[OpenBLAS] Update to v0.3.26

O/OpenBLAS/OpenBLAS32@0.3.26/build_tarballs.jl

@@ -0,0 +1,20 @@
+using BinaryBuilder
+
+include("../common.jl")
+
+# Collection of sources required to build OpenBLAS
+name = "OpenBLAS32"
+version = v"0.3.26"
+
+sources = openblas_sources(version)
+script = openblas_script(openblas32=true)
+platforms = openblas_platforms()
+products = openblas_products()
+dependencies = openblas_dependencies(platforms)
+
+# Build the tarballs
+build_tarballs(ARGS, name, version, sources, script, platforms, products, dependencies;
+               preferred_gcc_version=v"6", lock_microarchitecture=false,
+               julia_compat="1.11", preferred_llvm_version=v"13.0.1")
+
+# Build trigger: 1

O/OpenBLAS/OpenBLAS32@0.3.26/bundled/patches/10-neoverse-generic-kernels.patch

@@ -0,0 +1,19 @@
+diff --git a/kernel/arm64/KERNEL.NEOVERSEN1 b/kernel/arm64/KERNEL.NEOVERSEN1
+index ea010db4..074d7215 100644
+--- a/kernel/arm64/KERNEL.NEOVERSEN1
++++ b/kernel/arm64/KERNEL.NEOVERSEN1
+@@ -91,10 +91,10 @@ IDAMAXKERNEL   = iamax_thunderx2t99.c
+ ICAMAXKERNEL   = izamax_thunderx2t99.c
+ IZAMAXKERNEL   = izamax_thunderx2t99.c
+
+-SNRM2KERNEL    = scnrm2_thunderx2t99.c
+-DNRM2KERNEL    = dznrm2_thunderx2t99.c
+-CNRM2KERNEL    = scnrm2_thunderx2t99.c
+-ZNRM2KERNEL    = dznrm2_thunderx2t99.c
++SNRM2KERNEL    = nrm2.S
++DNRM2KERNEL    = nrm2.S
++CNRM2KERNEL    = znrm2.S
++ZNRM2KERNEL    = znrm2.S
+
+ DDOTKERNEL     = dot_thunderx2t99.c
+ SDOTKERNEL     = dot_thunderx2t99.c

O/OpenBLAS/OpenBLAS32@0.3.26/bundled/patches/20-Fix-older-version-of-gcc-missing-__has_builtin-cpuid.patch

@@ -0,0 +1,85 @@
+From d99aad8ee308600832da39105a6511275cfe32ad Mon Sep 17 00:00:00 2001
+From: Chip-Kerchner <[email protected]>
+Date: Tue, 14 Nov 2023 11:07:08 -0600
+Subject: [PATCH] Fix older version of gcc - missing __has_builtin, cpuid and
+ no support of P10.
+
+---
+ Makefile.power                | 12 ++++++++++++
+ driver/others/dynamic_power.c | 17 ++++++++++++-----
+ 2 files changed, 24 insertions(+), 5 deletions(-)
+
+diff --git a/Makefile.power b/Makefile.power
+index 95bada34f..aa1ca080a 100644
+--- a/Makefile.power
++++ b/Makefile.power
+@@ -11,7 +11,19 @@ endif
+
+ ifeq ($(CORE), POWER10)
+ ifneq ($(C_COMPILER), PGI)
++ifeq ($(C_COMPILER), GCC))
++ifeq ($(GCCVERSIONGTEQ10), 1)
+ CCOMMON_OPT += -Ofast -mcpu=power10 -mtune=power10 -mvsx -fno-fast-math
++else ifneq ($(GCCVERSIONGT4), 1)
++$(warning your compiler is too old to fully support POWER9, getting a newer version of gcc is recommended)
++CCOMMON_OPT += -Ofast -mcpu=power8 -mtune=power8 -mvsx -fno-fast-math
++else
++$(warning your compiler is too old to fully support POWER10, getting a newer version of gcc is recommended)
++CCOMMON_OPT += -Ofast -mcpu=power9 -mtune=power9 -mvsx -fno-fast-math
++endif
++else
++CCOMMON_OPT += -Ofast -mcpu=power10 -mtune=power10 -mvsx -fno-fast-math
++endif
+ ifeq ($(F_COMPILER), IBM)
+ FCOMMON_OPT += -O2 -qrecur -qnosave -qarch=pwr10 -qtune=pwr10 -qfloat=nomaf -qzerosize
+ else
+diff --git a/driver/others/dynamic_power.c b/driver/others/dynamic_power.c
+index f0faf2baf..0454f186c 100644
+--- a/driver/others/dynamic_power.c
++++ b/driver/others/dynamic_power.c
+@@ -66,8 +66,7 @@ static int cpuid(void)
+ #endif
+     return CPU_UNKNOWN;
+ }
+-#else
+-#if defined(C_PGI) || defined(__clang__)
++#elif defined(C_PGI) || defined(__clang__)
+ /*
+  * NV HPC compilers do not yet implement __builtin_cpu_is().
+  * Fake a version here for use in the CPU detection code below.
+@@ -196,13 +195,21 @@ static int cpuid(void)
+ 	cpu_type = pvrPOWER[i].cpu_type;
+ 	return (int)(cpu_type);
+ }
+-#endif  /* C_PGI */
++#elif !defined(__BUILTIN_CPU_SUPPORTS__)
++static int cpuid(void)
++{
++    return CPU_UNKNOWN;
++}
+ #endif  /* _AIX */
+
+ #ifndef __BUILTIN_CPU_SUPPORTS__
+ #include <string.h>
+
+-#if defined(_AIX) || (defined(__has_builtin) && !__has_builtin(__builtin_cpu_is))
++#ifndef __has_builtin
++#define __has_builtin(x)   0
++#endif
++
++#if defined(_AIX) || !__has_builtin(__builtin_cpu_is)
+ static int __builtin_cpu_is(const char *arg)
+ {
+     static int ipinfo = -1;
+@@ -227,7 +234,7 @@ static int __builtin_cpu_is(const char *arg)
+ }
+ #endif
+
+-#if defined(_AIX) || (defined(__has_builtin) && !__has_builtin(__builtin_cpu_supports))
++#if defined(_AIX) || !__has_builtin(__builtin_cpu_supports)
+ static int __builtin_cpu_supports(const char *arg)
+ {
+     return 0;
+-- 
+2.42.1
+

