Re: [PATCH] Fix middle-end/64061, ICE in gen_rtx_SUBREG

2014-12-01 Thread Andrew Pinski
On Mon, Dec 1, 2014 at 10:26 AM, Jeff Law wrote: > On 11/26/14 15:00, Andrew Pinski wrote: >> >> Hi, >>The problem here is lra_substitute_pseudo calls gen_rtx_SUBREG with >> a VOIDmode (const_int) argument but really it should not be calling >> gen_rtx_SUBREG directly instead it should be usin

Re: [PATCH] Fix middle-end/64061, ICE in gen_rtx_SUBREG

2014-12-01 Thread H.J. Lu
On Wed, Nov 26, 2014 at 2:01 PM, Andrew Pinski wrote: > On Wed, Nov 26, 2014 at 2:00 PM, Andrew Pinski wrote: >> Hi, >> The problem here is lra_substitute_pseudo calls gen_rtx_SUBREG with >> a VOIDmode (const_int) argument but really it should not be calling >> gen_rtx_SUBREG directly instead i

Re: [PATCH] Fix middle-end/64061, ICE in gen_rtx_SUBREG

2014-12-01 Thread Jeff Law
On 11/26/14 15:00, Andrew Pinski wrote: Hi, The problem here is lra_substitute_pseudo calls gen_rtx_SUBREG with a VOIDmode (const_int) argument but really it should not be calling gen_rtx_SUBREG directly instead it should be using gen_lowpart_if_possible. This patch fixes that and adds a test

Re: [PATCH] Fix middle-end/64061, ICE in gen_rtx_SUBREG

2014-11-26 Thread Andrew Pinski
On Wed, Nov 26, 2014 at 2:00 PM, Andrew Pinski wrote: > Hi, > The problem here is lra_substitute_pseudo calls gen_rtx_SUBREG with > a VOIDmode (const_int) argument but really it should not be calling > gen_rtx_SUBREG directly instead it should be using > gen_lowpart_if_possible. This patch fixe

[PATCH] Fix middle-end/64061, ICE in gen_rtx_SUBREG

2014-11-26 Thread Andrew Pinski
Hi, The problem here is lra_substitute_pseudo calls gen_rtx_SUBREG with a VOIDmode (const_int) argument but really it should not be calling gen_rtx_SUBREG directly instead it should be using gen_lowpart_if_possible. This patch fixes that and adds a testcase that had happened on x86_64. OK? I bo