summaryrefslogtreecommitdiffstats
path: root/drivers/accel/ivpu/ivpu_mmu_context.c
Commit message (Collapse)AuthorAgeFilesLines
* accel/ivpu: Make parts of FW image read-onlyWachowski, Karol2024-06-141-0/+86
| | | | | | | | | | | | Implement setting specified buffer ranges as read-only. In case if specified range is not 64K aligned and 64K contiguous MMU600 pages are turned on, split 64K mapping to allow 4K granularity for read-only configuration. Signed-off-by: Wachowski, Karol <karol.wachowski@intel.com> Reviewed-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com> Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240611120433.1012423-10-jacek.lawrynowicz@linux.intel.com
* fix missing vmalloc.h includesKent Overstreet2024-04-251-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Patch series "Memory allocation profiling", v6. Overview: Low overhead [1] per-callsite memory allocation profiling. Not just for debug kernels, overhead low enough to be deployed in production. Example output: root@moria-kvm:~# sort -rn /proc/allocinfo 127664128 31168 mm/page_ext.c:270 func:alloc_page_ext 56373248 4737 mm/slub.c:2259 func:alloc_slab_page 14880768 3633 mm/readahead.c:247 func:page_cache_ra_unbounded 14417920 3520 mm/mm_init.c:2530 func:alloc_large_system_hash 13377536 234 block/blk-mq.c:3421 func:blk_mq_alloc_rqs 11718656 2861 mm/filemap.c:1919 func:__filemap_get_folio 9192960 2800 kernel/fork.c:307 func:alloc_thread_stack_node 4206592 4 net/netfilter/nf_conntrack_core.c:2567 func:nf_ct_alloc_hashtable 4136960 1010 drivers/staging/ctagmod/ctagmod.c:20 [ctagmod] func:ctagmod_start 3940352 962 mm/memory.c:4214 func:alloc_anon_folio 2894464 22613 fs/kernfs/dir.c:615 func:__kernfs_new_node ... Usage: kconfig options: - CONFIG_MEM_ALLOC_PROFILING - CONFIG_MEM_ALLOC_PROFILING_ENABLED_BY_DEFAULT - CONFIG_MEM_ALLOC_PROFILING_DEBUG adds warnings for allocations that weren't accounted because of a missing annotation sysctl: /proc/sys/vm/mem_profiling Runtime info: /proc/allocinfo Notes: [1]: Overhead To measure the overhead we are comparing the following configurations: (1) Baseline with CONFIG_MEMCG_KMEM=n (2) Disabled by default (CONFIG_MEM_ALLOC_PROFILING=y && CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=n) (3) Enabled by default (CONFIG_MEM_ALLOC_PROFILING=y && CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=y) (4) Enabled at runtime (CONFIG_MEM_ALLOC_PROFILING=y && CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=n && /proc/sys/vm/mem_profiling=1) (5) Baseline with CONFIG_MEMCG_KMEM=y && allocating with __GFP_ACCOUNT (6) Disabled by default (CONFIG_MEM_ALLOC_PROFILING=y && CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=n) && CONFIG_MEMCG_KMEM=y (7) Enabled by default (CONFIG_MEM_ALLOC_PROFILING=y && CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=y) && CONFIG_MEMCG_KMEM=y Performance overhead: To evaluate performance we implemented an in-kernel test executing multiple get_free_page/free_page and kmalloc/kfree calls with allocation sizes growing from 8 to 240 bytes with CPU frequency set to max and CPU affinity set to a specific CPU to minimize the noise. Below are results from running the test on Ubuntu 22.04.2 LTS with 6.8.0-rc1 kernel on 56 core Intel Xeon: kmalloc pgalloc (1 baseline) 6.764s 16.902s (2 default disabled) 6.793s (+0.43%) 17.007s (+0.62%) (3 default enabled) 7.197s (+6.40%) 23.666s (+40.02%) (4 runtime enabled) 7.405s (+9.48%) 23.901s (+41.41%) (5 memcg) 13.388s (+97.94%) 48.460s (+186.71%) (6 def disabled+memcg) 13.332s (+97.10%) 48.105s (+184.61%) (7 def enabled+memcg) 13.446s (+98.78%) 54.963s (+225.18%) Memory overhead: Kernel size: text data bss dec diff (1) 26515311 18890222 17018880 62424413 (2) 26524728 19423818 16740352 62688898 264485 (3) 26524724 19423818 16740352 62688894 264481 (4) 26524728 19423818 16740352 62688898 264485 (5) 26541782 18964374 16957440 62463596 39183 Memory consumption on a 56 core Intel CPU with 125GB of memory: Code tags: 192 kB PageExts: 262144 kB (256MB) SlabExts: 9876 kB (9.6MB) PcpuExts: 512 kB (0.5MB) Total overhead is 0.2% of total memory. Benchmarks: Hackbench tests run 100 times: hackbench -s 512 -l 200 -g 15 -f 25 -P baseline disabled profiling enabled profiling avg 0.3543 0.3559 (+0.0016) 0.3566 (+0.0023) stdev 0.0137 0.0188 