RE: [PATCH v3 11/12] gdb, testsuite, lib: Add libipt version check.

["Willgerodt, Felix" <felix.willgerodt@intel.com>]

Hi,

Sorry for taking so long to get back to this. This time I will follow
through with this better. I will send the rest of my responses
for the other patches soon, but I have one question before doing
that.

See inline below.

Thanks,
Felix

> -----Original Message-----
> From: Metzger, Markus T <markus.t.metzger@intel.com>
> Sent: Freitag, 13. August 2021 12:36
> To: Willgerodt, Felix <felix.willgerodt@intel.com>
> Cc: gdb-patches@sourceware.org
> Subject: RE: [PATCH v3 11/12] gdb, testsuite, lib: Add libipt version check.
> 
> Thanks, Felix,
> 
> >-# Return 1 if version MAJOR.MINOR is at least
> >AT_LEAST_MAJOR.AT_LEAST_MINOR.
> >-proc version_at_least { major minor at_least_major at_least_minor} {
> >+# Return 1 if version MAJOR.MINOR.REVISION is at least
> >+# AT_LEAST_MAJOR.AT_LEAST_MINOR.AT_LEAST_REVISION.
> >+proc version_at_least { major minor revision at_least_major
> at_least_minor \
> >+			at_least_revision } {
> 
> In which case do we need to check for a particular patch version?  There
> shouldn't be any changes other than bug fixes between patch versions.
> 

I don't think it hurts if this can check for more than we need, it is a
generic function. But for our ULI tests, which I yet have to upstream,
we would actually check for 2.0.4 for UIRET.

> 
> >+# Run a test on the target to see if we have a minimum libipt version.
> >+# Return 0 if so, 1 if it does not.
> >+
> >+proc require_libipt_version { major minor revision } {
> 
> This is an indirect check for, I assume, ptwrite support.  Would it instead be
> possible to just try it?  E.g. compile a trivial program that contains a ptwrite
> and see if it runs and can be decoded without decode error.
> 
> This would cover compiler support for ptwrite as well as GDB support for PT
> in general and for ptwrite in particular.

I have tried implementing that, but I get no error (or other) indication in any
GDB command with libipt 1.6. I added ptwrite support to
"maintenance btrace packet-history" instead to check for ptwrite there.

Is that approach fine with you? It makes skip_btrace_ptw_tests() a bit more
complicated though.
Intel Deutschland GmbH
Registered Address: Am Campeon 10, 85579 Neubiberg, Germany
Tel: +49 89 99 8853-0, www.intel.de <http://www.intel.de>
Managing Directors: Christin Eisenschmid, Sharon Heck, Tiffany Doon Silva  
Chairperson of the Supervisory Board: Nicole Lau
Registered Office: Munich
Commercial Register: Amtsgericht Muenchen HRB 186928