On Tue, Apr 26, 2016 at 12:47 PM, Dylan Baker <baker.dyla...@gmail.com>
wrote:
> Quoting Jason Ekstrand (2016-04-25 21:39:19)
> > Previously, if an exception was encountered anywhere, nir_algebraic would
> > just die in a fire with no indication whatsoever as to where the actual
> bug
> > is. This commit makes it print out the particular search-and-replace
> > expression that is causing problems along with the exception. Also, it
> > will now report all of the errors it finds and then exit at the end like
> a
> > standard C compiler would do.
> >
> > Cc: Dylan Baker <baker.dyla...@gmail.com>
> > ---
> > src/compiler/nir/nir_algebraic.py | 17 ++++++++++++++++-
> > 1 file changed, 16 insertions(+), 1 deletion(-)
> >
> > diff --git a/src/compiler/nir/nir_algebraic.py
> b/src/compiler/nir/nir_algebraic.py
> > index 7697171..3f770b6 100644
> > --- a/src/compiler/nir/nir_algebraic.py
> > +++ b/src/compiler/nir/nir_algebraic.py
> > @@ -24,11 +24,13 @@
> > # Authors:
> > # Jason Ekstrand (ja...@jlekstrand.net)
> >
> > +from __future__ import print_function
> > import itertools
> > import struct
> > import sys
> > import mako.template
> > import re
> > +import traceback
> >
> > # Represents a set of variables, each with a unique id
> > class VarSet(object):
> > @@ -311,15 +313,28 @@ class AlgebraicPass(object):
> > self.xform_dict = {}
> > self.pass_name = pass_name
> >
> > + error = False
> > +
> > for xform in transforms:
> > if not isinstance(xform, SearchAndReplace):
> > - xform = SearchAndReplace(xform)
> > + try:
> > + xform = SearchAndReplace(xform)
> > + except:
> > + print("Failed to parse transformation:", file=sys.stderr)
> > + print(" " + str(xform), file=sys.stderr)
> > + traceback.print_exc(file=sys.stderr)
> > + print('', file=sys.stderr)
> > + error = True
> > + continue
> >
> > if xform.search.opcode not in self.xform_dict:
> > self.xform_dict[xform.search.opcode] = []
> >
> > self.xform_dict[xform.search.opcode].append(xform)
> >
> > + if error:
> > + raise RuntimeError("Error parsing transformations")
> > +
>
> I'm going to admit that I have a bias against using RuntimeError, but I
> think in the vein you're going for just calling "sys.exit(1)" here would
> be better, since you won't get a meaningless stack trace.
>
I'd rather use sys.exit(1) anyway. I just thought that might be "improper"
in python. Changed locally.
> > def render(self):
> > return _algebraic_pass_template.render(pass_name=self.pass_name,
> > xform_dict=self.xform_dict,
> > --
> > 2.5.0.400.gff86faf
> >
> Even with the one nit, either way:
> Reviewed-by: Dylan Baker <dy...@pnwbakers.com>
>
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
