import gcc-toolset-11-gcc-11.2.1-9.1.el8

This commit is contained in:
CentOS Sources 2022-05-10 02:59:12 -04:00 committed by Stepan Oksanichenko
parent ec219542a7
commit bc7dedae65
29 changed files with 3464 additions and 269 deletions

View File

@ -1,5 +1,5 @@
7f4348418dc3efefd357b32a2b5c8010211ab284 SOURCES/doxygen-1.8.0.src.tar.gz
8bb88a8a2cc72b6e4563e0e5e19f53ff09d72ee1 SOURCES/gcc-11.2.1-20210728.tar.xz
54f8e112180f2cad6cf9459aab4681c7f157aca9 SOURCES/gcc-11.2.1-20220127.tar.xz
db38c7b67f8eea9f2e5b8a48d219165b2fdab11f SOURCES/gmp-6.1.0.tar.bz2
bbffc5a2b05e4f0c97e882f96c448504491dc4ed SOURCES/isl-0.18.tar.bz2
b8be66396c726fdc36ebb0f692ed8a8cca3bcc66 SOURCES/mpc-1.0.3.tar.gz

2
.gitignore vendored
View File

@ -1,5 +1,5 @@
SOURCES/doxygen-1.8.0.src.tar.gz
SOURCES/gcc-11.2.1-20210728.tar.xz
SOURCES/gcc-11.2.1-20220127.tar.xz
SOURCES/gmp-6.1.0.tar.bz2
SOURCES/isl-0.18.tar.bz2
SOURCES/mpc-1.0.3.tar.gz

View File

@ -0,0 +1,85 @@
From 8a7f364afd86a4c4c2c747ae9cb4216fe992acc8 Mon Sep 17 00:00:00 2001
From: David Malcolm <dmalcolm@redhat.com>
Date: Wed, 25 Aug 2021 12:36:42 -0400
Subject: [PATCH 01/17] basic_string::reserve(n) semantics are not available in
DTS
Various tests were added upstream 2020-08-06 as part of:
"libstdc++: Implement P0966 std::string::reserve should not shrink"
https://gcc.gnu.org/git/?p=gcc.git;a=commitdiff;h=140cf935cd118f7208b7c3826a8b9d50936242f0
std::string and std::wstring are instantiated in the system libstdc++.so
via explicit instantiation definitions, so the new basic_string::reserve(n)
semantics are not available in DTS.
Update/disable the pertinent parts of the tests to reflect the behavior
when run against the system libstdc++.so.
---
.../testsuite/21_strings/basic_string/capacity/char/1.cc | 6 ------
.../21_strings/basic_string/capacity/char/18654.cc | 2 +-
.../testsuite/21_strings/basic_string/capacity/wchar_t/1.cc | 6 ------
.../21_strings/basic_string/capacity/wchar_t/18654.cc | 2 +-
4 files changed, 2 insertions(+), 14 deletions(-)
diff --git a/libstdc++-v3/testsuite/21_strings/basic_string/capacity/char/1.cc b/libstdc++-v3/testsuite/21_strings/basic_string/capacity/char/1.cc
index eea69771f..64187718d 100644
--- a/libstdc++-v3/testsuite/21_strings/basic_string/capacity/char/1.cc
+++ b/libstdc++-v3/testsuite/21_strings/basic_string/capacity/char/1.cc
@@ -41,12 +41,6 @@ void test01()
sz01 = str01.capacity();
VERIFY( sz01 < sz02 );
- // P0966: reserve should not shrink
- str01.reserve(100);
- sz01 = str01.capacity();
- str01.reserve(sz01 - 1);
- VERIFY( str01.capacity() == sz01 );
-
sz01 = str01.size() + 5;
str01.resize(sz01);
sz02 = str01.size();
diff --git a/libstdc++-v3/testsuite/21_strings/basic_string/capacity/char/18654.cc b/libstdc++-v3/testsuite/21_strings/basic_string/capacity/char/18654.cc
index 02ce78ea6..3a7352123 100644
--- a/libstdc++-v3/testsuite/21_strings/basic_string/capacity/char/18654.cc
+++ b/libstdc++-v3/testsuite/21_strings/basic_string/capacity/char/18654.cc
@@ -51,7 +51,7 @@ void test01()
VERIFY( cap >= 3 * i );
str.reserve(2 * i);
- VERIFY( str.capacity() == cap );
+ VERIFY( str.capacity() == 2 * i );
#if __cplusplus <= 201703L
str.reserve();
diff --git a/libstdc++-v3/testsuite/21_strings/basic_string/capacity/wchar_t/1.cc b/libstdc++-v3/testsuite/21_strings/basic_string/capacity/wchar_t/1.cc
index f01a27e8c..70915a94d 100644
--- a/libstdc++-v3/testsuite/21_strings/basic_string/capacity/wchar_t/1.cc
+++ b/libstdc++-v3/testsuite/21_strings/basic_string/capacity/wchar_t/1.cc
@@ -41,12 +41,6 @@ void test01()
sz01 = str01.capacity();
VERIFY( sz01 < sz02 );
- // P0966: reserve should not shrink
- str01.reserve(100);
- sz01 = str01.capacity();
- str01.reserve(sz01 - 1);
- VERIFY( str01.capacity() == sz01 );
-
sz01 = str01.size() + 5;
str01.resize(sz01);
sz02 = str01.size();
diff --git a/libstdc++-v3/testsuite/21_strings/basic_string/capacity/wchar_t/18654.cc b/libstdc++-v3/testsuite/21_strings/basic_string/capacity/wchar_t/18654.cc
index 267fd198b..c9711a294 100644
--- a/libstdc++-v3/testsuite/21_strings/basic_string/capacity/wchar_t/18654.cc
+++ b/libstdc++-v3/testsuite/21_strings/basic_string/capacity/wchar_t/18654.cc
@@ -51,7 +51,7 @@ void test01()
VERIFY( cap >= 3 * i );
str.reserve(2 * i);
- VERIFY( str.capacity() == cap );
+ VERIFY( str.capacity() == 2 * i );
#if __cplusplus <= 201703L
str.reserve();
--
2.31.1

View File

@ -0,0 +1,46 @@
From d7fae9f17479c44dae68483d457fc3fbb58c4b83 Mon Sep 17 00:00:00 2001
From: David Malcolm <dmalcolm@redhat.com>
Date: Wed, 25 Aug 2021 15:57:36 -0400
Subject: [PATCH 04/17] operator>>(istream&, char(&)[N]) eofbit fixes are not
available in DTS
The upstream commit on 2020-08--6:
Do not set eofbit eagerly in operator>>(istream&, char(&)[N])
https://gcc.gnu.org/git/?p=gcc.git;a=commitdiff;h=4e39f563c0cd25401f689c2093cb8c13692156ef
altered the behaviour of std::istream symbols defined in libstdc++.so,
but with DTS you get the old definitions from the system libstdc++.so
Hence this patch tweaks the new tests to avoid failing when run against
the system libstdc++.so
---
.../21_strings/basic_string/inserters_extractors/char/13.cc | 1 -
.../21_strings/basic_string/inserters_extractors/wchar_t/13.cc | 1 -
2 files changed, 2 deletions(-)
diff --git a/libstdc++-v3/testsuite/21_strings/basic_string/inserters_extractors/char/13.cc b/libstdc++-v3/testsuite/21_strings/basic_string/inserters_extractors/char/13.cc
index a25d8a93d..7d85e6ead 100644
--- a/libstdc++-v3/testsuite/21_strings/basic_string/inserters_extractors/char/13.cc
+++ b/libstdc++-v3/testsuite/21_strings/basic_string/inserters_extractors/char/13.cc
@@ -29,7 +29,6 @@ test01()
std::string buf;
in.width(4);
in >> buf;
- VERIFY( !in.eof() ); // should stop after reading 4 chars
VERIFY( buf == str );
}
diff --git a/libstdc++-v3/testsuite/21_strings/basic_string/inserters_extractors/wchar_t/13.cc b/libstdc++-v3/testsuite/21_strings/basic_string/inserters_extractors/wchar_t/13.cc
index f15294a44..fa5fe3d4a 100644
--- a/libstdc++-v3/testsuite/21_strings/basic_string/inserters_extractors/wchar_t/13.cc
+++ b/libstdc++-v3/testsuite/21_strings/basic_string/inserters_extractors/wchar_t/13.cc
@@ -29,7 +29,6 @@ test01()
std::wstring buf;
in.width(4);
in >> buf;
- VERIFY( !in.eof() ); // should stop after reading 4 chars
VERIFY( buf == str );
}
--
2.31.1

View File

@ -0,0 +1,53 @@
From 3ede89bd19328c26bcd881b873cf4a766ae0da3a Mon Sep 17 00:00:00 2001
From: David Malcolm <dmalcolm@redhat.com>
Date: Wed, 25 Aug 2021 17:04:02 -0400
Subject: [PATCH 05/17] Disable tests for PR libstdc++/79820 and PR
libstdc++/81751 under DTS
Upstream commit 2017-08-09
PR libstdc++/81751 don't call fflush(NULL)
https://gcc.gnu.org/git/?p=gcc.git;a=commitdiff;h=04d07b61cb80fd97e98eb39451ff6a8675a63d90
added these test cases as part of verifying the behavior of sys_open (that
it resets errno and doesn't call fflush on NULL).
These symbols are in the system stdlib when run from DTS and thus aren't
fixed by the above change:
1521: 000000000007c430 104 FUNC GLOBAL DEFAULT 12 std::__basic_file<char>::sys_open(int, std::_Ios_Openmode)@@GLIBCXX_3.4
2895: 000000000007c730 95 FUNC GLOBAL DEFAULT 12 std::__basic_file<char>::sys_open(_IO_FILE*, std::_Ios_Openmode)@@GLIBCXX_3.4
This patch disables the non-fixed assertions.
---
libstdc++-v3/testsuite/ext/stdio_filebuf/char/79820.cc | 2 --
libstdc++-v3/testsuite/ext/stdio_filebuf/char/81751.cc | 1 -
2 files changed, 3 deletions(-)
diff --git a/libstdc++-v3/testsuite/ext/stdio_filebuf/char/79820.cc b/libstdc++-v3/testsuite/ext/stdio_filebuf/char/79820.cc
index 278b99169..e2a8e3b3d 100644
--- a/libstdc++-v3/testsuite/ext/stdio_filebuf/char/79820.cc
+++ b/libstdc++-v3/testsuite/ext/stdio_filebuf/char/79820.cc
@@ -26,9 +26,7 @@ void
test01()
{
FILE* f = std::fopen("79820.txt", "w");
- errno = 127;
__gnu_cxx::stdio_filebuf<char> b(f, std::ios::out, BUFSIZ);
- VERIFY(errno == 127); // PR libstdc++/79820
b.close();
std::fclose(f);
}
diff --git a/libstdc++-v3/testsuite/ext/stdio_filebuf/char/81751.cc b/libstdc++-v3/testsuite/ext/stdio_filebuf/char/81751.cc
index 21aa06f78..6a231d7cf 100644
--- a/libstdc++-v3/testsuite/ext/stdio_filebuf/char/81751.cc
+++ b/libstdc++-v3/testsuite/ext/stdio_filebuf/char/81751.cc
@@ -31,7 +31,6 @@ test01()
FILE* in1 = std::fopen("81751.txt", "r");
__gnu_cxx::stdio_filebuf<char> buf1(in1, std::ios::in, BUFSIZ);
int c = buf1.sgetc();
- VERIFY( c == std::char_traits<char>::eof() ); // PR libstdc++/81751
std::fflush(out);
FILE* in2 = std::fopen("81751.txt", "r");
--
2.31.1

View File

@ -0,0 +1,118 @@
From 59388ee3dac5c737086ed1f83f552a44481a2213 Mon Sep 17 00:00:00 2001
From: David Malcolm <dmalcolm@redhat.com>
Date: Tue, 31 Aug 2021 16:08:57 -0400
Subject: [PATCH 06/17] Don't assume has_facet<codecvt_c16> when run against
system libstdc++
Upstream commit 2019-02-19:
P0482R5 char8_t: Standard library support
https://gcc.gnu.org/git/?p=gcc.git;a=commitdiff;h=c124af936b6b225eb548ccdd7f01400511d784dc
added new locale facets for char8_t.
has_facet<codecvt_c16> returns false, as it is using the std::locale the system libstdc++:
$ eu-readelf -s char16_t-char8_t.exe |c++filt |grep UNDEF|grep locale
3: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF std::locale::classic()@GLIBCXX_3.4 (2)
8: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF std::locale::locale(std::locale const&)@GLIBCXX_3.4 (2)
13: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF std::locale::id::_M_id() const@GLIBCXX_3.4 (2)
22: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF std::locale::~locale()@GLIBCXX_3.4 (2)
64: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF std::locale::classic()@@GLIBCXX_3.4
76: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF std::locale::locale(std::locale const&)@@GLIBCXX_3.4
89: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF std::locale::id::_M_id() const@@GLIBCXX_3.4
108: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF std::locale::~locale()@@GLIBCXX_3.4
Hence this patch tweaks tests so that they bail out if run against an older std::locale.
---
libstdc++-v3/testsuite/22_locale/codecvt/char16_t-char8_t.cc | 3 ++-
libstdc++-v3/testsuite/22_locale/codecvt/char16_t.cc | 3 ++-
libstdc++-v3/testsuite/22_locale/codecvt/char32_t-char8_t.cc | 3 ++-
libstdc++-v3/testsuite/22_locale/codecvt/char32_t.cc | 3 ++-
libstdc++-v3/testsuite/22_locale/codecvt/utf8-char8_t.cc | 2 ++
libstdc++-v3/testsuite/22_locale/codecvt/utf8.cc | 2 ++
6 files changed, 12 insertions(+), 4 deletions(-)
diff --git a/libstdc++-v3/testsuite/22_locale/codecvt/char16_t-char8_t.cc b/libstdc++-v3/testsuite/22_locale/codecvt/char16_t-char8_t.cc
index 71a23bef4..d4aa519a2 100644
--- a/libstdc++-v3/testsuite/22_locale/codecvt/char16_t-char8_t.cc
+++ b/libstdc++-v3/testsuite/22_locale/codecvt/char16_t-char8_t.cc
@@ -31,7 +31,8 @@ test01()
using namespace std;
typedef codecvt<char16_t, char8_t, mbstate_t> codecvt_c16;
locale loc_c = locale::classic();
- VERIFY(has_facet<codecvt_c16>(loc_c));
+ if (!has_facet<codecvt_c16>(loc_c))
+ return;
const codecvt_c16* const cvt = &use_facet<codecvt_c16>(loc_c);
VERIFY(!cvt->always_noconv());
diff --git a/libstdc++-v3/testsuite/22_locale/codecvt/char16_t.cc b/libstdc++-v3/testsuite/22_locale/codecvt/char16_t.cc
index 7b5ce126d..05c77b2b1 100644
--- a/libstdc++-v3/testsuite/22_locale/codecvt/char16_t.cc
+++ b/libstdc++-v3/testsuite/22_locale/codecvt/char16_t.cc
@@ -29,7 +29,8 @@ test01()
using namespace std;
typedef codecvt<char16_t, char, mbstate_t> codecvt_c16;
locale loc_c = locale::classic();
- VERIFY(has_facet<codecvt_c16>(loc_c));
+ if (!has_facet<codecvt_c16>(loc_c))
+ return;
const codecvt_c16* const cvt = &use_facet<codecvt_c16>(loc_c);
VERIFY(!cvt->always_noconv());
diff --git a/libstdc++-v3/testsuite/22_locale/codecvt/char32_t-char8_t.cc b/libstdc++-v3/testsuite/22_locale/codecvt/char32_t-char8_t.cc
index f30ae22e9..281969e9b 100644
--- a/libstdc++-v3/testsuite/22_locale/codecvt/char32_t-char8_t.cc
+++ b/libstdc++-v3/testsuite/22_locale/codecvt/char32_t-char8_t.cc
@@ -33,7 +33,8 @@ test01()
using namespace std;
typedef codecvt<char32_t, char8_t, mbstate_t> codecvt_c32;
locale loc_c = locale::classic();
- VERIFY(has_facet<codecvt_c32>(loc_c));
+ if (!has_facet<codecvt_c32>(loc_c))
+ return;
const codecvt_c32* const cvt = &use_facet<codecvt_c32>(loc_c);
VERIFY(!cvt->always_noconv());
diff --git a/libstdc++-v3/testsuite/22_locale/codecvt/char32_t.cc b/libstdc++-v3/testsuite/22_locale/codecvt/char32_t.cc
index 558ba9145..82980b702 100644
--- a/libstdc++-v3/testsuite/22_locale/codecvt/char32_t.cc
+++ b/libstdc++-v3/testsuite/22_locale/codecvt/char32_t.cc
@@ -31,7 +31,8 @@ test01()
using namespace std;
typedef codecvt<char32_t, char, mbstate_t> codecvt_c32;
locale loc_c = locale::classic();
- VERIFY(has_facet<codecvt_c32>(loc_c));
+ if (!has_facet<codecvt_c32>(loc_c))
+ return;
const codecvt_c32* const cvt = &use_facet<codecvt_c32>(loc_c);
VERIFY(!cvt->always_noconv());
diff --git a/libstdc++-v3/testsuite/22_locale/codecvt/utf8-char8_t.cc b/libstdc++-v3/testsuite/22_locale/codecvt/utf8-char8_t.cc
index 5eab05ba2..d6f663ce6 100644
--- a/libstdc++-v3/testsuite/22_locale/codecvt/utf8-char8_t.cc
+++ b/libstdc++-v3/testsuite/22_locale/codecvt/utf8-char8_t.cc
@@ -34,6 +34,8 @@ void test(const C* from)
std::mbstate_t state{};
char8_t buf[16] = { };
using test_type = std::codecvt<C, char8_t, std::mbstate_t>;
+ if (!std::has_facet<test_type>(std::locale::classic()))
+ return;
const test_type& cvt = std::use_facet<test_type>(std::locale::classic());
auto from_end = from + len;
auto from_next = from;
diff --git a/libstdc++-v3/testsuite/22_locale/codecvt/utf8.cc b/libstdc++-v3/testsuite/22_locale/codecvt/utf8.cc
index b1b3cff31..167bc0ca3 100644
--- a/libstdc++-v3/testsuite/22_locale/codecvt/utf8.cc
+++ b/libstdc++-v3/testsuite/22_locale/codecvt/utf8.cc
@@ -32,6 +32,8 @@ void test(const C* from)
std::mbstate_t state{};
char buf[16] = { };
using test_type = std::codecvt<C, char, std::mbstate_t>;
+ if (!std::has_facet<test_type>(std::locale::classic()))
+ return;
const test_type& cvt = std::use_facet<test_type>(std::locale::classic());
auto from_end = from + len;
auto from_next = from;
--
2.31.1

View File

@ -0,0 +1,42 @@
From 16694079a3f415f5e5683b0afe6978810ac259b1 Mon Sep 17 00:00:00 2001
From: David Malcolm <dmalcolm@redhat.com>
Date: Tue, 31 Aug 2021 17:01:05 -0400
Subject: [PATCH 08/17] testsuite: build plugins with -std=c++11
Various testsuite plugins fail when built within DTS with e.g.:
cc1: error: cannot load plugin ./diagnostic_plugin_test_paths.so: /builddir/build/BUILD/gcc-11.1.1-20210623/obj-x86_64-redhat-linux/x86_64-redhat-linux/libstdc++-v3/src/.libs/libstdc++.so.6: version `CXXABI_1.3.9' not found (required by ./diagnostic_plugin_test_paths.so)
These turn out to C++14's sized deletion
(see https://en.cppreference.com/w/cpp/memory/new/operator_delete):
14: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF _ZdlPvm@CXXABI_1.3.9 (4)
48: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF _ZdlPvm@CXXABI_1.3.9
14: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF operator delete(void*, unsigned long)@CXXABI_1.3.9 (4)
48: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF operator delete(void*, unsigned long)@CXXABI_1.3.9
Looks like plugin.exp is building the test plugins against the
freshly-built libstdc++, and then trying to dynamically load them
against the system libstdc++.
This patch forces the use of -std=c++11 when building these test plugins,
to sidestep the problem.
---
gcc/testsuite/lib/plugin-support.exp | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/gcc/testsuite/lib/plugin-support.exp b/gcc/testsuite/lib/plugin-support.exp
index 6d651901e..9943dbb37 100644
--- a/gcc/testsuite/lib/plugin-support.exp
+++ b/gcc/testsuite/lib/plugin-support.exp
@@ -103,7 +103,7 @@ proc plugin-test-execute { plugin_src plugin_tests } {
}
set optstr [concat $optstr "-DIN_GCC -fPIC -shared -fno-rtti -undefined dynamic_lookup"]
} else {
- set plug_cflags $PLUGINCFLAGS
+ set plug_cflags "$PLUGINCFLAGS -std=c++11"
set optstr "$includes $extra_flags -DIN_GCC -fPIC -shared -fno-rtti"
}
--
2.31.1

