<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<meta name="Generator" content="Microsoft Word 15 (filtered medium)">
<!--[if !mso]><style>v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
</style><![endif]--><style><!--
/* Font Definitions */
@font-face
{font-family:Helvetica;
panose-1:0 0 0 0 0 0 0 0 0 0;}
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
{font-family:Verdana;
panose-1:2 11 6 4 3 5 4 4 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri",sans-serif;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
span.apple-converted-space
{mso-style-name:apple-converted-space;}
span.EmailStyle19
{mso-style-type:personal-reply;
font-family:"Calibri",sans-serif;
color:windowtext;}
.MsoChpDefault
{mso-style-type:export-only;
font-size:10.0pt;}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang="EN-US" link="blue" vlink="purple">
<div class="WordSection1">
<p class="MsoNormal">Thanks for the reply.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Just picking out a few points:<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">> Module in the context of F18 means parser, semantics, evaluate etc.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Those are abstract things. What are the concrete things you want to capitalize? Namespaces? Directory names?<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">> But what we meant here was to make sure that std::string is not being used as a parameter to functions etc, these should be using StringRef (or possibly Twine where appropriate).<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">What's the benefit of using StringRef over std::string for parameters?<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Regarding "always use early exit":<o:p></o:p></p>
<p class="MsoNormal">> The LLVM Coding Standards have a strong preference for this, with justification given here:<o:p></o:p></p>
<p class="MsoNormal">> https://llvm.org/docs/CodingStandards.html#don-t-use-else-after-a-return<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">That's a pretty thin justification. It gives an example where else should not be used and generalizes it to all cases. When you encounter exceptional conditions early exit it certainly appropriate. But in non-exceptional cases it does not
always correctly express intent.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Here's an example from resolve-names.cpp:<o:p></o:p></p>
<p class="MsoNormal"><span style="font-family:"Courier New""> if (type.IsAssumedType()) {<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-family:"Courier New""> return currScope().MakeTypeStarType();<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-family:"Courier New""> } else if (type.IsUnlimitedPolymorphic()) {<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-family:"Courier New""> return currScope().MakeClassStarType();<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-family:"Courier New""> } else {<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-family:"Courier New""> return currScope().MakeDerivedType(...);<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-family:"Courier New""> }<o:p></o:p></span></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">You could write this as:<o:p></o:p></p>
<p class="MsoNormal"><span style="font-family:"Courier New""> if (type.IsAssumedType()) {<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-family:"Courier New""> return currScope().MakeTypeStarType();<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-family:"Courier New""> }<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-family:"Courier New""> if (type.IsUnlimitedPolymorphic()) {<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-family:"Courier New""> return currScope().MakeClassStarType();<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-family:"Courier New""> }<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-family:"Courier New""> return currScope().MakeDerivedType(...);<o:p></o:p></span></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">That makes it look like MakeDerivedType is the "normal" case and the other two are exceptions.<o:p></o:p></p>
<p class="MsoNormal">But that's wrong -- they are all normal and it's just a return based on a three-way condition.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Tim<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<div style="border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0in 0in 0in">
<p class="MsoNormal"><b><span style="font-size:12.0pt;color:black">From: </span></b><span style="font-size:12.0pt;color:black">David Truby <David.Truby@arm.com><br>
<b>Date: </b>Wednesday, January 29, 2020 at 6:51 AM<br>
<b>To: </b>Timothy Keith <tkeith@nvidia.com><br>
<b>Cc: </b>Richard Barton <Richard.Barton@arm.com>, "flang-dev@lists.llvm.org" <flang-dev@lists.llvm.org><br>
<b>Subject: </b>Re: [flang-dev] (Strawman) Costed plan for LLVM-ification of F18<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<table class="MsoNormalTable" border="1" cellpadding="0" style="background:#FFEB9C">
<tbody>
<tr>
<td style="padding:.75pt .75pt .75pt .75pt">
<p class="MsoNormal"><b><span style="font-size:7.5pt;font-family:"Verdana",sans-serif;color:black">External email: Use caution opening links or attachments</span></b><span style="font-size:7.5pt;font-family:"Verdana",sans-serif;color:black">
</span><o:p></o:p></p>
</td>
</tr>
</tbody>
</table>
<p class="MsoNormal"><o:p> </o:p></p>
<div>
<p class="MsoNormal">Hi Tim, <o:p></o:p></p>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">I had some input into putting this list together so I’ve added some clarifications to some things inline<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">Thanks<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">David Truby<o:p></o:p></p>
<div>
<p class="MsoNormal"><br>
<br>
<o:p></o:p></p>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal">On 29 Jan 2020, at 14:12, Timothy Keith via flang-dev <<a href="mailto:flang-dev@lists.llvm.org">flang-dev@lists.llvm.org</a>> wrote:<o:p></o:p></p>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
<div>
<div>
