* Fix cpuid kernel module build on ARM
* Return ERR_NO_CPUID when cpuid kernel module cannot be used on AArch32
* state
* Refactor the build of the bindings (#203)
* Fix build error on AArch64 when HWCAP_CPUID is not defined (#205)
* Fix build on Windows ARM with MSVC (206)
* Fix detection of Intel Meteor Lake
* Support for Intel Arrow Lake
* Support for AMD Turin
OBS-URL: https://build.opensuse.org/package/show/utilities/libcpuid?expand=0&rev=38
* Version 0.7.0 (2024-08-26):
* Fix handle leaks in rdmsr.c (#199)
* Fix cpuid_get_hypervisor when NULL data is provided (#199)
* Prevent intel_fn11 array overruns (#199)
* Support for AMD Hawk Point
* Support for more AMD Phoenix (8000 series)
* Add cpu_clock_by_tsc() function to the library (#124)
* Check x86 CPUs MSR support (#185)
* Add support for ARM CPUs (AArch32 + AArch64) (#200)
* Add cpu_feature_level_t enumerated values for x86 CPUs (#177)
* Support up to 4 subleaf entries for CPUID leaf 0x80000026
(#189)
* Support for Extended CPU topology subleaf in
cpuid_identify_purpose_amd() (#189)
* Support CPU purpose for AMD x86 CPUs (#189)
* Add cpuid Linux and FreeBSD kernel modules for ARM CPUs
* Improve errors handling in cpuid_get_all_raw_data() and
cpuid_get_raw_data_core() (#202)
* Support get_total_cpus() on DragonFly BSD
* Improve set_cpu_affinity() on NetBSD
* Fix build on OpenBSD
* Improve behavior when CPU affinity cannot be set
* Fix a regression in cpuid_tool about arguments doing nothing
(like --rdmsr or --cpuid)
* Fix a segmentation fault when using --quiet in cpuid_tool
* Improve error handling in cpu_identify_all()
* Add Python bindings (#197)
* Support for AMD Granite Ridge
* Support for AMD Strix Point
OBS-URL: https://build.opensuse.org/package/show/utilities/libcpuid?expand=0&rev=36
* Support for Intel Bay Trail-M
* Support for Intel Bay Trail-T
* Support for Intel Bay Trail-D
* Support for AMD Storm Peak
* Support for Intel Raport Lake Refresh
* Support heterogeneous RAW dumps in cpu_identify_all()
* Support for Intel Meteor Lake, including detection of LP E-Cores
* Support for Intel Emerald Rapids-SP
* Support for more AMD Van Gogh
OBS-URL: https://build.opensuse.org/package/show/utilities/libcpuid?expand=0&rev=34
- Update to version 0.6.3:
* Support for Intel Pentium and Celeron for Alder Lake-S
* Support for Intel Alder Lake-HX
* Support for Intel Alder Lake-X
* Fix detection of Intel Alder Lake-P
* Fix infinite loop in set_cpu_affinity() on macOS
* Fix a misprint of extended CPUID in cpuid_basic_identify()
* Restore previous thread CPU affinity before returning from cpuid_get_all_raw_data() (#184)
* Query CPU info at least once even if set_cpu_affinity() fails
* Support for AMD 19h family MSRs
* Fix detection of Intel Core i5 Lynnfield
* Rename set_error() to cpuid_set_error() and get_error() to cpuid_get_error() (#188)
* Support for Intel Alder Lake-N
* Support for AMD Rembrandt with Radeon Graphics
* Support for Intel Raptor Lake-S with "Golden Cove" cores
* Support for Intel Raptor Lake-P
* Support for Intel Raptor Lake-U
* Support for Intel Rocket Lake-E
OBS-URL: https://build.opensuse.org/request/show/1082737
OBS-URL: https://build.opensuse.org/package/show/openSUSE:Factory/libcpuid?expand=0&rev=14
* Support for Intel Alder Lake-X
* Fix detection of Intel Alder Lake-P
* Fix infinite loop in set_cpu_affinity() on macOS
* Fix a misprint of extended CPUID in cpuid_basic_identify()
* Restore previous thread CPU affinity before returning from cpuid_get_all_raw_data() (#184)
* Query CPU info at least once even if set_cpu_affinity() fails
* Support for AMD 19h family MSRs
* Fix detection of Intel Core i5 Lynnfield
* Rename set_error() to cpuid_set_error() and get_error() to cpuid_get_error() (#188)
* Support for Intel Alder Lake-N
* Support for AMD Rembrandt with Radeon Graphics
* Support for Intel Raptor Lake-S with "Golden Cove" cores
* Support for Intel Raptor Lake-P
* Support for Intel Raptor Lake-U
* Support for Intel Rocket Lake-E
OBS-URL: https://build.opensuse.org/package/show/utilities/libcpuid?expand=0&rev=30
* Support for Intel Pentium and Celeron for Alder Lake-S
* Support for Intel Alder Lake-HX
* Support for Intel Alder Lake-X