View File

@ -0,0 +1,36 @@
From 87e2a4f28b444f376ebe15f38e2743eb952ff355 Mon Sep 17 00:00:00 2001
From: David Malcolm <dmalcolm@redhat.com>
Date: Wed, 1 Sep 2021 11:02:53 -0400
Subject: [PATCH 09/17] Fix 22_locale/locale/cons/unicode.cc when run under DTS
Various has_facet calls return false in DTS, as DTS is using the
std::locale from the system libstdc++.
Hence this patch tweaks tests to remove the VERIFY from the result of
the call, so that they do not fail if run against an older
std::locale.
These VERIFY tests were added upstream 2015-01-16 in
https://gcc.gnu.org/git/?p=gcc.git;a=commitdiff;h=bb93f35da1612940266f5d159b6cc5a3e54fca14
---
libstdc++-v3/testsuite/22_locale/locale/cons/unicode.cc | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libstdc++-v3/testsuite/22_locale/locale/cons/unicode.cc b/libstdc++-v3/testsuite/22_locale/locale/cons/unicode.cc
index 15c621099..328145094 100644
--- a/libstdc++-v3/testsuite/22_locale/locale/cons/unicode.cc
+++ b/libstdc++-v3/testsuite/22_locale/locale/cons/unicode.cc
@@ -70,8 +70,8 @@ void test01()
VERIFY( has_facet<w_codecvt>(loc13) );
#endif
#ifdef _GLIBCXX_USE_C99_STDINT_TR1
- VERIFY( has_facet<u16_codecvt>(loc13) );
- VERIFY( has_facet<u32_codecvt>(loc13) );
+ (void)( has_facet<u16_codecvt>(loc13) );
+ (void)( has_facet<u32_codecvt>(loc13) );
#ifdef _GLIBCXX_USE_CHAR8_T
VERIFY( has_facet<u16u8_codecvt>(loc13) );
VERIFY( has_facet<u32u8_codecvt>(loc13) );
--
2.31.1

View File

@ -0,0 +1,30 @@
From d1555ffdd9b4fa6f3ceaa166bbfee0b3b9973ecf Mon Sep 17 00:00:00 2001
From: David Malcolm <dmalcolm@redhat.com>
Date: Wed, 1 Sep 2021 11:24:34 -0400
Subject: [PATCH 10/17] Don't verify exception handling in basic_filebuf::close
in DTS [PR81256]
27_io/basic_filebuf/close/81256.cc was added upstream 2018-05-14 in
PR libstdc++/81256 fix exception handling in basic_filebuf::close
https://gcc.gnu.org/git/?p=gcc.git;a=commitdiff;h=e77497ab777d3dfa19224648d658220807ab7419
When run under DTS,
std::basic_filebuf<wchar_t, std::char_traits<wchar_t> >::close
comes from the system libstdc++, and hence the test fails.
---
libstdc++-v3/testsuite/27_io/basic_filebuf/close/81256.cc | 1 -
1 file changed, 1 deletion(-)
--- a/libstdc++-v3/testsuite/27_io/basic_filebuf/close/wchar_t/81256.cc
+++ b/libstdc++-v3/testsuite/27_io/basic_filebuf/close/wchar_t/81256.cc
@@ -80,7 +80,6 @@ test01()
caught = true;
}
VERIFY( conv.exceptions_thrown == 1 );
- VERIFY( caught );
}
VERIFY( conv.exceptions_thrown == 1 );
}
--
2.31.1

View File

@ -0,0 +1,117 @@
From b6989e3a4acda2d75612f3f3847dbea4245ff536 Mon Sep 17 00:00:00 2001
From: David Malcolm <dmalcolm@redhat.com>
Date: Wed, 1 Sep 2021 15:39:45 -0400
Subject: [PATCH 11/17] Add dts.exp and use it to fix
22_locale/messages/13631.cc
This test was added upstream 2014-12-03:
"re PR libstdc++/13631 (Problems in messages)"
https://gcc.gnu.org/git/?p=gcc.git;a=commitdiff;h=d31008d7a0d53b431f176aad8dda5498de823122
as part of a fix that is present in the system libstdc++.so in
GCC 5 onwards.
When run in DTS against such a system library, this test will fail.
This patch introduces a dts.exp which detects the version of the
underlying system libstdc++.so and exposes it to tests via
-D__CXXSTDLIB_SO_VERSION__=, so that we can ifdef specific tests
away, conditionally on the base GCC.
---
.../testsuite/22_locale/messages/13631.cc | 7 +++++
libstdc++-v3/testsuite/lib/dts.exp | 31 +++++++++++++++++++
libstdc++-v3/testsuite/lib/libstdc++.exp | 6 ++++
3 files changed, 44 insertions(+)
create mode 100644 libstdc++-v3/testsuite/lib/dts.exp
diff --git a/libstdc++-v3/testsuite/22_locale/messages/13631.cc b/libstdc++-v3/testsuite/22_locale/messages/13631.cc
index b8ae3d4f1..5b20df382 100644
--- a/libstdc++-v3/testsuite/22_locale/messages/13631.cc
+++ b/libstdc++-v3/testsuite/22_locale/messages/13631.cc
@@ -50,7 +50,10 @@ void test01()
msgs_facet.close(fake_msgs);
msgs_facet.close(msgs);
+ // Fixed upstream in GCC 5
+#if __CXXSTDLIB_SO_VERSION__ >= 501000
VERIFY( translation1 == translation2 );
+#endif
}
void test02()
@@ -72,8 +75,12 @@ void test02()
std::wstring translation1 = msgs_facet.get(msgs, 0, 0, msgid);
// Without a real translation this test doesn't mean anything:
+
+ // Fixed upstream in GCC 5
+#if __CXXSTDLIB_SO_VERSION__ >= 501000
VERIFY( !translation1.empty() );
VERIFY( translation1 != msgid );
+#endif
// Opening another catalog was enough to show the problem, even a fake
// catalog.
diff --git a/libstdc++-v3/testsuite/lib/dts.exp b/libstdc++-v3/testsuite/lib/dts.exp
new file mode 100644
index 000000000..76ece66d3
--- /dev/null
+++ b/libstdc++-v3/testsuite/lib/dts.exp
@@ -0,0 +1,31 @@
+# For DTS testing, generate a number expressing the
+# system version of libstdc++.so
+#
+# Generate a version number equivalent to
+# #define GCC_VERSION (__GNUC__ * 10000 \
+# + __GNUC_MINOR__ * 100 \
+# + __GNUC_PATCHLEVEL__)
+#
+# For example, given an underlying version of gcc 4.8.5
+# this function will return 408050.
+
+proc get_dts_base_version { } {
+
+ # Invoke gcc in the PATH to get at the underlying GCC version
+ # in dotted form (e.g. "4.8.5").
+ set dotted_version [exec gcc -dumpversion]
+ verbose "dotted_version: '$dotted_version'" 2
+
+ # Extract major, minor, patchlevel
+ regexp {([0-9]+)\.([0-9]+)\.([0-9]+)} \
+ $dotted_version \
+ _ major minor patchlevel
+ verbose "major: '$major'" 2
+ verbose "minor: '$minor'" 2
+ verbose "patchlevel: '$patchlevel'" 2
+
+ set base_gcc_version [expr (($major * 10000) + ($minor * 100) + $patchlevel)]
+ verbose "base_gcc_version: '$base_gcc_version'" 2
+
+ return $base_gcc_version
+}
diff --git a/libstdc++-v3/testsuite/lib/libstdc++.exp b/libstdc++-v3/testsuite/lib/libstdc++.exp
index 7f9580db8..5e4b32f76 100644
--- a/libstdc++-v3/testsuite/lib/libstdc++.exp
+++ b/libstdc++-v3/testsuite/lib/libstdc++.exp
@@ -58,6 +58,7 @@ load_gcc_lib timeout.exp
load_gcc_lib timeout-dg.exp
load_gcc_lib wrapper.exp
load_gcc_lib target-utils.exp
+load_lib dts.exp
# Useful for debugging. Pass the name of a variable and the verbosity
# threshold (number of -v's on the command line).
@@ -323,6 +324,11 @@ proc libstdc++_init { testfile } {
set ccflags "$cxxflags -DLOCALEDIR=\".\""
set cxxflags "$cxxflags -DLOCALEDIR=\".\""
+ # For DTS testing, expose the system version of libstdc++.so as
+ # a preprocessor define.
+ set base_gcc_version [get_dts_base_version]
+ set cxxflags "$cxxflags -D__CXXSTDLIB_SO_VERSION__=$base_gcc_version"
+
# If a PCH file is available, use it. We must delay performing
# this check until $cxx and such have been initialized because we
# perform a test compilation. (Ideally, gcc --print-file-name would
--
2.31.1

View File

@ -0,0 +1,28 @@
From c3910bb945aba02e0c06dec041da9f1f148b0df1 Mon Sep 17 00:00:00 2001
From: David Malcolm <dmalcolm@redhat.com>
Date: Thu, 2 Sep 2021 11:40:42 -0400
Subject: [PATCH 12/17] dts.exp: use /usr/bin/gcc
---
libstdc++-v3/testsuite/lib/dts.exp | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libstdc++-v3/testsuite/lib/dts.exp b/libstdc++-v3/testsuite/lib/dts.exp
index 76ece66d3..dc5cf3e95 100644
--- a/libstdc++-v3/testsuite/lib/dts.exp
+++ b/libstdc++-v3/testsuite/lib/dts.exp
@@ -11,9 +11,9 @@
proc get_dts_base_version { } {
- # Invoke gcc in the PATH to get at the underlying GCC version
+ # Invoke /usr/bin/gcc to get at the underlying GCC version
# in dotted form (e.g. "4.8.5").
- set dotted_version [exec gcc -dumpversion]
+ set dotted_version [exec /usr/bin/gcc -dumpversion]
verbose "dotted_version: '$dotted_version'" 2
# Extract major, minor, patchlevel
--
2.31.1

View File

@ -0,0 +1,86 @@
From c922a56d4461c9e4ca2af07ecccb3626b99a35ae Mon Sep 17 00:00:00 2001
From: David Malcolm <dmalcolm@redhat.com>
Date: Thu, 2 Sep 2021 11:44:37 -0400
Subject: [PATCH 13/17] Rename __CXXSTDLIB_SO_VERSION__ to
__LIBSTDCXX_SO_VERSION and only use major version
---
.../testsuite/22_locale/messages/13631.cc | 4 ++--
libstdc++-v3/testsuite/lib/dts.exp | 18 +++++-------------
libstdc++-v3/testsuite/lib/libstdc++.exp | 4 ++--
3 files changed, 9 insertions(+), 17 deletions(-)
diff --git a/libstdc++-v3/testsuite/22_locale/messages/13631.cc b/libstdc++-v3/testsuite/22_locale/messages/13631.cc
index 5b20df382..f4c5eb7a2 100644
--- a/libstdc++-v3/testsuite/22_locale/messages/13631.cc
+++ b/libstdc++-v3/testsuite/22_locale/messages/13631.cc
@@ -51,7 +51,7 @@ void test01()
msgs_facet.close(msgs);
// Fixed upstream in GCC 5
-#if __CXXSTDLIB_SO_VERSION__ >= 501000
+#if __LIBSTDCXX_SO_VERSION >= 5
VERIFY( translation1 == translation2 );
#endif
}
@@ -77,7 +77,7 @@ void test02()
// Without a real translation this test doesn't mean anything:
// Fixed upstream in GCC 5
-#if __CXXSTDLIB_SO_VERSION__ >= 501000
+#if __LIBSTDCXX_SO_VERSION >= 5
VERIFY( !translation1.empty() );
VERIFY( translation1 != msgid );
#endif
diff --git a/libstdc++-v3/testsuite/lib/dts.exp b/libstdc++-v3/testsuite/lib/dts.exp
index dc5cf3e95..60f94bd6d 100644
--- a/libstdc++-v3/testsuite/lib/dts.exp
+++ b/libstdc++-v3/testsuite/lib/dts.exp
@@ -1,15 +1,10 @@
-# For DTS testing, generate a number expressing the
-# system version of libstdc++.so
-#
-# Generate a version number equivalent to
-# #define GCC_VERSION (__GNUC__ * 10000 \
-# + __GNUC_MINOR__ * 100 \
-# + __GNUC_PATCHLEVEL__)
+# For DTS testing, get the major version of the
+# system libstdc++.so
#
# For example, given an underlying version of gcc 4.8.5
-# this function will return 408050.
+# this function will return 4
-proc get_dts_base_version { } {
+proc get_dts_base_major_version { } {
# Invoke /usr/bin/gcc to get at the underlying GCC version
# in dotted form (e.g. "4.8.5").
@@ -24,8 +19,5 @@ proc get_dts_base_version { } {
verbose "minor: '$minor'" 2
verbose "patchlevel: '$patchlevel'" 2
- set base_gcc_version [expr (($major * 10000) + ($minor * 100) + $patchlevel)]
- verbose "base_gcc_version: '$base_gcc_version'" 2
-
- return $base_gcc_version
+ return $major
}
diff --git a/libstdc++-v3/testsuite/lib/libstdc++.exp b/libstdc++-v3/testsuite/lib/libstdc++.exp
index 5e4b32f76..297485b19 100644
--- a/libstdc++-v3/testsuite/lib/libstdc++.exp
+++ b/libstdc++-v3/testsuite/lib/libstdc++.exp
@@ -326,8 +326,8 @@ proc libstdc++_init { testfile } {
# For DTS testing, expose the system version of libstdc++.so as
# a preprocessor define.
- set base_gcc_version [get_dts_base_version]
- set cxxflags "$cxxflags -D__CXXSTDLIB_SO_VERSION__=$base_gcc_version"
+ set base_gcc_version [get_dts_base_major_version]
+ set cxxflags "$cxxflags -D__LIBSTDCXX_SO_VERSION=$base_gcc_version"
# If a PCH file is available, use it. We must delay performing
# this check until $cxx and such have been initialized because we
--
2.31.1

View File

@ -0,0 +1,84 @@
From fa770c25013df537c41d0929c4202b0a774a6a75 Mon Sep 17 00:00:00 2001
From: David Malcolm <dmalcolm@redhat.com>
Date: Thu, 2 Sep 2021 15:27:59 -0400
Subject: [PATCH 14/17] Conditionalize tests for PR libstdc++/98466 on
__LIBSTDCXX_SO_VERSION >= 11
The tests:
23_containers/deque/debug/98466.cc
23_containers/unordered_map/debug/98466.cc
were added upstream 2021-01-01 as part of:
libstdc++: Implement N3644 for _GLIBCXX_DEBUG iterators
https://gcc.gnu.org/git/?p=gcc.git;a=commitdiff;h=05a30af3f237984b4dcf1dbbc17fdac583c46506
but fail when run in DTS against a system libstdc++.so
from an older GCC.
In particular, they use the older implementation of _M_can_compare from
the dynamic library:
$ eu-readelf -s 98466.exe | grep can_compare | c++filt
11: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF __gnu_debug::_Safe_iterator_base::_M_can_compare(__gnu_debug::_Safe_iterator_base const&) const@GLIBCXX_3.4 (4)
157: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF __gnu_debug::_Safe_iterator_base::_M_can_compare(__gnu_debug::_Safe_iterator_base const&) const@@GLIBCXX_3.4
---
.../testsuite/23_containers/deque/debug/98466.cc | 6 ++++++
.../23_containers/unordered_map/debug/98466.cc | 12 ++++++++++++
2 files changed, 18 insertions(+)
diff --git a/libstdc++-v3/testsuite/23_containers/deque/debug/98466.cc b/libstdc++-v3/testsuite/23_containers/deque/debug/98466.cc
index c2d793374..e92d719bf 100644
--- a/libstdc++-v3/testsuite/23_containers/deque/debug/98466.cc
+++ b/libstdc++-v3/testsuite/23_containers/deque/debug/98466.cc
@@ -26,11 +26,17 @@ void test01()
{
typedef typename __gnu_debug::deque<int>::iterator It;
It it = It();
+ // Fixed upstream in GCC 11
+#if __LIBSTDCXX_SO_VERSION >= 11
VERIFY( it == it );
+#endif
typedef typename __gnu_debug::deque<int>::const_iterator Cit;
Cit cit = Cit();
+ // Fixed upstream in GCC 11
+#if __LIBSTDCXX_SO_VERSION >= 11
VERIFY( cit == cit );
+#endif
}
int main()
diff --git a/libstdc++-v3/testsuite/23_containers/unordered_map/debug/98466.cc b/libstdc++-v3/testsuite/23_containers/unordered_map/debug/98466.cc
index cc22b9ff8..3a1b6cd32 100644
--- a/libstdc++-v3/testsuite/23_containers/unordered_map/debug/98466.cc
+++ b/libstdc++-v3/testsuite/23_containers/unordered_map/debug/98466.cc
@@ -25,16 +25,28 @@
void test01()
{
__gnu_debug::unordered_map<int, int>::iterator it{};
+ // Fixed upstream in GCC 11
+#if __LIBSTDCXX_SO_VERSION >= 11
VERIFY( it == it );
+#endif
__gnu_debug::unordered_map<int, int>::const_iterator cit{};
+ // Fixed upstream in GCC 11
+#if __LIBSTDCXX_SO_VERSION >= 11
VERIFY( cit == cit );
+#endif
__gnu_debug::unordered_map<int, int>::local_iterator lit{};
+ // Fixed upstream in GCC 11
+#if __LIBSTDCXX_SO_VERSION >= 11
VERIFY( lit == lit );
+#endif
__gnu_debug::unordered_map<int, int>::const_local_iterator clit{};
+ // Fixed upstream in GCC 11
+#if __LIBSTDCXX_SO_VERSION >= 11
VERIFY( clit == clit );
+#endif
}
int main()
--
2.31.1

View File

@ -0,0 +1,41 @@
From 992665eab6c48d6a4819f42509346d24b277485d Mon Sep 17 00:00:00 2001
From: David Malcolm <dmalcolm@redhat.com>
Date: Thu, 2 Sep 2021 16:17:29 -0400
Subject: [PATCH 15/17] Conditionalize test for PR libstdc++/87135 on
__LIBSTDCXX_SO_VERSION >= 9
This VERIFY was added upstream 2018-09-18 as part of:
re PR libstdc++/87135 ([C++17] unordered containers violate iterator validity requirements)
https://gcc.gnu.org/git/?p=gcc.git;a=commitdiff;h=a521e62615e439aea7502a52fd0f8a21eaa6304f
but fails when run in DTS against a system libstdc++.so from an older GCC.
In particular, rehash from the header is using
std::__detail::_Prime_rehash_policy::_M_next_bkt
from the system .so:
12: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF std::__detail::_Prime_rehash_policy::_M_next_bkt(unsigned long) const@GLIBCXX_3.4.18 (5)
225: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF std::__detail::_Prime_rehash_policy::_M_next_bkt(unsigned long) const@@GLIBCXX_3.4.18
---
.../23_containers/unordered_map/modifiers/reserve.cc | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/libstdc++-v3/testsuite/23_containers/unordered_map/modifiers/reserve.cc b/libstdc++-v3/testsuite/23_containers/unordered_map/modifiers/reserve.cc
index 58c8924b9..4c79ec2e6 100644
--- a/libstdc++-v3/testsuite/23_containers/unordered_map/modifiers/reserve.cc
+++ b/libstdc++-v3/testsuite/23_containers/unordered_map/modifiers/reserve.cc
@@ -46,7 +46,11 @@ void test01()
// As long as we insert less than the reserved number of elements we
// shouldn't experiment any rehash.
+
+ // Fixed upstream in GCC 9
+#if __LIBSTDCXX_SO_VERSION >= 9
VERIFY( m.bucket_count() == bkts );
+#endif
VERIFY( m.load_factor() <= m.max_load_factor() );
}
--
2.31.1

