Subzero: Dump register numbers as signed quantities.

Subzero: Dump register numbers as signed quantities.

Undo an overly aggression application of unsigned to register numbers in 8aa396610b7baf728631a43ea16ad3d13e38397a (https://codereview.chromium.org/1676123002).

Now, instead of -verbose=regalloc output like:

++++++ Unhandled:
R=jpp@chromium.org, kschimpf@google.com, 4294967295  V=%__4  Range=[2:7), 4294967295  V=%__5  Range=[7:8), 4294967295  V=%__6  Range=[9:11), -1  V=%__4  Range=[2:7), -1  V=%__5  Range=[7:8), -1  V=%__6  Range=[9:11)

we have the originally intended:

++++++ Unhandled:

BUG= none

Committed: https://gerrit.chromium.org/gerrit/gitweb?p=native_client/pnacl-subzero.git;a=commit;h=11756b510456b215628124e7c13750f60b745f86

[Jim Stichnoth, 2016-04-06 15:48:03]

Description was changed from

==========
Subzero: Dump register numbers as signed quantities.

BUG= none
==========

to

==========
Subzero: Dump register numbers as signed quantities.

Undo an overly aggression application of unsigned to register numbers in
8aa396610b7baf728631a43ea16ad3d13e38397a
(https://codereview.chromium.org/1676123002).

Now, instead of -verbose=regalloc output like:

++++++ Unhandled:
R=4294967295  V=%__4  Range=[2:7)
R=4294967295  V=%__5  Range=[7:8)
R=4294967295  V=%__6  Range=[9:11)

we have the originally intended:

++++++ Unhandled:
R=-1  V=%__4  Range=[2:7)
R=-1  V=%__5  Range=[7:8)
R=-1  V=%__6  Range=[9:11)

BUG= none
==========

[Jim Stichnoth, 2016-04-06 15:48:03]

stichnot@chromium.org changed reviewers:
+ eholk@chromium.org, jpp@chromium.org, kschimpf@google.com, sehr@chromium.org

[Karl, 2016-04-06 15:52:18]

lgtm

[John, 2016-04-06 16:36:14]

lgtm

https://codereview.chromium.org/1868543002/diff/1/src/IceRegAlloc.cpp
File src/IceRegAlloc.cpp (right):

https://codereview.chromium.org/1868543002/diff/1/src/IceRegAlloc.cpp#newcode74
src/IceRegAlloc.cpp:74: snprintf(buf, llvm::array_lengthof(buf), "%2d",
What about printing None instead of -1?

[Jim Stichnoth, 2016-04-06 17:04:37]

https://codereview.chromium.org/1868543002/diff/1/src/IceRegAlloc.cpp
File src/IceRegAlloc.cpp (right):

https://codereview.chromium.org/1868543002/diff/1/src/IceRegAlloc.cpp#newcode74
src/IceRegAlloc.cpp:74: snprintf(buf, llvm::array_lengthof(buf), "%2d",
On 2016/04/06 16:36:14, John wrote:
> What about printing None instead of -1?

I tried that, but then I realized that the whole snprintf plumbing is to try to
make the output line up nicely in columns.  Practically all register numbers
(including -1) fit within 2 chars.

I could do something else, maybe "xx" or "--"?  What do you think?  One think I
noticed is that "R=-1" and "R= 1" are a bit hard to distinguish in the output.

[John, 2016-04-06 17:28:43]

On 2016/04/06 17:04:37, stichnot wrote:
> https://codereview.chromium.org/1868543002/diff/1/src/IceRegAlloc.cpp
> File src/IceRegAlloc.cpp (right):
> 
>
https://codereview.chromium.org/1868543002/diff/1/src/IceRegAlloc.cpp#newcode74
> src/IceRegAlloc.cpp:74: snprintf(buf, llvm::array_lengthof(buf), "%2d",
> On 2016/04/06 16:36:14, John wrote:
> > What about printing None instead of -1?
> 
> I tried that, but then I realized that the whole snprintf plumbing is to try
to
> make the output line up nicely in columns.  Practically all register numbers
> (including -1) fit within 2 chars.
> 
> I could do something else, maybe "xx" or "--"?  What do you think?  One think
I
> noticed is that "R=-1" and "R= 1" are a bit hard to distinguish in the output.

NA (not assigned)?

[Jim Stichnoth, 2016-04-06 18:41:04]

On 2016/04/06 17:28:43, John wrote:
> On 2016/04/06 17:04:37, stichnot wrote:
> > https://codereview.chromium.org/1868543002/diff/1/src/IceRegAlloc.cpp
> > File src/IceRegAlloc.cpp (right):
> > 
> >
>
https://codereview.chromium.org/1868543002/diff/1/src/IceRegAlloc.cpp#newcode74
> > src/IceRegAlloc.cpp:74: snprintf(buf, llvm::array_lengthof(buf), "%2d",
> > On 2016/04/06 16:36:14, John wrote:
> > > What about printing None instead of -1?
> > 
> > I tried that, but then I realized that the whole snprintf plumbing is to try
> to
> > make the output line up nicely in columns.  Practically all register numbers
> > (including -1) fit within 2 chars.
> > 
> > I could do something else, maybe "xx" or "--"?  What do you think?  One
think
> I
> > noticed is that "R=-1" and "R= 1" are a bit hard to distinguish in the
output.
> 
> NA (not assigned)?

I tried a few things and like NA best.

Also, I realized I can get rid of the icky snprintf stuff.

[Jim Stichnoth, 2016-04-06 19:20:28]

Description was changed from

==========
Subzero: Dump register numbers as signed quantities.

Undo an overly aggression application of unsigned to register numbers in
8aa396610b7baf728631a43ea16ad3d13e38397a
(https://codereview.chromium.org/1676123002).

Now, instead of -verbose=regalloc output like:

++++++ Unhandled:
R=4294967295  V=%__4  Range=[2:7)
R=4294967295  V=%__5  Range=[7:8)
R=4294967295  V=%__6  Range=[9:11)

we have the originally intended:

++++++ Unhandled:
R=-1  V=%__4  Range=[2:7)
R=-1  V=%__5  Range=[7:8)
R=-1  V=%__6  Range=[9:11)

BUG= none
==========

to

==========
Subzero: Dump register numbers as signed quantities.

Undo an overly aggression application of unsigned to register numbers in
8aa396610b7baf728631a43ea16ad3d13e38397a
(https://codereview.chromium.org/1676123002).

Now, instead of -verbose=regalloc output like:

++++++ Unhandled:
R=jpp@chromium.org, kschimpf@google.com, 4294967295  V=%__4  Range=[2:7),
4294967295  V=%__5  Range=[7:8), 4294967295  V=%__6  Range=[9:11), -1  V=%__4 
Range=[2:7), -1  V=%__5  Range=[7:8), -1  V=%__6  Range=[9:11)

we have the originally intended:

++++++ Unhandled:

BUG= none

Committed:
https://gerrit.chromium.org/gerrit/gitweb?p=native_client/pnacl-subzero.git;a...
==========

[Jim Stichnoth, 2016-04-06 19:20:37]

Committed patchset #2 (id:20001) manually as
11756b510456b215628124e7c13750f60b745f86 (presubmit successful).