[flang] Compile error instantiating std::clock_t on Apple clang version 11.0.0

[craig.rasmussen]

Summary

Apple clang gives error with initialization std::clock_t{-1}.

Error message is

/Users/rasmussen17/Compilers/llvm-project/flang/runtime/time-intrinsic.cpp:39:33: error: constant expression

    evaluates to -1 which cannot be narrowed to type 'std::clock_t' (aka 'unsigned long') [-Wc++11-narrowing]
if (timestamp != std::clock_t{-1}) {
                              ^~

---

Simple fix is to replace with a static cast: std::clock_t{static_cast<clock_t>(-1)}

[jhenderson]

It looks like you've got many changes in this diff that are unrelated? Have you rebased properly and generated your diff accordingly?

[craig.rasmussen]

Forgot to commit changes so patch/diff was not correct.

[craig.rasmussen]

Fixed patch and added reviewer.

[rovka]

flang/runtime/time-intrinsic.cpp

39

Can we simplify to this instead?

timestamp != static_cast<std::clock_t>(-1)

[craig.rasmussen]

Made simplification suggested by reviewer:

std::clock_t{static_cast<clock_t>(-1)} --> static_cast<clock_t>(-1)

[craig.rasmussen]

Requesting review (is there a better way to request a review)

[rovka]

LGTM, thanks!

[rovka]

In D105595#2894946, @craig.rasmussen wrote:

Requesting review (is there a better way to request a review)

I think people usually notice and have another look when you upload a new version of the patch. If there's no reaction, common practice is to ping about once a week.

[craig.rasmussen]

How do I ping? I have another pull request that is also in limbo.

Thanks,
Craig

From: Diana Picus via Phabricator <[e-mail]>
Date: Monday, July 26, 2021 at 8:45 AM
To: Rasmussen, Craig E. <[e-mail]>, [e-mail] <[e-mail]>, [e-mail] <[e-mail]>, [e-mail] <[e-mail]>
Cc: [e-mail] <[e-mail]>, [e-mail] <[e-mail]>, [e-mail] <[e-mail]>, [e-mail] <[e-mail]>, [e-mail] <[e-mail]>, [e-mail] <[e-mail]>, [e-mail] <[e-mail]>, [e-mail] <[e-mail]>, [e-mail] <[e-mail]>
Subject: [PATCH] D105595: [flang] Compile error instantiating std::clock_t on Apple clang version 11.0.0
rovka added a comment.

In D105595#2894946 <https://urldefense.us/v3/__https://reviews.llvm.org/D105595*2894946__;Iw!!G2kpM7uM-TzIFchu!j8TcPSbVUpRWmsqLEU_2NiAJlQkuOPOamifPnxDhHpBXqKbAyypxYlbd08Z4MU1Ca6WH$ >, @craig.rasmussen wrote:

Requesting review (is there a better way to request a review)

I think people usually notice and have another look when you upload a new version of the patch. If there's no reaction, common practice is to ping about once a week.

CHANGES SINCE LAST ACTION

https://urldefense.us/v3/__https://reviews.llvm.org/D105595/new/__;!!G2kpM7uM-TzIFchu!j8TcPSbVUpRWmsqLEU_2NiAJlQkuOPOamifPnxDhHpBXqKbAyypxYlbd08Z4MWbf3AVF$<https://urldefense.us/v3/__https:/reviews.llvm.org/D105595/new/__;!!G2kpM7uM-TzIFchu!j8TcPSbVUpRWmsqLEU_2NiAJlQkuOPOamifPnxDhHpBXqKbAyypxYlbd08Z4MWbf3AVF$>

https://urldefense.us/v3/__https://reviews.llvm.org/D105595__;!!G2kpM7uM-TzIFchu!j8TcPSbVUpRWmsqLEU_2NiAJlQkuOPOamifPnxDhHpBXqKbAyypxYlbd08Z4MfUDIKSg$https://urldefense.us/v3/__https:/reviews.llvm.org/D105595__;!!G2kpM7uM-TzIFchu!j8TcPSbVUpRWmsqLEU_2NiAJlQkuOPOamifPnxDhHpBXqKbAyypxYlbd08Z4MfUDIKSg$

[jhenderson]

In D105595#2905230, @craig.rasmussen wrote:

How do I ping? I have another pull request that is also in limbo.

Just literally post a comment saying "ping" or similar to grab people's attention.

[craig.rasmussen]

ping

[rovka]

Hi Craig,

I think there was some confusion related to this patch. I approved it long ago, which means that you could commit it upstream. See the docs explaining the review process. Since you didn't commit it or ask someone else to do it, it was fixed independently in the meantime. Unfortunately I was out of office during the time so I didn't catch it. In any case, thanks for the patch and next time please let people know during the review if you don't have commit rights.

flang/runtime/time-intrinsic.cpp

 // We will invoke our helper with an integer literal argument, so the overload
 // with the highest priority should have the type int.
 using fallback_implementation = double;
 using preferred_implementation = int;
 
 // This is the fallback implementation, which should work everywhere.
 template <typename Unused = void> double getCpuTime(fallback_implementation) {
   std::clock_t timestamp{std::clock()};
-  if (timestamp != std::clock_t{-1}) {
+  if (timestamp != static_cast<clock_t>(-1)) {
     return static_cast<double>(timestamp) / CLOCKS_PER_SEC;
   }
 
   // Return some negative value to represent failure.
   return -1.0;
 }
 
 // POSIX implementation using clock_gettime. This is only enabled if