kernel.git/arch/arm64/mm/init.c, branch linux-5.10.y

arm64: Do not defer reserve_crashkernel() for platforms with no DMA memory zones

2022-04-08T12:40:45Z

commit 031495635b4668f94e964e037ca93d0d38bfde58 upstream. The following patches resulted in deferring crash kernel reservation to mem_init(), mainly aimed at platforms with DMA memory zones (no IOMMU), in particular Raspberry Pi 4. commit 1a8e1cef7603 ("arm64: use both ZONE_DMA and ZONE_DMA32") commit 8424ecdde7df ("arm64: mm: Set ZONE_DMA size based on devicetree's dma-ranges") commit 0a30c53573b0 ("arm64: mm: Move reserve_crashkernel() into mem_init()") commit 2687275a5843 ("arm64: Force NO_BLOCK_MAPPINGS if crashkernel reservation is required") Above changes introduced boot slowdown due to linear map creation for all the memory banks with NO_BLOCK_MAPPINGS, see discussion[1]. The proposed changes restore crash kernel reservation to earlier behavior thus avoids slow boot, particularly for platforms with IOMMU (no DMA memory zones). Tested changes to confirm no ~150ms boot slowdown on our SoC with IOMMU and 8GB memory. Also tested with ZONE_DMA and/or ZONE_DMA32 configs to confirm no regression to deferring scheme of crash kernel memory reservation. In both cases successfully collected kernel crash dump. [1] https://lore.kernel.org/all/9436d033-579b-55fa-9b00-6f4b661c2dd7@linux.microsoft.com/ Signed-off-by: Vijay Balakrishna Cc: stable@vger.kernel.org Reviewed-by: Pasha Tatashin Link: https://lore.kernel.org/r/1646242689-20744-1-git-send-email-vijayb@linux.microsoft.com [will: Add #ifdef CONFIG_KEXEC_CORE guards to fix 'crashk_res' references in allnoconfig build] Signed-off-by: Will Deacon Signed-off-by: Greg Kroah-Hartman

arm64: Ignore any DMA offsets in the max_zone_phys() calculation

2021-06-30T12:47:16Z

commit 791ab8b2e3db0c6e4295467d10398800ec29144c upstream. Currently, the kernel assumes that if RAM starts above 32-bit (or zone_bits), there is still a ZONE_DMA/DMA32 at the bottom of the RAM and such constrained devices have a hardwired DMA offset. In practice, we haven't noticed any such hardware so let's assume that we can expand ZONE_DMA32 to the available memory if no RAM below 4GB. Similarly, ZONE_DMA is expanded to the 4GB limit if no RAM addressable by zone_bits. Signed-off-by: Catalin Marinas Tested-by: Nicolas Saenz Julienne Reviewed-by: Nicolas Saenz Julienne Cc: Nicolas Saenz Julienne Cc: Robin Murphy Link: https://lore.kernel.org/r/20201118185809.1078362-1-catalin.marinas@arm.com Signed-off-by: Catalin Marinas Cc: Kefeng Wang Signed-off-by: Greg Kroah-Hartman

arm64: Remove arm64_dma32_phys_limit and its uses

2021-05-14T07:50:46Z

commit d78050ee35440d7879ed94011c52994b8932e96e upstream. With the introduction of a dynamic ZONE_DMA range based on DT or IORT information, there's no need for CMA allocations from the wider ZONE_DMA32 since on most platforms ZONE_DMA will cover the 32-bit addressable range. Remove the arm64_dma32_phys_limit and set arm64_dma_phys_limit to cover the smallest DMA range required on the platform. CMA allocation and crashkernel reservation now go in the dynamically sized ZONE_DMA, allowing correct functionality on RPi4. Signed-off-by: Catalin Marinas Cc: Chen Zhou Reviewed-by: Nicolas Saenz Julienne Tested-by: Nicolas Saenz Julienne # On RPi4B Cc: Kefeng Wang Signed-off-by: Greg Kroah-Hartman

