<html dir="ltr"><head>
<meta http-equiv="Content-Type" content="text/html; charset=Windows-1252">
<style>@font-face {
font-family: Cambria Math;
}
@font-face {
font-family: Calibri;
}
@font-face {
font-family: Tahoma;
}
@page WordSection1 {margin: 1.0in 1.0in 1.0in 1.0in; }
P.MsoNormal {
MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman","serif"; FONT-SIZE: 12pt
}
LI.MsoNormal {
MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman","serif"; FONT-SIZE: 12pt
}
DIV.MsoNormal {
MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman","serif"; FONT-SIZE: 12pt
}
A:link {
COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlink {
COLOR: blue; TEXT-DECORATION: underline
}
A:visited {
COLOR: blue; TEXT-DECORATION: underline
}
SPAN.MsoHyperlinkFollowed {
COLOR: blue; TEXT-DECORATION: underline
}
P {
MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman","serif"; FONT-SIZE: 12pt
}
P.MsoAcetate {
MARGIN: 0in 0in 0pt; FONT-FAMILY: "Tahoma","sans-serif"; FONT-SIZE: 8pt
}
LI.MsoAcetate {
MARGIN: 0in 0in 0pt; FONT-FAMILY: "Tahoma","sans-serif"; FONT-SIZE: 8pt
}
DIV.MsoAcetate {
MARGIN: 0in 0in 0pt; FONT-FAMILY: "Tahoma","sans-serif"; FONT-SIZE: 8pt
}
SPAN.BalloonTextChar {
FONT-FAMILY: "Tahoma","sans-serif"
}
P.balloontext {
MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman","serif"; FONT-SIZE: 12pt
}
LI.balloontext {
MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman","serif"; FONT-SIZE: 12pt
}
DIV.balloontext {
MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman","serif"; FONT-SIZE: 12pt
}
P.balloontext0 {
MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman","serif"; FONT-SIZE: 12pt
}
LI.balloontext0 {
MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman","serif"; FONT-SIZE: 12pt
}
DIV.balloontext0 {
MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman","serif"; FONT-SIZE: 12pt
}
P.balloontext1 {
MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman","serif"; FONT-SIZE: 12pt
}
LI.balloontext1 {
MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman","serif"; FONT-SIZE: 12pt
}
DIV.balloontext1 {
MARGIN: 0in 0in 0pt; FONT-FAMILY: "Times New Roman","serif"; FONT-SIZE: 12pt
}
SPAN.balloontextchar0 {
FONT-FAMILY: "Tahoma","sans-serif"
}
SPAN.balloontextchar00 {
FONT-FAMILY: "Tahoma","sans-serif"
}
SPAN.balloontextchar000 {
FONT-FAMILY: "Tahoma","sans-serif"
}
SPAN.emailstyle24 {
FONT-FAMILY: "Arial","sans-serif"; COLOR: navy
}
SPAN.emailstyle25 {
FONT-FAMILY: "Calibri","sans-serif"; COLOR: #1f497d
}
SPAN.emailstyle26 {
FONT-FAMILY: "Arial","sans-serif"; COLOR: navy
}
SPAN.emailstyle27 {
FONT-FAMILY: "Arial","sans-serif"; COLOR: navy
}
SPAN.emailstyle28 {
FONT-FAMILY: "Calibri","sans-serif"; COLOR: #1f497d
}
SPAN.emailstyle29 {
FONT-FAMILY: "Calibri","sans-serif"; COLOR: #1f497d
}
SPAN.emailstyle30 {
FONT-FAMILY: "Arial","sans-serif"; COLOR: navy
}
SPAN.emailstyle32 {
FONT-FAMILY: "Calibri","sans-serif"; COLOR: #1f497d
}
SPAN.emailstyle33 {
FONT-FAMILY: "Arial","sans-serif"; COLOR: navy
}
SPAN.apple-converted-space {
}
SPAN.EmailStyle36 {
FONT-FAMILY: "Calibri","sans-serif"; COLOR: #1f497d
}
.MsoChpDefault {
FONT-SIZE: 10pt
}
DIV.WordSection1 {
}
</style>
<meta name="GENERATOR" content="MSHTML 8.00.6001.19088">
<style id="owaTempEditStyle"></style><style title="owaParaStyle"><!--P {
MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px
}
--></style>
</head>
<body lang="EN-US" link="blue" vlink="blue" ocsi="x">
<div style="FONT-FAMILY: Tahoma; DIRECTION: ltr; COLOR: #000000; FONT-SIZE: x-small">
<div>
<div>Of course, c<font face="tahoma">ommitted at r153694 with these changes, thank you!</font></div>
<div><font face="tahoma"></font> </div>
<div>Regards,</div>
<div><font face="tahoma">Danil</font></div>
</div>
<div dir="ltr"><font color="#000000" size="2" face="Tahoma"></font> </div>
<div style="DIRECTION: ltr" id="divRpF90265">
<hr tabindex="-1">
<font color="#000000" size="2" face="Tahoma"><b>From:</b> Kaylor, Andrew [andrew.kaylor@intel.com]<br>
<b>Sent:</b> Thursday, March 29, 2012 1:30 PM<br>
<b>To:</b> Danil Malyshev; Jim Grosbach; llvm-commits@cs.uiuc.edu<br>
<b>Subject:</b> RE: RuntimeDyLd new features<br>
</font><br>
</div>
<div></div>
<div>
<div class="WordSection1">
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">Hi Danil,</span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">When you use the Check() function, there’s no need to have the ‘if (Check(…) break;’ idiom because Check will execute a fatal error in the case of an error.
