Re: [PATCH 1/2] C++-ify dominance.c

2015-08-21 Thread Mikhail Maltsev
On 08/18/2015 10:00 PM, Jeff Law wrote: On 08/14/2015 10:02 PM, Mikhail Maltsev wrote: gcc/ChangeLog: 2015-08-15 Mikhail Maltsev malts...@gmail.com * dominance.c (new_zero_array): Define. (dom_info): Redefine as class with proper encapsulation.

Re: [PATCH 1/2] C++-ify dominance.c

2015-08-18 Thread Jeff Law
On 08/14/2015 10:02 PM, Mikhail Maltsev wrote: On 08/14/2015 10:54 AM, Richard Biener wrote: In fact initializing dom_info from that would allow it to work on SESE regions as well? Can't tell for sure. We would need a way to iterate through region's basic blocks (I'm not familiar with that

Re: [PATCH 1/2] C++-ify dominance.c

2015-08-15 Thread Richard Biener
On August 14, 2015 8:20:18 PM GMT+02:00, Jeff Law l...@redhat.com wrote: On 08/13/2015 06:55 PM, Mikhail Maltsev wrote: Hi all. These two patches are refactoring of dominator-related code. The comment in dominance.c says: We work in a poor-mans object oriented fashion, and carry an instance

Re: [PATCH 1/2] C++-ify dominance.c

2015-08-14 Thread Richard Biener
On Fri, Aug 14, 2015 at 2:55 AM, Mikhail Maltsev malts...@gmail.com wrote: Hi all. These two patches are refactoring of dominator-related code. The comment in dominance.c says: We work in a poor-mans object oriented fashion, and carry an instance of this structure through all our 'methods'.

Re: [PATCH 1/2] C++-ify dominance.c

2015-08-14 Thread Jeff Law
On 08/14/2015 01:54 AM, Richard Biener wrote: On Fri, Aug 14, 2015 at 2:55 AM, Mikhail Maltsev malts...@gmail.com wrote: Hi all. These two patches are refactoring of dominator-related code. The comment in dominance.c says: We work in a poor-mans object oriented fashion, and carry an instance

Re: [PATCH 1/2] C++-ify dominance.c

2015-08-14 Thread Jeff Law
On 08/13/2015 06:55 PM, Mikhail Maltsev wrote: Hi all. These two patches are refactoring of dominator-related code. The comment in dominance.c says: We work in a poor-mans object oriented fashion, and carry an instance of this structure through all our 'methods'. So, the first patch converts

Re: [PATCH 1/2] C++-ify dominance.c

2015-08-14 Thread Mikhail Maltsev
On 08/14/2015 10:54 AM, Richard Biener wrote: Putting in m_fn looks backwards to me - it looks like we only need to remember the entry and exit BBs and the number of blocks. Fixed. In fact initializing dom_info from that would allow it to work on SESE regions as well? Can't tell for sure.

[PATCH 1/2] C++-ify dominance.c

2015-08-13 Thread Mikhail Maltsev
Hi all. These two patches are refactoring of dominator-related code. The comment in dominance.c says: We work in a poor-mans object oriented fashion, and carry an instance of this structure through all our 'methods'. So, the first patch converts the mentioned structure (dom_info) into a class