arm64/mm: Fix pfn_valid() for ZONE_DEVICE based memory

2021-03-17T16:06:33Z

[ Upstream commit eeb0753ba27b26f609e61f9950b14f1b934fe429 ] pfn_valid() validates a pfn but basically it checks for a valid struct page backing for that pfn. It should always return positive for memory ranges backed with struct page mapping. But currently pfn_valid() fails for all ZONE_DEVICE based memory types even though they have struct page mapping. pfn_valid() asserts that there is a memblock entry for a given pfn without MEMBLOCK_NOMAP flag being set. The problem with ZONE_DEVICE based memory is that they do not have memblock entries. Hence memblock_is_map_memory() will invariably fail via memblock_search() for a ZONE_DEVICE based address. This eventually fails pfn_valid() which is wrong. memblock_is_map_memory() needs to be skipped for such memory ranges. As ZONE_DEVICE memory gets hotplugged into the system via memremap_pages() called from a driver, their respective memory sections will not have SECTION_IS_EARLY set. Normal hotplug memory will never have MEMBLOCK_NOMAP set in their memblock regions. Because the flag MEMBLOCK_NOMAP was specifically designed and set for firmware reserved memory regions. memblock_is_map_memory() can just be skipped as its always going to be positive and that will be an optimization for the normal hotplug memory. Like ZONE_DEVICE based memory, all normal hotplugged memory too will not have SECTION_IS_EARLY set for their sections Skipping memblock_is_map_memory() for all non early memory sections would fix pfn_valid() problem for ZONE_DEVICE based memory and also improve its performance for normal hotplug memory as well. Cc: Catalin Marinas Cc: Will Deacon Cc: Ard Biesheuvel Cc: Robin Murphy Cc: linux-arm-kernel@lists.infradead.org Cc: linux-kernel@vger.kernel.org Acked-by: David Hildenbrand Fixes: 73b20c84d42d ("arm64: mm: implement pte_devmap support") Signed-off-by: Anshuman Khandual Acked-by: Catalin Marinas Link: https://lore.kernel.org/r/1614921898-4099-2-git-send-email-anshuman.khandual@arm.com Signed-off-by: Will Deacon Signed-off-by: Sasha Levin

arm64: mm: Set ZONE_DMA size based on early IORT scan

2021-03-09T10:11:13Z

commit 2b8652936f0ca9ca2e6c984ae76c7bfcda1b3f22 upstream We recently introduced a 1 GB sized ZONE_DMA to cater for platforms incorporating masters that can address less than 32 bits of DMA, in particular the Raspberry Pi 4, which has 4 or 8 GB of DRAM, but has peripherals that can only address up to 1 GB (and its PCIe host bridge can only access the bottom 3 GB) Instructing the DMA layer about these limitations is straight-forward, even though we had to fix some issues regarding memory limits set in the IORT for named components, and regarding the handling of ACPI _DMA methods. However, the DMA layer also needs to be able to allocate memory that is guaranteed to meet those DMA constraints, for bounce buffering as well as allocating the backing for consistent mappings. This is why the 1 GB ZONE_DMA was introduced recently. Unfortunately, it turns out the having a 1 GB ZONE_DMA as well as a ZONE_DMA32 causes problems with kdump, and potentially in other places where allocations cannot cross zone boundaries. Therefore, we should avoid having two separate DMA zones when possible. So let's do an early scan of the IORT, and only create the ZONE_DMA if we encounter any devices that need it. This puts the burden on the firmware to describe such limitations in the IORT, which may be redundant (and less precise) if _DMA methods are also being provided. However, it should be noted that this situation is highly unusual for arm64 ACPI machines. Also, the DMA subsystem still gives precedence to the _DMA method if implemented, and so we will not lose the ability to perform streaming DMA outside the ZONE_DMA if the _DMA method permits it. [nsaenz: unified implementation with DT's counterpart] Signed-off-by: Ard Biesheuvel Signed-off-by: Nicolas Saenz Julienne Tested-by: Jeremy Linton Acked-by: Lorenzo Pieralisi Acked-by: Hanjun Guo Cc: Jeremy Linton Cc: Lorenzo Pieralisi Cc: Nicolas Saenz Julienne Cc: Rob Herring Cc: Christoph Hellwig Cc: Robin Murphy Cc: Hanjun Guo Cc: Sudeep Holla Cc: Anshuman Khandual Link: https://lore.kernel.org/r/20201119175400.9995-7-nsaenzjulienne@suse.de Signed-off-by: Catalin Marinas Cc: Signed-off-by: Jing Xiangfeng Signed-off-by: Greg Kroah-Hartman

arm64: mm: Set ZONE_DMA size based on devicetree's dma-ranges

2021-03-09T10:11:13Z

commit 8424ecdde7df99d5426e1a1fd9f0fb36f4183032 upstream We recently introduced a 1 GB sized ZONE_DMA to cater for platforms incorporating masters that can address less than 32 bits of DMA, in particular the Raspberry Pi 4, which has 4 or 8 GB of DRAM, but has peripherals that can only address up to 1 GB (and its PCIe host bridge can only access the bottom 3 GB) The DMA layer also needs to be able to allocate memory that is guaranteed to meet those DMA constraints, for bounce buffering as well as allocating the backing for consistent mappings. This is why the 1 GB ZONE_DMA was introduced recently. Unfortunately, it turns out the having a 1 GB ZONE_DMA as well as a ZONE_DMA32 causes problems with kdump, and potentially in other places where allocations cannot cross zone boundaries. Therefore, we should avoid having two separate DMA zones when possible. So, with the help of of_dma_get_max_cpu_address() get the topmost physical address accessible to all DMA masters in system and use that information to fine-tune ZONE_DMA's size. In the absence of addressing limited masters ZONE_DMA will span the whole 32-bit address space, otherwise, in the case of the Raspberry Pi 4 it'll only span the 30-bit address space, and have ZONE_DMA32 cover the rest of the 32-bit address space. Signed-off-by: Nicolas Saenz Julienne Link: https://lore.kernel.org/r/20201119175400.9995-6-nsaenzjulienne@suse.de Signed-off-by: Catalin Marinas Cc: Signed-off-by: Jing Xiangfeng Signed-off-by: Greg Kroah-Hartman

