similar to: RFC: Dynamic dominators

Hi Tobias, 1) Daniel and Chandler have for a long time been talking about computing > dominance and post-dominance in one shot to reduce the cost of > post-dominance and make it (freely) available everywhere. Is this > covered by your current (or planned) work? I'm not sure what you exactly mean by one shot; I'll ask around tomorrow. I wanted to play a little bit with your

Hi folks, For the past month I’ve been working on improving the DominatorTree and PostDominatorTree in LLVM. The RFC that explains the motivations and plans can be found here: http://lists.llvm.org/pipermail/llvm-dev/2017-June/114045.html . Here’s a short summary of what changed upstream since posting it: - We switched from the Simple Lengauer-Tarjan algorithm for computing dominators

This is release_50 branch of git, sha1: f1d5723be3f9456a6b16cdf687847ac2918846de Using gcc 7.2.0 from homebrew. $ CC=/usr/local/opt/gcc/bin/x86_64-apple-darwin16.7.0-gcc-7 CXX=/usr/local/opt/gcc/bin/x86_64-apple-darwin16.7.0-g++-7 cmake .. -DCMAKE_BUILD_TYPE=Release -DCMAKE_INSTALL_PREFIX=/Users/andy/local/llvm5 -DCMAKE_PREFIX_PATH=/Users/andy/local/llvm5 $ make VERBOSE=1 [ 92%] Linking CXX

Hello, I’m an undergraduate student studying CS in the South China University of Technology. I have been using clang compiler and related tools since I started studying C++ and I would like to work on LLVM in this year’s GSoC. I am interested in “Implement a single updater class for Dominators”. [1] I have achieved a bronze medal in the 2017 ACM-ICPC Asia Xian Regional Contest [2] (being a

Hi Kuba, Thanks for your advice in your previous letter. During last week, I have read the documents on Doxygen and the source code of the DomTreeBase/DomTree/PostDomTree/DeferredDominance class, I believe now I have a much better understanding on the relationship between these classes and how DeferredDominance class performs lazy updates. I have also learnt the current usage and drawbacks of

Btw, here is another interesting paper about post-dominators and control dependence: https://pdfs.semanticscholar.org/cbb2/9a0e4895025bd9df24f9263217df12f1ed1e.pdf I think a great outcome of your internship would be some precise documentation regarding the guarantees the LLVM dominators give -- possibly also considering classic and weak control dependence and the difference between

Hi Kuba, Thanks for your clarification on the project in the previous letter. I have submitted a proposal draft at the GSoC website, the draft has been shared with the LLVM organization. I will appreciate it if you can give me some advice on the proposal. This draft can be viewed by the organization. (If you do not have access, please mail me, and I will give you the link.) I am looking forward

Hi Chijun, Thanks for your interest in the project. I have gone through most of the LLVM Kaleidoscope tutorial and I have > watched the video of the presentation “Dominator Trees and incremental > updates that transcend time” presented on the 2017 LLVM Developers’ > Meeting. I have also started to understand the algorithm mentioned in > the comments of the code related to the

Hi Chijun, Great, seems like you did a lot of progress and understand the issues quite well! I have done some early sketch on the API of the new updater class. > From my current understanding, to solve the fragmentation problem of > the API, the new class first, need to maintain the DomTree and > PostDomTree class and deprecate the DefferredDominance class. > There are a couple of

Hi Chijun, I left you my feedback in the doc (+ some nitpicks). Overall, it looks very solid and shows you understand the problem well and know what to expect. Your proposed timeline is reasonable and I don't see any major things to improve there. Don't hesitate to reach out if you have any questions related to my comments (either here or in the document). Thanks, Kuba On Wed, Mar 21,

Hi Kuba, Thanks for your feedback. I have made some improvements on the proposal according to your feedback and clarify something on the time availability. I left comments on the questions you asked in the doc. Please check it out. Thanks, Chijun 2018-03-22 10:37 GMT+08:00 Jakub (Kuba) Kuderski <kubakuderski at gmail.com>: > Hi Chijun, > > I left you my feedback in the doc (+

Dear All, I'm having some problems using a function pass from a Module pass. My code is as follows: DominatorTree & domTree; ... Function &F = *I; // I is an interator from using Module::begin() ... domTree = getAnalysis<DominatorTree>(F); When I compile this code, I get the following error: /home/vadve/criswell/src/llvm22/include/llvm/Pass.h: In member function

Hi, Coincidentally I've been working to optimize this same case last week. I was struggling a bit to determine where to put this functionality and eventually went for the pragmatic approach of creating an experimental pass. Probably not the eventual solution, but it may provide some useful input to the discussion here. Basically, I experimented with a 'pre-inlining-transform' pass

All, I'm working on an improvement to the inline cost model, but I'm unsure how to proceed. Let me begin by first describing the problem I'm trying to solve. Consider the following pseudo C code: *typedef struct element { unsigned idx; } element_t; * *static inline unsigned char fn2 (element_t *dst_ptr, const element_t *a_ptr, const element_t *b_ptr,

On 8/4/2017 2:06 PM, Daniel Berlin wrote: > A few notes: > I'm a bit surprised IPO copy/constant propagation doesn't get this > case, but i didn't look if the lattice supports variables. > In particular, in your example, given no other call sites, it should > eliminate the dead code. > (In a real program, it may require cloning). In the actual program

Gordon Henriksen wrote: > What blocks would a phi node in %catch require for a case like this? > > define i8 @f(i1 %b) { > > entry: > > b label %try > > try: unwinds to %catch > > b i1 %b, label %then, label %else > > then: unwinds to %catch > > ret void > > else: unwinds to %catch > > ret

On 8/7/2017 1:02 PM, Daniel Berlin wrote: > Can someone fill me in on the issue with the dominator tree, > precisely, during inlining? > We now have the capability of quickly keeping it up to date without > too much trouble (it may require pushing it through a bunch of places, > but the actual changes to do should be easy). If I'm not mistaken (which I very well could be

Hello, I have code containing conditions and loops. Hence some BB execution are determined at run time depending on condition. Now I want to count only those BB that are always executed irrespective of condition result means reachable. and their execution is evident at compile time. How to do this? Please help Thank You Regards -------------- next part -------------- An HTML attachment was

Hi Andrew, Thanks a lot. But the function "DT->dominate(A,B)" decides the dominance relationship through comparing the DFS numbers, right? At least, in my example, when I check whether the newly split node (i.e., % *for.end.split*) DOMINATES the original node (I.e., %for.end), the answer is true, which is obviously wrong. Paul On Fri, Feb 14, 2014 at 1:59 AM, Andrew Trick

I think it would make sense for (1) and (2). I am not sure if (3) is feasible in instcombine. (I am not too familiar with LoopInfo) For the Octasic's Opus platform, I modified shouldMergeGEPs in our fork to: if (GEP.hasAllZeroIndices() && !Src.hasAllZeroIndices() && !Src.hasOneUse()) return false; return Src.hasAllConstantIndices(); // was return false;