Does this patch fix the problem of double fault/triple fault?
Leendert van Doorn <> wrote:
> This patch improves VMX exit handling while software interrupts are
> pending: These should not stored as a VMX entry exception but should
> instead just be restarted. This was the cause of the VMRESUME
> failure.
>
> Leendert
>
> Signed-Off-By: Leendert van Doorn <leendert@xxxxxxxxxxxxxx>
>
>
> --- xeno-unstable.orig/xen/arch/x86/vmx.c 2005-06-23
> 09:58:11.000000000 -0400 +++
> xeno-unstable.mine/xen/arch/x86/vmx.c 2005-06-27 21:46:12.000000000
> -0400 @@ -1159,15 +1159,19 @@
>
> __vmread(IDT_VECTORING_INFO_FIELD, &idtv_info_field);
> if (idtv_info_field & INTR_INFO_VALID_MASK) {
> - __vmwrite(VM_ENTRY_INTR_INFO_FIELD, idtv_info_field);
> - if ((idtv_info_field & 0xff) == 14) {
> - unsigned long error_code;
> + if ((idtv_info_field & 0x0700) != 0x400) { /* exclude soft ints */
> + __vmwrite(VM_ENTRY_INTR_INFO_FIELD, idtv_info_field);
>
> - __vmread(VM_EXIT_INTR_ERROR_CODE, &error_code);
> - printk("#PG error code: %lx\n", error_code);
> - }
> - VMX_DBG_LOG(DBG_LEVEL_1, "idtv_info_field=%x",
> - idtv_info_field);
> + if (idtv_info_field & 0x800) { /* valid error code */
> + unsigned long error_code;
> + printk("VMX exit %x: %x/%lx\n",
> + exit_reason, idtv_info_field, error_code);
> + __vmread(VM_EXIT_INTR_ERROR_CODE, &error_code);
> + __vmwrite(VM_ENTRY_EXCEPTION_ERROR_CODE, error_code);
> + } else
> + printk("VMX exit %x: %x\n", exit_reason, idtv_info_field);
> + }
> + VMX_DBG_LOG(DBG_LEVEL_1, "idtv_info_field=%x",
> +idtv_info_field);
> }
>
> /* don't bother H/W interrutps */
> @@ -1399,7 +1403,6 @@
> #else
> asm volatile("movq %0,%%cr2": :"r" (d->arch.arch_vmx.cpu_cr2));
> #endif -
> }
>
> #endif /* CONFIG_VMX */
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@xxxxxxxxxxxxxxxxxxx
> http://lists.xensource.com/xen-devel
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
Home