[PATCH] D66981: Fix driver tests when `LLVM_ENABLE_PER_TARGET_RUNTIME_DIR` is `ON`

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb added a comment.

In D66981#1651897 , @phosek wrote:

> LGTM


Can you please commit this? I don't have commit access.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D66981/new/

https://reviews.llvm.org/D66981



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D66834: Driver tests: set `--sysroot=""` to support clang with `DEFAULT_SYSROOT`

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb added a comment.

In D66834#1685260 , @sepavloff wrote:

> In D66834#1653334 , @broadwaylamb 
> wrote:
>
> > In D66834#1652756 , @compnerd 
> > wrote:
> >
> > > I think that this is pretty easy to forget.  Fortunately, last argument 
> > > wins.  Why not sink this into the `%clang` substitution in lit?  That 
> > > ensures that we run with an empty sysroot and then when the test needs to 
> > > adjust the sysroot, it can do so explicitly.
> >
> >
> > I've just tried to do it, but unfortunately some tests are failing, for 
> > example, `Driver/cc1-response-files.c`. The problem is that `%clang` is 
> > expanded to `/path/to/clang --sysroot=`, but the succeeding flags (such as 
> > `-cc1`) may be incompatible with `--sysroot`.
>
>
> Does the issue manifests itself with `-cc1` only? We usually use `%clang_cc1` 
> in such cases, so probably those tests require update.


Some Driver tests take a list of arguments from a file (for example 
`clang/test/Driver/cc1-response-files.c`), so we probably cannot use 
`%clang_cc1` there


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D66834/new/

https://reviews.llvm.org/D66834



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D66834: Driver tests: set `--sysroot=""` to support clang with `DEFAULT_SYSROOT`

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb added a comment.

Can someone please commit this?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D66834/new/

https://reviews.llvm.org/D66834



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D83490: [cmake] Support running compiler-rt tests in CrossWinToARMLinux.cmake

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb added a comment.

In D83490#2177256 , @vvereschaka wrote:

> - looks like `ssh.py` arguments have been changed since the diff file was 
> placed. Currently the correct string for `COMPILER_RT_EMULATOR` is the 
> following: ``` set(COMPILER_RT_EMULATOR
> "\\\"${Python3_EXECUTABLE}\\\" 
> \\\"${LLVM_PROJECT_DIR}/libcxx/utils/ssh.py\\\" --execdir %%T --host 
> ${REMOTE_TEST_USER}@${REMOTE_TEST_HOST} %%t" CACHE STRING "") ```


The usage of this `ssh.py` syntax depends on D83429 
, which is kind of controversial and probably 
won't be accepted as is.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D83490/new/

https://reviews.llvm.org/D83490



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78027: [TimeProfiler] Emit real process ID and thread names

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb created this revision.
broadwaylamb added reviewers: anton-afanasyev, russell.gallop.
Herald added subscribers: cfe-commits, MaskRay, hiraditya, emaste.
Herald added a reviewer: espindola.
Herald added a project: clang.
broadwaylamb added a parent revision: D78022: Introduce 
llvm::sys::Process::getProcessId() and adopt it.
broadwaylamb updated this revision to Diff 256993.
broadwaylamb added a comment.

Remove unrelated changes


https://reviews.llvm.org/D78027

Files:
  clang/test/Driver/check-time-trace.cpp
  lld/test/ELF/time-trace.s
  llvm/lib/Support/TimeProfiler.cpp

Index: llvm/lib/Support/TimeProfiler.cpp
===
--- llvm/lib/Support/TimeProfiler.cpp
+++ llvm/lib/Support/TimeProfiler.cpp
@@ -15,6 +15,7 @@
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Support/JSON.h"
 #include "llvm/Support/Path.h"
+#include "llvm/Support/Process.h"
 #include "llvm/Support/Threading.h"
 #include 
 #include 
@@ -73,10 +74,18 @@
   }
 };
 
+static std::string getThreadName() {
+  SmallString<64> Name;
+  llvm::get_thread_name(Name);
+  return std::string(std::move(Name));
+}
+
 struct TimeTraceProfiler {
   TimeTraceProfiler(unsigned TimeTraceGranularity = 0, StringRef ProcName = "")
   : StartTime(steady_clock::now()), ProcName(ProcName),
-Tid(llvm::get_threadid()), TimeTraceGranularity(TimeTraceGranularity) {}
+Pid(sys::Process::getProcessId()), ThreadName(getThreadName()),
+Tid(llvm::get_threadid()),
+TimeTraceGranularity(TimeTraceGranularity) {}
 
   void begin(std::string Name, llvm::function_ref Detail) {
 Stack.emplace_back(steady_clock::now(), TimePointType(), std::move(Name),
@@ -141,7 +150,7 @@
   auto DurUs = E.getFlameGraphDurUs();
 
   J.object([&]{
-J.attribute("pid", 1);
+J.attribute("pid", Pid);
 J.attribute("tid", int64_t(Tid));
 J.attribute("ph", "X");
 J.attribute("ts", StartUs);
@@ -205,7 +214,7 @@
   auto Count = AllCountAndTotalPerName[Total.first].first;
 
   J.object([&]{
-J.attribute("pid", 1);
+J.attribute("pid", Pid);
 J.attribute("tid", int64_t(TotalTid));
 J.attribute("ph", "X");
 J.attribute("ts", 0);
@@ -220,16 +229,25 @@
   ++TotalTid;
 }
 
-// Emit metadata event with process name.
-J.object([&] {
-  J.attribute("cat", "");
-  J.attribute("pid", 1);
-  J.attribute("tid", 0);
-  J.attribute("ts", 0);
-  J.attribute("ph", "M");
-  J.attribute("name", "process_name");
-  J.attributeObject("args", [&] { J.attribute("name", ProcName); });
-});
+auto writeMetadataEvent = [&](const char* Name, uint64_t Tid, auto arg) {
+  J.object([&] {
+J.attribute("cat", "");
+J.attribute("pid", Pid);
+J.attribute("tid", int64_t(Tid));
+J.attribute("ts", 0);
+J.attribute("ph", "M");
+J.attribute("name", Name);
+J.attributeObject("args", [&] { J.attribute("name", arg); });
+  });
+};
+
+writeMetadataEvent("process_name", Tid, ProcName);
+
+writeMetadataEvent("thread_name", Tid, ThreadName);
+
+for (const auto &TTP : ThreadTimeTraceProfilerInstances) {
+  writeMetadataEvent("thread_name", TTP->Tid, TTP->ThreadName);
+}
 
 J.arrayEnd();
 J.attributeEnd();
@@ -241,6 +259,8 @@
   StringMap CountAndTotalPerName;
   const TimePointType StartTime;
   const std::string ProcName;
+  const sys::Process::Pid Pid;
+  const std::string ThreadName;
   const uint64_t Tid;
 
   // Minimum time granularity (in microseconds)
Index: lld/test/ELF/time-trace.s
===
--- lld/test/ELF/time-trace.s
+++ lld/test/ELF/time-trace.s
@@ -34,6 +34,7 @@
 # Check process_name entry field
 # CHECK: "name": "ld.lld{{(.exe)?}}"
 # CHECK: "name": "process_name"
+# CHECK: "name": "thread_name"
 
 .globl _start
 _start:
Index: clang/test/Driver/check-time-trace.cpp
===
--- clang/test/Driver/check-time-trace.cpp
+++ clang/test/Driver/check-time-trace.cpp
@@ -14,6 +14,7 @@
 // CHECK-NEXT: "ts":
 // CHECK: "name": "clang{{.*}}"
 // CHECK: "name": "process_name"
+// CHECK: "name": "thread_name"
 
 template 
 struct Struct {
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78027: [TimeProfiler] Emit real process ID and thread names

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb updated this revision to Diff 256993.
broadwaylamb added a comment.

Remove unrelated changes


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78027/new/

https://reviews.llvm.org/D78027

Files:
  clang/test/Driver/check-time-trace.cpp
  lld/test/ELF/time-trace.s
  llvm/lib/Support/TimeProfiler.cpp

Index: llvm/lib/Support/TimeProfiler.cpp
===
--- llvm/lib/Support/TimeProfiler.cpp
+++ llvm/lib/Support/TimeProfiler.cpp
@@ -15,6 +15,7 @@
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Support/JSON.h"
 #include "llvm/Support/Path.h"
+#include "llvm/Support/Process.h"
 #include "llvm/Support/Threading.h"
 #include 
 #include 
@@ -73,10 +74,18 @@
   }
 };
 
+static std::string getThreadName() {
+  SmallString<64> Name;
+  llvm::get_thread_name(Name);
+  return std::string(std::move(Name));
+}
+
 struct TimeTraceProfiler {
   TimeTraceProfiler(unsigned TimeTraceGranularity = 0, StringRef ProcName = "")
   : StartTime(steady_clock::now()), ProcName(ProcName),
-Tid(llvm::get_threadid()), TimeTraceGranularity(TimeTraceGranularity) {}
+Pid(sys::Process::getProcessId()), ThreadName(getThreadName()),
+Tid(llvm::get_threadid()),
+TimeTraceGranularity(TimeTraceGranularity) {}
 
   void begin(std::string Name, llvm::function_ref Detail) {
 Stack.emplace_back(steady_clock::now(), TimePointType(), std::move(Name),
@@ -141,7 +150,7 @@
   auto DurUs = E.getFlameGraphDurUs();
 
   J.object([&]{
-J.attribute("pid", 1);
+J.attribute("pid", Pid);
 J.attribute("tid", int64_t(Tid));
 J.attribute("ph", "X");
 J.attribute("ts", StartUs);
@@ -205,7 +214,7 @@
   auto Count = AllCountAndTotalPerName[Total.first].first;
 
   J.object([&]{
-J.attribute("pid", 1);
+J.attribute("pid", Pid);
 J.attribute("tid", int64_t(TotalTid));
 J.attribute("ph", "X");
 J.attribute("ts", 0);
@@ -220,16 +229,25 @@
   ++TotalTid;
 }
 
-// Emit metadata event with process name.
-J.object([&] {
-  J.attribute("cat", "");
-  J.attribute("pid", 1);
-  J.attribute("tid", 0);
-  J.attribute("ts", 0);
-  J.attribute("ph", "M");
-  J.attribute("name", "process_name");
-  J.attributeObject("args", [&] { J.attribute("name", ProcName); });
-});
+auto writeMetadataEvent = [&](const char* Name, uint64_t Tid, auto arg) {
+  J.object([&] {
+J.attribute("cat", "");
+J.attribute("pid", Pid);
+J.attribute("tid", int64_t(Tid));
+J.attribute("ts", 0);
+J.attribute("ph", "M");
+J.attribute("name", Name);
+J.attributeObject("args", [&] { J.attribute("name", arg); });
+  });
+};
+
+writeMetadataEvent("process_name", Tid, ProcName);
+
+writeMetadataEvent("thread_name", Tid, ThreadName);
+
+for (const auto &TTP : ThreadTimeTraceProfilerInstances) {
+  writeMetadataEvent("thread_name", TTP->Tid, TTP->ThreadName);
+}
 
 J.arrayEnd();
 J.attributeEnd();
@@ -241,6 +259,8 @@
   StringMap CountAndTotalPerName;
   const TimePointType StartTime;
   const std::string ProcName;
+  const sys::Process::Pid Pid;
+  const std::string ThreadName;
   const uint64_t Tid;
 
   // Minimum time granularity (in microseconds)
Index: lld/test/ELF/time-trace.s
===
--- lld/test/ELF/time-trace.s
+++ lld/test/ELF/time-trace.s
@@ -34,6 +34,7 @@
 # Check process_name entry field
 # CHECK: "name": "ld.lld{{(.exe)?}}"
 # CHECK: "name": "process_name"
+# CHECK: "name": "thread_name"
 
 .globl _start
 _start:
Index: clang/test/Driver/check-time-trace.cpp
===
--- clang/test/Driver/check-time-trace.cpp
+++ clang/test/Driver/check-time-trace.cpp
@@ -14,6 +14,7 @@
 // CHECK-NEXT: "ts":
 // CHECK: "name": "clang{{.*}}"
 // CHECK: "name": "process_name"
+// CHECK: "name": "thread_name"
 
 template 
 struct Struct {
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78030: [TimeProfiler] Emit clock synchronization point

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb created this revision.
broadwaylamb added reviewers: anton-afanasyev, russell.gallop, espindola.
Herald added subscribers: cfe-commits, MaskRay, hiraditya, emaste.
Herald added a project: clang.
broadwaylamb added a parent revision: D78027: [TimeProfiler] Emit real process 
ID and thread names.

`TimeProfiler` emits relative timestamps for events (the number of microseconds 
passed since the start of the current process).

This patch allows combining events from different processes while preserving 
their relative timing by emitting a new attribute `beginningOfTime`. This 
attribute contains the system time that corresponds to the zero timestamp of 
the time profiler.

This has at least two use cases:

- Build systems can use this to merge time traces from multiple compiler 
invocations and generate statistics for the whole build. Tools like 
ClangBuildAnalyzer  could also 
leverage this feature.
- Compilers that use LLVM as their backend by invoking `llc`/`opt` in a child 
process. I'm currently working on supporting the `-ftime-trace` functionality 
in GHC. A single GHC invocation can emit GHC-specific events, but with this 
patch it could also include LLVM-specific events in its log.

I wrote a proof-of-concept script that merges multiple logs that contain a 
synchronization point into one log: 
https://github.com/broadwaylamb/merge_trace_events

This is how the result looks like for GHC:

F11724922: 79080338-ad673580-7d1c-11ea-9e30-5e6f72e77555.png 



Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D78030

Files:
  clang/test/Driver/check-time-trace-sections.py
  clang/test/Driver/check-time-trace.cpp
  lld/test/ELF/time-trace.s
  llvm/lib/Support/TimeProfiler.cpp


Index: llvm/lib/Support/TimeProfiler.cpp
===
--- llvm/lib/Support/TimeProfiler.cpp
+++ llvm/lib/Support/TimeProfiler.cpp
@@ -251,6 +251,21 @@
 
 J.arrayEnd();
 J.attributeEnd();
+
+// Emit synchronization point, i. e. the absolute time of StartTime.
+// When combining time profiler logs from different processes,
+// this attribute helps preserve relative timing.
+{
+  const auto SystemTime =
+  time_point_cast(system_clock::now());
+  const microseconds ProcessLocalTime =
+  time_point_cast(steady_clock::now()) -
+  time_point_cast(StartTime);
+  const auto BeginningOfTimeUs = SystemTime - ProcessLocalTime;
+  J.attribute("beginningOfTime",
+  BeginningOfTimeUs.time_since_epoch().count());
+}
+
 J.objectEnd();
   }
 
Index: lld/test/ELF/time-trace.s
===
--- lld/test/ELF/time-trace.s
+++ lld/test/ELF/time-trace.s
@@ -18,7 +18,8 @@
 # RUN:   | %python -c 'import json, sys; 
json.dump(json.loads(sys.stdin.read()), sys.stdout, sort_keys=True, indent=2)' \
 # RUN:   | FileCheck %s
 
-# CHECK: "traceEvents": [
+# CHECK: "beginningOfTime": {{[0-9]{16},}}
+# CHECK-NEXT: "traceEvents": [
 
 # Check one event has correct fields
 # CHECK:  "dur":
Index: clang/test/Driver/check-time-trace.cpp
===
--- clang/test/Driver/check-time-trace.cpp
+++ clang/test/Driver/check-time-trace.cpp
@@ -3,7 +3,8 @@
 // RUN:   | %python -c 'import json, sys; 
json.dump(json.loads(sys.stdin.read()), sys.stdout, sort_keys=True, indent=2)' \
 // RUN:   | FileCheck %s
 
-// CHECK: "traceEvents": [
+// CHECK: "beginningOfTime": {{[0-9]{16},}}
+// CHECK-NEXT: "traceEvents": [
 // CHECK: "args":
 // CHECK: "detail":
 // CHECK: "dur":
Index: clang/test/Driver/check-time-trace-sections.py
===
--- clang/test/Driver/check-time-trace-sections.py
+++ clang/test/Driver/check-time-trace-sections.py
@@ -1,6 +1,6 @@
 #!/usr/bin/env python
 
-import json, sys
+import json, sys, time
 
 def is_inside(range1, range2):
 a = range1["ts"]; b = a + range1["dur"]
@@ -11,11 +11,20 @@
 b = range1["ts"] + range1["dur"]; c = range2["ts"]
 return b <= c
 
-events = json.loads(sys.stdin.read())["traceEvents"]
+log_contents = json.loads(sys.stdin.read())
+events = log_contents["traceEvents"]
 codegens = [event for event in events if event["name"] == "CodeGen Function"]
 frontends = [event for event in events if event["name"] == "Frontend"]
 backends = [event for event in events if event["name"] == "Backend"]
 
+beginning_of_time = log_contents["beginningOfTime"] / 100
+seconds_since_epoch = time.time()
+
+# Make sure that the 'beginningOfTime' is not earlier than 10 seconds ago.
+if seconds_since_epoch - beginning_of_time > 10:
+sys.exit("'beginningOfTime' should represent the absolute time when the "
+ "process has started")
+
 if not all([any([is_inside(codegen, frontend) for frontend in frontends])
   

[PATCH] D78027: [TimeProfiler] Emit real process ID and thread names

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb updated this revision to Diff 257015.
broadwaylamb added a comment.

