This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: rfc: another switch optimization idea


On Tue, Mar 26, 2013 at 10:31 AM, Richard Biener wrote:
> On Mon, Mar 25, 2013 at 10:23 PM, Dinar Temirbulatov wrote:
>> The error in 252.eon was due to incorrect setup. Also "if (count >
>> 3*PARAM_VALUE (PARAM_SWITCH_JUMP_TABLES_BB_OPS_LIMIT))" does not look
>> correct, and probably it is better to move this code in the earlier
>> stage just before the gimple expand and keep preference expand state
>> (jump-tables or not) for every switch statement to avoid dealing with
>> the RTL altogether.
>
> Moving switch "expansion" to GIMPLE is an idea that is around since
> quite some time.  Basically you'd lower switches so that remaining
> switch statements directly map to jump-tables only.  Steven was working
> on this a bit and if I remember correctly 4.8 has some improvements
> here in the switch-conversion pass.

Right. I will move switch lowering to GIMPLE for GCC 4.9. Everything
not lowered will be expanded as a casesi or tablejump. New methods of
lowering switches can be added in the new GIMPLE lowering pass once
that's done.

Ciao!
Steven


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]