How to reinterpret data in GIMPLE.

1 msg	Serious Bugs In Gcc Builds
1 msg	problems when move one pseudo reg to another (n...
5 msg	Issue with fixincludes (?) and `limits.h'
1 msg	gcc-4.1-20071203 is now available

How to reinterpret data in GIMPLE.
\ Sjodin, Jan (3 Dec 2007)
. \ Andrew Pinski (3 Dec 2007)

3 msg	[RFC] Introduce middle-end expressions on arrays
84 msg	Rant about ChangeLog entries and commit messages
1 msg	[RFC] Cleaning up latch blocks
1 msg	ADOPT A PUP OR KITTEN THIS XMAS.
2 msg	problems compiling a program with -m32
1 msg	gcc-4.3-20071130 is now available
11 msg	volatile and R/M/W operations
8 msg	GNAT SVN trunk on PowerPC and SPARC
10 msg	Re: Bug in builtins.def, the execve. don't use ...
9 msg	Describing commercial support on our website
2 msg	Bug in builtins.def, the execve.
1 msg	gcc-4.2-20071128 is now available
4 msg	Testsuite infrastructure for comparing outputs
18 msg	Function specific optimizations call for discus...
12 msg	gnat1 huge time

Subject:	How to reinterpret data in GIMPLE.
Group:	Gcc
From:	Sjodin, Jan
Date:	3 Dec 2007

I would like to reinterpret (not convert/cast) a 32-bit integer to a
32-bit float in GIMPLE. Is using a NOP_EXPR with the wanted type the
correct way of doing this?

The reinterpretation of a value is needed to optimize reads and writes
to unions. I modified the value numbering pass which worked fine, but
changing PRE to use a NOP_EXPR to change the type still resulted in a
conversion. It may be because of a bug somewhere else, but before doing
more work I would like to make sure that NOP_EXPR is the correct
operator.

Thanks,
Jan