merge main into amd-staging (#701)

passed https://compiler-ci.amd.com/blue/organizations/jenkins/compiler-psdb-amd-staging/detail/compiler-psdb-amd-staging/3005/pipeline/722/
which failed trying to land PR , since approval missing

Author: ronlieb

bolt/lib/Passes/Inliner.cpp

@@ -491,6 +491,32 @@ bool Inliner::inlineCallsInFunction(BinaryFunction &Function) {
         }
       }
 
+      // AArch64 BTI:
+      // If the callee has an indirect tailcall (BR), we would transform it to
+      // an indirect call (BLR) in InlineCall. Because of this, we would have to
+      // update the BTI at the target of the tailcall. However, these targets
+      // are not known. Instead, we skip inlining blocks with indirect
+      // tailcalls.
+      auto HasIndirectTailCall = [&](const BinaryFunction &BF) -> bool {
+        for (const auto &BB : BF) {
+          for (const auto &II : BB) {
+            if (BC.MIB->isIndirectBranch(II) && BC.MIB->isTailCall(II)) {
+              return true;
+            }
+          }
+        }
+        return false;
+      };
+
+      if (BC.isAArch64() && BC.usesBTI() &&
+          HasIndirectTailCall(*TargetFunction)) {
+        ++InstIt;
+        LLVM_DEBUG(dbgs() << "BOLT-DEBUG: Skipping inlining block with tailcall"
+                          << " in " << Function << " : " << BB->getName()
+                          << " to keep BTIs consistent.\n");
+        continue;
+      }
+
       LLVM_DEBUG(dbgs() << "BOLT-DEBUG: inlining call to " << *TargetFunction
                         << " in " << Function << " : " << BB->getName()
                         << ". Count: " << BB->getKnownExecutionCount()

bolt/test/AArch64/inline-bti.s

@@ -0,0 +1,39 @@
+## This test checks that for AArch64 binaries with BTI, we do not inline blocks with indirect tailcalls.
+
+# REQUIRES: system-linux
+
+# RUN: llvm-mc -filetype=obj -triple aarch64-unknown-unknown %s -o %t.o
+# RUN: %clang %cflags -O0 %t.o -o %t.exe -Wl,-q -Wl,-z,force-bti
+# RUN: llvm-bolt --inline-all %t.exe -o %t.bolt --debug 2>&1 | FileCheck %s
+
+# For BTI, we should not inline foo.
+# CHECK: BOLT-DEBUG: Skipping inlining block with tailcall in _Z3barP1A : .LBB01 to keep BTIs consistent.
+# CHECK-NOT: BOLT-INFO: inlined {{[0-9]+}} calls at {{[0-9]+}} call sites in {{[0-9]+}} iteration(s). Change in binary size: {{[0-9]+}} bytes.
+
+	.text
+	.globl	_Z3fooP1A
+	.type	_Z3fooP1A,@function
+_Z3fooP1A:
+	ldr	x8, [x0]
+	ldr	w0, [x8]
+	br x30
+	.size	_Z3fooP1A, .-_Z3fooP1A
+
+	.globl	_Z3barP1A
+	.type	_Z3barP1A,@function
+_Z3barP1A:
+	stp	x29, x30, [sp, #-16]!
+	mov	x29, sp
+	bl	_Z3fooP1A
+	mul	w0, w0, w0
+	ldp	x29, x30, [sp], #16
+	ret
+	.size	_Z3barP1A, .-_Z3barP1A
+
+	.globl	main
+	.p2align	2
+	.type	main,@function
+main:
+	mov	w0, wzr
+	ret
+	.size	main, .-main

clang/docs/ClangStaticAnalyzer.rst

@@ -5,9 +5,9 @@ Clang Static Analyzer
 The Clang Static Analyzer is a source code analysis tool that finds bugs in C, C++, and Objective-C programs.
 It implements *path-sensitive*, *inter-procedural analysis* based on *symbolic execution* technique.
 
-This is the Static Analyzer documentation page.
+The Static Analyzer is a part of Clang; for downloading and installing Clang visit the `LLVM releases page <https://releases.llvm.org/>`_.
 
-See the `Official Tool Page <https://clang-analyzer.llvm.org/>`_.
+This is the documentation page of the Static Analyzer; there is also an old `Official Tool Page <https://clang-analyzer.llvm.org/>`_ which provides a short overview of features and limitations.
 
