v0.12.0

Added

• Add idiomatic abstraction for vGPU type information (#68)
• Add device::mig_device_by_index() (#93)
• Add device::mig_device_count() (#94)
• Add device::vgpu_scheduler_capabilities() (#95)
• Add device::mig_parent_device() (#96)
• Add device::vgpu_host_mode() (#98)
• Add device::mig_is_mig_device_handle() (#99)
• Add device::set_mig_mode() (#100)
• Add vGPU capabilities support (#101)
• Add device::possible_placements() (#102)
• Add NvLink bandwidth mode API (#103)
• Add NvLink::remote_device_type() (#106)
• Add device::set_temperature_threshold() (#107)
• Add Nvml::vgpu_version() and Nvml::set_vgpu_version() (#108)
• Add device::vgpu_accounting_pids() and device::vgpu_accounting_stats() (#109)
• Add device::vgpu_scheduler_log() (#110)
• Add device::vgpu_scheduler_state() and device::set_vgpu_scheduler_state() (#111)
• Add MPS version of running_compute_processes for >= Volta (#112)
• Add device::gsp_firmware_mode() and device::gsp_firmware_version() (#113)
• Add device::profile_info() (#115)
• Implement From<NvmlError> for nvmlReturn_t conversion (#117)
• Add device::numa_node_id() for NUMA-node-friendly GPUs (#118)
• Add missing AcousticThreshold enum variants (#121)
• Add GPU Performance Monitoring (GPM) support (#122)

Fixed

• Only return samples that are actually valid (#67)
• Fix get cc macros (#104)
• Fix attestation report size (#116)
• Fix spelling mistake of "occurred" (#120)

Internal

• Address cargo clippy warnings (#105)
• Fix lifetime warnings (#119)

What's Changed

• Provide idiomatic abstraction for vGPU type information by @christoph-heiss in #68
• adding device::mig_device_by_index(). by @devnexen in #93
• adding device::mig_device_count() by @devnexen in #94
• Adding device::vgpu_scheduler_capabilities() by @devnexen in #95
• adding device::mig_parent_device() by @devnexen in #96
• adding device::mig_is_mig_device_handle() by @devnexen in #99
• adding device::vgpu_host_mode() by @devnexen in #98
• adding vGPU capabilities support. by @devnexen in #101
• adding device::set_mig_mode() by @devnexen in #100
• adding device::possible_placements() by @devnexen in #102
• Fix get cc macros by @dcmiddle in #104
• adding bandwidth mode api for NvLink. by @devnexen in #103
• Address cargo clippy warnings by @thinkingfish in #105
• adding NvLink::remote_device_type from a link index attached to a dev… by @devnexen in #106
• adding device::set_temperature_threshold for a given type and C. temp. by @devnexen in #107
• adding Nvml::vgpu_version() and Nvml::set_vgpu_version(). by @devnexen in #108
• Vgpu instance pids by @devnexen in #109
• adding device::vgpu_scheduler_log() to fetch n scheduler entries log by @devnexen in #110
• adding device::vgpu_scheduler_state/set_vgpu_scheduler_state by @devnexen in #111
• adding device::numa_node_id() for NUMA node friendly GPUs. by @devnexen in #118
• adding device::profile_info() by @devnexen in #115
• adding the MPS version of running_compute_processes for >= Volta. by @devnexen in #112
• device: fix attestation report size by @fitzthum in #116
• feat(error): implement From for nvmlReturn_t conversion by @fslongjin in #117
• samples: Only return samples that are actually valid. by @umanwizard in #67
• adding device::gsp_firmware_mode/gsp_firmware_version() by @devnexen in #113
• Fix lifetime warnings by @swlynch99 in #119
• Fix spelling mistake of occurred by @bdrung in #120
• Add missing TemperatureThreshold enum variants by @ilya-zlobintsev in #121
• feat: add GPU Performance Monitoring (GPM) by @brayniac in #122
• release: v0.12.0 by @brayniac in #123

New Contributors

• @christoph-heiss made their first contribution in #68
• @dcmiddle made their first contribution in #104
• @thinkingfish made their first contribution in #105
• @fitzthum made their first contribution in #116
• @fslongjin made their first contribution in #117
• @umanwizard made their first contribution in #67
• @swlynch99 made their first contribution in #119
• @bdrung made their first contribution in #120

Full Changelog: v0.11.0...v0.12.0

v0.11.0

And nvml-wrapper-sys 0.9.0

Fixes, improvements, and updates for NVML 12.8.90.

Fixed

• Include license text in the packaged crates (#66)
• Fix EventLoop elided lifetime warnings (#69)
• Use proper api version from header. (#80)

Added

• Add Gpc and Mem clock offset methods (#61)
• Add nvmlDeviceGetP2PStatus (#62)
• Add Blackwell to DeviceArchitecture enum (#70)
• Add DeviceAttributes and device::attributes(). (#71)
• Add attestation report fetch functionality to device (#73)
• Add confidential compute gpu certificate functionality (#74)
• Add checks for confidential compute (#76)
• Add fan control functions (#77)
• Add device::cpu_affinity_within_scope. (#79)
• Add FanSpeedInfo and FanSpeedRPM wrappers. (#80)
• Add device::memory_affinity(). (#82)
• Add device::mig_mode(). (#83)
• Add wrappers for clock offset and pstate functions, min_max_fan_speed (#85)
• Add device::virtualization_mode() wrapper. (#86)
• Add vgpu_driver_capabilities() (#87)
• Add device::performance_modes() (#88)
• Add device::active_vgpus() (#89)
• Add util functions for cc and multi-gpu ppcie (#90)
• Add set fan control policy function (#91)
• Add last_seen_timestamp parameter to get more accurate process utilization counts. (#92)

Changed

• Update nvmlDeviceGetMemoryInfo to version 2 to be consistent with nvidia-smi (#58)
• Set LIB_PATH to libnvidia-ml.so.1 instead of libnvidia-ml.so on Linux (#63)