Explicitly specify argument types in `writeMetadataEvent` lambda.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78027/new/

https://reviews.llvm.org/D78027

Files:
  clang/test/Driver/check-time-trace.cpp
  lld/test/ELF/time-trace.s
  llvm/lib/Support/TimeProfiler.cpp

Index: llvm/lib/Support/TimeProfiler.cpp
===
--- llvm/lib/Support/TimeProfiler.cpp
+++ llvm/lib/Support/TimeProfiler.cpp
@@ -15,6 +15,7 @@
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Support/JSON.h"
 #include "llvm/Support/Path.h"
+#include "llvm/Support/Process.h"
 #include "llvm/Support/Threading.h"
 #include 
 #include 
@@ -73,10 +74,18 @@
   }
 };
 
+static std::string getThreadName() {
+  SmallString<64> Name;
+  llvm::get_thread_name(Name);
+  return std::string(std::move(Name));
+}
+
 struct TimeTraceProfiler {
   TimeTraceProfiler(unsigned TimeTraceGranularity = 0, StringRef ProcName = "")
   : StartTime(steady_clock::now()), ProcName(ProcName),
-Tid(llvm::get_threadid()), TimeTraceGranularity(TimeTraceGranularity) {}
+Pid(sys::Process::getProcessId()), ThreadName(getThreadName()),
+Tid(llvm::get_threadid()),
+TimeTraceGranularity(TimeTraceGranularity) {}
 
   void begin(std::string Name, llvm::function_ref Detail) {
 Stack.emplace_back(steady_clock::now(), TimePointType(), std::move(Name),
@@ -141,7 +150,7 @@
   auto DurUs = E.getFlameGraphDurUs();
 
   J.object([&]{
-J.attribute("pid", 1);
+J.attribute("pid", Pid);
 J.attribute("tid", int64_t(Tid));
 J.attribute("ph", "X");
 J.attribute("ts", StartUs);
@@ -205,7 +214,7 @@
   auto Count = AllCountAndTotalPerName[Total.first].first;
 
   J.object([&]{
-J.attribute("pid", 1);
+J.attribute("pid", Pid);
 J.attribute("tid", int64_t(TotalTid));
 J.attribute("ph", "X");
 J.attribute("ts", 0);
@@ -220,16 +229,26 @@
   ++TotalTid;
 }
 
-// Emit metadata event with process name.
-J.object([&] {
-  J.attribute("cat", "");
-  J.attribute("pid", 1);
-  J.attribute("tid", 0);
-  J.attribute("ts", 0);
-  J.attribute("ph", "M");
-  J.attribute("name", "process_name");
-  J.attributeObject("args", [&] { J.attribute("name", ProcName); });
-});
+auto writeMetadataEvent = [&](const char *Name, uint64_t Tid,
+  StringRef arg) {
+  J.object([&] {
+J.attribute("cat", "");
+J.attribute("pid", Pid);
+J.attribute("tid", int64_t(Tid));
+J.attribute("ts", 0);
+J.attribute("ph", "M");
+J.attribute("name", Name);
+J.attributeObject("args", [&] { J.attribute("name", arg); });
+  });
+};
+
+writeMetadataEvent("process_name", Tid, ProcName);
+
+writeMetadataEvent("thread_name", Tid, ThreadName);
+
+for (const auto &TTP : ThreadTimeTraceProfilerInstances) {
+  writeMetadataEvent("thread_name", TTP->Tid, TTP->ThreadName);
+}
 
 J.arrayEnd();
 J.attributeEnd();
@@ -241,6 +260,8 @@
   StringMap CountAndTotalPerName;
   const TimePointType StartTime;
   const std::string ProcName;
+  const sys::Process::Pid Pid;
+  const std::string ThreadName;
   const uint64_t Tid;
 
   // Minimum time granularity (in microseconds)
Index: lld/test/ELF/time-trace.s
===
--- lld/test/ELF/time-trace.s
+++ lld/test/ELF/time-trace.s
@@ -34,6 +34,7 @@
 # Check process_name entry field
 # CHECK: "name": "ld.lld{{(.exe)?}}"
 # CHECK: "name": "process_name"
+# CHECK: "name": "thread_name"
 
 .globl _start
 _start:
Index: clang/test/Driver/check-time-trace.cpp
===
--- clang/test/Driver/check-time-trace.cpp
+++ clang/test/Driver/check-time-trace.cpp
@@ -14,6 +14,7 @@
 // CHECK-NEXT: "ts":
 // CHECK: "name": "clang{{.*}}"
 // CHECK: "name": "process_name"
+// CHECK: "name": "thread_name"
 
 template 
 struct Struct {
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78030: [TimeProfiler] Emit clock synchronization point

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb updated this revision to Diff 257340.
broadwaylamb added a comment.

Indent patterns in tests


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78030/new/

https://reviews.llvm.org/D78030

Files:
  clang/test/Driver/check-time-trace-sections.py
  clang/test/Driver/check-time-trace.cpp
  lld/test/ELF/time-trace.s
  llvm/lib/Support/TimeProfiler.cpp

Index: llvm/lib/Support/TimeProfiler.cpp
===
--- llvm/lib/Support/TimeProfiler.cpp
+++ llvm/lib/Support/TimeProfiler.cpp
@@ -252,6 +252,21 @@
 
 J.arrayEnd();
 J.attributeEnd();
+
+// Emit synchronization point, i. e. the absolute time of StartTime.
+// When combining time profiler logs from different processes,
+// this attribute helps preserve relative timing.
+{
+  const auto SystemTime =
+  time_point_cast(system_clock::now());
+  const microseconds ProcessLocalTime =
+  time_point_cast(steady_clock::now()) -
+  time_point_cast(StartTime);
+  const auto BeginningOfTimeUs = SystemTime - ProcessLocalTime;
+  J.attribute("beginningOfTime",
+  BeginningOfTimeUs.time_since_epoch().count());
+}
+
 J.objectEnd();
   }
 
Index: lld/test/ELF/time-trace.s
===
--- lld/test/ELF/time-trace.s
+++ lld/test/ELF/time-trace.s
@@ -18,7 +18,8 @@
 # RUN:   | %python -c 'import json, sys; json.dump(json.loads(sys.stdin.read()), sys.stdout, sort_keys=True, indent=2)' \
 # RUN:   | FileCheck %s
 
-# CHECK: "traceEvents": [
+# CHECK:  "beginningOfTime": {{[0-9]{16},}}
+# CHECK-NEXT: "traceEvents": [
 
 # Check one event has correct fields
 # CHECK:  "dur":
Index: clang/test/Driver/check-time-trace.cpp
===
--- clang/test/Driver/check-time-trace.cpp
+++ clang/test/Driver/check-time-trace.cpp
@@ -3,18 +3,19 @@
 // RUN:   | %python -c 'import json, sys; json.dump(json.loads(sys.stdin.read()), sys.stdout, sort_keys=True, indent=2)' \
 // RUN:   | FileCheck %s
 
-// CHECK: "traceEvents": [
-// CHECK: "args":
-// CHECK: "detail":
-// CHECK: "dur":
-// CHECK: "name":
+// CHECK:  "beginningOfTime": {{[0-9]{16},}}
+// CHECK-NEXT: "traceEvents": [
+// CHECK:  "args":
+// CHECK:  "detail":
+// CHECK:  "dur":
+// CHECK:  "name":
 // CHECK-NEXT: "ph":
 // CHECK-NEXT: "pid":
 // CHECK-NEXT: "tid":
 // CHECK-NEXT: "ts":
-// CHECK: "name": "clang{{.*}}"
-// CHECK: "name": "process_name"
-// CHECK: "name": "thread_name"
+// CHECK:  "name": "clang{{.*}}"
+// CHECK:  "name": "process_name"
+// CHECK:  "name": "thread_name"
 
 template 
 struct Struct {
Index: clang/test/Driver/check-time-trace-sections.py
===
--- clang/test/Driver/check-time-trace-sections.py
+++ clang/test/Driver/check-time-trace-sections.py
@@ -1,6 +1,6 @@
 #!/usr/bin/env python
 
-import json, sys
+import json, sys, time
 
 def is_inside(range1, range2):
 a = range1["ts"]; b = a + range1["dur"]
@@ -11,11 +11,20 @@
 b = range1["ts"] + range1["dur"]; c = range2["ts"]
 return b <= c
 
-events = json.loads(sys.stdin.read())["traceEvents"]
+log_contents = json.loads(sys.stdin.read())
+events = log_contents["traceEvents"]
 codegens = [event for event in events if event["name"] == "CodeGen Function"]
 frontends = [event for event in events if event["name"] == "Frontend"]
 backends = [event for event in events if event["name"] == "Backend"]
 
+beginning_of_time = log_contents["beginningOfTime"] / 100
+seconds_since_epoch = time.time()
+
+# Make sure that the 'beginningOfTime' is not earlier than 10 seconds ago.
+if seconds_since_epoch - beginning_of_time > 10:
+sys.exit("'beginningOfTime' should represent the absolute time when the "
+ "process has started")
+
 if not all([any([is_inside(codegen, frontend) for frontend in frontends])
 for codegen in codegens]):
 sys.exit("Not all CodeGen sections are inside any Frontend section!")
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78030: [TimeProfiler] Emit clock synchronization point

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb marked an inline comment as done.
broadwaylamb added a comment.

It'd be good if I had one more LGTM.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78030/new/

https://reviews.llvm.org/D78030



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78027: [TimeProfiler] Emit real process ID and thread names

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb marked 2 inline comments as done.
broadwaylamb added inline comments.



Comment at: llvm/lib/Support/TimeProfiler.cpp:77
 
+static std::string getThreadName() {
+  SmallString<64> Name;

MaskRay wrote:
> You can define ThreadName as `SmallString<0>` to avoid this function.
Then the `ThreadName` field will lose `const`-ness, and I will also need to 
call `llvm::get_thread_name` in the constructor body, not the member 
initialization list. Would that be OK?



Comment at: llvm/lib/Support/TimeProfiler.cpp:249
+
+for (const auto &TTP : ThreadTimeTraceProfilerInstances) {
+  writeMetadataEvent("thread_name", TTP->Tid, TTP->ThreadName);

MaskRay wrote:
> Delete `{}` for simple statements. In LLVM code, such `{}` is not common.
The code above uses `{}`.

According to http://llvm.org/docs/CodingStandards.html:

>If you are extending, enhancing, or bug fixing already implemented code, use 
>the style that is already being used so that the source is uniform and easy to 
>follow.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78027/new/

https://reviews.llvm.org/D78027



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78030: [TimeProfiler] Emit clock synchronization point

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb marked an inline comment as done.
broadwaylamb added inline comments.



Comment at: llvm/lib/Support/TimeProfiler.cpp:266
+  const auto BeginningOfTimeUs = SystemTime - ProcessLocalTime;
+  J.attribute("beginningOfTime",
+  BeginningOfTimeUs.time_since_epoch().count());

MaskRay wrote:
> What is the logic here?
> 
> Two events in the same process may have different `beginningOfTime`. Is that 
> intended?
> 
> Noe that `system_clock::now()` and `steady_clock::now()` are not measured at 
> the same time.
>Two events in the same process may have different beginningOfTime. Is that 
>intended?

How would that be possible? We only compute `beginningOfTime` when writing the 
data of the time profiler to a JSON file. AFAICT it happens only once. So, this 
code is only called once.

>Noe that system_clock::now() and steady_clock::now() are not measured at the 
>same time.
I understand. But the difference should be very small (certainly less than a 
microsecond), since there are only a few arithmetic operations in between.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78030/new/

https://reviews.llvm.org/D78030



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78030: [TimeProfiler] Emit clock synchronization point

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb marked an inline comment as done.
broadwaylamb added inline comments.



Comment at: llvm/lib/Support/TimeProfiler.cpp:266
+  const auto BeginningOfTimeUs = SystemTime - ProcessLocalTime;
+  J.attribute("beginningOfTime",
+  BeginningOfTimeUs.time_since_epoch().count());

broadwaylamb wrote:
> MaskRay wrote:
> > What is the logic here?
> > 
> > Two events in the same process may have different `beginningOfTime`. Is 
> > that intended?
> > 
> > Noe that `system_clock::now()` and `steady_clock::now()` are not measured 
> > at the same time.
> >Two events in the same process may have different beginningOfTime. Is that 
> >intended?
> 
> How would that be possible? We only compute `beginningOfTime` when writing 
> the data of the time profiler to a JSON file. AFAICT it happens only once. 
> So, this code is only called once.
> 
> >Noe that system_clock::now() and steady_clock::now() are not measured at the 
> >same time.
> I understand. But the difference should be very small (certainly less than a 
> microsecond), since there are only a few arithmetic operations in between.
Although we could set the beginning of time in `TimeProfiler`'s constructor and 
assign it to a field. That would be a lot simpler, I guess. What do you think?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78030/new/

https://reviews.llvm.org/D78030



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D83490: [cmake] Support running compiler-rt tests in CrossWinToARMLinux.cmake

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb created this revision.
broadwaylamb added a reviewer: vvereschaka.
Herald added subscribers: cfe-commits, danielkiss, kristof.beyls, mgorny, 
dberris.
Herald added a project: clang.

Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D83490

Files:
  clang/cmake/caches/CrossWinToARMLinux.cmake


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -72,6 +72,7 @@
 set(COMPILER_RT_BUILD_LIBFUZZER OFF CACHE BOOL "")
 set(COMPILER_RT_BUILD_PROFILE   OFF CACHE BOOL "")
 set(COMPILER_RT_DEFAULT_TARGET_ONLY ON CACHE BOOL "")
+set(COMPILER_RT_INCLUDE_TESTS   ON CACHE BOOL "")
 
 set(LIBUNWIND_USE_COMPILER_RT   ON CACHE BOOL "")
 set(LIBUNWIND_TARGET_TRIPLE "${CMAKE_C_COMPILER_TARGET}" CACHE 
STRING "")
@@ -108,6 +109,10 @@
   set(DEFAULT_TEST_TARGET_INFO  
"libcxx.test.target_info.LinuxRemoteTI")
 
   # Allow override with the custom values.
+  if(NOT DEFINED COMPILER_RT_EMULATOR)
+set(COMPILER_RT_EMULATOR"\\\"${Python3_EXECUTABLE}\\\" 
\\\"${LLVM_PROJECT_DIR}/libcxx/utils/ssh.py\\\" --execdir %%T --test-executable 
%%t --host ${REMOTE_TEST_USER}@${REMOTE_TEST_HOST} --" CACHE STRING "")
+  endif()
+  
   if(NOT DEFINED LIBUNWIND_TARGET_INFO)
 set(LIBUNWIND_TARGET_INFO   "${DEFAULT_TEST_TARGET_INFO}" 
CACHE STRING "")
   endif()


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -72,6 +72,7 @@
 set(COMPILER_RT_BUILD_LIBFUZZER OFF CACHE BOOL "")
 set(COMPILER_RT_BUILD_PROFILE   OFF CACHE BOOL "")
 set(COMPILER_RT_DEFAULT_TARGET_ONLY ON CACHE BOOL "")
+set(COMPILER_RT_INCLUDE_TESTS   ON CACHE BOOL "")
 
 set(LIBUNWIND_USE_COMPILER_RT   ON CACHE BOOL "")
 set(LIBUNWIND_TARGET_TRIPLE "${CMAKE_C_COMPILER_TARGET}" CACHE STRING "")
@@ -108,6 +109,10 @@
   set(DEFAULT_TEST_TARGET_INFO  "libcxx.test.target_info.LinuxRemoteTI")
 
   # Allow override with the custom values.
+  if(NOT DEFINED COMPILER_RT_EMULATOR)
+set(COMPILER_RT_EMULATOR"\\\"${Python3_EXECUTABLE}\\\" \\\"${LLVM_PROJECT_DIR}/libcxx/utils/ssh.py\\\" --execdir %%T --test-executable %%t --host ${REMOTE_TEST_USER}@${REMOTE_TEST_HOST} --" CACHE STRING "")
+  endif()
+  
   if(NOT DEFINED LIBUNWIND_TARGET_INFO)
 set(LIBUNWIND_TARGET_INFO   "${DEFAULT_TEST_TARGET_INFO}" CACHE STRING "")
   endif()
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78030: [TimeProfiler] Emit clock synchronization point

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb marked an inline comment as done.
broadwaylamb added inline comments.



