grace-baremetal.txt

# NVIDIA Grace BML Kernel Patches
# Format: COMMIT_SHA | TITLE
#
# This list is organized by the same groupings found in
# the NVIDIA Grace and Vera Platform Support Software
# Patches and Configurations Guide.
#
# https://docs.nvidia.com/dccpu/patches-config-guide/index.html
#



# Bare Metal Support: Enablement
################################

4c207e7121fa | PCI: Mark some NVIDIA GPUs to avoid bus reset
6937b7dd4349 | PCI: Add support for ACPI _RST reset method
a9c38c5d267c | dma-mapping: remove bogus test for pfn_valid from dma_map_resource
bd2fdedbf2ba | i2c: tegra: Add the ACPI support
f89d2cc3967a | spi: tegra210-quad: use devm call for cdata memory
50665d58db05 | i2c: tegra: use i2c_timings for bus clock freq
5f84e73f9a8f | gpio: tegra186: Add IRQ per bank for Tegra241
d1056b771ddb | gpio: tegra186: Add support for Tegra241
d7583c8a5748 | i2c: tegra: Add SMBus block read function
ca0acb511c21 | device property: Add fwnode_irq_get_byname
a263a84088f6 | i2c: smbus: Use device_*() functions instead of of_*()
20e8ef5c7ffa | docs: firmware-guide: ACPI: Add named interrupt doc
ac982578e7d3 | spi: tegra210-quad: use device_reset method
ea23f0e148b8 | spi: tegra210-quad: add new chips to compatible
75a1b44a54bd | spi: tegra210-quad: add acpi support
1b8342cc4a38 | spi: tegra210-quad: combined sequence mode
b76134178168 | spi: tegra210-quad: Multi-cs support
0ddf10a319f1 | arm64: tegra: Enable Tegra SPI & QSPI in deconfig
721255b9826b | genirq: Use a maple tree for interrupt descriptor management
47c8846a49ba | PCI: Extend ACS configurability



# Bare Metal Support: Critical Fixes
####################################

5bf71889ad9a | i2c: tegra: Set ACPI node as primary fwnode
9f855779a387 | i2c: tegra: Fix PEC support for SMBUS block read
35727af2b15d | irqchip/gicv3: Workaround for NVIDIA erratum T241-FABRIC-4
4327a6137ed4 | drm/ast: Fix ARM compatibility
195aba96b854 | tpm_tis_spi: Account for SPI header when allocating TPM SPI xfer buffer
1a596ad00ffe | PCI: Use downstream bridges for distributing resources
e403e8538359 | arm64: errata: Assume that unknown CPUs _are_ vulnerable to Spectre BHB
a5951389e58d | arm64: errata: Add newer ARM cores to the spectre_bhb_loop_affected() lists
fee4d171451c | arm64: errata: Add missing sentinels to Spectre-BHB MIDR arrays
d42e6c20de61 | arm64/entry: Mask DAIF in cpu_switch_to(), call_on_irq_stack()
97c5550b7631 | arm64: contpte: fix set_access_flags() no-op check for SMMU/ATS faults



