Hi Michal, On Wed, 13 May 2026 at 02:53, Michal Simek <[email protected]> wrote: > > > > On 5/7/26 23:11, Simon Glass wrote: > > Hi Michal, > > > > On 2026-05-05T12:30:28, Michal Simek <[email protected]> wrote: > >> reset: Add sandbox tests for reset_reset() and reset_reset_bulk() > >> > >> Add DM test coverage for the new reset_reset() and reset_reset_bulk() > >> API functions. The tests exercise the assert + deassert fallback path > >> since the sandbox reset driver does not implement the rst_reset op. > >> > >> Signed-off-by: Michal Simek <[email protected]> > >> > >> arch/sandbox/include/asm/reset.h | 3 ++ > >> drivers/reset/sandbox-reset-test.c | 14 ++++++++ > >> drivers/reset/sandbox-reset.c | 31 ++++++++++++++++++ > >> test/dm/reset.c | 67 > >> ++++++++++++++++++++++++++++++++++++++ > >> 4 files changed, 115 insertions(+) > > > >> diff --git a/drivers/reset/sandbox-reset.c b/drivers/reset/sandbox-reset.c > >> @@ -66,6 +69,21 @@ static int sandbox_reset_deassert(struct reset_ctl > >> *reset_ctl) > >> return 0; > >> } > >> > >> +static int sandbox_reset_reset(struct reset_ctl *reset_ctl, ulong > >> delay_us) > >> +{ > >> + struct sandbox_reset *sbr = dev_get_priv(reset_ctl->dev); > >> + > >> + debug("%s(reset_ctl=%p, delay_us=%lu)\n", __func__, reset_ctl, > >> + delay_us); > >> + > >> + sbr->signals[reset_ctl->id].asserted = true; > >> + udelay(delay_us); > >> + sbr->signals[reset_ctl->id].asserted = false; > >> + sbr->signals[reset_ctl->id].reset_count++; > >> + > >> + return 0; > >> +} > > > > Reviewed-by: Simon Glass <[email protected]> > > > > Some optional nits / thoughts below. > > > > The commit message says the tests exercise the fallback path since the > > sandbox reset driver doesn't have rst_reset op, this patch adds that > > method below - so the commit message could use an update. > > Will update. > > > > > If you want coverage for the fallback you could add a second > > sandbox-reset driver without rst_reset, or hack the rest to clear and > > restart ops->rst_reset, > > > >> diff --git a/drivers/reset/sandbox-reset-test.c > >> b/drivers/reset/sandbox-reset-test.c > >> @@ -96,6 +96,20 @@ int sandbox_reset_test_deassert_bulk(struct udevice > >> *dev) > >> +int sandbox_reset_test_reset(struct udevice *dev) > >> +{ > >> + struct sandbox_reset_test *sbrt = dev_get_priv(dev); > >> + > >> + return reset_reset(sbrt->ctlp, 0); > >> +} > >> + > >> +int sandbox_reset_test_reset_bulk(struct udevice *dev) > >> +{ > >> + struct sandbox_reset_test *sbrt = dev_get_priv(dev); > >> + > >> + return reset_reset_bulk(sbrt->bulkp, 0); > >> +} > > > > You could pass a non-zero value here if you want to test the delay. > > udelay is called and because there is missing check for 0 value it does timer > reading once. Don't think that make sense to delay tests.
We try to avoid delays in tests, although 10us wouldn't matter. We do have timer_test_add_offset() but I don't think it helps with udelay() Regards, Simon