* Fix detection of Intel Alder Lake-P
* Fix infinite loop in set_cpu_affinity() on macOS
* Fix a misprint of extended CPUID in cpuid_basic_identify()
* Restore previous thread CPU affinity before returning from cpuid_get_all_raw_data() (#184)
* Query CPU info at least once even if set_cpu_affinity() fails
* Support for AMD 19h family MSRs
* Fix detection of Intel Core i5 Lynnfield
* Rename set_error() to cpuid_set_error() and get_error() to cpuid_get_error() (#188)
* Support for Intel Alder Lake-N
* Support for AMD Rembrandt with Radeon Graphics
* Support for Intel Raptor Lake-S with "Golden Cove" cores
* Support for Intel Raptor Lake-P
* Support for Intel Raptor Lake-U
* Support for Intel Rocket Lake-E
OBS-URL: https://build.opensuse.org/package/show/utilities/libcpuid?expand=0&rev=29
- update to 0.6.2:
* Support for AMD Raphael
* Support for AMD Dali
* Support for AMD Van Gogh
* Fix stuck cpuid_tool due to set_cpu_affinity() on Windows (#172)
* Remove AMD Warhol from DB (Zen3+ cancelled)
* Fix physical core count computed by cpu_identify_all() when HT is disabled (#175)
* Fix shared library symlinks with CMake (#174)
* Support for Intel Raptor Lake-S
* Fix segmentation fault in cpu_identify_all() for single-core CPUs
* Support for Intel Penryn L
* Support for Intel Tremont
* Support for AMD Mendocino
* Support for Intel Ice Lake (Xeon-D)
* Support for AMD Zen 2 Desktop Kit CPUs (4700S + 4800S)
* Support for AMD Athlon 64 Sherman
OBS-URL: https://build.opensuse.org/request/show/1039652
OBS-URL: https://build.opensuse.org/package/show/utilities/libcpuid?expand=0&rev=27
- Add missing dependency
- update to 0.6.0:
* Support for AMD Rembrandt
* Support for AMD Warhol
* Remove Debian package from source tree (#165)
* Fix build under Clang 15 (#167)
* Support for AMD Athlon Godavari
* Support for hybrid CPU like Intel Alder Lake (#166)
* Detect presence of hypervisor (#169)
* Decode deterministic cache info for AMD CPUs (#168)
* Add cache instances field in cpu_id_t and system_id_t (#168)
* Support AMD Bald Eagle
* Support for more AMD Godavari (Athlon)
* Rename AMD Bulldozer to Zambezi
* Support for AMD Interlagos
* Support for AMD Abu Dhabi
* Support for AMD Beema
* Support for AMD Steppe Eagle
* Support for more AMD Kabini (Sempron + Athlon)
* Improve msr_serialize_raw_data()
* Support for AMD Zen 2 custom APU for Steam Deck
OBS-URL: https://build.opensuse.org/request/show/1010188
OBS-URL: https://build.opensuse.org/package/show/openSUSE:Factory/libcpuid?expand=0&rev=12
- update to 0.6.0:
* Support for AMD Rembrandt
* Support for AMD Warhol
* Remove Debian package from source tree (#165)
* Fix build under Clang 15 (#167)
* Support for AMD Athlon Godavari
* Support for hybrid CPU like Intel Alder Lake (#166)
* Detect presence of hypervisor (#169)
* Decode deterministic cache info for AMD CPUs (#168)
* Add cache instances field in cpu_id_t and system_id_t (#168)
* Support AMD Bald Eagle
* Support for more AMD Godavari (Athlon)
* Rename AMD Bulldozer to Zambezi
* Support for AMD Interlagos
* Support for AMD Abu Dhabi
* Support for AMD Beema
* Support for AMD Steppe Eagle
* Support for more AMD Kabini (Sempron + Athlon)
* Improve msr_serialize_raw_data()
* Support for AMD Zen 2 custom APU for Steam Deck
OBS-URL: https://build.opensuse.org/request/show/1007396
OBS-URL: https://build.opensuse.org/package/show/utilities/libcpuid?expand=0&rev=24
- Update to version 0.5.1+git.1626502835:
* CMake: reduce min cmake requirement 3.14 -> 3.13
* Fix failing CI builds introduced by cb5fdd1
* Use popcount64 from libc when available (#152)
* allow to build either static or shared (#156)
* cmake: allow libcpuid to be added as a CMake subproject (#155)
* fix installation of BUNDLE if iOS (#154)
* cmake: add an option to build tests (#153)
* Fix#150: CPU Family/Model is used as Ext.Family/Model
* DB: add Tiger Lake
OBS-URL: https://build.opensuse.org/request/show/914539
OBS-URL: https://build.opensuse.org/package/show/openSUSE:Factory/libcpuid?expand=0&rev=10
* CMake: reduce min cmake requirement 3.14 -> 3.13
* Fix failing CI builds introduced by cb5fdd1
* Use popcount64 from libc when available (#152)