Comment at: llvm/lib/Support/TimeProfiler.cpp:266
+  const auto BeginningOfTimeUs = SystemTime - ProcessLocalTime;
+  J.attribute("beginningOfTime",
+  BeginningOfTimeUs.time_since_epoch().count());

MaskRay wrote:
> broadwaylamb wrote:
> > broadwaylamb wrote:
> > > MaskRay wrote:
> > > > What is the logic here?
> > > > 
> > > > Two events in the same process may have different `beginningOfTime`. Is 
> > > > that intended?
> > > > 
> > > > Noe that `system_clock::now()` and `steady_clock::now()` are not 
> > > > measured at the same time.
> > > >Two events in the same process may have different beginningOfTime. Is 
> > > >that intended?
> > > 
> > > How would that be possible? We only compute `beginningOfTime` when 
> > > writing the data of the time profiler to a JSON file. AFAICT it happens 
> > > only once. So, this code is only called once.
> > > 
> > > >Noe that system_clock::now() and steady_clock::now() are not measured at 
> > > >the same time.
> > > I understand. But the difference should be very small (certainly less 
> > > than a microsecond), since there are only a few arithmetic operations in 
> > > between.
> > Although we could set the beginning of time in `TimeProfiler`'s constructor 
> > and assign it to a field. That would be a lot simpler, I guess. What do you 
> > think?
> OK, but I am still confused by the logic and the comment. Does the code just 
> want to compute the start time in system_clock?
Exactly, yes.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78030/new/

https://reviews.llvm.org/D78030



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78027: [TimeProfiler] Emit real process ID and thread names

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb updated this revision to Diff 258059.
broadwaylamb added a comment.

Fix coding style


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78027/new/

https://reviews.llvm.org/D78027

Files:
  clang/test/Driver/check-time-trace.cpp
  lld/test/ELF/time-trace.s
  llvm/lib/Support/TimeProfiler.cpp

Index: llvm/lib/Support/TimeProfiler.cpp
===
--- llvm/lib/Support/TimeProfiler.cpp
+++ llvm/lib/Support/TimeProfiler.cpp
@@ -16,6 +16,7 @@
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Support/JSON.h"
 #include "llvm/Support/Path.h"
+#include "llvm/Support/Process.h"
 #include "llvm/Support/Threading.h"
 #include 
 #include 
@@ -75,7 +76,10 @@
 struct llvm::TimeTraceProfiler {
   TimeTraceProfiler(unsigned TimeTraceGranularity = 0, StringRef ProcName = "")
   : StartTime(steady_clock::now()), ProcName(ProcName),
-Tid(llvm::get_threadid()), TimeTraceGranularity(TimeTraceGranularity) {}
+Pid(sys::Process::getProcessId()), Tid(llvm::get_threadid()),
+TimeTraceGranularity(TimeTraceGranularity) {
+llvm::get_thread_name(ThreadName);
+  }
 
   void begin(std::string Name, llvm::function_ref Detail) {
 Stack.emplace_back(steady_clock::now(), TimePointType(), std::move(Name),
@@ -138,8 +142,8 @@
   auto StartUs = E.getFlameGraphStartUs(StartTime);
   auto DurUs = E.getFlameGraphDurUs();
 
-  J.object([&]{
-J.attribute("pid", 1);
+  J.object([&] {
+J.attribute("pid", Pid);
 J.attribute("tid", int64_t(Tid));
 J.attribute("ph", "X");
 J.attribute("ts", StartUs);
@@ -194,8 +198,8 @@
   auto DurUs = duration_cast(Total.second.second).count();
   auto Count = AllCountAndTotalPerName[Total.first].first;
 
-  J.object([&]{
-J.attribute("pid", 1);
+  J.object([&] {
+J.attribute("pid", Pid);
 J.attribute("tid", int64_t(TotalTid));
 J.attribute("ph", "X");
 J.attribute("ts", 0);
@@ -210,16 +214,23 @@
   ++TotalTid;
 }
 
-// Emit metadata event with process name.
-J.object([&] {
-  J.attribute("cat", "");
-  J.attribute("pid", 1);
-  J.attribute("tid", 0);
-  J.attribute("ts", 0);
-  J.attribute("ph", "M");
-  J.attribute("name", "process_name");
-  J.attributeObject("args", [&] { J.attribute("name", ProcName); });
-});
+auto writeMetadataEvent = [&](const char *Name, uint64_t Tid,
+  StringRef arg) {
+  J.object([&] {
+J.attribute("cat", "");
+J.attribute("pid", Pid);
+J.attribute("tid", int64_t(Tid));
+J.attribute("ts", 0);
+J.attribute("ph", "M");
+J.attribute("name", Name);
+J.attributeObject("args", [&] { J.attribute("name", arg); });
+  });
+};
+
+writeMetadataEvent("process_name", Tid, ProcName);
+writeMetadataEvent("thread_name", Tid, ThreadName);
+for (const TimeTraceProfiler *TTP : ThreadTimeTraceProfilerInstances)
+  writeMetadataEvent("thread_name", TTP->Tid, TTP->ThreadName);
 
 J.arrayEnd();
 J.attributeEnd();
@@ -231,6 +242,8 @@
   StringMap CountAndTotalPerName;
   const TimePointType StartTime;
   const std::string ProcName;
+  const sys::Process::Pid Pid;
+  SmallString<0> ThreadName;
   const uint64_t Tid;
 
   // Minimum time granularity (in microseconds)
Index: lld/test/ELF/time-trace.s
===
--- lld/test/ELF/time-trace.s
+++ lld/test/ELF/time-trace.s
@@ -34,6 +34,7 @@
 # Check process_name entry field
 # CHECK: "name": "ld.lld{{(.exe)?}}"
 # CHECK: "name": "process_name"
+# CHECK: "name": "thread_name"
 
 .globl _start
 _start:
Index: clang/test/Driver/check-time-trace.cpp
===
--- clang/test/Driver/check-time-trace.cpp
+++ clang/test/Driver/check-time-trace.cpp
@@ -14,6 +14,7 @@
 // CHECK-NEXT: "ts":
 // CHECK: "name": "clang{{.*}}"
 // CHECK: "name": "process_name"
+// CHECK: "name": "thread_name"
 
 template 
 struct Struct {
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78030: [TimeProfiler] Emit clock synchronization point

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb updated this revision to Diff 258061.
broadwaylamb added a comment.

A simpler generation of the beginning of time.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78030/new/

https://reviews.llvm.org/D78030

Files:
  clang/test/Driver/check-time-trace-sections.py
  clang/test/Driver/check-time-trace.cpp
  lld/test/ELF/time-trace.s
  llvm/lib/Support/TimeProfiler.cpp

Index: llvm/lib/Support/TimeProfiler.cpp
===
--- llvm/lib/Support/TimeProfiler.cpp
+++ llvm/lib/Support/TimeProfiler.cpp
@@ -75,9 +75,9 @@
 
 struct llvm::TimeTraceProfiler {
   TimeTraceProfiler(unsigned TimeTraceGranularity = 0, StringRef ProcName = "")
-  : StartTime(steady_clock::now()), ProcName(ProcName),
-Pid(sys::Process::getProcessId()), Tid(llvm::get_threadid()),
-TimeTraceGranularity(TimeTraceGranularity) {
+  : BeginningOfTime(system_clock::now()), StartTime(steady_clock::now()),
+ProcName(ProcName), Pid(sys::Process::getProcessId()),
+Tid(llvm::get_threadid()), TimeTraceGranularity(TimeTraceGranularity) {
 llvm::get_thread_name(ThreadName);
   }
 
@@ -234,12 +234,22 @@
 
 J.arrayEnd();
 J.attributeEnd();
+
+// Emit the absolute time of the moment when this TimeProfiler started
+// measurements. This can be used to combine the profiling data from
+// multiple processes and preserve actual time intervals.
+J.attribute("beginningOfTime",
+time_point_cast(BeginningOfTime)
+.time_since_epoch()
+.count());
+
 J.objectEnd();
   }
 
   SmallVector Stack;
   SmallVector Entries;
   StringMap CountAndTotalPerName;
+  const time_point BeginningOfTime;
   const TimePointType StartTime;
   const std::string ProcName;
   const sys::Process::Pid Pid;
Index: lld/test/ELF/time-trace.s
===
--- lld/test/ELF/time-trace.s
+++ lld/test/ELF/time-trace.s
@@ -18,7 +18,8 @@
 # RUN:   | %python -c 'import json, sys; json.dump(json.loads(sys.stdin.read()), sys.stdout, sort_keys=True, indent=2)' \
 # RUN:   | FileCheck %s
 
-# CHECK: "traceEvents": [
+# CHECK:  "beginningOfTime": {{[0-9]{16},}}
+# CHECK-NEXT: "traceEvents": [
 
 # Check one event has correct fields
 # CHECK:  "dur":
Index: clang/test/Driver/check-time-trace.cpp
===
--- clang/test/Driver/check-time-trace.cpp
+++ clang/test/Driver/check-time-trace.cpp
@@ -3,18 +3,19 @@
 // RUN:   | %python -c 'import json, sys; json.dump(json.loads(sys.stdin.read()), sys.stdout, sort_keys=True, indent=2)' \
 // RUN:   | FileCheck %s
 
-// CHECK: "traceEvents": [
-// CHECK: "args":
-// CHECK: "detail":
-// CHECK: "dur":
-// CHECK: "name":
+// CHECK:  "beginningOfTime": {{[0-9]{16},}}
+// CHECK-NEXT: "traceEvents": [
+// CHECK:  "args":
+// CHECK:  "detail":
+// CHECK:  "dur":
+// CHECK:  "name":
 // CHECK-NEXT: "ph":
 // CHECK-NEXT: "pid":
 // CHECK-NEXT: "tid":
 // CHECK-NEXT: "ts":
-// CHECK: "name": "clang{{.*}}"
-// CHECK: "name": "process_name"
-// CHECK: "name": "thread_name"
+// CHECK:  "name": "clang{{.*}}"
+// CHECK:  "name": "process_name"
+// CHECK:  "name": "thread_name"
 
 template 
 struct Struct {
Index: clang/test/Driver/check-time-trace-sections.py
===
--- clang/test/Driver/check-time-trace-sections.py
+++ clang/test/Driver/check-time-trace-sections.py
@@ -1,6 +1,6 @@
 #!/usr/bin/env python
 
-import json, sys
+import json, sys, time
 
 def is_inside(range1, range2):
 a = range1["ts"]; b = a + range1["dur"]
@@ -11,11 +11,20 @@
 b = range1["ts"] + range1["dur"]; c = range2["ts"]
 return b <= c
 
-events = json.loads(sys.stdin.read())["traceEvents"]
+log_contents = json.loads(sys.stdin.read())
+events = log_contents["traceEvents"]
 codegens = [event for event in events if event["name"] == "CodeGen Function"]
 frontends = [event for event in events if event["name"] == "Frontend"]
 backends = [event for event in events if event["name"] == "Backend"]
 
+beginning_of_time = log_contents["beginningOfTime"] / 100
+seconds_since_epoch = time.time()
+
+# Make sure that the 'beginningOfTime' is not earlier than 10 seconds ago.
+if seconds_since_epoch - beginning_of_time > 10:
+sys.exit("'beginningOfTime' should represent the absolute time when the "
+ "process has started")
+
 if not all([any([is_inside(codegen, frontend) for frontend in frontends])
 for codegen in codegens]):
 sys.exit("Not all CodeGen sections are inside any Frontend section!")
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78030: [TimeProfiler] Emit clock synchronization point

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb added a comment.

Since we're adding a new attribute here that we won't ever be able to remove in 
the future, I'd still like to get a LGTM from some of the folks that are 
familiar with the time profiler. @russell.gallop @anton-afanasyev?


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78030/new/

https://reviews.llvm.org/D78030



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78027: [TimeProfiler] Emit real process ID and thread names

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb marked an inline comment as done.
broadwaylamb added inline comments.



Comment at: llvm/lib/Support/TimeProfiler.cpp:222
+J.attribute("pid", Pid);
+J.attribute("tid", int64_t(Tid));
+J.attribute("ts", 0);

MaskRay wrote:
> This is not great but I guess it is ok...
What is?


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78027/new/

https://reviews.llvm.org/D78027



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78404: [clang] Implement P0692R1 from C++20 (access checking on specializations)

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb created this revision.
broadwaylamb added reviewers: asl, rsmith, doug.gregor, rjmccall, triton.
broadwaylamb added a project: clang.
Herald added subscribers: cfe-commits, dexonsmith.

This patch implements paper P0692R1 
 from the 
C++20 standard.

This also fixes a bug (https://llvm.org/PR37424) where explicit instantiations 
of templates parameterized by overloaded private member functions were 
incorrectly rejected.

