<html><head><meta http-equiv="Content-Type" content="text/html charset=us-ascii"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;">On Jun 20, 2013, at 9:21 AM, Quentin Colombet <<a href="mailto:qcolombet@apple.com">qcolombet@apple.com</a>> wrote:<br><div><br class="Apple-interchange-newline"><blockquote type="cite"><div style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><div><br class="Apple-interchange-newline">On Jun 20, 2013, at 5:14 AM, Duncan Sands <<a href="mailto:duncan.sands@gmail.com">duncan.sands@gmail.com</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div style="font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;">Hi Quentin,<br><br>On 18/06/13 22:14, Quentin Colombet wrote:<br><blockquote type="cite">Author: qcolombet<br>Date: Tue Jun 18 15:14:39 2013<br>New Revision: 184222<br><br>URL:<span class="Apple-converted-space"> </span><a href="http://llvm.org/viewvc/llvm-project?rev=184222&view=rev">http://llvm.org/viewvc/llvm-project?rev=184222&view=rev</a><br>Log:<br>During SelectionDAG building explicitly set a node to constant zero when the<br>value is zero.<br></blockquote><br>you could also set it explicitly to -1 when all bits are sign bits.<br></div></blockquote>Yes, you are right.<br></div></blockquote><div dir="auto">No, in fact, this is not true.</div><div dir="auto">The sign bits can be all zeros. We do not know statically.</div><br><blockquote type="cite"><div style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><blockquote type="cite"><div style="font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><br><blockquote type="cite">This allows optmizations to kick in more easily.<br>Fix some test cases so that they remain meaningful (i.e., not completely dead<br>coded) when optimizations apply.<br><br><<a href="rdar://problem/14096009">rdar://problem/14096009</a>> superfluous multiply by high part of zero-extended<br>value.<br></blockquote><br>...<br><br><blockquote type="cite">Modified: llvm/trunk/test/CodeGen/ARM/mvn.ll<br>URL:<span class="Apple-converted-space"> </span><a href="http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/ARM/mvn.ll?rev=184222&r1=184221&r2=184222&view=diff">http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/ARM/mvn.ll?rev=184222&r1=184221&r2=184222&view=diff</a><br>==============================================================================<br>--- llvm/trunk/test/CodeGen/ARM/mvn.ll (original)<br>+++ llvm/trunk/test/CodeGen/ARM/mvn.ll Tue Jun 18 15:14:39 2013<br>@@ -1,4 +1,4 @@<br>-; RUN: llc < %s -march=arm | grep mvn | count 8<br>+; RUN: llc < %s -march=arm | grep mvn | count 9<br></blockquote><br>Is this an improvement?<br></div></blockquote><div dir="auto">Yes, but it is not obvious from the way the check is written though.</div><div dir="auto">Indeed, the sequence:</div><div dir="auto"><div dir="auto">mov<span class="Apple-tab-span" style="white-space: pre;"> </span>r0, #0</div><div dir="auto">sub<span class="Apple-tab-span" style="white-space: pre;"> </span>r0, r0, #3</div><div dir="auto">becomes</div><div dir="auto">mvn<span class="Apple-tab-span" style="white-space: pre;"> </span>r0, #2</div></div><br><blockquote type="cite"><div style="font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><br>...<br><blockquote type="cite">Modified: llvm/trunk/test/CodeGen/X86/2007-10-12-CoalesceExtSubReg.ll<br>URL:<span class="Apple-converted-space"> </span><a href="http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/X86/2007-10-12-CoalesceExtSubReg.ll?rev=184222&r1=184221&r2=184222&view=diff">http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/X86/2007-10-12-CoalesceExtSubReg.ll?rev=184222&r1=184221&r2=184222&view=diff</a><br>==============================================================================<br>--- llvm/trunk/test/CodeGen/X86/2007-10-12-CoalesceExtSubReg.ll (original)<br>+++ llvm/trunk/test/CodeGen/X86/2007-10-12-CoalesceExtSubReg.ll Tue Jun 18 15:14:39 2013<br>@@ -7,8 +7,10 @@ entry:<br> cond_next127:<span class="Apple-tab-span" style="white-space: pre;"> </span><span class="Apple-tab-span" style="white-space: pre;"> </span>; preds = %cond_next391, %entry<br> <span class="Apple-tab-span" style="white-space: pre;"> </span>%v.1 = phi i32 [ undef, %entry ], [ %tmp411, %cond_next391 ]<span class="Apple-tab-span" style="white-space: pre;"> </span><span class="Apple-tab-span" style="white-space: pre;"> </span>; <i32> [#uses=1]<br> <span class="Apple-tab-span" style="white-space: pre;"> </span>%tmp149 = mul i32 0, %v.1<span class="Apple-tab-span" style="white-space: pre;"> </span><span class="Apple-tab-span" style="white-space: pre;"> </span>; <i32> [#uses=0]<br>-<span class="Apple-tab-span" style="white-space: pre;"> </span>%tmp254 = and i32 0, 15<span class="Apple-tab-span" style="white-space: pre;"> </span><span class="Apple-tab-span" style="white-space: pre;"> </span>; <i32> [#uses=1]<br>-<span class="Apple-tab-span" style="white-space: pre;"> </span>%tmp256 = and i32 0, 15<span class="Apple-tab-span" style="white-space: pre;"> </span><span class="Apple-tab-span" style="white-space: pre;"> </span>; <i32> [#uses=2]<br>+<span class="Apple-tab-span" style="white-space: pre;"> </span>%tmpss = load i32* %ss, align 4<span class="Apple-tab-span" style="white-space: pre;"> </span><span class="Apple-tab-span" style="white-space: pre;"> </span>; <i32> [#uses=1]<br>+<span class="Apple-tab-span" style="white-space: pre;"> </span>%tmpbp = load i32* %bp, align 4<span class="Apple-tab-span" style="white-space: pre;"> </span><span class="Apple-tab-span" style="white-space: pre;"> </span>; <i32> [#uses=2]<br>+<span class="Apple-tab-span" style="white-space: pre;"> </span>%tmp254 = and i32 %tmpss, 15<span class="Apple-tab-span" style="white-space: pre;"> </span><span class="Apple-tab-span" style="white-space: pre;"> </span>; <i32> [#uses=1]<br>+<span class="Apple-tab-span" style="white-space: pre;"> </span>%tmp256 = and i32 %tmpbp, 15<span class="Apple-tab-span" style="white-space: pre;"> </span><span class="Apple-tab-span" style="white-space: pre;"> </span>; <i32> [#uses=2]<br></blockquote><br>This looks like a regression.<br></div></blockquote><div>No, it is not, I have to modify the input of test to prevent dead code eliminate to remove everything.</div><div>Thus, this is a complexification (is that a word?) of the test to exerce something... I have to admit what I was about to remove this test has the code was completely dead before the change in the input.</div></div></blockquote>Oh, my orthography is terrible... I need more coffee! (I guess you have got the idea :)).<br><blockquote type="cite"><div style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><div><br></div><div>-Quentin</div><br><blockquote type="cite"><div style="font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><br>Ciao, Duncan.<br>_______________________________________________<br>llvm-commits mailing list<br><a href="mailto:llvm-commits@cs.uiuc.edu">llvm-commits@cs.uiuc.edu</a><br><a href="http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits">http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits</a></div></blockquote></div><br style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><span style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px; float: none; display: inline !important;">_______________________________________________</span><br style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><span style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px; float: none; display: inline !important;">llvm-commits mailing list</span><br style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><a href="mailto:llvm-commits@cs.uiuc.edu" style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;">llvm-commits@cs.uiuc.edu</a><br style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><a href="http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits" style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;">http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits</a></blockquote></div><br></body></html>