Re: [Open64-devel] Code review for WHIRL SSA and CFG changes

2011-03-04 Thread Jian-Xin Lai
That's a typo. I'll correct it in the code. Thank you very much. 2011/3/5 Sun Chan > May be use consistent iterator or iter or something throughout the > entire wssa code > Sun > > On Sat, Mar 5, 2011 at 7:21 AM, Ye, Mei wrote: > > Hi Jian-Xin > > > > Looks good to me.The name of “dfs_fwd_i

Re: [Open64-devel] Code review for WHIRL SSA and CFG changes

2011-03-04 Thread Sun Chan
May be use consistent iterator or iter or something throughout the entire wssa code Sun On Sat, Mar 5, 2011 at 7:21 AM, Ye, Mei wrote: > Hi Jian-Xin > > Looks good to me.    The name of “dfs_fwd_itrator” will look better if it > can be changed to: “dfs_fwd_iterator”. > > It exists before this pat

Re: [Open64-devel] Code review for WHIRL SSA and CFG changes

2011-03-04 Thread Ye, Mei
Hi Jian-Xin Looks good to me.The name of "dfs_fwd_itrator" will look better if it can be changed to: "dfs_fwd_iterator". It exists before this patch, but should be easy to change since there are not many places that use it. -Mei From: Jian-Xin Lai [mailto:laij...@gmail.com] Sent: Friday, Ma

Re: [Open64-devel] Code review for WHIRL SSA and CFG changes

2011-03-02 Thread Jian-Xin Lai
Hi, Can a gate keeper review the patch for WHIRL SSA and CFG changes? Thank you very much. This patch fixed the following issues: 1. in WSSA_UPDATER::ssa_rename_tree(), BLOCK is not handled in previous code. 2. in WN_CFG_VERIFY_ACTION::Finalize(), dfs_iterator is used instead of BB_iterator becaus

Re: [Open64-devel] Code review for WHIRL SSA WOPT emitter

2010-11-29 Thread Jian-Xin Lai
For the first comment, it->second is right. It's the reference of the 'value' part of the pair in the map. I'll change the code according to your comments and check in. Thank you very much. 2010/11/29 Sun Chan > looks fine, except a couple of minor things > Sun > > in emitter.cxx, is this right

Re: [Open64-devel] Code review for WHIRL SSA WOPT emitter

2010-11-29 Thread Sun Chan
looks fine, except a couple of minor things Sun in emitter.cxx, is this right else { new_idx = it->second + 1; } ?? shouldn't this be it->second()? WSSA_VSYM_TYPE, I would add FmtAssert outside of the switch. If people add new cases and forgot to return inside the case stmt, you will have

Re: [Open64-devel] Code review for WHIRL SSA WOPT emitter

2010-11-17 Thread Jian-Xin Lai
Sorry, forgot to attach another patch. This patch contains the changes in osprey/common/com, including: * read/write WSSA information from/to .B/.I/etc file * enhancements to ir_b2a to dump the WSSA information * options to control WSSA. It's turned off by default so far. Thank you very much. 201

Re: [Open64-devel] Code review for WHIRL SSA

2010-11-09 Thread Jian-Xin Lai
I just want to check in the code. If any one are interested to do some analysis/optimizations based on WHIRL SSA, he can start here. 2010/11/9 Tianwei > For the testing, the most straightforward way is using client > optimizations to verify its effects. What's plan for this? or you just > wan

Re: [Open64-devel] Code review for WHIRL SSA

2010-11-09 Thread Tianwei
For the testing, the most straightforward way is using client optimizations to verify its effects. What's plan for this? or you just want to checkin the files now? Tianwei On Wed, Nov 10, 2010 at 5:23 AM, Jian-Xin Lai wrote: > Hi, > > I plan to check in the core code for WHIRL SSA recently.