O/OpenBLAS/OpenBLAS32@0.3.26/bundled/patches/30-openblas-ofast-power.patch

@@ -0,0 +1,43 @@
+diff --git a/Makefile.power b/Makefile.power
+index aa1ca080a..42c417a78 100644
+--- a/Makefile.power
++++ b/Makefile.power
+@@ -13,16 +13,16 @@ ifeq ($(CORE), POWER10)
+ ifneq ($(C_COMPILER), PGI)
+ ifeq ($(C_COMPILER), GCC))
+ ifeq ($(GCCVERSIONGTEQ10), 1)
+-CCOMMON_OPT += -Ofast -mcpu=power10 -mtune=power10 -mvsx -fno-fast-math
++CCOMMON_OPT += -mcpu=power10 -mtune=power10 -mvsx -fno-fast-math
+ else ifneq ($(GCCVERSIONGT4), 1)
+ $(warning your compiler is too old to fully support POWER9, getting a newer version of gcc is recommended)
+-CCOMMON_OPT += -Ofast -mcpu=power8 -mtune=power8 -mvsx -fno-fast-math
++CCOMMON_OPT += -mcpu=power8 -mtune=power8 -mvsx -fno-fast-math
+ else
+ $(warning your compiler is too old to fully support POWER10, getting a newer version of gcc is recommended)
+-CCOMMON_OPT += -Ofast -mcpu=power9 -mtune=power9 -mvsx -fno-fast-math
++CCOMMON_OPT += -mcpu=power9 -mtune=power9 -mvsx -fno-fast-math
+ endif
+ else
+-CCOMMON_OPT += -Ofast -mcpu=power10 -mtune=power10 -mvsx -fno-fast-math
++CCOMMON_OPT += -mcpu=power10 -mtune=power10 -mvsx -fno-fast-math
+ endif
+ ifeq ($(F_COMPILER), IBM)
+ FCOMMON_OPT += -O2 -qrecur -qnosave -qarch=pwr10 -qtune=pwr10 -qfloat=nomaf -qzerosize
+@@ -34,7 +34,7 @@ endif
+
+ ifeq ($(CORE), POWER9)
+ ifneq ($(C_COMPILER), PGI)
+-CCOMMON_OPT += -Ofast -mvsx -fno-fast-math
++CCOMMON_OPT += -mvsx -fno-fast-math
+ ifeq ($(C_COMPILER), GCC)
+ ifneq ($(GCCVERSIONGT4), 1)
+ $(warning your compiler is too old to fully support POWER9, getting a newer version of gcc is recommended)
+@@ -70,7 +70,7 @@ endif
+
+ ifeq ($(CORE), POWER8)
+ ifneq ($(C_COMPILER), PGI)
+-CCOMMON_OPT += -Ofast -mcpu=power8 -mtune=power8 -mvsx  -fno-fast-math
++CCOMMON_OPT += -mcpu=power8 -mtune=power8 -mvsx  -fno-fast-math
+ else
+ CCOMMON_OPT += -fast -Mvect=simd -Mcache_align
+ endif