In retrospect, there’s no reason that Check should not be a void function.</span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">Otherwise, the 01 and 02 patches look good.</span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">-Andy</span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt"></span> </p>
<div>
<div style="BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; FONT-SIZE: 10pt">From:</span></b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; FONT-SIZE: 10pt"> Danil Malyshev [mailto:dmalyshev@accesssoftek.com]
<br>
<b>Sent:</b> Thursday, March 29, 2012 11:40 AM<br>
<b>To:</b> Danil Malyshev; Kaylor, Andrew; Jim Grosbach; llvm-commits@cs.uiuc.edu; Sean Callanan<br>
<b>Cc:</b> Chandler Carruth<br>
<b>Subject:</b> RE: RuntimeDyLd new features</span></p>
</div>
</div>
<p class="MsoNormal"> </p>
<div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">Hi everyone,</span></p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">These patches was reverted by some reason, so please review changed patches 01-RuntimeDyLd-04.patch and 02-MCJIT_tests-02.patch.</span></p>
</div>
<p><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU"></span> </p>
<p><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">The 01-RuntimeDyLd-04.patch has some style and name changes with Chandler wrote me about.<br>
It's also has changes in MachO relocations committed by Sean in r153466.<br>
Unfortunately, I again cannot add to this patch Sean's changes from r153147 because the ExecutionEngine/MCJIT tests fails with these changes.<br>
Sean, can you review attached the relocation_referrers.patch? It's has these changes prepared for my patch, maybe I missed something.</span></p>
<p><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">02-MCJIT_tests-02.patch tests has changes for cmake, so the host_os filter now correctly works.</span></p>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU"></span> </p>
</div>
<p><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">Regards,</span></p>
<p><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">Danil</span></p>
<p><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU"></span> </p>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU"></span> </p>
</div>
<div id="divRpF648036">
<div style="TEXT-ALIGN: center" class="MsoNormal" align="center"><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">
<hr align="center" size="2" width="100%">
</span></div>
<p style="MARGIN-BOTTOM: 12pt" class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">From:</span></b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU"> Danil Malyshev<br>
<b>Sent:</b> Wednesday, March 21, 2012 2:20 PM<br>
<b>To:</b> Kaylor, Andrew; Jim Grosbach; <a href="mailto:llvm-commits@cs.uiuc.edu">
llvm-commits@cs.uiuc.edu</a><br>
<b>Subject:</b> RE: RuntimeDyLd new features</span></p>
</div>
<div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Hello,</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Committed at r153221, thank you!</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">I tried to add changes from r153147 to the patch, but unfortunately, the test ExecutionEngine/MCJIT/test-global-init-nonzero.ll failed after it.</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Attached the patch does these changes, maybe Spyffe can fix it.</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Regards,</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Danil</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<div style="TEXT-ALIGN: center" class="MsoNormal" align="center"><span style="COLOR: black" lang="RU">
<hr align="center" size="2" width="100%">
</span></div>
<p class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">From:</span></b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU"> Kaylor, Andrew [<a href="mailto:andrew.kaylor@intel.com">mailto:andrew.kaylor@intel.com</a>]
<br>
<b>Sent:</b> Wednesday, March 21, 2012 2:28 AM<br>
<b>To:</b> Danil Malyshev; Jim Grosbach; <a href="mailto:llvm-commits@cs.uiuc.edu">
llvm-commits@cs.uiuc.edu</a><br>
<b>Subject:</b> RE: RuntimeDyLd new features</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">Looks good to me!</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<div style="BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt">From:</span></b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt"> Danil Malyshev [<a href="mailto:dmalyshev@accesssoftek.com">mailto:dmalyshev@accesssoftek.com</a>]
<br>
<b>Sent:</b> Tuesday, March 20, 2012 12:00 PM<br>
<b>To:</b> Kaylor, Andrew; Jim Grosbach; <a href="mailto:llvm-commits@cs.uiuc.edu">
llvm-commits@cs.uiuc.edu</a><br>
<b>Subject:</b> RE: RuntimeDyLd new features</span><span style="COLOR: black" lang="RU"></span></p>
</div>
</div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Hello Andrew,</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">I'm very sorry for big delay.</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">I'm just yet submitted the patch move_getPointerToNamedFunction-01.patch:</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">It's will moves getPointerToNamedFunction() from JIT/MCJIT to JITMemoryManager, so we can use it in MCJITMemoryManager.</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Please review attached the patch 01-RuntimeDyLd-03.patch</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">It's uses MCJITMemoryManager.getPointerToNamedFunction() for lookup external functions.</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">But until the patch move_getPointerToNamedFunction-01.patch not committed, MCJITMemoryManager.getPointerToNamedFunction() implementation uses sys::DynamicLibrary::SearchForAddressOfSymbol,
so I can commit it right now and it's will work normally and will pass tests from 02-MCJIT_tests-01.patch.</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Also I add changes RuntimeDyldMachO from March 8 to this patch.</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Regards,</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Danil</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<div style="TEXT-ALIGN: center" class="MsoNormal" align="center"><span style="COLOR: black" lang="RU">
<hr align="center" size="2" width="100%">
</span></div>
