[lldb] Reenable test HostTest.GetProcessInfo with relaxed constraints.

[feg208]

@jimingham I am wondering if you are ok removing this test? It caused failures in some of the build bots because the user time was less than a microsecond. Alternatively we can increase the number of loops or maybe I need some other approach? I had commented it out just to not impact others

[llvmbot]

@llvm/pr-subscribers-lldb

Author: Fred Grim (feg208)

Changes

@jimingham I am wondering if you are ok removing this test? It caused failures in some of the build bots because the user time was less than a microsecond. Alternatively we can increase the number of loops or maybe I need some other approach? I had commented it out just to not impact others

---

Full diff: https://github.com/llvm/llvm-project/pull/89637.diff

1 Files Affected:

• (modified) lldb/unittests/Host/linux/HostTest.cpp (-14)

diff --git a/lldb/unittests/Host/linux/HostTest.cpp b/lldb/unittests/Host/linux/HostTest.cpp
index 733909902474d7..5599e4349c8291 100644
--- a/lldb/unittests/Host/linux/HostTest.cpp
+++ b/lldb/unittests/Host/linux/HostTest.cpp
@@ -68,18 +68,4 @@ TEST_F(HostTest, GetProcessInfo) {
   EXPECT_TRUE(Info.GetArchitecture().IsValid());
   EXPECT_EQ(HostInfo::GetArchitecture(HostInfo::eArchKindDefault),
             Info.GetArchitecture());
-  // Test timings
-  /*
-   * This is flaky in the buildbots on all archs
-  ASSERT_TRUE(Host::GetProcessInfo(getpid(), Info));
-  ProcessInstanceInfo::timespec user_time = Info.GetUserTime();
-  static volatile unsigned u = 0;
-  for (unsigned i = 0; i < 10'000'000; i++) {
-    u = i;
-  }
-  ASSERT_TRUE(Host::GetProcessInfo(getpid(), Info));
-  ProcessInstanceInfo::timespec next_user_time = Info.GetUserTime();
-  ASSERT_TRUE(user_time.tv_sec < next_user_time.tv_sec ||
-              user_time.tv_usec < next_user_time.tv_usec);
-  */
 }

[JDevlieghere]

A potential compromise would be to check that the time is great or equal?

[feg208]

A potential compromise would be to check that the time is great or equal?

That's a good idea. I'll make that change and add a note

[felipepiovezan]

Hi @feg208 , I noticed that your previous commit has a very.. unusual commit title and message. The same thing applies to this PR.

I'd appreciate it if you could follow a more standard commit title and message, so that the logs are more easily parseable.
Here's the official LLVM docs on the matter: https://llvm.org/docs/DeveloperPolicy.html#commit-messages

You can also get a good sample by doing a git log --one-line inside the llvm repo.

[feg208]

Hi @feg208 , I noticed that your previous commit has a very.. unusual commit title and message. The same thing applies to this PR.

I'd appreciate it if you could follow a more standard commit title and message, so that the logs are more easily parseable. Here's the official LLVM docs on the matter: https://llvm.org/docs/DeveloperPolicy.html#commit-messages

You can also get a good sample by doing a git log --one-line inside the llvm repo.

Sure thing. I apologize for deviating from expectations.