0.0077 hackbench -l 10000 baseline disabled profiling enabled profiling avg 6.4218 6.4306 (+0.0088) 6.5077 (+0.0859) stdev 0.0933 0.0286 0.0489 stress-ng tests: stress-ng --class memory --seq 4 -t 60 stress-ng --class cpu --seq 4 -t 60 Results posted at: https://evilpiepirate.org/~kent/memalloc_prof_v4_stress-ng/ [2] https://lore.kernel.org/all/20240306182440.2003814-1-surenb@google.com/ This patch (of 37): The next patch drops vmalloc.h from a system header in order to fix a circular dependency; this adds it to all the files that were pulling it in implicitly. [kent.overstreet@linux.dev: fix arch/alpha/lib/memcpy.c] Link: https://lkml.kernel.org/r/20240327002152.3339937-1-kent.overstreet@linux.dev [surenb@google.com: fix arch/x86/mm/numa_32.c] Link: https://lkml.kernel.org/r/20240402180933.1663992-1-surenb@google.com [kent.overstreet@linux.dev: a few places were depending on sizes.h] Link: https://lkml.kernel.org/r/20240404034744.1664840-1-kent.overstreet@linux.dev [arnd@arndb.de: fix mm/kasan/hw_tags.c] Link: https://lkml.kernel.org/r/20240404124435.3121534-1-arnd@kernel.org [surenb@google.com: fix arc build] Link: https://lkml.kernel.org/r/20240405225115.431056-1-surenb@google.com Link: https://lkml.kernel.org/r/20240321163705.3067592-1-surenb@google.com Link: https://lkml.kernel.org/r/20240321163705.3067592-2-surenb@google.com Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev> Signed-off-by: Suren Baghdasaryan <surenb@google.com> Signed-off-by: Arnd Bergmann <arnd@arndb.de> Reviewed-by: Pasha Tatashin <pasha.tatashin@soleen.com> Tested-by: Kees Cook <keescook@chromium.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Alex Gaynor <alex.gaynor@gmail.com> Cc: Alice Ryhl <aliceryhl@google.com> Cc: Andreas Hindborg <a.hindborg@samsung.com> Cc: Benno Lossin <benno.lossin@proton.me> Cc: "Björn Roy Baron" <bjorn3_gh@protonmail.com> Cc: Boqun Feng <boqun.feng@gmail.com> Cc: Christoph Lameter <cl@linux.com> Cc: Dennis Zhou <dennis@kernel.org> Cc: Gary Guo <gary@garyguo.net> Cc: Miguel Ojeda <ojeda@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Tejun Heo <tj@kernel.org> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Wedson Almeida Filho <wedsonaf@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
* accel/ivpu: Add debug prints for MMU map/unmap operationsWachowski, Karol2024-01-221-0/+9
| | | | | | | | | | | | | | It is common need to be able to see IOVA/physical to VPU addresses mappings. Especially when debugging different kind of memory related issues. Lack of such logs forces user to modify and recompile KMD manually. This commit adds those logs under MMU debug mask which can be turned on dynamically with module param during KMD load. Signed-off-by: Wachowski, Karol <karol.wachowski@intel.com> Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com> Reviewed-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240115134434.493839-4-jacek.lawrynowicz@linux.intel.com
* Merge drm/drm-next into drm-misc-nextMaxime Ripard2023-11-151-6/+3
|\ | | | | | | | | | | Let's kickstart the v6.8 release cycle. Signed-off-by: Maxime Ripard <mripard@kernel.org>
| * BackMerge tag 'v6.6-rc7' into drm-nextDave Airlie2023-10-231-6/+3
| |\ | | | | | | | | | | | | | | | This is needed to add the msm pr which is based on a higher base. Signed-off-by: Dave Airlie <airlied@redhat.com>
| | * accel/ivpu: Extend address range for MMU mmapWludzik, Jozef2023-10-191-6/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Allow to use whole address range in MMU context mmap which is up to 48 bits. Return invalid argument from MMU context mmap in case address is not aligned to MMU page size, address is below MMU page size or address is greater then 47 bits. This fixes problem disallowing to run large models on VPU4 Signed-off-by: Wludzik, Jozef <jozef.wludzik@intel.com> Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20231018110113.547208-1-stanislaw.gruszka@linux.intel.com
