On 08/06/2020 10.13, Pierre Morel wrote:
> After a channel is enabled we start a SENSE_ID command using
> the SSCH instruction to recognize the control unit and device.
>
> This tests the success of SSCH, the I/O interruption and the TSCH
> instructions.
>
> The SENSE_ID command response is tested to report 0xff inside
> its reserved field and to report the same control unit type
> as the cu_type kernel argument.
>
> Without the cu_type kernel argument, the test expects a device
> with a default control unit type of 0x3832, a.k.a virtio-net-ccw.
>
> Signed-off-by: Pierre Morel <pmorel@xxxxxxxxxxxxx>
> ---
> lib/s390x/css.h | 20 ++++++
> lib/s390x/css_lib.c | 46 ++++++++++++++
> s390x/css.c | 149 +++++++++++++++++++++++++++++++++++++++++++-
> 3 files changed, 214 insertions(+), 1 deletion(-)
[...]
> +}
> diff --git a/s390x/css.c b/s390x/css.c
> index 6f58d4a..79c997d 100644
> --- a/s390x/css.c
> +++ b/s390x/css.c
> @@ -16,10 +16,26 @@
> #include <string.h>
> #include <interrupt.h>
> #include <asm/arch_def.h>
> +#include <kernel-args.h>
>
> #include <css.h>
>
> +#define DEFAULT_CU_TYPE 0x3832
> +static unsigned long cu_type = DEFAULT_CU_TYPE;
> +
> +struct lowcore *lowcore = (void *)0x0;
> +
> static int test_device_sid;
> +static struct irb irb;
> +static struct senseid senseid;
> +
> +static void set_io_irq_subclass_mask(uint64_t const new_mask)
> +{
> + asm volatile (
> + "lctlg %%c6, %%c6, %[source]\n"
> + : /* No outputs */
> + : [source] "R" (new_mask));
I think the "R" constraint is wrong here - this instruction does not use
an index register. "Q" is likely the better choice. But it might be
easier to use the lctlg() wrapper from lib/s390x/asm/arch_def.h instead.
[...]
> +
> + report((senseid.cu_type == cu_type),
Please drop the innermost parentheses here.
> + "cu_type: expect 0x%04x got 0x%04x",
> + (uint16_t) cu_type, senseid.cu_type);
> +
> +unreg_cb:
> + unregister_io_int_func(irq_io);
> +}
Thomas