View File

@ -0,0 +1,102 @@
From db8f53df0be1daeda3159c1413549ff40696c710 Mon Sep 17 00:00:00 2001
From: David Malcolm <dmalcolm@redhat.com>
Date: Thu, 2 Sep 2021 17:02:33 -0400
Subject: [PATCH 16/17] Conditionalize test for hashtable bucket sizes on
__LIBSTDCXX_SO_VERSION >= 11
These tests were added upstream 2020-01-20 as part of:
libstdc++: Do not over-size hashtable buckets on range insertion
https://gcc.gnu.org/git/?p=gcc.git;a=commitdiff;h=6dcf042368012e2d7ce1626ee5d378bf3ad0ccfc
but fail when run in DTS against a system libstdc++.so from an older GCC.
In particular, _M_insert_unique_node from the header is using the older
implementation of
std::__detail::_Prime_rehash_policy::_M_need_rehash
from the dynamic library.
23: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF std::__detail::_Prime_rehash_policy::_M_need_rehash(unsigned long, unsigned long, unsigned long) const@GLIBCXX_3.4.18 (5)
412: 0000000000000000 0 FUNC GLOBAL DEFAULT UNDEF std::__detail::_Prime_rehash_policy::_M_need_rehash(unsigned long, unsigned long, unsigned long) const@@GLIBCXX_3.4.18
---
.../23_containers/unordered_set/cons/bucket_hint.cc | 10 ++++++++++
.../23_containers/unordered_set/modifiers/insert.cc | 9 +++++++++
2 files changed, 19 insertions(+)
diff --git a/libstdc++-v3/testsuite/23_containers/unordered_set/cons/bucket_hint.cc b/libstdc++-v3/testsuite/23_containers/unordered_set/cons/bucket_hint.cc
index a3b014a3a..af231e54e 100644
--- a/libstdc++-v3/testsuite/23_containers/unordered_set/cons/bucket_hint.cc
+++ b/libstdc++-v3/testsuite/23_containers/unordered_set/cons/bucket_hint.cc
@@ -29,7 +29,11 @@ void test01()
a.reserve(2);
std::unordered_set<int> b({ 0, 1, 0, 1, 0, 1, 0, 1 }, a.bucket_count());
+
+ // Fixed upstream in GCC 11
+#if __LIBSTDCXX_SO_VERSION >= 11
VERIFY( b.bucket_count() == a.bucket_count() );
+#endif
}
void test02()
@@ -40,7 +44,10 @@ void test02()
std::vector<int> v { 0, 1, 0, 1, 0, 1, 0, 1, 0, 1 };
std::unordered_set<int> b(v.begin(), v.end(), a.bucket_count());
+ // Fixed upstream in GCC 11
+#if __LIBSTDCXX_SO_VERSION >= 11
VERIFY( b.bucket_count() == a.bucket_count() );
+#endif
}
void test03()
@@ -51,7 +58,10 @@ void test03()
std::forward_list<int> fl { 0, 1, 0, 1, 0, 1, 0, 1, 0, 1 };
std::unordered_set<int> b(fl.begin(), fl.end(), a.bucket_count());
+ // Fixed upstream in GCC 11
+#if __LIBSTDCXX_SO_VERSION >= 11
VERIFY( b.bucket_count() == a.bucket_count() );
+#endif
}
int main()
diff --git a/libstdc++-v3/testsuite/23_containers/unordered_set/modifiers/insert.cc b/libstdc++-v3/testsuite/23_containers/unordered_set/modifiers/insert.cc
index 015c2f872..aae8298ae 100644
--- a/libstdc++-v3/testsuite/23_containers/unordered_set/modifiers/insert.cc
+++ b/libstdc++-v3/testsuite/23_containers/unordered_set/modifiers/insert.cc
@@ -30,7 +30,10 @@ void test01()
auto bkt_count = a.bucket_count();
a.insert({ 0, 1, 0, 1, 0, 1, 0, 1 });
+ // Fixed upstream in GCC 11
+#if __LIBSTDCXX_SO_VERSION >= 11
VERIFY( a.bucket_count() == bkt_count );
+#endif
}
void test02()
@@ -42,7 +45,10 @@ void test02()
auto bkt_count = a.bucket_count();
a.insert(v.begin(), v.end());
+ // Fixed upstream in GCC 11
+#if __LIBSTDCXX_SO_VERSION >= 11
VERIFY( a.bucket_count() == bkt_count );
+#endif
}
void test03()
@@ -54,7 +60,10 @@ void test03()
auto bkt_count = a.bucket_count();
a.insert(fl.begin(), fl.end());
+ // Fixed upstream in GCC 11
+#if __LIBSTDCXX_SO_VERSION >= 11
VERIFY( a.bucket_count() == bkt_count );
+#endif
}
int main()
--
2.31.1

View File

@ -0,0 +1,33 @@
From 102adc148612b50a4e3f12a88871ddc47bfb469e Mon Sep 17 00:00:00 2001
From: David Malcolm <dmalcolm@redhat.com>
Date: Thu, 2 Sep 2021 17:17:50 -0400
Subject: [PATCH 17/17] Conditionalize test for PR libstdc++/71181 on
__LIBSTDCXX_SO_VERSION >= 7
This test was added upstream 2016-06-20 as part of:
re PR libstdc++/71181 (Reserving in unordered_map doesn't reserve enough)
https://gcc.gnu.org/git/?p=gcc.git;a=commitdiff;h=29dbb034cb3199167a9d0aaed040733c72326eed
but fails when run in DTS against a system libstdc++.so from an older GCC.
---
.../testsuite/23_containers/unordered_set/hash_policy/71181.cc | 3 +++
1 file changed, 3 insertions(+)
diff --git a/libstdc++-v3/testsuite/23_containers/unordered_set/hash_policy/71181.cc b/libstdc++-v3/testsuite/23_containers/unordered_set/hash_policy/71181.cc
index 324d735b8..b2dcd0137 100644
--- a/libstdc++-v3/testsuite/23_containers/unordered_set/hash_policy/71181.cc
+++ b/libstdc++-v3/testsuite/23_containers/unordered_set/hash_policy/71181.cc
@@ -39,7 +39,10 @@ template<typename _USet>
us.insert(nb_insert);
+ // Fixed upstream in GCC 7
+#if __LIBSTDCXX_SO_VERSION >= 7
VERIFY( us.bucket_count() == bkts );
+#endif
}
}
--
2.31.1

View File

@ -0,0 +1,32 @@
commit 14c7757e9b751781360737f53b71f851fc356d3d
Author: Jeff Law <jeffreyalaw@gmail.com>
Date: Fri Oct 29 11:30:15 2021 -0400
Avoid overly-greedy match in dejagnu regexp.
Occasionally I've been seeing failures with the multi-line diagnostics. It's never been clear what's causing the spurious failures, though I have long suspected a greedy regexp match.
It happened again yesterday with a local change that in no way should affect diagnostics, so I finally went searching and found that sure enough the multi-line diagnostics had a ".*" in their regexp. According to the comments, the .* is primarily to catch any dg directives that may appear -- ie it should eat to EOL, but not multiple lines. But a .* can indeed match a newline and cause it to eat multiple lines.
The fix is simple. [^\r\n]* will eat to EOL, but not further.
Regression tested on x86_64 and on our internal target.
gcc/testsuite
* lib/multiline.exp (_build_multiline_regex): Use a better
regexp than .* to match up to EOL.
diff --git a/gcc/testsuite/lib/multiline.exp b/gcc/testsuite/lib/multiline.exp
index 0e151b6d222..86387f8209b 100644
--- a/gcc/testsuite/lib/multiline.exp
+++ b/gcc/testsuite/lib/multiline.exp
@@ -331,7 +331,7 @@ proc _build_multiline_regex { multiline index } {
# Support arbitrary followup text on each non-empty line,
# to deal with comments containing containing DejaGnu
# directives.
- append rexp ".*"
+ append rexp "\[^\\n\\r\]*"
}
}
append rexp "\n"

View File

@ -0,0 +1,83 @@
diff --git a/gcc/testsuite/c-c++-common/goacc/kernels-decompose-ice-1.c b/gcc/testsuite/c-c++-common/goacc/kernels-decompose-ice-1.c
index 8c3884bdc00..141f089d4f8 100644
--- a/gcc/testsuite/c-c++-common/goacc/kernels-decompose-ice-1.c
+++ b/gcc/testsuite/c-c++-common/goacc/kernels-decompose-ice-1.c
@@ -1,7 +1,7 @@
/* Test OpenACC 'kernels' construct decomposition. */
/* { dg-additional-options "-fopt-info-omp-all" } */
-/* { dg-additional-options "-fchecking --param=openacc-kernels=decompose" } */
+/* { dg-additional-options "-fno-report-bug -fchecking --param=openacc-kernels=decompose" } */
/* { dg-ice "TODO" }
{ dg-prune-output "during GIMPLE pass: omplower" } */
diff --git a/gcc/testsuite/c-c++-common/goacc/kernels-decompose-ice-2.c b/gcc/testsuite/c-c++-common/goacc/kernels-decompose-ice-2.c
index 8bf60a9a509..a07e0a42116 100644
--- a/gcc/testsuite/c-c++-common/goacc/kernels-decompose-ice-2.c
+++ b/gcc/testsuite/c-c++-common/goacc/kernels-decompose-ice-2.c
@@ -1,6 +1,6 @@
/* Test OpenACC 'kernels' construct decomposition. */
-/* { dg-additional-options "-fchecking --param=openacc-kernels=decompose" } */
+/* { dg-additional-options "-fno-report-bug -fchecking --param=openacc-kernels=decompose" } */
/* { dg-ice "TODO" }
{ dg-prune-output "during GIMPLE pass: omplower" } */
diff --git a/gcc/testsuite/g++.dg/cpp0x/constexpr-52830.C b/gcc/testsuite/g++.dg/cpp0x/constexpr-52830.C
index eae0d8c377b..ed37f95e865 100644
--- a/gcc/testsuite/g++.dg/cpp0x/constexpr-52830.C
+++ b/gcc/testsuite/g++.dg/cpp0x/constexpr-52830.C
@@ -1,6 +1,6 @@
// PR c++/52830
// { dg-do compile { target c++11 } }
-// { dg-additional-options "-fchecking" }
+// { dg-additional-options "-fchecking -fno-report-bug" }
// { dg-ice "comptypes" }
template<bool b> struct eif { typedef void type; };
diff --git a/gcc/testsuite/g++.dg/cpp0x/vt-88982.C b/gcc/testsuite/g++.dg/cpp0x/vt-88982.C
index 7a868233d73..45307a63483 100644
--- a/gcc/testsuite/g++.dg/cpp0x/vt-88982.C
+++ b/gcc/testsuite/g++.dg/cpp0x/vt-88982.C
@@ -1,6 +1,6 @@
// PR c++/88982
// { dg-do compile { target c++11 } }
-// { dg-additional-options "-fchecking" }
+// { dg-additional-options "-fchecking -fno-report-bug" }
// { dg-ice "tsubst_pack_expansion" }
template<typename...Ts> struct A {
diff --git a/gcc/testsuite/g++.dg/cpp1y/auto-fn61.C b/gcc/testsuite/g++.dg/cpp1y/auto-fn61.C
index bed5ea0cfc0..cb235bf891a 100644
--- a/gcc/testsuite/g++.dg/cpp1y/auto-fn61.C
+++ b/gcc/testsuite/g++.dg/cpp1y/auto-fn61.C
@@ -1,6 +1,6 @@
// PR c++/88003
// { dg-do compile { target c++14 } }
-// { dg-additional-options "-fchecking" }
+// { dg-additional-options "-fchecking -fno-report-bug" }
// { dg-ice "poplevel_class" }
auto test() {
diff --git a/gcc/testsuite/g++.dg/cpp1z/constexpr-lambda26.C b/gcc/testsuite/g++.dg/cpp1z/constexpr-lambda26.C
index 0cdb400d21c..8f17cca31dd 100644
--- a/gcc/testsuite/g++.dg/cpp1z/constexpr-lambda26.C
+++ b/gcc/testsuite/g++.dg/cpp1z/constexpr-lambda26.C
@@ -1,6 +1,6 @@
// PR c++/87765
// { dg-do compile { target c++17 } }
-// { dg-additional-options "-fchecking" }
+// { dg-additional-options "-fchecking -fno-report-bug" }
// { dg-ice "cxx_eval_constant_expression" }
template <int N>
diff --git a/libgomp/testsuite/libgomp.oacc-c-c++-common/declare-vla-kernels-decompose-ice-1.c b/libgomp/testsuite/libgomp.oacc-c-c++-common/declare-vla-kernels-decompose-ice-1.c
index 0777b612b63..53abbe2db83 100644
--- a/libgomp/testsuite/libgomp.oacc-c-c++-common/declare-vla-kernels-decompose-ice-1.c
+++ b/libgomp/testsuite/libgomp.oacc-c-c++-common/declare-vla-kernels-decompose-ice-1.c
@@ -1,4 +1,4 @@
-/* { dg-additional-options "--param=openacc-kernels=decompose" } */
+/* { dg-additional-options "--param=openacc-kernels=decompose -fno-report-bug" } */
/* Hopefully, this is the same issue as '../../../gcc/testsuite/c-c++-common/goacc/kernels-decompose-ice-1.c'.
{ dg-ice "TODO" }
TODO { dg-prune-output "during GIMPLE pass: omplower" }

View File

@ -0,0 +1,57 @@
libgcc: Honor LDFLAGS_FOR_TARGET when linking libgcc_s
When building gcc with some specific LDFLAGS_FOR_TARGET, e.g.
LDFLAGS_FOR_TARGET=-Wl,-z,relro,-z,now
those flags propagate info linking of target shared libraries,
e.g. lib{ubsan,tsan,stdc++,quadmath,objc,lsan,itm,gphobos,gdruntime,gomp,go,gfortran,atomic,asan}.so.*
but there is one important exception, libgcc_s.so.* linking ignores it.
The following patch fixes that.
Bootstrapped/regtested on x86_64-linux with LDFLAGS_FOR_TARGET=-Wl,-z,relro,-z,now
and verified that libgcc_s.so.* is BIND_NOW when it previously wasn't, and
without any LDFLAGS_FOR_TARGET on x86_64-linux and i686-linux.
There on x86_64-linux I've verified that the libgcc_s.so.1 linking command
line for -m64 is identical except for whitespace to one without the patch,
and for -m32 multilib $(LDFLAGS) actually do supply there an extra -m32
that also repeats later in the @multilib_flags@, which should be harmless.
2021-08-04 Jakub Jelinek <jakub@redhat.com>
* config/t-slibgcc (SHLIB_LINK): Add $(LDFLAGS).
* config/t-slibgcc-darwin (SHLIB_LINK): Likewise.
* config/t-slibgcc-vms (SHLIB_LINK): Likewise.
--- libgcc/config/t-slibgcc
+++ libgcc/config/t-slibgcc
@@ -32,7 +32,7 @@ SHLIB_INSTALL_SOLINK = $(LN_S) $(SHLIB_SONAME) \
$(DESTDIR)$(slibdir)$(SHLIB_SLIBDIR_QUAL)/$(SHLIB_SOLINK)
SHLIB_LINK = $(CC) $(LIBGCC2_CFLAGS) -shared -nodefaultlibs \
- $(SHLIB_LDFLAGS) \
+ $(SHLIB_LDFLAGS) $(LDFLAGS) \
-o $(SHLIB_DIR)/$(SHLIB_SONAME).tmp @multilib_flags@ \
$(SHLIB_OBJS) $(SHLIB_LC) && \
rm -f $(SHLIB_DIR)/$(SHLIB_SOLINK) && \
--- libgcc/config/t-slibgcc-darwin
+++ libgcc/config/t-slibgcc-darwin
@@ -15,7 +15,7 @@ SHLIB_LC = -lc
# Note that this version is used for the loader, not the linker; the linker
# uses the stub versions named by the versioned members of $(INSTALL_FILES).
-SHLIB_LINK = $(CC) $(LIBGCC2_CFLAGS) -dynamiclib -nodefaultlibs \
+SHLIB_LINK = $(CC) $(LIBGCC2_CFLAGS) $(LDFLAGS) -dynamiclib -nodefaultlibs \
-install_name @shlib_slibdir@/$(SHLIB_INSTALL_NAME) \
-single_module -o $(SHLIB_DIR)/$(SHLIB_SONAME) \
-Wl,-exported_symbols_list,$(SHLIB_MAP) \
--- libgcc/config/t-slibgcc-vms
+++ libgcc/config/t-slibgcc-vms
@@ -22,7 +22,7 @@ SHLIB_LINK = \
objdump --syms $(SHLIB_OBJS) | \
$(SHLIB_SYMVEC) >> SYMVEC_$$$$.opt ; \
echo "case_sensitive=NO" >> SYMVEC_$$$$.opt; \
- $(CC) $(LIBGCC2_CFLAGS) -nodefaultlibs \
+ $(CC) $(LIBGCC2_CFLAGS) $(LDFLAGS) -nodefaultlibs \
-shared --for-linker=/noinform -o $(SHLIB_NAME) $(SHLIB_OBJS) \
--for-linker=SYMVEC_$$$$.opt \
--for-linker=gsmatch=equal,$(shlib_version)

View File

@ -165,9 +165,9 @@
__at_thread_exit(__at_thread_exit_elt*);
--- libstdc++-v3/src/c++11/system_error.cc.libstdc++-compat~ 2021-05-12 09:34:49.000000000 -0400
+++ libstdc++-v3/src/c++11/system_error.cc 2021-05-17 12:05:15.766954365 -0400
@@ -346,7 +346,9 @@ _GLIBCXX_BEGIN_NAMESPACE_VERSION
@@ -372,7 +372,9 @@ _GLIBCXX_BEGIN_NAMESPACE_VERSION
const error_category&
_V2::generic_category() noexcept { return generic_category_instance; }
_V2::generic_category() noexcept { return generic_category_instance.obj; }
+#ifndef _GLIBCXX_NONSHARED_CXX11
system_error::~system_error() noexcept = default;
@ -175,7 +175,7 @@
error_condition
error_category::default_error_condition(int __i) const noexcept
@@ -361,9 +363,11 @@ _GLIBCXX_BEGIN_NAMESPACE_VERSION
@@ -387,9 +389,11 @@ _GLIBCXX_BEGIN_NAMESPACE_VERSION
error_category::equivalent(const error_code& __code, int __i) const noexcept
{ return *this == __code.category() && __code.value() == __i; }
@ -13463,8 +13463,8 @@
lt_dlunknown=0; lt_dlno_uscore=1; lt_dlneed_uscore=2
lt_status=$lt_dlunknown
cat > conftest.$ac_ext <<_LT_EOF
-#line 12133 "configure"
+#line 12135 "configure"
-#line 12179 "configure"
+#line 12181 "configure"
#include "confdefs.h"
#if HAVE_DLFCN_H
@ -13472,8 +13472,8 @@
lt_dlunknown=0; lt_dlno_uscore=1; lt_dlneed_uscore=2
lt_status=$lt_dlunknown
cat > conftest.$ac_ext <<_LT_EOF
-#line 12239 "configure"
+#line 12241 "configure"
-#line 12285 "configure"
+#line 12287 "configure"
#include "confdefs.h"
#if HAVE_DLFCN_H
@ -13481,8 +13481,8 @@
# Fake what AC_TRY_COMPILE does.
cat > conftest.$ac_ext << EOF
-#line 15921 "configure"
+#line 15923 "configure"
-#line 15967 "configure"
+#line 15969 "configure"
int main()
{
typedef bool atomic_type;
@ -13490,8 +13490,8 @@
rm -f conftest*
cat > conftest.$ac_ext << EOF
-#line 15956 "configure"
+#line 15958 "configure"
-#line 16002 "configure"
+#line 16004 "configure"
int main()
{
typedef short atomic_type;
@ -13499,8 +13499,8 @@
rm -f conftest*
cat > conftest.$ac_ext << EOF
-#line 15991 "configure"
+#line 15993 "configure"
-#line 16037 "configure"
+#line 16039 "configure"
int main()
{
// NB: _Atomic_word not necessarily int.
@ -13508,8 +13508,8 @@
rm -f conftest*
cat > conftest.$ac_ext << EOF
-#line 16027 "configure"
+#line 16029 "configure"
-#line 16073 "configure"
+#line 16075 "configure"
int main()
{
typedef long long atomic_type;
@ -13517,8 +13517,8 @@
# unnecessary for this test.
cat > conftest.$ac_ext << EOF
-#line 16180 "configure"
+#line 16182 "configure"
-#line 16226 "configure"
+#line 16228 "configure"
int main()
{
_Decimal32 d1;
@ -13526,8 +13526,8 @@
# unnecessary for this test.
cat > conftest.$ac_ext << EOF
-#line 16222 "configure"
+#line 16224 "configure"
-#line 16268 "configure"
+#line 16270 "configure"
template<typename T1, typename T2>
struct same
{ typedef T2 type; };
@ -13535,8 +13535,8 @@
rm -f conftest*
cat > conftest.$ac_ext << EOF
-#line 16256 "configure"
+#line 16258 "configure"
-#line 16302 "configure"
+#line 16304 "configure"
template<typename T1, typename T2>
struct same
{ typedef T2 type; };

