Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add proper CPUID eax checking #3026

Draft
wants to merge 2 commits into
base: main
Choose a base branch
from
Draft

Add proper CPUID eax checking #3026

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
e4ddec6
Add proper CPUID eax checking
isuruf Dec 17, 2024
7c180fc
remove __daal_serv_cpu_extensions_available and add comments
isuruf Dec 19, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
38 changes: 23 additions & 15 deletions cpp/daal/src/services/compiler/generic/env_detect_features.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -71,6 +71,24 @@ void run_cpuid(uint32_t eax, uint32_t ecx, uint32_t * abcd)
#endif
}

uint32_t __daal_internal_get_max_extension_support()
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This function is not necessary, it can be folded in below.

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This follows the pattern of __daal_internal_is_intel_cpu and daal_check_is_intel_cpu where the former is the one that does the work and the latter is there to cache the value in a static variable to avoid running the former multiple times.

{
// Running cpuid with a value other than eax=0 and 0x8000000 is an extension
// To check that a particular eax value is supported we need to check
// maximum extension that is supported by checking the value returned by
// cpuid when eax=0x80000000 is given.
uint32_t abcd[4];
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

clearer naming is necessary.

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The naming used is abcd throughout this file. This variable stores the values of the eax, ebx, ecx, edx registers just like the other functions. Any ideas on what to rename all of them to?

run_cpuid(0x80000000, 0, abcd);
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

commentary for this would be nice.

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Added comments.

return abcd[0];
}

uint32_t daal_get_max_extension_support()
{
// We cache the result in a static variable here.
static const uint32_t result = __daal_internal_get_max_extension_support();
return result;
}

bool __daal_internal_is_intel_cpu()
{
const uint32_t genu = 0x756e6547, inei = 0x49656e69, ntel = 0x6c65746e;
Expand All @@ -87,6 +105,11 @@ DAAL_EXPORT bool daal_check_is_intel_cpu()

static int check_cpuid(uint32_t eax, uint32_t ecx, int abcd_index, uint32_t mask)
{
if (daal_get_max_extension_support() < eax)
{
// need to check that the eax we run here is supported.
return 0;
}
uint32_t abcd[4];

run_cpuid(eax, ecx, abcd);
Expand Down Expand Up @@ -191,11 +214,6 @@ static int check_sse42_features()
return 1;
}

DAAL_EXPORT bool __daal_serv_cpu_extensions_available()
{
return daal_check_is_intel_cpu();
}

DAAL_EXPORT int __daal_serv_cpu_detect(int enable)
{
#if defined(__APPLE__)
Expand Down Expand Up @@ -226,11 +244,6 @@ static bool check_sve_features()
return (hwcap & HWCAP_SVE) != 0;
}

DAAL_EXPORT bool __daal_serv_cpu_extensions_available()
{
return 0;
}

DAAL_EXPORT int __daal_serv_cpu_detect(int enable)
{
if (check_sve_features())
Expand All @@ -250,11 +263,6 @@ bool daal_check_is_intel_cpu()
return false;
}
#elif defined(TARGET_RISCV64)
DAAL_EXPORT bool __daal_serv_cpu_extensions_available()
{
return 0;
}

DAAL_EXPORT int __daal_serv_cpu_detect(int enable)
{
return daal::rv64;
Expand Down
7 changes: 5 additions & 2 deletions cpp/daal/src/services/service_defines.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,7 +28,6 @@
#include <stdint.h>
#include "services/env_detect.h"

DAAL_EXPORT bool __daal_serv_cpu_extensions_available();
DAAL_EXPORT int __daal_serv_cpu_detect(int);

void run_cpuid(uint32_t eax, uint32_t ecx, uint32_t * abcd);
Expand All @@ -42,7 +41,11 @@ DAAL_EXPORT bool daal_check_is_intel_cpu();
#define DAAL_BASE_CPU daal::rv64
#endif

#define DAAL_CHECK_CPU_ENVIRONMENT (daal_check_is_intel_cpu())
#if defined(TARGET_X86_64)
#define DAAL_CHECK_CPU_ENVIRONMENT 1
#else
#define DAAL_CHECK_CPU_ENVIRONMENT 0
#endif

#if defined(__INTEL_COMPILER)
#define PRAGMA_IVDEP _Pragma("ivdep")
Expand Down
10 changes: 3 additions & 7 deletions cpp/oneapi/dal/detail/cpu.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,15 +38,11 @@ cpu_extension from_daal_cpu_type(int cpu_type) {
}

cpu_extension detect_top_cpu_extension() {
if (!__daal_serv_cpu_extensions_available()) {
#if defined(TARGET_X86_64)
return detail::cpu_extension::sse2;
#elif defined(TARGET_ARM)
return detail::cpu_extension::sve;
#if defined(TARGET_ARM)
return detail::cpu_extension::sve;
#elif defined(TARGET_RISCV64)
return detail::cpu_extension::rv64;
return detail::cpu_extension::rv64;
#endif
}
const auto daal_cpu = __daal_serv_cpu_detect(0);

return from_daal_cpu_type(daal_cpu);
Expand Down