search for: end12

...o a different function, and the function itself is missing a DISubprogram metadata attachment. > >> (lldb) p OldFunc->dump() >> >> ; Function Attrs: nounwind optsize >> define internal void @f_1.extracted_region(i32, i32*, %struct.t_c*, %struct.t_d*) #0 { >> if.end12.extracted_entry: >> %and14 = and i32 %0, 2, !dbg !89 >> %tobool15 = icmp eq i32 %and14, 0, !dbg !89 >> br i1 %tobool15, label %exit, label %if.then16, !dbg !185 >> >> if.then16: ; preds = %if.end12.extracted_entry >> %4...

...on itself is missing a DISubprogram metadata attachment. >> >>> (lldb) p OldFunc->dump() >>> >>> ; Function Attrs: nounwind optsize >>> define internal void @f_1.extracted_region(i32, i32*, %struct.t_c*, >>> %struct.t_d*) #0 { >>> if.end12.extracted_entry: >>> %and14 = and i32 %0, 2, !dbg !89 >>> %tobool15 = icmp eq i32 %and14, 0, !dbg !89 br i1 %tobool15, label >>> %exit, label %if.then16, !dbg !185 >>> >>> if.then16: ; preds = %if.end12.extracted...

...exit28:                                ; preds = %polly.stmt.for.body.i, %polly.loop_if25   br label %polly.merge_new_and_old polly.merge_new_and_old:                          ; preds = %polly.loop_exit28, %for.body.i   ...   %cmp = icmp slt i32 %2, 0   ...   br i1 %cmp, label %for.body, label %for.end12 for.end12:                                        ; preds = %loop.exit   ...   ret i32 0    Unfortunately, the new basic block "polly.merge_new_and_old" is marked as unreachable after a serial of immediate Loop optimizations in the following pass order:           Polly - Create LLVM-IR fr...

The if you are cloning into the same LLVM module the CU should not cloned. If don't mind sharing your code, I can try to help diagnose why the CU gets cloned... just send me a patch that applies to trunk and instructions. -- adrian > On Jun 16, 2017, at 1:54 PM, Sergei Larin <slarin at codeaurora.org> wrote: > > Sorry… It takes a pass that was not accepted for upstreaming….

...; preds = %polly.stmt.for.body.i, %polly.loop_if25 > br label %polly.merge_new_and_old > polly.merge_new_and_old: ; preds = %polly.loop_exit28, %for.body.i > ... > %cmp = icmp slt i32 %2, 0 > ... > br i1 %cmp, label %for.body, label %for.end12 > for.end12: ; preds = %loop.exit > ... > ret i32 0 > > Unfortunately, the new basic block "polly.merge_new_and_old" is marked as unreachable after a serial of immediate Loop optimizations in the following pass order: >...

...ec.ptr1.9.i.i, align 2, !tbaa !5 %265 = load i16* %incdec.ptr.10.i.i, align 2, !tbaa !5 store i16 %265, i16* %incdec.ptr1.10.i.i, align 2, !tbaa !5 store i16 0, i16* %scevgep.i596.i, align 2, !tbaa !5 br label %for.body.i599.i for.body.i599.i: ; preds = %if.end12.i.for.body.i599_crit_edge.i, %if.end153.i %266 = phi i16 [ %256, %if.end153.i ], [ %.pre1004.i, %if.end12.i.for.body.i599_crit_edge.i ] %i.025.i.i = phi i32 [ 3, %if.end153.i ], [ %phitmp1011.i, %if.end12.i.for.body.i599_crit_edge.i ] %bits.024.i.i = phi i16 [ 0, %if.end153.i ], [ %conv14.i.i...

Hi Tobias, At 2013-09-19 22:59:25,"Tobias Grosser" <tobias at grosser.es> wrote: >On 09/19/2013 04:46 PM, Star Tan wrote: >> Hi Tobias, >> >> >> I am trying to move Polly later. >> >> >> LLVM provides some predefined ExtensionPointTy: >> EP_EarlyAsPossible, >> EP_ModuleOptimizerEarly, >>

> On Mar 2, 2015, at 4:19 PM, Diego Novillo <dnovillo at google.com> wrote: > > On Thu, Feb 26, 2015 at 6:54 PM, Diego Novillo <dnovillo at google.com <mailto:dnovillo at google.com>> wrote: > > I've created a few bugzilla issues with details of some of the things I'll be looking into. I'm not yet done wordsmithing the overall design document.