<div dir="ltr">No tests fail with the patch below, so I would say it's pretty useless. It seems that the C bindings are the only user but we can probably just have them return IPSCCP instead.<div><br></div><div>Any idea why this wasn't removed when IPSCCP was introduced? Probably worth understanding that before ripping it out, but in the current state of things I don't think removing it will be problematic.<br><div><br></div><div><div>diff --git a/llvm/lib/Transforms/IPO/IPConstantPropagation.cpp b/llvm/lib/Transforms/IPO/IPConstantPropagation.cpp</div><div>index a1533b3..24aea5c 100644</div><div>--- a/llvm/lib/Transforms/IPO/IPConstantPropagation.cpp</div><div>+++ b/llvm/lib/Transforms/IPO/IPConstantPropagation.cpp</div><div>@@ -51,7 +51,9 @@ char IPCP::ID = 0;</div><div> INITIALIZE_PASS(IPCP, "ipconstprop",</div><div>                 "Interprocedural constant propagation", false, false)</div><div> </div><div>-ModulePass *llvm::createIPConstantPropagationPass() { return new IPCP(); }</div><div>+ModulePass *llvm::createIPConstantPropagationPass() {</div><div>+  return createIPSCCPPass();</div><div>+}</div><div> </div><div> bool IPCP::runOnModule(Module &M) {</div><div>   if (skipModule(M))</div><div><br></div><div>-- Sean Silva</div><div><br></div></div></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Tue, May 3, 2016 at 3:06 PM, Davide Italiano via llvm-dev <span dir="ltr"><<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">The pass is pretty rudimental (as the comment at the top of the file<br>
hints), and it seems LLVM already has IPSCCP (which should do a better<br>
job at interprocedural constant propagation).<br>
I'm also not entirely sure it's used anywhere.<br>
Is there any reason to keep it around?<br>
<br>
Thanks,<br>
<span class="HOEnZb"><font color="#888888"><br>
--<br>
Davide<br>
<br>
"There are no solved problems; there are only problems that are more<br>
or less solved" -- Henri Poincare<br>
_______________________________________________<br>
LLVM Developers mailing list<br>
<a href="mailto:llvm-dev@lists.llvm.org">llvm-dev@lists.llvm.org</a><br>
<a href="http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" rel="noreferrer" target="_blank">http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br>
</font></span></blockquote></div><br></div>