<p class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">From:</span></b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU"> Kaylor, Andrew [<a href="mailto:andrew.kaylor@intel.com">mailto:andrew.kaylor@intel.com</a>]
<br>
<b>Sent:</b> Tuesday, March 20, 2012 10:13 PM<br>
<b>To:</b> Kaylor, Andrew; Jim Grosbach; Danil Malyshev; <a href="mailto:llvm-commits@cs.uiuc.edu">
llvm-commits@cs.uiuc.edu</a><br>
<b>Subject:</b> RE: RuntimeDyLd new features</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">Ping.</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<div style="BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt">From:</span></b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt">
<a href="mailto:llvm-commits-bounces@cs.uiuc.edu">llvm-commits-bounces@cs.uiuc.edu</a> [<a href="mailto:llvm-commits-bounces@cs.uiuc.edu">mailto:llvm-commits-bounces@cs.uiuc.edu</a>]
<b>On Behalf Of </b>Kaylor, Andrew<br>
<b>Sent:</b> Tuesday, March 13, 2012 3:28 PM<br>
<b>To:</b> Jim Grosbach; Danil Malyshev; <a href="mailto:llvm-commits@cs.uiuc.edu">
llvm-commits@cs.uiuc.edu</a><br>
<b>Subject:</b> Re: [llvm-commits] RuntimeDyLd new features</span><span style="COLOR: black" lang="RU"></span></p>
</div>
</div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">What is the status of this patch?</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">I have some changes to go into this code area which layer on top of this patch. Should I post a contingent version of my changes to the list so that people
can get a preview of what I’ve got in the works?</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">Thanks,</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">Andy</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<div style="BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt">From:</span></b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt"> Jim Grosbach [<a href="mailto:grosbach@apple.com">mailto:grosbach@apple.com</a>]
<br>
<b>Sent:</b> Tuesday, March 06, 2012 1:40 PM<br>
<b>To:</b> Danil Malyshev<br>
<b>Cc:</b> Kaylor, Andrew; <a href="mailto:llvm-commits@cs.uiuc.edu">llvm-commits@cs.uiuc.edu</a><br>
<b>Subject:</b> Re: RuntimeDyLd new features</span><span style="COLOR: black" lang="RU"></span></p>
</div>
</div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="COLOR: black">Hi Danil,</span><span style="COLOR: black" lang="RU"></span></p>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black">Yes, I agree getPointerToNamedFunction() should move to the memory manager.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<div>
<p class="MsoNormal"><span style="COLOR: black">On Mar 6, 2012, at 12:31 PM, Danil Malyshev <<a href="mailto:dmalyshev@accesssoftek.com">dmalyshev@accesssoftek.com</a>> wrote:</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<p style="MARGIN-BOTTOM: 12pt" class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Hi Jim,</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Thank you for the update. You wrote an interesting idea. If we continue to follow it, I think the best way is move an implementation of the getPointerToNamedFunction()
from JIT/MCJIT to DefaultJITMemoryManager.</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">And JIT/MCJIT::getPointerToNamedFunction() just will do something like:</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt"> if (!isSymbolSearchingDisabled()) {</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt"> JMM->getPointerToNamedFunction(</span><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt" lang="RU">…);</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<blockquote style="MARGIN-TOP: 5pt; MARGIN-BOTTOM: 5pt">
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt"> }</span><span style="COLOR: black" lang="RU"></span></p>
</div>
</blockquote>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black">Right, with the minor nit that this will call MemMgr-> getPointerToNamedFunction() since that's what the MCJIT has access to. The JMM will be called by MCJITMemoryManager().</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<p style="MARGIN-BOTTOM: 12pt" class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">So we can delete MCJIT/Intercept.cpp and store only JIT/Intercept.cpp.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black">Exactly. Eventually, JIT/Intercept.cpp will go away too, which the old JIT is removed entirely. We could refactor this part of it to use the JMM as well if we want, though.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black">Regards,</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black"> Jim</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<p style="MARGIN-BOTTOM: 12pt" class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">That do you think about it?</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Regards,</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Danil</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<div style="TEXT-ALIGN: center" class="MsoNormal" align="center"><span style="COLOR: black" lang="RU">
<hr align="center" size="2" width="100%">
</span></div>
<p class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt">From:</span></b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt"> Jim Grosbach [<a href="mailto:grosbach@apple.com">mailto:grosbach@apple.com</a>]
<br>
<b>Sent:</b> Tuesday, March 06, 2012 10:40 PM<br>
<b>To:</b> Danil Malyshev<br>
<b>Cc:</b> Kaylor, A</span><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">ndrew;
<a href="mailto:llvm-commits@cs.uiuc.edu">llvm-commits@cs.uiuc.edu</a><br>
<b>Subject:</b> Re: RuntimeDyLd new features</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">Hi Danil,</span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">Ack, sorry! I lost track of where things stood and didn't realize you were waiting for me. Mea culpa.</span></p>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">I like the general idea of putting the guts of the logic for symbol in the memory manager interface, as you're right, that's the piece that actually knows about the target. The RTDyldMemoryManager is
just an interface specification, not an implementation of anything. I like the idea of using that for the conduit. The only thing I don't care for is the idea that the memory manager should call into the ExecutionEngine. I don't think that's right. I would