This is my first contribution to CFE, so please let me know if I did something 
horribly wrong.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D78404

Files:
  clang/lib/Parse/ParseDeclCXX.cpp
  clang/test/CXX/temp/temp.decls/temp.class.spec/p10.cpp
  clang/test/CXX/temp/temp.spec/p6.cpp
  clang/test/CXX/temp/temp.spec/temp.explicit/p11.cpp
  clang/www/cxx_status.html

Index: clang/www/cxx_status.html
===
--- clang/www/cxx_status.html
+++ clang/www/cxx_status.html
@@ -966,7 +966,7 @@
 
   Access checking on specializations
   https://wg21.link/p0692r1";>P0692R1
-  Partial
+  Clang 11
 
 
   Default constructible and assignable stateless lambdas
Index: clang/test/CXX/temp/temp.spec/temp.explicit/p11.cpp
===
--- clang/test/CXX/temp/temp.spec/temp.explicit/p11.cpp
+++ /dev/null
@@ -1,19 +0,0 @@
-// RUN: %clang_cc1 -fsyntax-only -verify %s
-// expected-no-diagnostics
-
-class X {
-  template  class Y {};
-};
-
-class A {
-  class B {};
-  class C {};
-};
-
-// C++0x [temp.explicit] 14.7.2/11:
-//   The usual access checking rules do not apply to names used to specify
-//   explicit instantiations.
-template class X::Y;
-
-// As an extension, this rule is applied to explicit specializations as well.
-template <> class X::Y {};
Index: clang/test/CXX/temp/temp.spec/p6.cpp
===
--- /dev/null
+++ clang/test/CXX/temp/temp.spec/p6.cpp
@@ -0,0 +1,44 @@
+// RUN: %clang_cc1 -fsyntax-only -verify %s
+// expected-no-diagnostics
+
+class X {
+  template  class Y {};
+};
+
+class A {
+  class B {};
+  class C {};
+
+  void func();
+  static void staticFunc();
+
+  // See https://llvm.org/PR37424
+  void funcOverloaded();
+  void funcOverloaded(int);
+  static void staticFuncOverloaded();
+  static void staticFuncOverloaded(int);
+
+  int field;
+};
+
+// C++20 [temp.spec] 17.8/6:
+//   The usual access checking rules do not apply to names in a declaration of
+//   an explicit instantiation or explicit specialization, with the exception
+//   of names appearing in a function body, default argument, base-clause,
+//   member-specification, enumerator-list, or static data member or variable
+//   template initializer.
+template class X::Y;
+
+template  class D {};
+template class D<&A::func>;
+template class D<&A::funcOverloaded>;
+
+template  class E { };
+template class E<&A::staticFunc>;
+template class E<&A::staticFuncOverloaded>;
+
+template  class G {};
+template class G<&A::field>;
+
+// As an extension, this rule is applied to explicit specializations as well.
+template <> class X::Y {};
Index: clang/test/CXX/temp/temp.decls/temp.class.spec/p10.cpp
===
--- /dev/null
+++ clang/test/CXX/temp/temp.decls/temp.class.spec/p10.cpp
@@ -0,0 +1,61 @@
+// RUN: %clang_cc1 -fsyntax-only -verify %s
+// expected-no-diagnostics
+
+// C++20 [temp.class.spec] 17.6.5/10:
+//   The usual access checking rules do not apply to non-dependent names used
+//   to specify template arguments of the simple-template-id of the partial
+//   specialization.
+
+class TestClass {
+private:
+  void func();
+  void funcOverloaded();
+  void funcOverloaded(int);
+
+  static void staticFunc();
+  static void staticFuncOverloaded();
+  static void staticFuncOverloaded(int);
+
+  class Nested {};
+
+  int field;
+};
+
+template  class TemplateClass {};
+template <> class TemplateClass<&TestClass::func> {};
+template <> class TemplateClass<&TestClass::funcOverloaded> {};
+
+template  class TemplateClass2 { };
+template <> class TemplateClass2<&TestClass::staticFunc> {};
+template <> class TemplateClass2<&TestClass::staticFuncOverloaded> {};
+
+template class TemplateClass3 {};
+template class TemplateClass3 {};
+template class TemplateClass3 {};
+
+template class TemplateClass4 {};
+template class TemplateClass4 {};
+template class TemplateClass4 {};
+
+template class TemplateClass5 {};
+template<> class TemplateClass5 {};
+
+template class TemplateClass6 {};
+template class TemplateClass6 {};
+
+template  class TemplateClass7 {};
+template <> class TemplateClass7<&TestClass::field> {};
+
+template  class TemplateClass8 {};
+template  class TemplateClass8 {};
+
+template
+struct trait;
+
+class class_ {
+  template
+  struct impl;
+};
+
+template
+struct trai

[PATCH] D78390: [dfsan] Add "DataFlow" option to LLVM_USE_SANITIZER

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb added inline comments.



Comment at: libcxx/utils/libcxx/test/config.py:914
+self.cxx.flags += ['-fsanitize=dataflow']
+self.config.available_features.add('dfsan')
 else:

I'm not sure we need a new feature if none of the tests actually use the 
feature.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78390/new/

https://reviews.llvm.org/D78390



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78404: [clang] Implement P0692R1 from C++20 (access checking on specializations)

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb added a comment.

In D78404#1990192 , @rsmith wrote:

> ... please also test ...
>
>   template class TemplateClass3 
> varTemplate3{};
>
>
> ... which we should diagnose, because that's a primary variable template 
> definition, not a partial / explicit specialization or explicit instantiation.


Interestingly, the latest GCC doesn't diagnose here 
, but we do. Do we need to remain compatible with 
GCC in this case?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78404/new/

https://reviews.llvm.org/D78404



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78404: [clang] Implement P0692R1 from C++20 (access checking on specializations)

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb updated this revision to Diff 258555.
broadwaylamb added a comment.

- Add more tests
- Allow class template member explicit specializations
- Inherit TypeAliasDecl from DeclContext (this is needed so that we could 
perform access checks when parsing 'using' declaration templates)


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78404/new/

https://reviews.llvm.org/D78404

Files:
  clang/include/clang/AST/Decl.h
  clang/include/clang/AST/DeclBase.h
  clang/include/clang/Basic/DeclNodes.td
  clang/lib/AST/DeclBase.cpp
  clang/lib/Parse/ParseDecl.cpp
  clang/lib/Parse/ParseDeclCXX.cpp
  clang/lib/Parse/ParseTemplate.cpp
  clang/test/CXX/drs/dr1xx.cpp
  clang/test/CXX/temp/temp.decls/temp.class.spec/p10.cpp
  clang/test/CXX/temp/temp.spec/p6.cpp
  clang/test/CXX/temp/temp.spec/temp.explicit/p11.cpp
  clang/www/cxx_status.html

Index: clang/www/cxx_status.html
===
--- clang/www/cxx_status.html
+++ clang/www/cxx_status.html
@@ -966,7 +966,7 @@
 
   Access checking on specializations
   https://wg21.link/p0692r1";>P0692R1
-  Partial
+  Clang 11
 
 
   Default constructible and assignable stateless lambdas
Index: clang/test/CXX/temp/temp.spec/temp.explicit/p11.cpp
===
--- clang/test/CXX/temp/temp.spec/temp.explicit/p11.cpp
+++ /dev/null
@@ -1,19 +0,0 @@
-// RUN: %clang_cc1 -fsyntax-only -verify %s
-// expected-no-diagnostics
-
-class X {
-  template  class Y {};
-};
-
-class A {
-  class B {};
-  class C {};
-};
-
-// C++0x [temp.explicit] 14.7.2/11:
-//   The usual access checking rules do not apply to names used to specify
-//   explicit instantiations.
-template class X::Y;
-
-// As an extension, this rule is applied to explicit specializations as well.
-template <> class X::Y {};
Index: clang/test/CXX/temp/temp.spec/p6.cpp
===
--- /dev/null
+++ clang/test/CXX/temp/temp.spec/p6.cpp
@@ -0,0 +1,71 @@
+// RUN: %clang_cc1 -fsyntax-only -verify %s
+// expected-no-diagnostics
+
+class X {
+  template  class Y {};
+};
+
+class A {
+  class B {};
+  class C {};
+
+  void func();
+  static void staticFunc();
+
+  // See https://llvm.org/PR37424
+  void funcOverloaded();
+  void funcOverloaded(int);
+  static void staticFuncOverloaded();
+  static void staticFuncOverloaded(int);
+
+  int field;
+};
+
+// C++20 [temp.spec] 17.8/6:
+//   The usual access checking rules do not apply to names in a declaration of
+//   an explicit instantiation or explicit specialization, with the exception
+//   of names appearing in a function body, default argument, base-clause,
+//   member-specification, enumerator-list, or static data member or variable
+//   template initializer.
+template class X::Y;
+
+template  class D {};
+template class D<&A::func>;
+template class D<&A::funcOverloaded>;
+
+template  class E { };
+template class E<&A::staticFunc>;
+template class E<&A::staticFuncOverloaded>;
+
+template  class G {};
+template class G<&A::field>;
+
+template <> class X::Y {};
+
+namespace member_spec {
+
+  template 
+  struct X {
+struct A {};
+void f();
+enum E : int;
+static int var;
+  };
+
+  class Y {
+using Z = int;
+  };
+
+  template <>
+  struct X::A {};
+
+  template <>
+  void X::f() {}
+
+  template <>
+  enum X::E : int {};
+
+  template <>
+  int X::var = 76;
+
+}
Index: clang/test/CXX/temp/temp.decls/temp.class.spec/p10.cpp
===
--- /dev/null
+++ clang/test/CXX/temp/temp.decls/temp.class.spec/p10.cpp
@@ -0,0 +1,119 @@
+// RUN: %clang_cc1 -fsyntax-only -verify %s
+
+// C++20 [temp.class.spec] 17.6.5/10:
+//   The usual access checking rules do not apply to non-dependent names used
+//   to specify template arguments of the simple-template-id of the partial
+//   specialization.
+
+class TestClass {
+  // expected-note@+1 4 {{declared private here}}
+  void func();
+
+  // expected-note@+1 4 {{declared private here}}
+  void funcOverloaded();
+
+  void funcOverloaded(int);
+
+  // expected-note@+1 2 {{declared private here}}
+  static void staticFunc();
+
+  // expected-note@+1 2 {{declared private here}}
+  static void staticFuncOverloaded();
+
+  static void staticFuncOverloaded(int);
+
+  // expected-note@+1 {{declared private here}}
+  class Nested {};
+
+  // expected-note@+1 {{declared private here}}
+  int field;
+};
+
+template  class TemplateClass {};
+template <> class TemplateClass<&TestClass::func> {};
+template <> class TemplateClass<&TestClass::funcOverloaded> {};
+
+// expected-error@+1 {{'func' is a private member of 'TestClass'}}
+using alias1_1 = TemplateClass<&TestClass::func>;
+
+// expected-error@+1 {{'funcOverloaded' is a private member of 'TestClass'}}
+using alias1_2 = TemplateClass<&TestClass::funcOverloaded>;
+
+template  class TemplateClass2 { };
+template <> class Temp

[PATCH] D78404: [clang] Implement P0692R1 from C++20 (access checking on specializations)

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb marked 3 inline comments as done.
broadwaylamb added inline comments.



Comment at: clang/include/clang/AST/Decl.h:3198
 /// alias-declaration.
-class TypeAliasDecl : public TypedefNameDecl {
+class TypeAliasDecl : public TypedefNameDecl, public DeclContext {
   /// The template for which this is the pattern, if any.

I'm not sure about inheriting `TypeAliasDecl` from `DeclContext`, but (see 
below)



Comment at: clang/lib/Parse/ParseTemplate.cpp:211
+Decl *Decl = usingDeclPtr.get().getSingleDecl();
+ParsingDeclRAII.complete(Decl);
+return Decl;

…but otherwise I couldn't make it print access level diagnostics for a 
particular kind of `using` declaration template (see in the next inline 
comment).

Why? Because of [this 
line](https://github.com/llvm/llvm-project/blob/8e0c9e21bf5f3e7a427b07e3eaf3bc80d2c74cb6/clang/lib/Sema/SemaAccess.cpp#L1479)
 — we need to be able to cast the `TypeAliasDecl` to `DeclContext` in order for 
delayed access check to be actually performed.



Comment at: clang/test/CXX/temp/temp.decls/temp.class.spec/p10.cpp:58
+template 
+using alias3_1 = TemplateClass3;
+

I'm talking about declarations like this.

Previously, we didn't reject it (which I believe was incorrect), and now we do.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78404/new/

https://reviews.llvm.org/D78404



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78404: [clang] Implement P0692R1 from C++20 (access checking on specializations)

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb marked an inline comment as done.
broadwaylamb added inline comments.



Comment at: clang/lib/Parse/ParseDecl.cpp:5672
+  //   initializer.
+  SuppressAccessChecks diagsFromTag(*this);
+

This is for things like

```
template<> void X::f() {}
```

not to be rejected (here `Z` is a private member of class `Y`)

I wasn't sure how to suppress it only when we're parsing template parameter 
list, so we suppress it unconditionally here. All the tests pass though, but 
I'd appreciate any hints.

Note that testing that `D.getContext() == 
DeclaratorContext::TemplateParamContext` doesn't work — when we get here, we're 
actually in a `FileContext`. 


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78404/new/

https://reviews.llvm.org/D78404



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78030: [TimeProfiler] Emit clock synchronization point

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb added a comment.

ping


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78030/new/

https://reviews.llvm.org/D78030



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78030: [TimeProfiler] Emit clock synchronization point

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb added a comment.

@russell.gallop thanks! I'll address your comments and commit the change.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78030/new/

https://reviews.llvm.org/D78030



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78027: [TimeProfiler] Emit real process ID and thread names

[Sergej Jaskiewicz via Phabricator via cfe-commits]

This revision was automatically updated to reflect the committed changes.
Closed by commit rGa5bf02815d8b: [TimeProfiler] Emit real process ID and thread 
names (authored by broadwaylamb).

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78027/new/

https://reviews.llvm.org/D78027

Files:
  clang/test/Driver/check-time-trace.cpp
  lld/test/ELF/time-trace.s
  llvm/lib/Support/TimeProfiler.cpp

Index: llvm/lib/Support/TimeProfiler.cpp
===
--- llvm/lib/Support/TimeProfiler.cpp
+++ llvm/lib/Support/TimeProfiler.cpp
@@ -16,6 +16,7 @@
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Support/JSON.h"
 #include "llvm/Support/Path.h"
+#include "llvm/Support/Process.h"
 #include "llvm/Support/Threading.h"
 #include 
 #include 
@@ -75,7 +76,10 @@
 struct llvm::TimeTraceProfiler {
   TimeTraceProfiler(unsigned TimeTraceGranularity = 0, StringRef ProcName = "")
   : StartTime(steady_clock::now()), ProcName(ProcName),
-Tid(llvm::get_threadid()), TimeTraceGranularity(TimeTraceGranularity) {}
+Pid(sys::Process::getProcessId()), Tid(llvm::get_threadid()),
+TimeTraceGranularity(TimeTraceGranularity) {
+llvm::get_thread_name(ThreadName);
+  }
 
   void begin(std::string Name, llvm::function_ref Detail) {
 Stack.emplace_back(steady_clock::now(), TimePointType(), std::move(Name),
@@ -138,8 +142,8 @@
   auto StartUs = E.getFlameGraphStartUs(StartTime);
   auto DurUs = E.getFlameGraphDurUs();
 
-  J.object([&]{
-J.attribute("pid", 1);
+  J.object([&] {
+J.attribute("pid", Pid);
 J.attribute("tid", int64_t(Tid));
 J.attribute("ph", "X");
 J.attribute("ts", StartUs);
@@ -194,8 +198,8 @@
   auto DurUs = duration_cast(Total.second.second).count();
   auto Count = AllCountAndTotalPerName[Total.first].first;
 
-  J.object([&]{
-J.attribute("pid", 1);
+  J.object([&] {
+J.attribute("pid", Pid);
 J.attribute("tid", int64_t(TotalTid));
 J.attribute("ph", "X");
 J.attribute("ts", 0);
@@ -210,16 +214,23 @@
   ++TotalTid;
 }
 
-// Emit metadata event with process name.
-J.object([&] {
-  J.attribute("cat", "");
-  J.attribute("pid", 1);
-  J.attribute("tid", 0);
-  J.attribute("ts", 0);
-  J.attribute("ph", "M");
-  J.attribute("name", "process_name");
-  J.attributeObject("args", [&] { J.attribute("name", ProcName); });
-});
+auto writeMetadataEvent = [&](const char *Name, uint64_t Tid,
+  StringRef arg) {
+  J.object([&] {
+J.attribute("cat", "");
+J.attribute("pid", Pid);
+J.attribute("tid", int64_t(Tid));
+J.attribute("ts", 0);
+J.attribute("ph", "M");
+J.attribute("name", Name);
+J.attributeObject("args", [&] { J.attribute("name", arg); });
+  });
+};
+
+writeMetadataEvent("process_name", Tid, ProcName);
+writeMetadataEvent("thread_name", Tid, ThreadName);
+for (const TimeTraceProfiler *TTP : ThreadTimeTraceProfilerInstances)
+  writeMetadataEvent("thread_name", TTP->Tid, TTP->ThreadName);
 
 J.arrayEnd();
 J.attributeEnd();
@@ -231,6 +242,8 @@
   StringMap CountAndTotalPerName;
   const TimePointType StartTime;
   const std::string ProcName;
+  const sys::Process::Pid Pid;
+  SmallString<0> ThreadName;
   const uint64_t Tid;
 
   // Minimum time granularity (in microseconds)
Index: lld/test/ELF/time-trace.s
===
--- lld/test/ELF/time-trace.s
+++ lld/test/ELF/time-trace.s
@@ -34,6 +34,7 @@
 # Check process_name entry field
 # CHECK: "name": "ld.lld{{(.exe)?}}"
 # CHECK: "name": "process_name"
+# CHECK: "name": "thread_name"
 
 .globl _start
 _start:
Index: clang/test/Driver/check-time-trace.cpp
===
--- clang/test/Driver/check-time-trace.cpp
+++ clang/test/Driver/check-time-trace.cpp
@@ -14,6 +14,7 @@
 // CHECK-NEXT: "ts":
 // CHECK: "name": "clang{{.*}}"
 // CHECK: "name": "process_name"
+// CHECK: "name": "thread_name"
 
 template 
 struct Struct {
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D78030: [TimeProfiler] Emit clock synchronization point

[Sergej Jaskiewicz via Phabricator via cfe-commits]

This revision was automatically updated to reflect the committed changes.
Closed by commit rG2899103108d3: [TimeProfiler] Emit clock synchronization 
point (authored by broadwaylamb).