* allow to build either static or shared (#156)
* cmake: allow libcpuid to be added as a CMake subproject (#155)
* fix installation of BUNDLE if iOS (#154)
* cmake: add an option to build tests (#153)
* Fix#150: CPU Family/Model is used as Ext.Family/Model
* DB: add Tiger Lake
OBS-URL: https://build.opensuse.org/package/show/utilities/libcpuid?expand=0&rev=22
- Update to version v0.5.0+git.20200526:
* CI: remove 'v' prefix in assets
* CI: checkout sources before making release
* Release version 0.5.0 (#146)
* Add GitHub workflows for CI/CD - CI: it will check code consistency and run tests for all events (except for tags) - CD: it will build all assets and create a draft Close#122
* check-consistency: return error count
* Fix code consistency Result before this patch:
* CMake: fix include directory
* CMake: fix build on Windows
* CMake: fix install target's export
* tests: fix unused-result warning in convert_instlatx64 tool
* Update .gitignore
* CMake: fix Unix install and format
* Add config file for cmake-format It formats CMakeLists.txt files See https://github.com/cheshirekow/cmake_format
* Doxygen: upgrade Doxyfile to avoid warnings warning: Tag 'PERL_PATH' at line 1032 of file '/libcpuid/build/libcpuid/Doxyfile' has become obsolete. To avoid this warning please remove this line from your configuration file or upgrade it using "doxygen -u" warning: argument 'a4wide' for option PAPER_TYPE is not a valid enum value Using the default: a4!
* Doxygen: turn on quiet mode It is too noisy with CMake
* Detect AVX512VBMI and AVX512VBMI2 features on Intel CPUs More information: https://en.wikichip.org/wiki/x86/avx-512Resolve#134
* Detect ABM feature on Intel CPUs Resolve#144
* Detect RDSEED/ADX/SHA_NI features on AMD CPUs These x86 instruction set extensions are present since Zen micro-architecture Resolve#145
* Update cpuid_main.c
* DB: add Ivy Bridge-E (Xeon)
* Tests: update all tests to add fields for L1I
* Tests: update to add L1I information Related to 25d0614811991c855ce7db0d898dbc6200dfa840 Dump of Core i5 520m from CPU-X#119
* Add L1 Instruction Cache information Some CPUs does not have the same associativity for L1D and L1I, as reported in X0rg/CPU-X#119 It adds l1_instruction_assoc and l1_instruction_cacheline in cpu_id_t To avoid confusing, also adds l1_data_assoc and l1_data_cacheline l1_assoc and l1_cacheline are leave untouched for backward compatibility
* Ignore .vscode directory Yes, 0b05f45e03b0aa39a65eba9451b59c9381e8474c was about VS Code
* Tests: add amd_fn8000001dh subleaf See e562798cecf4af852fdfef4b0e7bf159a5d9b4de
* Tests: parse subleafs in convert_instlatx64 Also, it adds 0xffffffff when data is not available, so all lines are presents
* Re-fix L3 cache associativity detection on AMD Zen 2 CPUs Previous commit: 848394ee460c70298f91569d33f2c156bddb0f6c
* Applied a patch from @tavplubix
* Use constant for registers name It helps when reading technical documentation and it avoids 'magic values'
OBS-URL: https://build.opensuse.org/request/show/809020
OBS-URL: https://build.opensuse.org/package/show/openSUSE:Factory/libcpuid?expand=0&rev=5
* CI: remove 'v' prefix in assets
* CI: checkout sources before making release
* Release version 0.5.0 (#146)
* Add GitHub workflows for CI/CD - CI: it will check code consistency and run tests for all events (except for tags) - CD: it will build all assets and create a draft Close#122
* check-consistency: return error count
* Fix code consistency Result before this patch:
* CMake: fix include directory
* CMake: fix build on Windows
* CMake: fix install target's export
* tests: fix unused-result warning in convert_instlatx64 tool
* Update .gitignore
* CMake: fix Unix install and format
* Add config file for cmake-format It formats CMakeLists.txt files See https://github.com/cheshirekow/cmake_format
* Doxygen: upgrade Doxyfile to avoid warnings warning: Tag 'PERL_PATH' at line 1032 of file '/libcpuid/build/libcpuid/Doxyfile' has become obsolete. To avoid this warning please remove this line from your configuration file or upgrade it using "doxygen -u" warning: argument 'a4wide' for option PAPER_TYPE is not a valid enum value Using the default: a4!
