Re: From SSA back to GIMPLE.

2 msg	How to avoid common subexpression elimination u...
1 msg	Fwd: [Announcing OpenMP 3.0 draft for public co...
4 msg	builtin_frame_address for stack pointer
2 msg	GCC 4.1.1 unwind support for arm-none-linux-gnu...
1 msg	gcc-4.1-20071022 is now available
19 msg	What is a regression?
2 msg	modified x86 ABI
1 msg	I need your fast reply!

Re: From SSA back to GIMPLE.
\ J.C. Pizarro (22 Oct 2007)
. \ Paolo Bonzini (22 Oct 2007)
. . \ skaller (22 Oct 2007)
. . . \ Dave Korn (22 Oct 2007)
. . . . \ Jamie Lokier (22 Oct 2007)
. . \ J.C. Pizarro (22 Oct 2007)
. \ Zdenek Dvorak (22 Oct 2007)
. . \ J.C. Pizarro (22 Oct 2007)
. \ David Edelsohn (22 Oct 2007)
. . \ J.C. Pizarro (22 Oct 2007)
. . . \ Joe Buck (22 Oct 2007)
. . . \ Diego Novillo (25 Oct 2007)
. \ David Edelsohn (22 Oct 2007)
. . \ J.C. Pizarro (22 Oct 2007)
. . . \ Jamie Lokier (22 Oct 2007)

2 msg	Safe optimization flags for x86 processors.
5 msg	From SSA back to GIMPLE
2 msg	Bug in C++ FE?
208 msg	Optimization of conditional access to globals: ...
3 msg	Fw[1]:
1 msg	(libgcc) what does _Unwind_Resume do ?
4 msg	one question: tree-ssa vs no tree-ssa? no such ...
3 msg	Optimizing for code size: new PR about issues w...
1 msg	gcc-4.3-20071019 is now available
4 msg	SPARC Ada compilation hangs
31 msg	Removal of pre-ISO C++ items from include/backw...

Subject:	Re: From SSA back to GIMPLE.
Group:	Gcc
From:	Diego Novillo
Date:	25 Oct 2007

J.C. Pizarro wrote:

> why is hard to optimize unrolling loop, inlining code, instructions
> scheduling, etc because of the SSA's presence?

None of these things are particluarly hard with SSA. I'm not sure I
understand what you are trying to get at.

> Don't forget, "Premature optimization is the root of all evil".

Indeed. That is one of the reasons why GCC has been restructured this
way. And the job is still not done.