instead do it the other way around. We'd add the interface to RTDyldMemoryManger and JITMemoryManager, both. The MCJITMemoryManager implementation would tie the two together via:</span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">uint64_t getExternalSymbolAddr(…) {</span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"> return JMM->getExternalSymbolAddr(…);</span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">}</span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">The DefaultJITMemoryManager() would, as it does now, consider itself a hosted environment and do whatever appropriate symbol lookup is needed. There's some legacy horribleness in MCJIT/Intercept.cpp
that would all go away via this, which would be fantastic.</span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">It seems OK to me for RuntimeDyld to call this function whenever it sees a symbol it doesn't know. Theoretically, that should allow everything to "just work" for external symbols in a hosted environment
without any extra fiddling by the client process. For a remote target, there will be a custom JITMemoryManager implementation anyway to handle the remote allocation and such, so that can be handled there, too.</span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">Regards,</span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">-Jim</span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">On Mar 6, 2012, at 9:00 AM, Danil Malyshev <<a href="mailto:dmalyshev@accesssoftek.com">dmalyshev@accesssoftek.com</a>> wrote:</span></p>
</div>
<p style="MARGIN-BOTTOM: 12pt" class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Hi Andrew,</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">I'm waiting for Jim's response, but can submit the changed patch tomorrow.</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Regards,</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Danil</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<div style="TEXT-ALIGN: center" class="MsoNormal" align="center"><span style="COLOR: black" lang="RU">
<hr align="center" size="2" width="100%">
</span></div>
<p class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">From:</span></b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU"> Kaylor, Andrew [<a href="mailto:andrew.kaylor@intel.com">mailto:andrew.kaylor@intel.com</a>]
<br>
<b>Sent:</b> Tuesday, March 06, 2012 8:46 PM<br>
<b>To:</b> Danil Malyshev; Jim Grosbach<br>
<b>Cc:</b> <a href="mailto:llvm-commits@cs.uiuc.edu">llvm-commits@cs.uiuc.edu</a><br>
<b>Subject:</b> RE: RuntimeDyLd new features</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">Hi Danil,</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">What is the current status of your patch? Are you waiting for a response from Jim or will you be re-submitting something soon?</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">I have a set of changes merged with your latest patch and I’m hoping to minimize any conflicts in our ongoing work.</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">Thanks,</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">Andy</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<div style="BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<p class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt">From:</span></b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt"> Danil Malyshev [<a href="mailto:dmalyshev@accesssoftek.com">mailto:dmalyshev@accesssoftek.com</a>]
<br>
<b>Sent:</b> Tuesday, February 28, 2012 12:46 PM<br>
<b>To:</b> Jim Grosbach; Kaylor, Andrew<br>
<b>Cc:</b> <a href="mailto:llvm-commits@cs.uiuc.edu">llvm-commits@cs.uiuc.edu</a><br>
<b>Subject:</b> RE: RuntimeDyLd new features</span><span style="COLOR: black" lang="RU"></span></p>
</div>
</div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Hello Jim,</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">As variant, add the GetExternalSymbolAddr() to the RTDyldMemoryManager.</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">The memory manager linked to a target system, its emit sections to the target system, so its can known external symbol addresses.</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">The MCJITMemoryManager will use ExecutionEngine for getting addresses, another memory manager will use their methods, or just return 0 if this feature not supported.</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">I can add these changes to the patch quickly.</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Thank you,</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Danil</span><span style="COLOR: black" lang="RU"></span></p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<div style="TEXT-ALIGN: center" class="MsoNormal" align="center"><span style="COLOR: black" lang="RU">
<hr align="center" size="2" width="100%">
</span></div>
<p class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">From:</span></b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU"> Jim Grosbach [<a href="mailto:grosbach@apple.com">mailto:grosbach@apple.com</a>]
<br>
<b>Sent:</b> Tuesday, February 28, 2012 4:36 AM<br>
<b>To:</b> Kaylor, Andrew<br>
<b>Cc:</b> Danil Malyshev; <a href="mailto:llvm-commits@cs.uiuc.edu">llvm-commits@cs.uiuc.edu</a><br>
<b>Subject:</b> Re: RuntimeDyLd new features</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">On Feb 23, 2012, at 2:08 PM, "Kaylor, Andrew" <<a href="mailto:andrew.kaylor@intel.com">andrew.kaylor@intel.com</a>> wrote:</span></p>
</div>
<p style="MARGIN-BOTTOM: 12pt" class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">These changes look good to me.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">The test-globall.ll test is failing because your code doesn’t handle sections without bits in the object image and needing zero-initialization (at least,
that’s why it fails on ELF). I believe that the stubs.ll and test-common-symbols.ll tests are both failing because common symbols aren’t being properly handled.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">As I mentioned before, I think the common symbol issue requires a bit of restructuring to allow format specific handling of part of the common symbol processing.
