Re: Setting precision for a PSImode type

2012-04-11 Thread Peter Bigot
On Wed, Apr 11, 2012 at 11:55 AM, Bernd Schmidt wrote: > On 04/11/2012 06:53 PM, Peter Bigot wrote: >> On Mon, Mar 5, 2012 at 10:38 AM, Bernd Schmidt >> wrote: >>> On 03/05/2012 05:24 PM, Peter Bigot wrote: And is there any reason (other than it doesn't seem to have been done before) t

Re: Setting precision for a PSImode type

2012-04-11 Thread Bernd Schmidt
On 04/11/2012 06:53 PM, Peter Bigot wrote: > On Mon, Mar 5, 2012 at 10:38 AM, Bernd Schmidt > wrote: >> On 03/05/2012 05:24 PM, Peter Bigot wrote: >>> And is there any reason (other than it doesn't seem to have been done >>> before) to believe PSImode is the wrong way to support a >>> general-pur

Re: Setting precision for a PSImode type

2012-04-11 Thread Peter Bigot
On Mon, Mar 5, 2012 at 10:38 AM, Bernd Schmidt wrote: > On 03/05/2012 05:24 PM, Peter Bigot wrote: >> And is there any reason (other than it doesn't seem to have been done >> before) to believe PSImode is the wrong way to support a >> general-purpose 20-bit integral type in gcc? > > If you're usin

Re: Setting precision for a PSImode type

2012-03-26 Thread Peter Bigot
On Mon, Mar 5, 2012 at 10:38 AM, Bernd Schmidt wrote: > On 03/05/2012 05:24 PM, Peter Bigot wrote: >> And is there any reason (other than it doesn't seem to have been done >> before) to believe PSImode is the wrong way to support a >> general-purpose 20-bit integral type in gcc? > > If you're usin

Re: Setting precision for a PSImode type

2012-03-05 Thread Georg-Johann Lay
Bernd Schmidt wrote: > On 03/05/2012 05:24 PM, Peter Bigot wrote: >> And is there any reason (other than it doesn't seem to have been done >> before) to believe PSImode is the wrong way to support a >> general-purpose 20-bit integral type in gcc? > > If you're using 4.7.0, it should be possible to

Re: Setting precision for a PSImode type

2012-03-05 Thread Peter Bigot
On Mon, Mar 5, 2012 at 10:38 AM, Bernd Schmidt wrote: > On 03/05/2012 05:24 PM, Peter Bigot wrote: >> And is there any reason (other than it doesn't seem to have been done >> before) to believe PSImode is the wrong way to support a >> general-purpose 20-bit integral type in gcc? > > If you're usin

Re: Setting precision for a PSImode type

2012-03-05 Thread Bernd Schmidt
On 03/05/2012 05:24 PM, Peter Bigot wrote: > And is there any reason (other than it doesn't seem to have been done > before) to believe PSImode is the wrong way to support a > general-purpose 20-bit integral type in gcc? If you're using 4.7.0, it should be possible to use FRACTIONAL_INT_MODE and g

Setting precision for a PSImode type

2012-03-05 Thread Peter Bigot
I'm enhancing the out-of-tree msp430 back end to support the 20-bit extended registers of the 430X CPU. I'd like to do this using a native type rather than special-casing its use in addresses. It seems the best approach is to use PSImode, since the values are 20 bits when in a register, and 32 bi