* | | accel/ivpu: Fix locking in ivpu_bo_remove_all_bos_from_context()Jacek Lawrynowicz2023-11-081-15/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ivpu_bo_remove_all_bos_from_context() could race with ivpu_bo_free() when prime buffer was closed after vpu device was closed. Move the bo_list from context to vdev and use a dedicated lock to sync it. This list is not modified when BO is added/removed from a context. Also rename ivpu_bo_free_vpu_addr() to ivpu_bo_unbind() because this function does more then just free vpu_addr. Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com> Reviewed-by: Jeffrey Hugo <quic_jhugo@quicinc.com> Link: https://patchwork.freedesktop.org/patch/msgid/20231031073156.1301669-3-stanislaw.gruszka@linux.intel.com
* | | accel/ivpu: Allocate vpu_addr in gem->open() callbackJacek Lawrynowicz2023-11-081-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Use gem->open() callback to simplify the code and prepare for gem_shmem conversion. It is called during handle creation for a gem object, during prime import and in BO_CREATE ioctl. Hence can be used for vpu_addr allocation. On the way remove unused bo->user_ptr field. Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com> Reviewed-by: Jeffrey Hugo <quic_jhugo@quicinc.com> Link: https://patchwork.freedesktop.org/patch/msgid/20231031073156.1301669-2-stanislaw.gruszka@linux.intel.com
* | | accel/ivpu: Make DMA allocations for MMU600 write combinedKarol Wachowski2023-10-311-52/+63
|/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Previously using dma_alloc_wc() API we created cache coherent (mapped as write-back) mappings. Because we disable MMU600 snooping it was required to do costly page walk and cache flushes after each page table modification. With write-combined buffers it's possible to do a single write memory barrier to flush write-combined buffer to memory which simplifies the driver and significantly reduce time of map/unmap operations. Mapping time of 255 MB is reduced from 2.5 ms to 500 us. Signed-off-by: Karol Wachowski <karol.wachowski@linux.intel.com> Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Reviewed-by: Jeffrey Hugo <quic_jhugo@quicinc.com> Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20231028155936.1183342-7-stanislaw.gruszka@linux.intel.com
* | accel/ivpu: Initialize context with SSID = 1Karol Wachowski2023-09-041-0/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Context with SSID = 1 is reserved and accesses on that context happen only when context is uninitialized on the VPU side. Such access triggers MMU fault (0xa) "Invalid CD Fetch", which doesn't contain any useful information besides context ID. This commit will change that state, now (0x10) "Translation fault" will be triggered and accessed address will shown in the log. Signed-off-by: Karol Wachowski <karol.wachowski@linux.intel.com> Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Reviewed-by: Jeffrey Hugo <quic_jhugo@quicinc.com> Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20230901094957.168898-7-stanislaw.gruszka@linux.intel.com
* | accel/ivpu: Add information about context on failureStanislaw Gruszka2023-09-041-2/+2
| | | | | | | | | | | | | | | | | | | | Identify the mmu context that failed to initialize in the error messages. This allows the error to be correlated with a specific user during debug. Reviewed-by: Karol Wachowski <karol.wachowski@linux.intel.com> Reviewed-by: Jeffrey Hugo <quic_jhugo@quicinc.com> Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20230901094957.168898-5-stanislaw.gruszka@linux.intel.com
* | accel/ivpu: Remove duplicated error messagesJacek Lawrynowicz2023-09-041-1/+3
|/ | | | | | | | | | | | | | | Reduce the number of error messages per single failure in ivpu_dev_init() and ivpu_probe(). Most error messages are already printed by functions called from ivpu_dev_init(). Add missed error prints in ivpu_ipc_init() and ivpu_mmu_context_init(). Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com> Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Reviewed-by: Jeffrey Hugo <quic_jhugo@quicinc.com> Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20230901094957.168898-3-stanislaw.gruszka@linux.intel.com
* accel/ivpu: Refactor memory ranges logicKarol Wachowski2023-08-091-4/+4
| | | | | | | | | | | | | | | | | Add new dma range and change naming convention for virtual address memory ranges managed by KMD. New available ranges are named as follows: * global range - global context accessible by FW * aliased range - user context accessible by FW * dma range - user context accessible by DMA * shave range - user context accessible by shaves * global shave range - global context accessible by shave nn Signed-off-by: Karol Wachowski <karol.wachowski@linux.intel.com> Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20230731161258.2987564-6-stanislaw.gruszka@linux.intel.com