View File

@ -0,0 +1,457 @@
From 54ccc52ba3f842cd94718967465a6015a752ca47 Mon Sep 17 00:00:00 2001
From: "Cui,Lili" <lili.cui@intel.com>
Date: Thu, 4 Nov 2021 10:38:56 +0800
Subject: [PATCH] x86: Update -mtune=alderlake
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Update mtune for alderlake, Alder Lake Intel Hybrid Technology will not support
Intel® AVX-512. ISA features such as Intel® AVX, AVX-VNNI, Intel® AVX2, and
UMONITOR/UMWAIT/TPAUSE are supported.
gcc/ChangeLog
* config/i386/i386-options.c (m_CORE_AVX2): Remove Alderlake
from m_CORE_AVX2.
(processor_cost_table): Use alderlake_cost for Alderlake.
* config/i386/i386.c (ix86_sched_init_global): Handle Alderlake.
* config/i386/x86-tune-costs.h (struct processor_costs): Add alderlake
cost.
* config/i386/x86-tune-sched.c (ix86_issue_rate): Change Alderlake
issue rate to 4.
(ix86_adjust_cost): Handle Alderlake.
* config/i386/x86-tune.def (X86_TUNE_SCHEDULE): Enable for Alderlake.
(X86_TUNE_PARTIAL_REG_DEPENDENCY): Likewise.
(X86_TUNE_SSE_PARTIAL_REG_DEPENDENCY): Likewise.
(X86_TUNE_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY): Likewise.
(X86_TUNE_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY): Likewise.
(X86_TUNE_MEMORY_MISMATCH_STALL): Likewise.
(X86_TUNE_USE_LEAVE): Likewise.
(X86_TUNE_PUSH_MEMORY): Likewise.
(X86_TUNE_USE_INCDEC): Likewise.
(X86_TUNE_INTEGER_DFMODE_MOVES): Likewise.
(X86_TUNE_MISALIGNED_MOVE_STRING_PRO_EPILOGUES): Likewise.
(X86_TUNE_USE_SAHF): Likewise.
(X86_TUNE_USE_BT): Likewise.
(X86_TUNE_AVOID_FALSE_DEP_FOR_BMI): Likewise.
(X86_TUNE_ONE_IF_CONV_INSN): Likewise.
(X86_TUNE_AVOID_MFENCE): Likewise.
(X86_TUNE_USE_SIMODE_FIOP): Likewise.
(X86_TUNE_EXT_80387_CONSTANTS): Likewise.
(X86_TUNE_SSE_UNALIGNED_LOAD_OPTIMAL): Likewise.
(X86_TUNE_SSE_UNALIGNED_STORE_OPTIMAL): Likewise.
(X86_TUNE_SSE_TYPELESS_STORES): Likewise.
(X86_TUNE_SSE_LOAD0_BY_PXOR): Likewise.
(X86_TUNE_AVOID_4BYTE_PREFIXES): Likewise.
(X86_TUNE_USE_GATHER): Disable for Alderlake.
---
gcc/config/i386/i386-options.c | 4 +-
gcc/config/i386/i386.c | 1 +
gcc/config/i386/x86-tune-costs.h | 120 +++++++++++++++++++++++++++++++
gcc/config/i386/x86-tune-sched.c | 2 +
gcc/config/i386/x86-tune.def | 58 +++++++--------
5 files changed, 155 insertions(+), 30 deletions(-)
diff --git a/gcc/config/i386/i386-options.c b/gcc/config/i386/i386-options.c
index e7a3bd4aaea..a8cc0664f11 100644
--- a/gcc/config/i386/i386-options.c
+++ b/gcc/config/i386/i386-options.c
@@ -131,7 +131,7 @@ along with GCC; see the file COPYING3. If not see
| m_ICELAKE_CLIENT | m_ICELAKE_SERVER | m_CASCADELAKE \
| m_TIGERLAKE | m_COOPERLAKE | m_SAPPHIRERAPIDS \
| m_ROCKETLAKE)
-#define m_CORE_AVX2 (m_HASWELL | m_SKYLAKE | m_ALDERLAKE | m_CORE_AVX512)
+#define m_CORE_AVX2 (m_HASWELL | m_SKYLAKE | m_CORE_AVX512)
#define m_CORE_ALL (m_CORE2 | m_NEHALEM | m_SANDYBRIDGE | m_CORE_AVX2)
#define m_GOLDMONT (HOST_WIDE_INT_1U<<PROCESSOR_GOLDMONT)
#define m_GOLDMONT_PLUS (HOST_WIDE_INT_1U<<PROCESSOR_GOLDMONT_PLUS)
@@ -736,7 +736,7 @@ static const struct processor_costs *processor_cost_table[] =
&icelake_cost,
&skylake_cost,
&icelake_cost,
- &icelake_cost,
+ &alderlake_cost,
&icelake_cost,
&intel_cost,
&geode_cost,
diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
index e94efdf39fb..73c4d5115bb 100644
--- a/gcc/config/i386/i386.c
+++ b/gcc/config/i386/i386.c
@@ -17014,6 +17014,7 @@ ix86_sched_init_global (FILE *, int, int)
case PROCESSOR_SANDYBRIDGE:
case PROCESSOR_HASWELL:
case PROCESSOR_TREMONT:
+ case PROCESSOR_ALDERLAKE:
case PROCESSOR_GENERIC:
/* Do not perform multipass scheduling for pre-reload schedule
to save compile time. */
diff --git a/gcc/config/i386/x86-tune-costs.h b/gcc/config/i386/x86-tune-costs.h
index 93644be9cb3..dd5563d2e64 100644
--- a/gcc/config/i386/x86-tune-costs.h
+++ b/gcc/config/i386/x86-tune-costs.h
@@ -2070,6 +2070,126 @@ struct processor_costs icelake_cost = {
"16", /* Func alignment. */
};
+/* alderlake_cost should produce code tuned for alderlake family of CPUs. */
+static stringop_algs alderlake_memcpy[2] = {
+ {libcall,
+ {{256, rep_prefix_1_byte, true},
+ {256, loop, false},
+ {-1, libcall, false}}},
+ {libcall,
+ {{256, rep_prefix_1_byte, true},
+ {256, loop, false},
+ {-1, libcall, false}}}};
+static stringop_algs alderlake_memset[2] = {
+ {libcall,
+ {{256, rep_prefix_1_byte, true},
+ {256, loop, false},
+ {-1, libcall, false}}},
+ {libcall,
+ {{256, rep_prefix_1_byte, true},
+ {256, loop, false},
+ {-1, libcall, false}}}};
+static const
+struct processor_costs alderlake_cost = {
+ {
+ /* Start of register allocator costs. integer->integer move cost is 2. */
+ 6, /* cost for loading QImode using movzbl */
+ {6, 6, 6}, /* cost of loading integer registers
+ in QImode, HImode and SImode.
+ Relative to reg-reg move (2). */
+ {6, 6, 6}, /* cost of storing integer registers */
+ 4, /* cost of reg,reg fld/fst */
+ {6, 6, 12}, /* cost of loading fp registers
+ in SFmode, DFmode and XFmode */
+ {6, 6, 12}, /* cost of storing fp registers
+ in SFmode, DFmode and XFmode */
+ 2, /* cost of moving MMX register */
+ {6, 6}, /* cost of loading MMX registers
+ in SImode and DImode */
+ {6, 6}, /* cost of storing MMX registers
+ in SImode and DImode */
+ 2, 3, 4, /* cost of moving XMM,YMM,ZMM register */
+ {6, 6, 6, 10, 15}, /* cost of loading SSE registers
+ in 32,64,128,256 and 512-bit */
+ {6, 6, 6, 10, 15}, /* cost of storing SSE registers
+ in 32,64,128,256 and 512-bit */
+ 6, 6, /* SSE->integer and integer->SSE moves */
+ 6, 6, /* mask->integer and integer->mask moves */
+ {6, 6, 6}, /* cost of loading mask register
+ in QImode, HImode, SImode. */
+ {6, 6, 6}, /* cost if storing mask register
+ in QImode, HImode, SImode. */
+ 2, /* cost of moving mask register. */
+ /* End of register allocator costs. */
+ },
+
+ COSTS_N_INSNS (1), /* cost of an add instruction */
+ COSTS_N_INSNS (1) + 1, /* cost of a lea instruction */
+ COSTS_N_INSNS (1), /* variable shift costs */
+ COSTS_N_INSNS (1), /* constant shift costs */
+ {COSTS_N_INSNS (3), /* cost of starting multiply for QI */
+ COSTS_N_INSNS (4), /* HI */
+ COSTS_N_INSNS (3), /* SI */
+ COSTS_N_INSNS (4), /* DI */
+ COSTS_N_INSNS (4)}, /* other */
+ 0, /* cost of multiply per each bit set */
+ {COSTS_N_INSNS (16), /* cost of a divide/mod for QI */
+ COSTS_N_INSNS (22), /* HI */
+ COSTS_N_INSNS (30), /* SI */
+ COSTS_N_INSNS (74), /* DI */
+ COSTS_N_INSNS (74)}, /* other */
+ COSTS_N_INSNS (1), /* cost of movsx */
+ COSTS_N_INSNS (1), /* cost of movzx */
+ 8, /* "large" insn */
+ 17, /* MOVE_RATIO */
+ 17, /* CLEAR_RATIO */
+ {6, 6, 6}, /* cost of loading integer registers
+ in QImode, HImode and SImode.
+ Relative to reg-reg move (2). */
+ {6, 6, 6}, /* cost of storing integer registers */
+ {6, 6, 6, 10, 15}, /* cost of loading SSE register
+ in 32bit, 64bit, 128bit, 256bit and 512bit */
+ {6, 6, 6, 10, 15}, /* cost of storing SSE register
+ in 32bit, 64bit, 128bit, 256bit and 512bit */
+ {6, 6, 6, 10, 15}, /* cost of unaligned loads. */
+ {6, 6, 6, 10, 15}, /* cost of unaligned storess. */
+ 2, 3, 4, /* cost of moving XMM,YMM,ZMM register */
+ 6, /* cost of moving SSE register to integer. */
+ 18, 6, /* Gather load static, per_elt. */
+ 18, 6, /* Gather store static, per_elt. */
+ 32, /* size of l1 cache. */
+ 512, /* size of l2 cache. */
+ 64, /* size of prefetch block */
+ 6, /* number of parallel prefetches */
+ 3, /* Branch cost */
+ COSTS_N_INSNS (3), /* cost of FADD and FSUB insns. */
+ COSTS_N_INSNS (5), /* cost of FMUL instruction. */
+ COSTS_N_INSNS (17), /* cost of FDIV instruction. */
+ COSTS_N_INSNS (1), /* cost of FABS instruction. */
+ COSTS_N_INSNS (1), /* cost of FCHS instruction. */
+ COSTS_N_INSNS (14), /* cost of FSQRT instruction. */
+
+ COSTS_N_INSNS (1), /* cost of cheap SSE instruction. */
+ COSTS_N_INSNS (3), /* cost of ADDSS/SD SUBSS/SD insns. */
+ COSTS_N_INSNS (4), /* cost of MULSS instruction. */
+ COSTS_N_INSNS (5), /* cost of MULSD instruction. */
+ COSTS_N_INSNS (5), /* cost of FMA SS instruction. */
+ COSTS_N_INSNS (5), /* cost of FMA SD instruction. */
+ COSTS_N_INSNS (13), /* cost of DIVSS instruction. */
+ COSTS_N_INSNS (17), /* cost of DIVSD instruction. */
+ COSTS_N_INSNS (14), /* cost of SQRTSS instruction. */
+ COSTS_N_INSNS (18), /* cost of SQRTSD instruction. */
+ 1, 4, 3, 3, /* reassoc int, fp, vec_int, vec_fp. */
+ alderlake_memcpy,
+ alderlake_memset,
+ COSTS_N_INSNS (4), /* cond_taken_branch_cost. */
+ COSTS_N_INSNS (2), /* cond_not_taken_branch_cost. */
+ "16:11:8", /* Loop alignment. */
+ "16:11:8", /* Jump alignment. */
+ "0:0:8", /* Label alignment. */
+ "16", /* Func alignment. */
+};
+
/* BTVER1 has optimized REP instruction for medium sized blocks, but for
very small blocks it is better to use loop. For large blocks, libcall can
do nontemporary accesses and beat inline considerably. */
diff --git a/gcc/config/i386/x86-tune-sched.c b/gcc/config/i386/x86-tune-sched.c
index 56ada99a450..0c149a09531 100644
--- a/gcc/config/i386/x86-tune-sched.c
+++ b/gcc/config/i386/x86-tune-sched.c
@@ -72,6 +72,7 @@ ix86_issue_rate (void)
case PROCESSOR_SANDYBRIDGE:
case PROCESSOR_HASWELL:
case PROCESSOR_TREMONT:
+ case PROCESSOR_ALDERLAKE:
case PROCESSOR_GENERIC:
return 4;
@@ -431,6 +432,7 @@ ix86_adjust_cost (rtx_insn *insn, int dep_type, rtx_insn *dep_insn, int cost,
case PROCESSOR_SANDYBRIDGE:
case PROCESSOR_HASWELL:
case PROCESSOR_TREMONT:
+ case PROCESSOR_ALDERLAKE:
case PROCESSOR_GENERIC:
/* Stack engine allows to execute push&pop instructions in parall. */
if ((insn_type == TYPE_PUSH || insn_type == TYPE_POP)
diff --git a/gcc/config/i386/x86-tune.def b/gcc/config/i386/x86-tune.def
index 58e8ead56b4..4ae0b569841 100644
--- a/gcc/config/i386/x86-tune.def
+++ b/gcc/config/i386/x86-tune.def
@@ -42,7 +42,7 @@ see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
DEF_TUNE (X86_TUNE_SCHEDULE, "schedule",
m_PENT | m_LAKEMONT | m_PPRO | m_CORE_ALL | m_BONNELL | m_SILVERMONT
| m_INTEL | m_KNL | m_KNM | m_K6_GEODE | m_AMD_MULTIPLE | m_GOLDMONT
- | m_GOLDMONT_PLUS | m_TREMONT | m_GENERIC)
+ | m_GOLDMONT_PLUS | m_TREMONT | m_ALDERLAKE |m_GENERIC)
/* X86_TUNE_PARTIAL_REG_DEPENDENCY: Enable more register renaming
on modern chips. Preffer stores affecting whole integer register
@@ -51,7 +51,7 @@ DEF_TUNE (X86_TUNE_SCHEDULE, "schedule",
DEF_TUNE (X86_TUNE_PARTIAL_REG_DEPENDENCY, "partial_reg_dependency",
m_P4_NOCONA | m_CORE2 | m_NEHALEM | m_SANDYBRIDGE | m_CORE_AVX2
| m_BONNELL | m_SILVERMONT | m_GOLDMONT | m_GOLDMONT_PLUS | m_INTEL
- | m_KNL | m_KNM | m_AMD_MULTIPLE | m_TREMONT
+ | m_KNL | m_KNM | m_AMD_MULTIPLE | m_TREMONT | m_ALDERLAKE
| m_GENERIC)
/* X86_TUNE_SSE_PARTIAL_REG_DEPENDENCY: This knob promotes all store
@@ -62,7 +62,7 @@ DEF_TUNE (X86_TUNE_PARTIAL_REG_DEPENDENCY, "partial_reg_dependency",
that can be partly masked by careful scheduling of moves. */
DEF_TUNE (X86_TUNE_SSE_PARTIAL_REG_DEPENDENCY, "sse_partial_reg_dependency",
m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_BONNELL | m_AMDFAM10
- | m_BDVER | m_ZNVER | m_TREMONT | m_GENERIC)
+ | m_BDVER | m_ZNVER | m_TREMONT | m_ALDERLAKE | m_GENERIC)
/* X86_TUNE_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY: This knob avoids
partial write to the destination in scalar SSE conversion from FP
@@ -70,14 +70,14 @@ DEF_TUNE (X86_TUNE_SSE_PARTIAL_REG_DEPENDENCY, "sse_partial_reg_dependency",
DEF_TUNE (X86_TUNE_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY,
"sse_partial_reg_fp_converts_dependency",
m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_BONNELL | m_AMDFAM10
- | m_BDVER | m_ZNVER | m_GENERIC)
+ | m_BDVER | m_ZNVER | m_ALDERLAKE| m_GENERIC)
/* X86_TUNE_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY: This knob avoids partial
write to the destination in scalar SSE conversion from integer to FP. */
DEF_TUNE (X86_TUNE_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY,
"sse_partial_reg_converts_dependency",
m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_BONNELL | m_AMDFAM10
- | m_BDVER | m_ZNVER | m_GENERIC)
+ | m_BDVER | m_ZNVER | m_ALDERLAKE | m_GENERIC)
/* X86_TUNE_SSE_SPLIT_REGS: Set for machines where the type and dependencies
are resolved on SSE register parts instead of whole registers, so we may
@@ -103,14 +103,14 @@ DEF_TUNE (X86_TUNE_MOVX, "movx",
m_PPRO | m_P4_NOCONA | m_CORE2 | m_NEHALEM | m_SANDYBRIDGE
| m_BONNELL | m_SILVERMONT | m_GOLDMONT | m_KNL | m_KNM | m_INTEL
| m_GOLDMONT_PLUS | m_GEODE | m_AMD_MULTIPLE
- | m_CORE_AVX2 | m_TREMONT | m_GENERIC)
+ | m_CORE_AVX2 | m_TREMONT | m_ALDERLAKE | m_GENERIC)
/* X86_TUNE_MEMORY_MISMATCH_STALL: Avoid partial stores that are followed by
full sized loads. */
DEF_TUNE (X86_TUNE_MEMORY_MISMATCH_STALL, "memory_mismatch_stall",
m_P4_NOCONA | m_CORE_ALL | m_BONNELL | m_SILVERMONT | m_INTEL
| m_KNL | m_KNM | m_GOLDMONT | m_GOLDMONT_PLUS | m_AMD_MULTIPLE
- | m_TREMONT | m_GENERIC)
+ | m_TREMONT | m_ALDERLAKE | m_GENERIC)
/* X86_TUNE_FUSE_CMP_AND_BRANCH_32: Fuse compare with a subsequent
conditional jump instruction for 32 bit TARGET. */
@@ -166,14 +166,14 @@ DEF_TUNE (X86_TUNE_EPILOGUE_USING_MOVE, "epilogue_using_move",
/* X86_TUNE_USE_LEAVE: Use "leave" instruction in epilogues where it fits. */
DEF_TUNE (X86_TUNE_USE_LEAVE, "use_leave",
m_386 | m_CORE_ALL | m_K6_GEODE | m_AMD_MULTIPLE | m_TREMONT
- | m_GENERIC)
+ | m_ALDERLAKE | m_GENERIC)
/* X86_TUNE_PUSH_MEMORY: Enable generation of "push mem" instructions.
Some chips, like 486 and Pentium works faster with separate load
and push instructions. */
DEF_TUNE (X86_TUNE_PUSH_MEMORY, "push_memory",
m_386 | m_P4_NOCONA | m_CORE_ALL | m_K6_GEODE | m_AMD_MULTIPLE
- | m_TREMONT | m_GENERIC)
+ | m_TREMONT | m_ALDERLAKE | m_GENERIC)
/* X86_TUNE_SINGLE_PUSH: Enable if single push insn is preferred
over esp subtraction. */
@@ -243,14 +243,14 @@ DEF_TUNE (X86_TUNE_READ_MODIFY, "read_modify", ~(m_PENT | m_LAKEMONT | m_PPRO))
DEF_TUNE (X86_TUNE_USE_INCDEC, "use_incdec",
~(m_P4_NOCONA | m_CORE2 | m_NEHALEM | m_SANDYBRIDGE
| m_BONNELL | m_SILVERMONT | m_INTEL | m_KNL | m_KNM | m_GOLDMONT
- | m_GOLDMONT_PLUS | m_TREMONT | m_GENERIC))
+ | m_GOLDMONT_PLUS | m_TREMONT | m_ALDERLAKE | m_GENERIC))
/* X86_TUNE_INTEGER_DFMODE_MOVES: Enable if integer moves are preferred
for DFmode copies */
DEF_TUNE (X86_TUNE_INTEGER_DFMODE_MOVES, "integer_dfmode_moves",
~(m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_BONNELL | m_SILVERMONT
| m_KNL | m_KNM | m_INTEL | m_GEODE | m_AMD_MULTIPLE | m_GOLDMONT
- | m_GOLDMONT_PLUS | m_TREMONT | m_GENERIC))
+ | m_GOLDMONT_PLUS | m_TREMONT | m_ALDERLAKE | m_GENERIC))
/* X86_TUNE_OPT_AGU: Optimize for Address Generation Unit. This flag
will impact LEA instruction selection. */
@@ -298,14 +298,14 @@ DEF_TUNE (X86_TUNE_PREFER_KNOWN_REP_MOVSB_STOSB,
DEF_TUNE (X86_TUNE_MISALIGNED_MOVE_STRING_PRO_EPILOGUES,
"misaligned_move_string_pro_epilogues",
m_386 | m_486 | m_CORE_ALL | m_AMD_MULTIPLE | m_TREMONT
- | m_GENERIC)
+ | m_ALDERLAKE |m_GENERIC)
/* X86_TUNE_USE_SAHF: Controls use of SAHF. */
DEF_TUNE (X86_TUNE_USE_SAHF, "use_sahf",
m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_BONNELL | m_SILVERMONT
| m_KNL | m_KNM | m_INTEL | m_K6_GEODE | m_K8 | m_AMDFAM10 | m_BDVER
| m_BTVER | m_ZNVER | m_GOLDMONT | m_GOLDMONT_PLUS | m_TREMONT
- | m_GENERIC)
+ | m_ALDERLAKE | m_GENERIC)
/* X86_TUNE_USE_CLTD: Controls use of CLTD and CTQO instructions. */
DEF_TUNE (X86_TUNE_USE_CLTD, "use_cltd",
@@ -316,12 +316,12 @@ DEF_TUNE (X86_TUNE_USE_CLTD, "use_cltd",
DEF_TUNE (X86_TUNE_USE_BT, "use_bt",
m_CORE_ALL | m_BONNELL | m_SILVERMONT | m_KNL | m_KNM | m_INTEL
| m_LAKEMONT | m_AMD_MULTIPLE | m_GOLDMONT | m_GOLDMONT_PLUS
- | m_TREMONT | m_GENERIC)
+ | m_TREMONT | m_ALDERLAKE | m_GENERIC)
/* X86_TUNE_AVOID_FALSE_DEP_FOR_BMI: Avoid false dependency
for bit-manipulation instructions. */
DEF_TUNE (X86_TUNE_AVOID_FALSE_DEP_FOR_BMI, "avoid_false_dep_for_bmi",
- m_SANDYBRIDGE | m_CORE_AVX2 | m_TREMONT | m_GENERIC)
+ m_SANDYBRIDGE | m_CORE_AVX2 | m_TREMONT | m_ALDERLAKE | m_GENERIC)
/* X86_TUNE_ADJUST_UNROLL: This enables adjusting the unroll factor based
on hardware capabilities. Bdver3 hardware has a loop buffer which makes
@@ -333,11 +333,11 @@ DEF_TUNE (X86_TUNE_ADJUST_UNROLL, "adjust_unroll_factor", m_BDVER3 | m_BDVER4)
if-converted sequence to one. */
DEF_TUNE (X86_TUNE_ONE_IF_CONV_INSN, "one_if_conv_insn",
m_SILVERMONT | m_KNL | m_KNM | m_INTEL | m_CORE_ALL | m_GOLDMONT
- | m_GOLDMONT_PLUS | m_TREMONT | m_GENERIC)
+ | m_GOLDMONT_PLUS | m_TREMONT | m_ALDERLAKE | m_GENERIC)
/* X86_TUNE_AVOID_MFENCE: Use lock prefixed instructions instead of mfence. */
DEF_TUNE (X86_TUNE_AVOID_MFENCE, "avoid_mfence",
- m_CORE_ALL | m_BDVER | m_ZNVER | m_TREMONT | m_GENERIC)
+ m_CORE_ALL | m_BDVER | m_ZNVER | m_TREMONT | m_ALDERLAKE | m_GENERIC)
/* X86_TUNE_EXPAND_ABS: This enables a new abs pattern by
generating instructions for abs (x) = (((signed) x >> (W-1) ^ x) -
@@ -361,7 +361,8 @@ DEF_TUNE (X86_TUNE_USE_HIMODE_FIOP, "use_himode_fiop",
DEF_TUNE (X86_TUNE_USE_SIMODE_FIOP, "use_simode_fiop",
~(m_PENT | m_LAKEMONT | m_PPRO | m_CORE_ALL | m_BONNELL
| m_SILVERMONT | m_KNL | m_KNM | m_INTEL | m_AMD_MULTIPLE
- | m_GOLDMONT | m_GOLDMONT_PLUS | m_TREMONT | m_GENERIC))
+ | m_GOLDMONT | m_GOLDMONT_PLUS | m_TREMONT | m_ALDERLAKE
+ | m_GENERIC))
/* X86_TUNE_USE_FFREEP: Use freep instruction instead of fstp. */
DEF_TUNE (X86_TUNE_USE_FFREEP, "use_ffreep", m_AMD_MULTIPLE)
@@ -370,7 +371,7 @@ DEF_TUNE (X86_TUNE_USE_FFREEP, "use_ffreep", m_AMD_MULTIPLE)
DEF_TUNE (X86_TUNE_EXT_80387_CONSTANTS, "ext_80387_constants",
m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_BONNELL | m_SILVERMONT
| m_KNL | m_KNM | m_INTEL | m_K6_GEODE | m_ATHLON_K8 | m_GOLDMONT
- | m_GOLDMONT_PLUS | m_TREMONT | m_GENERIC)
+ | m_GOLDMONT_PLUS | m_TREMONT | m_ALDERLAKE | m_GENERIC)
/*****************************************************************************/
/* SSE instruction selection tuning */
@@ -385,15 +386,15 @@ DEF_TUNE (X86_TUNE_GENERAL_REGS_SSE_SPILL, "general_regs_sse_spill",
of a sequence loading registers by parts. */
DEF_TUNE (X86_TUNE_SSE_UNALIGNED_LOAD_OPTIMAL, "sse_unaligned_load_optimal",
m_NEHALEM | m_SANDYBRIDGE | m_CORE_AVX2 | m_SILVERMONT | m_KNL | m_KNM
- | m_INTEL | m_GOLDMONT | m_GOLDMONT_PLUS
- | m_TREMONT | m_AMDFAM10 | m_BDVER | m_BTVER | m_ZNVER | m_GENERIC)
+ | m_INTEL | m_GOLDMONT | m_GOLDMONT_PLUS | m_TREMONT | m_ALDERLAKE
+ | m_AMDFAM10 | m_BDVER | m_BTVER | m_ZNVER | m_GENERIC)
/* X86_TUNE_SSE_UNALIGNED_STORE_OPTIMAL: Use movups for misaligned stores
instead of a sequence loading registers by parts. */
DEF_TUNE (X86_TUNE_SSE_UNALIGNED_STORE_OPTIMAL, "sse_unaligned_store_optimal",
m_NEHALEM | m_SANDYBRIDGE | m_CORE_AVX2 | m_SILVERMONT | m_KNL | m_KNM
| m_INTEL | m_GOLDMONT | m_GOLDMONT_PLUS
- | m_TREMONT | m_BDVER | m_ZNVER | m_GENERIC)
+ | m_TREMONT | m_ALDERLAKE | m_BDVER | m_ZNVER | m_GENERIC)
/* X86_TUNE_SSE_PACKED_SINGLE_INSN_OPTIMAL: Use packed single
precision 128bit instructions instead of double where possible. */
@@ -402,13 +403,13 @@ DEF_TUNE (X86_TUNE_SSE_PACKED_SINGLE_INSN_OPTIMAL, "sse_packed_single_insn_optim
/* X86_TUNE_SSE_TYPELESS_STORES: Always movaps/movups for 128bit stores. */
DEF_TUNE (X86_TUNE_SSE_TYPELESS_STORES, "sse_typeless_stores",
- m_AMD_MULTIPLE | m_CORE_ALL | m_TREMONT | m_GENERIC)
+ m_AMD_MULTIPLE | m_CORE_ALL | m_TREMONT | m_ALDERLAKE | m_GENERIC)
/* X86_TUNE_SSE_LOAD0_BY_PXOR: Always use pxor to load0 as opposed to
xorps/xorpd and other variants. */
DEF_TUNE (X86_TUNE_SSE_LOAD0_BY_PXOR, "sse_load0_by_pxor",
m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_BDVER | m_BTVER | m_ZNVER
- | m_TREMONT | m_GENERIC)
+ | m_TREMONT | m_ALDERLAKE | m_GENERIC)
/* X86_TUNE_INTER_UNIT_MOVES_TO_VEC: Enable moves in from integer
to SSE registers. If disabled, the moves will be done by storing
@@ -454,11 +455,12 @@ DEF_TUNE (X86_TUNE_SLOW_PSHUFB, "slow_pshufb",
/* X86_TUNE_AVOID_4BYTE_PREFIXES: Avoid instructions requiring 4+ bytes of prefixes. */
DEF_TUNE (X86_TUNE_AVOID_4BYTE_PREFIXES, "avoid_4byte_prefixes",
- m_SILVERMONT | m_GOLDMONT | m_GOLDMONT_PLUS | m_TREMONT | m_INTEL)
+ m_SILVERMONT | m_GOLDMONT | m_GOLDMONT_PLUS | m_TREMONT | m_ALDERLAKE
+ | m_INTEL)
/* X86_TUNE_USE_GATHER: Use gather instructions. */
DEF_TUNE (X86_TUNE_USE_GATHER, "use_gather",
- ~(m_ZNVER1 | m_ZNVER2 | m_GENERIC))
+ ~(m_ZNVER1 | m_ZNVER2 | m_GENERIC | m_ALDERLAKE))
/* X86_TUNE_AVOID_128FMA_CHAINS: Avoid creating loops with tight 128bit or
smaller FMA chain. */
--
2.17.1

