Re: [Qemu-devel] [PATCH v27 8/8] target/avr: Add tests

[Michael Rolnik <mrolnik@gmail.com>]

Hi Thomas.
where should I specify this command?

On Sun, Jul 21, 2019 at 10:13 AM Thomas Huth <thuth@redhat.com> wrote:

> On 19/07/2019 15.26, Philippe Mathieu-Daudé wrote:
> > On 7/19/19 10:26 AM, Michael Rolnik wrote:
> [...]
> >> diff --git a/tests/boot-serial-test.c b/tests/boot-serial-test.c
> >> index 24852d4c7d..73d273b73f 100644
> >> --- a/tests/boot-serial-test.c
> >> +++ b/tests/boot-serial-test.c
> >> @@ -16,6 +16,17 @@
> >>  #include "qemu/osdep.h"
> >>  #include "libqtest.h"
> >>
> >> +static const uint8_t bios_avr[] = {
> >> +    0x89, 0xe1,             /* ldi r24, 0x19   */
> >> +    0x80, 0x93, 0xc5, 0x00, /* sts 0x00C5, r24 ; set baud rate to
> 38400 */
> >
> > FWIW we can remove the previous two lines, we don't care about the
> > baudrate in this test.
> >
> >> +    0x88, 0xe0,             /* ldi r24, 0x08   */
> >> +    0x80, 0x93, 0xc1, 0x00, /* sts 0x00C1, r24 ; Enable tx */
> >> +    0x86, 0xe0,             /* ldi r24, 0x06   */
> >> +    0x80, 0x93, 0xc2, 0x00, /* sts 0x00C2, r24 ; Set the data bits to
> 8 */
> >> +    0x84, 0xe5,             /* ldi r24, 0x54   */
> >> +    0x80, 0x93, 0xc6, 0x00, /* sts 0x00C6, r24 ; Output 'T' */
> >> +};
> >> +
> >>  static const uint8_t kernel_mcf5208[] = {
> >>      0x41, 0xf9, 0xfc, 0x06, 0x00, 0x00,     /* lea 0xfc060000,%a0 */
> >>      0x10, 0x3c, 0x00, 0x54,                 /* move.b #'T',%d0 */
> >> @@ -92,6 +103,7 @@ typedef struct testdef {
> >>
> >>  static testdef_t tests[] = {
> >>      { "alpha", "clipper", "", "PCI:" },
> >> +    { "avr", "sample", "", "T", sizeof(bios_avr), NULL, bios_avr },
> >>      { "ppc", "ppce500", "", "U-Boot" },
> >>      { "ppc", "40p", "-vga none -boot d", "Trying cd:," },
> >>      { "ppc", "g3beige", "", "PowerPC,750" },
> >>
> >
> > Testing shows:
> >
> >   TEST    check-qtest-avr: tests/boot-serial-test
> > qemu-system-avr: Unable to load /tmp/qtest-boot-serial-cOndewD as ELF,
> > trying again as raw binary
> >
> > I wonder if this might fail Peter's testing, so Cc'ing Thomas.
>
> Such messages are quite a bit anoying during "make check", indeed. Could
> you please fence the message with qtest_enabled() ?
>
>  Thanks,
>    Thomas
>


-- 
Best Regards,
Michael Rolnik