Changed prior to commit:
  https://reviews.llvm.org/D78030?vs=258061&id=259413#toc

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78030/new/

https://reviews.llvm.org/D78030

Files:
  clang/test/Driver/check-time-trace-sections.py
  clang/test/Driver/check-time-trace.cpp
  lld/test/ELF/time-trace.s
  llvm/lib/Support/TimeProfiler.cpp

Index: llvm/lib/Support/TimeProfiler.cpp
===
--- llvm/lib/Support/TimeProfiler.cpp
+++ llvm/lib/Support/TimeProfiler.cpp
@@ -75,9 +75,9 @@
 
 struct llvm::TimeTraceProfiler {
   TimeTraceProfiler(unsigned TimeTraceGranularity = 0, StringRef ProcName = "")
-  : StartTime(steady_clock::now()), ProcName(ProcName),
-Pid(sys::Process::getProcessId()), Tid(llvm::get_threadid()),
-TimeTraceGranularity(TimeTraceGranularity) {
+  : BeginningOfTime(system_clock::now()), StartTime(steady_clock::now()),
+ProcName(ProcName), Pid(sys::Process::getProcessId()),
+Tid(llvm::get_threadid()), TimeTraceGranularity(TimeTraceGranularity) {
 llvm::get_thread_name(ThreadName);
   }
 
@@ -234,12 +234,22 @@
 
 J.arrayEnd();
 J.attributeEnd();
+
+// Emit the absolute time when this TimeProfiler started.
+// This can be used to combine the profiling data from
+// multiple processes and preserve actual time intervals.
+J.attribute("beginningOfTime",
+time_point_cast(BeginningOfTime)
+.time_since_epoch()
+.count());
+
 J.objectEnd();
   }
 
   SmallVector Stack;
   SmallVector Entries;
   StringMap CountAndTotalPerName;
+  const time_point BeginningOfTime;
   const TimePointType StartTime;
   const std::string ProcName;
   const sys::Process::Pid Pid;
Index: lld/test/ELF/time-trace.s
===
--- lld/test/ELF/time-trace.s
+++ lld/test/ELF/time-trace.s
@@ -18,7 +18,8 @@
 # RUN:   | %python -c 'import json, sys; json.dump(json.loads(sys.stdin.read()), sys.stdout, sort_keys=True, indent=2)' \
 # RUN:   | FileCheck %s
 
-# CHECK: "traceEvents": [
+# CHECK:  "beginningOfTime": {{[0-9]{16},}}
+# CHECK-NEXT: "traceEvents": [
 
 # Check one event has correct fields
 # CHECK:  "dur":
Index: clang/test/Driver/check-time-trace.cpp
===
--- clang/test/Driver/check-time-trace.cpp
+++ clang/test/Driver/check-time-trace.cpp
@@ -3,18 +3,19 @@
 // RUN:   | %python -c 'import json, sys; json.dump(json.loads(sys.stdin.read()), sys.stdout, sort_keys=True, indent=2)' \
 // RUN:   | FileCheck %s
 
-// CHECK: "traceEvents": [
-// CHECK: "args":
-// CHECK: "detail":
-// CHECK: "dur":
-// CHECK: "name":
+// CHECK:  "beginningOfTime": {{[0-9]{16},}}
+// CHECK-NEXT: "traceEvents": [
+// CHECK:  "args":
+// CHECK:  "detail":
+// CHECK:  "dur":
+// CHECK:  "name":
 // CHECK-NEXT: "ph":
 // CHECK-NEXT: "pid":
 // CHECK-NEXT: "tid":
 // CHECK-NEXT: "ts":
-// CHECK: "name": "clang{{.*}}"
-// CHECK: "name": "process_name"
-// CHECK: "name": "thread_name"
+// CHECK:  "name": "clang{{.*}}"
+// CHECK:  "name": "process_name"
+// CHECK:  "name": "thread_name"
 
 template 
 struct Struct {
Index: clang/test/Driver/check-time-trace-sections.py
===
--- clang/test/Driver/check-time-trace-sections.py
+++ clang/test/Driver/check-time-trace-sections.py
@@ -1,6 +1,6 @@
 #!/usr/bin/env python
 
-import json, sys
+import json, sys, time
 
 def is_inside(range1, range2):
 a = range1["ts"]; b = a + range1["dur"]
@@ -11,11 +11,22 @@
 b = range1["ts"] + range1["dur"]; c = range2["ts"]
 return b <= c
 
-events = json.loads(sys.stdin.read())["traceEvents"]
+log_contents = json.loads(sys.stdin.read())
+events = log_contents["traceEvents"]
 codegens = [event for event in events if event["name"] == "CodeGen Function"]
 frontends = [event for event in events if event["name"] == "Frontend"]
 backends = [event for event in events if event["name"] == "Backend"]
 
+beginning_of_time = log_contents["beginningOfTime"] / 100
+seconds_since_epoch = time.time()
+
+# Make sure that the 'beginningOfTime' is not earlier than 10 seconds ago
+# and not later than now.
+if beginning_of_time > seconds_since_epoch or \
+seconds_since_epoch - beginning_of_time > 10:
+sys.exit("'beginningOfTime' should represent the absolute time when the "
+ "process has started")
+
 if not all([any([is_inside(codegen, frontend) for frontend in frontends])
 for codegen in codegens]):
 sys.exit("Not all CodeGen sections are inside any Frontend section!")
__

[PATCH] D66834: Driver tests: set `--sysroot` to "" to support toolchains with default sysroot

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb created this revision.
broadwaylamb added reviewers: rsmith, sfantao, sepavloff.
Herald added a project: clang.
Herald added a subscriber: cfe-commits.

When testing clang that has been compiled with `-DDEFAULT_SYSROOT` set to some 
path,
the test `VFSGCCInstallation` would fail, because the search always starts 
either
in the default sysroot (which doesn't exist in the just-created in-memory VFS), 
or in
`${CLANG_BINARY_DIR}/../lib`, while the virtual GCC installation is in
`${CLANG_BINARY_DIR}/../usr/lib`.

Override sysroot to be empty string for the test to succeed.

(Note that a very similar test `VFSGCCInstallationRelativeDir` was passing 
because
the GCC installation was, in fact, in `${CLANG_BINARY_DIR}/../lib`.)


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D66834

Files:
  clang/unittests/Driver/ToolChainTest.cpp


Index: clang/unittests/Driver/ToolChainTest.cpp
===
--- clang/unittests/Driver/ToolChainTest.cpp
+++ clang/unittests/Driver/ToolChainTest.cpp
@@ -60,7 +60,7 @@
 llvm::MemoryBuffer::getMemBuffer("\n"));
 
   std::unique_ptr C(TheDriver.BuildCompilation(
-  {"-fsyntax-only", "--gcc-toolchain=", "foo.cpp"}));
+  {"-fsyntax-only", "--gcc-toolchain=", "--sysroot=", "foo.cpp"}));
   EXPECT_TRUE(C);
 
   std::string S;


Index: clang/unittests/Driver/ToolChainTest.cpp
===
--- clang/unittests/Driver/ToolChainTest.cpp
+++ clang/unittests/Driver/ToolChainTest.cpp
@@ -60,7 +60,7 @@
 llvm::MemoryBuffer::getMemBuffer("\n"));
 
   std::unique_ptr C(TheDriver.BuildCompilation(
-  {"-fsyntax-only", "--gcc-toolchain=", "foo.cpp"}));
+  {"-fsyntax-only", "--gcc-toolchain=", "--sysroot=", "foo.cpp"}));
   EXPECT_TRUE(C);
 
   std::string S;
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D66834: Driver tests: set `--sysroot` to "" to support toolchains with default sysroot

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb updated this revision to Diff 217987.
broadwaylamb added a comment.
Herald added subscribers: atanasyan, jrtc27.

Add `--sysroot=""` to some driver regression tests

These tests otherwise fail if clang is configured with DEFAULT_SYSROOT.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D66834/new/

https://reviews.llvm.org/D66834

Files:
  clang/test/Driver/darwin-sdkroot.c
  clang/test/Driver/gcc-toolchain.cpp
  clang/test/Driver/mips-mti.cpp
  clang/test/Driver/msp430-toolchain.c
  clang/test/Driver/print-libgcc-file-name-clangrt.c
  clang/unittests/Driver/ToolChainTest.cpp

Index: clang/unittests/Driver/ToolChainTest.cpp
===
--- clang/unittests/Driver/ToolChainTest.cpp
+++ clang/unittests/Driver/ToolChainTest.cpp
@@ -60,7 +60,7 @@
 llvm::MemoryBuffer::getMemBuffer("\n"));
 
   std::unique_ptr C(TheDriver.BuildCompilation(
-  {"-fsyntax-only", "--gcc-toolchain=", "foo.cpp"}));
+  {"-fsyntax-only", "--gcc-toolchain=", "--sysroot=", "foo.cpp"}));
   EXPECT_TRUE(C);
 
   std::string S;
Index: clang/test/Driver/print-libgcc-file-name-clangrt.c
===
--- clang/test/Driver/print-libgcc-file-name-clangrt.c
+++ clang/test/Driver/print-libgcc-file-name-clangrt.c
@@ -2,11 +2,13 @@
 
 // RUN: %clang -rtlib=compiler-rt -print-libgcc-file-name 2>&1 \
 // RUN: --target=x86_64-pc-linux \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
 // RUN:   | FileCheck --check-prefix=CHECK-CLANGRT-X8664 %s
 // CHECK-CLANGRT-X8664: libclang_rt.builtins-x86_64.a
 
 // RUN: %clang -rtlib=compiler-rt -print-libgcc-file-name 2>&1 \
 // RUN: --target=i386-pc-linux \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
 // RUN:   | FileCheck --check-prefix=CHECK-CLANGRT-I386 %s
 // CHECK-CLANGRT-I386: libclang_rt.builtins-i386.a
 
@@ -14,24 +16,29 @@
 //
 // RUN: %clang -rtlib=compiler-rt -print-libgcc-file-name 2>&1 \
 // RUN: --target=i686-pc-linux \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
 // RUN:   | FileCheck --check-prefix=CHECK-CLANGRT-I386 %s
 
 // RUN: %clang -rtlib=compiler-rt -print-libgcc-file-name 2>&1 \
 // RUN: --target=arm-linux-gnueabi \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
 // RUN:   | FileCheck --check-prefix=CHECK-CLANGRT-ARM %s
 // CHECK-CLANGRT-ARM: libclang_rt.builtins-arm.a
 
 // RUN: %clang -rtlib=compiler-rt -print-libgcc-file-name 2>&1 \
 // RUN: --target=arm-linux-androideabi \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
 // RUN:   | FileCheck --check-prefix=CHECK-CLANGRT-ARM-ANDROID %s
 // CHECK-CLANGRT-ARM-ANDROID: libclang_rt.builtins-arm-android.a
 
 // RUN: %clang -rtlib=compiler-rt -print-libgcc-file-name 2>&1 \
 // RUN: --target=arm-linux-gnueabihf \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
 // RUN:   | FileCheck --check-prefix=CHECK-CLANGRT-ARMHF %s
 // CHECK-CLANGRT-ARMHF: libclang_rt.builtins-armhf.a
 
 // RUN: %clang -rtlib=compiler-rt -print-libgcc-file-name 2>&1 \
 // RUN: --target=arm-linux-gnueabi -mfloat-abi=hard \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
 // RUN:   | FileCheck --check-prefix=CHECK-CLANGRT-ARM-ABI %s
 // CHECK-CLANGRT-ARM-ABI: libclang_rt.builtins-armhf.a
Index: clang/test/Driver/msp430-toolchain.c
===
--- clang/test/Driver/msp430-toolchain.c
+++ clang/test/Driver/msp430-toolchain.c
@@ -1,11 +1,11 @@
 // A basic clang -cc1 command-line, and simple environment check.
 
-// RUN: %clang %s -### -no-canonical-prefixes -target msp430 2>&1 \
+// RUN: %clang %s -### -no-canonical-prefixes -target msp430 --sysroot="" 2>&1 \
 // RUN:   | FileCheck -check-prefix=CC1 %s
 // CC1: clang{{.*}} "-cc1" "-triple" "msp430"
 
 // RUN: %clang %s -### -no-canonical-prefixes -target msp430 \
-// RUN:   --gcc-toolchain=%S/Inputs/basic_msp430_tree 2>&1 \
+// RUN:   --gcc-toolchain=%S/Inputs/basic_msp430_tree --sysroot="" 2>&1 \
 // RUN:   | FileCheck -check-prefix=MSP430 %s
 
 // MSP430: "{{.*}}Inputs/basic_msp430_tree/lib/gcc/msp430-elf/7.3.1/../../..{{/|}}..{{/|}}bin{{/|}}msp430-elf-ld"
@@ -18,7 +18,7 @@
 // MSP430: "{{.*}}/Inputs/basic_msp430_tree/lib/gcc/msp430-elf/7.3.1/../../..{{/|}}..{{/|}}msp430-elf{{/|}}lib/430{{/|}}crtn.o"
 
 // RUN: %clang %s -### -no-canonical-prefixes -target msp430 -nodefaultlibs \
-// RUN:   --gcc-toolchain=%S/Inputs/basic_msp430_tree 2>&1 \
+// RUN:   --gcc-toolchain=%S/Inputs/basic_msp430_tree --sysroot="" 2>&1 \
 // RUN:   | FileCheck -check-prefix=MSP430-NO-DFT-LIB %s
 
 // MSP430-NO-DFT-LIB: "{{.*}}Inputs/basic_msp430_tree/lib/gcc/msp430-elf/7.3.1/../../..{{/|}}..{{/|}}bin{{/|}}msp430-elf-ld"
@@ -31,7 +31,7 @@
 // MSP430-NO-DFT-LIB: "{{.*}}

[PATCH] D66834: Driver tests: set `--sysroot=""` to support clang with `DEFAULT_SYSROOT`

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb updated this revision to Diff 217991.
broadwaylamb added a comment.

Remove unrelated change


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D66834/new/

https://reviews.llvm.org/D66834

Files:
  clang/test/Driver/darwin-sdkroot.c
  clang/test/Driver/gcc-toolchain.cpp
  clang/test/Driver/mips-mti.cpp
  clang/test/Driver/msp430-toolchain.c
  clang/unittests/Driver/ToolChainTest.cpp

Index: clang/unittests/Driver/ToolChainTest.cpp
===
--- clang/unittests/Driver/ToolChainTest.cpp
+++ clang/unittests/Driver/ToolChainTest.cpp
@@ -60,7 +60,7 @@
 llvm::MemoryBuffer::getMemBuffer("\n"));
 
   std::unique_ptr C(TheDriver.BuildCompilation(
-  {"-fsyntax-only", "--gcc-toolchain=", "foo.cpp"}));
+  {"-fsyntax-only", "--gcc-toolchain=", "--sysroot=", "foo.cpp"}));
   EXPECT_TRUE(C);
 
   std::string S;
Index: clang/test/Driver/msp430-toolchain.c
===
--- clang/test/Driver/msp430-toolchain.c
+++ clang/test/Driver/msp430-toolchain.c
@@ -1,11 +1,11 @@
 // A basic clang -cc1 command-line, and simple environment check.
 