O/OpenBLAS/OpenBLAS32@0.3.26/bundled/patches/50-openblas-winexit.patch

@@ -0,0 +1,170 @@
+diff --git a/driver/others/memory.c b/driver/others/memory.c
+index 6e654ccf..1d2f9f12 100644
+--- a/driver/others/memory.c
++++ b/driver/others/memory.c
+@@ -1534,7 +1534,7 @@ void CONSTRUCTOR gotoblas_init(void) {
+
+ }
+
+-void DESTRUCTOR gotoblas_quit(void) {
++void gotoblas_quit(void) {
+
+   if (gotoblas_initialized == 0) return;
+
+@@ -1572,75 +1572,11 @@ void DESTRUCTOR gotoblas_quit(void) {
+ }
+
+ #if defined(_MSC_VER) && !defined(__clang__)
+-BOOL APIENTRY DllMain(HMODULE hModule, DWORD  ul_reason_for_call, LPVOID lpReserved)
+-{
+-  switch (ul_reason_for_call)
+-  {
+-    case DLL_PROCESS_ATTACH:
+-      gotoblas_init();
+-      break;
+-    case DLL_THREAD_ATTACH:
+-      break;
+-    case DLL_THREAD_DETACH:
+-#if defined(SMP)
+-      blas_thread_memory_cleanup();
+-#endif
+-      break;
+-    case DLL_PROCESS_DETACH:
+-      gotoblas_quit();
+-      break;
+-    default:
+-      break;
+-  }
+-  return TRUE;
+-}
+-
+-/*
+-  This is to allow static linking.
+-  Code adapted from Google performance tools:
+-  https://gperftools.googlecode.com/git-history/perftools-1.0/src/windows/port.cc
+-  Reference:
+-  https://sourceware.org/ml/pthreads-win32/2008/msg00028.html
+-  http://ci.boost.org/svn-trac/browser/trunk/libs/thread/src/win32/tss_pe.cpp
+-*/
+-static int on_process_term(void)
+-{
+-  gotoblas_quit();
+-  return 0;
+-}
+ #ifdef _WIN64
+ #pragma comment(linker, "/INCLUDE:_tls_used")
+ #else
+ #pragma comment(linker, "/INCLUDE:__tls_used")
+ #endif
+-
+-#ifdef _WIN64
+-#pragma const_seg(".CRT$XLB")
+-#else
+-#pragma data_seg(".CRT$XLB")
+-#endif
+-
+-#ifdef _WIN64
+-static const PIMAGE_TLS_CALLBACK dll_callback(HINSTANCE h, DWORD ul_reason_for_call, PVOID pv) = DllMain;
+-#pragma const_seg()
+-#else
+-static void (APIENTRY *dll_callback)(HINSTANCE h, DWORD ul_reason_for_call, PVOID pv) = DllMain;
+-#pragma data_seg()
+-#endif
+-
+-#ifdef _WIN64
+-#pragma const_seg(".CRT$XTU")
+-#else
+-#pragma data_seg(".CRT$XTU")
+-#endif
+-
+-#ifdef _WIN64
+-static const int(*p_process_term)(void) = on_process_term;
+-#pragma const_seg()
+-#else
+-static int(*p_process_term)(void) = on_process_term;
+-#pragma data_seg()
+-#endif
+ #endif
+
+ #if (defined(C_PGI) || (!defined(C_SUN) && defined(F_INTERFACE_SUN))) && (defined(ARCH_X86) || defined(ARCH_X86_64))
+@@ -3146,7 +3082,7 @@ void CONSTRUCTOR gotoblas_init(void) {
+
+ }
+
+-void DESTRUCTOR gotoblas_quit(void) {
++void gotoblas_quit(void) {
+
+   if (gotoblas_initialized == 0) return;
+
+@@ -3175,71 +3111,6 @@ void DESTRUCTOR gotoblas_quit(void) {
+ #endif
+ }
+
+-#if defined(_MSC_VER) && !defined(__clang__)
+-BOOL APIENTRY DllMain(HMODULE hModule, DWORD  ul_reason_for_call, LPVOID lpReserved)
+-{
+-  switch (ul_reason_for_call)
+-  {
+-    case DLL_PROCESS_ATTACH:
+-      gotoblas_init();
+-      break;
+-    case DLL_THREAD_ATTACH:
+-      break;
+-    case DLL_THREAD_DETACH:
+-      break;
+-    case DLL_PROCESS_DETACH:
+-      gotoblas_quit();
+-      break;
+-    default:
+-      break;
+-  }
+-  return TRUE;
+-}
+-
+-/*
+-  This is to allow static linking.
+-  Code adapted from Google performance tools:
+-  https://gperftools.googlecode.com/git-history/perftools-1.0/src/windows/port.cc
+-  Reference:
+-  https://sourceware.org/ml/pthreads-win32/2008/msg00028.html
+-  http://ci.boost.org/svn-trac/browser/trunk/libs/thread/src/win32/tss_pe.cpp
+-*/
+-static int on_process_term(void)
+-{
+-  gotoblas_quit();
+-  return 0;
+-}
+-#ifdef _WIN64
+-#pragma comment(linker, "/INCLUDE:_tls_used")
+-#else
+-#pragma comment(linker, "/INCLUDE:__tls_used")
+-#endif
+-
+-#ifdef _WIN64
+-#pragma const_seg(".CRT$XLB")
+-#else
+-#pragma data_seg(".CRT$XLB")
+-#endif
+-static void (APIENTRY *dll_callback)(HINSTANCE h, DWORD ul_reason_for_call, PVOID pv) = DllMain;
+-#ifdef _WIN64
+-#pragma const_seg()
+-#else
+-#pragma data_seg()
+-#endif
+-
+-#ifdef _WIN64
+-#pragma const_seg(".CRT$XTU")
+-#else
+-#pragma data_seg(".CRT$XTU")
+-#endif
+-static int(*p_process_term)(void) = on_process_term;
+-#ifdef _WIN64
+-#pragma const_seg()
+-#else
+-#pragma data_seg()
+-#endif
+-#endif
+-
+ #if (defined(C_PGI) || (!defined(C_SUN) && defined(F_INTERFACE_SUN))) && (defined(ARCH_X86) || defined(ARCH_X86_64))
+ /* Don't call me; this is just work around for PGI / Sun bug */
+ void gotoblas_dummy_for_PGI(void) {