 .. toctree::
    :caption: Table of Contents

clang/docs/HIPSupport.rst

@@ -376,6 +376,43 @@ Example Usage
       basePtr->virtualFunction(); // Allowed since obj is constructed in device code
    }
 
+Alias Attribute Support
+=======================
+
+Clang supports alias attributes in HIP code, allowing creation of alternative names for functions and variables. 
+ - Aliases work with ``__host__``, ``__device__``, and ``__host__ __device__`` functions and variables.
+ - The alias attribute uses the syntax ``__attribute__((alias("target_name")))``. Both weak and strong aliases are supported.
+ - Outside of ``extern "C"``, the alias target must use the mangled name of the aliasee
+ - The alias is only emitted if the aliasee is emitted on the same side (ie __host__ or __device__), otherwise it is ignored.
+
+Example Usage
+-------------
+
+.. code-block:: c++
+
+   extern "C" {
+     // Host function alias
+     int __HostFunc(void) { return 0; }
+     int HostFunc(void) __attribute__((weak, alias("__HostFunc")));
+
+     // Device function alias
+     __device__ int __DeviceFunc(void) { return 1; }
+     __device__ int DeviceFunc(void) __attribute__((weak, alias("__DeviceFunc")));
+
+     // Host-device function alias
+     __host__ __device__ int __BothFunc(void) { return 2; }
+     __host__ __device__ int BothFunc(void) __attribute__((alias("__BothFunc")));
+
+     // Variable alias
+     int __host_var = 3;
+     extern int __attribute__((weak, alias("__host_var"))) host_var;
+   }
+   // Mangled / overload alias
+   __host__ __device__ float __Four(float f) { return 2.0f * f; }
+   __host__ __device__ int Four(void) __attribute__((weak, alias("_Z6__Fourv")));
+   __host__ __device__ float Four(float f) __attribute__((weak, alias("_Z6__Fourf")));
+
+
 Host and Device Attributes of Default Destructors
 ===================================================
 

clang/docs/analyzer/user-docs.rst

@@ -6,7 +6,6 @@ Contents:
 .. toctree::
    :maxdepth: 2
 
-   user-docs/Installation
    user-docs/CommandLineUsage
    user-docs/Options
    user-docs/UsingWithXCode

clang/docs/analyzer/user-docs/CommandLineUsage.rst

@@ -16,18 +16,19 @@ It is possible, however, to invoke the static analyzer from the command line in
 The following tools are used commonly to run the analyzer from the command line.
 Both tools are wrapper scripts to drive the analysis and the underlying invocations of the Clang compiler:
 
-1. scan-build_ is an old and simple command line tool that emits static analyzer warnings as HTML files while compiling your project. You can view the analysis results in your web browser.
+1. scan-build_ is an old and simple command line tool that emits static analyzer warnings as HTML files while compiling your project. You can view the analysis results in your web browser; the utility script ``scan-view`` can provide a trivial HTTP server that servers these result files.
+    - Is available as a part of the LLVM project (together with ``scan-view``).
     - Useful for individual developers who simply want to view static analysis results at their desk, or in a very simple collaborative environment.
     - Works on all major platforms (Windows, Linux, macOS) and is available as a package in many Linux distributions.
     - Does not include support for cross-translation-unit analysis.
 
 2. CodeChecker_ is a driver and web server that runs the static analyzer on your projects on demand and maintains a database of issues.
+    - Open source, but out-of-tree, i.e. not part of the LLVM project.
     - Perfect for managing large amounts of thee static analyzer warnings in a collaborative environment.
     - Generally much more feature-rich than scan-build.
     - Supports incremental analysis: Results can be stored in a database, subsequent analysis runs can be compared to list the newly added defects.
     - :doc:`CrossTranslationUnit` is supported fully on Linux via CodeChecker.
-    - Can run clang-tidy checkers too.
-    - Open source, but out-of-tree, i.e. not part of the LLVM project.
+    - Can also run clang-tidy checks and various other analysis tools.
 
 scan-build
 ----------

clang/docs/analyzer/user-docs/Installation.rst

@@ -1,37 +1,6 @@
+:orphan:
+
 Obtaining the Static Analyzer
 =============================
 