-// RUN: %clang %s -### -no-canonical-prefixes -target msp430 2>&1 \
+// RUN: %clang %s -### -no-canonical-prefixes -target msp430 --sysroot="" 2>&1 \
 // RUN:   | FileCheck -check-prefix=CC1 %s
 // CC1: clang{{.*}} "-cc1" "-triple" "msp430"
 
 // RUN: %clang %s -### -no-canonical-prefixes -target msp430 \
-// RUN:   --gcc-toolchain=%S/Inputs/basic_msp430_tree 2>&1 \
+// RUN:   --gcc-toolchain=%S/Inputs/basic_msp430_tree --sysroot="" 2>&1 \
 // RUN:   | FileCheck -check-prefix=MSP430 %s
 
 // MSP430: "{{.*}}Inputs/basic_msp430_tree/lib/gcc/msp430-elf/7.3.1/../../..{{/|}}..{{/|}}bin{{/|}}msp430-elf-ld"
@@ -18,7 +18,7 @@
 // MSP430: "{{.*}}/Inputs/basic_msp430_tree/lib/gcc/msp430-elf/7.3.1/../../..{{/|}}..{{/|}}msp430-elf{{/|}}lib/430{{/|}}crtn.o"
 
 // RUN: %clang %s -### -no-canonical-prefixes -target msp430 -nodefaultlibs \
-// RUN:   --gcc-toolchain=%S/Inputs/basic_msp430_tree 2>&1 \
+// RUN:   --gcc-toolchain=%S/Inputs/basic_msp430_tree --sysroot="" 2>&1 \
 // RUN:   | FileCheck -check-prefix=MSP430-NO-DFT-LIB %s
 
 // MSP430-NO-DFT-LIB: "{{.*}}Inputs/basic_msp430_tree/lib/gcc/msp430-elf/7.3.1/../../..{{/|}}..{{/|}}bin{{/|}}msp430-elf-ld"
@@ -31,7 +31,7 @@
 // MSP430-NO-DFT-LIB: "{{.*}}/Inputs/basic_msp430_tree/lib/gcc/msp430-elf/7.3.1/../../..{{/|}}..{{/|}}msp430-elf{{/|}}lib/430{{/|}}crtn.o"
 
 // RUN: %clang %s -### -no-canonical-prefixes -target msp430 -nostartfiles \
-// RUN:   --gcc-toolchain=%S/Inputs/basic_msp430_tree 2>&1 \
+// RUN:   --gcc-toolchain=%S/Inputs/basic_msp430_tree --sysroot="" 2>&1 \
 // RUN:   | FileCheck -check-prefix=MSP430-NO-START %s
 
 // MSP430-NO-START: "{{.*}}Inputs/basic_msp430_tree/lib/gcc/msp430-elf/7.3.1/../../..{{/|}}..{{/|}}bin{{/|}}msp430-elf-ld"
@@ -40,7 +40,7 @@
 // MSP430-NO-START: "--start-group" "-lmul_none" "-lgcc" "-lc" "-lcrt" "-lnosys" "--end-group"
 
 // RUN: %clang %s -### -no-canonical-prefixes -target msp430 -nostdlib \
-// RUN:   --gcc-toolchain=%S/Inputs/basic_msp430_tree 2>&1 \
+// RUN:   --gcc-toolchain=%S/Inputs/basic_msp430_tree --sysroot="" 2>&1 \
 // RUN:   | FileCheck -check-prefix=MSP430-NO-STD-LIB %s
 
 // MSP430-NO-STD-LIB: "{{.*}}Inputs/basic_msp430_tree/lib/gcc/msp430-elf/7.3.1/../../..{{/|}}..{{/|}}bin{{/|}}msp430-elf-ld"
@@ -48,31 +48,31 @@
 // MSP430-NO-STD-LIB: "-L{{.*}}/Inputs/basic_msp430_tree/lib/gcc/msp430-elf/7.3.1/../../..{{/|}}..{{/|}}msp430-elf{{/|}}lib/430"
 // MSP430-NO-STD-LIB: "--start-group" "-lmul_none" "-lgcc" "--end-group"
 
-// RUN: %clang %s -### -no-canonical-prefixes -target msp430 -mmcu=msp430f147 2>&1 \
+// RUN: %clang %s -### -no-canonical-prefixes -target msp430 -mmcu=msp430f147 --sysroot="" 2>&1 \
 // RUN:   | FileCheck -check-prefix=MSP430-HWMult-16BIT %s
-// RUN: %clang %s -### -no-canonical-prefixes -target msp430 -mmcu=msp430f147 -mhwmult=auto 2>&1 \
+// RUN: %clang %s -### -no-canonical-prefixes -target msp430 -mmcu=msp430f147 -mhwmult=auto --sysroot="" 2>&1 \
 // RUN:   | FileCheck -check-prefix=MSP430-HWMult-16BIT %s
-// RUN: %clang %s -### -no-canonical-prefixes -target msp430 -mhwmult=16bit 2>&1 \
+// RUN: %clang %s -### -no-canonical-prefixes -target msp430 -mhwmult=16bit --sysroot="" 2>&1 \
 // RUN:   | FileCheck -check-prefix=MSP430-HWMult-16BIT %s
 
 // MSP430-HWMult-16BIT: "--start-group" "-lmul_16"
 
-// RUN: %clang %s -### -no-canonical-prefixes -target msp430 -mmcu=msp430f4783 2>&1 \
+// RUN: %clang %s -### -no-canonical-prefixes -target msp430 -mmcu=msp430f4783 --sysroot="" 2>&1 \
 // RUN:   | FileCheck -check-prefix=MSP430-HWMult-32BIT %s
-// RUN: %clang %s -### -no-canonical-prefixes -target msp430 -mmcu=msp430f4783 -mhwmult=auto 2>&1 \
+// RUN: %clang %s -### -no-canonical-prefi

[PATCH] D66981: Fix driver tests when `LLVM_ENABLE_PER_TARGET_RUNTIME_DIR` is `ON`

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb created this revision.
broadwaylamb added reviewers: phosek, rnk, mcgrathr, vitalybuka, beanz, hans, 
filcab, rsmith, compnerd.
Herald added subscribers: javed.absar, srhines.
Herald added a project: clang.
broadwaylamb edited the summary of this revision.

Some Driver tests relied on the default resource direcory having per-os 
per-arch subdirectory layout, and when clang is built with 
`-DLLVM_ENABLE_PER_TARGET_RUNTIME_DIR=ON`, those test fail, because clang by 
default assumes per-target subdirectories.

Explicitly set `-resource-dir` flag to point to a tree with per-os per-arch 
layout.

See also: D45604 , D62469 



Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D66981

Files:
  clang/test/Driver/arm-compiler-rt.c
  clang/test/Driver/print-libgcc-file-name-clangrt.c

Index: clang/test/Driver/print-libgcc-file-name-clangrt.c
===
--- clang/test/Driver/print-libgcc-file-name-clangrt.c
+++ clang/test/Driver/print-libgcc-file-name-clangrt.c
@@ -2,11 +2,13 @@
 
 // RUN: %clang -rtlib=compiler-rt -print-libgcc-file-name 2>&1 \
 // RUN: --target=x86_64-pc-linux \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
 // RUN:   | FileCheck --check-prefix=CHECK-CLANGRT-X8664 %s
 // CHECK-CLANGRT-X8664: libclang_rt.builtins-x86_64.a
 
 // RUN: %clang -rtlib=compiler-rt -print-libgcc-file-name 2>&1 \
 // RUN: --target=i386-pc-linux \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
 // RUN:   | FileCheck --check-prefix=CHECK-CLANGRT-I386 %s
 // CHECK-CLANGRT-I386: libclang_rt.builtins-i386.a
 
@@ -14,24 +16,29 @@
 //
 // RUN: %clang -rtlib=compiler-rt -print-libgcc-file-name 2>&1 \
 // RUN: --target=i686-pc-linux \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
 // RUN:   | FileCheck --check-prefix=CHECK-CLANGRT-I386 %s
 
 // RUN: %clang -rtlib=compiler-rt -print-libgcc-file-name 2>&1 \
 // RUN: --target=arm-linux-gnueabi \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
 // RUN:   | FileCheck --check-prefix=CHECK-CLANGRT-ARM %s
 // CHECK-CLANGRT-ARM: libclang_rt.builtins-arm.a
 
 // RUN: %clang -rtlib=compiler-rt -print-libgcc-file-name 2>&1 \
 // RUN: --target=arm-linux-androideabi \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
 // RUN:   | FileCheck --check-prefix=CHECK-CLANGRT-ARM-ANDROID %s
 // CHECK-CLANGRT-ARM-ANDROID: libclang_rt.builtins-arm-android.a
 
 // RUN: %clang -rtlib=compiler-rt -print-libgcc-file-name 2>&1 \
 // RUN: --target=arm-linux-gnueabihf \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
 // RUN:   | FileCheck --check-prefix=CHECK-CLANGRT-ARMHF %s
 // CHECK-CLANGRT-ARMHF: libclang_rt.builtins-armhf.a
 
 // RUN: %clang -rtlib=compiler-rt -print-libgcc-file-name 2>&1 \
 // RUN: --target=arm-linux-gnueabi -mfloat-abi=hard \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
 // RUN:   | FileCheck --check-prefix=CHECK-CLANGRT-ARM-ABI %s
 // CHECK-CLANGRT-ARM-ABI: libclang_rt.builtins-armhf.a
Index: clang/test/Driver/arm-compiler-rt.c
===
--- clang/test/Driver/arm-compiler-rt.c
+++ clang/test/Driver/arm-compiler-rt.c
@@ -1,21 +1,42 @@
-// RUN: %clang -target arm-linux-gnueabi -rtlib=compiler-rt -### %s 2>&1 | FileCheck %s -check-prefix ARM-GNUEABI
+// RUN: %clang -target arm-linux-gnueabi \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
+// RUN: -rtlib=compiler-rt -### %s 2>&1 \
+// RUN:   | FileCheck %s -check-prefix ARM-GNUEABI
 // ARM-GNUEABI: "{{.*[/\\]}}libclang_rt.builtins-arm.a"
 
-// RUN: %clang -target arm-linux-gnueabi -rtlib=compiler-rt -mfloat-abi=hard -### %s 2>&1 | FileCheck %s -check-prefix ARM-GNUEABI-ABI
+// RUN: %clang -target arm-linux-gnueabi \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
+// RUN: -rtlib=compiler-rt -mfloat-abi=hard -### %s 2>&1 \
+// RUN:   | FileCheck %s -check-prefix ARM-GNUEABI-ABI
 // ARM-GNUEABI-ABI: "{{.*[/\\]}}libclang_rt.builtins-armhf.a"
 
-// RUN: %clang -target arm-linux-gnueabihf -rtlib=compiler-rt -### %s 2>&1 | FileCheck %s -check-prefix ARM-GNUEABIHF
+// RUN: %clang -target arm-linux-gnueabihf \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
+// RUN: -rtlib=compiler-rt -### %s 2>&1 \
+// RUN:   | FileCheck %s -check-prefix ARM-GNUEABIHF
 // ARM-GNUEABIHF: "{{.*[/\\]}}libclang_rt.builtins-armhf.a"
 
-// RUN: %clang -target arm-linux-gnueabihf -rtlib=compiler-rt -mfloat-abi=soft -### %s 2>&1 | FileCheck %s -check-prefix ARM-GNUEABIHF-ABI
+// RUN: %clang -target arm-linux-gnueabihf \
+// RUN: -resource-dir=%S/Inputs/resource_dir_with_arch_subdir \
+// RUN: -rtlib=compiler-rt -mfloat-abi=soft -### %s 2>&1 \
+// RUN:   | FileCheck %s -check-prefix ARM-GNUEABIHF-ABI
 // ARM-

[PATCH] D66834: Driver tests: set `--sysroot=""` to support clang with `DEFAULT_SYSROOT`

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb added a comment.

In D66834#1652756 , @compnerd wrote:

> I think that this is pretty easy to forget.  Fortunately, last argument wins. 
>  Why not sink this into the `%clang` substitution in lit?  That ensures that 
> we run with an empty sysroot and then when the test needs to adjust the 
> sysroot, it can do so explicitly.


This totally makes sense.

I've just tried to do it, but unfortunately some tests are failing, for 
example, `Driver/cc1-response-files.c`. The problem is that `%clang` is 
expanded to `/path/to/clang --sysroot=`, but the succeeding flags (such as 
`-cc1`) may be incompatible with `--sysroot`.

Or am I missing something?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D66834/new/

https://reviews.llvm.org/D66834



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D72459: [CMake] Support running libc++abi tests in CrossWinToARMLinux cache file

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb created this revision.
Herald added subscribers: cfe-commits, ldionne, kristof.beyls, mgorny.
Herald added a reviewer: EricWF.
Herald added a project: clang.
broadwaylamb edited the summary of this revision.
broadwaylamb edited reviewers, added: andreil99, vvereschaka, aorlov; removed: 
EricWF.
broadwaylamb edited the summary of this revision.

Now that D71894  has landed, we're able to run 
libc++abi tests remotely.

For that we can use the same CMake command as before. The tests can be run 
using `ninja check-cxxabi`.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D72459

Files:
  clang/cmake/caches/CrossWinToARMLinux.cmake


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -76,10 +76,13 @@
 set(LIBCXXABI_ENABLE_NEW_DELETE_DEFINITIONS OFF CACHE BOOL "")
 set(LIBCXXABI_TARGET_TRIPLE "${CMAKE_C_COMPILER_TARGET}" CACHE 
STRING "")
 set(LIBCXXABI_SYSROOT   "${DEFAULT_SYSROOT}" CACHE STRING 
"")
+set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXABI OFF CACHE BOOL "")
+set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXOFF CACHE BOOL "")
 
 set(LIBCXX_USE_COMPILER_RT  ON CACHE BOOL "")
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE 
STRING "")
 set(LIBCXX_SYSROOT  "${DEFAULT_SYSROOT}" CACHE STRING 
"")
+set(LIBCXX_ENABLE_SHAREDOFF CACHE BOOL "")
 
 set(BUILTINS_CMAKE_ARGS 
"-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
 set(RUNTIMES_CMAKE_ARGS 
"-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -76,10 +76,13 @@
 set(LIBCXXABI_ENABLE_NEW_DELETE_DEFINITIONS OFF CACHE BOOL "")
 set(LIBCXXABI_TARGET_TRIPLE "${CMAKE_C_COMPILER_TARGET}" CACHE STRING "")
 set(LIBCXXABI_SYSROOT   "${DEFAULT_SYSROOT}" CACHE STRING "")
+set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXABI OFF CACHE BOOL "")
+set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXOFF CACHE BOOL "")
 
 set(LIBCXX_USE_COMPILER_RT  ON CACHE BOOL "")
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE STRING "")
 set(LIBCXX_SYSROOT  "${DEFAULT_SYSROOT}" CACHE STRING "")
+set(LIBCXX_ENABLE_SHAREDOFF CACHE BOOL "")
 
 set(BUILTINS_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
 set(RUNTIMES_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D72459: [CMake] Support running libc++abi tests in CrossWinToARMLinux cache file

[Sergej Jaskiewicz via Phabricator via cfe-commits]

This revision was automatically updated to reflect the committed changes.
Closed by commit rGe44dedd3631c: [CMake] Support running libc++abi tests in 
CrossWinToARMLinux cache file (authored by broadwaylamb).

