Howto disable Quartus infering M4Ks??

I have some problems with a design I am porting from Xilinx to Altera. The fitter dies with a message about the design not fitting into the device.Further investiogation shows that Quartus tries to move a lot of small shiftregisters (32-bit x 4) into M4Ks, which is a not the best use of my embedded memories...

Can anyone explain to me why this happens? I am using 75% of the FPGA right now and should be able to get the small shift registers into LEs. Is there any way to let Quartus infer none or _some_ of the shift registers into M4Ks and use LEs for the rest??

(I am using Quartus Webpack version 8.0, build 231 07/10/2008)

thanks for any help

Reply to
tgau3qk4
Loading thread data ...

Try running quartus with no constraints other than the device family.

-- Mike Treseler

Reply to
Mike Treseler

In the Quartus UI use the Assignments menu as folows : Assiginmnets-

in the dialog t a set of Maximum Setting entries which look like Maximm Number of M512 memory blocks. Set it to 0 or any non zero number other than -1 (which is default).

Hope tis helps, Subroto Datta Altera Corp.

Reply to
Subroto Datta

M512's or M4k's? Or to make it use M512's instead of M4k's?

Reply to
cs_posting

There are separate settings for both, and -1 means unlimited. But I don't think this will help.

The OP said

I expect that quartus was out of LEs on the specified device, and the excess shifters did not match the available blockram.

The quickest way to test this theory is to spec a larger device.

If this fails I would either break up the design, revert to version 7.2, or open a case with Altera.

-- Mike Treseler

Reply to
Mike Treseler

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.