Benjamin Ylvisaker wrote:
> On Oct 26, 2009, at 11:34 AM, Misha Aizatulin wrote:
>
>> It would be really nice if I could tell CIL not to remove some of the
>> definitions, for instance by using an attribute. Any comments on that?
>
> Do you mean something other than the following?
>
> #pragma ci
On Oct 26, 2009, at 11:34 AM, Misha Aizatulin wrote:
> It would be really nice if I could tell CIL not to remove some of
> the definitions, for instance by using an attribute. Any comments on
> that?
Do you mean something other than the following?
#pragma cilnoremove("type sarray_t", "sarray
> Here is how we solve the very same problem for CPC:
>
> # Makefile
> .cpc.cpi:
> gcc -E -x c $(CFLAGS) -include cpc/cpc_runtime.h \
> -o $@ $<
>
> And then, process the .cpi files with CIL.
I realised that this is not quite the right thing for me, because I
run CIL on a big and
Gabriel Kerneis wrote:
> On Thu, Oct 22, 2009 at 08:21:59AM +0100, Misha Aizatulin wrote:
>> Does anyone have a suggestion for dealing with this? Is there a way to
>> add something to the file before the first preprocessing pass?
>
> Here is how we solve the very same problem for CPC:
>
> # Mak
On Thu, Oct 22, 2009 at 08:21:59AM +0100, Misha Aizatulin wrote:
> Does anyone have a suggestion for dealing with this? Is there a way to
> add something to the file before the first preprocessing pass?
Here is how we solve the very same problem for CPC:
# Makefile
.cpc.cpi:
gcc -E -x c
hi,
I'm facing the following problem: I'm inserting some function calls
into the code using CIL. The new functions are defined in a separate
header, say funs.h. The naive idea would be to add
GText "#include \"funs.h\""
to the list of globals, however this doesn't work: if funs.h itself
#inclu