# Bare Metal Support: Fixes
###########################

8777dd9dff40 | spi: tegra210-quad: Fix combined sequence
ae4b3c1252f0 | spi: tegra210-quad: Don't initialise DMA if not supported
2197aa6b0aa2 | spi: tegra210-quad: Fix duplicate resource error
2e3468778dbe | mm: remember young/dirty bit for page migrations
101ca8d05913 | rtc: efi: Enable SET/GET WAKEUP services as optional
ef3691683d7b | KVM: arm64: GICv4.1: Fix race with doorbell on VPE activation/deactivation
05da178ce0aa | ACPI/IORT: Update SMMUv3 DeviceID support
047ee71ae4f4 | spi: tegra210-quad: Fix validate combined sequence
2449d436681d | spi: tegra210-quad: Fix iterator outside loop
f7482d8285b6 | spi: tegra210-quad: set half duplex flag
010338d729c1 | arm64: kaslr: don't pretend KASLR is enabled if offset < MIN_KIMG_ALIGN
c0e0421a60bf | ACPI: processor: Reorder acpi_processor_driver_init()
c43198af05cf | thermal: core: Introduce thermal_cooling_device_present()
790930f44289 | thermal: core: Introduce thermal_cooling_device_update()
22c52fa5155a | ACPI: processor: thermal: Update CPU cooling devices on cpufreq policy changes
b57841fb0b56 | thermal: core: Drop excessive lockdep_assert_held() calls
bba5065963f8 | PCI/AER: Configure ECRC only if AER is native
67a142dc9eb9 | spi: Add TPM HW flow flag
967ca91a996f | spi: tegra210-quad: Enable TPM wait polling
3e35d303ab7d | arm64: module: rework module VA range selection
a86a42ac2bd6 | tpm_tis_spi: Add hardware wait polling
d5afb4b47e13 | iommu/arm-smmu-v3: Fix soft lockup triggered by arm_smmu_mm_invalidate_range
416ef04fe00c | mm/page_alloc: fix min_free_kbytes calculation regarding ZONE_MOVABLE
081488051d28 | mm/mglru: fix underprotected page cache
5095a2b23987 | mm/mglru: try to stop at high watermarks
8aa420617918 | mm/mglru: respect min_ttl_ms with memcgs
4376807bf2d5 | mm/mglru: reclaim offlined memcgs harder
db744ddd59be | PCI/MSI: Prevent MSI hardware interrupt number truncation
6d4266675279 | PCI/ASPM: Update save_state when configuration changes
d806f474a9a7 | gpio: tegra186: Fix tegra186_gpio_is_accessible() check
04867a7a3332 | swiotlb: Fix double-allocation of slots due to broken alignment handling
cbf53074a528 | swiotlb: Honour dma_alloc_coherent() alignment in swiotlb_alloc()
51b30ecb73b4 | swiotlb: Fix alignment checks when both allocation and DMA masks are present
afc5aa46ed56 | iommu/dma: Force swiotlb_max_mapping_size on an untrusted device
af133562d5af | swiotlb: extend buffer pre-padding to alloc_align_mask if necessary
e3ba51ab24fd | arm64: tlb: Fix TLBI RANGE operand
e07255d69702 | arm64: tlb: Improve __TLBI_VADDR_RANGE()
73301e464a72 | arm64: tlb: Allow range operation for MAX_TLBI_RANGE_PAGES
