[PATCHv3 4/4] test: dm: tee: extend with RPC test

[Simon Glass]

Hi Igor,

On Thu, 14 Jan 2021 at 02:41, Igor Opaniuk <igor.opaniuk at foundries.io> wrote:
>
> Hi Simon,
>
> On Wed, Jan 13, 2021 at 6:12 PM Simon Glass <sjg at chromium.org> wrote:
> >
> > On Tue, 12 Jan 2021 at 01:44, Jorge Ramirez-Ortiz <jorge at foundries.io> wrote:
> > >
> > > From: Igor Opaniuk <igor.opaniuk at foundries.io>
> > >
> > > Extend existing DM tee tests adding test coverage for reverse RPC calls.
> > > Currently this commit only adds tests for I2C requests from TEE driver
> > > to TEE supplicant, for instance reading/writing data to emulated i2c
> > > eeprom defines in standard sandbox test device tree
> > > (arch/sandbox/dts/test.dtb):
> > >
> > > => i2c bus
> > > Bus 0:  i2c at 0  (active 0)
> > >    2c: eeprom at 2c, offset len 1, flags 0
> > >    ...
> > >
> > > Running TEE tests:
> > > => ut dm tee
> > > Test: dm_test_tee: tee.c
> > > Test: dm_test_tee: tee.c (flat tree)
> > > Failures: 0
> > >
> > > Signed-off-by: Igor Opaniuk <igor.opaniuk at foundries.io>
> > > ---
> > >  test/dm/tee.c | 109 +++++++++++++++++++++++++++++++++++++++++++++++---
> > >  1 file changed, 104 insertions(+), 5 deletions(-)
> >
> > Reviewed-by: Simon Glass <sjg at chromium.org>
> >
> > I wonder if the #ifdefs could go away, or at least be reduced. Perhaps
> > just assuming that the option is enabled on sandbox?
>
> I've switched to "if (IS_ENABLED(CONFIG_*))" in test/dm/tee.c, however
> for dropping all ifdefs in sandbox.c likely we have to do a complete overhaul
> of tee sandbox driver structure, moving all TA specific code to separate files
> (`sandbox-avb.c` etc), which IMHO is out of scope of the current patchset.
> I'll address that in another patchset if you don't have objections.

I don't see a great need for it as we do have #ifdefs in tests. But it
might work out nicely. See how you go.

>
> Jorge will send another version of this patchset including my changes
> in test/dm/tee.c.

OK.

Regards,
Simon