middle-end/3545: ICE on emit_move_insn

peter@frix.fri.utc.sk peter@frix.fri.utc.sk
Tue Jul 3 08:07:00 GMT 2001


>Number:         3545
>Category:       middle-end
>Synopsis:       ICE on emit_move_insn
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          ice-on-legal-code
>Submitter-Id:   net
>Arrival-Date:   Tue Jul 03 08:06:01 PDT 2001
>Closed-Date:
>Last-Modified:
>Originator:     Peter Skvarenina
>Release:        2.95.2, 2.95.3, 3.0 prerelease 20010523
>Organization:
>Environment:
DJGPP 2.03, i486DX4, Win95 SR1
>Description:
While trying to compile attached test.cpp that uses cislo.h with inlined asm
functions, i've encountered this problem:

gcc -c -O2 test.cpp ===>

cislo.h: In function 'int main()':
cislo.h:104: Internal compiler error in emit_move_insn, at expr.c:2724
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL: http://www.gnu.org/software/gcc/bugs.html > for instructions.

and RHIDE (DJGPP IDE) focused to the related asm block.

Problem arises when the non-default (my own) constructor and two copy
constructors are uncommented in the source.

Source is an excerpt of a collection of inlined asm functions to provide
32/16/16 fixed point math in the fastest way. I've cut off all the functions
expect some that provide you with the same results.

Three types of sources are included:
 - preprocessed correct - correct.cpp
 - preprocessed incorrect - invalid.cpp
 - and originals: test.cpp & cislo.h

These problems have been encountered in GCC 2.95.2, 2.95.3 (finals) and also 3.
20010523.

Reading specs from c:/program/djgpp/lib/gcc-lib/djgpp/3.0/specs
Configured with: ../gcc/configure i586-pc-msdosdjgpp --prefix=/dev/env/DJDIR --
able languages=c,c++,f77,objc --disable-nls --with-gxx-include-dir=\${prefix}/l
ang-cxx-v3
Thread model: single
gcc version 3.0 20010523 (prerelease)

GCC 3.0 (as opposed to 2.95.x) in larger project also dumps errors like
non-lvalue in unary `&', when constructors are inside comments, at the places,
where when uncommented, it generates those internal compiler errors.
(preprocessor output for this is in hlbka.cpp, using the same switches
as above, errors are related to 'PevnyBod' union, but the names of functions
are in the slovak)

I tried to provide 'templates' for compiler to be able to generate very fast
code that can be inlined directly into the code without any loss of time
caused by useless relocations of registers. If this is not working, how can
I reach that goal ? I hoped that it was compiler error, but maybe my coding
style or understanding of C++ is not very good.

>How-To-Repeat:

>Fix:

>Release-Note:
>Audit-Trail:
>Unformatted:
----gnatsweb-attachment----
Content-Type: application/octet-stream; name="cislo.zip"
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="cislo.zip"
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More information about the Gcc-bugs mailing list