Merge tag 'perf-core-for-mingo-20160712' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core

Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo:

User visible changes:

- Add demangling of symbols in programs written in the Rust language (David Tolnay)

- Add support for tracepoints in the python binding, including an example, that
  sets up and parses sched:sched_switch events, tools/perf/python/tracepoint.py
  (Jiri Olsa)

- Introduce --stdio-color to set up the color output mode selection in
  'annotate' and 'report', allowing emit color escape sequences when
  redirecting the output of these tools (Arnaldo Carvalho de Melo)

Infrastructure changes:

- Various tweaks to allow the 'perf trace' beautifiers to build without using
  kernel headers and in a wider range of Linux distributions/releases
  (Arnaldo Carvalho de Melo)

- Stop using kernel source files, instead copy what is needed and
  check when the original kernel source file gets modified, warning
  the developers about it. This helps in building the tool in older
  systems and even in recent ones, for just added kernel features
  for which ABI details (struct changes, defines, etc) still are not
  available on system headers (Arnaldo Carvalho de Melo)

- Be consistent in how to use strerror_r, adding a wrapper that makes sure that
  it returns a pointer to passed buffer, and using the XSI variant, that is
  available in more libc implementations (Arnaldo Carvalho de Melo)

- Avoid checking code drift on busibox's diff perf intel-pt-decoder, as it
  doesn't have the '-I' command line switch to check for regexps
  (Arnaldo Carvalho de Melo)

- Add missing headers in various places (Arnaldo Carvalho de Melo)

- Remove unneeded headers from various other places (Arnaldo Carvalho de Melo)

- Add feature detection for gelf_getnote(), disabling SDT support if not
  present (Arnaldo Carvalho de Melo)

- Fix oddities with GCC 5.3.0 by initializing some variables
  (Arnaldo Carvalho de Melo)

- With those changes in place perf now builds on Alpine Linux 3.4, in addition to
  on centos (5, 6, 7), debian (7, 8, experimental), fedora (21, 22, 23, 24, rawhide),
  mageia 5, opensuse (13.2, 42.1) and ubuntu (12.04.5, 14.04.4, 15.10, 16.04) and
  will be test build on those systems prior to future pull requests.

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>

100 files changed, 1138 insertions, 299 deletions

diff --git a/tools/perf/Documentation/perf-annotate.txt b/tools/perf/Documentation/perf-annotate.txt
index 778f54d4d0bd..8ffbd272952d 100644
--- a/tools/perf/Documentation/perf-annotate.txt
+++ b/tools/perf/Documentation/perf-annotate.txt
@@ -61,6 +61,13 @@ OPTIONS
 
 --stdio:: Use the stdio interface.
 
+--stdio-color::
+        'always', 'never' or 'auto', allowing configuring color output
+        via the command line, in addition to via "color.ui" .perfconfig.
+        Use '--stdio-color always' to generate color even when redirecting
+        to a pipe or file. Using just '--stdio-color' is equivalent to
+        using 'always'.
+
 --tui:: Use the TUI interface. Use of --tui requires a tty, if one is not
         present, as when piping to other commands, the stdio interface is
         used. This interfaces starts by centering on the line with more
diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt
index 9cbddc290aff..2d1746295abf 100644
--- a/tools/perf/Documentation/perf-report.txt
+++ b/tools/perf/Documentation/perf-report.txt
@@ -265,6 +265,13 @@ OPTIONS
 
 --stdio:: Use the stdio interface.
 
+--stdio-color::
+        'always', 'never' or 'auto', allowing configuring color output
+        via the command line, in addition to via "color.ui" .perfconfig.
+        Use '--stdio-color always' to generate color even when redirecting
+        to a pipe or file. Using just '--stdio-color' is equivalent to
+        using 'always'.
+
 --tui:: Use the TUI interface, that is integrated with annotate and allows
         zooming into DSOs or threads, among other features. Use of --tui
         requires a tty, if one is not present, as when piping to other
diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST
index 8c8c6b9ce915..923eda2e7d52 100644
--- a/tools/perf/MANIFEST
+++ b/tools/perf/MANIFEST
@@ -12,13 +12,23 @@ tools/arch/sparc/include/asm/barrier_32.h
 tools/arch/sparc/include/asm/barrier_64.h
 tools/arch/tile/include/asm/barrier.h
 tools/arch/x86/include/asm/barrier.h
+tools/arch/x86/include/asm/cpufeatures.h
+tools/arch/x86/include/asm/disabled-features.h
+tools/arch/x86/include/asm/required-features.h
+tools/arch/x86/include/uapi/asm/svm.h
+tools/arch/x86/include/uapi/asm/vmx.h
+tools/arch/x86/include/uapi/asm/kvm.h
+tools/arch/x86/include/uapi/asm/kvm_perf.h
+tools/arch/x86/lib/memcpy_64.S
+tools/arch/x86/lib/memset_64.S
+tools/arch/s390/include/uapi/asm/kvm_perf.h
+tools/arch/s390/include/uapi/asm/sie.h
 tools/arch/xtensa/include/asm/barrier.h
 tools/scripts
 tools/build
 tools/arch/x86/include/asm/atomic.h
 tools/arch/x86/include/asm/rmwcc.h
 tools/lib/traceevent
-tools/lib/bpf
 tools/lib/api
 tools/lib/bpf
 tools/lib/subcmd
@@ -29,6 +39,9 @@ tools/lib/symbol/kallsyms.c
 tools/lib/symbol/kallsyms.h
 tools/lib/find_bit.c
 tools/lib/bitmap.c
+tools/lib/str_error_r.c
+tools/lib/vsprintf.c
+tools/include/asm/alternative-asm.h
 tools/include/asm/atomic.h
 tools/include/asm/barrier.h
 tools/include/asm/bug.h
@@ -52,10 +65,15 @@ tools/include/linux/hash.h
 tools/include/linux/kernel.h
 tools/include/linux/list.h
 tools/include/linux/log2.h
+tools/include/uapi/linux/bpf.h
+tools/include/uapi/linux/bpf_common.h
+tools/include/uapi/linux/hw_breakpoint.h
+tools/include/uapi/linux/perf_event.h
 tools/include/linux/poison.h
 tools/include/linux/rbtree.h
 tools/include/linux/rbtree_augmented.h
 tools/include/linux/string.h
+tools/include/linux/stringify.h
 tools/include/linux/types.h
 tools/include/linux/err.h
 tools/include/linux/bitmap.h
@@ -64,31 +82,12 @@ include/asm-generic/bitops/const_hweight.h
 include/asm-generic/bitops/fls64.h
 include/asm-generic/bitops/__fls.h
 include/asm-generic/bitops/fls.h
-include/linux/perf_event.h
 include/linux/list.h
 include/linux/hash.h
-include/linux/stringify.h
 include/linux/swab.h
 arch/*/include/asm/unistd*.h
 arch/*/include/uapi/asm/unistd*.h
-arch/*/include/uapi/asm/perf_regs.h
-arch/*/lib/memcpy*.S
-arch/*/lib/memset*.S
-arch/*/include/asm/*features.h
+tools/arch/*/include/uapi/asm/perf_regs.h
 include/linux/poison.h
-include/linux/hw_breakpoint.h
-include/uapi/linux/perf_event.h
-include/uapi/linux/bpf.h
-include/uapi/linux/bpf_common.h
 include/uapi/linux/const.h
 include/uapi/linux/swab.h
-include/uapi/linux/hw_breakpoint.h
-arch/x86/include/asm/svm.h
-arch/x86/include/asm/vmx.h
-arch/x86/include/asm/kvm_host.h
-arch/x86/include/uapi/asm/svm.h
-arch/x86/include/uapi/asm/vmx.h
-arch/x86/include/uapi/asm/kvm.h
-arch/x86/include/uapi/asm/kvm_perf.h
-arch/s390/include/uapi/asm/sie.h
-arch/s390/include/uapi/asm/kvm_perf.h
diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
index d0a2cb11296f..feb2c66b110b 100644
--- a/tools/perf/Makefile.perf
+++ b/tools/perf/Makefile.perf
@@ -345,6 +345,69 @@ export srctree OUTPUT RM CC LD AR CFLAGS V BISON FLEX AWK
 include $(srctree)/tools/build/Makefile.include
 
 $(PERF_IN): prepare FORCE
+        @(test -f ../../include/uapi/linux/perf_event.h && ( \
+        (diff -B ../include/uapi/linux/perf_event.h ../../include/uapi/linux/perf_event.h >/dev/null) \
+        || echo "Warning: tools/include/uapi/linux/perf_event.h differs from kernel" >&2 )) || true
+        @(test -f ../../include/uapi/linux/hw_breakpoint.h && ( \
+        (diff -B ../include/uapi/linux/hw_breakpoint.h ../../include/uapi/linux/hw_breakpoint.h >/dev/null) \
+        || echo "Warning: tools/include/uapi/linux/hw_breakpoint.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/x86/include/asm/disabled-features.h && ( \
+        (diff -B ../arch/x86/include/asm/disabled-features.h ../../arch/x86/include/asm/disabled-features.h >/dev/null) \
+        || echo "Warning: tools/arch/x86/include/asm/disabled-features.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/x86/include/asm/required-features.h && ( \
+        (diff -B ../arch/x86/include/asm/required-features.h ../../arch/x86/include/asm/required-features.h >/dev/null) \
+        || echo "Warning: tools/arch/x86/include/asm/required-features.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/x86/include/asm/cpufeatures.h && ( \
+        (diff -B ../arch/x86/include/asm/cpufeatures.h ../../arch/x86/include/asm/cpufeatures.h >/dev/null) \
+        || echo "Warning: tools/arch/x86/include/asm/cpufeatures.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/x86/lib/memcpy_64.S && ( \
+        (diff -B ../arch/x86/lib/memcpy_64.S ../../arch/x86/lib/memcpy_64.S >/dev/null) \
+        || echo "Warning: tools/arch/x86/lib/memcpy_64.S differs from kernel" >&2 )) || true
+        @(test -f ../../arch/x86/lib/memset_64.S && ( \
+        (diff -B ../arch/x86/lib/memset_64.S ../../arch/x86/lib/memset_64.S >/dev/null) \
+        || echo "Warning: tools/arch/x86/lib/memset_64.S differs from kernel" >&2 )) || true
+        @(test -f ../../arch/arm/include/uapi/asm/perf_regs.h && ( \
+        (diff -B ../arch/arm/include/uapi/asm/perf_regs.h ../../arch/arm/include/uapi/asm/perf_regs.h >/dev/null) \
+        || echo "Warning: tools/arch/arm/include/uapi/asm/perf_regs.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/arm64/include/uapi/asm/perf_regs.h && ( \
+        (diff -B ../arch/arm64/include/uapi/asm/perf_regs.h ../../arch/arm64/include/uapi/asm/perf_regs.h >/dev/null) \
+        || echo "Warning: tools/arch/arm64/include/uapi/asm/perf_regs.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/powerpc/include/uapi/asm/perf_regs.h && ( \
+        (diff -B ../arch/powerpc/include/uapi/asm/perf_regs.h ../../arch/powerpc/include/uapi/asm/perf_regs.h >/dev/null) \
+        || echo "Warning: tools/arch/powerpc/include/uapi/asm/perf_regs.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/x86/include/uapi/asm/perf_regs.h && ( \
+        (diff -B ../arch/x86/include/uapi/asm/perf_regs.h ../../arch/x86/include/uapi/asm/perf_regs.h >/dev/null) \
+        || echo "Warning: tools/arch/x86/include/uapi/asm/perf_regs.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/x86/include/uapi/asm/kvm.h && ( \
+        (diff -B ../arch/x86/include/uapi/asm/kvm.h ../../arch/x86/include/uapi/asm/kvm.h >/dev/null) \
+        || echo "Warning: tools/arch/x86/include/uapi/asm/kvm.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/x86/include/uapi/asm/kvm_perf.h && ( \
+        (diff -B ../arch/x86/include/uapi/asm/kvm_perf.h ../../arch/x86/include/uapi/asm/kvm_perf.h >/dev/null) \
+        || echo "Warning: tools/arch/x86/include/uapi/asm/kvm_perf.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/x86/include/uapi/asm/svm.h && ( \
+        (diff -B ../arch/x86/include/uapi/asm/svm.h ../../arch/x86/include/uapi/asm/svm.h >/dev/null) \
+        || echo "Warning: tools/arch/x86/include/uapi/asm/svm.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/x86/include/uapi/asm/vmx.h && ( \
+        (diff -B ../arch/x86/include/uapi/asm/vmx.h ../../arch/x86/include/uapi/asm/vmx.h >/dev/null) \
+        || echo "Warning: tools/arch/x86/include/uapi/asm/vmx.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/powerpc/include/uapi/asm/kvm.h && ( \
+        (diff -B ../arch/powerpc/include/uapi/asm/kvm.h ../../arch/powerpc/include/uapi/asm/kvm.h >/dev/null) \
+        || echo "Warning: tools/arch/powerpc/include/uapi/asm/kvm.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/s390/include/uapi/asm/kvm.h && ( \
+        (diff -B ../arch/s390/include/uapi/asm/kvm.h ../../arch/s390/include/uapi/asm/kvm.h >/dev/null) \
+        || echo "Warning: tools/arch/s390/include/uapi/asm/kvm.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/s390/include/uapi/asm/kvm_perf.h && ( \
+        (diff -B ../arch/s390/include/uapi/asm/kvm_perf.h ../../arch/s390/include/uapi/asm/kvm_perf.h >/dev/null) \
+        || echo "Warning: tools/arch/s390/include/uapi/asm/kvm_perf.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/s390/include/uapi/asm/sie.h && ( \
+        (diff -B ../arch/s390/include/uapi/asm/sie.h ../../arch/s390/include/uapi/asm/sie.h >/dev/null) \
+        || echo "Warning: tools/arch/s390/include/uapi/asm/sie.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/arm/include/uapi/asm/kvm.h && ( \
+        (diff -B ../arch/arm/include/uapi/asm/kvm.h ../../arch/arm/include/uapi/asm/kvm.h >/dev/null) \
+        || echo "Warning: tools/arch/arm/include/uapi/asm/kvm.h differs from kernel" >&2 )) || true
+        @(test -f ../../arch/arm64/include/uapi/asm/kvm.h && ( \
+        (diff -B ../arch/arm64/include/uapi/asm/kvm.h ../../arch/arm64/include/uapi/asm/kvm.h >/dev/null) \
+        || echo "Warning: tools/arch/arm64/include/uapi/asm/kvm.h differs from kernel" >&2 )) || true
         $(Q)$(MAKE) $(build)=perf
 
 $(OUTPUT)perf: $(PERFLIBS) $(PERF_IN) $(LIBTRACEEVENT_DYNAMIC_LIST)
diff --git a/tools/perf/arch/common.c b/tools/perf/arch/common.c
index ee6966812a5a..886dd2aaff0d 100644
--- a/tools/perf/arch/common.c
+++ b/tools/perf/arch/common.c
@@ -1,6 +1,7 @@
 #include <stdio.h>
 #include <sys/utsname.h>
 #include "common.h"
+#include "../util/util.h"
 #include "../util/debug.h"
 
 const char *const arm_triplets[] = {
diff --git a/tools/perf/arch/x86/tests/rdpmc.c b/tools/perf/arch/x86/tests/rdpmc.c
index 72193f19d6d7..500cf96db979 100644
--- a/tools/perf/arch/x86/tests/rdpmc.c
+++ b/tools/perf/arch/x86/tests/rdpmc.c
@@ -1,12 +1,16 @@
+#include <errno.h>
 #include <unistd.h>
 #include <stdlib.h>
 #include <signal.h>
 #include <sys/mman.h>
+#include <sys/types.h>
+#include <sys/wait.h>
 #include <linux/types.h>
 #include "perf.h"
 #include "debug.h"
 #include "tests/tests.h"
 #include "cloexec.h"
+#include "util.h"
 #include "arch-tests.h"
 
 static u64 rdpmc(unsigned int counter)
@@ -111,14 +115,14 @@ static int __test__rdpmc(void)
         if (fd < 0) {
                 pr_err("Error: sys_perf_event_open() syscall returned "
                        "with %d (%s)\n", fd,
-                       strerror_r(errno, sbuf, sizeof(sbuf)));
+                       str_error_r(errno, sbuf, sizeof(sbuf)));
                 return -1;
         }
 
         addr = mmap(NULL, page_size, PROT_READ, MAP_SHARED, fd, 0);
         if (addr == (void *)(-1)) {
                 pr_err("Error: mmap() syscall returned with (%s)\n",
-                       strerror_r(errno, sbuf, sizeof(sbuf)));
+                       str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out_close;
         }
 
diff --git a/tools/perf/bench/futex-hash.c b/tools/perf/bench/futex-hash.c
index 0999ac536d86..8024cd5febd2 100644
--- a/tools/perf/bench/futex-hash.c
+++ b/tools/perf/bench/futex-hash.c
@@ -8,18 +8,23 @@
  * many threads and futexes as possible.
  */
 
