[Git][ghc/ghc][wip/marge_bot_batch_merge_job] 4 commits: Fix SCC grouping example

[Marge Bot (@marge-bot)]

Marge Bot pushed to branch wip/marge_bot_batch_merge_job at Glasgow Haskell Compiler / GHC


Commits:
d0c7bbed by Tom Ellis at 2023-02-27T20:04:07-05:00
Fix SCC grouping example

- - - - -
f6caeca1 by Bryan Richter at 2023-02-28T03:08:31-05:00
Mark setnumcapabilities001 fragile

- - - - -
9dfbcf44 by Bryan Richter at 2023-02-28T03:08:31-05:00
Allow nightly-x86_64-linux-deb10-validate+thread_sanitizer to fail

See #22520

- - - - -
61a15e3c by Cheng Shao at 2023-02-28T03:08:32-05:00
ghc-prim: fix hs_cmpxchg64 function prototype

hs_cmpxchg64 must return a StgWord64, otherwise incorrect runtime
results of 64-bit MO_Cmpxchg will appear in 32-bit unregisterised
builds, which go unnoticed at compile-time due to C implicit casting
in .hc files.

- - - - -


6 changed files:

- .gitlab/gen_ci.hs
- .gitlab/jobs.yaml
- docs/users_guide/profiling.rst
- libraries/ghc-prim/cbits/atomic.c
- rts/include/stg/Prim.h
- testsuite/tests/concurrent/should_run/all.T


Changes:

=====================================
.gitlab/gen_ci.hs
=====================================
@@ -857,7 +857,9 @@ job_groups =
      , validateBuilds Amd64 (Linux Debian10) nativeInt
      , fastCI (validateBuilds Amd64 (Linux Debian10) unreg)
      , fastCI (validateBuilds Amd64 (Linux Debian10) debug)
-     , modifyValidateJobs manual tsan_jobs
+     , -- Nightly allowed to fail: #22520
+       modifyNightlyJobs allowFailure
+         (modifyValidateJobs manual tsan_jobs)
      , -- Nightly allowed to fail: #22343
        modifyNightlyJobs allowFailure
         (modifyValidateJobs manual (validateBuilds Amd64 (Linux Debian10) noTntc))


=====================================
.gitlab/jobs.yaml
=====================================
@@ -1333,7 +1333,7 @@
       ".gitlab/ci.sh clean",
       "cat ci_timings"
     ],
-    "allow_failure": false,
+    "allow_failure": true,
     "artifacts": {
       "expire_in": "8 weeks",
       "paths": [


=====================================
docs/users_guide/profiling.rst
=====================================
@@ -214,7 +214,7 @@ may not appear in a position where it would change the grouping of
 subexpressions::
 
   a = 1 / 2 / 2                          -- accepted (a=0.25)
-  b = 1 / {-# SCC "name" #-} / 2 / 2     -- rejected (instead of b=1.0)
+  b = 1 / {-# SCC "name" #-} 2 / 2       -- rejected (instead of b=1.0)
 
 This restriction is required to maintain the property that inserting a pragma,
 just like inserting a comment, does not have unintended effects on the


=====================================
libraries/ghc-prim/cbits/atomic.c
=====================================
@@ -308,8 +308,8 @@ hs_cmpxchg32(StgWord x, StgWord old, StgWord new)
   return __sync_val_compare_and_swap((volatile StgWord32 *) x, (StgWord32) old, (StgWord32) new);
 }
 
-extern StgWord hs_cmpxchg64(StgWord x, StgWord64 old, StgWord64 new);
-StgWord
+extern StgWord64 hs_cmpxchg64(StgWord x, StgWord64 old, StgWord64 new);
+StgWord64
 hs_cmpxchg64(StgWord x, StgWord64 old, StgWord64 new)
 {
   return __sync_val_compare_and_swap((volatile StgWord64 *) x, old, new);


=====================================
rts/include/stg/Prim.h
=====================================
@@ -41,7 +41,7 @@ StgWord64 hs_atomic_xor64(StgWord x, StgWord64 val);
 StgWord hs_cmpxchg8(StgWord x, StgWord old, StgWord new_);
 StgWord hs_cmpxchg16(StgWord x, StgWord old, StgWord new_);
 StgWord hs_cmpxchg32(StgWord x, StgWord old, StgWord new_);
-StgWord hs_cmpxchg64(StgWord x, StgWord64 old, StgWord64 new_);
+StgWord64 hs_cmpxchg64(StgWord x, StgWord64 old, StgWord64 new_);
 StgWord hs_atomicread8(StgWord x);
 StgWord hs_atomicread16(StgWord x);
 StgWord hs_atomicread32(StgWord x);


=====================================
testsuite/tests/concurrent/should_run/all.T
=====================================
@@ -241,7 +241,8 @@ test('setnumcapabilities001',
        extra_run_opts('8 12 2000'),
        when(have_thread_sanitizer(), expect_broken(18808)),
        req_target_smp,
-       req_ghc_smp
+       req_ghc_smp,
+       fragile(22989)
      ],
      compile_and_run, [''])
 



View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/38e4ecfac0514a1d6f01f8a45da61a06796dab51...61a15e3c6efa8f6419e73e2a68d3ddebd0bf814f

-- 
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/38e4ecfac0514a1d6f01f8a45da61a06796dab51...61a15e3c6efa8f6419e73e2a68d3ddebd0bf814f
You're receiving this email because of your account on gitlab.haskell.org.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/ghc-commits/attachments/20230228/e3f90ecc/attachment-0001.html>