Re: [RFC PATCH] kvm-unit-tests : Basic architecture of VMX nested test case

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



And what about this version:

static int vmx_run()
{
    u32 ret = 0;

    asm volatile(
        "mov %%rsp, %%rsi\n\t"
        "mov %2, %%edi\n\t"
        "call vmcs_write\n\t"

        "0: "
        LOAD_GPR_C
        "cmpl $0, %1\n\t"
        "jne 1f\n\t"
        "vmlaunch;seta %1\n\t"
        /* vmlaunch error, return VMX_TEST_LAUNCH_ERR */
        "movl %3, %0\n\t"
        SAVE_GPR_C
        "jmp 2f\n\t"
        "1: "
        "vmresume;seta %1\n\t"
        /* vmresume error, return VMX_TEST_RESUME_ERR */
        "movl %4, %0\n\t"
        SAVE_GPR_C
        "jmp 2f\n\t"

        "vmx_return:\n\t"
        SAVE_GPR_C
        "call exit_handler\n\t"
        /* set launched = 1 */
        "movl $0x1, %1\n\t"
        /* jump to resume when VMX_TEST_RESUME */
        "cmp %5, %%eax\n\t"
        "je 0b\n\t"
        "mov %%eax, %0\n\t"
        "2: "
        : "=m"(ret), "=m"(launched)
        : "i"(HOST_RSP), "i"(VMX_TEST_LAUNCH_ERR),
          "i"(VMX_TEST_RESUME_ERR), "i"(VMX_TEST_RESUME)
        : "rax", "rbx", "rdi", "rsi",
         "r8", "r9", "r10", "r11", "r12", "r13", "r14", "r15",
          "memory", "cc"
    );
    switch (ret) {
    case VMX_TEST_VMEXIT:
        launched = 0;
        return 0;
    case VMX_TEST_LAUNCH_ERR:
        printf("%s : vmlaunch failed.\n", __func__);
        if (launched != 0)
            printf("\tvmlaunch set flags error\n");
        report("test vmlaunch", 0);
        break;
    case VMX_TEST_RESUME_ERR:
        printf("%s : vmresume failed.\n", __func__);
        if (launched != 0)
            printf("\tvmlaunch set flags error\n");
        report("test vmresume", 0);
        break;
    default:
        launched = 0;
        printf("%s : unhandled ret from exit_handler, ret=%d.\n",
__func__, ret);
        break;
    }
    return 1;
}

On Wed, Jul 24, 2013 at 6:24 PM, Jan Kiszka <jan.kiszka@xxxxxx> wrote:
> On 2013-07-24 12:16, Arthur Chunqi Li wrote:
>> On Wed, Jul 24, 2013 at 6:03 PM, Jan Kiszka <jan.kiszka@xxxxxx> wrote:
>>> On 2013-07-24 11:56, Arthur Chunqi Li wrote:
>>>> So what about this one. I merged all the exit reason to "ret" and
>>>> remove the flag detection after vmlaunch/vmresume (because I think
>>>> this detection is useless). Currently we support only one guest, so
>>>> variant "launched" is located in vmx_run(). If we want to support
>>>> multiple guest, we could move it to some structures (e.g.
>>>> environment_ctxt). Now I just put it here.
>>>>
>>>> static int vmx_run()
>>>> {
>>>>     u32 ret = 0;
>>>>     bool launched = 0;
>>>>
>>>>     asm volatile(
>>>>         "mov %%rsp, %%rsi\n\t"
>>>>         "mov %2, %%edi\n\t"
>>>>         "call vmcs_write\n\t"
>>>>
>>>>         "0: "
>>>>         LOAD_GPR_C
>>>>         "cmp $0, %1\n\t"
>>>>         "jne 1f\n\t"
>>>>         "vmlaunch\n\t"
>>>>         SAVE_GPR_C
>>>>         /* vmlaunch error, return VMX_TEST_LAUNCH_ERR */
>>>>         "mov %3, %0\n\t"
>>>>         "jmp 2f\n\t"
>>>>         "1: "
>>>>         "vmresume\n\t"
>>>>         SAVE_GPR_C
>>>>         /* vmresume error, return VMX_TEST_RESUME_ERR */
>>>>         "mov %4, %0\n\t"
>>>>         "jmp 2f\n\t"
>>>
>>> Where do you store the flags now? You may want to differentiate / test
>>> if ZF of CF is set.
>> I store the flags as a global variant. You mean I need to detect ZF/CF
>> after vmlaunch/vmresume?
>
> Yes - if you want to check correct emulation of those instructions
> completely.
>
> Jan
>
>



-- 
Arthur Chunqi Li
Department of Computer Science
School of EECS
Peking University
Beijing, China
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html




[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]
  Powered by Linux