[PATCH] sim: cgen: namespace custom trace functions

[PATCH] sim: cgen: namespace custom trace functions

[Mike Frysinger]

[-- Attachment #1.1: Type: text/plain, Size: 2563 bytes --]

The cgen code declares some macros/funcs using the trace_xxx prefix, but
the code isn't generic and only works with cgen targets.  This is blocking
the creation of new common trace functions.

Let's blindly add cgen_xxx prefixes to all these symbols.  Some already
use this convention to avoid conflicts, so it makes sense to align them.
In the future we might want to move some to the common trace core, but
one thing at a time.

I'm attaching the patch compressed because it's like 2MB in size.  It's
all pretty mechanical -- the large files were updated using `sed`.

Committed.
---
 sim/common/ChangeLog          |   41 +
 sim/common/cgen-run.c         |    2 +-
 sim/common/cgen-trace.c       |   21 +-
 sim/common/cgen-trace.h       |   39 +-
 sim/common/genmloop.sh        |    8 +-
 sim/cris/ChangeLog            |   10 +
 sim/cris/decodev10.c          |  226 ++--
 sim/cris/decodev32.c          |  216 ++--
 sim/cris/mloop.in             |    6 +-
 sim/cris/semcrisv10f-switch.c | 2466 ++++++++++++++++++------------------
 sim/cris/semcrisv32f-switch.c | 2476 ++++++++++++++++++------------------
 sim/frv/ChangeLog             |    7 +
 sim/frv/decode.c              |  546 ++++----
 sim/frv/mloop.in              |    6 +-
 sim/frv/sem.c                 | 2770 ++++++++++++++++++++---------------------
 sim/iq2000/ChangeLog          |    8 +
 sim/iq2000/decode.c           |   60 +-
 sim/iq2000/mloop.in           |    6 +-
 sim/iq2000/sem-switch.c       |  176 +--
 sim/iq2000/sem.c              |  174 +--
 sim/lm32/ChangeLog            |    8 +
 sim/lm32/decode.c             |   46 +-
 sim/lm32/mloop.in             |    6 +-
 sim/lm32/sem-switch.c         |  132 +-
 sim/lm32/sem.c                |  130 +-
 sim/m32r/ChangeLog            |   12 +
 sim/m32r/decode.c             |  118 +-
 sim/m32r/decode2.c            |  144 +--
 sim/m32r/decodex.c            |  144 +--
 sim/m32r/mloop.in             |    6 +-
 sim/m32r/sem-switch.c         |  254 ++--
 sim/m32r/sem.c                |  252 ++--
 sim/m32r/sem2-switch.c        |  522 ++++----
 sim/m32r/semx-switch.c        |  508 ++++----
 sim/sh64/ChangeLog            |   16 +
 sim/sh64/decode-compact.c     |  274 ++--
 sim/sh64/decode-media.c       |  170 +--
 sim/sh64/mloop-compact.c      |   14 +-
 sim/sh64/mloop-media.c        |   14 +-
 sim/sh64/sem-compact-switch.c |  600 ++++-----
 sim/sh64/sem-compact.c        |  598 ++++-----
 sim/sh64/sem-media-switch.c   |  476 +++----
 sim/sh64/sem-media.c          |  474 +++----
 43 files changed, 7143 insertions(+), 7039 deletions(-)

[-- Attachment #1.2: 0001-sim-cgen-namespace-custom-trace-functions.patch.bz2 --]
[-- Type: application/x-bzip2, Size: 67060 bytes --]

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

Re: [PATCH] sim: cgen: namespace custom trace functions

[Doug Evans]

On Fri, Jun 12, 2015 at 2:11 AM, Mike Frysinger <vapier@gentoo.org> wrote:
> The cgen code declares some macros/funcs using the trace_xxx prefix, but
> the code isn't generic and only works with cgen targets.  This is blocking
> the creation of new common trace functions.
>
> Let's blindly add cgen_xxx prefixes to all these symbols.  Some already
> use this convention to avoid conflicts, so it makes sense to align them.
> In the future we might want to move some to the common trace core, but
> one thing at a time.
>
> I'm attaching the patch compressed because it's like 2MB in size.  It's
> all pretty mechanical -- the large files were updated using `sed`.
>
> Committed.

Heads up, a lot of these are generated files.
I'm happy to update the generator and regenerate these.

Re: [PATCH] sim: cgen: namespace custom trace functions

[Mike Frysinger]

[-- Attachment #1: Type: text/plain, Size: 1052 bytes --]

On 14 Jun 2015 06:49, Doug Evans wrote:
> On Fri, Jun 12, 2015 at 2:11 AM, Mike Frysinger wrote:
> > The cgen code declares some macros/funcs using the trace_xxx prefix, but
> > the code isn't generic and only works with cgen targets.  This is blocking
> > the creation of new common trace functions.
> >
> > Let's blindly add cgen_xxx prefixes to all these symbols.  Some already
> > use this convention to avoid conflicts, so it makes sense to align them.
> > In the future we might want to move some to the common trace core, but
> > one thing at a time.
> >
> > I'm attaching the patch compressed because it's like 2MB in size.  It's
> > all pretty mechanical -- the large files were updated using `sed`.
> >
> > Committed.
> 
> Heads up, a lot of these are generated files.
> I'm happy to update the generator and regenerate these.

sorry about that.  i'm not terribly familiar with cgen, and the heavy use of 
scheme makes it unlikely to improve the situation :/.  looks like you posted
a patch though at this point.
-mike

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

Re: [PATCH] sim: cgen: namespace custom trace functions

[Doug Evans]

On Sun, Jun 21, 2015 at 1:40 PM, Mike Frysinger <vapier@gentoo.org> wrote:
> On 14 Jun 2015 06:49, Doug Evans wrote:
>> On Fri, Jun 12, 2015 at 2:11 AM, Mike Frysinger wrote:
>> > The cgen code declares some macros/funcs using the trace_xxx prefix, but
>> > the code isn't generic and only works with cgen targets.  This is blocking
>> > the creation of new common trace functions.
>> >
>> > Let's blindly add cgen_xxx prefixes to all these symbols.  Some already
>> > use this convention to avoid conflicts, so it makes sense to align them.
>> > In the future we might want to move some to the common trace core, but
>> > one thing at a time.
>> >
>> > I'm attaching the patch compressed because it's like 2MB in size.  It's
>> > all pretty mechanical -- the large files were updated using `sed`.
>> >
>> > Committed.
>>
>> Heads up, a lot of these are generated files.
>> I'm happy to update the generator and regenerate these.
>
> sorry about that.  i'm not terribly familiar with cgen, and the heavy use of
> scheme makes it unlikely to improve the situation :/.  looks like you posted
> a patch though at this point.

Feel free to ask for help.
Plus for simple renamings like this it's not that hard.

[If I can keep my lunch down from the Perl that's been
dropped on gdb, I just can't feel bad for having asked people
to learn some Scheme. :-)]