I have a fix ready which addresses this for ELF and would provide most of the fix and the necessary interface to fix it on MachO. Basically, my fix makes a list of common symbols, allocates memory for the common symbols and calls a virtual function to update
the symbol address in the SymbolRef. MachO would just need an implementation of the function to test a symbol to see if it is a common symbol, and an implementation of the function to update the address.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">I also have a general fix ready for the nobits/zero-init problem.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">Therefore, I think it would be OK to have this patch committed as is, with the understanding that these problems would be fixed shortly afterward. Again,
that is assuming that the changes are otherwise acceptable to Jim.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">I'm generally OK with that, with one exception. Adding a back-reference to the ExecutionEngine to RuntimeDyld is a pretty bad layering violation, as alluded to in the code comments. I'm not comfortable
with that part going in as-is.</span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">We need an interface to the RuntimeDyld which will allow the client to query for unresolved symbols and to</span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">a) provide an address for them if they are true external symbols, or</span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">b) provide a module which supplies the symbols.</span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">In a normal hosted JIT environment, the second option is our vector to lazy compilation. The JIT builds the entry point, then queries for unresolved symbols. It creates a module containing each that
has a compile-stub as an instantiation.</span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">The key is that the behavior is driven top-down from the client. There shouldn't be calls back up from the RuntimeDyld to either the JIT or the ExecutionEngine.</span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU">-Jim</span></p>
</div>
<p style="MARGIN-BOTTOM: 12pt" class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
<div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">I would like to discuss the way that the MCJIT-specific tests are handled, but we can talk about that later.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">-Andy</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<div style="BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<div>
<p class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt">From:</span></b><span class="apple-converted-space"><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt"> </span></span><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt">Danil
Malyshev [<a href="mailto:dmalyshev@accesssoftek.com">mailto:dmalyshev@accesssoftek.com</a>]<span class="apple-converted-space"> </span><br>
<b>Sent:</b><span class="apple-converted-space"> </span>Wednesday, February 22, 2012 4:08 PM<br>
<b>To:</b><span class="apple-converted-space"> </span>Jim Grosbach; Kaylor, Andrew;
<a href="mailto:llvm-commits@cs.uiuc.edu">llvm-commits@cs.uiuc.edu</a><br>
<b>Subject:</b><span class="apple-converted-space"> </span>RE: RuntimeDyLd new features</span><span style="COLOR: black" lang="RU"></span></p>
</div>
</div>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Hi Jim,</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">The MCJIT with changed RuntimeDyLd passes most of ExecutionEngine tests on Mac OS, except of 3 tests: stubs.ll, test-common-symbols.ll and test-global.ll.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">And I will work to ensure that these three tests are also passed.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Please look 01-RuntimeDyLd-02.patch of the previous letter, can I commit it?</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Regards,</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Danil</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<div style="TEXT-ALIGN: center" class="MsoNormal" align="center"><span style="COLOR: black" lang="RU">
<hr align="center" size="2" width="100%">
</span></div>
<div>
<p class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">From:</span></b><span class="apple-converted-space"><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU"> </span></span><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">Kaylor,
Andrew [<a href="mailto:andrew.kaylor@intel.com">mailto:andrew.kaylor@intel.com</a>]<span class="apple-converted-space"> </span><br>
<b>Sent:</b><span class="apple-converted-space"> </span>Saturday, February 18, 2012 5:41 AM<br>
<b>To:</b><span class="apple-converted-space"> </span>Danil Malyshev; <a href="mailto:llvm-commits@cs.uiuc.edu">
llvm-commits@cs.uiuc.edu</a><br>
<b>Cc:</b><span class="apple-converted-space"> </span>Jim Grosbach<br>
<b>Subject:</b><span class="apple-converted-space"> </span>RE: RuntimeDyLd new features</span><span style="COLOR: black" lang="RU"></span></p>
</div>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">Hi Danil,</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">I’d like to make clear that if Jim is OK with the impact of your patch on MachO JIT loading then I’d be happy for you to proceed with your patch, incorporating
my comments below, and we will withdraw our previous uncommitted patch and I will merge the GDB JIT debugging integration code we have with your changes after they have been committed.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">Thanks,</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">Andy</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<div style="BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<div>
<p class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt">From:</span></b><span class="apple-converted-space"><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt"> </span></span><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt"><a href="mailto:llvm-commits-bounces@cs.uiuc.edu">llvm-commits-bounces@cs.uiuc.edu</a>
[<a href="mailto:llvm-commits-bounces@cs.uiuc.edu">mailto:llvm-commits-bounces@cs.uiuc.edu</a>]<span class="apple-converted-space"> </span><b>On Behalf Of<span class="apple-converted-space"> </span></b>Kaylor, Andrew<br>
<b>Sent:</b><span class="apple-converted-space"> </span>Tuesday, February 14, 2012 4:53 PM<br>
<b>To:</b><span class="apple-converted-space"> </span>Danil Malyshev; <a href="mailto:llvm-commits@cs.uiuc.edu">
llvm-commits@cs.uiuc.edu</a><br>
<b>Subject:</b><span class="apple-converted-space"> </span>Re: [llvm-commits] RuntimeDyLd new features</span><span style="COLOR: black" lang="RU"></span></p>
</div>
</div>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">Hi Danil,</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">I've been working with your patch also. Mostly, it looks pretty good. I've been able to layer most of our added functionality on top of it, and I have
GDB integration working and our integration tests passing. There are a few issues I'd like to bring up, however.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">The first thing I saw with your patch is that I had problems loading images that were generated with debug information included. In particular, applying
relocations to debug sections using the actual addresses of the debug sections causes problems. This is really a small issue because, as you mentioned, debug sections don't need to be loaded. I worked around this problem by adding a new member function to
SectionRef to query whether or not the section is required for execution. This is fairly trivial for ELF.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">A related problem is that your code does not properly handle sections which are only represented by a header in the ELF image (that is, SHT_NOBITS sections)
or sections which need to be zero-initialized. I worked around these problems using the existing isBSS() member in SectionRef, but it would probably be better to split these two concepts.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">Similarly, your code does not handle symbols which are marked as relative to the SHN_COMMON section. Fixing this is a bit more complicated. I think we
can probably build a table of these symbols in the RuntimeDyldImpl class by adding a new member to SectionRef, but we'll need deeper access to the format-specific classes to update the symbol addresses. I'm working on an implementation of this for ELF.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">The code which loads the sections is currently assuming pointer-sized alignment for all sections. I ran into a case using floating point numbers where 16-byte
alignment was required. This is a fairly simple fix as the alignment value is accessible through SectionRef.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">I think it would be worthwhile adding an function to SectionRef to query whether or not a section contains code rather than the current algorithmic method
you are using.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">In a few places you are using uintptr_t or uint8_t* in places that will restrict cross-architecture JITing. The TargetAddr in the SectionEntry structure
is an example that comes to mind.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">The distinction between code and data for memory allocation will be important going forward. I think it would be better to fix the existing bug that to
disable the specific allocations.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">In RuntimeDyld::loadObject, in the "parse and process relocations loop" "SectionID == 0" is used to identify sections that need to be looked up, but zero
is a valid SectionID. For that section, it goes into the lookup for every relocation.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">I'm continuing to work on integrating our code, including some that hasn't yet been submitted to the list for review, with your changes. I need to go over
what I've done with the rest of my co-workers, but I wanted to give you a progress update now.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: #1f497d; FONT-SIZE: 11pt">-Andy</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<div style="BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<div>
<p class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt">From:</span></b><span class="apple-converted-space"><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt"> </span></span><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt">Danil
Malyshev [<a href="mailto:dmalyshev@accesssoftek.com">mailto:dmalyshev@accesssoftek.com</a>]<span class="apple-converted-space"> </span><br>
<b>Sent:</b><span class="apple-converted-space"> </span>Tuesday, February 14, 2012 10:30 AM<br>
<b>To:</b><span class="apple-converted-space"> </span>Kaylor, Andrew; <a href="mailto:llvm-commits@cs.uiuc.edu">
llvm-commits@cs.uiuc.edu</a><br>
<b>Cc:</b><span class="apple-converted-space"> </span>Jim Grosbach<br>
<b>Subject:</b><span class="apple-converted-space"> </span>RE: RuntimeDyLd new features</span><span style="COLOR: black" lang="RU"></span></p>
</div>
</div>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Hi Andrew,</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Thank you for the detailed explanation.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">I have carefully studied your patch and DyldELFObject.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">A gdb support is very important and it's actually a big step for MCJIT. I like your patch, its loadObject() looks faster than my. But I see one big conceptual
problem.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">The RuntimeDyLd was developed as a linker, which is able to prepare data for running on another platform. The RuntimeDyldMachO largely implements it.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">In the RuntimeDyldELF you went the other way. Now it is very far from remote execution. And even if the object file will be emitted with RTDyldMemoryManager,
it is not a good idea: (1) remote execution doesn't need many parts of object file, such as debugging information, headers, section table and etc, (2) some sections is code and must has execution permit, some sections is data without execution permit, so the
best way is use a different methods for emits different type of sections.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">If your patch will be committed, we obtain a RuntimeDyldMachO and RuntimeDyldELF, which different principles and a different results. And most likely, for ELF
will never be fully realization for remote execution.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">I think another ways will be better, just for instance, two possible solutions:</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">1. Always emits the sections required to execution with a RTDyldMemoryManager. Resolve relocation in these sections. And if the isDebugging flag is set, then
in addition, DyldELFObject::rebaseObject make his job, except that for the sections stored by the RTDyldMemoryManager write the correct address.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">2. Check the debug flag at first, before load object file. If it set, just use your realization of loadObject, otherwise - use normal way with RTDyldMemoryManager.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Both alternatives have their shortcomings, perhaps you find another solution. Also, I would be glad to hear that Jim thinks about it.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Now I trying to merge your patch with my.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Regards,</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: navy; FONT-SIZE: 10pt">Danil</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<div style="TEXT-ALIGN: center" class="MsoNormal" align="center"><span style="COLOR: black" lang="RU">
<hr align="center" size="2" width="100%">
</span></div>
<div>
<p class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">From:</span></b><span class="apple-converted-space"><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU"> </span></span><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt" lang="RU">Kaylor,
Andrew [<a href="mailto:andrew.kaylor@intel.com">mailto:andrew.kaylor@intel.com</a>]<span class="apple-converted-space"> </span><br>
<b>Sent:</b><span class="apple-converted-space"> </span>Thursday, February 09, 2012 4:56 AM<br>
<b>To:</b><span class="apple-converted-space"> </span>Danil Malyshev; <a href="mailto:llvm-commits@cs.uiuc.edu">
llvm-commits@cs.uiuc.edu</a><br>
<b>Subject:</b><span class="apple-converted-space"> </span>RE: RuntimeDyLd new features</span><span style="COLOR: black" lang="RU"></span></p>
</div>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: black; FONT-SIZE: 11pt">Hi Danil,</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: black; FONT-SIZE: 11pt">Thanks for your efforts in this area. As Eli Bendersky mentioned, we have a patch out for review in this area also. I'm hopeful that we can find a convergence
between our code, your code and the code that Jim Grosbach has put in place.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: black; FONT-SIZE: 11pt">Unfortunately, we have been submitting our code in small chunks for ease of review and to keep things stable, and it may not be obvious from what we've put
out for review what our intentions were or our intended solutions to the problems that were left open. I'd like to take this opportunity to discuss the direction we we're heading and see how it might align with what you and Jim have done.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: black; FONT-SIZE: 11pt">Let me first explain what we have done, and then I'll offer specific comments on your code and possible next steps.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: black; FONT-SIZE: 11pt">We have had two primary goals: (1) to get MCJIT generated code to work correctly on Intel architecture and help pave the way for other implementations and
(2) enable source-level debugging of JITed code with GDB. This second goal seems to be dropping from view, but it places a few constraints on the eventual implementation. We've had GDB integration working, BTW.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: black; FONT-SIZE: 11pt">In order to get GDB to handle JITed code, we need to register an ELF object image through an interface GDB defines. As you might expect, GDB has some peculiar
expectations for what this ELF object image should look like. In particular, we need to set a flag in the ELF header and update the sh_addr members in the section headers to reflect the address where the section contents reside in memory.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: black; FONT-SIZE: 11pt">Our most recent patch (<a href="http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20120130/135997.html" target="_blank"><span style="COLOR: purple">http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20120130/135997.html</span></a>,
not yet committed) begins by copying the entire ELF image emitted by the MC code generator into an executable buffer. This was intended as a temporary step toward our eventual solution. It enabled us to perform relocations in-place on the object and execute
functions in place (thus eliminating an extra copy that was previously being done). We were in the process of implementing a smarter section-based approach, but Jim Grosbach was implementing a similar approach in parallel and our submission ended up appearing
out of step in this regard.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: black; FONT-SIZE: 11pt">So that's our background. Now, returning to your patch....</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: black; FONT-SIZE: 11pt">I like the idea of combining as much common code as possible into the RuntimeDyldImpl class. I'm interested to hear from users of the MachO loader if your