arm64: mm: Move zone_dma_bits initialization into zone_sizes_init()

2021-03-09T10:11:13Z

commit 9804f8c69b04a39d0ba41d19e6bdc6aa91c19725 upstream zone_dma_bits's initialization happens earlier that it's actually needed, in arm64_memblock_init(). So move it into the more suitable zone_sizes_init(). Signed-off-by: Nicolas Saenz Julienne Tested-by: Jeremy Linton Link: https://lore.kernel.org/r/20201119175400.9995-3-nsaenzjulienne@suse.de Signed-off-by: Catalin Marinas Cc: Signed-off-by: Jing Xiangfeng Signed-off-by: Greg Kroah-Hartman

arm64: mm: Move reserve_crashkernel() into mem_init()

2021-03-09T10:11:13Z

commit 0a30c53573b07d5561457e41fb0ab046cd857da5 upstream crashkernel might reserve memory located in ZONE_DMA. We plan to delay ZONE_DMA's initialization after unflattening the devicetree and ACPI's boot table initialization, so move it later in the boot process. Specifically into bootmem_init() since request_standard_resources() depends on it. Signed-off-by: Nicolas Saenz Julienne Tested-by: Jeremy Linton Link: https://lore.kernel.org/r/20201119175400.9995-2-nsaenzjulienne@suse.de Signed-off-by: Catalin Marinas Cc: Signed-off-by: Jing Xiangfeng Signed-off-by: Greg Kroah-Hartman

arm64: mm: Fix ARCH_LOW_ADDRESS_LIMIT when !CONFIG_ZONE_DMA

2021-01-17T13:17:02Z

commit 095507dc1350b3a2b8b39fdc05edba0c10859eca upstream. Systems configured with CONFIG_ZONE_DMA32, CONFIG_ZONE_NORMAL and !CONFIG_ZONE_DMA will fail to properly setup ARCH_LOW_ADDRESS_LIMIT. The limit will default to ~0ULL, effectively spanning the whole memory, which is too high for a configuration that expects low memory to be capped at 4GB. Fix ARCH_LOW_ADDRESS_LIMIT by falling back to arm64_dma32_phys_limit when arm64_dma_phys_limit isn't set. arm64_dma32_phys_limit will honour CONFIG_ZONE_DMA32, or span the entire memory when not enabled. Fixes: 1a8e1cef7603 ("arm64: use both ZONE_DMA and ZONE_DMA32") Signed-off-by: Nicolas Saenz Julienne Link: https://lore.kernel.org/r/20201218163307.10150-1-nsaenzjulienne@suse.de Signed-off-by: Catalin Marinas Signed-off-by: Greg Kroah-Hartman

Merge tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux

2020-10-23T16:46:16Z

Pull more arm64 updates from Will Deacon: "A small selection of further arm64 fixes and updates. Most of these are fixes that came in during the merge window, with the exception of the HAVE_MOVE_PMD mremap() speed-up which we discussed back in 2018 and somehow forgot to enable upstream. - Improve performance of Spectre-v2 mitigation on Falkor CPUs (if you're lucky enough to have one) - Select HAVE_MOVE_PMD. This has been shown to improve mremap() performance, which is used heavily by the Android runtime GC, and it seems we forgot to enable this upstream back in 2018. - Ensure linker flags are consistent between LLVM and BFD - Fix stale comment in Spectre mitigation rework - Fix broken copyright header - Fix KASLR randomisation of the linear map - Prevent arm64-specific prctl()s from compat tasks (return -EINVAL)" Link: https://lore.kernel.org/kvmarm/20181108181201.88826-3-joelaf@google.com/ * tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: arm64: proton-pack: Update comment to reflect new function name arm64: spectre-v2: Favour CPU-specific mitigation at EL2 arm64: link with -z norelro regardless of CONFIG_RELOCATABLE arm64: Fix a broken copyright header in gen_vdso_offsets.sh arm64: mremap speedup - Enable HAVE_MOVE_PMD arm64: mm: use single quantity to represent the PA to VA translation arm64: reject prctl(PR_PAC_RESET_KEYS) on compat tasks