-#include "../perf.h"
-#include "../util/util.h"
+/* For the CLR_() macros */
+#include <pthread.h>
+
+#include <errno.h>
+#include <signal.h>
+#include <stdlib.h>
+#include <linux/compiler.h>
+#include <linux/kernel.h>
+#include <sys/time.h>
+
 #include "../util/stat.h"
 #include <subcmd/parse-options.h>
-#include "../util/header.h"
 #include "bench.h"
 #include "futex.h"
 
 #include <err.h>
-#include <stdlib.h>
 #include <sys/time.h>
-#include <pthread.h>
 
 static unsigned int nthreads = 0;
 static unsigned int nsecs    = 10;
diff --git a/tools/perf/bench/futex-lock-pi.c b/tools/perf/bench/futex-lock-pi.c
index 6952db65508a..936d89d30483 100644
--- a/tools/perf/bench/futex-lock-pi.c
+++ b/tools/perf/bench/futex-lock-pi.c
@@ -2,18 +2,21 @@
  * Copyright (C) 2015 Davidlohr Bueso.
  */
 
-#include "../perf.h"
-#include "../util/util.h"
+/* For the CLR_() macros */
+#include <pthread.h>
+
+#include <signal.h>
 #include "../util/stat.h"
 #include <subcmd/parse-options.h>
-#include "../util/header.h"
+#include <linux/compiler.h>
+#include <linux/kernel.h>
+#include <errno.h>
 #include "bench.h"
 #include "futex.h"
 
 #include <err.h>
 #include <stdlib.h>
 #include <sys/time.h>
-#include <pthread.h>
 
 struct worker {
         int tid;
diff --git a/tools/perf/bench/futex-requeue.c b/tools/perf/bench/futex-requeue.c
index 718238683013..f96e22ed9f87 100644
--- a/tools/perf/bench/futex-requeue.c
+++ b/tools/perf/bench/futex-requeue.c
@@ -8,18 +8,21 @@
  * requeues without waking up any tasks -- thus mimicking a regular futex_wait.
  */
 
-#include "../perf.h"
-#include "../util/util.h"
+/* For the CLR_() macros */
+#include <pthread.h>
+
+#include <signal.h>
 #include "../util/stat.h"
 #include <subcmd/parse-options.h>
-#include "../util/header.h"
+#include <linux/compiler.h>
+#include <linux/kernel.h>
+#include <errno.h>
 #include "bench.h"
 #include "futex.h"
 
 #include <err.h>
 #include <stdlib.h>
 #include <sys/time.h>
-#include <pthread.h>
 
 static u_int32_t futex1 = 0, futex2 = 0;
 
diff --git a/tools/perf/bench/futex-wake-parallel.c b/tools/perf/bench/futex-wake-parallel.c
index 91aaf2a1fa90..4a2ecd7438ca 100644
--- a/tools/perf/bench/futex-wake-parallel.c
+++ b/tools/perf/bench/futex-wake-parallel.c
@@ -7,18 +7,21 @@
  * it can be used to measure futex_wake() changes.
  */
 
-#include "../perf.h"
-#include "../util/util.h"
+/* For the CLR_() macros */
+#include <pthread.h>
+
+#include <signal.h>
 #include "../util/stat.h"
 #include <subcmd/parse-options.h>
-#include "../util/header.h"
+#include <linux/compiler.h>
+#include <linux/kernel.h>
+#include <errno.h>
 #include "bench.h"
 #include "futex.h"
 
 #include <err.h>
 #include <stdlib.h>
 #include <sys/time.h>
-#include <pthread.h>
 
 struct thread_data {
         pthread_t worker;
diff --git a/tools/perf/bench/futex-wake.c b/tools/perf/bench/futex-wake.c
index f416bd705f66..87d8f4f292d9 100644
--- a/tools/perf/bench/futex-wake.c
+++ b/tools/perf/bench/futex-wake.c
@@ -8,18 +8,21 @@
  * one or more tasks, and thus the waitqueue is never empty.
  */
 
-#include "../perf.h"
-#include "../util/util.h"
+/* For the CLR_() macros */
+#include <pthread.h>
+
+#include <signal.h>
 #include "../util/stat.h"
 #include <subcmd/parse-options.h>
-#include "../util/header.h"
+#include <linux/compiler.h>
+#include <linux/kernel.h>
+#include <errno.h>
 #include "bench.h"
 #include "futex.h"
 
 #include <err.h>
 #include <stdlib.h>
 #include <sys/time.h>
-#include <pthread.h>
 
 /* all threads will block on the same futex */
 static u_int32_t futex1 = 0;
diff --git a/tools/perf/bench/mem-memcpy-x86-64-asm.S b/tools/perf/bench/mem-memcpy-x86-64-asm.S
index 5c3cce082cb8..f700369bb0f6 100644
--- a/tools/perf/bench/mem-memcpy-x86-64-asm.S
+++ b/tools/perf/bench/mem-memcpy-x86-64-asm.S
@@ -6,7 +6,7 @@
 #define globl p2align 4; .globl
 #define _ASM_EXTABLE_FAULT(x, y)
 
-#include "../../../arch/x86/lib/memcpy_64.S"
+#include "../../arch/x86/lib/memcpy_64.S"
 /*
  * We need to provide note.GNU-stack section, saying that we want
  * NOT executable stack. Otherwise the final linking will assume that
diff --git a/tools/perf/bench/mem-memset-x86-64-asm.S b/tools/perf/bench/mem-memset-x86-64-asm.S
index de278784c866..58407aa24c1b 100644
--- a/tools/perf/bench/mem-memset-x86-64-asm.S
+++ b/tools/perf/bench/mem-memset-x86-64-asm.S
@@ -1,7 +1,7 @@
 #define memset MEMSET /* don't hide glibc's memset() */
 #define altinstr_replacement text
 #define globl p2align 4; .globl
-#include "../../../arch/x86/lib/memset_64.S"
+#include "../../arch/x86/lib/memset_64.S"
 
 /*
  * We need to provide note.GNU-stack section, saying that we want
diff --git a/tools/perf/bench/numa.c b/tools/perf/bench/numa.c
index 7500d959d7eb..f7f530081aa9 100644
--- a/tools/perf/bench/numa.c
+++ b/tools/perf/bench/numa.c
@@ -4,6 +4,9 @@
  * numa: Simulate NUMA-sensitive workload and measure their NUMA performance
  */
 
+/* For the CLR_() macros */
+#include <pthread.h>
+
 #include "../perf.h"
 #include "../builtin.h"
 #include "../util/util.h"
@@ -21,7 +24,6 @@
 #include <stdlib.h>
 #include <string.h>
 #include <unistd.h>
-#include <pthread.h>
 #include <sys/mman.h>
 #include <sys/time.h>
 #include <sys/resource.h>
diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c
index b15e7683f38c..9c1034d81b4f 100644
--- a/tools/perf/builtin-annotate.c
+++ b/tools/perf/builtin-annotate.c
@@ -339,6 +339,9 @@ int cmd_annotate(int argc, const char **argv, const char *prefix __maybe_unused)
                     "Show event group information together"),
         OPT_BOOLEAN(0, "show-total-period", &symbol_conf.show_total_period,
                     "Show a column with the sum of periods"),
+        OPT_CALLBACK_DEFAULT(0, "stdio-color", NULL, "mode",
+                             "'always' (default), 'never' or 'auto' only applicable to --stdio mode",
+                             stdio__config_color, "always"),
         OPT_END()
         };
         int ret = hists__init();
diff --git a/tools/perf/builtin-buildid-cache.c b/tools/perf/builtin-buildid-cache.c
index 76a4d03c7cd0..30e2b2cb2421 100644
--- a/tools/perf/builtin-buildid-cache.c
+++ b/tools/perf/builtin-buildid-cache.c
@@ -351,7 +351,7 @@ int cmd_buildid_cache(int argc, const char **argv,
                                                 continue;
                                         }
                                         pr_warning("Couldn't add %s: %s\n",
-                                                   pos->s, strerror_r(errno, sbuf, sizeof(sbuf)));
+                                                   pos->s, str_error_r(errno, sbuf, sizeof(sbuf)));
                                 }
 
                         strlist__delete(list);
@@ -369,7 +369,7 @@ int cmd_buildid_cache(int argc, const char **argv,
                                                 continue;
                                         }
                                         pr_warning("Couldn't remove %s: %s\n",
-                                                   pos->s, strerror_r(errno, sbuf, sizeof(sbuf)));
+                                                   pos->s, str_error_r(errno, sbuf, sizeof(sbuf)));
                                 }
 
                         strlist__delete(list);
@@ -387,7 +387,7 @@ int cmd_buildid_cache(int argc, const char **argv,
                                                 continue;
                                         }
                                         pr_warning("Couldn't remove %s: %s\n",
-                                                   pos->s, strerror_r(errno, sbuf, sizeof(sbuf)));
+                                                   pos->s, str_error_r(errno, sbuf, sizeof(sbuf)));
                                 }
 
                         strlist__delete(list);
@@ -408,7 +408,7 @@ int cmd_buildid_cache(int argc, const char **argv,
                                                 continue;
                                         }
                                         pr_warning("Couldn't update %s: %s\n",
-                                                   pos->s, strerror_r(errno, sbuf, sizeof(sbuf)));
+                                                   pos->s, str_error_r(errno, sbuf, sizeof(sbuf)));
                                 }
 
                         strlist__delete(list);
diff --git a/tools/perf/builtin-help.c b/tools/perf/builtin-help.c
index 268ab732b8aa..3bdb2c78a21b 100644
--- a/tools/perf/builtin-help.c
+++ b/tools/perf/builtin-help.c
@@ -117,7 +117,7 @@ static void exec_woman_emacs(const char *path, const char *page)
                         free(man_page);
                 }
                 warning("failed to exec '%s': %s", path,
-                        strerror_r(errno, sbuf, sizeof(sbuf)));
+                        str_error_r(errno, sbuf, sizeof(sbuf)));
         }
 }
 
@@ -150,7 +150,7 @@ static void exec_man_konqueror(const char *path, const char *page)
                         free(man_page);
                 }
                 warning("failed to exec '%s': %s", path,
-                        strerror_r(errno, sbuf, sizeof(sbuf)));
+                        str_error_r(errno, sbuf, sizeof(sbuf)));
         }
 }
 
@@ -162,7 +162,7 @@ static void exec_man_man(const char *path, const char *page)
                 path = "man";
         execlp(path, "man", page, NULL);
         warning("failed to exec '%s': %s", path,
-                strerror_r(errno, sbuf, sizeof(sbuf)));
+                str_error_r(errno, sbuf, sizeof(sbuf)));
 }
 
 static void exec_man_cmd(const char *cmd, const char *page)
@@ -175,7 +175,7 @@ static void exec_man_cmd(const char *cmd, const char *page)
                 free(shell_cmd);
         }
         warning("failed to exec '%s': %s", cmd,
-                strerror_r(errno, sbuf, sizeof(sbuf)));
+                str_error_r(errno, sbuf, sizeof(sbuf)));
 }
 
 static void add_man_viewer(const char *name)
diff --git a/tools/perf/builtin-kvm.c b/tools/perf/builtin-kvm.c
index f4efef9d1eb3..5e2127e04f83 100644
--- a/tools/perf/builtin-kvm.c
+++ b/tools/perf/builtin-kvm.c
@@ -1018,13 +1018,13 @@ static int kvm_live_open_events(struct perf_kvm_stat *kvm)
         err = perf_evlist__open(evlist);
         if (err < 0) {
                 printf("Couldn't create the events: %s\n",
-                       strerror_r(errno, sbuf, sizeof(sbuf)));
+                       str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out;
         }
 
         if (perf_evlist__mmap(evlist, kvm->opts.mmap_pages, false) < 0) {
                 ui__error("Failed to mmap the events: %s\n",
-                          strerror_r(errno, sbuf, sizeof(sbuf)));
+                          str_error_r(errno, sbuf, sizeof(sbuf)));
                 perf_evlist__close(evlist);
                 goto out;
         }
diff --git a/tools/perf/builtin-probe.c b/tools/perf/builtin-probe.c
index a1a5cd1b8d60..c6d890ad2c1a 100644
--- a/tools/perf/builtin-probe.c
+++ b/tools/perf/builtin-probe.c
@@ -308,7 +308,7 @@ static void pr_err_with_code(const char *msg, int err)
 
         pr_err("%s", msg);
         pr_debug(" Reason: %s (Code: %d)",
-                 strerror_r(-err, sbuf, sizeof(sbuf)), err);
+                 str_error_r(-err, sbuf, sizeof(sbuf)), err);
         pr_err("\n");
 }
 
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index b2b3b600adf5..d9f5cc3a3667 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -361,7 +361,7 @@ static int record__mmap_evlist(struct record *rec,
                         return -errno;
                 } else {
                         pr_err("failed to mmap with %d (%s)\n", errno,
-                                strerror_r(errno, msg, sizeof(msg)));
+                                str_error_r(errno, msg, sizeof(msg)));
                         if (errno)
                                 return -errno;
                         else
@@ -407,7 +407,7 @@ try_again:
         if (perf_evlist__apply_filters(evlist, &pos)) {
                 error("failed to set filter \"%s\" on event %s with %d (%s)\n",
                         pos->filter, perf_evsel__name(pos), errno,
-                        strerror_r(errno, msg, sizeof(msg)));
+                        str_error_r(errno, msg, sizeof(msg)));
                 rc = -1;
                 goto out;
         }
@@ -1003,7 +1003,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 
         if (forks && workload_exec_errno) {
                 char msg[STRERR_BUFSIZE];
-                const char *emsg = strerror_r(workload_exec_errno, msg, sizeof(msg));
+                const char *emsg = str_error_r(workload_exec_errno, msg, sizeof(msg));
                 pr_err("Workload failed: %s\n", emsg);
                 err = -1;
                 goto out_child;
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index f6cb357986c6..949e5a15c960 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -817,6 +817,9 @@ int cmd_report(int argc, const char **argv, const char *prefix __maybe_unused)
                     "Show raw trace event output (do not use print fmt or plugins)"),
         OPT_BOOLEAN(0, "hierarchy", &symbol_conf.report_hierarchy,
                     "Show entries in a hierarchy"),
+        OPT_CALLBACK_DEFAULT(0, "stdio-color", NULL, "mode",
+                             "'always' (default), 'never' or 'auto' only applicable to --stdio mode",
+                             stdio__config_color, "always"),
         OPT_END()
         };
         struct perf_data_file file = {
diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c
index afa057666c2a..0dfe8df2ab9b 100644
--- a/tools/perf/builtin-sched.c
+++ b/tools/perf/builtin-sched.c
@@ -494,7 +494,7 @@ force_again:
                 }
                 pr_err("Error: sys_perf_event_open() syscall returned "
                        "with %d (%s)\n%s", fd,
-                       strerror_r(errno, sbuf, sizeof(sbuf)), info);
+                       str_error_r(errno, sbuf, sizeof(sbuf)), info);
                 exit(EXIT_FAILURE);
         }
         return fd;
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index c367a43525e6..8c5a3bfdfdd7 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -596,7 +596,7 @@ try_again:
         if (perf_evlist__apply_filters(evsel_list, &counter)) {
                 error("failed to set filter \"%s\" on event %s with %d (%s)\n",
                         counter->filter, perf_evsel__name(counter), errno,
-                        strerror_r(errno, msg, sizeof(msg)));
+                        str_error_r(errno, msg, sizeof(msg)));
                 return -1;
         }
 
@@ -637,7 +637,7 @@ try_again:
                 wait(&status);
 
                 if (workload_exec_errno) {
-                        const char *emsg = strerror_r(workload_exec_errno, msg, sizeof(msg));
+                        const char *emsg = str_error_r(workload_exec_errno, msg, sizeof(msg));
                         pr_err("Workload failed: %s\n", emsg);
                         return -1;
                 }
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index 07fc7921980c..bd108683fcb8 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -907,7 +907,7 @@ try_again:
 
         if (perf_evlist__mmap(evlist, opts->mmap_pages, false) < 0) {
                 ui__error("Failed to mmap with %d (%s)\n",
-                            errno, strerror_r(errno, msg, sizeof(msg)));
+                            errno, str_error_r(errno, msg, sizeof(msg)));
                 goto out_err;
         }
 
@@ -1028,7 +1028,7 @@ out_delete:
 
 out_err_cpu_topo: {
         char errbuf[BUFSIZ];
-        const char *err = strerror_r(-ret, errbuf, sizeof(errbuf));
+        const char *err = str_error_r(-ret, errbuf, sizeof(errbuf));
 
         ui__error("Could not read the CPU topology map: %s\n", err);
         goto out_delete;
@@ -1295,7 +1295,7 @@ int cmd_top(int argc, const char **argv, const char *prefix __maybe_unused)
 
         if (perf_evlist__create_maps(top.evlist, target) < 0) {
                 ui__error("Couldn't create thread/CPU maps: %s\n",
-                          errno == ENOENT ? "No such process" : strerror_r(errno, errbuf, sizeof(errbuf)));
+                          errno == ENOENT ? "No such process" : str_error_r(errno, errbuf, sizeof(errbuf)));
                 goto out_delete_evlist;
         }
 
diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index cf90de811523..b8c6766301db 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -43,7 +43,6 @@
 #include <linux/err.h>
 #include <linux/filter.h>
 #include <linux/audit.h>
-#include <sys/ptrace.h>
 #include <linux/random.h>
 #include <linux/stringify.h>
 
