On Fri, 6 Dec 2019 17:26:25 +0100
Pierre Morel <pmorel@xxxxxxxxxxxxx> wrote:
> First step for testing the channel subsystem is to enumerate the css and
> retrieve the css devices.
>
> This tests the success of STSCH I/O instruction.
>
> Signed-off-by: Pierre Morel <pmorel@xxxxxxxxxxxxx>
> ---
> lib/s390x/css.h | 1 +
> s390x/Makefile | 2 ++
> s390x/css.c | 82 +++++++++++++++++++++++++++++++++++++++++++++
> s390x/unittests.cfg | 4 +++
> 4 files changed, 89 insertions(+)
> create mode 100644 s390x/css.c
>
> +static void test_enumerate(void)
> +{
> + struct pmcw *pmcw = &schib.pmcw;
> + int scn;
> + int cc, i;
> + int found = 0;
> +
> + for (scn = 0; scn < 0xffff; scn++) {
> + cc = stsch(scn|SID_ONE, &schib);
> + if (!cc && (pmcw->flags & PMCW_DNV)) {
Not sure when dnv is actually applicable... it is used for I/O
subchannels; chsc subchannels don't have a device; message subchannels
use a different bit IIRC; not sure about EADM subchannels.
[Not very relevant as long as we run under KVM, but should be
considered if you plan to run this test under z/VM or LPAR as well.]
> + report_info("SID %04x Type %s PIM %x", scn,
> + Channel_type[PMCW_CHANNEL_TYPE(pmcw)],
> + pmcw->pim);
> + for (i = 0; i < 8; i++) {
> + if ((pmcw->pim << i) & 0x80) {
> + report_info("CHPID[%d]: %02x", i,
> + pmcw->chpid[i]);
> + break;
That 'break;' seems odd -- won't you end up printing the first chpid in
the pim only?
Maybe modify this loop to print the chpid if the path is in the pim,
and 'n/a' or so if not?
> + }
> + }
> + found++;
> + }
> + if (cc == 3) /* cc = 3 means no more channel in CSS */
s/channel/subchannels/
> + break;
> + if (found && !test_device_sid)
> + test_device_sid = scn|SID_ONE;
> + }
> + if (!found) {
> + report("Tested %d devices, none found", 0, scn);
> + return;
> + }
> + report("Tested %d devices, %d found", 1, scn, found);
> +}