* Doxygen: turn on quiet mode It is too noisy with CMake
* Detect AVX512VBMI and AVX512VBMI2 features on Intel CPUs More information: https://en.wikichip.org/wiki/x86/avx-512Resolve#134
* Detect ABM feature on Intel CPUs Resolve#144
* Detect RDSEED/ADX/SHA_NI features on AMD CPUs These x86 instruction set extensions are present since Zen micro-architecture Resolve#145
* Update cpuid_main.c
* DB: add Ivy Bridge-E (Xeon)
* Tests: update all tests to add fields for L1I
* Tests: update to add L1I information Related to 25d0614811991c855ce7db0d898dbc6200dfa840 Dump of Core i5 520m from CPU-X#119
* Add L1 Instruction Cache information Some CPUs does not have the same associativity for L1D and L1I, as reported in X0rg/CPU-X#119 It adds l1_instruction_assoc and l1_instruction_cacheline in cpu_id_t To avoid confusing, also adds l1_data_assoc and l1_data_cacheline l1_assoc and l1_cacheline are leave untouched for backward compatibility
* Ignore .vscode directory Yes, 0b05f45e03b0aa39a65eba9451b59c9381e8474c was about VS Code
* Tests: add amd_fn8000001dh subleaf See e562798cecf4af852fdfef4b0e7bf159a5d9b4de
* Tests: parse subleafs in convert_instlatx64 Also, it adds 0xffffffff when data is not available, so all lines are presents
* Re-fix L3 cache associativity detection on AMD Zen 2 CPUs Previous commit: 848394ee460c70298f91569d33f2c156bddb0f6c
* Applied a patch from @tavplubix
* Use constant for registers name It helps when reading technical documentation and it avoids 'magic values'
OBS-URL: https://build.opensuse.org/package/show/utilities/libcpuid?expand=0&rev=12
- Update to version 0.4.1+git.20200102:
* DB: Add Threadripper (Castle Peak)
* Fix compilation on non-x86/ARM architectures.
* Add support for get_total_cpus on Haiku.
* Some typo fixes in human readable text.
* Add Xeon CLX (Cascade lake-based) using data from PR #129
* add support to feature intel avx512_vnni
* AARCH64 stub
* Ignore convert_instlatx64 binary
* add Hygon Dhyana C86 7seris test file
* Add Hygon Dhyana detect support
- Switch to _service
OBS-URL: https://build.opensuse.org/request/show/766548
OBS-URL: https://build.opensuse.org/package/show/openSUSE:Factory/libcpuid?expand=0&rev=4
* DB: Add Threadripper (Castle Peak)
* Fix compilation on non-x86/ARM architectures.
* Add support for get_total_cpus on Haiku.
* Some typo fixes in human readable text.
* Add Xeon CLX (Cascade lake-based) using data from PR #129
* add support to feature intel avx512_vnni
* AARCH64 stub
* Ignore convert_instlatx64 binary
* add Hygon Dhyana C86 7seris test file
* Add Hygon Dhyana detect support
- Switch to _service
OBS-URL: https://build.opensuse.org/package/show/utilities/libcpuid?expand=0&rev=10
* Better support for Skylake Core i5 (#76)
* Misdiagnosis microarchitecture for i3-3220T (#81)
* Ability to dump MSR values to a file (PR #82)
* AMD Ryzen support (#86)
* Support for Coffee and Kaby Lake (#104)
* Support for Raven Ridge and Threadripper (#106)
* Support for Pinnacle Ridge (#111)
* Fix P-III Celeron misdetection
* Support for Skylake-X (#116)
* Support for Zen+ Threadripper
OBS-URL: https://build.opensuse.org/package/show/utilities/libcpuid?expand=0&rev=7
