Patch to fix PR83712

2018-03-13 Thread Vladimir Makarov
  This is one more try to fix PR83712: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83712   The patch was successfully bootstrapped and tested on x86-64, i686, and ppc64.   Committed as rev. 258504. Index: ChangeLog === ---

Re: patch to fix PR83712

2018-03-10 Thread Jeff Law
On 03/10/2018 09:40 AM, Vladimir Makarov wrote: > A few people reported that the patch broke i686.  I am going to work on > the patch more.  Meanwhile I've reverted the patch. Just a note, none of my other builds failed. Though i686 probably stresses the class-likely-spilled bits than any other.

Re: patch to fix PR83712

2018-03-10 Thread Vladimir Makarov
A few people reported that the patch broke i686.  I am going to work on the patch more.  Meanwhile I've reverted the patch. On 03/09/2018 11:16 AM, Vladimir Makarov wrote: The following patch fixes https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83712 It is another "cannot find a spill reg for

Re: patch to fix PR83712

2018-03-10 Thread H.J. Lu
On Fri, Mar 9, 2018 at 8:16 AM, Vladimir Makarov wrote: > The following patch fixes > > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83712 > > It is another "cannot find a spill reg for reload" problem. LRA has already > a code splitting hard reg live ranges to avoid such

Re: patch to fix PR83712

2018-03-09 Thread Jeff Law
On 03/09/2018 09:16 AM, Vladimir Makarov wrote: > The following patch fixes > > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83712 > > It is another "cannot find a spill reg for reload" problem.  LRA has > already a code splitting hard reg live ranges to avoid such problem.  > This code is in

patch to fix PR83712

2018-03-09 Thread Vladimir Makarov
The following patch fixes https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83712 It is another "cannot find a spill reg for reload" problem.  LRA has already a code splitting hard reg live ranges to avoid such problem.  This code is in LRA inheritance pass.  Unfortunately, the code does