View File

@ -0,0 +1,66 @@
c++: Optimize constinit thread_local vars [PR101786]
The paper that introduced constinit mentioned in rationale that constinit
can be used on externs as well and that it can be used to avoid the
thread_local initialization wrappers, because the standard requires that
if constinit is present on any declaration, it is also present on the
initialization declaration, even if it is in some other TU etc.
There is a small problem though, we use the tls wrappers not just if
the thread_local variable needs dynamic initialization, but also when
it has static initialization, but non-trivial destructor, as the
"dynamic initialization" in that case needs to register the destructor.
So, the following patch optimizes constinit thread_local vars only
if we can prove they will not have non-trivial destructors. That includes
the case where we have incomplete type where we don't know and need to
conservatively assume the type will have non-trivial destructor at the
initializing declaration side.
2021-08-11 Jakub Jelinek <jakub@redhat.com>
PR c++/101786
* decl2.c (var_defined_without_dynamic_init): Return true for
DECL_DECLARED_CONSTINIT_P with complete type and trivial destructor.
* g++.dg/cpp2a/constinit16.C: New test.
--- gcc/cp/decl2.c
+++ gcc/cp/decl2.c
@@ -3447,6 +3447,12 @@ set_guard (tree guard)
static bool
var_defined_without_dynamic_init (tree var)
{
+ /* constinit vars are guaranteed to not have dynamic initializer,
+ but still registering the destructor counts as dynamic initialization. */
+ if (DECL_DECLARED_CONSTINIT_P (var)
+ && COMPLETE_TYPE_P (TREE_TYPE (var))
+ && !TYPE_HAS_NONTRIVIAL_DESTRUCTOR (TREE_TYPE (var)))
+ return true;
/* If it's defined in another TU, we can't tell. */
if (DECL_EXTERNAL (var))
return false;
--- gcc/testsuite/g++.dg/cpp2a/constinit16.C
+++ gcc/testsuite/g++.dg/cpp2a/constinit16.C
@@ -0,0 +1,21 @@
+// PR c++/101786
+// { dg-do compile { target c++20 } }
+// { dg-add-options tls }
+// { dg-require-alias "" }
+// { dg-require-effective-target tls_runtime }
+// { dg-final { scan-assembler-not "_ZTH17mythreadlocalvar1" } }
+// { dg-final { scan-assembler "_ZTH17mythreadlocalvar2" } }
+// { dg-final { scan-assembler-not "_ZTH17mythreadlocalvar3" } }
+// { dg-final { scan-assembler "_ZTH17mythreadlocalvar4" } }
+
+extern thread_local constinit int mythreadlocalvar1;
+struct S;
+extern thread_local constinit S mythreadlocalvar2;
+struct T { int t; };
+extern thread_local constinit T mythreadlocalvar3;
+struct U { int u; ~U (); };
+extern thread_local constinit U mythreadlocalvar4;
+int foo () { return mythreadlocalvar1; }
+S *bar () { return &mythreadlocalvar2; }
+T *baz () { return &mythreadlocalvar3; }
+U *qux () { return &mythreadlocalvar4; }

View File

@ -0,0 +1,197 @@
libcpp: Fix up #__VA_OPT__ handling [PR103415]
stringify_arg uses pfile->u_buff to create the string literal.
Unfortunately, paste_tokens -> _cpp_lex_direct -> lex_number -> _cpp_unaligned_alloc
can in some cases use pfile->u_buff too, which results in losing everything
prepared for the string literal until the token pasting.
The following patch fixes that by not calling paste_token during the
construction of the string literal, but doing that before. All the tokens
we are processing have been pushed into a token buffer using
tokens_buff_add_token so it is fine if we paste some of them in that buffer
(successful pasting creates a new token in that buffer), move following
tokens if any to make it contiguous, pop (throw away) the extra tokens at
the end and then do stringify_arg.
Also, paste_tokens now copies over PREV_WHITE and PREV_FALLTHROUGH flags
from the original lhs token to the replacement token. Copying that way
the PREV_WHITE flag is needed for the #__VA_OPT__ handling and copying
over PREV_FALLTHROUGH fixes the new Wimplicit-fallthrough-38.c test.
2021-12-01 Jakub Jelinek <jakub@redhat.com>
PR preprocessor/103415
libcpp/
* macro.c (stringify_arg): Remove va_opt argument and va_opt handling.
(paste_tokens): On successful paste or in PREV_WHITE and
PREV_FALLTHROUGH flags from the *plhs token to the new token.
(replace_args): Adjust stringify_arg callers. For #__VA_OPT__,
perform token pasting in a separate loop before stringify_arg call.
gcc/testsuite/
* c-c++-common/cpp/va-opt-8.c: New test.
* c-c++-common/Wimplicit-fallthrough-38.c: New test.
--- libcpp/macro.c.jj
+++ libcpp/macro.c
@@ -295,7 +295,7 @@ static cpp_context *next_context (cpp_re
static const cpp_token *padding_token (cpp_reader *, const cpp_token *);
static const cpp_token *new_string_token (cpp_reader *, uchar *, unsigned int);
static const cpp_token *stringify_arg (cpp_reader *, const cpp_token **,
- unsigned int, bool);
+ unsigned int);
static void paste_all_tokens (cpp_reader *, const cpp_token *);
static bool paste_tokens (cpp_reader *, location_t,
const cpp_token **, const cpp_token *);
@@ -826,8 +826,7 @@ cpp_quote_string (uchar *dest, const uch
/* Convert a token sequence FIRST to FIRST+COUNT-1 to a single string token
according to the rules of the ISO C #-operator. */
static const cpp_token *
-stringify_arg (cpp_reader *pfile, const cpp_token **first, unsigned int count,
- bool va_opt)
+stringify_arg (cpp_reader *pfile, const cpp_token **first, unsigned int count)
{
unsigned char *dest;
unsigned int i, escape_it, backslash_count = 0;
@@ -844,24 +843,6 @@ stringify_arg (cpp_reader *pfile, const
{
const cpp_token *token = first[i];
- if (va_opt && (token->flags & PASTE_LEFT))
- {
- location_t virt_loc = pfile->invocation_location;
- const cpp_token *rhs;
- do
- {
- if (i == count)
- abort ();
- rhs = first[++i];
- if (!paste_tokens (pfile, virt_loc, &token, rhs))
- {
- --i;
- break;
- }
- }
- while (rhs->flags & PASTE_LEFT);
- }
-
if (token->type == CPP_PADDING)
{
if (source == NULL
@@ -995,6 +976,7 @@ paste_tokens (cpp_reader *pfile, locatio
return false;
}
+ lhs->flags |= (*plhs)->flags & (PREV_WHITE | PREV_FALLTHROUGH);
*plhs = lhs;
_cpp_pop_buffer (pfile);
return true;
@@ -1937,8 +1919,7 @@ replace_args (cpp_reader *pfile, cpp_has
if (src->flags & STRINGIFY_ARG)
{
if (!arg->stringified)
- arg->stringified = stringify_arg (pfile, arg->first, arg->count,
- false);
+ arg->stringified = stringify_arg (pfile, arg->first, arg->count);
}
else if ((src->flags & PASTE_LEFT)
|| (src != macro->exp.tokens && (src[-1].flags & PASTE_LEFT)))
@@ -2065,11 +2046,46 @@ replace_args (cpp_reader *pfile, cpp_has
{
unsigned int count
= start ? paste_flag - start : tokens_buff_count (buff);
- const cpp_token *t
- = stringify_arg (pfile,
- start ? start + 1
- : (const cpp_token **) (buff->base),
- count, true);
+ const cpp_token **first
+ = start ? start + 1
+ : (const cpp_token **) (buff->base);
+ unsigned int i, j;
+
+ /* Paste any tokens that need to be pasted before calling
+ stringify_arg, because stringify_arg uses pfile->u_buff
+ which paste_tokens can use as well. */
+ for (i = 0, j = 0; i < count; i++, j++)
+ {
+ const cpp_token *token = first[i];
+
+ if (token->flags & PASTE_LEFT)
+ {
+ location_t virt_loc = pfile->invocation_location;
+ const cpp_token *rhs;
+ do
+ {
+ if (i == count)
+ abort ();
+ rhs = first[++i];
+ if (!paste_tokens (pfile, virt_loc, &token, rhs))
+ {
+ --i;
+ break;
+ }
+ }
+ while (rhs->flags & PASTE_LEFT);
+ }
+
+ first[j] = token;
+ }
+ if (j != i)
+ {
+ while (i-- != j)
+ tokens_buff_remove_last_token (buff);
+ count = j;
+ }
+
+ const cpp_token *t = stringify_arg (pfile, first, count);
while (count--)
tokens_buff_remove_last_token (buff);
if (src->flags & PASTE_LEFT)
--- gcc/testsuite/c-c++-common/cpp/va-opt-8.c.jj
+++ gcc/testsuite/c-c++-common/cpp/va-opt-8.c
@@ -0,0 +1,18 @@
+/* PR preprocessor/103415 */
+/* { dg-do run } */
+/* { dg-options "-std=gnu99" { target c } } */
+/* { dg-options "-std=c++20" { target c++ } } */
+
+#define n(x, ...) = #__VA_OPT__(x##3)
+#define o(x, ...) #__VA_OPT__(x##__VA_ARGS__##9)
+const char *c n(1 2, 4);
+const char *d = o(5 6, 7 8);
+
+int
+main ()
+{
+ if (__builtin_strcmp (c, "1 23")
+ || __builtin_strcmp (d, "5 67 89"))
+ __builtin_abort ();
+ return 0;
+}
--- gcc/testsuite/c-c++-common/Wimplicit-fallthrough-38.c.jj
+++ gcc/testsuite/c-c++-common/Wimplicit-fallthrough-38.c
@@ -0,0 +1,24 @@
+/* { dg-do compile } */
+/* { dg-options "-Wimplicit-fallthrough=3" } */
+
+#define FOO \
+int \
+foo (int a) \
+{ \
+ switch (a) \
+ { \
+ case 1: \
+ ++a; \
+ /* FALLTHRU */ \
+ case 2: \
+ ++a; \
+ /* FALLTHRU */ \
+ ca##se 3: \
+ ++a; \
+ default: \
+ break; \
+ } \
+ return a; \
+}
+
+FOO