@@ -334,6 +333,10 @@ static size_t syscall_arg__scnprintf_fd(char *bf, size_t size,
 
 #define SCA_FD syscall_arg__scnprintf_fd
 
+#ifndef AT_FDCWD
+#define AT_FDCWD        -100
+#endif
+
 static size_t syscall_arg__scnprintf_fd_at(char *bf, size_t size,
                                            struct syscall_arg *arg)
 {
@@ -1601,7 +1604,7 @@ signed_print:
                 fprintf(trace->output, ") = %ld", ret);
         } else if (ret < 0 && (sc->fmt->errmsg || sc->fmt->errpid)) {
                 char bf[STRERR_BUFSIZE];
-                const char *emsg = strerror_r(-ret, bf, sizeof(bf)),
+                const char *emsg = str_error_r(-ret, bf, sizeof(bf)),
                            *e = audit_errno_to_name(-ret);
 
                 fprintf(trace->output, ") = -1 %s %s", e, emsg);
@@ -2402,7 +2405,7 @@ out_error_apply_filters:
         fprintf(trace->output,
                 "Failed to set filter \"%s\" on event %s with %d (%s)\n",
                 evsel->filter, perf_evsel__name(evsel), errno,
-                strerror_r(errno, errbuf, sizeof(errbuf)));
+                str_error_r(errno, errbuf, sizeof(errbuf)));
         goto out_delete_evlist;
 }
 out_error_mem:
diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile
index c7e269a7ca37..5ac428060779 100644
--- a/tools/perf/config/Makefile
+++ b/tools/perf/config/Makefile
@@ -115,7 +115,7 @@ endif
 FEATURE_CHECK_CFLAGS-libbabeltrace := $(LIBBABELTRACE_CFLAGS)
 FEATURE_CHECK_LDFLAGS-libbabeltrace := $(LIBBABELTRACE_LDFLAGS) -lbabeltrace-ctf
 
-FEATURE_CHECK_CFLAGS-bpf = -I. -I$(srctree)/tools/include -I$(srctree)/arch/$(ARCH)/include/uapi -I$(srctree)/include/uapi
+FEATURE_CHECK_CFLAGS-bpf = -I. -I$(srctree)/tools/include -I$(srctree)/tools/arch/$(ARCH)/include/uapi -I$(srctree)/tools/include/uapi
 # include ARCH specific config
 -include $(src-perf)/arch/$(ARCH)/Makefile
 
@@ -206,11 +206,11 @@ endif
 
 CFLAGS += -I$(src-perf)/util/include
 CFLAGS += -I$(src-perf)/arch/$(ARCH)/include
+CFLAGS += -I$(srctree)/tools/include/uapi
 CFLAGS += -I$(srctree)/tools/include/
-CFLAGS += -I$(srctree)/arch/$(ARCH)/include/uapi
-CFLAGS += -I$(srctree)/arch/$(ARCH)/include
-CFLAGS += -I$(srctree)/include/uapi
-CFLAGS += -I$(srctree)/include
+CFLAGS += -I$(srctree)/tools/arch/$(ARCH)/include/uapi
+CFLAGS += -I$(srctree)/tools/arch/$(ARCH)/include/
+CFLAGS += -I$(srctree)/tools/arch/$(ARCH)/
 
 # $(obj-perf)      for generated common-cmds.h
 # $(obj-perf)/util for generated bison/flex headers
@@ -309,6 +309,12 @@ ifndef NO_LIBELF
     CFLAGS += -DHAVE_ELF_GETPHDRNUM_SUPPORT
   endif
 
+  ifeq ($(feature-libelf-gelf_getnote), 1)
+    CFLAGS += -DHAVE_GELF_GETNOTE_SUPPORT
+  else
+    msg := $(warning gelf_getnote() not found on libelf, SDT support disabled);
+  endif
+
   ifeq ($(feature-libelf-getshdrstrndx), 1)
     CFLAGS += -DHAVE_ELF_GETSHDRSTRNDX_SUPPORT
   endif
diff --git a/tools/perf/perf-sys.h b/tools/perf/perf-sys.h
index 83a25cef82fd..5cee8a3d0455 100644
--- a/tools/perf/perf-sys.h
+++ b/tools/perf/perf-sys.h
@@ -11,29 +11,11 @@
 #if defined(__i386__)
 #define cpu_relax()     asm volatile("rep; nop" ::: "memory");
 #define CPUINFO_PROC    {"model name"}
-#ifndef __NR_perf_event_open
-# define __NR_perf_event_open 336
-#endif
-#ifndef __NR_futex
-# define __NR_futex 240
-#endif
-#ifndef __NR_gettid
-# define __NR_gettid 224
-#endif
 #endif
 
 #if defined(__x86_64__)
 #define cpu_relax()     asm volatile("rep; nop" ::: "memory");
 #define CPUINFO_PROC    {"model name"}
-#ifndef __NR_perf_event_open
-# define __NR_perf_event_open 298
-#endif
-#ifndef __NR_futex
-# define __NR_futex 202
-#endif
-#ifndef __NR_gettid
-# define __NR_gettid 186
-#endif
 #endif
 
 #ifdef __powerpc__
diff --git a/tools/perf/perf.c b/tools/perf/perf.c
index 8f219223f305..4b2ff021434c 100644
--- a/tools/perf/perf.c
+++ b/tools/perf/perf.c
@@ -374,7 +374,7 @@ static int run_builtin(struct cmd_struct *p, int argc, const char **argv)
         /* Check for ENOSPC and EIO errors.. */
         if (fflush(stdout)) {
                 fprintf(stderr, "write failure on standard output: %s",
-                        strerror_r(errno, sbuf, sizeof(sbuf)));
+                        str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out;
         }
         if (ferror(stdout)) {
@@ -383,7 +383,7 @@ static int run_builtin(struct cmd_struct *p, int argc, const char **argv)
         }
         if (fclose(stdout)) {
                 fprintf(stderr, "close failed on standard output: %s",
-                        strerror_r(errno, sbuf, sizeof(sbuf)));
+                        str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out;
         }
         status = 0;
@@ -497,6 +497,16 @@ void pthread__unblock_sigwinch(void)
         pthread_sigmask(SIG_UNBLOCK, &set, NULL);
 }
 
+#ifdef _SC_LEVEL1_DCACHE_LINESIZE
+#define cache_line_size(cacheline_sizep) *cacheline_sizep = sysconf(_SC_LEVEL1_DCACHE_LINESIZE)
+#else
+static void cache_line_size(int *cacheline_sizep)
+{
+        if (sysfs__read_int("devices/system/cpu/cpu0/cache/index0/coherency_line_size", cacheline_sizep))
+                perror("cannot determine cache line size");
+}
+#endif
+
 int main(int argc, const char **argv)
 {
         const char *cmd;
@@ -509,7 +519,7 @@ int main(int argc, const char **argv)
 
         /* The page_size is placed in util object. */
         page_size = sysconf(_SC_PAGE_SIZE);
-        cacheline_size = sysconf(_SC_LEVEL1_DCACHE_LINESIZE);
+        cache_line_size(&cacheline_size);
 
         if (sysctl__read_int("kernel/perf_event_max_stack", &value) == 0)
                 sysctl_perf_event_max_stack = value;
@@ -615,7 +625,7 @@ int main(int argc, const char **argv)
         }
 
         fprintf(stderr, "Failed to run command '%s': %s\n",
-                cmd, strerror_r(errno, sbuf, sizeof(sbuf)));
+                cmd, str_error_r(errno, sbuf, sizeof(sbuf)));
 out:
         return 1;
 }
diff --git a/tools/perf/python/tracepoint.py b/tools/perf/python/tracepoint.py
new file mode 100755
index 000000000000..eb4dbed57de7
--- /dev/null
+++ b/tools/perf/python/tracepoint.py
@@ -0,0 +1,47 @@
+#! /usr/bin/python
+# -*- python -*-
+# -*- coding: utf-8 -*-
+
+import perf
+
+class tracepoint(perf.evsel):
+    def __init__(self, sys, name):
+        config = perf.tracepoint(sys, name)
+        perf.evsel.__init__(self,
+                            type   = perf.TYPE_TRACEPOINT,
+                            config = config,
+                            freq = 0, sample_period = 1, wakeup_events = 1,
+                            sample_type = perf.SAMPLE_PERIOD | perf.SAMPLE_TID | perf.SAMPLE_CPU | perf.SAMPLE_RAW | perf.SAMPLE_TIME)
+
+def main():
+    tp      = tracepoint("sched", "sched_switch")
+    cpus    = perf.cpu_map()
+    threads = perf.thread_map(-1)
+
+    evlist = perf.evlist(cpus, threads)
+    evlist.add(tp)
+    evlist.open()
+    evlist.mmap()
+
+    while True:
+        evlist.poll(timeout = -1)
+        for cpu in cpus:
+            event = evlist.read_on_cpu(cpu)
+            if not event:
+                continue
+
+            if not isinstance(event, perf.sample_event):
+                continue
+
+            print "time %u prev_comm=%s prev_pid=%d prev_prio=%d prev_state=0x%x ==> next_comm=%s next_pid=%d next_prio=%d" % (
+                   event.sample_time,
+                   event.prev_comm,
+                   event.prev_pid,
+                   event.prev_prio,
+                   event.prev_state,
+                   event.next_comm,
+                   event.next_pid,
+                   event.next_prio)
+
+if __name__ == '__main__':
+    main()
diff --git a/tools/perf/tests/backward-ring-buffer.c b/tools/perf/tests/backward-ring-buffer.c
index e70313fac5a5..f20ea4c0d0cb 100644
--- a/tools/perf/tests/backward-ring-buffer.c
+++ b/tools/perf/tests/backward-ring-buffer.c
@@ -60,7 +60,7 @@ static int do_test(struct perf_evlist *evlist, int mmap_pages,
         err = perf_evlist__mmap(evlist, mmap_pages, true);
         if (err < 0) {
                 pr_debug("perf_evlist__mmap: %s\n",
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 return TEST_FAIL;
         }
 
@@ -124,7 +124,7 @@ int test__backward_ring_buffer(int subtest __maybe_unused)
         err = perf_evlist__open(evlist);
         if (err < 0) {
                 pr_debug("perf_evlist__open: %s\n",
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out_delete_evlist;
         }
 
diff --git a/tools/perf/tests/bpf-script-example.c b/tools/perf/tests/bpf-script-example.c
index 0ec9c2c03164..e53bc91fa260 100644
--- a/tools/perf/tests/bpf-script-example.c
+++ b/tools/perf/tests/bpf-script-example.c
@@ -31,8 +31,8 @@ struct bpf_map_def SEC("maps") flip_table = {
         .max_entries = 1,
 };
 
-SEC("func=sys_epoll_pwait")
-int bpf_func__sys_epoll_pwait(void *ctx)
+SEC("func=sys_epoll_wait")
+int bpf_func__sys_epoll_wait(void *ctx)
 {
         int ind =0;
         int *flag = bpf_map_lookup_elem(&flip_table, &ind);
diff --git a/tools/perf/tests/bpf.c b/tools/perf/tests/bpf.c
index f31eed31c1a9..fc54064b9186 100644
--- a/tools/perf/tests/bpf.c
+++ b/tools/perf/tests/bpf.c
@@ -13,13 +13,13 @@
 
 #ifdef HAVE_LIBBPF_SUPPORT
 
-static int epoll_pwait_loop(void)
+static int epoll_wait_loop(void)
 {
         int i;
 
         /* Should fail NR_ITERS times */
         for (i = 0; i < NR_ITERS; i++)
-                epoll_pwait(-(i + 1), NULL, 0, 0, NULL);
+                epoll_wait(-(i + 1), NULL, 0, 0);
         return 0;
 }
 
@@ -61,7 +61,7 @@ static struct {
                 "[basic_bpf_test]",
                 "fix 'perf test LLVM' first",
                 "load bpf object failed",
-                &epoll_pwait_loop,
+                &epoll_wait_loop,
                 (NR_ITERS + 1) / 2,
         },
 #ifdef HAVE_BPF_PROLOGUE
@@ -143,14 +143,14 @@ static int do_test(struct bpf_object *obj, int (*func)(void),
         err = perf_evlist__open(evlist);
         if (err < 0) {
                 pr_debug("perf_evlist__open: %s\n",
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out_delete_evlist;
         }
 
         err = perf_evlist__mmap(evlist, opts.mmap_pages, false);
         if (err < 0) {
                 pr_debug("perf_evlist__mmap: %s\n",
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out_delete_evlist;
         }
 
diff --git a/tools/perf/tests/builtin-test.c b/tools/perf/tests/builtin-test.c
index 07c14e9f6546..c23cbf733549 100644
--- a/tools/perf/tests/builtin-test.c
+++ b/tools/perf/tests/builtin-test.c
@@ -258,7 +258,7 @@ static int run_test(struct test *test, int subtest)
 
         if (child < 0) {
                 pr_err("failed to fork test: %s\n",
-                        strerror_r(errno, sbuf, sizeof(sbuf)));
+                        str_error_r(errno, sbuf, sizeof(sbuf)));
                 return -1;
         }
 
diff --git a/tools/perf/tests/cpumap.c b/tools/perf/tests/cpumap.c
index c9ec5f83e42c..f168a85992d0 100644
--- a/tools/perf/tests/cpumap.c
+++ b/tools/perf/tests/cpumap.c
@@ -1,5 +1,12 @@
 #include "tests.h"
+#include <stdio.h>
 #include "cpumap.h"
+#include "event.h"
+#include <string.h>
+#include <linux/bitops.h>
+#include "debug.h"
+
+struct machine;
 
 static int process_event_mask(struct perf_tool *tool __maybe_unused,
                          union perf_event *event,
diff --git a/tools/perf/tests/event-times.c b/tools/perf/tests/event-times.c
index 9f5698ac81ae..19ef77bd6eb4 100644
--- a/tools/perf/tests/event-times.c
+++ b/tools/perf/tests/event-times.c
@@ -37,7 +37,7 @@ static int attach__enable_on_exec(struct perf_evlist *evlist)
         err = perf_evlist__open(evlist);
         if (err < 0) {
                 pr_debug("perf_evlist__open: %s\n",
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 return err;
         }
 
diff --git a/tools/perf/tests/fdarray.c b/tools/perf/tests/fdarray.c
index 59dbd0550c51..a2b5ff9bf83d 100644
--- a/tools/perf/tests/fdarray.c
+++ b/tools/perf/tests/fdarray.c
@@ -1,4 +1,5 @@
 #include <api/fd/array.h>
+#include <poll.h>
 #include "util/debug.h"
 #include "tests/tests.h"
 
diff --git a/tools/perf/tests/llvm.c b/tools/perf/tests/llvm.c
index cff564fb4b66..b798a4bfd238 100644
--- a/tools/perf/tests/llvm.c
+++ b/tools/perf/tests/llvm.c
@@ -5,6 +5,7 @@
 #include "llvm.h"
 #include "tests.h"
 #include "debug.h"
+#include "util.h"
 
 #ifdef HAVE_LIBBPF_SUPPORT
 static int test__bpf_parsing(void *obj_buf, size_t obj_buf_sz)
diff --git a/tools/perf/tests/mmap-basic.c b/tools/perf/tests/mmap-basic.c
index aea33f5589c5..634bce9caebd 100644
--- a/tools/perf/tests/mmap-basic.c
+++ b/tools/perf/tests/mmap-basic.c
@@ -1,3 +1,6 @@
+/* For the CLR_() macros */
+#include <pthread.h>
+
 #include "evlist.h"
 #include "evsel.h"
 #include "thread_map.h"
@@ -49,7 +52,7 @@ int test__basic_mmap(int subtest __maybe_unused)
         sched_setaffinity(0, sizeof(cpu_set), &cpu_set);
         if (sched_setaffinity(0, sizeof(cpu_set), &cpu_set) < 0) {
                 pr_debug("sched_setaffinity() failed on CPU %d: %s ",
-                         cpus->map[0], strerror_r(errno, sbuf, sizeof(sbuf)));
+                         cpus->map[0], str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out_free_cpus;
         }
 
@@ -79,7 +82,7 @@ int test__basic_mmap(int subtest __maybe_unused)
                 if (perf_evsel__open(evsels[i], cpus, threads) < 0) {
                         pr_debug("failed to open counter: %s, "
                                  "tweak /proc/sys/kernel/perf_event_paranoid?\n",
-                                 strerror_r(errno, sbuf, sizeof(sbuf)));
+                                 str_error_r(errno, sbuf, sizeof(sbuf)));
                         goto out_delete_evlist;
                 }
 