7bf9d2af7e89 | PCI: Clear Secondary Status errors after enumeration
eebab7e3eb4b | PCI/DOE: Support discovery version 2
90e4ed6bb02a | cpufreq/cppc: Don't compare desired_perf in target()
4cd7ba16a0af | mm: fix old/young bit handling in the faulting path
14d069d92951 | i2c: tegra: Do not mark ACPI devices as irq safe
088984c8d54c | ACPI: PRM: Find EFI_MEMORY_RUNTIME block for PRM handler and context
aa6f8b2593b5 | mm/gup: stop leaking pinned pages in low memory conditions
94efde1d1539 | mm/gup: avoid an unnecessary allocation call for FOLL_LONGTERM cases
a1268be280d8 | mm/gup: handle NULL pages in unpin_user_pages()
f3c3ccc4fe49 | PCI: Fix pci_enable_acs() support for the ACS quirks
c47195631960 | cppc_cpufreq: Use desired perf if feedback ctrs are 0 or unchanged
ea1829d4d413 | cppc_cpufreq: Remove HiSilicon CPPC workaround
1bc542c6a0d1 | mm/vmscan: wake up flushers conditionally to avoid cgroup OOM
c8e88de1b44e | ACPI/HMAT: Move HMAT messages to pr_debug()
7f5704b6a143 | ACPI: PRM: Remove unnecessary strict handler address checks
abc7b3f1f056 | RDMA/mlx5: Fix a WARN during dereg_mr for DM type
f7edb07ad7c6 | Fix mmu notifiers for range-based invalidates
9cf8a952d57b | PCI/ACS: Fix 'pci=config_acs=' parameter
b3f6fcd8404f | iommu: Skip PASID validation for devices without PASID capability
41c721fc0939 | spi: tegra210-quad: use WARN_ON_ONCE instead of WARN_ON for timeouts
21f4314e66ed | spi: tegra210-quad: add rate limiting and simplify timeout error message
a6e04f05ce0b | i2c: tegra: check msg length in SMBUS block read
56344e241c54 | i2c: tegra: Fix reset error handling with ACPI
d8966b654133 | spi: tegra210-quad: modify chip select (CS) deactivation
635bf3c88533 | i2c: tegra: Use internal reset when reset property is not available
c7c31f8dc54a | drm/ast: Use msleep instead of mdelay for edid read
7526183cfdbe | net: aquantia: Add missing descriptor cache invalidation on ATL2
5ffcb7b890f6 | net: atlantic: fix fragment overflow handling in RX path
b4e002d8a7ce | spi: tegra210-quad: Fix timeout handling
6022eacdda8b | spi: tegra210-quad: Refactor error handling into helper functions
380fd29d57ab | spi: tegra210-quad: Check hardware status on timeout
aabd8ea0aa25 | spi: tegra210-quad: Return IRQ_HANDLED when timeout already processed transfer
ef13ba357656 | spi: tegra210-quad: Move curr_xfer read inside spinlock
f5a4d7f5e32b | spi: tegra210-quad: Protect curr_xfer assignment in tegra_qspi_setup_transfer_one
bf4528ab28e2 | spi: tegra210-quad: Protect curr_xfer in tegra_qspi_combined_seq_xfer
6d7723e8161f | spi: tegra210-quad: Protect curr_xfer clearing in tegra_qspi_non_combined_seq_xfer
edf9088b6e1d | spi: tegra210-quad: Protect curr_xfer check in IRQ handler



# Bare Metal Support: Performance
#################################

a2ee7581afd5 | ACPI: thermal: Add Thermal fast Sampling Period (_TFP) support
310293a2b941 | ACPI: processor: reduce CPUFREQ thermal reduction pctg for Tegra241
ccb45b34d440 | ACPI: arm64: export acpi_arch_thermal_cpufreq_pctg()
7dc7c5ef6463 | mm: allow deferred splitting of arbitrary anon large folios
372cbd4d5a06 | mm: non-pmd-mappable, large folios for folio_add_new_anon_rmap()
3485b88390b0 | mm: thp: introduce multi-size THP sysfs interface
19eaf44954df | mm: thp: support allocation of anonymous multi-size THP
b6aab3384caf | selftests/mm/kugepaged: restore thp settings at exit
00679a183ac6 | selftests/mm: factor out thp settings management
4f5070a5e40d | selftests/mm: support multi-size THP interface in thp_settings
9f0704eae8a4 | selftests/mm/khugepaged: enlighten for multi-size THP
12dc16b38463 | selftests/mm/cow: generalize do_run_with_thp() helper
c0f79103322c | selftests/mm/cow: add tests for anonymous multi-size THP
6280d7317cca | mm: clarify the spec for set_ptes()
2bdba9868a4f | mm: thp: batch-collapse PMD with set_ptes()
583ceaaa3399 | mm: introduce pte_advance_pfn() and use for pte_next_pfn()
c1bd2b4028ae | arm64/mm: convert pte_next_pfn() to pte_advance_pfn()
506b586769ec | x86/mm: convert pte_next_pfn() to pte_advance_pfn()
fb23bf6bd288 | mm: tidy up pte_next_pfn() definition
532736558e8e | arm64/mm: convert READ_ONCE(*ptep) to ptep_get(ptep)
659e19302791 | arm64/mm: convert set_pte_at() to set_ptes(..., 1)
cbb0294fdd72 | arm64/mm: convert ptep_clear() to ptep_get_and_clear()
5a00bfd6a52c | arm64/mm: new ptep layer to manage contig bit
d9d8dc2bd3fb | arm64/mm: dplit __flush_tlb_range() to elide trailing DSB
4602e5757bcc | arm64/mm: wire up PTE_CONT for user mappings
311a6cf29690 | arm64/mm: implement new wrprotect_ptes() batch API
6b1e4efb6f54 | arm64/mm: implement new [get_and_]clear_full_ptes() batch APIs
c6ec76a2ebc5 | mm: add pte_batch_hint() to reduce scanning in folio_pte_batch()
fb5451e5f72b | arm64/mm: implement pte_batch_hint()
b972fc6afba0 | arm64/mm: __always_inline to improve fork() perf
f0c2264958e1 | arm64/mm: automatically fold contpte mappings
ead79118dae6 | arm64/io: Provide a WC friendly __iowriteXX_copy()
2b7a5e1fe022 | net: hns3: Remove io_stop_wc() calls after __iowrite64_copy()
ef302283ddfc | IB/mlx5: Use __iowrite64_copy() for write combining stores