View File

@ -0,0 +1,307 @@
c++: Add C++20 #__VA_OPT__ support
The following patch implements C++20 # __VA_OPT__ (...) support.
Testcases cover what I came up with myself and what LLVM has for #__VA_OPT__
in its testsuite and the string literals are identical between the two
compilers on the va-opt-5.c testcase.
2021-08-17 Jakub Jelinek <jakub@redhat.com>
libcpp/
* macro.c (vaopt_state): Add m_stringify member.
(vaopt_state::vaopt_state): Initialize it.
(vaopt_state::update): Overwrite it.
(vaopt_state::stringify): New method.
(stringify_arg): Replace arg argument with first, count arguments
and add va_opt argument. Use first instead of arg->first and
count instead of arg->count, for va_opt add paste_tokens handling.
(paste_tokens): Fix up len calculation. Don't spell rhs twice,
instead use %.*s to supply lhs and rhs spelling lengths. Don't call
_cpp_backup_tokens here.
(paste_all_tokens): Call it here instead.
(replace_args): Adjust stringify_arg caller. For vaopt_state::END
if stringify is true handle __VA_OPT__ stringification.
(create_iso_definition): Handle # __VA_OPT__ similarly to # macro_arg.
gcc/testsuite/
* c-c++-common/cpp/va-opt-5.c: New test.
* c-c++-common/cpp/va-opt-6.c: New test.
--- libcpp/macro.c
+++ libcpp/macro.c
@@ -118,6 +118,7 @@ class vaopt_state {
m_arg (arg),
m_variadic (is_variadic),
m_last_was_paste (false),
+ m_stringify (false),
m_state (0),
m_paste_location (0),
m_location (0),
@@ -145,6 +146,7 @@ class vaopt_state {
}
++m_state;
m_location = token->src_loc;
+ m_stringify = (token->flags & STRINGIFY_ARG) != 0;
return BEGIN;
}
else if (m_state == 1)
@@ -234,6 +236,12 @@ class vaopt_state {
return m_state == 0;
}
+ /* Return true for # __VA_OPT__. */
+ bool stringify () const
+ {
+ return m_stringify;
+ }
+
private:
/* The cpp_reader. */
@@ -247,6 +255,8 @@ class vaopt_state {
/* If true, the previous token was ##. This is used to detect when
a paste occurs at the end of the sequence. */
bool m_last_was_paste;
+ /* True for #__VA_OPT__. */
+ bool m_stringify;
/* The state variable:
0 means not parsing
@@ -284,7 +294,8 @@ static _cpp_buff *collect_args (cpp_read
static cpp_context *next_context (cpp_reader *);
static const cpp_token *padding_token (cpp_reader *, const cpp_token *);
static const cpp_token *new_string_token (cpp_reader *, uchar *, unsigned int);
-static const cpp_token *stringify_arg (cpp_reader *, macro_arg *);
+static const cpp_token *stringify_arg (cpp_reader *, const cpp_token **,
+ unsigned int, bool);
static void paste_all_tokens (cpp_reader *, const cpp_token *);
static bool paste_tokens (cpp_reader *, location_t,
const cpp_token **, const cpp_token *);
@@ -812,10 +823,11 @@ cpp_quote_string (uchar *dest, const uch
return dest;
}
-/* Convert a token sequence ARG to a single string token according to
- the rules of the ISO C #-operator. */
+/* Convert a token sequence FIRST to FIRST+COUNT-1 to a single string token
+ according to the rules of the ISO C #-operator. */
static const cpp_token *
-stringify_arg (cpp_reader *pfile, macro_arg *arg)
+stringify_arg (cpp_reader *pfile, const cpp_token **first, unsigned int count,
+ bool va_opt)
{
unsigned char *dest;
unsigned int i, escape_it, backslash_count = 0;
@@ -828,9 +840,27 @@ stringify_arg (cpp_reader *pfile, macro_
*dest++ = '"';
/* Loop, reading in the argument's tokens. */
- for (i = 0; i < arg->count; i++)
+ for (i = 0; i < count; i++)
{
- const cpp_token *token = arg->first[i];
+ const cpp_token *token = first[i];
+
+ if (va_opt && (token->flags & PASTE_LEFT))
+ {
+ location_t virt_loc = pfile->invocation_location;
+ const cpp_token *rhs;
+ do
+ {
+ if (i == count)
+ abort ();
+ rhs = first[++i];
+ if (!paste_tokens (pfile, virt_loc, &token, rhs))
+ {
+ --i;
+ break;
+ }
+ }
+ while (rhs->flags & PASTE_LEFT);
+ }
if (token->type == CPP_PADDING)
{
@@ -917,7 +947,7 @@ paste_tokens (cpp_reader *pfile, locatio
cpp_token *lhs;
unsigned int len;
- len = cpp_token_len (*plhs) + cpp_token_len (rhs) + 1;
+ len = cpp_token_len (*plhs) + cpp_token_len (rhs) + 2;
buf = (unsigned char *) alloca (len);
end = lhsend = cpp_spell_token (pfile, *plhs, buf, true);
@@ -943,8 +973,10 @@ paste_tokens (cpp_reader *pfile, locatio
location_t saved_loc = lhs->src_loc;
_cpp_pop_buffer (pfile);
- _cpp_backup_tokens (pfile, 1);
- *lhsend = '\0';
+
+ unsigned char *rhsstart = lhsend;
+ if ((*plhs)->type == CPP_DIV && rhs->type != CPP_EQ)
+ rhsstart++;
/* We have to remove the PASTE_LEFT flag from the old lhs, but
we want to keep the new location. */
@@ -956,8 +988,10 @@ paste_tokens (cpp_reader *pfile, locatio
/* Mandatory error for all apart from assembler. */
if (CPP_OPTION (pfile, lang) != CLK_ASM)
cpp_error_with_line (pfile, CPP_DL_ERROR, location, 0,
- "pasting \"%s\" and \"%s\" does not give a valid preprocessing token",
- buf, cpp_token_as_text (pfile, rhs));
+ "pasting \"%.*s\" and \"%.*s\" does not give "
+ "a valid preprocessing token",
+ (int) (lhsend - buf), buf,
+ (int) (end - rhsstart), rhsstart);
return false;
}
@@ -1033,7 +1067,10 @@ paste_all_tokens (cpp_reader *pfile, con
abort ();
}
if (!paste_tokens (pfile, virt_loc, &lhs, rhs))
- break;
+ {
+ _cpp_backup_tokens (pfile, 1);
+ break;
+ }
}
while (rhs->flags & PASTE_LEFT);
@@ -1900,7 +1937,8 @@ replace_args (cpp_reader *pfile, cpp_has
if (src->flags & STRINGIFY_ARG)
{
if (!arg->stringified)
- arg->stringified = stringify_arg (pfile, arg);
+ arg->stringified = stringify_arg (pfile, arg->first, arg->count,
+ false);
}
else if ((src->flags & PASTE_LEFT)
|| (src != macro->exp.tokens && (src[-1].flags & PASTE_LEFT)))
@@ -2023,6 +2061,24 @@ replace_args (cpp_reader *pfile, cpp_has
paste_flag = tokens_buff_last_token_ptr (buff);
}
+ if (vaopt_tracker.stringify ())
+ {
+ unsigned int count
+ = start ? paste_flag - start : tokens_buff_count (buff);
+ const cpp_token *t
+ = stringify_arg (pfile,
+ start ? start + 1
+ : (const cpp_token **) (buff->base),
+ count, true);
+ while (count--)
+ tokens_buff_remove_last_token (buff);
+ if (src->flags & PASTE_LEFT)
+ copy_paste_flag (pfile, &t, src);
+ tokens_buff_add_token (buff, virt_locs,
+ t, t->src_loc, t->src_loc,
+ NULL, 0);
+ continue;
+ }
if (start && paste_flag == start && (*start)->flags & PASTE_LEFT)
/* If __VA_OPT__ expands to nothing (either because __VA_ARGS__
is empty or because it is __VA_OPT__() ), drop PASTE_LEFT
@@ -3584,7 +3640,10 @@ create_iso_definition (cpp_reader *pfile
function-like macros when lexing the subsequent token. */
if (macro->count > 1 && token[-1].type == CPP_HASH && macro->fun_like)
{
- if (token->type == CPP_MACRO_ARG)
+ if (token->type == CPP_MACRO_ARG
+ || (macro->variadic
+ && token->type == CPP_NAME
+ && token->val.node.node == pfile->spec_nodes.n__VA_OPT__))
{
if (token->flags & PREV_WHITE)
token->flags |= SP_PREV_WHITE;
--- gcc/testsuite/c-c++-common/cpp/va-opt-5.c
+++ gcc/testsuite/c-c++-common/cpp/va-opt-5.c
@@ -0,0 +1,67 @@
+/* { dg-do run } */
+/* { dg-options "-std=gnu99" { target c } } */
+/* { dg-options "-std=c++20" { target c++ } } */
+
+#define lparen (
+#define a0 fooa0
+#define a1 fooa1 a0
+#define a2 fooa2 a1
+#define a3 fooa3 a2
+#define a() b lparen )
+#define b() c lparen )
+#define c() d lparen )
+#define g h
+#define i(j) j
+#define f(...) #__VA_OPT__(g i(0))
+#define k(x,...) # __VA_OPT__(x) #x #__VA_OPT__(__VA_ARGS__)
+#define l(x,...) #__VA_OPT__(a1 x)
+#define m(x,...) "a()" #__VA_OPT__(a3 __VA_ARGS__ x ## __VA_ARGS__ ## x ## c a3) "a()"
+#define n(x,...) = #__VA_OPT__(a3 __VA_ARGS__ x ## __VA_ARGS__ ## x ## c a3) #x #__VA_OPT__(a0 __VA_ARGS__ x ## __VA_ARGS__ ## x ## c a0) ;
+#define o(x, ...) #__VA_OPT__(x##x x##x)
+#define p(x, ...) #__VA_OPT__(_Pragma ("foobar"))
+#define q(...) #__VA_OPT__(/* foo */x/* bar */)
+const char *v1 = f();
+const char *v2 = f(123);
+const char *v3 = k(1);
+const char *v4 = k(1, 2, 3 );
+const char *v5 = l(a());
+const char *v6 = l(a1 a(), 1);
+const char *v7 = m();
+const char *v8 = m(,);
+const char *v9 = m(,a3);
+const char *v10 = m(a3,a(),a0);
+const char *v11 n()
+const char *v12 n(,)
+const char *v13 n(,a0)
+const char *v14 n(a0, a(),a0)
+const char *v15 = o(, 0);
+const char *v16 = p(0);
+const char *v17 = p(0, 1);
+const char *v18 = q();
+const char *v19 = q(1);
+
+int
+main ()
+{
+ if (__builtin_strcmp (v1, "")
+ || __builtin_strcmp (v2, "g i(0)")
+ || __builtin_strcmp (v3, "1")
+ || __builtin_strcmp (v4, "112, 3")
+ || __builtin_strcmp (v5, "")
+ || __builtin_strcmp (v6, "a1 fooa1 fooa0 b ( )")
+ || __builtin_strcmp (v7, "a()a()")
+ || __builtin_strcmp (v8, "a()a()")
+ || __builtin_strcmp (v9, "a()a3 fooa3 fooa2 fooa1 fooa0 a3c a3a()")
+ || __builtin_strcmp (v10, "a()a3 b ( ),fooa0 a3a(),a0a3c a3a()")
+ || __builtin_strcmp (v11, "")
+ || __builtin_strcmp (v12, "")
+ || __builtin_strcmp (v13, "a3 fooa0 a0c a3a0 fooa0 a0c a0")
+ || __builtin_strcmp (v14, "a3 b ( ),fooa0 a0a(),a0a0c a3a0a0 b ( ),fooa0 a0a(),a0a0c a0")
+ || __builtin_strcmp (v15, "")
+ || __builtin_strcmp (v16, "")
+ || __builtin_strcmp (v17, "_Pragma (\"foobar\")")
+ || __builtin_strcmp (v18, "")
+ || __builtin_strcmp (v19, "x"))
+ __builtin_abort ();
+ return 0;
+}
--- gcc/testsuite/c-c++-common/cpp/va-opt-6.c
+++ gcc/testsuite/c-c++-common/cpp/va-opt-6.c
@@ -0,0 +1,17 @@
+/* { dg-do preprocess } */
+/* { dg-options "-std=gnu99" { target c } } */
+/* { dg-options "-std=c++20" { target c++ } } */
+
+#define a ""
+#define b(...) a ## #__VA_OPT__(1) /* { dg-error "pasting \"a\" and \"\"\"\" does not give a valid preprocessing token" } */
+#define c(...) a ## #__VA_OPT__(1) /* { dg-error "pasting \"a\" and \"\"1\"\" does not give a valid preprocessing token" } */
+#define d(...) #__VA_OPT__(1) ## !
+#define e(...) #__VA_OPT__(1) ## !
+#define f(...) #__VA_OPT__(. ## !)
+#define g(...) #__VA_OPT__(. ## !)
+b()
+c(1)
+d( ) /* { dg-error "pasting \"\"\"\" and \"!\" does not give a valid preprocessing token" } */
+e( 1 ) /* { dg-error "pasting \"\"1\"\" and \"!\" does not give a valid preprocessing token" } */
+f()
+g(0) /* { dg-error "pasting \".\" and \"!\" does not give a valid preprocessing token" } */

View File

@ -0,0 +1,294 @@
From cbd145b132c79c36e990a0eaf10c86159009f18d Mon Sep 17 00:00:00 2001
From: "H.J. Lu" <hjl.tools@gmail.com>
Date: Wed, 15 Sep 2021 14:15:10 +0800
Subject: [PATCH 1/3] x86: Update -mtune=tremont
Initial -mtune=tremont update
1. Use Haswell scheduling model.
2. Assume that stack engine allows to execute push&pop instructions in
parall.
3. Prepare for scheduling pass as -mtune=generic.
4. Use the same issue rate as -mtune=generic.
5. Enable partial_reg_dependency.
6. Disable accumulate_outgoing_args
7. Enable use_leave
8. Enable push_memory
9. Disable four_jump_limit
10. Disable opt_agu
11. Disable avoid_lea_for_addr
12. Disable avoid_mem_opnd_for_cmove
13. Enable misaligned_move_string_pro_epilogues
14. Enable use_cltd
16. Enable avoid_false_dep_for_bmi
17. Enable avoid_mfence
18. Disable expand_abs
19. Enable sse_typeless_stores
20. Enable sse_load0_by_pxor
21. Disable split_mem_opnd_for_fp_converts
22. Disable slow_pshufb
23. Enable partial_reg_dependency
This is the first patch to tune for Tremont. With all patches applied,
performance impacts on SPEC CPU 2017 are:
500.perlbench_r 1.81%
502.gcc_r 0.57%
505.mcf_r 1.16%
520.omnetpp_r 0.00%
523.xalancbmk_r 0.00%
525.x264_r 4.55%
531.deepsjeng_r 0.00%
541.leela_r 0.39%
548.exchange2_r 1.13%
557.xz_r 0.00%
geomean for intrate 0.95%
503.bwaves_r 0.00%
507.cactuBSSN_r 6.94%
508.namd_r 12.37%
510.parest_r 1.01%
511.povray_r 3.70%
519.lbm_r 36.61%
521.wrf_r 8.79%
526.blender_r 2.91%
527.cam4_r 6.23%
538.imagick_r 0.28%
544.nab_r 21.99%
549.fotonik3d_r 3.63%
554.roms_r -1.20%
geomean for fprate 7.50%
gcc/ChangeLog
* common/config/i386/i386-common.c: Use Haswell scheduling model
for Tremont.
* config/i386/i386.c (ix86_sched_init_global): Prepare for Tremont
scheduling pass.
* config/i386/x86-tune-sched.c (ix86_issue_rate): Change Tremont
issue rate to 4.
(ix86_adjust_cost): Handle Tremont.
* config/i386/x86-tune.def (X86_TUNE_SSE_PARTIAL_REG_DEPENDENCY):
Enable for Tremont.
(X86_TUNE_USE_LEAVE): Likewise.
(X86_TUNE_PUSH_MEMORY): Likewise.
(X86_TUNE_MISALIGNED_MOVE_STRING_PRO_EPILOGUES): Likewise.
(X86_TUNE_USE_CLTD): Likewise.
(X86_TUNE_AVOID_FALSE_DEP_FOR_BMI): Likewise.
(X86_TUNE_AVOID_MFENCE): Likewise.
(X86_TUNE_SSE_TYPELESS_STORES): Likewise.
(X86_TUNE_SSE_LOAD0_BY_PXOR): Likewise.
(X86_TUNE_ACCUMULATE_OUTGOING_ARGS): Disable for Tremont.
(X86_TUNE_FOUR_JUMP_LIMIT): Likewise.
(X86_TUNE_OPT_AGU): Likewise.
(X86_TUNE_AVOID_LEA_FOR_ADDR): Likewise.
(X86_TUNE_AVOID_MEM_OPND_FOR_CMOVE): Likewise.
(X86_TUNE_EXPAND_ABS): Likewise.
(X86_TUNE_SPLIT_MEM_OPND_FOR_FP_CONVERTS): Likewise.
(X86_TUNE_SLOW_PSHUFB): Likewise.
---
gcc/common/config/i386/i386-common.c | 2 +-
gcc/config/i386/i386.c | 1 +
gcc/config/i386/x86-tune-sched.c | 2 ++
gcc/config/i386/x86-tune.def | 37 ++++++++++++++--------------
4 files changed, 23 insertions(+), 19 deletions(-)
diff --git a/gcc/common/config/i386/i386-common.c b/gcc/common/config/i386/i386-common.c
index 38dbb9d9263..ef382ec9a22 100644
--- a/gcc/common/config/i386/i386-common.c
+++ b/gcc/common/config/i386/i386-common.c
@@ -1916,7 +1916,7 @@ const pta processor_alias_table[] =
M_CPU_TYPE (INTEL_GOLDMONT), P_PROC_SSE4_2},
{"goldmont-plus", PROCESSOR_GOLDMONT_PLUS, CPU_GLM, PTA_GOLDMONT_PLUS,
M_CPU_TYPE (INTEL_GOLDMONT_PLUS), P_PROC_SSE4_2},
- {"tremont", PROCESSOR_TREMONT, CPU_GLM, PTA_TREMONT,
+ {"tremont", PROCESSOR_TREMONT, CPU_HASWELL, PTA_TREMONT,
M_CPU_TYPE (INTEL_TREMONT), P_PROC_SSE4_2},
{"knl", PROCESSOR_KNL, CPU_SLM, PTA_KNL,
M_CPU_TYPE (INTEL_KNL), P_PROC_AVX512F},
diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
index 42c47d2b12b..fc2a27b5cbf 100644
--- a/gcc/config/i386/i386.c
+++ b/gcc/config/i386/i386.c
@@ -16732,6 +16732,7 @@ ix86_sched_init_global (FILE *, int, int)
case PROCESSOR_NEHALEM:
case PROCESSOR_SANDYBRIDGE:
case PROCESSOR_HASWELL:
+ case PROCESSOR_TREMONT:
case PROCESSOR_GENERIC:
/* Do not perform multipass scheduling for pre-reload schedule
to save compile time. */
diff --git a/gcc/config/i386/x86-tune-sched.c b/gcc/config/i386/x86-tune-sched.c
index 2bcc64b865a..278035eec0b 100644
--- a/gcc/config/i386/x86-tune-sched.c
+++ b/gcc/config/i386/x86-tune-sched.c
@@ -71,6 +71,7 @@ ix86_issue_rate (void)
case PROCESSOR_NEHALEM:
case PROCESSOR_SANDYBRIDGE:
case PROCESSOR_HASWELL:
+ case PROCESSOR_TREMONT:
case PROCESSOR_GENERIC:
return 4;
@@ -430,6 +431,7 @@ ix86_adjust_cost (rtx_insn *insn, int dep_type, rtx_insn *dep_insn, int cost,
case PROCESSOR_NEHALEM:
case PROCESSOR_SANDYBRIDGE:
case PROCESSOR_HASWELL:
+ case PROCESSOR_TREMONT:
case PROCESSOR_GENERIC:
/* Stack engine allows to execute push&pop instructions in parall. */
if ((insn_type == TYPE_PUSH || insn_type == TYPE_POP)
diff --git a/gcc/config/i386/x86-tune.def b/gcc/config/i386/x86-tune.def
index eb057a67750..6bd7087a03f 100644
--- a/gcc/config/i386/x86-tune.def
+++ b/gcc/config/i386/x86-tune.def
@@ -62,7 +62,7 @@ DEF_TUNE (X86_TUNE_PARTIAL_REG_DEPENDENCY, "partial_reg_dependency",
that can be partly masked by careful scheduling of moves. */
DEF_TUNE (X86_TUNE_SSE_PARTIAL_REG_DEPENDENCY, "sse_partial_reg_dependency",
m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_BONNELL | m_AMDFAM10
- | m_BDVER | m_ZNVER | m_GENERIC)
+ | m_BDVER | m_ZNVER | m_TREMONT | m_GENERIC)
/* X86_TUNE_SSE_SPLIT_REGS: Set for machines where the type and dependencies
are resolved on SSE register parts instead of whole registers, so we may
@@ -136,7 +136,7 @@ DEF_TUNE (X86_TUNE_FUSE_ALU_AND_BRANCH, "fuse_alu_and_branch",
DEF_TUNE (X86_TUNE_ACCUMULATE_OUTGOING_ARGS, "accumulate_outgoing_args",
m_PPRO | m_P4_NOCONA | m_BONNELL | m_SILVERMONT | m_KNL | m_KNM | m_INTEL
- | m_GOLDMONT | m_GOLDMONT_PLUS | m_TREMONT | m_ATHLON_K8)
+ | m_GOLDMONT | m_GOLDMONT_PLUS | m_ATHLON_K8)
/* X86_TUNE_PROLOGUE_USING_MOVE: Do not use push/pop in prologues that are
considered on critical path. */
@@ -150,14 +150,15 @@ DEF_TUNE (X86_TUNE_EPILOGUE_USING_MOVE, "epilogue_using_move",
/* X86_TUNE_USE_LEAVE: Use "leave" instruction in epilogues where it fits. */
DEF_TUNE (X86_TUNE_USE_LEAVE, "use_leave",
- m_386 | m_CORE_ALL | m_K6_GEODE | m_AMD_MULTIPLE | m_GENERIC)
+ m_386 | m_CORE_ALL | m_K6_GEODE | m_AMD_MULTIPLE | m_TREMONT
+ | m_GENERIC)
/* X86_TUNE_PUSH_MEMORY: Enable generation of "push mem" instructions.
Some chips, like 486 and Pentium works faster with separate load
and push instructions. */
DEF_TUNE (X86_TUNE_PUSH_MEMORY, "push_memory",
m_386 | m_P4_NOCONA | m_CORE_ALL | m_K6_GEODE | m_AMD_MULTIPLE
- | m_GENERIC)
+ | m_TREMONT | m_GENERIC)
/* X86_TUNE_SINGLE_PUSH: Enable if single push insn is preferred
over esp subtraction. */
@@ -198,8 +199,7 @@ DEF_TUNE (X86_TUNE_PAD_RETURNS, "pad_returns",
than 4 branch instructions in the 16 byte window. */
DEF_TUNE (X86_TUNE_FOUR_JUMP_LIMIT, "four_jump_limit",
m_PPRO | m_P4_NOCONA | m_BONNELL | m_SILVERMONT | m_KNL | m_KNM
- | m_GOLDMONT | m_GOLDMONT_PLUS | m_TREMONT | m_INTEL | m_ATHLON_K8
- | m_AMDFAM10)
+ | m_GOLDMONT | m_GOLDMONT_PLUS | m_INTEL | m_ATHLON_K8 | m_AMDFAM10)
/*****************************************************************************/
/* Integer instruction selection tuning */
@@ -240,11 +240,11 @@ DEF_TUNE (X86_TUNE_INTEGER_DFMODE_MOVES, "integer_dfmode_moves",
/* X86_TUNE_OPT_AGU: Optimize for Address Generation Unit. This flag
will impact LEA instruction selection. */
DEF_TUNE (X86_TUNE_OPT_AGU, "opt_agu", m_BONNELL | m_SILVERMONT | m_KNL
- | m_KNM | m_GOLDMONT | m_GOLDMONT_PLUS | m_TREMONT | m_INTEL)
+ | m_KNM | m_GOLDMONT | m_GOLDMONT_PLUS | m_INTEL)
/* X86_TUNE_AVOID_LEA_FOR_ADDR: Avoid lea for address computation. */
DEF_TUNE (X86_TUNE_AVOID_LEA_FOR_ADDR, "avoid_lea_for_addr",
- m_BONNELL | m_SILVERMONT | m_GOLDMONT | m_GOLDMONT_PLUS | m_TREMONT
+ m_BONNELL | m_SILVERMONT | m_GOLDMONT | m_GOLDMONT_PLUS
| m_KNL | m_KNM)
/* X86_TUNE_SLOW_IMUL_IMM32_MEM: Imul of 32-bit constant and memory is
@@ -263,7 +263,7 @@ DEF_TUNE (X86_TUNE_SLOW_IMUL_IMM8, "slow_imul_imm8",
a conditional move. */
DEF_TUNE (X86_TUNE_AVOID_MEM_OPND_FOR_CMOVE, "avoid_mem_opnd_for_cmove",
m_BONNELL | m_SILVERMONT | m_GOLDMONT | m_GOLDMONT_PLUS | m_KNL
- | m_KNM | m_TREMONT | m_INTEL)
+ | m_KNM | m_INTEL)
/* X86_TUNE_SINGLE_STRINGOP: Enable use of single string operations, such
as MOVS and STOS (without a REP prefix) to move/set sequences of bytes. */
@@ -282,7 +282,8 @@ DEF_TUNE (X86_TUNE_PREFER_KNOWN_REP_MOVSB_STOSB,
FIXME: This may actualy be a win on more targets than listed here. */
DEF_TUNE (X86_TUNE_MISALIGNED_MOVE_STRING_PRO_EPILOGUES,
"misaligned_move_string_pro_epilogues",
- m_386 | m_486 | m_CORE_ALL | m_AMD_MULTIPLE | m_GENERIC)
+ m_386 | m_486 | m_CORE_ALL | m_AMD_MULTIPLE | m_TREMONT
+ | m_GENERIC)
/* X86_TUNE_USE_SAHF: Controls use of SAHF. */
DEF_TUNE (X86_TUNE_USE_SAHF, "use_sahf",
@@ -294,7 +295,7 @@ DEF_TUNE (X86_TUNE_USE_SAHF, "use_sahf",
/* X86_TUNE_USE_CLTD: Controls use of CLTD and CTQO instructions. */
DEF_TUNE (X86_TUNE_USE_CLTD, "use_cltd",
~(m_PENT | m_LAKEMONT | m_BONNELL | m_SILVERMONT | m_KNL | m_KNM | m_INTEL
- | m_K6 | m_GOLDMONT | m_GOLDMONT_PLUS | m_TREMONT))
+ | m_K6 | m_GOLDMONT | m_GOLDMONT_PLUS))
/* X86_TUNE_USE_BT: Enable use of BT (bit test) instructions. */
DEF_TUNE (X86_TUNE_USE_BT, "use_bt",
@@ -305,7 +306,7 @@ DEF_TUNE (X86_TUNE_USE_BT, "use_bt",
/* X86_TUNE_AVOID_FALSE_DEP_FOR_BMI: Avoid false dependency
for bit-manipulation instructions. */
DEF_TUNE (X86_TUNE_AVOID_FALSE_DEP_FOR_BMI, "avoid_false_dep_for_bmi",
- m_SANDYBRIDGE | m_CORE_AVX2 | m_GENERIC)
+ m_SANDYBRIDGE | m_CORE_AVX2 | m_TREMONT | m_GENERIC)
/* X86_TUNE_ADJUST_UNROLL: This enables adjusting the unroll factor based
on hardware capabilities. Bdver3 hardware has a loop buffer which makes
@@ -321,14 +322,14 @@ DEF_TUNE (X86_TUNE_ONE_IF_CONV_INSN, "one_if_conv_insn",
/* X86_TUNE_AVOID_MFENCE: Use lock prefixed instructions instead of mfence. */
DEF_TUNE (X86_TUNE_AVOID_MFENCE, "avoid_mfence",
- m_CORE_ALL | m_BDVER | m_ZNVER | m_GENERIC)
+ m_CORE_ALL | m_BDVER | m_ZNVER | m_TREMONT | m_GENERIC)
/* X86_TUNE_EXPAND_ABS: This enables a new abs pattern by
generating instructions for abs (x) = (((signed) x >> (W-1) ^ x) -
(signed) x >> (W-1)) instead of cmove or SSE max/abs instructions. */
DEF_TUNE (X86_TUNE_EXPAND_ABS, "expand_abs",
m_CORE_ALL | m_SILVERMONT | m_KNL | m_KNM | m_GOLDMONT
- | m_GOLDMONT_PLUS | m_TREMONT )
+ | m_GOLDMONT_PLUS)
/*****************************************************************************/
/* 387 instruction selection tuning */
@@ -386,13 +387,13 @@ DEF_TUNE (X86_TUNE_SSE_PACKED_SINGLE_INSN_OPTIMAL, "sse_packed_single_insn_optim
/* X86_TUNE_SSE_TYPELESS_STORES: Always movaps/movups for 128bit stores. */
DEF_TUNE (X86_TUNE_SSE_TYPELESS_STORES, "sse_typeless_stores",
- m_AMD_MULTIPLE | m_CORE_ALL | m_GENERIC)
+ m_AMD_MULTIPLE | m_CORE_ALL | m_TREMONT | m_GENERIC)
/* X86_TUNE_SSE_LOAD0_BY_PXOR: Always use pxor to load0 as opposed to
xorps/xorpd and other variants. */
DEF_TUNE (X86_TUNE_SSE_LOAD0_BY_PXOR, "sse_load0_by_pxor",
m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_BDVER | m_BTVER | m_ZNVER
- | m_GENERIC)
+ | m_TREMONT | m_GENERIC)
/* X86_TUNE_INTER_UNIT_MOVES_TO_VEC: Enable moves in from integer
to SSE registers. If disabled, the moves will be done by storing
@@ -419,7 +420,7 @@ DEF_TUNE (X86_TUNE_INTER_UNIT_CONVERSIONS, "inter_unit_conversions",
fp converts to destination register. */
DEF_TUNE (X86_TUNE_SPLIT_MEM_OPND_FOR_FP_CONVERTS, "split_mem_opnd_for_fp_converts",
m_SILVERMONT | m_KNL | m_KNM | m_GOLDMONT | m_GOLDMONT_PLUS
- | m_TREMONT | m_INTEL)
+ | m_INTEL)
/* X86_TUNE_USE_VECTOR_FP_CONVERTS: Prefer vector packed SSE conversion
from FP to FP. This form of instructions avoids partial write to the
@@ -434,7 +435,7 @@ DEF_TUNE (X86_TUNE_USE_VECTOR_CONVERTS, "use_vector_converts", m_AMDFAM10)
/* X86_TUNE_SLOW_SHUFB: Indicates tunings with slow pshufb instruction. */
DEF_TUNE (X86_TUNE_SLOW_PSHUFB, "slow_pshufb",
m_BONNELL | m_SILVERMONT | m_KNL | m_KNM | m_GOLDMONT
- | m_GOLDMONT_PLUS | m_TREMONT | m_INTEL)
+ | m_GOLDMONT_PLUS | m_INTEL)
/* X86_TUNE_AVOID_4BYTE_PREFIXES: Avoid instructions requiring 4+ bytes of prefixes. */
DEF_TUNE (X86_TUNE_AVOID_4BYTE_PREFIXES, "avoid_4byte_prefixes",
--
2.18.2

