|
|
|
|
|
|
|
|
|
|
xen-devel
RE: [Xen-devel] RE: x86_64 SMP support (status update)
Chris Wright wrote:
> Yes, here it is, there's only one line change (sorry, I
> thought this
> change was in the last one). This should get you to
> faulting on the
> first CPU (which is actually progress ;-). I think the
> second CPU is
> happliy idling in cpu_idle, and the first one is back to
> finishing up
> booting when it dies.
Chris, cpu_gdt_table in latest bk has only 15 entries, I think that's
not correct.
______________________________________________________________
ENTRY(cpu_gdt_table)
/* The TLS descriptors are currently at a different place compared to
i386.
Hopefully nobody expects them at a fixed place (Wine?) */
.quad 0x0000000000000000 /* NULL descriptor */
.quad 0x008ffa000000ffff /* __KERNEL_COMPAT32_CS */
.quad 0x00affa000000ffff /* __KERNEL_CS */
.quad 0x00cff2000000ffff /* __KERNEL_DS */
.quad 0x00cffa000000ffff /* __USER32_CS */
.quad 0x00cff2000000ffff /* __USER_DS, __USER32_DS */
.quad 0x00affa000000ffff /* __USER_CS */
.quad 0x00cffa000000ffff /* __KERNEL32_CS */
.quad 0,0 /* TSS */
.quad 0 /* LDT */
.quad 0,0,0 /* three TLS descriptors */
.quad 0 /* unused now */
gdt_end:
/* asm/segment.h:GDT_ENTRIES must match this */
/* This should be a multiple of the cache line size */
/* GDTs of other CPUs: */
.fill (GDT_SIZE * NR_CPUS) - (gdt_end - cpu_gdt_table)
______________________________________________________________
On x86_64, GDT_SIZE = GDT_ENTRIES * 8, and GDT_ENTRIES = 16. seems we
missed one LDT entry here comparing to native x86_64 linux kernel.
-Xin
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
|
|
|
|