@@ -89,7 +92,7 @@ int test__basic_mmap(int subtest __maybe_unused)
 
         if (perf_evlist__mmap(evlist, 128, true) < 0) {
                 pr_debug("failed to mmap events: %d (%s)\n", errno,
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out_delete_evlist;
         }
 
diff --git a/tools/perf/tests/openat-syscall-all-cpus.c b/tools/perf/tests/openat-syscall-all-cpus.c
index ad1cb63139a7..c8d9592eb142 100644
--- a/tools/perf/tests/openat-syscall-all-cpus.c
+++ b/tools/perf/tests/openat-syscall-all-cpus.c
@@ -1,3 +1,6 @@
+/* For the CPU_* macros */
+#include <pthread.h>
+
 #include <api/fs/fs.h>
 #include <linux/err.h>
 #include "evsel.h"
@@ -41,7 +44,7 @@ int test__openat_syscall_event_on_all_cpus(int subtest __maybe_unused)
         if (perf_evsel__open(evsel, cpus, threads) < 0) {
                 pr_debug("failed to open counter: %s, "
                          "tweak /proc/sys/kernel/perf_event_paranoid?\n",
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out_evsel_delete;
         }
 
@@ -62,7 +65,7 @@ int test__openat_syscall_event_on_all_cpus(int subtest __maybe_unused)
                 if (sched_setaffinity(0, sizeof(cpu_set), &cpu_set) < 0) {
                         pr_debug("sched_setaffinity() failed on CPU %d: %s ",
                                  cpus->map[cpu],
-                                 strerror_r(errno, sbuf, sizeof(sbuf)));
+                                 str_error_r(errno, sbuf, sizeof(sbuf)));
                         goto out_close_fd;
                 }
                 for (i = 0; i < ncalls; ++i) {
diff --git a/tools/perf/tests/openat-syscall-tp-fields.c b/tools/perf/tests/openat-syscall-tp-fields.c
index 4344fe482c1d..f52239fed361 100644
--- a/tools/perf/tests/openat-syscall-tp-fields.c
+++ b/tools/perf/tests/openat-syscall-tp-fields.c
@@ -6,6 +6,13 @@
 #include "tests.h"
 #include "debug.h"
 
+#ifndef O_DIRECTORY
+#define O_DIRECTORY    00200000
+#endif
+#ifndef AT_FDCWD
+#define AT_FDCWD       -100
+#endif
+
 int test__syscall_openat_tp_fields(int subtest __maybe_unused)
 {
         struct record_opts opts = {
@@ -51,14 +58,14 @@ int test__syscall_openat_tp_fields(int subtest __maybe_unused)
         err = perf_evlist__open(evlist);
         if (err < 0) {
                 pr_debug("perf_evlist__open: %s\n",
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out_delete_evlist;
         }
 
         err = perf_evlist__mmap(evlist, UINT_MAX, false);
         if (err < 0) {
                 pr_debug("perf_evlist__mmap: %s\n",
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out_delete_evlist;
         }
 
diff --git a/tools/perf/tests/openat-syscall.c b/tools/perf/tests/openat-syscall.c
index 1184f9ba6499..d7414128d7fe 100644
--- a/tools/perf/tests/openat-syscall.c
+++ b/tools/perf/tests/openat-syscall.c
@@ -29,7 +29,7 @@ int test__openat_syscall_event(int subtest __maybe_unused)
         if (perf_evsel__open_per_thread(evsel, threads) < 0) {
                 pr_debug("failed to open counter: %s, "
                          "tweak /proc/sys/kernel/perf_event_paranoid?\n",
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out_evsel_delete;
         }
 
diff --git a/tools/perf/tests/perf-record.c b/tools/perf/tests/perf-record.c
index b836ee6a8d9b..8f2e1de6d0ea 100644
--- a/tools/perf/tests/perf-record.c
+++ b/tools/perf/tests/perf-record.c
@@ -1,3 +1,6 @@
+/* For the CLR_() macros */
+#include <pthread.h>
+
 #include <sched.h>
 #include "evlist.h"
 #include "evsel.h"
@@ -104,7 +107,7 @@ int test__PERF_RECORD(int subtest __maybe_unused)
         err = sched__get_first_possible_cpu(evlist->workload.pid, &cpu_mask);
         if (err < 0) {
                 pr_debug("sched__get_first_possible_cpu: %s\n",
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out_delete_evlist;
         }
 
@@ -115,7 +118,7 @@ int test__PERF_RECORD(int subtest __maybe_unused)
          */
         if (sched_setaffinity(evlist->workload.pid, cpu_mask_size, &cpu_mask) < 0) {
                 pr_debug("sched_setaffinity: %s\n",
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out_delete_evlist;
         }
 
@@ -126,7 +129,7 @@ int test__PERF_RECORD(int subtest __maybe_unused)
         err = perf_evlist__open(evlist);
         if (err < 0) {
                 pr_debug("perf_evlist__open: %s\n",
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out_delete_evlist;
         }
 
@@ -138,7 +141,7 @@ int test__PERF_RECORD(int subtest __maybe_unused)
         err = perf_evlist__mmap(evlist, opts.mmap_pages, false);
         if (err < 0) {
                 pr_debug("perf_evlist__mmap: %s\n",
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out_delete_evlist;
         }
 
diff --git a/tools/perf/tests/sw-clock.c b/tools/perf/tests/sw-clock.c
index 36e8ce1550e3..4c9fd046d57b 100644
--- a/tools/perf/tests/sw-clock.c
+++ b/tools/perf/tests/sw-clock.c
@@ -70,7 +70,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id)
 
                 err = -errno;
                 pr_debug("Couldn't open evlist: %s\nHint: check %s, using %" PRIu64 " in this test.\n",
-                         strerror_r(errno, sbuf, sizeof(sbuf)),
+                         str_error_r(errno, sbuf, sizeof(sbuf)),
                          knob, (u64)attr.sample_freq);
                 goto out_delete_evlist;
         }
@@ -78,7 +78,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id)
         err = perf_evlist__mmap(evlist, 128, true);
         if (err < 0) {
                 pr_debug("failed to mmap event: %d (%s)\n", errno,
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out_delete_evlist;
         }
 
diff --git a/tools/perf/tests/task-exit.c b/tools/perf/tests/task-exit.c
index 2dfff7ac8ef3..01a5ba2788c6 100644
--- a/tools/perf/tests/task-exit.c
+++ b/tools/perf/tests/task-exit.c
@@ -91,13 +91,13 @@ int test__task_exit(int subtest __maybe_unused)
         err = perf_evlist__open(evlist);
         if (err < 0) {
                 pr_debug("Couldn't open the evlist: %s\n",
-                         strerror_r(-err, sbuf, sizeof(sbuf)));
+                         str_error_r(-err, sbuf, sizeof(sbuf)));
                 goto out_delete_evlist;
         }
 
         if (perf_evlist__mmap(evlist, 128, true) < 0) {
                 pr_debug("failed to mmap events: %d (%s)\n", errno,
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 goto out_delete_evlist;
         }
 
diff --git a/tools/perf/trace/beauty/flock.c b/tools/perf/trace/beauty/flock.c
index 021bb48c6336..74613703a14e 100644
--- a/tools/perf/trace/beauty/flock.c
+++ b/tools/perf/trace/beauty/flock.c
@@ -1,3 +1,20 @@
+#include <fcntl.h>
+
+#ifndef LOCK_MAND
+#define LOCK_MAND        32
+#endif
+
+#ifndef LOCK_READ
+#define LOCK_READ        64
+#endif
+
+#ifndef LOCK_WRITE
+#define LOCK_WRITE      128
+#endif
+
+#ifndef LOCK_RW
+#define LOCK_RW         192
+#endif
 
 static size_t syscall_arg__scnprintf_flock(char *bf, size_t size,
                                            struct syscall_arg *arg)
diff --git a/tools/perf/trace/beauty/futex_op.c b/tools/perf/trace/beauty/futex_op.c
index e2476211f22d..bfd3359b09b6 100644
--- a/tools/perf/trace/beauty/futex_op.c
+++ b/tools/perf/trace/beauty/futex_op.c
@@ -1,5 +1,21 @@
 #include <linux/futex.h>
 
+#ifndef FUTEX_WAIT_BITSET
+#define FUTEX_WAIT_BITSET         9
+#endif
+#ifndef FUTEX_WAKE_BITSET
+#define FUTEX_WAKE_BITSET        10
+#endif
+#ifndef FUTEX_WAIT_REQUEUE_PI
+#define FUTEX_WAIT_REQUEUE_PI    11
+#endif
+#ifndef FUTEX_CMP_REQUEUE_PI
+#define FUTEX_CMP_REQUEUE_PI     12
+#endif
+#ifndef FUTEX_CLOCK_REALTIME
+#define FUTEX_CLOCK_REALTIME    256
+#endif
+
 static size_t syscall_arg__scnprintf_futex_op(char *bf, size_t size, struct syscall_arg *arg)
 {
         enum syscall_futex_args {
diff --git a/tools/perf/trace/beauty/mmap.c b/tools/perf/trace/beauty/mmap.c
index 3444a4d5382d..d0a3a8e402e7 100644
--- a/tools/perf/trace/beauty/mmap.c
+++ b/tools/perf/trace/beauty/mmap.c
@@ -1,5 +1,9 @@
 #include <sys/mman.h>
 
+#ifndef PROT_SEM
+#define PROT_SEM 0x8
+#endif
+
 static size_t syscall_arg__scnprintf_mmap_prot(char *bf, size_t size,
                                                struct syscall_arg *arg)
 {
@@ -16,9 +20,7 @@ static size_t syscall_arg__scnprintf_mmap_prot(char *bf, size_t size,
         P_MMAP_PROT(EXEC);
         P_MMAP_PROT(READ);
         P_MMAP_PROT(WRITE);
-#ifdef PROT_SEM
         P_MMAP_PROT(SEM);
-#endif
         P_MMAP_PROT(GROWSDOWN);
         P_MMAP_PROT(GROWSUP);
 #undef P_MMAP_PROT
@@ -31,10 +33,31 @@ static size_t syscall_arg__scnprintf_mmap_prot(char *bf, size_t size,
 
 #define SCA_MMAP_PROT syscall_arg__scnprintf_mmap_prot
 
+#ifndef MAP_FIXED
+#define MAP_FIXED                    0x10
+#endif
+
+#ifndef MAP_ANONYMOUS
+#define MAP_ANONYMOUS                0x20
+#endif
+
+#ifndef MAP_32BIT
+#define MAP_32BIT                    0x40
+#endif
+
 #ifndef MAP_STACK
-# define MAP_STACK              0x20000
+#define MAP_STACK                 0x20000
 #endif
 
+#ifndef MAP_HUGETLB
+#define MAP_HUGETLB               0x40000
+#endif
+
+#ifndef MAP_UNINITIALIZED
+#define MAP_UNINITIALIZED       0x4000000
+#endif
+
+
 static size_t syscall_arg__scnprintf_mmap_flags(char *bf, size_t size,
                                                 struct syscall_arg *arg)
 {
@@ -48,26 +71,20 @@ static size_t syscall_arg__scnprintf_mmap_flags(char *bf, size_t size,
 
         P_MMAP_FLAG(SHARED);
         P_MMAP_FLAG(PRIVATE);
-#ifdef MAP_32BIT
         P_MMAP_FLAG(32BIT);
-#endif
         P_MMAP_FLAG(ANONYMOUS);
         P_MMAP_FLAG(DENYWRITE);
         P_MMAP_FLAG(EXECUTABLE);
         P_MMAP_FLAG(FILE);
         P_MMAP_FLAG(FIXED);
         P_MMAP_FLAG(GROWSDOWN);
-#ifdef MAP_HUGETLB
         P_MMAP_FLAG(HUGETLB);
-#endif
         P_MMAP_FLAG(LOCKED);
         P_MMAP_FLAG(NONBLOCK);
         P_MMAP_FLAG(NORESERVE);
         P_MMAP_FLAG(POPULATE);
         P_MMAP_FLAG(STACK);
-#ifdef MAP_UNINITIALIZED
         P_MMAP_FLAG(UNINITIALIZED);
-#endif
 #undef P_MMAP_FLAG
 
         if (flags)
@@ -78,6 +95,13 @@ static size_t syscall_arg__scnprintf_mmap_flags(char *bf, size_t size,
 
 #define SCA_MMAP_FLAGS syscall_arg__scnprintf_mmap_flags
 
+#ifndef MREMAP_MAYMOVE
+#define MREMAP_MAYMOVE 1
+#endif
+#ifndef MREMAP_FIXED
+#define MREMAP_FIXED 2
+#endif
+
 static size_t syscall_arg__scnprintf_mremap_flags(char *bf, size_t size,
                                                   struct syscall_arg *arg)
 {
@@ -90,9 +114,7 @@ static size_t syscall_arg__scnprintf_mremap_flags(char *bf, size_t size,
         }
 
         P_MREMAP_FLAG(MAYMOVE);
-#ifdef MREMAP_FIXED
         P_MREMAP_FLAG(FIXED);
-#endif
 #undef P_MREMAP_FLAG
 
         if (flags)
@@ -107,6 +129,10 @@ static size_t syscall_arg__scnprintf_mremap_flags(char *bf, size_t size,
 #define MADV_HWPOISON           100
 #endif
 
+#ifndef MADV_SOFT_OFFLINE
+#define MADV_SOFT_OFFLINE       101
+#endif
+
 #ifndef MADV_MERGEABLE
 #define MADV_MERGEABLE           12
 #endif
@@ -115,6 +141,23 @@ static size_t syscall_arg__scnprintf_mremap_flags(char *bf, size_t size,
 #define MADV_UNMERGEABLE         13
 #endif
 
+#ifndef MADV_HUGEPAGE
+#define MADV_HUGEPAGE            14
+#endif
+
+#ifndef MADV_NOHUGEPAGE
+#define MADV_NOHUGEPAGE          15
+#endif
+
+#ifndef MADV_DONTDUMP
+#define MADV_DONTDUMP            16
+#endif
+
+#ifndef MADV_DODUMP
+#define MADV_DODUMP              17
+#endif
+
+
 static size_t syscall_arg__scnprintf_madvise_behavior(char *bf, size_t size,
                                                       struct syscall_arg *arg)
 {
@@ -131,24 +174,14 @@ static size_t syscall_arg__scnprintf_madvise_behavior(char *bf, size_t size,
         P_MADV_BHV(DONTFORK);
         P_MADV_BHV(DOFORK);
         P_MADV_BHV(HWPOISON);
-#ifdef MADV_SOFT_OFFLINE
         P_MADV_BHV(SOFT_OFFLINE);
-#endif
         P_MADV_BHV(MERGEABLE);
         P_MADV_BHV(UNMERGEABLE);
-#ifdef MADV_HUGEPAGE
         P_MADV_BHV(HUGEPAGE);
-#endif
-#ifdef MADV_NOHUGEPAGE
         P_MADV_BHV(NOHUGEPAGE);
-#endif
-#ifdef MADV_DONTDUMP
         P_MADV_BHV(DONTDUMP);
-#endif
-#ifdef MADV_DODUMP
         P_MADV_BHV(DODUMP);
-#endif
-#undef P_MADV_PHV
+#undef P_MADV_BHV
         default: break;
         }
 
diff --git a/tools/perf/trace/beauty/msg_flags.c b/tools/perf/trace/beauty/msg_flags.c
index 07fa8a0acad6..1106c8960cc4 100644
--- a/tools/perf/trace/beauty/msg_flags.c
+++ b/tools/perf/trace/beauty/msg_flags.c
@@ -33,7 +33,6 @@ static size_t syscall_arg__scnprintf_msg_flags(char *bf, size_t size,
         P_MSG_FLAG(OOB);
         P_MSG_FLAG(PEEK);
         P_MSG_FLAG(DONTROUTE);
-        P_MSG_FLAG(TRYHARD);
         P_MSG_FLAG(CTRUNC);
         P_MSG_FLAG(PROBE);
         P_MSG_FLAG(TRUNC);
diff --git a/tools/perf/trace/beauty/open_flags.c b/tools/perf/trace/beauty/open_flags.c
index 0f3679e0cdcf..f55a4597fc38 100644
--- a/tools/perf/trace/beauty/open_flags.c
+++ b/tools/perf/trace/beauty/open_flags.c
@@ -1,3 +1,18 @@
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+
+#ifndef O_DIRECT
+#define O_DIRECT        00040000
+#endif
+
+#ifndef O_DIRECTORY
+#define O_DIRECTORY     00200000
+#endif
+
+#ifndef O_NOATIME
+#define O_NOATIME       01000000
+#endif
 
 static size_t syscall_arg__scnprintf_open_flags(char *bf, size_t size,
                                                struct syscall_arg *arg)
diff --git a/tools/perf/trace/beauty/seccomp.c b/tools/perf/trace/beauty/seccomp.c
index 213c5a7e3e92..356441bce27d 100644
--- a/tools/perf/trace/beauty/seccomp.c
+++ b/tools/perf/trace/beauty/seccomp.c
@@ -1,5 +1,3 @@
-#include <linux/seccomp.h>
-
 #ifndef SECCOMP_SET_MODE_STRICT
 #define SECCOMP_SET_MODE_STRICT 0
 #endif
diff --git a/tools/perf/ui/browsers/hists.c b/tools/perf/ui/browsers/hists.c
index e08b8f7b6d3f..13d414384739 100644
--- a/tools/perf/ui/browsers/hists.c
+++ b/tools/perf/ui/browsers/hists.c
@@ -2029,7 +2029,7 @@ static int hist_browser__dump(struct hist_browser *browser)
         fp = fopen(filename, "w");
         if (fp == NULL) {
                 char bf[64];
-                const char *err = strerror_r(errno, bf, sizeof(bf));
+                const char *err = str_error_r(errno, bf, sizeof(bf));
                 ui_helpline__fpush("Couldn't write to %s: %s", filename, err);
                 return -1;
         }
diff --git a/tools/perf/ui/gtk/util.c b/tools/perf/ui/gtk/util.c
index 52e7fc48af9f..00b91921edb1 100644
--- a/tools/perf/ui/gtk/util.c
+++ b/tools/perf/ui/gtk/util.c
@@ -1,4 +1,5 @@
 #include "../util.h"
+#include "../../util/util.h"
 #include "../../util/debug.h"
 #include "gtk.h"
 
diff --git a/tools/perf/ui/helpline.c b/tools/perf/ui/helpline.c
index 700fb3cfa1c7..5b74a7eba210 100644
--- a/tools/perf/ui/helpline.c
+++ b/tools/perf/ui/helpline.c
@@ -5,6 +5,7 @@
 #include "../debug.h"
 #include "helpline.h"
 #include "ui.h"
+#include "../util.h"
 
 char ui_helpline__current[512];
 
diff --git a/tools/perf/ui/setup.c b/tools/perf/ui/setup.c
index ba51fa8a1176..1f6b0994f4f4 100644
--- a/tools/perf/ui/setup.c
+++ b/tools/perf/ui/setup.c
@@ -60,6 +60,13 @@ static inline int setup_gtk_browser(void) { return -1; }
 static inline void exit_gtk_browser(bool wait_for_ok __maybe_unused) {}
 #endif
 
+int stdio__config_color(const struct option *opt __maybe_unused,
+                        const char *mode, int unset __maybe_unused)
+{
+        perf_use_color_default = perf_config_colorbool("color.ui", mode, -1);
+        return 0;
+}
+
 void setup_browser(bool fallback_to_pager)
 {
         if (use_browser < 2 && (!isatty(1) || dump_trace))
diff --git a/tools/perf/ui/tui/setup.c b/tools/perf/ui/tui/setup.c
index 7dfeba0a91f3..4ea2ba861fc2 100644
--- a/tools/perf/ui/tui/setup.c
+++ b/tools/perf/ui/tui/setup.c
@@ -1,3 +1,4 @@
+#include <errno.h>
 #include <signal.h>
 #include <stdbool.h>
 #ifdef HAVE_BACKTRACE_SUPPORT
@@ -6,6 +7,7 @@
 
 #include "../../util/cache.h"
 #include "../../util/debug.h"
+#include "../../util/util.h"
 #include "../browser.h"
 #include "../helpline.h"
 #include "../ui.h"
diff --git a/tools/perf/ui/ui.h b/tools/perf/ui/ui.h
index ab88383f8be8..4b6fb6c7a542 100644
--- a/tools/perf/ui/ui.h
+++ b/tools/perf/ui/ui.h
@@ -26,4 +26,8 @@ static inline void ui__exit(bool wait_for_ok __maybe_unused) {}
 
 void ui__refresh_dimensions(bool force);
 
+struct option;
+
+int stdio__config_color(const struct option *opt, const char *mode, int unset);
+
 #endif /* _PERF_UI_H_ */
diff --git a/tools/perf/util/Build b/tools/perf/util/Build
index fced8336e5fd..2fa7d8b69873 100644
--- a/tools/perf/util/Build
+++ b/tools/perf/util/Build
@@ -70,6 +70,7 @@ libperf-y += stat.o
 libperf-y += stat-shadow.o
 libperf-y += record.o
 libperf-y += srcline.o
+libperf-y += str_error_r.o
 libperf-y += data.o
 libperf-y += tsc.o
 libperf-y += cloexec.o
@@ -84,6 +85,7 @@ libperf-y += parse-regs-options.o
 libperf-y += term.o
 libperf-y += help-unknown-cmd.o
 libperf-y += mem-events.o
+libperf-y += vsprintf.o
 
 libperf-$(CONFIG_LIBBPF) += bpf-loader.o
 libperf-$(CONFIG_BPF_PROLOGUE) += bpf-prologue.o
@@ -111,6 +113,7 @@ libperf-y += scripting-engines/
 libperf-$(CONFIG_ZLIB) += zlib.o
 libperf-$(CONFIG_LZMA) += lzma.o
 libperf-y += demangle-java.o
+libperf-y += demangle-rust.o
 
 ifdef CONFIG_JITDUMP
 libperf-$(CONFIG_LIBELF) += jitdump.o
@@ -173,6 +176,14 @@ $(OUTPUT)util/libstring.o: ../lib/string.c FORCE
         $(call rule_mkdir)
         $(call if_changed_dep,cc_o_c)
 
+$(OUTPUT)util/str_error_r.o: ../lib/str_error_r.c FORCE
+        $(call rule_mkdir)
+        $(call if_changed_dep,cc_o_c)
+
 $(OUTPUT)util/hweight.o: ../lib/hweight.c FORCE
         $(call rule_mkdir)
         $(call if_changed_dep,cc_o_c)
+
+$(OUTPUT)util/vsprintf.o: ../lib/vsprintf.c FORCE
+        $(call rule_mkdir)
+        $(call if_changed_dep,cc_o_c)
diff --git a/tools/perf/util/alias.c b/tools/perf/util/alias.c
index 6c80f830358b..6455471d9cd1 100644
--- a/tools/perf/util/alias.c
+++ b/tools/perf/util/alias.c
@@ -1,4 +1,5 @@
 #include "cache.h"
+#include "util.h"
 #include "config.h"
 
 static const char *alias_key;
diff --git a/tools/perf/util/bpf-loader.c b/tools/perf/util/bpf-loader.c
index dcc8845881ae..8445e89621fe 100644
--- a/tools/perf/util/bpf-loader.c
+++ b/tools/perf/util/bpf-loader.c
@@ -1589,7 +1589,7 @@ bpf_loader_strerror(int err, char *buf, size_t size)
                 snprintf(buf, size, "Unknown bpf loader error %d", err);
         else
                 snprintf(buf, size, "%s",
-                         strerror_r(err, sbuf, sizeof(sbuf)));
+                         str_error_r(err, sbuf, sizeof(sbuf)));
 
         buf[size - 1] = '\0';
         return -1;
diff --git a/tools/perf/util/build-id.c b/tools/perf/util/build-id.c
index e1a16408da9c..1e504e40dac8 100644
--- a/tools/perf/util/build-id.c
+++ b/tools/perf/util/build-id.c
@@ -533,7 +533,7 @@ int build_id_cache__list_build_ids(const char *pathname,
         return ret;
 }
 
-#ifdef HAVE_LIBELF_SUPPORT
+#if defined(HAVE_LIBELF_SUPPORT) && defined(HAVE_GELF_GETNOTE_SUPPORT)
 static int build_id_cache__add_sdt_cache(const char *sbuild_id,
                                           const char *realname)
 {
diff --git a/tools/perf/util/cache.h b/tools/perf/util/cache.h
index 9f90e36b56fb..512c0c83fbc6 100644
--- a/tools/perf/util/cache.h
+++ b/tools/perf/util/cache.h
@@ -1,11 +1,8 @@
 #ifndef __PERF_CACHE_H
 #define __PERF_CACHE_H
 
-#include <stdbool.h>
-#include "util.h"
 #include "strbuf.h"
 #include <subcmd/pager.h>
-#include "../perf.h"
 #include "../ui/ui.h"
 
 #include <linux/string.h>
diff --git a/tools/perf/util/cloexec.c b/tools/perf/util/cloexec.c
index 2babddaa2481..f0dcd0ee0afa 100644
--- a/tools/perf/util/cloexec.c
+++ b/tools/perf/util/cloexec.c
@@ -4,18 +4,24 @@
 #include "cloexec.h"
 #include "asm/bug.h"
 #include "debug.h"
+#include <unistd.h>
+#include <asm/unistd.h>
+#include <sys/syscall.h>
 
 static unsigned long flag = PERF_FLAG_FD_CLOEXEC;
 
-#ifdef __GLIBC_PREREQ
-#if !__GLIBC_PREREQ(2, 6)
 int __weak sched_getcpu(void)
 {
+#ifdef __NR_getcpu
+        unsigned cpu;
+        int err = syscall(__NR_getcpu, &cpu, NULL, NULL);
+        if (!err)
+                return cpu;
+#else
         errno = ENOSYS;
+#endif
         return -1;
 }
-#endif
-#endif
 
 static int perf_flag_probe(void)
 {
@@ -58,7 +64,7 @@ static int perf_flag_probe(void)
 
         WARN_ONCE(err != EINVAL && err != EBUSY,
                   "perf_event_open(..., PERF_FLAG_FD_CLOEXEC) failed with unexpected error %d (%s)\n",
-                  err, strerror_r(err, sbuf, sizeof(sbuf)));
+                  err, str_error_r(err, sbuf, sizeof(sbuf)));
 
         /* not supported, confirm error related to PERF_FLAG_FD_CLOEXEC */
         while (1) {
@@ -76,7 +82,7 @@ static int perf_flag_probe(void)
 
         if (WARN_ONCE(fd < 0 && err != EBUSY,
                       "perf_event_open(..., 0) failed unexpectedly with error %d (%s)\n",
-                      err, strerror_r(err, sbuf, sizeof(sbuf))))
+                      err, str_error_r(err, sbuf, sizeof(sbuf))))
                 return -1;
 
         return 0;
diff --git a/tools/perf/util/color.c b/tools/perf/util/color.c
index 1210ba555d07..dbbf89b050a5 100644
--- a/tools/perf/util/color.c
+++ b/tools/perf/util/color.c
@@ -1,8 +1,11 @@
 #include <linux/kernel.h>
 #include "cache.h"
 #include "config.h"
+#include <stdlib.h>
+#include <stdio.h>
 #include "color.h"
 #include <math.h>
+#include <unistd.h>
 
 int perf_use_color_default = -1;
 
diff --git a/tools/perf/util/data.c b/tools/perf/util/data.c
index be83516155ee..60bfc9ca1e22 100644
--- a/tools/perf/util/data.c
+++ b/tools/perf/util/data.c
@@ -57,7 +57,7 @@ static int open_file_read(struct perf_data_file *file)
                 int err = errno;
 
                 pr_err("failed to open %s: %s", file->path,
-                        strerror_r(err, sbuf, sizeof(sbuf)));
+                        str_error_r(err, sbuf, sizeof(sbuf)));
                 if (err == ENOENT && !strcmp(file->path, "perf.data"))
                         pr_err("  (try 'perf record' first)");
                 pr_err("\n");
@@ -99,7 +99,7 @@ static int open_file_write(struct perf_data_file *file)
 
         if (fd < 0)
                 pr_err("failed to open %s : %s\n", file->path,
-                        strerror_r(errno, sbuf, sizeof(sbuf)));
+                        str_error_r(errno, sbuf, sizeof(sbuf)));
 
         return fd;
 }
diff --git a/tools/perf/util/debug.h b/tools/perf/util/debug.h
index 14bafda79eda..d242adc3d5a2 100644
--- a/tools/perf/util/debug.h
+++ b/tools/perf/util/debug.h
@@ -38,7 +38,7 @@ extern int debug_data_convert;
 #define pr_oe_time(t, fmt, ...)  pr_time_N(1, debug_ordered_events, t, pr_fmt(fmt), ##__VA_ARGS__)
 #define pr_oe_time2(t, fmt, ...) pr_time_N(2, debug_ordered_events, t, pr_fmt(fmt), ##__VA_ARGS__)
 
-#define STRERR_BUFSIZE  128     /* For the buffer size of strerror_r */
+#define STRERR_BUFSIZE  128     /* For the buffer size of str_error_r */
 
 int dump_printf(const char *fmt, ...) __attribute__((format(printf, 1, 2)));
 void trace_event(union perf_event *event);
diff --git a/tools/perf/util/demangle-rust.c b/tools/perf/util/demangle-rust.c
new file mode 100644
index 000000000000..f9dafa888c06
--- /dev/null
+++ b/tools/perf/util/demangle-rust.c
@@ -0,0 +1,269 @@
+#include <string.h>
+#include "util.h"
+#include "debug.h"
+
+#include "demangle-rust.h"
+
+/*
+ * Mangled Rust symbols look like this:
+ *
+ *     _$LT$std..sys..fd..FileDesc$u20$as$u20$core..ops..Drop$GT$::drop::hc68340e1baa4987a
+ *
+ * The original symbol is:
+ *
+ *     <std::sys::fd::FileDesc as core::ops::Drop>::drop
+ *
+ * The last component of the path is a 64-bit hash in lowercase hex, prefixed
+ * with "h". Rust does not have a global namespace between crates, an illusion
+ * which Rust maintains by using the hash to distinguish things that would
+ * otherwise have the same symbol.
+ *
+ * Any path component not starting with a XID_Start character is prefixed with
+ * "_".
+ *
+ * The following escape sequences are used:
+ *
+ *     ","  =>  $C$
+ *     "@"  =>  $SP$
+ *     "*"  =>  $BP$
+ *     "&"  =>  $RF$
+ *     "<"  =>  $LT$
+ *     ">"  =>  $GT$
+ *     "("  =>  $LP$
+ *     ")"  =>  $RP$
+ *     " "  =>  $u20$
+ *     "'"  =>  $u27$
+ *     "["  =>  $u5b$
+ *     "]"  =>  $u5d$
+ *     "~"  =>  $u7e$
+ *
+ * A double ".." means "::" and a single "." means "-".
+ *
+ * The only characters allowed in the mangled symbol are a-zA-Z0-9 and _.:$
+ */
+
+static const char *hash_prefix = "::h";
+static const size_t hash_prefix_len = 3;
+static const size_t hash_len = 16;
+
+static bool is_prefixed_hash(const char *start);
+static bool looks_like_rust(const char *sym, size_t len);
+static bool unescape(const char **in, char **out, const char *seq, char value);
+
+/*
+ * INPUT:
+ *     sym: symbol that has been through BFD-demangling
+ *
+ * This function looks for the following indicators:
+ *
+ *  1. The hash must consist of "h" followed by 16 lowercase hex digits.
+ *
+ *  2. As a sanity check, the hash must use between 5 and 15 of the 16 possible
+ *     hex digits. This is true of 99.9998% of hashes so once in your life you
+ *     may see a false negative. The point is to notice path components that
+ *     could be Rust hashes but are probably not, like "haaaaaaaaaaaaaaaa". In
+ *     this case a false positive (non-Rust symbol has an important path
+ *     component removed because it looks like a Rust hash) is worse than a
+ *     false negative (the rare Rust symbol is not demangled) so this sets the
+ *     balance in favor of false negatives.
+ *
+ *  3. There must be no characters other than a-zA-Z0-9 and _.:$
+ *
+ *  4. There must be no unrecognized $-sign sequences.
+ *
+ *  5. There must be no sequence of three or more dots in a row ("...").
+ */
+bool
+rust_is_mangled(const char *sym)
+{
+        size_t len, len_without_hash;
+
+        if (!sym)
+                return false;
+
+        len = strlen(sym);
+        if (len <= hash_prefix_len + hash_len)
+                /* Not long enough to contain "::h" + hash + something else */
+                return false;
+
+        len_without_hash = len - (hash_prefix_len + hash_len);
+        if (!is_prefixed_hash(sym + len_without_hash))
+                return false;
+
+        return looks_like_rust(sym, len_without_hash);
+}
+
+/*
+ * A hash is the prefix "::h" followed by 16 lowercase hex digits. The hex
+ * digits must comprise between 5 and 15 (inclusive) distinct digits.
+ */
+static bool is_prefixed_hash(const char *str)
+{
+        const char *end;
+        bool seen[16];
+        size_t i;
+        int count;
+
+        if (strncmp(str, hash_prefix, hash_prefix_len))
+                return false;
+        str += hash_prefix_len;
+
+        memset(seen, false, sizeof(seen));
+        for (end = str + hash_len; str < end; str++)
+                if (*str >= '0' && *str <= '9')
+                        seen[*str - '0'] = true;
+                else if (*str >= 'a' && *str <= 'f')
+                        seen[*str - 'a' + 10] = true;
+                else
+                        return false;
+
+        /* Count how many distinct digits seen */
+        count = 0;
+        for (i = 0; i < 16; i++)
+                if (seen[i])
+                        count++;
+
+        return count >= 5 && count <= 15;
+}
+
+static bool looks_like_rust(const char *str, size_t len)
+{
+        const char *end = str + len;
+
+        while (str < end)
+                switch (*str) {
+                case '$':
+                        if (!strncmp(str, "$C$", 3))
+                                str += 3;
+                        else if (!strncmp(str, "$SP$", 4)
+                                        || !strncmp(str, "$BP$", 4)
+                                        || !strncmp(str, "$RF$", 4)
+                                        || !strncmp(str, "$LT$", 4)
+                                        || !strncmp(str, "$GT$", 4)
+                                        || !strncmp(str, "$LP$", 4)
+                                        || !strncmp(str, "$RP$", 4))
+                                str += 4;
+                        else if (!strncmp(str, "$u20$", 5)
+                                        || !strncmp(str, "$u27$", 5)
+                                        || !strncmp(str, "$u5b$", 5)
+                                        || !strncmp(str, "$u5d$", 5)
+                                        || !strncmp(str, "$u7e$", 5))
+                                str += 5;
+                        else
+                                return false;
+                        break;
+                case '.':
+                        /* Do not allow three or more consecutive dots */
+                        if (!strncmp(str, "...", 3))
+                                return false;
+                        /* Fall through */
+                case 'a' ... 'z':
+                case 'A' ... 'Z':
+                case '0' ... '9':
+                case '_':
+                case ':':
+                        str++;
+                        break;
+                default:
+                        return false;
+                }
+
+        return true;
+}
+
+/*
+ * INPUT:
+ *     sym: symbol for which rust_is_mangled(sym) returns true
+ *
+ * The input is demangled in-place because the mangled name is always longer
+ * than the demangled one.
+ */
+void
+rust_demangle_sym(char *sym)
+{
+        const char *in;
+        char *out;
+        const char *end;
+
+        if (!sym)
+                return;
+
+        in = sym;
+        out = sym;
+        end = sym + strlen(sym) - (hash_prefix_len + hash_len);
+
+        while (in < end)
+                switch (*in) {
+                case '$':
+                        if (!(unescape(&in, &out, "$C$", ',')
+                                        || unescape(&in, &out, "$SP$", '@')
+                                        || unescape(&in, &out, "$BP$", '*')
+                                        || unescape(&in, &out, "$RF$", '&')
+                                        || unescape(&in, &out, "$LT$", '<')
+                                        || unescape(&in, &out, "$GT$", '>')
+                                        || unescape(&in, &out, "$LP$", '(')
+                                        || unescape(&in, &out, "$RP$", ')')
+                                        || unescape(&in, &out, "$u20$", ' ')
+                                        || unescape(&in, &out, "$u27$", '\'')
+                                        || unescape(&in, &out, "$u5b$", '[')
+                                        || unescape(&in, &out, "$u5d$", ']')
+                                        || unescape(&in, &out, "$u7e$", '~'))) {
+                                pr_err("demangle-rust: unexpected escape sequence");
+                                goto done;
+                        }
+                        break;
+                case '_':
+                        /*
+                         * If this is the start of a path component and the next
+                         * character is an escape sequence, ignore the
+                         * underscore. The mangler inserts an underscore to make
+                         * sure the path component begins with a XID_Start
+                         * character.
+                         */
+                        if ((in == sym || in[-1] == ':') && in[1] == '$')
+                                in++;
+                        else
+                                *out++ = *in++;
+                        break;
+                case '.':
+                        if (in[1] == '.') {
+                                /* ".." becomes "::" */
+                                *out++ = ':';
+                                *out++ = ':';
+                                in += 2;
+                        } else {
+                                /* "." becomes "-" */
+                                *out++ = '-';
+                                in++;
+                        }
+                        break;
+                case 'a' ... 'z':
+                case 'A' ... 'Z':
+                case '0' ... '9':
+                case ':':
+                        *out++ = *in++;
+                        break;
+                default:
+                        pr_err("demangle-rust: unexpected character '%c' in symbol\n",
+                                *in);
+                        goto done;
+                }
+
+done:
+        *out = '\0';
+}
+
+static bool unescape(const char **in, char **out, const char *seq, char value)
+{
+        size_t len = strlen(seq);
+
+        if (strncmp(*in, seq, len))
+                return false;
+
+        **out = value;
+
+        *in += len;
+        *out += 1;
+
+        return true;
+}
diff --git a/tools/perf/util/demangle-rust.h b/tools/perf/util/demangle-rust.h
new file mode 100644
index 000000000000..7b41ead7e0dd
--- /dev/null
+++ b/tools/perf/util/demangle-rust.h
@@ -0,0 +1,7 @@
+#ifndef __PERF_DEMANGLE_RUST
+#define __PERF_DEMANGLE_RUST 1
+
+bool rust_is_mangled(const char *str);
+void rust_demangle_sym(char *str);
+
+#endif /* __PERF_DEMANGLE_RUST */
diff --git a/tools/perf/util/dso.c b/tools/perf/util/dso.c
index e1de6cc4863e..774f6ec884d5 100644
--- a/tools/perf/util/dso.c
+++ b/tools/perf/util/dso.c
@@ -335,7 +335,7 @@ static int do_open(char *name)
                         return fd;
 
                 pr_debug("dso open failed: %s\n",
-                         strerror_r(errno, sbuf, sizeof(sbuf)));
+                         str_error_r(errno, sbuf, sizeof(sbuf)));
                 if (!dso__data_open_cnt || errno != EMFILE)
                         break;
 
@@ -786,7 +786,7 @@ static int data_file_size(struct dso *dso, struct machine *machine)
         if (fstat(dso->data.fd, &st) < 0) {
                 ret = -errno;
                 pr_err("dso cache fstat failed: %s\n",
-                       strerror_r(errno, sbuf, sizeof(sbuf)));
+                       str_error_r(errno, sbuf, sizeof(sbuf)));
                 dso->data.status = DSO_DATA_STATUS_ERROR;
                 goto out;
         }
@@ -1366,7 +1366,7 @@ int dso__strerror_load(struct dso *dso, char *buf, size_t buflen)
         BUG_ON(buflen == 0);
 
         if (errnum >= 0) {
-                const char *err = strerror_r(errnum, buf, buflen);
+                const char *err = str_error_r(errnum, buf, buflen);
 
                 if (err != buf)
                         scnprintf(buf, buflen, "%s", err);
diff --git a/tools/perf/util/dso.h b/tools/perf/util/dso.h
index a571f24895ca..ecc4bbd3f82e 100644
--- a/tools/perf/util/dso.h
+++ b/tools/perf/util/dso.h
@@ -4,6 +4,7 @@
 #include <linux/atomic.h>
 #include <linux/types.h>
 #include <linux/rbtree.h>
+#include <sys/types.h>
 #include <stdbool.h>
 #include <pthread.h>
 #include <linux/types.h>
diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h
index 8d363d5e65a2..b32464b353aa 100644
--- a/tools/perf/util/event.h
+++ b/tools/perf/util/event.h
@@ -8,6 +8,7 @@
 #include "map.h"
 #include "build-id.h"
 #include "perf_regs.h"
+#include <asm/perf_regs.h>
 
 struct mmap_event {
         struct perf_event_header header;
diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index 113507716044..862e69c2690d 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -651,8 +651,8 @@ static int perf_evlist__event2id(struct perf_evlist *evlist,
         return 0;
 }
 
-static struct perf_evsel *perf_evlist__event2evsel(struct perf_evlist *evlist,
-                                                   union perf_event *event)
+struct perf_evsel *perf_evlist__event2evsel(struct perf_evlist *evlist,
+                                            union perf_event *event)
 {
         struct perf_evsel *first = perf_evlist__first(evlist);
         struct hlist_head *head;
@@ -1790,7 +1790,7 @@ int perf_evlist__strerror_open(struct perf_evlist *evlist,
                                int err, char *buf, size_t size)
 {
         int printed, value;
-        char sbuf[STRERR_BUFSIZE], *emsg = strerror_r(err, sbuf, sizeof(sbuf));
+        char sbuf[STRERR_BUFSIZE], *emsg = str_error_r(err, sbuf, sizeof(sbuf));
 
         switch (err) {
         case EACCES:
@@ -1842,7 +1842,7 @@ out_default:
 
 int perf_evlist__strerror_mmap(struct perf_evlist *evlist, int err, char *buf, size_t size)
 {
-        char sbuf[STRERR_BUFSIZE], *emsg = strerror_r(err, sbuf, sizeof(sbuf));
+        char sbuf[STRERR_BUFSIZE], *emsg = str_error_r(err, sbuf, sizeof(sbuf));
         int pages_attempted = evlist->mmap_len / 1024, pages_max_per_user, printed = 0;
 
         switch (err) {
diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h
index 872912b392c9..afd087761a47 100644
--- a/tools/perf/util/evlist.h
+++ b/tools/perf/util/evlist.h
@@ -323,4 +323,7 @@ void perf_event_attr__set_max_precise_ip(struct perf_event_attr *attr);
 
 struct perf_evsel *
 perf_evlist__find_evsel_by_str(struct perf_evlist *evlist, const char *str);
+
+struct perf_evsel *perf_evlist__event2evsel(struct perf_evlist *evlist,
+                                            union perf_event *event);
 #endif /* __PERF_EVLIST_H */
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 0fea724e735c..ba0f59fa3d5d 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -200,6 +200,24 @@ void perf_evsel__set_sample_id(struct perf_evsel *evsel,
         evsel->attr.read_format |= PERF_FORMAT_ID;
 }
 
+/**
+ * perf_evsel__is_function_event - Return whether given evsel is a function
+ * trace event
+ *
+ * @evsel - evsel selector to be tested
+ *
+ * Return %true if event is function trace event
+ */
+bool perf_evsel__is_function_event(struct perf_evsel *evsel)
+{
+#define FUNCTION_EVENT "ftrace:function"
+
+        return evsel->name &&
+               !strncmp(FUNCTION_EVENT, evsel->name, sizeof(FUNCTION_EVENT));
+
+#undef FUNCTION_EVENT
+}
+
 void perf_evsel__init(struct perf_evsel *evsel,
                       struct perf_event_attr *attr, int idx)
 {
@@ -2419,7 +2437,7 @@ int perf_evsel__open_strerror(struct perf_evsel *evsel, struct target *target,
         "The sys_perf_event_open() syscall returned with %d (%s) for event (%s).\n"
         "/bin/dmesg may provide additional information.\n"
         "No CONFIG_PERF_EVENTS=y kernel support configured?",
-                         err, strerror_r(err, sbuf, sizeof(sbuf)),
+                         err, str_error_r(err, sbuf, sizeof(sbuf)),
                          perf_evsel__name(evsel));
 }
 
diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
index 86fed7a2932b..d73391e8740e 100644
--- a/tools/perf/util/evsel.h
+++ b/tools/perf/util/evsel.h
@@ -355,23 +355,7 @@ static inline bool perf_evsel__is_group_event(struct perf_evsel *evsel)
         return perf_evsel__is_group_leader(evsel) && evsel->nr_members > 1;
 }
 
-/**
- * perf_evsel__is_function_event - Return whether given evsel is a function
- * trace event
- *
- * @evsel - evsel selector to be tested
- *
- * Return %true if event is function trace event
- */
-static inline bool perf_evsel__is_function_event(struct perf_evsel *evsel)
-{
-#define FUNCTION_EVENT "ftrace:function"
-
-        return evsel->name &&
-               !strncmp(FUNCTION_EVENT, evsel->name, sizeof(FUNCTION_EVENT));
-
-#undef FUNCTION_EVENT
-}
+bool perf_evsel__is_function_event(struct perf_evsel *evsel);
 
 static inline bool perf_evsel__is_bpf_output(struct perf_evsel *evsel)
 {
diff --git a/tools/perf/util/help-unknown-cmd.c b/tools/perf/util/help-unknown-cmd.c
index 776e28562345..2821f8d77e52 100644
--- a/tools/perf/util/help-unknown-cmd.c
+++ b/tools/perf/util/help-unknown-cmd.c
@@ -1,5 +1,6 @@
 #include "cache.h"
 #include "config.h"
+#include <stdio.h>
 #include <subcmd/help.h>
 #include "../builtin.h"
 #include "levenshtein.h"
diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
index e1fcc8d7c01a..a18d142cdca3 100644
--- a/tools/perf/util/hist.c
+++ b/tools/perf/util/hist.c
@@ -352,86 +352,114 @@ void hists__delete_entries(struct hists *hists)
  * histogram, sorted on item, collects periods
  */
 
-static struct hist_entry *hist_entry__new(struct hist_entry *template,
-                                          bool sample_self)
-{
-        size_t callchain_size = 0;
-        struct hist_entry *he;
+static int hist_entry__init(struct hist_entry *he,
+                            struct hist_entry *template,
+                            bool sample_self)
+{
+        *he = *template;
+
+        if (symbol_conf.cumulate_callchain) {
+                he->stat_acc = malloc(sizeof(he->stat));
+                if (he->stat_acc == NULL)
+                        return -ENOMEM;
+                memcpy(he->stat_acc, &he->stat, sizeof(he->stat));
+                if (!sample_self)
+                        memset(&he->stat, 0, sizeof(he->stat));
+        }
 
-        if (symbol_conf.use_callchain)
-                callchain_size = sizeof(struct callchain_root);
+        map__get(he->ms.map);
+
+        if (he->branch_info) {
+                /*
+                 * This branch info is (a part of) allocated from
+                 * sample__resolve_bstack() and will be freed after
+                 * adding new entries.  So we need to save a copy.
+                 */
+                he->branch_info = malloc(sizeof(*he->branch_info));
+                if (he->branch_info == NULL) {
+                        map__zput(he->ms.map);
+                        free(he->stat_acc);
+                        return -ENOMEM;
+                }
+
+                memcpy(he->branch_info, template->branch_info,
+                       sizeof(*he->branch_info));
+
+                map__get(he->branch_info->from.map);
+                map__get(he->branch_info->to.map);
+        }
+
+        if (he->mem_info) {
+                map__get(he->mem_info->iaddr.map);
+                map__get(he->mem_info->daddr.map);
+        }
 
-        he = zalloc(sizeof(*he) + callchain_size);
+        if (symbol_conf.use_callchain)
+                callchain_init(he->callchain);
 
-        if (he != NULL) {
-                *he = *template;
+        if (he->raw_data) {
+                he->raw_data = memdup(he->raw_data, he->raw_size);
 
-                if (symbol_conf.cumulate_callchain) {
-                        he->stat_acc = malloc(sizeof(he->stat));
-                        if (he->stat_acc == NULL) {
-                                free(he);
-                                return NULL;
+                if (he->raw_data == NULL) {
+                        map__put(he->ms.map);
+                        if (he->branch_info) {
+                                map__put(he->branch_info->from.map);
+                                map__put(he->branch_info->to.map);
+                                free(he->branch_info);
                         }
-                        memcpy(he->stat_acc, &he->stat, sizeof(he->stat));
-                        if (!sample_self)
-                                memset(&he->stat, 0, sizeof(he->stat));
+                        if (he->mem_info) {
+                                map__put(he->mem_info->iaddr.map);
+                                map__put(he->mem_info->daddr.map);
+                        }
+                        free(he->stat_acc);
+                        return -ENOMEM;
                 }
+        }
+        INIT_LIST_HEAD(&he->pairs.node);
+        thread__get(he->thread);
 
-                map__get(he->ms.map);
+        if (!symbol_conf.report_hierarchy)
+                he->leaf = true;
 
-                if (he->branch_info) {
-                        /*
-                         * This branch info is (a part of) allocated from
-                         * sample__resolve_bstack() and will be freed after
-                         * adding new entries.  So we need to save a copy.
-                         */
-                        he->branch_info = malloc(sizeof(*he->branch_info));
-                        if (he->branch_info == NULL) {
-                                map__zput(he->ms.map);
-                                free(he->stat_acc);
-                                free(he);
-                                return NULL;
-                        }
+        return 0;
+}
 
-                        memcpy(he->branch_info, template->branch_info,
-                               sizeof(*he->branch_info));
+static void *hist_entry__zalloc(size_t size)
+{
+        return zalloc(size + sizeof(struct hist_entry));
+}
 
-                        map__get(he->branch_info->from.map);
-                        map__get(he->branch_info->to.map);
-                }
+static void hist_entry__free(void *ptr)
+{
+        free(ptr);
+}
 
-                if (he->mem_info) {
-                        map__get(he->mem_info->iaddr.map);
-                        map__get(he->mem_info->daddr.map);
-                }
+static struct hist_entry_ops default_ops = {
+        .new    = hist_entry__zalloc,
+        .free   = hist_entry__free,
+};
 
-                if (symbol_conf.use_callchain)
-                        callchain_init(he->callchain);
+static struct hist_entry *hist_entry__new(struct hist_entry *template,
+                                          bool sample_self)
+{
+        struct hist_entry_ops *ops = template->ops;
+        size_t callchain_size = 0;
+        struct hist_entry *he;
+        int err = 0;
 
-                if (he->raw_data) {
-                        he->raw_data = memdup(he->raw_data, he->raw_size);
+        if (!ops)
+                ops = template->ops = &default_ops;
 
-                        if (he->raw_data == NULL) {
-                                map__put(he->ms.map);
-                                if (he->branch_info) {
-                                        map__put(he->branch_info->from.map);
-                                        map__put(he->branch_info->to.map);
-                                        free(he->branch_info);
-                                }
-                                if (he->mem_info) {
-                                        map__put(he->mem_info->iaddr.map);
-                                        map__put(he->mem_info->daddr.map);
-                                }
-                                free(he->stat_acc);
-                                free(he);
-                                return NULL;
-                        }
-                }
-                INIT_LIST_HEAD(&he->pairs.node);
-                thread__get(he->thread);
+        if (symbol_conf.use_callchain)
+                callchain_size = sizeof(struct callchain_root);
 
-                if (!symbol_conf.report_hierarchy)
-                        he->leaf = true;
+        he = ops->new(callchain_size);
+        if (he) {
+                err = hist_entry__init(he, template, sample_self);
+                if (err) {
+                        ops->free(he);
+                        he = NULL;
+                }
         }
 
         return he;
@@ -531,13 +559,15 @@ out:
         return he;
 }
 
-struct hist_entry *hists__add_entry(struct hists *hists,
-                                    struct addr_location *al,
-                                    struct symbol *sym_parent,
-                                    struct branch_info *bi,
-                                    struct mem_info *mi,
-                                    struct perf_sample *sample,
-                                    bool sample_self)
+static struct hist_entry*
+__hists__add_entry(struct hists *hists,
+                   struct addr_location *al,
+                   struct symbol *sym_parent,
+                   struct branch_info *bi,
+                   struct mem_info *mi,
+                   struct perf_sample *sample,
+                   bool sample_self,
+                   struct hist_entry_ops *ops)
 {
         struct hist_entry entry = {
                 .thread = al->thread,
@@ -564,11 +594,37 @@ struct hist_entry *hists__add_entry(struct hists *hists,
                 .transaction = sample->transaction,
                 .raw_data = sample->raw_data,
                 .raw_size = sample->raw_size,
+                .ops = ops,
         };
 
         return hists__findnew_entry(hists, &entry, al, sample_self);
 }
 
+struct hist_entry *hists__add_entry(struct hists *hists,
+                                    struct addr_location *al,
+                                    struct symbol *sym_parent,
+                                    struct branch_info *bi,
+                                    struct mem_info *mi,
+                                    struct perf_sample *sample,
+                                    bool sample_self)
+{
+        return __hists__add_entry(hists, al, sym_parent, bi, mi,
+                                  sample, sample_self, NULL);
+}
+
+struct hist_entry *hists__add_entry_ops(struct hists *hists,
+                                        struct hist_entry_ops *ops,
+                                        struct addr_location *al,
+                                        struct symbol *sym_parent,
+                                        struct branch_info *bi,
+                                        struct mem_info *mi,
+                                        struct perf_sample *sample,
+                                        bool sample_self)
+{
+        return __hists__add_entry(hists, al, sym_parent, bi, mi,
+                                  sample, sample_self, ops);
+}
+
 static int
 iter_next_nop_entry(struct hist_entry_iter *iter __maybe_unused,
                     struct addr_location *al __maybe_unused)
@@ -1043,6 +1099,8 @@ hist_entry__collapse(struct hist_entry *left, struct hist_entry *right)
 
 void hist_entry__delete(struct hist_entry *he)
 {
+        struct hist_entry_ops *ops = he->ops;
+
         thread__zput(he->thread);
         map__zput(he->ms.map);
 
@@ -1067,7 +1125,7 @@ void hist_entry__delete(struct hist_entry *he)
         free_callchain(he->callchain);
         free(he->trace_output);
         free(he->raw_data);
-        free(he);
+        ops->free(he);
 }
 
 /*
diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h
index 0a03e08be503..49aa4fac148f 100644
--- a/tools/perf/util/hist.h
+++ b/tools/perf/util/hist.h
@@ -10,6 +10,7 @@
 #include "ui/progress.h"
 
 struct hist_entry;
+struct hist_entry_ops;
 struct addr_location;
 struct symbol;
 
@@ -127,6 +128,16 @@ struct hist_entry *hists__add_entry(struct hists *hists,
                                     struct mem_info *mi,
                                     struct perf_sample *sample,
                                     bool sample_self);
+
+struct hist_entry *hists__add_entry_ops(struct hists *hists,
+                                        struct hist_entry_ops *ops,
+                                        struct addr_location *al,
+                                        struct symbol *sym_parent,
+                                        struct branch_info *bi,
+                                        struct mem_info *mi,
+                                        struct perf_sample *sample,
+                                        bool sample_self);
+
 int hist_entry_iter__add(struct hist_entry_iter *iter, struct addr_location *al,
                          int max_stack_depth, void *arg);
 
diff --git a/tools/perf/util/include/asm/alternative-asm.h b/tools/perf/util/include/asm/alternative-asm.h
deleted file mode 100644
index 3a3a0f16456a..000000000000
--- a/tools/perf/util/include/asm/alternative-asm.h
+++ /dev/null
@@ -1,9 +0,0 @@
-#ifndef _PERF_ASM_ALTERNATIVE_ASM_H
-#define _PERF_ASM_ALTERNATIVE_ASM_H
-
-/* Just disable it so we can build arch/x86/lib/memcpy_64.S for perf bench: */
-
-#define altinstruction_entry #
-#define ALTERNATIVE_2 #
-
-#endif
diff --git a/tools/perf/util/include/asm/unistd_32.h b/tools/perf/util/include/asm/unistd_32.h
deleted file mode 100644
index 8b137891791f..000000000000
--- a/tools/perf/util/include/asm/unistd_32.h
+++ /dev/null
@@ -1 +0,0 @@
-
diff --git a/tools/perf/util/include/asm/unistd_64.h b/tools/perf/util/include/asm/unistd_64.h
deleted file mode 100644
index 8b137891791f..000000000000
--- a/tools/perf/util/include/asm/unistd_64.h
+++ /dev/null
@@ -1 +0,0 @@
-
diff --git a/tools/perf/util/intel-pt-decoder/Build b/tools/perf/util/intel-pt-decoder/Build
index 0611d619a42e..9b742ea8bfe8 100644
--- a/tools/perf/util/intel-pt-decoder/Build
+++ b/tools/perf/util/intel-pt-decoder/Build
@@ -7,8 +7,11 @@ $(OUTPUT)util/intel-pt-decoder/inat-tables.c: $(inat_tables_script) $(inat_table
         $(call rule_mkdir)
         @$(call echo-cmd,gen)$(AWK) -f $(inat_tables_script) $(inat_tables_maps) > $@ || rm -f $@
 
+# Busybox's diff doesn't have -I, avoid warning in the case
+
 $(OUTPUT)util/intel-pt-decoder/intel-pt-insn-decoder.o: util/intel-pt-decoder/intel-pt-insn-decoder.c util/intel-pt-decoder/inat.c $(OUTPUT)util/intel-pt-decoder/inat-tables.c
-        @(test -d ../../kernel -a -d ../../tools -a -d ../perf && (( \
+        @(diff -I 2>&1 | grep -q 'option requires an argument' && \
+        test -d ../../kernel -a -d ../../tools -a -d ../perf && (( \
         diff -B -I'^#include' util/intel-pt-decoder/insn.c ../../arch/x86/lib/insn.c >/dev/null && \
         diff -B -I'^#include' util/intel-pt-decoder/inat.c ../../arch/x86/lib/inat.c >/dev/null && \
         diff -B util/intel-pt-decoder/x86-opcode-map.txt ../../arch/x86/lib/x86-opcode-map.txt >/dev/null && \
diff --git a/tools/perf/util/levenshtein.c b/tools/perf/util/levenshtein.c
index e521d1516df6..f616e4f65b67 100644
--- a/tools/perf/util/levenshtein.c
+++ b/tools/perf/util/levenshtein.c
@@ -1,5 +1,7 @@
-#include "cache.h"
 #include "levenshtein.h"
+#include <errno.h>
+#include <stdlib.h>
+#include <string.h>
 
 /*
  * This function implements the Damerau-Levenshtein algorithm to
diff --git a/tools/perf/util/llvm-utils.c b/tools/perf/util/llvm-utils.c
index 40b6f7269cb4..bf7216b8731d 100644
--- a/tools/perf/util/llvm-utils.c
+++ b/tools/perf/util/llvm-utils.c
@@ -3,12 +3,14 @@
  * Copyright (C) 2015, Huawei Inc.
  */
 
+#include <errno.h>
 #include <limits.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include "debug.h"
 #include "llvm-utils.h"
 #include "config.h"
+#include "util.h"
 
 #define CLANG_BPF_CMD_DEFAULT_TEMPLATE                          \
                 "$CLANG_EXEC -D__KERNEL__ -D__NR_CPUS__=$NR_CPUS "\
@@ -106,7 +108,7 @@ read_from_pipe(const char *cmd, void **p_buf, size_t *p_read_sz)
         file = popen(cmd, "r");
         if (!file) {
                 pr_err("ERROR: unable to popen cmd: %s\n",
-                       strerror_r(errno, serr, sizeof(serr)));
+                       str_error_r(errno, serr, sizeof(serr)));
                 return -EINVAL;
         }
 
@@ -140,7 +142,7 @@ read_from_pipe(const char *cmd, void **p_buf, size_t *p_read_sz)
 
         if (ferror(file)) {
                 pr_err("ERROR: error occurred when reading from pipe: %s\n",
-                       strerror_r(errno, serr, sizeof(serr)));
+                       str_error_r(errno, serr, sizeof(serr)));
                 err = -EIO;
                 goto errout;
         }
@@ -382,7 +384,7 @@ int llvm__compile_bpf(const char *path, void **p_obj_buf,
         if (path[0] != '-' && realpath(path, abspath) == NULL) {
                 err = errno;
                 pr_err("ERROR: problems with path %s: %s\n",
-                       path, strerror_r(err, serr, sizeof(serr)));
+                       path, str_error_r(err, serr, sizeof(serr)));
                 return -err;
         }
 
@@ -410,7 +412,7 @@ int llvm__compile_bpf(const char *path, void **p_obj_buf,
         if (nr_cpus_avail <= 0) {
                 pr_err(
 "WARNING:\tunable to get available CPUs in this system: %s\n"
-"        \tUse 128 instead.\n", strerror_r(errno, serr, sizeof(serr)));
+"        \tUse 128 instead.\n", str_error_r(errno, serr, sizeof(serr)));
                 nr_cpus_avail = 128;
         }
         snprintf(nr_cpus_avail_str, sizeof(nr_cpus_avail_str), "%d",
diff --git a/tools/perf/util/path.c b/tools/perf/util/path.c
index cff8bf0f87e8..7c7630be5a89 100644
--- a/tools/perf/util/path.c
+++ b/tools/perf/util/path.c
@@ -11,6 +11,8 @@
  * which is what it's designed for.
  */
 #include "cache.h"
+#include "util.h"
+#include <limits.h>
 
 static char bad_path[] = "/bad-path/";
 /*
diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index 0201f661ccb8..2b222a7955c9 100644
--- a/tools/perf/util/probe-event.c
+++ b/tools/perf/util/probe-event.c
@@ -468,7 +468,7 @@ static struct debuginfo *open_debuginfo(const char *module, bool silent)
                 err = kernel_get_module_dso(module, &dso);
                 if (err < 0) {
                         if (!dso || dso->load_errno == 0) {
-                                if (!strerror_r(-err, reason, STRERR_BUFSIZE))
+                                if (!str_error_r(-err, reason, STRERR_BUFSIZE))
                                         strcpy(reason, "(unknown)");
                         } else
                                 dso__strerror_load(dso, reason, STRERR_BUFSIZE);
@@ -806,7 +806,7 @@ static int __show_one_line(FILE *fp, int l, bool skip, bool show_num)
 error:
         if (ferror(fp)) {
                 pr_warning("File read error: %s\n",
-                           strerror_r(errno, sbuf, sizeof(sbuf)));
+                           str_error_r(errno, sbuf, sizeof(sbuf)));
                 return -1;
         }
         return 0;
@@ -886,7 +886,7 @@ static int __show_line_range(struct line_range *lr, const char *module,
         fp = fopen(lr->path, "r");
         if (fp == NULL) {
                 pr_warning("Failed to open %s: %s\n", lr->path,
-                           strerror_r(errno, sbuf, sizeof(sbuf)));
+                           str_error_r(errno, sbuf, sizeof(sbuf)));
                 return -errno;
         }
         /* Skip to starting line number */
diff --git a/tools/perf/util/probe-file.c b/tools/perf/util/probe-file.c
index 5b563b2e8b1d..e705a742ee1e 100644
--- a/tools/perf/util/probe-file.c
+++ b/tools/perf/util/probe-file.c
@@ -50,7 +50,7 @@ static void print_open_warning(int err, bool uprobe)
         else
                 pr_warning("Failed to open %cprobe_events: %s\n",
                            uprobe ? 'u' : 'k',
-                           strerror_r(-err, sbuf, sizeof(sbuf)));
+                           str_error_r(-err, sbuf, sizeof(sbuf)));
 }
 
 static void print_both_open_warning(int kerr, int uerr)
@@ -64,9 +64,9 @@ static void print_both_open_warning(int kerr, int uerr)
         else {
                 char sbuf[STRERR_BUFSIZE];
                 pr_warning("Failed to open kprobe events: %s.\n",
-                           strerror_r(-kerr, sbuf, sizeof(sbuf)));
+                           str_error_r(-kerr, sbuf, sizeof(sbuf)));
                 pr_warning("Failed to open uprobe events: %s.\n",
-                           strerror_r(-uerr, sbuf, sizeof(sbuf)));
+                           str_error_r(-uerr, sbuf, sizeof(sbuf)));
         }
 }
 
@@ -224,7 +224,7 @@ int probe_file__add_event(int fd, struct probe_trace_event *tev)
                 if (write(fd, buf, strlen(buf)) < (int)strlen(buf)) {
                         ret = -errno;
                         pr_warning("Failed to write event: %s\n",
-                                   strerror_r(errno, sbuf, sizeof(sbuf)));
+                                   str_error_r(errno, sbuf, sizeof(sbuf)));
                 }
         }
         free(buf);
@@ -262,7 +262,7 @@ static int __del_trace_probe_event(int fd, struct str_node *ent)
         return 0;
 error:
         pr_warning("Failed to delete event: %s\n",
-                   strerror_r(-ret, buf, sizeof(buf)));
+                   str_error_r(-ret, buf, sizeof(buf)));
         return ret;
 }
 
@@ -624,6 +624,7 @@ out_err:
         return ret;
 }
 
+#ifdef HAVE_GELF_GETNOTE_SUPPORT
 static unsigned long long sdt_note__get_addr(struct sdt_note *note)
 {
         return note->bit32 ? (unsigned long long)note->addr.a32[0]
@@ -682,6 +683,7 @@ int probe_cache__scan_sdt(struct probe_cache *pcache, const char *pathname)
         cleanup_sdt_note_list(&sdtlist);
         return ret;
 }
+#endif
 
 static int probe_cache_entry__write(struct probe_cache_entry *entry, int fd)
 {
diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
index 1259839dbf6d..f2d9ff064e2d 100644
--- a/tools/perf/util/probe-finder.c
+++ b/tools/perf/util/probe-finder.c
@@ -381,7 +381,7 @@ formatted:
                 if (ret >= 16)
                         ret = -E2BIG;
                 pr_warning("Failed to convert variable type: %s\n",
-                           strerror_r(-ret, sbuf, sizeof(sbuf)));
+                           str_error_r(-ret, sbuf, sizeof(sbuf)));
                 return ret;
         }
         tvar->type = strdup(buf);
@@ -809,7 +809,7 @@ static int find_lazy_match_lines(struct intlist *list,
         fp = fopen(fname, "r");
         if (!fp) {
                 pr_warning("Failed to open %s: %s\n", fname,
-                           strerror_r(errno, sbuf, sizeof(sbuf)));
+                           str_error_r(errno, sbuf, sizeof(sbuf)));
                 return -errno;
         }
 
diff --git a/tools/perf/util/python-ext-sources b/tools/perf/util/python-ext-sources
index 36c6862119e3..5065ec98049c 100644
--- a/tools/perf/util/python-ext-sources
+++ b/tools/perf/util/python-ext-sources
@@ -13,6 +13,8 @@ util/cpumap.c
 ../lib/bitmap.c
 ../lib/find_bit.c
 ../lib/hweight.c
+../lib/str_error_r.c
+../lib/vsprintf.c
 util/thread_map.c
 util/util.c
 util/xyarray.c
diff --git a/tools/perf/util/python.c b/tools/perf/util/python.c
index 65c6c73d28fe..d32f97033718 100644
--- a/tools/perf/util/python.c
+++ b/tools/perf/util/python.c
@@ -2,6 +2,7 @@
 #include <structmember.h>
 #include <inttypes.h>
 #include <poll.h>
+#include <linux/err.h>
 #include "evlist.h"
 #include "evsel.h"
 #include "event.h"
@@ -47,6 +48,7 @@ PyMODINIT_FUNC initperf(void);
 
 struct pyrf_event {
         PyObject_HEAD
+        struct perf_evsel *evsel;
         struct perf_sample sample;
         union perf_event   event;
 };
@@ -288,6 +290,97 @@ static PyObject *pyrf_sample_event__repr(struct pyrf_event *pevent)
         return ret;
 }
 