View File

@ -0,0 +1,196 @@
From 80c2ed8228817fb6438120997227811a746272ba Mon Sep 17 00:00:00 2001
From: "H.J. Lu" <hjl.tools@gmail.com>
Date: Wed, 15 Sep 2021 14:17:08 +0800
Subject: [PATCH 2/3] x86: Update memcpy/memset inline strategies for
-mtune=tremont
Simply memcpy and memset inline strategies to avoid branches for
-mtune=tremont:
1. Create Tremont cost model from generic cost model.
2. With MOVE_RATIO and CLEAR_RATIO == 17, GCC will use integer/vector
load and store for up to 16 * 16 (256) bytes when the data size is
fixed and known.
3. Inline only if data size is known to be <= 256.
a. Use "rep movsb/stosb" with simple code sequence if the data size
is a constant.
b. Use loop if data size is not a constant.
4. Use memcpy/memset libray function if data size is unknown or > 256.
* config/i386/i386-options.c (processor_cost_table): Use
tremont_cost for Tremont.
* config/i386/x86-tune-costs.h (tremont_memcpy): New.
(tremont_memset): Likewise.
(tremont_cost): Likewise.
* config/i386/x86-tune.def (X86_TUNE_PREFER_KNOWN_REP_MOVSB_STOSB):
Enable for Tremont.
---
gcc/config/i386/i386-options.c | 2 +-
gcc/config/i386/x86-tune-costs.h | 124 +++++++++++++++++++++++++++++++
gcc/config/i386/x86-tune.def | 2 +-
3 files changed, 126 insertions(+), 2 deletions(-)
diff --git a/gcc/config/i386/i386-options.c b/gcc/config/i386/i386-options.c
index 19632b5fd6b..4b77d62926f 100644
--- a/gcc/config/i386/i386-options.c
+++ b/gcc/config/i386/i386-options.c
@@ -719,7 +719,7 @@ static const struct processor_costs *processor_cost_table[] =
&slm_cost,
&slm_cost,
&slm_cost,
- &slm_cost,
+ &tremont_cost,
&slm_cost,
&slm_cost,
&skylake_cost,
diff --git a/gcc/config/i386/x86-tune-costs.h b/gcc/config/i386/x86-tune-costs.h
index ffe810f2bcb..93644be9cb3 100644
--- a/gcc/config/i386/x86-tune-costs.h
+++ b/gcc/config/i386/x86-tune-costs.h
@@ -2734,6 +2734,130 @@ struct processor_costs slm_cost = {
"16", /* Func alignment. */
};
+static stringop_algs tremont_memcpy[2] = {
+ {libcall,
+ {{256, rep_prefix_1_byte, true},
+ {256, loop, false},
+ {-1, libcall, false}}},
+ {libcall,
+ {{256, rep_prefix_1_byte, true},
+ {256, loop, false},
+ {-1, libcall, false}}}};
+static stringop_algs tremont_memset[2] = {
+ {libcall,
+ {{256, rep_prefix_1_byte, true},
+ {256, loop, false},
+ {-1, libcall, false}}},
+ {libcall,
+ {{256, rep_prefix_1_byte, true},
+ {256, loop, false},
+ {-1, libcall, false}}}};
+static const
+struct processor_costs tremont_cost = {
+ {
+ /* Start of register allocator costs. integer->integer move cost is 2. */
+ 6, /* cost for loading QImode using movzbl */
+ {6, 6, 6}, /* cost of loading integer registers
+ in QImode, HImode and SImode.
+ Relative to reg-reg move (2). */
+ {6, 6, 6}, /* cost of storing integer registers */
+ 4, /* cost of reg,reg fld/fst */
+ {6, 6, 12}, /* cost of loading fp registers
+ in SFmode, DFmode and XFmode */
+ {6, 6, 12}, /* cost of storing fp registers
+ in SFmode, DFmode and XFmode */
+ 2, /* cost of moving MMX register */
+ {6, 6}, /* cost of loading MMX registers
+ in SImode and DImode */
+ {6, 6}, /* cost of storing MMX registers
+ in SImode and DImode */
+ 2, 3, 4, /* cost of moving XMM,YMM,ZMM register */
+ {6, 6, 6, 10, 15}, /* cost of loading SSE registers
+ in 32,64,128,256 and 512-bit */
+ {6, 6, 6, 10, 15}, /* cost of storing SSE registers
+ in 32,64,128,256 and 512-bit */
+ 6, 6, /* SSE->integer and integer->SSE moves */
+ 6, 6, /* mask->integer and integer->mask moves */
+ {6, 6, 6}, /* cost of loading mask register
+ in QImode, HImode, SImode. */
+ {6, 6, 6}, /* cost if storing mask register
+ in QImode, HImode, SImode. */
+ 2, /* cost of moving mask register. */
+ /* End of register allocator costs. */
+ },
+
+ COSTS_N_INSNS (1), /* cost of an add instruction */
+ /* Setting cost to 2 makes our current implementation of synth_mult result in
+ use of unnecessary temporary registers causing regression on several
+ SPECfp benchmarks. */
+ COSTS_N_INSNS (1) + 1, /* cost of a lea instruction */
+ COSTS_N_INSNS (1), /* variable shift costs */
+ COSTS_N_INSNS (1), /* constant shift costs */
+ {COSTS_N_INSNS (3), /* cost of starting multiply for QI */
+ COSTS_N_INSNS (4), /* HI */
+ COSTS_N_INSNS (3), /* SI */
+ COSTS_N_INSNS (4), /* DI */
+ COSTS_N_INSNS (4)}, /* other */
+ 0, /* cost of multiply per each bit set */
+ {COSTS_N_INSNS (16), /* cost of a divide/mod for QI */
+ COSTS_N_INSNS (22), /* HI */
+ COSTS_N_INSNS (30), /* SI */
+ COSTS_N_INSNS (74), /* DI */
+ COSTS_N_INSNS (74)}, /* other */
+ COSTS_N_INSNS (1), /* cost of movsx */
+ COSTS_N_INSNS (1), /* cost of movzx */
+ 8, /* "large" insn */
+ 17, /* MOVE_RATIO */
+ 17, /* CLEAR_RATIO */
+ {6, 6, 6}, /* cost of loading integer registers
+ in QImode, HImode and SImode.
+ Relative to reg-reg move (2). */
+ {6, 6, 6}, /* cost of storing integer registers */
+ {6, 6, 6, 10, 15}, /* cost of loading SSE register
+ in 32bit, 64bit, 128bit, 256bit and 512bit */
+ {6, 6, 6, 10, 15}, /* cost of storing SSE register
+ in 32bit, 64bit, 128bit, 256bit and 512bit */
+ {6, 6, 6, 10, 15}, /* cost of unaligned loads. */
+ {6, 6, 6, 10, 15}, /* cost of unaligned storess. */
+ 2, 3, 4, /* cost of moving XMM,YMM,ZMM register */
+ 6, /* cost of moving SSE register to integer. */
+ 18, 6, /* Gather load static, per_elt. */
+ 18, 6, /* Gather store static, per_elt. */
+ 32, /* size of l1 cache. */
+ 512, /* size of l2 cache. */
+ 64, /* size of prefetch block */
+ 6, /* number of parallel prefetches */
+ /* Benchmarks shows large regressions on K8 sixtrack benchmark when this
+ value is increased to perhaps more appropriate value of 5. */
+ 3, /* Branch cost */
+ COSTS_N_INSNS (3), /* cost of FADD and FSUB insns. */
+ COSTS_N_INSNS (5), /* cost of FMUL instruction. */
+ COSTS_N_INSNS (17), /* cost of FDIV instruction. */
+ COSTS_N_INSNS (1), /* cost of FABS instruction. */
+ COSTS_N_INSNS (1), /* cost of FCHS instruction. */
+ COSTS_N_INSNS (14), /* cost of FSQRT instruction. */
+
+ COSTS_N_INSNS (1), /* cost of cheap SSE instruction. */
+ COSTS_N_INSNS (3), /* cost of ADDSS/SD SUBSS/SD insns. */
+ COSTS_N_INSNS (4), /* cost of MULSS instruction. */
+ COSTS_N_INSNS (5), /* cost of MULSD instruction. */
+ COSTS_N_INSNS (5), /* cost of FMA SS instruction. */
+ COSTS_N_INSNS (5), /* cost of FMA SD instruction. */
+ COSTS_N_INSNS (13), /* cost of DIVSS instruction. */
+ COSTS_N_INSNS (17), /* cost of DIVSD instruction. */
+ COSTS_N_INSNS (14), /* cost of SQRTSS instruction. */
+ COSTS_N_INSNS (18), /* cost of SQRTSD instruction. */
+ 1, 4, 3, 3, /* reassoc int, fp, vec_int, vec_fp. */
+ tremont_memcpy,
+ tremont_memset,
+ COSTS_N_INSNS (4), /* cond_taken_branch_cost. */
+ COSTS_N_INSNS (2), /* cond_not_taken_branch_cost. */
+ "16:11:8", /* Loop alignment. */
+ "16:11:8", /* Jump alignment. */
+ "0:0:8", /* Label alignment. */
+ "16", /* Func alignment. */
+};
+
static stringop_algs intel_memcpy[2] = {
{libcall, {{11, loop, false}, {-1, rep_prefix_4_byte, false}}},
{libcall, {{32, loop, false}, {64, rep_prefix_4_byte, false},
diff --git a/gcc/config/i386/x86-tune.def b/gcc/config/i386/x86-tune.def
index 6bd7087a03f..636e0c788bf 100644
--- a/gcc/config/i386/x86-tune.def
+++ b/gcc/config/i386/x86-tune.def
@@ -273,7 +273,7 @@ DEF_TUNE (X86_TUNE_SINGLE_STRINGOP, "single_stringop", m_386 | m_P4_NOCONA)
move/set sequences of bytes with known size. */
DEF_TUNE (X86_TUNE_PREFER_KNOWN_REP_MOVSB_STOSB,
"prefer_known_rep_movsb_stosb",
- m_SKYLAKE | m_ALDERLAKE | m_CORE_AVX512)
+ m_SKYLAKE | m_ALDERLAKE | m_TREMONT | m_CORE_AVX512)
/* X86_TUNE_MISALIGNED_MOVE_STRING_PRO_EPILOGUES: Enable generation of
compact prologues and epilogues by issuing a misaligned moves. This
--
2.18.2