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D72459/new/

https://reviews.llvm.org/D72459

Files:
  clang/cmake/caches/CrossWinToARMLinux.cmake


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -76,10 +76,13 @@
 set(LIBCXXABI_ENABLE_NEW_DELETE_DEFINITIONS OFF CACHE BOOL "")
 set(LIBCXXABI_TARGET_TRIPLE "${CMAKE_C_COMPILER_TARGET}" CACHE 
STRING "")
 set(LIBCXXABI_SYSROOT   "${DEFAULT_SYSROOT}" CACHE STRING 
"")
+set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXABI OFF CACHE BOOL "")
+set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXOFF CACHE BOOL "")
 
 set(LIBCXX_USE_COMPILER_RT  ON CACHE BOOL "")
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE 
STRING "")
 set(LIBCXX_SYSROOT  "${DEFAULT_SYSROOT}" CACHE STRING 
"")
+set(LIBCXX_ENABLE_SHAREDOFF CACHE BOOL "")
 
 set(BUILTINS_CMAKE_ARGS 
"-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
 set(RUNTIMES_CMAKE_ARGS 
"-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -76,10 +76,13 @@
 set(LIBCXXABI_ENABLE_NEW_DELETE_DEFINITIONS OFF CACHE BOOL "")
 set(LIBCXXABI_TARGET_TRIPLE "${CMAKE_C_COMPILER_TARGET}" CACHE STRING "")
 set(LIBCXXABI_SYSROOT   "${DEFAULT_SYSROOT}" CACHE STRING "")
+set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXABI OFF CACHE BOOL "")
+set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXOFF CACHE BOOL "")
 
 set(LIBCXX_USE_COMPILER_RT  ON CACHE BOOL "")
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE STRING "")
 set(LIBCXX_SYSROOT  "${DEFAULT_SYSROOT}" CACHE STRING "")
+set(LIBCXX_ENABLE_SHAREDOFF CACHE BOOL "")
 
 set(BUILTINS_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
 set(RUNTIMES_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D72717: [CMake] Disable libc++ filesystem tests in CrossWinToARMLinux cache file

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb created this revision.
broadwaylamb added reviewers: aorlov, vvereschaka, andreil99.
Herald added subscribers: cfe-commits, ldionne, kristof.beyls, mgorny.
Herald added a reviewer: EricWF.
Herald added a project: clang.

`filesystem` tests are not yet supported when running tests on a remote board, 
because the testing infrastructure isn't quite ready for it yet. Supporting 
`filesystem` in the infrastructure is non-trivial, so I suggest disabling these 
tests for now.

Also, pass through LIT arguments to builtins and runtimes (they were not passed 
through before). I'm not sure about this change though, as we may want to 
differentiate between LIT arguments for llvm/clang tests from those for 
libunwind/libc++abi/libc++ tests. For example, we may want to specify a higher 
number of threads for the tests that are run remotely, as it significantly 
speeds up the test suite. Please let me know what you think.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D72717

Files:
  clang/cmake/caches/CrossWinToARMLinux.cmake


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -54,7 +54,9 @@
 set(LLVM_ENABLE_RUNTIMES
"compiler-rt;libunwind;libcxxabi;libcxx" CACHE STRING "")
 set(LLVM_DEFAULT_TARGET_TRIPLE  "${CMAKE_C_COMPILER_TARGET}" CACHE 
STRING "")
 set(LLVM_TARGET_ARCH"${CMAKE_C_COMPILER_TARGET}" CACHE 
STRING "")
-set(LLVM_LIT_ARGS   "-vv ${LLVM_LIT_ARGS}" CACHE 
STRING "" FORCE)
+
+# Note that testing the 'filesystem' header from libc++ is not yet supported 
in remote configuration.
+set(LLVM_LIT_ARGS   "-vv --param 
enable_filesystem=False ${LLVM_LIT_ARGS}" CACHE STRING "" FORCE)
 
 set(CLANG_DEFAULT_LINKER"lld" CACHE STRING "")
 
@@ -84,8 +86,15 @@
 set(LIBCXX_SYSROOT  "${DEFAULT_SYSROOT}" CACHE STRING 
"")
 set(LIBCXX_ENABLE_SHAREDOFF CACHE BOOL "")
 
-set(BUILTINS_CMAKE_ARGS 
"-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
-set(RUNTIMES_CMAKE_ARGS 
"-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
+set(BUILTINS_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;\
+ -DCMAKE_AR=${CMAKE_AR};\
+ -DLLVM_LIT_ARGS=${LLVM_LIT_ARGS}" 
+ CACHE STRING "")
+
+set(RUNTIMES_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;\
+ -DCMAKE_AR=${CMAKE_AR};\
+ -DLLVM_LIT_ARGS=${LLVM_LIT_ARGS}"
+ CACHE STRING "")
 
 # Remote test configuration.
 if(DEFINED REMOTE_TEST_HOST)


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -54,7 +54,9 @@
 set(LLVM_ENABLE_RUNTIMES"compiler-rt;libunwind;libcxxabi;libcxx" CACHE STRING "")
 set(LLVM_DEFAULT_TARGET_TRIPLE  "${CMAKE_C_COMPILER_TARGET}" CACHE STRING "")
 set(LLVM_TARGET_ARCH"${CMAKE_C_COMPILER_TARGET}" CACHE STRING "")
-set(LLVM_LIT_ARGS   "-vv ${LLVM_LIT_ARGS}" CACHE STRING "" FORCE)
+
+# Note that testing the 'filesystem' header from libc++ is not yet supported in remote configuration.
+set(LLVM_LIT_ARGS   "-vv --param enable_filesystem=False ${LLVM_LIT_ARGS}" CACHE STRING "" FORCE)
 
 set(CLANG_DEFAULT_LINKER"lld" CACHE STRING "")
 
@@ -84,8 +86,15 @@
 set(LIBCXX_SYSROOT  "${DEFAULT_SYSROOT}" CACHE STRING "")
 set(LIBCXX_ENABLE_SHAREDOFF CACHE BOOL "")
 
-set(BUILTINS_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
-set(RUNTIMES_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
+set(BUILTINS_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;\
+ -DCMAKE_AR=${CMAKE_AR};\
+ -DLLVM_LIT_ARGS=${LLVM_LIT_ARGS}" 
+ CACHE STRING "")
+
+set(RUNTIMES_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;\
+ -DCMAKE_AR=${CMAKE_AR};\
+ -DLLVM_LIT_ARGS=${LLVM_LIT_ARGS}"
+ CACHE STRING "")
 
 # Remote test configuration.
 if(DEFINED REMOTE_TEST_HOST)
_

[PATCH] D89793: [libc++] Explicitly request new/delete in libc++ for Win to ARM Linux builds

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb accepted this revision.
broadwaylamb added a comment.
This revision is now accepted and ready to land.

LGTM


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D89793/new/

https://reviews.llvm.org/D89793

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D76774: [cmake] Link libc++ tests against static libc++/libc++abi in CrossWinToARMLinux.cmake

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb created this revision.
broadwaylamb added a reviewer: vvereschaka.
Herald added subscribers: cfe-commits, kristof.beyls, mgorny.
Herald added a project: clang.

Now that D72687  has landed, we can enable 
this setting in our cache file.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D76774

Files:
  clang/cmake/caches/CrossWinToARMLinux.cmake


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -86,6 +86,8 @@
 set(LIBCXXABI_SYSROOT   "${DEFAULT_SYSROOT}" CACHE STRING 
"")
 set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXABI OFF CACHE BOOL "")
 set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXOFF CACHE BOOL "")
+set(LIBCXX_LINK_TESTS_WITH_SHARED_LIBCXXABIOFF CACHE BOOL "")
+set(LIBCXX_LINK_TESTS_WITH_SHARED_LIBCXX   OFF CACHE BOOL "")
 
 set(LIBCXX_USE_COMPILER_RT  ON CACHE BOOL "")
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE 
STRING "")


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -86,6 +86,8 @@
 set(LIBCXXABI_SYSROOT   "${DEFAULT_SYSROOT}" CACHE STRING "")
 set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXABI OFF CACHE BOOL "")
 set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXOFF CACHE BOOL "")
+set(LIBCXX_LINK_TESTS_WITH_SHARED_LIBCXXABIOFF CACHE BOOL "")
+set(LIBCXX_LINK_TESTS_WITH_SHARED_LIBCXX   OFF CACHE BOOL "")
 
 set(LIBCXX_USE_COMPILER_RT  ON CACHE BOOL "")
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE STRING "")
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D76774: [cmake] Link libc++ tests against static libc++/libc++abi in CrossWinToARMLinux.cmake

[Sergej Jaskiewicz via Phabricator via cfe-commits]

This revision was automatically updated to reflect the committed changes.
Closed by commit rG9a6a696817bd: [cmake] Link libc++ tests against static 
libc++/libc++abi in CrossWinToARMLinux. (authored by broadwaylamb).

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D76774/new/

https://reviews.llvm.org/D76774

Files:
  clang/cmake/caches/CrossWinToARMLinux.cmake


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -86,6 +86,8 @@
 set(LIBCXXABI_SYSROOT   "${DEFAULT_SYSROOT}" CACHE STRING 
"")
 set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXABI OFF CACHE BOOL "")
 set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXOFF CACHE BOOL "")
+set(LIBCXX_LINK_TESTS_WITH_SHARED_LIBCXXABIOFF CACHE BOOL "")
+set(LIBCXX_LINK_TESTS_WITH_SHARED_LIBCXX   OFF CACHE BOOL "")
 
 set(LIBCXX_USE_COMPILER_RT  ON CACHE BOOL "")
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE 
STRING "")


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -86,6 +86,8 @@
 set(LIBCXXABI_SYSROOT   "${DEFAULT_SYSROOT}" CACHE STRING "")
 set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXABI OFF CACHE BOOL "")
 set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXOFF CACHE BOOL "")
+set(LIBCXX_LINK_TESTS_WITH_SHARED_LIBCXXABIOFF CACHE BOOL "")
+set(LIBCXX_LINK_TESTS_WITH_SHARED_LIBCXX   OFF CACHE BOOL "")
 
 set(LIBCXX_USE_COMPILER_RT  ON CACHE BOOL "")
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE STRING "")
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D70499: [clang] Fix the path to CrossWinToARMLinux.cmake CMake cache

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb created this revision.
broadwaylamb added reviewers: vvereschaka, andreil99.
broadwaylamb added a project: clang.
Herald added subscribers: cfe-commits, kristof.beyls, mgorny.

The comment was slightly misleading.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D70499

Files:
  clang/cmake/caches/CrossWinToARMLinux.cmake


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -14,7 +14,7 @@
 #   -DDEFAULT_SYSROOT= ^
 #   -DLLVM_AR=/bin/llvm-ar[.exe] ^
 #   -DCMAKE_CXX_FLAGS="-D__OPTIMIZE__" ^
-#   -C/llvm-project/clang/caches/CrossWinToARMLinux.cmake ^
+#   
-C/llvm-project/clang/cmake/caches/CrossWinToARMLinux.cmake ^
 #   /llvm-project/llvm
 # Build:
 #  cmake --build . --target install


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -14,7 +14,7 @@
 #   -DDEFAULT_SYSROOT= ^
 #   -DLLVM_AR=/bin/llvm-ar[.exe] ^
 #   -DCMAKE_CXX_FLAGS="-D__OPTIMIZE__" ^
-#   -C/llvm-project/clang/caches/CrossWinToARMLinux.cmake ^
+#   -C/llvm-project/clang/cmake/caches/CrossWinToARMLinux.cmake ^
 #   /llvm-project/llvm
 # Build:
 #  cmake --build . --target install
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D70499: [clang] Fix the path to CrossWinToARMLinux.cmake CMake cache

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb added a comment.

Can you commit this for me?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D70499/new/

https://reviews.llvm.org/D70499



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D71625: [CMake] Added remote test execution support into CrossWinToARMLinux CMake cache file.

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb added a comment.

LGTM




Comment at: clang/cmake/caches/CrossWinToARMLinux.cmake:88-91
+set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXXABI  OFF CACHE BOOL "")
+set(LIBCXXABI_LINK_TESTS_WITH_SHARED_LIBCXX OFF CACHE BOOL "")
+set(LIBCXX_LINK_TESTS_WITH_SHARED_LIBCXXOFF CACHE BOOL "")
+set(LIBCXX_LINK_TESTS_WITH_SHARED_LIBCXXABI OFF CACHE BOOL "")

These options haven't yet made it into the upstream, maybe it's better remove 
them for now and add them later when they're stabilized.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D71625/new/

https://reviews.llvm.org/D71625



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D74347: [CMake] Explicitly specify paths to libc++abi in CrossWinToARMLinux.cmake

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb updated this revision to Diff 249779.
broadwaylamb added a comment.

Add one more libc++abi header search path when building libc++

@vvereschaka this is supposed to fix the buildbot failure. It's more of a sweep 
under the rug type of fix. The real problem is that the value of the 
`CMAKE_SOURCE_DIR` variable is unexpected.

See for yourself.

Before 0197eac3330c04a49519f3e4dac38c4de605c654 
 I had the 
following in the build log:

  -- Looking for cxxabi.h in 
C:/buildbot/as-builder-1/llvm-clang-win-x-armv7l/llvm-project/libcxxabi/include 
- found

After (note the missing 'llvm-project' component in the path):

  -- Looking for cxxabi.h in 
C:/buildbot/as-builder-1/llvm-clang-win-x-armv7l/libcxxabi/include - not found

Maybe you have an idea of why this is happening? This all works fine on my 
machine even if I reproduce the directory layout used on the buildbot and 
invoke CMake from the `build` directory:

  llvm-clang-win-x-armv7l
  ├───build
  └───llvm-project


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D74347/new/

https://reviews.llvm.org/D74347

Files:
  clang/cmake/caches/CrossWinToARMLinux.cmake


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -83,6 +83,9 @@
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE 
STRING "")
 set(LIBCXX_SYSROOT  "${DEFAULT_SYSROOT}" CACHE STRING 
"")
 set(LIBCXX_ENABLE_SHAREDOFF CACHE BOOL "")
+set(LIBCXX_CXX_ABI  "libcxxabi" CACHE STRING "")
+set(LIBCXX_CXX_ABI_INCLUDE_PATHS
"${CMAKE_SOURCE_DIR}/../libcxxabi/include;${CMAKE_SOURCE_DIR}/../llvm-project/libcxxabi/include"
 CACHE PATH "")
+set(LIBCXX_CXX_ABI_LIBRARY_PATH 
"${CMAKE_BINARY_DIR}/lib/${CMAKE_C_COMPILER_TARGET}/c++" CACHE PATH "")
 
 set(BUILTINS_CMAKE_ARGS 
"-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
 set(RUNTIMES_CMAKE_ARGS 
"-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -83,6 +83,9 @@
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE STRING "")
 set(LIBCXX_SYSROOT  "${DEFAULT_SYSROOT}" CACHE STRING "")
 set(LIBCXX_ENABLE_SHAREDOFF CACHE BOOL "")
+set(LIBCXX_CXX_ABI  "libcxxabi" CACHE STRING "")
+set(LIBCXX_CXX_ABI_INCLUDE_PATHS"${CMAKE_SOURCE_DIR}/../libcxxabi/include;${CMAKE_SOURCE_DIR}/../llvm-project/libcxxabi/include" CACHE PATH "")
+set(LIBCXX_CXX_ABI_LIBRARY_PATH "${CMAKE_BINARY_DIR}/lib/${CMAKE_C_COMPILER_TARGET}/c++" CACHE PATH "")
 
 set(BUILTINS_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
 set(RUNTIMES_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D74347: [CMake] Explicitly specify paths to libc++abi in CrossWinToARMLinux.cmake

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb added a comment.

@vvereschaka very interesting. I've downloaded CMake 3.10.2 and finally been 
able to reproduce the issue locally. So it //is// a CMake bug which does not 
appear in the latest versions. I'll update this patch to use 
`CMAKE_CURRENT_LIST_DIR` and add a comment.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D74347/new/

https://reviews.llvm.org/D74347



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D74347: [CMake] Explicitly specify paths to libc++abi in CrossWinToARMLinux.cmake

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb updated this revision to Diff 249916.
broadwaylamb added a comment.

