Re: [PATCH 4/6] xen/x86/pvh: Set up descriptors for 32-bit PVH guests

From: Boris Ostrovsky
Date: Tue Jul 07 2015 - 16:44:42 EST


On 07/07/2015 03:54 PM, Konrad Rzeszutek Wilk wrote:
On Mon, Jul 06, 2015 at 11:34:23PM -0400, Boris Ostrovsky wrote:
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@xxxxxxxxxx>
---
arch/x86/xen/enlighten.c | 16 ++++++++++++----
1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index 0b95c9b..7953e68 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -1362,12 +1362,12 @@ static void __init xen_boot_params_init_edd(void)
static void __ref xen_setup_gdt(int cpu)
{
if (xen_feature(XENFEAT_auto_translated_physmap)) {
-#ifdef CONFIG_X86_64
unsigned long dummy;
- load_percpu_segment(cpu); /* We need to access per-cpu area */
+ setup_stack_canary_segment(cpu);
switch_to_new_gdt(cpu); /* GDT and GS set */
+#ifdef CONFIG_X86_64
/* We are switching of the Xen provided GDT to our HVM mode
* GDT. The new GDT has __KERNEL_CS with CS.L = 1
* and we are jumping to reload it.
@@ -1392,8 +1392,16 @@ static void __ref xen_setup_gdt(int cpu)
loadsegment(ds, 0);
loadsegment(fs, 0);
#else
- /* PVH: TODO Implement. */
- BUG();
+ asm volatile ("ljmp %1,$1f\n"
+ "1:\n"
+ "movl %2,%0\n"
+ "movl %0,%%ss\n"
+ "movl %3,%0\n"
+ "movl %0,%%ds\n"
+ "movl %0,%%es\n"
+ : "=&r" (dummy)
+ : "i" (__KERNEL_CS), "i" (__KERNEL_DS),
+ "i" (__USER_DS));
#endif
Which comes out to:
ljmp $96,$1f #[KERNEL_CS]
1:
movl $104,%eax #, dummy #[KERNEL_DS]
movl %eax,%ss # dummy
movl $123,%eax #, dummy #[USER_DS]
movl %eax,%ds # dummy
movl %eax,%es # dummy

And that looks right

Actually, now that look at this again I think I should just use loadsegment(), just like 64-bit code above.

-boris



Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx>

return; /* PVH does not need any PV GDT ops. */
}
--
1.8.1.4


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/