-This page describes how to download and install the analyzer. Once the analyzer is installed, follow the :doc:`CommandLineUsage` on using the command line to get started analyzing your code.
-
-.. contents::
-   :local:
-
-
-Building the Analyzer from Source
----------------------------------
-
-Currently there are no officially supported binary distributions for the static analyzer.
-You must build Clang and LLVM manually.
-To do so, please follow the instructions for `building Clang from source code <https://clang.llvm.org/get_started.html#build>`_.
-
-Once the Clang is built, you need to add the location of the ``clang`` binary and the locations of the command line utilities (`CodeChecker` or ``scan-build`` and ``scan-view``) to you PATH for :doc:`CommandLineUsage`.
-
-[Legacy] Packaged Builds (Mac OS X)
------------------------------------
-
-Semi-regular pre-built binaries of the analyzer used to be available on Mac OS X. These were built to run on OS X 10.7 and later.
-
-For older builds for MacOS visit https://clang-analyzer.llvm.org/release_notes.html.
-
-Packaged builds for other platforms may eventually be provided, but we need volunteers who are willing to help provide such regular builds. If you wish to help contribute regular builds of the analyzer on other platforms, please get in touch via `LLVM Discourse <https://discourse.llvm.org/>`_.
-
-[Legacy] Using Packaged Builds
-------------------------------
-
-To use the legacy package builds, simply unpack it anywhere. If the build archive has the name **``checker-XXX.tar.bz2``** then the archive will expand to a directory called **``checker-XXX``**. You do not need to place this directory or the contents of this directory in any special place. Uninstalling the analyzer is as simple as deleting this directory.
-
-Most of the files in the **``checker-XXX``** directory will be supporting files for the analyzer that you can simply ignore. Most users will only care about two files, which are located at the top of the **``checker-XXX``** directory:
-
-* **scan-build**: ``scan-build`` is the high-level command line utility for running the analyzer
-* **scan-view**: ``scan-view`` a companion command line utility to ``scan-build``, ``scan-view`` is used to view analysis results generated by ``scan-build``. There is an option that one can pass to ``scan-build`` to cause ``scan-view`` to run as soon as it the analysis of a build completes
-
+The Static Analyzer can be obtained as a part of Clang; for downloading and installing Clang visit the `LLVM releases page <https://releases.llvm.org/>`_. Once the analyzer is installed, follow the :doc:`CommandLineUsage` on using the command line to get started analyzing your code.

clang/include/clang/Basic/BuiltinsX86.td

@@ -210,17 +210,6 @@ let Header = "emmintrin.h", Attributes = [NoThrow, RequireDeclaration] in {
   def _mm_pause : X86LibBuiltin<"void()">;
 }
 
-let Features = "sse2", Attributes = [NoThrow, Const, RequiredVectorWidth<128>] in {
-  def psraw128 : X86Builtin<"_Vector<8, short>(_Vector<8, short>, _Vector<8, short>)">;
-  def psrad128 : X86Builtin<"_Vector<4, int>(_Vector<4, int>, _Vector<4, int>)">;
-  def psrlw128 : X86Builtin<"_Vector<8, short>(_Vector<8, short>, _Vector<8, short>)">;
-  def psrld128 : X86Builtin<"_Vector<4, int>(_Vector<4, int>, _Vector<4, int>)">;
-  def psrlq128 : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, _Vector<2, long long int>)">;
-  def psllw128 : X86Builtin<"_Vector<8, short>(_Vector<8, short>, _Vector<8, short>)">;
-  def pslld128 : X86Builtin<"_Vector<4, int>(_Vector<4, int>, _Vector<4, int>)">;
-  def psllq128 : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, _Vector<2, long long int>)">;
-}
-
 let Features = "sse2", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<128>] in {
   def movmskpd : X86Builtin<"int(_Vector<2, double>)">;
   def pmovmskb128 : X86Builtin<"int(_Vector<16, char>)">;
@@ -261,6 +250,15 @@ let Features = "sse2", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWi
   def vec_ext_v4sf : X86Builtin<"float(_Vector<4, float>, _Constant int)">;
   def vec_ext_v8hi : X86Builtin<"short(_Vector<8, short>, _Constant int)">;
   def vec_set_v8hi : X86Builtin<"_Vector<8, short>(_Vector<8, short>, short, _Constant int)">;
+
+  def psraw128 : X86Builtin<"_Vector<8, short>(_Vector<8, short>, _Vector<8, short>)">;
+  def psrad128 : X86Builtin<"_Vector<4, int>(_Vector<4, int>, _Vector<4, int>)">;
+  def psrlw128 : X86Builtin<"_Vector<8, short>(_Vector<8, short>, _Vector<8, short>)">;
+  def psrld128 : X86Builtin<"_Vector<4, int>(_Vector<4, int>, _Vector<4, int>)">;
+  def psrlq128 : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, _Vector<2, long long int>)">;
+  def psllw128 : X86Builtin<"_Vector<8, short>(_Vector<8, short>, _Vector<8, short>)">;
+  def pslld128 : X86Builtin<"_Vector<4, int>(_Vector<4, int>, _Vector<4, int>)">;
+  def psllq128 : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, _Vector<2, long long int>)">;
 }
 
 let Features = "sse3", Attributes = [NoThrow] in {
@@ -579,14 +577,6 @@ let Features = "avx2", Attributes = [NoThrow, Const, RequiredVectorWidth<256>] i
   def psadbw256
       : X86Builtin<
             "_Vector<4, long long int>(_Vector<32, char>, _Vector<32, char>)">;
-  def psllw256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, _Vector<8, short>)">;
-  def pslld256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Vector<4, int>)">;
-  def psllq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Vector<2, long long int>)">;
-  def psraw256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, _Vector<8, short>)">;
-  def psrad256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Vector<4, int>)">;
-  def psrlw256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, _Vector<8, short>)">;
-  def psrld256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Vector<4, int>)">;
-  def psrlq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Vector<2, long long int>)">;
   def permdf256 : X86Builtin<"_Vector<4, double>(_Vector<4, double>, _Constant int)">;
   def permti256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Vector<4, long long int>, _Constant int)">;
   def permdi256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Constant int)">;
@@ -663,6 +653,15 @@ let Features = "avx2", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWi
 
   def permvarsi256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Vector<8, int>)">;
   def permvarsf256 : X86Builtin<"_Vector<8, float>(_Vector<8, float>, _Vector<8, int>)">;
+
+  def psllw256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, _Vector<8, short>)">;
+  def pslld256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Vector<4, int>)">;
+  def psllq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Vector<2, long long int>)">;
+  def psraw256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, _Vector<8, short>)">;
+  def psrad256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Vector<4, int>)">;
+  def psrlw256 : X86Builtin<"_Vector<16, short>(_Vector<16, short>, _Vector<8, short>)">;
+  def psrld256 : X86Builtin<"_Vector<8, int>(_Vector<8, int>, _Vector<4, int>)">;
+  def psrlq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Vector<2, long long int>)">;
 }
 
 let Features = "avx2", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<128>] in {
@@ -1926,16 +1925,13 @@ let Features = "avx512vl", Attributes = [NoThrow, Const, Constexpr, RequiredVect
   def prorq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Constant int)">;
 }
 
-let Features = "avx512bw", Attributes = [NoThrow, Const, RequiredVectorWidth<512>] in {
-  def psllw512 : X86Builtin<"_Vector<32, short>(_Vector<32, short>, _Vector<8, short>)">;
-}
-
 let Features = "avx512bw", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<512>] in {
   def pmaddubsw512 : X86Builtin<"_Vector<32, short>(_Vector<64, char>, _Vector<64, char>)">;
   def pmaddwd512 : X86Builtin<"_Vector<16, int>(_Vector<32, short>, _Vector<32, short>)">;
   def psllv32hi : X86Builtin<"_Vector<32, short>(_Vector<32, short>, _Vector<32, short>)">;
   def pshufhw512 : X86Builtin<"_Vector<32, short>(_Vector<32, short>, _Constant int)">;
   def pshuflw512 : X86Builtin<"_Vector<32, short>(_Vector<32, short>, _Constant int)">;
+  def psllw512 : X86Builtin<"_Vector<32, short>(_Vector<32, short>, _Vector<8, short>)">;
 }
 
 let Features = "avx512bw,avx512vl", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<256>] in {
@@ -1991,7 +1987,7 @@ let Features = "avx512vl", Attributes = [NoThrow, Const, Constexpr, RequiredVect
   def psravq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Vector<4, long long int>)">;
 }
 
-let Features = "avx512bw", Attributes = [NoThrow, Const, RequiredVectorWidth<512>] in {
+let Features = "avx512bw", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<512>] in {
   def psraw512
       : X86Builtin<"_Vector<32, short>(_Vector<32, short>, _Vector<8, short>)">;
   def psrlw512
@@ -2308,25 +2304,17 @@ let Features = "avx512f",
   def psraqi512 : X86Builtin<"_Vector<8, long long int>(_Vector<8, long long int>, int)">;
 }
 
-let Features = "avx512vl", Attributes = [NoThrow, Const, RequiredVectorWidth<128>] in {
+let Features = "avx512vl", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<128>] in {
   def psraq128 : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, _Vector<2, long long int>)">;
-}
-
-let Features = "avx512vl", Attributes = [NoThrow, Const, RequiredVectorWidth<256>] in {
-  def psraq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Vector<2, long long int>)">;
-}
-
-let Features = "avx512vl",
-    Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<128>] in {
   def psraqi128 : X86Builtin<"_Vector<2, long long int>(_Vector<2, long long int>, int)">;
 }
 
-let Features = "avx512vl",
-    Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<256>] in {
+let Features = "avx512vl", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<256>] in {
+  def psraq256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, _Vector<2, long long int>)">;
   def psraqi256 : X86Builtin<"_Vector<4, long long int>(_Vector<4, long long int>, int)">;
 }
 
-let Features = "avx512f", Attributes = [NoThrow, Const, RequiredVectorWidth<512>] in {
+let Features = "avx512f", Attributes = [NoThrow, Const, Constexpr, RequiredVectorWidth<512>] in {
   def pslld512 : X86Builtin<"_Vector<16, int>(_Vector<16, int>, _Vector<4, int>)">;
   def psllq512 : X86Builtin<"_Vector<8, long long int>(_Vector<8, long long int>, _Vector<2, long long int>)">;
   def psrad512 : X86Builtin<"_Vector<16, int>(_Vector<16, int>, _Vector<4, int>)">;