Use `CMAKE_CURRENT_LIST_DIR` instead of `CMAKE_SOURCE_DIR` for locating the 
llvm-project directory.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D74347/new/

https://reviews.llvm.org/D74347

Files:
  clang/cmake/caches/CrossWinToARMLinux.cmake


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -25,6 +25,14 @@
 #  cmake --build . --target check-clang
 #  cmake --build . --target check-lld
 
+# LLVM_PROJECT_DIR is the path to the llvm-project directory.
+# The right way to compute it would probably be to use 
"${CMAKE_SOURCE_DIR}/../",
+# but CMAKE_SOURCE_DIR is set to the wrong value on earlier CMake versions
+# that we still need to support (for instance, 3.10.2).
+get_filename_component(LLVM_PROJECT_DIR
+   "${CMAKE_CURRENT_LIST_DIR}/../../../"
+   ABSOLUTE)
+
 if (NOT DEFINED DEFAULT_SYSROOT)
   message(WARNING "DEFAULT_SYSROOT must be specified for the cross toolchain 
build.")
 endif()
@@ -83,6 +91,9 @@
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE 
STRING "")
 set(LIBCXX_SYSROOT  "${DEFAULT_SYSROOT}" CACHE STRING 
"")
 set(LIBCXX_ENABLE_SHAREDOFF CACHE BOOL "")
+set(LIBCXX_CXX_ABI  "libcxxabi" CACHE STRING "")
+set(LIBCXX_CXX_ABI_INCLUDE_PATHS
"${LLVM_PROJECT_DIR}/libcxxabi/include" CACHE PATH "")
+set(LIBCXX_CXX_ABI_LIBRARY_PATH 
"${CMAKE_BINARY_DIR}/lib/${LIBCXX_TARGET_TRIPLE}/c++" CACHE PATH "")
 
 set(BUILTINS_CMAKE_ARGS 
"-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
 set(RUNTIMES_CMAKE_ARGS 
"-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -25,6 +25,14 @@
 #  cmake --build . --target check-clang
 #  cmake --build . --target check-lld
 
+# LLVM_PROJECT_DIR is the path to the llvm-project directory.
+# The right way to compute it would probably be to use "${CMAKE_SOURCE_DIR}/../",
+# but CMAKE_SOURCE_DIR is set to the wrong value on earlier CMake versions
+# that we still need to support (for instance, 3.10.2).
+get_filename_component(LLVM_PROJECT_DIR
+   "${CMAKE_CURRENT_LIST_DIR}/../../../"
+   ABSOLUTE)
+
 if (NOT DEFINED DEFAULT_SYSROOT)
   message(WARNING "DEFAULT_SYSROOT must be specified for the cross toolchain build.")
 endif()
@@ -83,6 +91,9 @@
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE STRING "")
 set(LIBCXX_SYSROOT  "${DEFAULT_SYSROOT}" CACHE STRING "")
 set(LIBCXX_ENABLE_SHAREDOFF CACHE BOOL "")
+set(LIBCXX_CXX_ABI  "libcxxabi" CACHE STRING "")
+set(LIBCXX_CXX_ABI_INCLUDE_PATHS"${LLVM_PROJECT_DIR}/libcxxabi/include" CACHE PATH "")
+set(LIBCXX_CXX_ABI_LIBRARY_PATH "${CMAKE_BINARY_DIR}/lib/${LIBCXX_TARGET_TRIPLE}/c++" CACHE PATH "")
 
 set(BUILTINS_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
 set(RUNTIMES_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D74347: [CMake] Explicitly specify paths to libc++abi in CrossWinToARMLinux.cmake

[Sergej Jaskiewicz via Phabricator via cfe-commits]

This revision was not accepted when it landed; it landed in state "Needs 
Review".
This revision was automatically updated to reflect the committed changes.
Closed by commit rG03993f3b6f46: [CMake] Explicitly specify paths to libc++abi 
in CrossWinToARMLinux.cmake (authored by broadwaylamb).

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D74347/new/

https://reviews.llvm.org/D74347

Files:
  clang/cmake/caches/CrossWinToARMLinux.cmake


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -25,6 +25,14 @@
 #  cmake --build . --target check-clang
 #  cmake --build . --target check-lld
 
+# LLVM_PROJECT_DIR is the path to the llvm-project directory.
+# The right way to compute it would probably be to use 
"${CMAKE_SOURCE_DIR}/../",
+# but CMAKE_SOURCE_DIR is set to the wrong value on earlier CMake versions
+# that we still need to support (for instance, 3.10.2).
+get_filename_component(LLVM_PROJECT_DIR
+   "${CMAKE_CURRENT_LIST_DIR}/../../../"
+   ABSOLUTE)
+
 if (NOT DEFINED DEFAULT_SYSROOT)
   message(WARNING "DEFAULT_SYSROOT must be specified for the cross toolchain 
build.")
 endif()
@@ -83,6 +91,9 @@
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE 
STRING "")
 set(LIBCXX_SYSROOT  "${DEFAULT_SYSROOT}" CACHE STRING 
"")
 set(LIBCXX_ENABLE_SHAREDOFF CACHE BOOL "")
+set(LIBCXX_CXX_ABI  "libcxxabi" CACHE STRING "")
+set(LIBCXX_CXX_ABI_INCLUDE_PATHS
"${LLVM_PROJECT_DIR}/libcxxabi/include" CACHE PATH "")
+set(LIBCXX_CXX_ABI_LIBRARY_PATH 
"${CMAKE_BINARY_DIR}/lib/${LIBCXX_TARGET_TRIPLE}/c++" CACHE PATH "")
 
 set(BUILTINS_CMAKE_ARGS 
"-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
 set(RUNTIMES_CMAKE_ARGS 
"-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -25,6 +25,14 @@
 #  cmake --build . --target check-clang
 #  cmake --build . --target check-lld
 
+# LLVM_PROJECT_DIR is the path to the llvm-project directory.
+# The right way to compute it would probably be to use "${CMAKE_SOURCE_DIR}/../",
+# but CMAKE_SOURCE_DIR is set to the wrong value on earlier CMake versions
+# that we still need to support (for instance, 3.10.2).
+get_filename_component(LLVM_PROJECT_DIR
+   "${CMAKE_CURRENT_LIST_DIR}/../../../"
+   ABSOLUTE)
+
 if (NOT DEFINED DEFAULT_SYSROOT)
   message(WARNING "DEFAULT_SYSROOT must be specified for the cross toolchain build.")
 endif()
@@ -83,6 +91,9 @@
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE STRING "")
 set(LIBCXX_SYSROOT  "${DEFAULT_SYSROOT}" CACHE STRING "")
 set(LIBCXX_ENABLE_SHAREDOFF CACHE BOOL "")
+set(LIBCXX_CXX_ABI  "libcxxabi" CACHE STRING "")
+set(LIBCXX_CXX_ABI_INCLUDE_PATHS"${LLVM_PROJECT_DIR}/libcxxabi/include" CACHE PATH "")
+set(LIBCXX_CXX_ABI_LIBRARY_PATH "${CMAKE_BINARY_DIR}/lib/${LIBCXX_TARGET_TRIPLE}/c++" CACHE PATH "")
 
 set(BUILTINS_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
 set(RUNTIMES_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D74347: [CMake] Explicitly specify paths to libc++abi in CrossWinToARMLinux.cmake

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb created this revision.
broadwaylamb added reviewers: vvereschaka, aorlov, andreil99.
Herald added subscribers: cfe-commits, ldionne, kristof.beyls, mgorny.
Herald added a reviewer: EricWF.
Herald added a project: clang.

D69169 , which was necessary for running 
libc++ tests on remote host, got reverted. I couldn't think of a less invasive 
way to achieve this behavior but specify libc++abi paths in our cache file.

I was wondering though if the assumptions I made in this patch regarding 
directory layout always hold.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D74347

Files:
  clang/cmake/caches/CrossWinToARMLinux.cmake


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -83,6 +83,9 @@
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE 
STRING "")
 set(LIBCXX_SYSROOT  "${DEFAULT_SYSROOT}" CACHE STRING 
"")
 set(LIBCXX_ENABLE_SHAREDOFF CACHE BOOL "")
+set(LIBCXX_CXX_ABI  "libcxxabi" CACHE STRING "")
+set(LIBCXX_CXX_ABI_INCLUDE_PATHS
"${CMAKE_SOURCE_DIR}/../libcxxabi/include" CACHE PATH "")
+set(LIBCXX_CXX_ABI_LIBRARY_PATH 
"${CMAKE_BINARY_DIR}/lib/${CMAKE_C_COMPILER_TARGET}/c++" CACHE PATH "")
 
 set(BUILTINS_CMAKE_ARGS 
"-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
 set(RUNTIMES_CMAKE_ARGS 
"-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -83,6 +83,9 @@
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE STRING "")
 set(LIBCXX_SYSROOT  "${DEFAULT_SYSROOT}" CACHE STRING "")
 set(LIBCXX_ENABLE_SHAREDOFF CACHE BOOL "")
+set(LIBCXX_CXX_ABI  "libcxxabi" CACHE STRING "")
+set(LIBCXX_CXX_ABI_INCLUDE_PATHS"${CMAKE_SOURCE_DIR}/../libcxxabi/include" CACHE PATH "")
+set(LIBCXX_CXX_ABI_LIBRARY_PATH "${CMAKE_BINARY_DIR}/lib/${CMAKE_C_COMPILER_TARGET}/c++" CACHE PATH "")
 
 set(BUILTINS_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
 set(RUNTIMES_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D74347: [CMake] Explicitly specify paths to libc++abi in CrossWinToARMLinux.cmake

[Sergej Jaskiewicz via Phabricator via cfe-commits]

This revision was automatically updated to reflect the committed changes.
Closed by commit rG801627ee490e: [CMake] Explicitly specify paths to libc++abi 
in CrossWinToARMLinux.cmake (authored by broadwaylamb).

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D74347/new/

https://reviews.llvm.org/D74347

Files:
  clang/cmake/caches/CrossWinToARMLinux.cmake


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -83,6 +83,9 @@
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE 
STRING "")
 set(LIBCXX_SYSROOT  "${DEFAULT_SYSROOT}" CACHE STRING 
"")
 set(LIBCXX_ENABLE_SHAREDOFF CACHE BOOL "")
+set(LIBCXX_CXX_ABI  "libcxxabi" CACHE STRING "")
+set(LIBCXX_CXX_ABI_INCLUDE_PATHS
"${CMAKE_SOURCE_DIR}/../libcxxabi/include" CACHE PATH "")
+set(LIBCXX_CXX_ABI_LIBRARY_PATH 
"${CMAKE_BINARY_DIR}/lib/${CMAKE_C_COMPILER_TARGET}/c++" CACHE PATH "")
 
 set(BUILTINS_CMAKE_ARGS 
"-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
 set(RUNTIMES_CMAKE_ARGS 
"-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -83,6 +83,9 @@
 set(LIBCXX_TARGET_TRIPLE"${CMAKE_C_COMPILER_TARGET}" CACHE STRING "")
 set(LIBCXX_SYSROOT  "${DEFAULT_SYSROOT}" CACHE STRING "")
 set(LIBCXX_ENABLE_SHAREDOFF CACHE BOOL "")
+set(LIBCXX_CXX_ABI  "libcxxabi" CACHE STRING "")
+set(LIBCXX_CXX_ABI_INCLUDE_PATHS"${CMAKE_SOURCE_DIR}/../libcxxabi/include" CACHE PATH "")
+set(LIBCXX_CXX_ABI_LIBRARY_PATH "${CMAKE_BINARY_DIR}/lib/${CMAKE_C_COMPILER_TARGET}/c++" CACHE PATH "")
 
 set(BUILTINS_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
 set(RUNTIMES_CMAKE_ARGS "-DCMAKE_SYSTEM_NAME=Linux;-DCMAKE_AR=${CMAKE_AR}" CACHE STRING "")
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D74347: [CMake] Explicitly specify paths to libc++abi in CrossWinToARMLinux.cmake

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb added a comment.

Reverted this because it broke our buildbot: 
http://lab.llvm.org:8011/builders/llvm-clang-win-x-armv7l/builds/4704/steps/build-llvm-project/logs/stdio.

Going to investigate.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D74347/new/

https://reviews.llvm.org/D74347



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D72850: [CMake] Use LinuxRemoteTI instead of LinuxLocalTI in CrossWinToARMLinux cmake cache

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb created this revision.
broadwaylamb added reviewers: vvereschaka, aorlov, andreil99.
Herald added subscribers: cfe-commits, kristof.beyls, mgorny.
Herald added a project: clang.

Depends on D72847 


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D72850

Files:
  clang/cmake/caches/CrossWinToARMLinux.cmake


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -99,7 +99,7 @@
 # Remote test configuration.
 if(DEFINED REMOTE_TEST_HOST)
   set(DEFAULT_TEST_EXECUTOR 
"SSHExecutor('${REMOTE_TEST_HOST}', '${REMOTE_TEST_USER}')")
-  set(DEFAULT_TEST_TARGET_INFO  
"libcxx.test.target_info.LinuxLocalTI")
+  set(DEFAULT_TEST_TARGET_INFO  
"libcxx.test.target_info.LinuxRemoteTI")
 
   # Allow override with the custom values.
   if(NOT DEFINED LIBUNWIND_TARGET_INFO)


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -99,7 +99,7 @@
 # Remote test configuration.
 if(DEFINED REMOTE_TEST_HOST)
   set(DEFAULT_TEST_EXECUTOR "SSHExecutor('${REMOTE_TEST_HOST}', '${REMOTE_TEST_USER}')")
-  set(DEFAULT_TEST_TARGET_INFO  "libcxx.test.target_info.LinuxLocalTI")
+  set(DEFAULT_TEST_TARGET_INFO  "libcxx.test.target_info.LinuxRemoteTI")
 
   # Allow override with the custom values.
   if(NOT DEFINED LIBUNWIND_TARGET_INFO)
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D72717: [CMake] Disable libc++ filesystem tests in CrossWinToARMLinux cache file

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb added a comment.

In D72717#1824858 , @vvereschaka wrote:

> I think it is optimal to pass this parameter directly in the builder 
> configuration only for libcxx as '-DLLVM_LIT_ARGS="--param 
> enable_filesystem=False"'.


You mean, instead of `ninja check-cxx` just invoke LIT directly and pass this 
argument?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D72717/new/

https://reviews.llvm.org/D72717



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D72850: [CMake] Use LinuxRemoteTI instead of LinuxLocalTI in CrossWinToARMLinux cmake cache

[Sergej Jaskiewicz via Phabricator via cfe-commits]

This revision was automatically updated to reflect the committed changes.
Closed by commit rG383ff4eac1db: [CMake] Use LinuxRemoteTI instead of 
LinuxLocalTI in CrossWinToARMLinux cmake… (authored by broadwaylamb).

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D72850/new/

https://reviews.llvm.org/D72850

Files:
  clang/cmake/caches/CrossWinToARMLinux.cmake


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -90,7 +90,7 @@
 # Remote test configuration.
 if(DEFINED REMOTE_TEST_HOST)
   set(DEFAULT_TEST_EXECUTOR 
"SSHExecutor('${REMOTE_TEST_HOST}', '${REMOTE_TEST_USER}')")
-  set(DEFAULT_TEST_TARGET_INFO  
"libcxx.test.target_info.LinuxLocalTI")
+  set(DEFAULT_TEST_TARGET_INFO  
"libcxx.test.target_info.LinuxRemoteTI")
 
   # Allow override with the custom values.
   if(NOT DEFINED LIBUNWIND_TARGET_INFO)


Index: clang/cmake/caches/CrossWinToARMLinux.cmake
===
--- clang/cmake/caches/CrossWinToARMLinux.cmake
+++ clang/cmake/caches/CrossWinToARMLinux.cmake
@@ -90,7 +90,7 @@
 # Remote test configuration.
 if(DEFINED REMOTE_TEST_HOST)
   set(DEFAULT_TEST_EXECUTOR "SSHExecutor('${REMOTE_TEST_HOST}', '${REMOTE_TEST_USER}')")
-  set(DEFAULT_TEST_TARGET_INFO  "libcxx.test.target_info.LinuxLocalTI")
+  set(DEFAULT_TEST_TARGET_INFO  "libcxx.test.target_info.LinuxRemoteTI")
 
   # Allow override with the custom values.
   if(NOT DEFINED LIBUNWIND_TARGET_INFO)
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D72717: [CMake] Disable libc++ filesystem tests in CrossWinToARMLinux cache file

[Sergej Jaskiewicz via Phabricator via cfe-commits]

broadwaylamb planned changes to this revision.
broadwaylamb added a comment.

In D72717#1848509 , @ldionne wrote:

> It's weird to mix Clang and libc++ test options because they test 
> fundamentally different things. But I don't mind about this patch since it 
> doesn't touch libc++ directly.


I understand that. This was supposed to be a temporary workaround.

I'm going to go with what @vvereschaka has proposed, although that will 
probably require some changes in libc++ CMake configuration.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D72717/new/

https://reviews.llvm.org/D72717



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits