Re: Fix profiledbootstrap with release checking

2015-01-16 Thread Markus Trippelsdorf
On 2015.01.16 at 12:03 +0100, Richard Biener wrote: > On Fri, Jan 16, 2015 at 9:34 AM, Markus Trippelsdorf > wrote: > > On 2015.01.16 at 00:15 +0100, Jan Hubicka wrote: > >> Hi, > >> can_remove_node_now_p assumes that the node in question has no direct > >> calls, this > >> however is not checked

Re: Fix profiledbootstrap with release checking

2015-01-16 Thread Richard Biener
On Fri, Jan 16, 2015 at 9:34 AM, Markus Trippelsdorf wrote: > On 2015.01.16 at 00:15 +0100, Jan Hubicka wrote: >> Hi, >> can_remove_node_now_p assumes that the node in question has no direct calls, >> this >> however is not checked in inline_call that leads to alias to be removed from >> comdat g

Re: Fix profiledbootstrap with release checking

2015-01-16 Thread Markus Trippelsdorf
On 2015.01.16 at 00:15 +0100, Jan Hubicka wrote: > Hi, > can_remove_node_now_p assumes that the node in question has no direct calls, > this > however is not checked in inline_call that leads to alias to be removed from > comdat group while it should not. > > Bootstrapped/regtested x86_64-linux,

Fix profiledbootstrap with release checking

2015-01-15 Thread Jan Hubicka
Hi, can_remove_node_now_p assumes that the node in question has no direct calls, this however is not checked in inline_call that leads to alias to be removed from comdat group while it should not. Bootstrapped/regtested x86_64-linux, comitted. Honza PR ipa/64612 * ipa-inline-tra