View File

@ -0,0 +1,185 @@
From c4c7244349999f91ef2a7cd2108eee0372490be9 Mon Sep 17 00:00:00 2001
From: "H.J. Lu" <hjl.tools@gmail.com>
Date: Wed, 15 Sep 2021 14:18:21 +0800
Subject: [PATCH 3/3] x86: Add TARGET_SSE_PARTIAL_REG_[FP_]CONVERTS_DEPENDENCY
1. Replace TARGET_SSE_PARTIAL_REG_DEPENDENCY with
TARGET_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY in SSE FP to FP splitters.
2. Replace TARGET_SSE_PARTIAL_REG_DEPENDENCY with
TARGET_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY in SSE INT to FP splitters.
gcc/
* config/i386/i386.h (TARGET_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY):
New.
(TARGET_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY): Likewise.
* config/i386/i386.md (SSE FP to FP splitters): Replace
TARGET_SSE_PARTIAL_REG_DEPENDENCY with
TARGET_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY.
(SSE INT to FP splitter): Replace TARGET_SSE_PARTIAL_REG_DEPENDENCY
with TARGET_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY.
* config/i386/x86-tune.def
(X86_TUNE_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY): New.
(X86_TUNE_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY): Likewise.
gcc/testsuite/
* gcc.target/i386/sse-covert-1.c: Likewise.
* gcc.target/i386/sse-fp-covert-1.c: Likewise.
* gcc.target/i386/sse-int-covert-1.c: Likewise.
---
gcc/config/i386/i386.h | 4 ++++
gcc/config/i386/i386.md | 9 ++++++---
gcc/config/i386/x86-tune.def | 15 +++++++++++++++
gcc/testsuite/gcc.target/i386/sse-covert-1.c | 19 +++++++++++++++++++
.../gcc.target/i386/sse-fp-covert-1.c | 15 +++++++++++++++
.../gcc.target/i386/sse-int-covert-1.c | 14 ++++++++++++++
6 files changed, 73 insertions(+), 3 deletions(-)
create mode 100644 gcc/testsuite/gcc.target/i386/sse-covert-1.c
create mode 100644 gcc/testsuite/gcc.target/i386/sse-fp-covert-1.c
create mode 100644 gcc/testsuite/gcc.target/i386/sse-int-covert-1.c
diff --git a/gcc/config/i386/i386.h b/gcc/config/i386/i386.h
index 73e118900f7..5b992195df7 100644
--- a/gcc/config/i386/i386.h
+++ b/gcc/config/i386/i386.h
@@ -553,6 +553,10 @@ extern unsigned char ix86_tune_features[X86_TUNE_LAST];
ix86_tune_features[X86_TUNE_PARTIAL_REG_DEPENDENCY]
#define TARGET_SSE_PARTIAL_REG_DEPENDENCY \
ix86_tune_features[X86_TUNE_SSE_PARTIAL_REG_DEPENDENCY]
+#define TARGET_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY \
+ ix86_tune_features[X86_TUNE_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY]
+#define TARGET_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY \
+ ix86_tune_features[X86_TUNE_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY]
#define TARGET_SSE_UNALIGNED_LOAD_OPTIMAL \
ix86_tune_features[X86_TUNE_SSE_UNALIGNED_LOAD_OPTIMAL]
#define TARGET_SSE_UNALIGNED_STORE_OPTIMAL \
diff --git a/gcc/config/i386/i386.md b/gcc/config/i386/i386.md
index 97325e38676..053bec1c1e1 100644
--- a/gcc/config/i386/i386.md
+++ b/gcc/config/i386/i386.md
@@ -4378,7 +4378,8 @@
(float_extend:DF
(match_operand:SF 1 "nonimmediate_operand")))]
"!TARGET_AVX
- && TARGET_SSE_PARTIAL_REG_DEPENDENCY && epilogue_completed
+ && TARGET_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY
+ && epilogue_completed
&& optimize_function_for_speed_p (cfun)
&& (!REG_P (operands[1])
|| (!TARGET_AVX && REGNO (operands[0]) != REGNO (operands[1])))
@@ -4540,7 +4541,8 @@
(float_truncate:SF
(match_operand:DF 1 "nonimmediate_operand")))]
"!TARGET_AVX
- && TARGET_SSE_PARTIAL_REG_DEPENDENCY && epilogue_completed
+ && TARGET_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY
+ && epilogue_completed
&& optimize_function_for_speed_p (cfun)
&& (!REG_P (operands[1])
|| (!TARGET_AVX && REGNO (operands[0]) != REGNO (operands[1])))
@@ -5053,7 +5055,8 @@
[(set (match_operand:MODEF 0 "sse_reg_operand")
(float:MODEF (match_operand:SWI48 1 "nonimmediate_operand")))]
"!TARGET_AVX
- && TARGET_SSE_PARTIAL_REG_DEPENDENCY && epilogue_completed
+ && TARGET_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY
+ && epilogue_completed
&& optimize_function_for_speed_p (cfun)
&& (!EXT_REX_SSE_REG_P (operands[0])
|| TARGET_AVX512VL)"
diff --git a/gcc/config/i386/x86-tune.def b/gcc/config/i386/x86-tune.def
index 636e0c788bf..b5166fb1316 100644
--- a/gcc/config/i386/x86-tune.def
+++ b/gcc/config/i386/x86-tune.def
@@ -64,6 +64,21 @@ DEF_TUNE (X86_TUNE_SSE_PARTIAL_REG_DEPENDENCY, "sse_partial_reg_dependency",
m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_BONNELL | m_AMDFAM10
| m_BDVER | m_ZNVER | m_TREMONT | m_GENERIC)
+/* X86_TUNE_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY: This knob avoids
+ partial write to the destination in scalar SSE conversion from FP
+ to FP. */
+DEF_TUNE (X86_TUNE_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY,
+ "sse_partial_reg_fp_converts_dependency",
+ m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_BONNELL | m_AMDFAM10
+ | m_BDVER | m_ZNVER | m_GENERIC)
+
+/* X86_TUNE_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY: This knob avoids partial
+ write to the destination in scalar SSE conversion from integer to FP. */
+DEF_TUNE (X86_TUNE_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY,
+ "sse_partial_reg_converts_dependency",
+ m_PPRO | m_P4_NOCONA | m_CORE_ALL | m_BONNELL | m_AMDFAM10
+ | m_BDVER | m_ZNVER | m_GENERIC)
+
/* X86_TUNE_SSE_SPLIT_REGS: Set for machines where the type and dependencies
are resolved on SSE register parts instead of whole registers, so we may
maintain just lower part of scalar values in proper format leaving the
diff --git a/gcc/testsuite/gcc.target/i386/sse-covert-1.c b/gcc/testsuite/gcc.target/i386/sse-covert-1.c
new file mode 100644
index 00000000000..c30af694505
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse-covert-1.c
@@ -0,0 +1,19 @@
+/* { dg-do compile } */
+/* { dg-options "-O2 -march=x86-64 -mfpmath=sse -mtune-ctrl=^sse_partial_reg_fp_converts_dependency,^sse_partial_reg_converts_dependency" } */
+
+extern float f;
+extern double d;
+extern int i;
+
+void
+foo (void)
+{
+ d = f;
+ f = i;
+}
+
+/* { dg-final { scan-assembler "cvtss2sd" } } */
+/* { dg-final { scan-assembler "cvtsi2ssl" } } */
+/* { dg-final { scan-assembler-not "cvtps2pd" } } */
+/* { dg-final { scan-assembler-not "cvtdq2ps" } } */
+/* { dg-final { scan-assembler-not "pxor" } } */
diff --git a/gcc/testsuite/gcc.target/i386/sse-fp-covert-1.c b/gcc/testsuite/gcc.target/i386/sse-fp-covert-1.c
new file mode 100644
index 00000000000..b6567e60e3e
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse-fp-covert-1.c
@@ -0,0 +1,15 @@
+/* { dg-do compile } */
+/* { dg-options "-O2 -march=x86-64 -mfpmath=sse -mtune-ctrl=^sse_partial_reg_fp_converts_dependency" } */
+
+extern float f;
+extern double d;
+
+void
+foo (void)
+{
+ d = f;
+}
+
+/* { dg-final { scan-assembler "cvtss2sd" } } */
+/* { dg-final { scan-assembler-not "cvtps2pd" } } */
+/* { dg-final { scan-assembler-not "pxor" } } */
diff --git a/gcc/testsuite/gcc.target/i386/sse-int-covert-1.c b/gcc/testsuite/gcc.target/i386/sse-int-covert-1.c
new file mode 100644
index 00000000000..107f7241def
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse-int-covert-1.c
@@ -0,0 +1,14 @@
+/* { dg-do compile } */
+/* { dg-options "-O2 -march=x86-64 -mfpmath=sse -mtune-ctrl=^sse_partial_reg_converts_dependency" } */
+
+extern float f;
+extern int i;
+
+void
+foo (void)
+{
+ f = i;
+}
+
+/* { dg-final { scan-assembler "cvtsi2ssl" } } */
+/* { dg-final { scan-assembler-not "pxor" } } */
--
2.18.2

View File

@ -2,13 +2,13 @@
%{?scl:%global __strip %%{_scl_root}/usr/bin/strip}
%{?scl:%global __objdump %%{_scl_root}/usr/bin/objdump}
%{?scl:%scl_package gcc}
%global DATE 20210728
%global gitrev 134ab8155c937122663513b76afa8e64ad61fe99
%global DATE 20220127
%global gitrev 2fa6e5c54e782377faa4c9c1f0e0b16db27f266c
%global gcc_version 11.2.1
%global gcc_major 11
# Note, gcc_release must be integer, if you want to add suffixes to
# %%{release}, append them after %%{gcc_release} on Release: line.
%global gcc_release 1
%global gcc_release 9
%global nvptx_tools_gitrev 5f6f343a302d620b0868edab376c00b15741e39e
%global newlib_cygwin_gitrev 50e2a63b04bdd018484605fbb954fd1bd5147fa0
%global mpc_version 1.0.3
@ -18,6 +18,7 @@
%global doxygen_version 1.8.0
%global _unpackaged_files_terminate_build 0
%if 0%{?fedora} > 27 || 0%{?rhel} > 7
# Until annobin is fixed (#1519165).
%undefine _annotated_build
%endif
# Strip will fail on nvptx-none *.a archives and the brp-* scripts will
@ -28,7 +29,11 @@
%if 0%{?__brp_strip_lto:1}
%global __brp_strip_lto %{__brp_strip_lto} || :
%endif
%if 0%{?fedora} < 32 && 0%{?rhel} < 8
%global multilib_64_archs sparc64 ppc64 ppc64p7 s390x x86_64
%else
%global multilib_64_archs sparc64 ppc64 ppc64p7 x86_64
%endif
%if 0%{?rhel} > 7
%global build_ada 0
%global build_objc 0
@ -133,7 +138,7 @@
Summary: GCC version 11
Name: %{?scl_prefix}gcc
Version: %{gcc_version}
Release: %{gcc_release}.2%{?dist}
Release: %{gcc_release}.1%{?dist}
# libgcc, libgfortran, libgomp, libstdc++ and crtstuff have
# GCC Runtime Exception.
License: GPLv3+ and GPLv3+ with exceptions and GPLv2+ with exceptions and LGPLv2+ and BSD
@ -193,7 +198,7 @@ BuildRequires: texinfo, texinfo-tex, /usr/bin/pod2man
#BuildRequires: gmp-devel >= 4.1.2-8, mpfr-devel >= 3.1.0, libmpc-devel >= 0.8.1
#BuildRequires: python3-devel, /usr/bin/python
BuildRequires: gcc, gcc-c++, make
%if 0%{?rhel} <= 7
%if 0%{?rhel} == 7
BuildRequires: python3
%endif
# For VTA guality testing
@ -333,7 +338,12 @@ Patch9: gcc11-foffload-default.patch
Patch10: gcc11-Wno-format-security.patch
Patch11: gcc11-rh1574936.patch
Patch12: gcc11-d-shared-libphobos.patch
Patch13: gcc11-add-Wbidirectional.patch
Patch14: gcc11-libgcc-link.patch
Patch15: gcc11-pr101786.patch
Patch16: gcc11-stringify-__VA_OPT__.patch
Patch17: gcc11-stringify-__VA_OPT__-2.patch
Patch18: gcc11-Wbidi-chars.patch
Patch19: gcc11-dejagnu-multiline.patch
Patch100: gcc11-fortran-fdec-duplicates.patch
Patch101: gcc11-fortran-flogical-as-integer.patch
@ -346,6 +356,11 @@ Patch107: gcc11-fortran-fdec-promotion.patch
Patch108: gcc11-fortran-fdec-sequence.patch
Patch109: gcc11-fortran-fdec-add-missing-indexes.patch
Patch200: gcc11-tremont1.patch
Patch201: gcc11-tremont2.patch
Patch202: gcc11-tremont3.patch
Patch203: gcc11-mtune-alderlake.patch
Patch1000: gcc11-libstdc++-compat.patch
Patch1001: gcc11-alt-compat-test.patch
Patch1002: gcc11-libgfortran-compat.patch
@ -354,6 +369,22 @@ Patch2001: doxygen-1.7.1-config.patch
Patch2002: doxygen-1.7.5-timestamp.patch
Patch2003: doxygen-1.8.0-rh856725.patch
Patch3000: 0001-basic_string-reserve-n-semantics-are-not-available-i.patch
Patch3001: 0004-operator-istream-char-N-eofbit-fixes-are-not-availab.patch
Patch3002: 0005-Disable-tests-for-PR-libstdc-79820-and-PR-libstdc-81.patch
Patch3003: 0006-Don-t-assume-has_facet-codecvt_c16-when-run-against-.patch
Patch3004: 0008-testsuite-build-plugins-with-std-c-11.patch
Patch3005: 0009-Fix-22_locale-locale-cons-unicode.cc-when-run-under-.patch
Patch3006: 0010-Don-t-verify-exception-handling-in-basic_filebuf-clo.patch
Patch3007: 0011-Add-dts.exp-and-use-it-to-fix-22_locale-messages-136.patch
Patch3008: 0012-dts.exp-use-usr-bin-gcc.patch
Patch3009: 0013-Rename-__CXXSTDLIB_SO_VERSION__-to-__LIBSTDCXX_SO_VE.patch
Patch3010: 0014-Conditionalize-tests-for-PR-libstdc-98466-on-__LIBST.patch
Patch3011: 0015-Conditionalize-test-for-PR-libstdc-87135-on-__LIBSTD.patch
Patch3012: 0016-Conditionalize-test-for-hashtable-bucket-sizes-on-__.patch
Patch3013: 0017-Conditionalize-test-for-PR-libstdc-71181-on-__LIBSTD.patch
Patch3014: gcc11-dg-ice-fixes.patch
%if 0%{?rhel} > 7
%global nonsharedver 80
%else
@ -682,7 +713,12 @@ to NVidia PTX capable devices if available.
%patch11 -p0 -b .rh1574936~
%endif
%patch12 -p0 -b .d-shared-libphobos~
%patch13 -p1 -b .bidi~
%patch14 -p0 -b .libgcc-link~
%patch15 -p0 -b .pr101786~
%patch16 -p0 -b .stringify-__VA_OPT__~
%patch17 -p0 -b .stringify-__VA_OPT__-2~
%patch18 -p1 -b .bidi~
%patch19 -p1 -b .dejagnu-multiline~
%patch100 -p1 -b .fortran-fdec-duplicates~
%patch101 -p1 -b .fortran-flogical-as-integer~
@ -695,6 +731,11 @@ to NVidia PTX capable devices if available.
%patch108 -p1 -b .fortran-fdec-sequence~
%patch109 -p1 -b .fortran-fdec-add-missing-indexes~
%patch200 -p1 -b .tremont1~
%patch201 -p1 -b .tremont2~
%patch202 -p1 -b .tremont3~
%patch203 -p1 -b .alderlake~
%patch1000 -p0 -b .libstdc++-compat~
%ifarch %{ix86} x86_64
%if 0%{?rhel} < 7
@ -715,6 +756,22 @@ cd ..
%endif
%endif
# Apply DTS-specific testsuite patches.
%patch3000 -p1 -b .dts-test-0~
%patch3001 -p1 -b .dts-test-1~
%patch3002 -p1 -b .dts-test-2~
%patch3003 -p1 -b .dts-test-3~
%patch3004 -p1 -b .dts-test-4~
%patch3005 -p1 -b .dts-test-5~
%patch3006 -p1 -b .dts-test-6~
%patch3007 -p1 -b .dts-test-7~
%patch3008 -p1 -b .dts-test-8~
%patch3009 -p1 -b .dts-test-9~
%patch3010 -p1 -b .dts-test-10~
%patch3011 -p1 -b .dts-test-11~
%patch3012 -p1 -b .dts-test-12~
%patch3013 -p1 -b .dts-test-13~
%patch3014 -p1 -b .dts-test-14~
find gcc/testsuite -name \*.pr96939~ | xargs rm -f
@ -1052,7 +1109,7 @@ CONFIGURE_OPTS="\
%ifnarch sparc sparcv9 ppc
--build=%{gcc_target_platform} \
%endif
%if 0%{?fedora} >= 35
%if 0%{?fedora} >= 35 || 0%{?rhel} >= 9
%ifarch x86_64 %{ix86} ppc64le s390x
--with-build-config=bootstrap-lto --enable-link-serialization=1 \
%endif
@ -1068,9 +1125,9 @@ CC="$CC" CXX="$CXX" CFLAGS="$OPT_FLAGS" \
$CONFIGURE_OPTS
%ifarch sparc sparcv9 sparc64
make %{?_smp_mflags} BOOT_CFLAGS="$OPT_FLAGS" bootstrap
make %{?_smp_mflags} BOOT_CFLAGS="$OPT_FLAGS" LDFLAGS_FOR_TARGET=-Wl,-z,relro,-z,now bootstrap
%else
make %{?_smp_mflags} BOOT_CFLAGS="$OPT_FLAGS" profiledbootstrap
make %{?_smp_mflags} BOOT_CFLAGS="$OPT_FLAGS" LDFLAGS_FOR_TARGET=-Wl,-z,relro,-z,now profiledbootstrap
%endif
%if 0%{?rhel} <= 8
@ -1784,7 +1841,8 @@ ln -sf lib64/libasan_preinit.o libasan_preinit.o
%endif
%if %{build_libubsan}
ln -sf ../lib32/libubsan.a 32/libubsan.a
ln -sf lib64/libubsan.a libubsan.a
# BZ #2027391
mv -f lib64/libubsan.a libubsan.a
%endif
%else
%ifarch %{multilib_64_archs}
@ -2288,6 +2346,7 @@ fi
%{_prefix}/lib/gcc/%{gcc_target_platform}/%{gcc_major}/include/hresetintrin.h
%{_prefix}/lib/gcc/%{gcc_target_platform}/%{gcc_major}/include/keylockerintrin.h
%{_prefix}/lib/gcc/%{gcc_target_platform}/%{gcc_major}/include/avxvnniintrin.h
%{_prefix}/lib/gcc/%{gcc_target_platform}/%{gcc_major}/include/mwaitintrin.h
%endif
%ifarch ia64
%{_prefix}/lib/gcc/%{gcc_target_platform}/%{gcc_major}/include/ia64intrin.h
@ -2833,8 +2892,91 @@ fi
%endif
%changelog
* Fri Oct 29 2021 Marek Polacek <polacek@redhat.com> 11.2.1-1.2
- add -Wbidirectional patch (#2017819)
* Wed Feb 2 2022 Marek Polacek <polacek@redhat.com> 11.2.1-9.1
- avoid overly-greedy match in dejagnu regexp (#2049712)
* Fri Jan 28 2022 Marek Polacek <polacek@redhat.com> 11.2.1-9
- update from releases/gcc-11-branch (#2047286)
- PRs fortran/104127, fortran/104212, fortran/104227, target/101529
- fix up va-opt-6.c testcase
* Fri Jan 28 2022 Marek Polacek <polacek@redhat.com> 11.2.1-8
- update from releases/gcc-11-branch (#2047286)
- PRs ada/103538, analyzer/101962, bootstrap/103688, c++/85846, c++/95009,
c++/98394, c++/99911, c++/100493, c++/101715, c++/102229, c++/102933,
c++/103012, c++/103198, c++/103480, c++/103703, c++/103714,
c++/103758, c++/103783, c++/103831, c++/103912, c++/104055, c/97548,
c/101289, c/101537, c/103587, c/103881, d/103604, debug/103838,
debug/103874, fortran/67804, fortran/83079, fortran/101329,
fortran/101762, fortran/102332, fortran/102717, fortran/102787,
fortran/103411, fortran/103412, fortran/103418, fortran/103473,
fortran/103505, fortran/103588, fortran/103591, fortran/103606,
fortran/103607, fortran/103609, fortran/103610, fortran/103692,
fortran/103717, fortran/103718, fortran/103719, fortran/103776,
fortran/103777, fortran/103778, fortran/103782, fortran/103789,
ipa/101354, jit/103562, libfortran/103634, libstdc++/100017,
libstdc++/102994, libstdc++/103453, libstdc++/103501,
libstdc++/103549, libstdc++/103877, libstdc++/103919,
middle-end/101751, middle-end/102860, middle-end/103813, objc/103639,
preprocessor/89971, preprocessor/102432, rtl-optimization/102478,
rtl-optimization/103837, rtl-optimization/103860,
rtl-optimization/103908, sanitizer/102911, target/102347,
target/103465, target/103661, target/104172, target/104188,
tree-optimization/101615, tree-optimization/103523,
tree-optimization/103603, tree-optimization/103995
* Wed Jan 5 2022 Marek Polacek <polacek@redhat.com> 11.2.1-7.2
- fix dg-ice tests (#2037072)
* Fri Dec 10 2021 Marek Polacek <polacek@redhat.com> 11.2.1-7.1
- update Intel Tremont tuning patches (#2014276)
- backport Intel Alderlake tuning (#2023553)
* Tue Dec 7 2021 Marek Polacek <polacek@redhat.com> 11.2.1-7
- update from releases/gcc-11-branch (#1996862)
- PRs ada/100486, c++/70796, c++/92746, c++/93286, c++/94490, c++/102642,
c++/102786, debug/101378, debug/103046, debug/103315, fortran/87711,
fortran/87851, fortran/97896, fortran/99061, fortran/99348,
fortran/102521, fortran/102685, fortran/102715, fortran/102745,
fortran/102816, fortran/102817, fortran/102917, fortran/103137,
fortran/103138, fortran/103392, gcov-profile/100520, ipa/102714,
ipa/102762, ipa/103052, ipa/103246, ipa/103267, libstdc++/96416,
libstdc++/98421, libstdc++/100117, libstdc++/100153, libstdc++/100748,
libstdc++/101571, libstdc++/101608, libstdc++/102894,
libstdc++/103022, libstdc++/103086, libstdc++/103133,
libstdc++/103240, libstdc++/103381, middle-end/64888,
middle-end/101480, middle-end/102431, middle-end/102518,
middle-end/103059, middle-end/103181, middle-end/103248,
middle-end/103384, preprocessor/103130, rtl-optimization/102356,
rtl-optimization/102842, target/101985, target/102976, target/102991,
target/103205, target/103274, target/103275, testsuite/102690,
tree-optimization/100393, tree-optimization/102139,
tree-optimization/102505, tree-optimization/102572,
tree-optimization/102788, tree-optimization/102789,
tree-optimization/102798, tree-optimization/102970,
tree-optimization/103192, tree-optimization/103204,
tree-optimization/103237, tree-optimization/103255,
tree-optimization/103435
- fix up #__VA_OPT__ handling (PR preprocessor/103415)
* Wed Nov 17 2021 Marek Polacek <polacek@redhat.com> 11.2.1-6.3
- backport Intel Tremont tuning (#2014276)
* Wed Nov 17 2021 Marek Polacek <polacek@redhat.com> 11.2.1-6.2
- drop -Wbidirectional patch, use newer -Wbidi-chars (#2017820)
* Fri Oct 29 2021 Marek Polacek <polacek@redhat.com> 11.2.1-6.1
- add -Wbidirectional patch (#2017820)
* Tue Oct 26 2021 Marek Polacek <polacek@redhat.com> 11.2.1-6
- update from releases/gcc-11-branch (#1996862)
- PRs target/100208, target/100316, target/102761
- build target shared libraries with -Wl,-z,relro,-z,now
- add mwaitintrin.h on x86 (#2013860)
- improve generated code with extern thread_local constinit vars
with trivial dtors
- add support for C++20 #__VA_OPT__
- apply DTS-specific testsuite patches (#1996085)
* Tue Aug 17 2021 Marek Polacek <polacek@redhat.com> 11.2.1-1.1
- add .hidden for _ZNSt10filesystem9_Dir_base7advanceEbRSt10error_code