<p class="MsoNormal">My comments and questions are below in line.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">Tim<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> Hi list<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> <span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> As discussed on previous threads, we need to propose a plan for making F18 more LLVM-like in style and use of LLVM APIs and facilities. David is working on getting the permissions to put this into a github project in LLVM so we can collaborate
. In order to make progress ahead of that happening, I'd like to have a go on email.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> <span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> For each area I would like to capture a list of work items we'll commit to doing before merge to monorepo. I want us to be happy we can achieve these in a short time period so we can propose a new merging date. I'd also like to create
a list of items we'll commit to fixing after merging to the monorepo and a timeline for getting that done.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> <span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> Note that I'll only cover API and coding style work items in this thread. The other initiatives we can discuss in other threads. I'll just say that we'll propose cmake changes and porting test suite to lit in the pre-merge stuff and setting
up build bots in the post-merge stuff.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> <span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> When he has access, David can create two projects one for pre-merge and one for post-merge and we'll put everything in there.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> <span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> Please consider the below a strawman. All dates can change, all list items can change position, new ones can be added, items can be deleted, etc.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> <span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> What do people think about these lists? Do we think those two dates are do-able?<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> <span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> Ta<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> Rich<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> <span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> Proposal:<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> We will make the following style changes before merging to the monorepo<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> <span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> F18 changes to make it more LLVM-like in code style<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> 1. Rationalise headers to put public headers in /include and not /lib<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> 2. Unify F18's clang-format file to match LLVMs<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">We should discuss what should be in .clang-format with an eye to make it more like LLVM's.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> 3. Rename all .cc files to .cpp<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> 4. Switch module names to starting with capital letters<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">What does "module" mean in this context?<o:p></o:p></p>
</div>
</div>
</blockquote>
<p class="MsoNormal">Module in the context of F18 means parser, semantics, evaluate etc.<br>
<br>
<o:p></o:p></p>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> Increase use of LLVM APIs and utilities in F18<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> a. Switch F18 custom File handling to LLVM's File handling (helps with non-POSIX platform support)<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> b. Change uses of <iostream> to LLVM's streams<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">F18 doesn't use <iostream>.<o:p></o:p></p>
</div>
</div>
</blockquote>
<p class="MsoNormal">I think we are (perhaps ambiguously) using <iostream> here to refer to the entire standard stream IO library. E.g. I do see a lot of uses of <ostream> in F18. LLVM has its own replacement for these for various reasons and highly discourages
the use of the standard ones.<br>
<br>
<o:p></o:p></p>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> c. Migrate use of std::list to a suitable alternative in LLVM's API<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">I would say: replace std::list with a better data structure where appropriate.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> d. Use llvm_unreachable with an error message for unreachable cases<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">What does llvm_unreachable do for us?<o:p></o:p></p>
</div>
</div>
</blockquote>
<p class="MsoNormal">llvm_unreachable effectively behaves like an abort in debug builds, however in release builds it adds optimiser hints abstracted across the various supported compilers that the code in question isn’t reachable (so no need to generate a
branch there or whatever).<br>
<br>
<o:p></o:p></p>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> e. Use llvm::Error instead of error codes if and when error codes are used.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">I don't know of any use of error codes.<o:p></o:p></p>
</div>
</div>
</blockquote>
<p class="MsoNormal">What error handling facility does F18 currently use if error codes and exceptions aren’t used? I am familiar with a few places that optionals are returned, we could consider looking at changing these to using llvm::Expected to provide more
context about what has gone wrong.<br>
<br>
<o:p></o:p></p>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> We would like to aim for a merge date of Monday 24th Feb to merge to the monorepo with all of the above changes committed to F18 master.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> <span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> We then propose to make the following changes after merging to the monorepo.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> <span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> F18 changes to make it more LLVM-like in code style<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> We will commit to a perform a one-off exercise where we automatically audit the code to find these instances and bring them in line.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> 1. Eliminate braces from all single-line if statements<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">This seems like a really bad idea and would like to see a reason for it.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> 2. Eliminate all uses of else-after-return<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">Doing this blindly is also a bad idea.<o:p></o:p></p>