* accel/ivpu: Mark 64 kB contiguous areas as contiguous in PTEsKarol Wachowski2023-06-081-16/+59
| | | | | | | | | | | | | | | Whenever KMD maps region larger than 64kB that is both aligned and contiguous, set contiguous bit (52) in MMU PTE descriptor for each page in that region. This allows to treat 16 contiguous pages as one and reduce number of MMU page walks required which results in lower latency. Signed-off-by: Karol Wachowski <karol.wachowski@linux.intel.com> Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Reviewed-by: Jeffrey Hugo <quic_jhugo@quicinc.com> Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20230518131605.650622-6-stanislaw.gruszka@linux.intel.com
* accel/ivpu: Rename and cleanup MMU600 page tablesKarol Wachowski2023-06-081-88/+93
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Simplify and unify naming convention in MMU600 page tables configuration. All DMA addresses in page tables directly accessed by VPU are called with _dma sufix and all CPU pointers to those page tables have _ptr sufix. Base pointers used to do a page walk on the CPU have corresponding names: pud_ptrs (pointers used to get access to PUD DMA) pmd_ptrs (pointers used to get access to PMD DMA) pte_ptrs (pointers used to get access to PTE DMA) with the following convention: u64 *pud_dma_ptr = pud_ptrs[pgd_idx]; *pud_dma_ptr = pud_dma; u64 *pmd_dma_ptr = pmd_ptrs[pgd_idx][pud_idx]; *pmd_dma_ptr = pmd_dma; u64 *pte_dma_ptr = pte_ptrs[pgd_idx][pud_idx][pmd_idx]; *pte_dma_ptr = pte_dma; On the way change to coherent dma allocation, _wc is only valid on ARM and was used by mistake. Signed-off-by: Karol Wachowski <karol.wachowski@linux.intel.com> Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Reviewed-by: Jeffrey Hugo <quic_jhugo@quicinc.com> Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20230518131605.650622-5-stanislaw.gruszka@linux.intel.com
* accel/ivpu: Add MMU support for 4 level page mappingsKarol Wachowski2023-06-081-49/+117
| | | | | | | | | | | | Program additional fourth level required for mappings with VA above 38bits. Co-developed-by: Raymond Tan <raymond.tan@intel.com> Signed-off-by: Raymond Tan <raymond.tan@intel.com> Signed-off-by: Karol Wachowski <karol.wachowski@linux.intel.com> Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Reviewed-by: Jeffrey Hugo <quic_jhugo@quicinc.com> Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20230518131605.650622-3-stanislaw.gruszka@linux.intel.com
* accel/ivpu: Add Intel VPU MMU supportJacek Lawrynowicz2023-01-191-0/+398
VPU Memory Management Unit is based on ARM MMU-600. It allows the creation of multiple virtual address spaces for the device and map noncontinuous host memory (there is no dedicated memory on the VPU). Address space is implemented as a struct ivpu_mmu_context, it has an ID, drm_mm allocator for VPU addresses and struct ivpu_mmu_pgtable that holds actual 3-level, 4KB page table. Context with ID 0 (global context) is created upon driver initialization and it's mainly used for mapping memory required to execute the firmware. Contexts with non-zero IDs are user contexts allocated each time the devices is open()-ed and they map command buffers and other workload-related memory. Workloads executing in a given contexts have access only to the memory mapped in this context. This patch is has two main files: - ivpu_mmu_context.c handles MMU page tables and memory mapping - ivpu_mmu.c implements a driver that programs the MMU device Co-developed-by: Karol Wachowski <karol.wachowski@linux.intel.com> Signed-off-by: Karol Wachowski <karol.wachowski@linux.intel.com> Co-developed-by: Krystian Pradzynski <krystian.pradzynski@linux.intel.com> Signed-off-by: Krystian Pradzynski <krystian.pradzynski@linux.intel.com> Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Reviewed-by: Jeffrey Hugo <quic_jhugo@quicinc.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: https://patchwork.freedesktop.org/patch/msgid/20230117092723.60441-3-jacek.lawrynowicz@linux.intel.com
generated by cgit v1.2.3 (git 2.25.1) at 2024-07-29 19:15:26 +0000