+= $(c-obj-y) $(s-obj-y)
 obj-m	+= $(c-obj-m)

Re: movl issue

[Vincent Hanquez]

On Fri, Jul 01, 2005 at 02:12:42PM -0700, Chris Wright wrote:
> Last patch and this one have been working fine for me.  This one spews
> warnings such as:
> 
> [snip warnings]

Didn't you forget to move the file x86_64-linux.patch in
patches/linux-2.6.12/ ?

this patch should specificly fix thoses warnings.

> Also, below if rediffed portion of xen/i386/kernel/Makefile which no
> longer applied due to recent changes. (untested)

ok thanks

-- 
Vincent Hanquez

Re: movl issue

[Chris Wright]

* Vincent Hanquez (vincent.hanquez@cl.cam.ac.uk) wrote:
> On Fri, Jul 01, 2005 at 02:12:42PM -0700, Chris Wright wrote:
> > Last patch and this one have been working fine for me.  This one spews
> > warnings such as:
> > 
> > [snip warnings]
> 
> Didn't you forget to move the file x86_64-linux.patch in
> patches/linux-2.6.12/ ?
> 
> this patch should specificly fix thoses warnings.

hmm, entirely possible, i'll double check.  indeed, looks like I missed that
one this time.  thanks vincent.

thanks,
-chris