</div>
</div>
</blockquote>
<p class="MsoNormal">The LLVM Coding Standards have a strong preference for this, with justification given here:<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><a href="https://llvm.org/docs/CodingStandards.html#don-t-use-else-after-a-return">https://llvm.org/docs/CodingStandards.html#don-t-use-else-after-a-return</a><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal">> 3. Add doxygen infrastructure so we can generate docs<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> <span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> Increase use of LLVM APIs and utilities in F18<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> a. std::string → StringRef where appropriate<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">These aren't similar data structures. std::string owns the string data and StringRef does not. std::string_view is now a standard class for the latter case.<o:p></o:p></p>
</div>
</blockquote>
<p class="MsoNormal">LLVM still currently prefers the use of StringRef over std::string_view. I am not sure if there are plans to change this when LLVM switches the majority of its code to C++17. But what we meant here was to make sure that std::string is not
being used as a parameter to functions etc, these should be using StringRef (or possibly Twine where appropriate).<o:p></o:p></p>
</div>
<div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> b. std::vector → llvm::SmallVector where appropriate<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> c. std::set → llvm::SmallSet/llvm::StringSet/llvm::DenseSet where appropriate<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> d. std::map → llvm::StringMap/llvm::DenseMap where appropriate<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">As long as "where appropriate" means there is tangible benefit to changing from a standard type to a non-standard one. Not just change for the sake of change.<o:p></o:p></p>
</div>
</blockquote>
<p class="MsoNormal">This is indeed what we meant by where appropriate. The LLVM guidelines are perfectly happy for the use of standard data structures in general, however the LLVM specific ones are much better optimised for certain use cases. For example,
the DenseMap class in particular is much more efficient than std::map when the key and value are both small objects.<o:p></o:p></p>
</div>
<div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> e. Audit functions in include/flang/common and port to LLVM equivalents (e.g. builtin_popcount)<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> <span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> Assuming we hit the above merge date, we think we can commit to making these changes before the LLVM11 branch is taken in June.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> <span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> After that date, we will continue to make progress towards LLVM style and API usage by fixing things as we find them during development and enforce the new style through code review.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> <span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> A few specific areas that have been mentioned before that we will tackle in this way are:<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> - Add doxygen style comments and file comments<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> - Find more expressive names for classes, files, etc.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">Names are already supposed to be expressive. Improvements are always welcome.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">> - Refactor code to use early exits<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">Where it improves the code, not blindly. As with improving names, there is no reason to wait for any specific date to make improvements to the code.<o:p></o:p></p>
</div>
</blockquote>
<p class="MsoNormal">Again, early exists are something the LLVM Coding Standards have a strong preference for, with justification given here:<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><a href="https://llvm.org/docs/CodingStandards.html#early-exits">https://llvm.org/docs/CodingStandards.html#early-exits</a><br>
<br>
<o:p></o:p></p>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div class="MsoNormal" align="center" style="text-align:center"><span style="font-size:9.0pt;font-family:Helvetica">
<hr size="0" width="100%" align="center">
</span></div>
</div>
<div>
<p class="MsoNormal"><span style="font-size:9.0pt;font-family:Helvetica">This email message is for the sole use of the intended recipient(s) and may contain confidential information. Any unauthorized review, use, disclosure or distribution is prohibited.
If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message.<o:p></o:p></span></p>
</div>
<div>
<div class="MsoNormal" align="center" style="text-align:center"><span style="font-size:9.0pt;font-family:Helvetica">
<hr size="0" width="100%" align="center">
</span></div>
</div>
<p class="MsoNormal"><span style="font-size:9.0pt;font-family:Helvetica">_______________________________________________<br>
flang-dev mailing list<br>
</span><a href="mailto:flang-dev@lists.llvm.org"><span style="font-size:9.0pt;font-family:Helvetica">flang-dev@lists.llvm.org</span></a><span style="font-size:9.0pt;font-family:Helvetica"><br>
</span><a href="https://lists.llvm.org/cgi-bin/mailman/listinfo/flang-dev"><span style="font-size:9.0pt;font-family:Helvetica">https://lists.llvm.org/cgi-bin/mailman/listinfo/flang-dev</span></a><o:p></o:p></p>
</blockquote>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
</div>
</div>
</body>
</html>