[Intel-gfx] [PATCH igt] lib: Check and report if a subtest triggers a new kernel taint

[Szwichtenberg, Radoslaw]

On Wed, 2017-11-29 at 12:40 +0000, Chris Wilson wrote:
> Quoting Chris Wilson (2017-11-29 12:30:23)
> > Checking for a tainted kernel is a convenient way to see if the test
> > generated a critical error such as a oops, or machine check.
> > 
> > Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> > Cc: Daniel Vetter <daniel.vetter at ffwll.ch>
> > Cc: Radoslaw Szwichtenberg <radoslaw.szwichtenberg at intel.com>
> > ---
> > diff --git a/lib/igt_kernel_taint.c b/lib/igt_kernel_taint.c
> > new file mode 100644
> > index 00000000..86d9cd20
> > --- /dev/null
> > +++ b/lib/igt_kernel_taint.c
> > @@ -0,0 +1,95 @@
> > +/*
> > + * Copyright 2017 Intel Corporation
> > + *
> > + * Permission is hereby granted, free of charge, to any person obtaining a
> > + * copy of this software and associated documentation files (the
> > "Software"),
> > + * to deal in the Software without restriction, including without
> > limitation
> > + * the rights to use, copy, modify, merge, publish, distribute, sublicense,
> > + * and/or sell copies of the Software, and to permit persons to whom the
> > + * Software is furnished to do so, subject to the following conditions:
> > + *
> > + * The above copyright notice and this permission notice (including the
> > next
> > + * paragraph) shall be included in all copies or substantial portions of
> > the
> > + * Software.
> > + *
> > + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
> > OR
> > + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
> > + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
> > + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
> > OTHER
> > + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
> > + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
> > DEALINGS
> > + * IN THE SOFTWARE.
> > + */
> > +
> > +#include <unistd.h>
> > +#include <fcntl.h>
> > +
> > +#include "igt.h"
> > +#include "igt_kernel_taint.h"
> > +#include "igt_sysfs.h"
> > +
> > +#define BIT(x) (1ul << (x))
> > +
> > +static const struct kernel_taint {
> > +       const char *msg;
> > +       unsigned int flags;
> > +} taints[] = {
> > +       { "Non-GPL module loaded" },
> > +       { "Forced module load" },
> > +       { "Unsafe SMP processor" },
> > +       { "Forced module unload" },
> > +       { "Machine Check Exception", TAINT_WARN },
> > +       { "Bad page detected", TAINT_ERROR },
> > +       { "Tainted by user request", TAINT_WARN },
> 
> Since unsafe modparams generate these and we are still using them
> extensively, we should probably ignore this one.
> 
> > +       { "System is on fire", TAINT_ERROR },
> > +       { "ACPI DSDT has been overridden by user" },
> > +       { "OOPS", TAINT_ERROR },
> > +       { "Staging driver loaded; are you mad?" },
> > +       { "Severe firmware bug workaround active", TAINT_WARN },
> > +       { "Out-of-tree module loaded" },
> > +       { "Unsigned module loaded" },
> > +       { "Soft-lockup detected", TAINT_WARN },
> > +       { "Kernel has been live patched" },
> > +};
> > +
> > +unsigned long igt_read_kernel_taint(void)
> 
> One thing I haven't checked is whether we can clear the kernel taints.
> At the moment, once we see an oops, we never report a second test
> generating another oops.
> -Chris

I guess that clearing kernel taints is not needed when you hit oops - you
probably should stop executing tests and reboot the machine, right?

Radek