Re: LTO streaming of TARGET_OPTIMIZE_NODE

2015-01-12 Thread Jan Hubicka
> On 09 Jan 12:45, Jakub Jelinek wrote: > > --- gcc/cgraphunit.c.jj 2015-01-09 12:01:33.0 +0100 > > +++ gcc/cgraphunit.c2015-01-09 12:22:27.742692667 +0100 > > @@ -2108,11 +2108,14 @@ ipa_passes (void) > >if (g->have_offload) > > { > > section_name_prefix = OFF

Re: LTO streaming of TARGET_OPTIMIZE_NODE

2015-01-11 Thread Ilya Verbin
On 09 Jan 12:45, Jakub Jelinek wrote: > --- gcc/cgraphunit.c.jj 2015-01-09 12:01:33.0 +0100 > +++ gcc/cgraphunit.c 2015-01-09 12:22:27.742692667 +0100 > @@ -2108,11 +2108,14 @@ ipa_passes (void) >if (g->have_offload) > { > section_name_prefix = OFFLOAD_SECTION_N

Re: LTO streaming of TARGET_OPTIMIZE_NODE

2015-01-11 Thread Jan Hubicka
> On Fri, Jan 09, 2015 at 12:07:26PM +0100, Thomas Schwinge wrote: > > On Thu, 8 Jan 2015 15:11:49 +0100, Jakub Jelinek wrote: > > > On Thu, Nov 20, 2014 at 01:27:08PM +0100, Bernd Schmidt wrote: > > > > On 11/13/2014 05:06 AM, Jan Hubicka wrote: > > > > >this patch adds infrastructure for proper

Re: LTO streaming of TARGET_OPTIMIZE_NODE

2015-01-09 Thread Richard Biener
On Fri, 9 Jan 2015, Jakub Jelinek wrote: > On Fri, Jan 09, 2015 at 12:07:26PM +0100, Thomas Schwinge wrote: > > On Thu, 8 Jan 2015 15:11:49 +0100, Jakub Jelinek wrote: > > > On Thu, Nov 20, 2014 at 01:27:08PM +0100, Bernd Schmidt wrote: > > > > On 11/13/2014 05:06 AM, Jan Hubicka wrote: > > > > >

Re: LTO streaming of TARGET_OPTIMIZE_NODE

2015-01-09 Thread Jakub Jelinek
On Fri, Jan 09, 2015 at 12:07:26PM +0100, Thomas Schwinge wrote: > On Thu, 8 Jan 2015 15:11:49 +0100, Jakub Jelinek wrote: > > On Thu, Nov 20, 2014 at 01:27:08PM +0100, Bernd Schmidt wrote: > > > On 11/13/2014 05:06 AM, Jan Hubicka wrote: > > > >this patch adds infrastructure for proper streaming

Re: LTO streaming of TARGET_OPTIMIZE_NODE

2015-01-09 Thread Thomas Schwinge
Hi! On Thu, 8 Jan 2015 15:11:49 +0100, Jakub Jelinek wrote: > On Thu, Nov 20, 2014 at 01:27:08PM +0100, Bernd Schmidt wrote: > > On 11/13/2014 05:06 AM, Jan Hubicka wrote: > > >this patch adds infrastructure for proper streaming and merging of > > >TREE_TARGET_OPTION. > > > > This breaks the off

Re: LTO streaming of TARGET_OPTIMIZE_NODE

2015-01-08 Thread Jakub Jelinek
On Thu, Nov 20, 2014 at 01:27:08PM +0100, Bernd Schmidt wrote: > On 11/13/2014 05:06 AM, Jan Hubicka wrote: > >this patch adds infrastructure for proper streaming and merging of > >TREE_TARGET_OPTION. > > This breaks the offloading path via LTO since it introduces an > incompatibility in LTO forma

Re: LTO streaming of TARGET_OPTIMIZE_NODE

2014-11-20 Thread Jan Hubicka
> On 11/20/2014 02:20 PM, Richard Biener wrote: > >On Thu, 20 Nov 2014, Bernd Schmidt wrote: > > > >>On 11/13/2014 05:06 AM, Jan Hubicka wrote: > >>>this patch adds infrastructure for proper streaming and merging of > >>>TREE_TARGET_OPTION. > >> > >>This breaks the offloading path via LTO since it

Re: LTO streaming of TARGET_OPTIMIZE_NODE

2014-11-20 Thread Bernd Schmidt
On 11/20/2014 02:20 PM, Richard Biener wrote: On Thu, 20 Nov 2014, Bernd Schmidt wrote: On 11/13/2014 05:06 AM, Jan Hubicka wrote: this patch adds infrastructure for proper streaming and merging of TREE_TARGET_OPTION. This breaks the offloading path via LTO since it introduces an incompatibi

Re: LTO streaming of TARGET_OPTIMIZE_NODE

2014-11-20 Thread Richard Biener
On Thu, 20 Nov 2014, Bernd Schmidt wrote: > On 11/13/2014 05:06 AM, Jan Hubicka wrote: > > this patch adds infrastructure for proper streaming and merging of > > TREE_TARGET_OPTION. > > This breaks the offloading path via LTO since it introduces an incompatibility > in LTO format between host and

Re: LTO streaming of TARGET_OPTIMIZE_NODE

2014-11-20 Thread Bernd Schmidt
On 11/13/2014 05:06 AM, Jan Hubicka wrote: this patch adds infrastructure for proper streaming and merging of TREE_TARGET_OPTION. This breaks the offloading path via LTO since it introduces an incompatibility in LTO format between host and offload machine. A very quick patch to fix it is bel

Re: LTO streaming of TARGET_OPTIMIZE_NODE

2014-11-14 Thread Richard Biener
On Fri, 14 Nov 2014, Jan Hubicka wrote: > Hi, > here is upated version with bitfields and also tested on PPC64-linux/aix. > I hacked configury to use system awk instead of gawk, so the changes are > hopefully safe. > > OK? Ok. Thanks, Richard. > Honza > > * optc-save-gen.awk: Output cl

Re: LTO streaming of TARGET_OPTIMIZE_NODE

2014-11-13 Thread Jan Hubicka
Hi, here is upated version with bitfields and also tested on PPC64-linux/aix. I hacked configury to use system awk instead of gawk, so the changes are hopefully safe. OK? Honza * optc-save-gen.awk: Output cl_target_option_eq, cl_target_option_hash, cl_target_option_stream_out,

Re: LTO streaming of TARGET_OPTIMIZE_NODE

2014-11-13 Thread Jan Hubicka
> > > > Incrementally I would like to handle optimizations nodes same way and change > > streaming format to be a set of assignments field_name=value. > > This way we could behave sanely when some field is introduced or removed by > > either defaulting it or ignoring it. > > I'm not sure that ide

Re: LTO streaming of TARGET_OPTIMIZE_NODE

2014-11-13 Thread Richard Biener
On Thu, 13 Nov 2014, Jan Hubicka wrote: > Hi, > this patch adds infrastructure for proper streaming and merging of > TREE_TARGET_OPTION. The catch is that TREE_TARGET_OPTION is autogenerated > structure. For x86_64 it looks as follows: > > /* Structure to save/restore selected target specific o