What to do with new-ra for GCC 4.0

["Björn Haase" <Bjoern dot M dot Haase at web dot de>]

Steven wrote,
>Hi,
>
>Perhaps not so surprisingly, new-ra is completely broken on mainline
>at the moment.  Literally not a single non-empty file will compile
>with -fnew-ra.
>
>First of all there is a bug in ra-build.c:live_in() which has been
>there since the merge from the edge-vector-branch.  Then we hit an
>ICE on sibcalls in ra.c:reg_alloc(), which also must have been there
>since ages.  The patch below sort-of fixes these two issues.  But
>then we hit SUBREG problems, probably due to rth's stricter subreg
>patches.  At that point I gave up.
>
>So, any nontrivial code will ICE with -fnew-ra.  I hate to bring up
>painful issues, but I would like to propose we remove all of new-ra
>for GCC 4.0 as it is complete and utterly broken anyway, and given
>that it has been broken for months without anyone complaining, it's
>not like anyone will miss it.
>
>Thoughts?
>
>Gr.
>Steven


Hi,

concerning 'it's not like anyone will miss it' :

For the avr target, new-ra used to be *extremely* useful. There have been discussions about the -fnew-ra issue. Unfortunately not on this list but on some of the specialized lists for the avr targets, iirc. Maybe we should cross post summaries of our discussions also at this list.?

It is true that also for 3.4.3 the use of -fnew-ra used to crash the compiler sometimes. Depends on the source file. I never succeeded in figuring out what was the origin of the failures. If, however, your code compiles for the avr target with -fnew-ra, you will be able to observe a code size reduction of up to 10 per cent! This could be essential if you are short on memory!

Yours,

Björn

________________________________________________________________
Verschicken Sie romantische, coole und witzige Bilder per SMS!
Jetzt neu bei WEB.DE FreeMail: http://freemail.web.de/?mc=021193