+static bool is_tracepoint(struct pyrf_event *pevent)
+{
+        return pevent->evsel->attr.type == PERF_TYPE_TRACEPOINT;
+}
+
+static int is_printable_array(char *p, unsigned int len)
+{
+        unsigned int i;
+
+        for (i = 0; i < len; i++) {
+                if (!isprint(p[i]) && !isspace(p[i]))
+                        return 0;
+        }
+
+        return 1;
+}
+
+static PyObject*
+tracepoint_field(struct pyrf_event *pe, struct format_field *field)
+{
+        struct pevent *pevent = field->event->pevent;
+        void *data = pe->sample.raw_data;
+        PyObject *ret = NULL;
+        unsigned long long val;
+        unsigned int offset, len;
+
+        if (field->flags & FIELD_IS_ARRAY) {
+                offset = field->offset;
+                len    = field->size;
+                if (field->flags & FIELD_IS_DYNAMIC) {
+                        val     = pevent_read_number(pevent, data + offset, len);
+                        offset  = val;
+                        len     = offset >> 16;
+                        offset &= 0xffff;
+                }
+                if (field->flags & FIELD_IS_STRING &&
+                    is_printable_array(data + offset, len)) {
+                        ret = PyString_FromString((char *)data + offset);
+                } else {
+                        ret = PyByteArray_FromStringAndSize((const char *) data + offset, len);
+                        field->flags &= ~FIELD_IS_STRING;
+                }
+        } else {
+                val = pevent_read_number(pevent, data + field->offset,
+                                         field->size);
+                if (field->flags & FIELD_IS_POINTER)
+                        ret = PyLong_FromUnsignedLong((unsigned long) val);
+                else if (field->flags & FIELD_IS_SIGNED)
+                        ret = PyLong_FromLong((long) val);
+                else
+                        ret = PyLong_FromUnsignedLong((unsigned long) val);
+        }
+
+        return ret;
+}
+
+static PyObject*
+get_tracepoint_field(struct pyrf_event *pevent, PyObject *attr_name)
+{
+        const char *str = PyString_AsString(PyObject_Str(attr_name));
+        struct perf_evsel *evsel = pevent->evsel;
+        struct format_field *field;
+
+        if (!evsel->tp_format) {
+                struct event_format *tp_format;
+
+                tp_format = trace_event__tp_format_id(evsel->attr.config);
+                if (!tp_format)
+                        return NULL;
+
+                evsel->tp_format = tp_format;
+        }
+
+        field = pevent_find_any_field(evsel->tp_format, str);
+        if (!field)
+                return NULL;
+
+        return tracepoint_field(pevent, field);
+}
+
+static PyObject*
+pyrf_sample_event__getattro(struct pyrf_event *pevent, PyObject *attr_name)
+{
+        PyObject *obj = NULL;
+
+        if (is_tracepoint(pevent))
+                obj = get_tracepoint_field(pevent, attr_name);
+
+        return obj ?: PyObject_GenericGetAttr((PyObject *) pevent, attr_name);
+}
+
 static PyTypeObject pyrf_sample_event__type = {
         PyVarObject_HEAD_INIT(NULL, 0)
         .tp_name        = "perf.sample_event",
@@ -296,6 +389,7 @@ static PyTypeObject pyrf_sample_event__type = {
         .tp_doc         = pyrf_sample_event__doc,
         .tp_members     = pyrf_sample_event__members,
         .tp_repr        = (reprfunc)pyrf_sample_event__repr,
+        .tp_getattro    = (getattrofunc) pyrf_sample_event__getattro,
 };
 
 static char pyrf_context_switch_event__doc[] = PyDoc_STR("perf context_switch event object.");
@@ -653,6 +747,7 @@ static int pyrf_evsel__init(struct pyrf_evsel *pevsel,
         attr.precise_ip     = precise_ip;
         attr.mmap_data      = mmap_data;
         attr.sample_id_all  = sample_id_all;
+        attr.size           = sizeof(attr);
 
         perf_evsel__init(&pevsel->evsel, &attr, idx);
         return 0;
@@ -863,13 +958,22 @@ static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist,
         if (event != NULL) {
                 PyObject *pyevent = pyrf_event__new(event);
                 struct pyrf_event *pevent = (struct pyrf_event *)pyevent;
-
-                perf_evlist__mmap_consume(evlist, cpu);
+                struct perf_evsel *evsel;
 
                 if (pyevent == NULL)
                         return PyErr_NoMemory();
 
-                err = perf_evlist__parse_sample(evlist, event, &pevent->sample);
+                evsel = perf_evlist__event2evsel(evlist, event);
+                if (!evsel)
+                        return Py_None;
+
+                pevent->evsel = evsel;
+
+                err = perf_evsel__parse_sample(evsel, event, &pevent->sample);
+
+                /* Consume the even only after we parsed it out. */
+                perf_evlist__mmap_consume(evlist, cpu);
+
                 if (err)
                         return PyErr_Format(PyExc_OSError,
                                             "perf: can't parse sample, err=%d", err);
@@ -1073,7 +1177,32 @@ static struct {
         { .name = NULL, },
 };
 
+static PyObject *pyrf__tracepoint(struct pyrf_evsel *pevsel,
+                                  PyObject *args, PyObject *kwargs)
+{
+        struct event_format *tp_format;
+        static char *kwlist[] = { "sys", "name", NULL };
+        char *sys  = NULL;
+        char *name = NULL;
+
+        if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|ss", kwlist,
+                                         &sys, &name))
+                return NULL;
+
+        tp_format = trace_event__tp_format(sys, name);
+        if (IS_ERR(tp_format))
+                return PyInt_FromLong(-1);
+
+        return PyInt_FromLong(tp_format->id);
+}
+
 static PyMethodDef perf__methods[] = {
+        {
+                .ml_name  = "tracepoint",
+                .ml_meth  = (PyCFunction) pyrf__tracepoint,
+                .ml_flags = METH_VARARGS | METH_KEYWORDS,
+                .ml_doc   = PyDoc_STR("Get tracepoint config.")
+        },
         { .ml_name = NULL, }
 };
 
@@ -1100,6 +1229,33 @@ PyMODINIT_FUNC initperf(void)
         Py_INCREF(&pyrf_evsel__type);
         PyModule_AddObject(module, "evsel", (PyObject*)&pyrf_evsel__type);
 
+        Py_INCREF(&pyrf_mmap_event__type);
+        PyModule_AddObject(module, "mmap_event", (PyObject *)&pyrf_mmap_event__type);
+
+        Py_INCREF(&pyrf_lost_event__type);
+        PyModule_AddObject(module, "lost_event", (PyObject *)&pyrf_lost_event__type);
+
+        Py_INCREF(&pyrf_comm_event__type);
+        PyModule_AddObject(module, "comm_event", (PyObject *)&pyrf_comm_event__type);
+
+        Py_INCREF(&pyrf_task_event__type);
+        PyModule_AddObject(module, "task_event", (PyObject *)&pyrf_task_event__type);
+
+        Py_INCREF(&pyrf_throttle_event__type);
+        PyModule_AddObject(module, "throttle_event", (PyObject *)&pyrf_throttle_event__type);
+
+        Py_INCREF(&pyrf_task_event__type);
+        PyModule_AddObject(module, "task_event", (PyObject *)&pyrf_task_event__type);
+
+        Py_INCREF(&pyrf_read_event__type);
+        PyModule_AddObject(module, "read_event", (PyObject *)&pyrf_read_event__type);
+
+        Py_INCREF(&pyrf_sample_event__type);
+        PyModule_AddObject(module, "sample_event", (PyObject *)&pyrf_sample_event__type);
+
+        Py_INCREF(&pyrf_context_switch_event__type);
+        PyModule_AddObject(module, "switch_event", (PyObject *)&pyrf_context_switch_event__type);
+
         Py_INCREF(&pyrf_thread_map__type);
         PyModule_AddObject(module, "thread_map", (PyObject*)&pyrf_thread_map__type);
 
diff --git a/tools/perf/util/quote.c b/tools/perf/util/quote.c
index c6d4ee2de752..639d1da2f978 100644
--- a/tools/perf/util/quote.c
+++ b/tools/perf/util/quote.c
@@ -1,5 +1,7 @@
-#include "cache.h"
+#include <stdlib.h>
+#include "strbuf.h"
 #include "quote.h"
+#include "util.h"
 
 /* Help to copy the thing properly quoted for the shell safety.
  * any single quote is replaced with '\'', any exclamation point
diff --git a/tools/perf/util/quote.h b/tools/perf/util/quote.h
index e1ec19146fb0..055ca45bed99 100644
--- a/tools/perf/util/quote.h
+++ b/tools/perf/util/quote.h
@@ -2,7 +2,6 @@
 #define __PERF_QUOTE_H
 
 #include <stddef.h>
-#include <stdio.h>
 
 /* Help to copy the thing properly quoted for the shell safety.
  * any single quote is replaced with '\'', any exclamation point
@@ -24,6 +23,8 @@
  * sq_quote() in a real application.
  */
 
+struct strbuf;
+
 int sq_quote_argv(struct strbuf *, const char **argv, size_t maxlen);
 
 #endif /* __PERF_QUOTE_H */
diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c
index ff134700bf30..6ac6b7a33f42 100644
--- a/tools/perf/util/scripting-engines/trace-event-python.c
+++ b/tools/perf/util/scripting-engines/trace-event-python.c
@@ -273,7 +273,7 @@ static PyObject *get_field_numeric_entry(struct event_format *event,
                 struct format_field *field, void *data)
 {
         bool is_array = field->flags & FIELD_IS_ARRAY;
-        PyObject *obj, *list = NULL;
+        PyObject *obj = NULL, *list = NULL;
         unsigned long long val;
         unsigned int item_size, n_items, i;
 
@@ -392,7 +392,7 @@ static void python_process_tracepoint(struct perf_sample *sample,
                                       struct addr_location *al)
 {
         struct event_format *event = evsel->tp_format;
-        PyObject *handler, *context, *t, *obj, *callchain;
+        PyObject *handler, *context, *t, *obj = NULL, *callchain;
         PyObject *dict = NULL;
         static char handler_name[256];
         struct format_field *field;
diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h
index ebb59cacd092..7ca37ea17395 100644
--- a/tools/perf/util/sort.h
+++ b/tools/perf/util/sort.h
@@ -67,6 +67,11 @@ struct hist_entry_diff {
         };
 };
 
+struct hist_entry_ops {
+        void    *(*new)(size_t size);
+        void    (*free)(void *ptr);
+};
+
 /**
  * struct hist_entry - histogram entry
  *
@@ -125,6 +130,7 @@ struct hist_entry {
         void                    *trace_output;
         struct perf_hpp_list    *hpp_list;
         struct hist_entry       *parent_he;
+        struct hist_entry_ops   *ops;
         union {
                 /* this is for hierarchical entry structure */
                 struct {
diff --git a/tools/perf/util/strbuf.c b/tools/perf/util/strbuf.c
index f95f682aa2b2..817593908d47 100644
--- a/tools/perf/util/strbuf.c
+++ b/tools/perf/util/strbuf.c
@@ -1,5 +1,5 @@
 #include "debug.h"
-#include "cache.h"
+#include "util.h"
 #include <linux/kernel.h>
 
 int prefixcmp(const char *str, const char *prefix)
diff --git a/tools/perf/util/strbuf.h b/tools/perf/util/strbuf.h
index 54b409297d4a..b268a6648a5d 100644
--- a/tools/perf/util/strbuf.h
+++ b/tools/perf/util/strbuf.h
@@ -40,6 +40,9 @@
 
 #include <assert.h>
 #include <stdarg.h>
+#include <stddef.h>
+#include <string.h>
+#include <sys/types.h>
 
 extern char strbuf_slopbuf[];
 struct strbuf {
diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c
index 6f15b92cbf70..a34321e9b44d 100644
--- a/tools/perf/util/symbol-elf.c
+++ b/tools/perf/util/symbol-elf.c
@@ -7,6 +7,7 @@
 
 #include "symbol.h"
 #include "demangle-java.h"
+#include "demangle-rust.h"
 #include "machine.h"
 #include "vdso.h"
 #include <symbol/kallsyms.h>
@@ -16,6 +17,7 @@
 #define EM_AARCH64      183  /* ARM 64 bit */
 #endif
 
+typedef Elf64_Nhdr GElf_Nhdr;
 
 #ifdef HAVE_CPLUS_DEMANGLE_SUPPORT
 extern char *cplus_demangle(const char *, int);
@@ -1080,6 +1082,13 @@ new_symbol:
                         demangled = bfd_demangle(NULL, elf_name, demangle_flags);
                         if (demangled == NULL)
                                 demangled = java_demangle_sym(elf_name, JAVA_DEMANGLE_NORET);
+                        else if (rust_is_mangled(demangled))
+                                /*
+                                 * Input to Rust demangling is the BFD-demangled
+                                 * name which it Rust-demangles in place.
+                                 */
+                                rust_demangle_sym(demangled);
+
                         if (demangled != NULL)
                                 elf_name = demangled;
                 }
@@ -1789,6 +1798,7 @@ void kcore_extract__delete(struct kcore_extract *kce)
         unlink(kce->extract_filename);
 }
 
+#ifdef HAVE_GELF_GETNOTE_SUPPORT
 /**
  * populate_sdt_note : Parse raw data and identify SDT note
  * @elf: elf of the opened file
@@ -2040,6 +2050,7 @@ int sdt_notes__get_count(struct list_head *start)
                 count++;
         return count;
 }
+#endif
 
 void symbol__elf_init(void)
 {
diff --git a/tools/perf/util/target.c b/tools/perf/util/target.c
index a53603b27e52..8cdcf4641c51 100644
--- a/tools/perf/util/target.c
+++ b/tools/perf/util/target.c
@@ -7,6 +7,7 @@
  */
 
 #include "target.h"
+#include "util.h"
 #include "debug.h"
 
 #include <pwd.h>
@@ -121,7 +122,7 @@ int target__strerror(struct target *target, int errnum,
         BUG_ON(buflen == 0);
 
         if (errnum >= 0) {
-                const char *err = strerror_r(errnum, buf, buflen);
+                const char *err = str_error_r(errnum, buf, buflen);
 
                 if (err != buf)
                         scnprintf(buf, buflen, "%s", err);
diff --git a/tools/perf/util/trace-event.c b/tools/perf/util/trace-event.c
index 8ae051e0ec79..c330780674fc 100644
--- a/tools/perf/util/trace-event.c
+++ b/tools/perf/util/trace-event.c
@@ -105,3 +105,11 @@ trace_event__tp_format(const char *sys, const char *name)
 
         return tp_format(sys, name);
 }
+
+struct event_format *trace_event__tp_format_id(int id)
+{
+        if (!tevent_initialized && trace_event__init2())
+                return ERR_PTR(-ENOMEM);
+
+        return pevent_find_event(tevent.pevent, id);
+}
diff --git a/tools/perf/util/trace-event.h b/tools/perf/util/trace-event.h
index bce5b1dac268..b0af9c81bb0d 100644
--- a/tools/perf/util/trace-event.h
+++ b/tools/perf/util/trace-event.h
@@ -23,6 +23,8 @@ int trace_event__register_resolver(struct machine *machine,
 struct event_format*
 trace_event__tp_format(const char *sys, const char *name);
 
+struct event_format *trace_event__tp_format_id(int id);
+
 int bigendian(void);
 
 void event_format__fprintf(struct event_format *event,
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index 1e8c3167b9fb..6178cab82374 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -72,7 +72,6 @@
 #include <sys/ioctl.h>
 #include <inttypes.h>
 #include <linux/kernel.h>
-#include <linux/magic.h>
 #include <linux/types.h>
 #include <sys/ttydefaults.h>
 #include <api/fs/tracing_path.h>
@@ -360,4 +359,9 @@ typedef void (*print_binary_t)(enum binary_printer_ops,
 void print_binary(unsigned char *data, size_t len,
                   size_t bytes_per_line, print_binary_t printer,
                   void *extra);
+
+#ifndef __GLIBC__
+extern int sched_getcpu(void);
+#endif
+
 #endif /* GIT_COMPAT_UTIL_H */