implementation has lost any of the specialization that they need. I think it's a promising approach. There are some ELF-specific details that we will need to have incorporated to re-enable GDB integration, but I expect that we'll be able to find a way to
work that in with a few well-placed overloaded function calls.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: black; FONT-SIZE: 11pt">I have some reservations about the use of the basic ObjectFile interface, which has some serious limitations. We've been working toward exposing the ELFObjectFile
template for use in the runtime loading process (as well as other unrelated uses). It may be that this is something that can be generalized enough to fit with your approach. My main concern in this regard is that we need to be able to update specific entries
in the ELF image, as described above.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: black; FONT-SIZE: 11pt">A related issue is that section loading can be refined with some ELF-specific details. Some sections need to have memory allocated for their contents. Other
sections can be left in place in the originally generated image. There is a good bit of unnecessary copying going on in the existing implementation, and I'm not clear to what extent your patch addresses that. Before the object is loaded, it is copied into
a new buffer and then the contents of each section are copied again as we go. What I'd like is for the runtime loaders to use the buffer into which the object is originally generated and only make copies where it is strictly necessary. This isn't necessarily
something you need to do for your work to be acceptable, but I mention it as a likely next step.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: black; FONT-SIZE: 11pt">Over the next few days I intend to apply your patch locally and try to merge our work into it. I'll provide additional feedback as I get a better feel for
what you've done.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Calibri','sans-serif'; COLOR: black; FONT-SIZE: 11pt">-Andy</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<div style="BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<div>
<p class="MsoNormal"><b><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt">From:</span></b><span class="apple-converted-space"><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt"> </span></span><span style="FONT-FAMILY: 'Tahoma','sans-serif'; COLOR: black; FONT-SIZE: 10pt"><a href="mailto:llvm-commits-bounces@cs.uiuc.edu">llvm-commits-bounces@cs.uiuc.edu</a>
[<a href="mailto:llvm-commits-bounces@cs.uiuc.edu">mailto:llvm-commits-bounces@cs.uiuc.edu</a>]<span class="apple-converted-space"> </span><b>On Behalf Of<span class="apple-converted-space"> </span></b>Danil Malyshev<br>
<b>Sent:</b><span class="apple-converted-space"> </span>Tuesday, February 07, 2012 12:24 PM<br>
<b>To:</b><span class="apple-converted-space"> </span><a href="mailto:llvm-commits@cs.uiuc.edu">llvm-commits@cs.uiuc.edu</a><br>
<b>Subject:</b><span class="apple-converted-space"> </span>[llvm-commits] RuntimeDyLd new features</span><span style="COLOR: black" lang="RU"></span></p>
</div>
</div>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: black; FONT-SIZE: 10pt">Hello everyone,</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: black; FONT-SIZE: 10pt">Please review the RuntimeDyLd-01.patch.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: black; FONT-SIZE: 10pt">This patch makes the following changes:</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: black; FONT-SIZE: 10pt">1. The main works will made in the RuntimeDyLdImpl with uses the ObjectFile class. RuntimeDyLdMachO and RuntimeDyLdELF now only parses relocations and resolve
it. This is allows to make improvements of the RuntimeDyLd more easily. In addition the support for COFF can be easily added.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: black; FONT-SIZE: 10pt">2. Added ARM relocations to RuntimeDyLdELF.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: black; FONT-SIZE: 10pt">3. Added support for stub functions for the ARM, allowing to do a long branch.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: black; FONT-SIZE: 10pt">4. Added support for external functions that are not loaded from the object files, but can be loaded from external libraries. Now MCJIT can correctly execute
the code containing the printf, putc, and etc.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: black; FONT-SIZE: 10pt">5. The sections emitted instead functions, thanks Jim Grosbach. MemoryManager.startFunctionBody() and MemoryManager.endFunctionBody() have been removed.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: black; FONT-SIZE: 10pt">6. MCJITMemoryManager.allocateDataSection() and MCJITMemoryManager. allocateCodeSection() used JMM->allocateSpace() instead of JMM->allocateCodeSection() and
JMM->allocateDataSection(), because I got an error: "Cannot allocate an allocated block!" with object file contains more than one code or data sections.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: black; FONT-SIZE: 10pt">7. Fixed ELF::R_X86_64_PC32 relocation for the case when RealOffset is negative value.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: black; FONT-SIZE: 10pt">8. Added new testing folder: ExecutionEngine/MCJIT because mcjit tests can be running only for x86 and arm and it's can be filtered with dg.exp.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: black; FONT-SIZE: 10pt">Tested in Ubuntu x86_64, Ubuntu armv7 and MacOS 64.</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: black; FONT-SIZE: 10pt">Thank you,</span><span style="COLOR: black" lang="RU"></span></p>
</div>
<div>
<p class="MsoNormal"><span style="FONT-FAMILY: 'Arial','sans-serif'; COLOR: black; FONT-SIZE: 10pt">Danil</span><span style="COLOR: black" lang="RU"></span></p>
</div>
</div>
</div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
</div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
</div>
</div>
</div>
<p class="MsoNormal"><span style="COLOR: black" lang="RU"></span> </p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</body>
</html>