O/OpenBLAS/OpenBLAS@0.3.26/build_tarballs.jl

@@ -0,0 +1,21 @@
+using BinaryBuilder
+
+include("../common.jl")
+
+# Collection of sources required to build OpenBLAS
+name = "OpenBLAS"
+version = v"0.3.26"
+
+sources = openblas_sources(version)
+script = openblas_script(;aarch64_ilp64=true, num_64bit_threads=512)
+platforms = openblas_platforms(;experimental=true)
+push!(platforms, Platform("x86_64", "linux"; sanitize="memory"))
+products = openblas_products()
+dependencies = openblas_dependencies(platforms)
+
+# Build the tarballs
+build_tarballs(ARGS, name, version, sources, script, platforms, products, dependencies;
+               preferred_gcc_version=v"6", lock_microarchitecture=false,
+               julia_compat="1.11", preferred_llvm_version=v"13.0.1")
+
+# Build trigger: 1

O/OpenBLAS/OpenBLAS@0.3.26/bundled/patches/10-neoverse-generic-kernels.patch

@@ -0,0 +1,19 @@
+diff --git a/kernel/arm64/KERNEL.NEOVERSEN1 b/kernel/arm64/KERNEL.NEOVERSEN1
+index ea010db4..074d7215 100644
+--- a/kernel/arm64/KERNEL.NEOVERSEN1
++++ b/kernel/arm64/KERNEL.NEOVERSEN1
+@@ -91,10 +91,10 @@ IDAMAXKERNEL   = iamax_thunderx2t99.c
+ ICAMAXKERNEL   = izamax_thunderx2t99.c
+ IZAMAXKERNEL   = izamax_thunderx2t99.c
+
+-SNRM2KERNEL    = scnrm2_thunderx2t99.c
+-DNRM2KERNEL    = dznrm2_thunderx2t99.c
+-CNRM2KERNEL    = scnrm2_thunderx2t99.c
+-ZNRM2KERNEL    = dznrm2_thunderx2t99.c
++SNRM2KERNEL    = nrm2.S
++DNRM2KERNEL    = nrm2.S
++CNRM2KERNEL    = znrm2.S
++ZNRM2KERNEL    = znrm2.S
+
+ DDOTKERNEL     = dot_thunderx2t99.c
+ SDOTKERNEL     = dot_thunderx2t99.c