# Bare Metal Support: Performance Tools
#######################################

1c5d62f5dd97 | ACPICA: Add support for ARM Performance Monitoring Unit Table.
6251d38059ae | ACPI: ARM Performance Monitoring Unit Table (APMT) initial support
e37dfd65731d | perf: arm_cspmu: Add support for ARM CoreSight PMU driver
84481be7167e | perf: arm_cspmu: Add support for NVIDIA SCF and MCF attribute
39522031798d | ACPI: APMT: Fix kerneldoc and indentation
1830902eb896 | perf: arm_cspmu: Fix modular builds due to missing MODULE_LICENSE()s
e72dbf9085b5 | perf: arm_cspmu: Fix build failure on x86_64
a91bbd5c9984 | perf: arm_cspmu: Fix module cyclic dependency
16e15834659e | perf: arm_cspmu: Fix variable dereference warning
f462eb1d0df3 | arm64: defconfig: Enable ARM CoreSight PMU driver
225d757012e0 | perf: arm_cspmu: Set irq affinitiy only if overflow interrupt is used
71e0cb32d5fc | perf/arm_cspmu: Fix event attribute type
7e51d05e43f1 | perf: arm_cspmu: Add missing MODULE_DEVICE_TABLE
87b3b6d53efc | ACPI/APMT: Don't register invalid resource
d2e3bb512818 | perf/arm_cspmu: Decouple APMT dependency
f9bd34e3753e | perf/arm_cspmu: Clean up ACPI dependency
3095e90eee5e | coresight: etm4x: Allocate and device assign 'struct etmv4_drvdata' earlier
4e3b9a6eae98 | coresight: etm4x: Drop iomem 'base' argument from etm4_probe()
5a1c7097472f | coresight: etm4x: Drop pid argument from etm4_probe()
73d779a03a76 | coresight: etm4x: Change etm4_platform_driver driver for MMIO devices
3a2888aa1f96 | coresight: platform: acpi: Ignore the absence of graph
134124acb57f | coresight: etm4x: Add ACPI support in platform driver
1aa3d0274a4a | arm_pmu: acpi: Add a representative platform device for TRBE
81e5ee471609 | arm_pmu: acpi: Refactor arm_spe_acpi_register_device()
4473949074c3 | perf vendor events arm64: Update N2 and V2 metrics and events using Arm telemetry repo
15c7ef7341a2 | perf: arm_cspmu: Reject events meant for other PMUs
a16afcc58a8c | perf cs-etm: Fix incorrect or missing decoder for raw trace
bfc653aa89cb | perf: arm_cspmu: Separate Arm and vendor module
17f8b216e026 | coresight: trbe: Enable ACPI based TRBE devices
4277f035d227 | coresight: trbe: Add a representative coresight_platform_data for TRBE
a24d9d9dc096 | perf parse-events: Make legacy events lower priority than sysfs/JSON
f4d9d9dcc70b | arm64: Add Neoverse-V2 part
dc6abbbde4b0 | tools headers arm64: Sync arm64's cputype.h with the kernel sources
b6aa0de9a53a | perf cs-etm: Create decoders after both AUX and HW_ID search passes
57880a7966be | perf: cs-etm: Allocate queues for all CPUs
77c123f53e97 | perf: cs-etm: Move traceid_list to each queue
19c3e4db38c5 | perf: cs-etm: Create decoders based on the trace ID mappings
940007cee539 | perf: cs-etm: Only save valid trace IDs into files
1506af6db8c4 | perf: cs-etm: Support version 0.1 of HW_ID packets
022aa67b5ab9 | perf: cs-etm: Print queue number in raw trace dump
34172002bdac | coresight: Remove unused ETM Perf stubs
eda1d11979c0 | coresight: Clarify comments around the PID of the sink owner
acb0184fe9bc | coresight: Move struct coresight_trace_id_map to common header
7e52877868ae | coresight: Expose map arguments in trace ID API
d53c8253c782 | coresight: Make CPU id map a property of a trace ID map
5ad628a76176 | coresight: Use per-sink trace ID maps for Perf sessions
de0029fdde86 | coresight: Remove pending trace ID release mechanism
487eec8da80a | coresight: Emit sink ID in the HW_ID packets
988d40a4d4e7 | coresight: Make trace ID map spinlock local to the map
fb98fa3bf868 | perf arm-spe: Rename arm_spe__synth_data_source_generic()
50b8f1d5bf4a | perf arm-spe: Rename the common data source encoding
56ae663e7636 | perf arm-spe: Introduce arm_spe__is_homogeneous()
ba5e7169e548 | perf arm-spe: Use metadata to decide the data source feature
6bcf54c89b3d | perf arm-spe: Remove the unused 'midr' field
041c0e5715a6 | perf arm-spe: Add Neoverse-V2 to common data source encoding list
ea2ead4224fd | perf arm-spe: Add Cortex CPUs to common data source encoding list
ac4c52956f62 | perf: arm_cspmu: nvidia: remove unsupported SCF events
5f7cd0dc98a6 | perf: arm_cspmu: nvidia: fix sysfs path in the kernel doc
ca26df4b1036 | perf: arm_cspmu: nvidia: enable NVLINK-C2C port filtering
bce61d5c5764 | perf: arm_cspmu: nvidia: monitor all ports by default
d923782b0412 | arm64: amu: Delay allocating cpumask for AMU FIE support
004b500a9031 | arch_topology: init capacity_freq_ref to 0
38e480d4fcac | cpufreq: Allow arch_freq_get_on_cpu to return an error
fbb4a4759b54 | cpufreq: Introduce an optional cpuinfo_avg_freq sysfs entry
16d1e27475f6 | arm64: Provide an AMU-based version of arch_freq_get_on_cpu
39b19974982e | arm64: Update AMU-based freq scale factor on entering idle
20711efa91e8 | arm64: Utilize for_each_cpu_wrap for reference lookup



# Bare Metal Support: CUDA
##########################

18b2db3b0385 | mm: Convert page kmemcg type to a page memcg flag
368c0159d492 | RDMA/umem: Support importing dma-buf as user memory region
b392a1989170 | vfio/pci: remove vfio_pci_nvlink2
319561669a59 | mm/mmu_notifier.c: fix race in mmu_interval_notifier_remove()
fd35ca3d12cc | mm/migrate_device.c: copy pte dirty bit to page
a3589e1d5fe3 | mm/migrate_device.c: add missing flush_cache_page()
60bae7370896 | mm/migrate_device.c: flush TLB while holding PTL



# End of list