kernel/linux.git/arch/microblaze/include/asm/page.h, branch linux-7.0.y

treewide: provide a generic clear_user_page() variant

2026-01-21T03:24:39+00:00

Patch series "mm: folio_zero_user: clear page ranges", v11. This series adds clearing of contiguous page ranges for hugepages. The series improves on the current discontiguous clearing approach in two ways: - clear pages in a contiguous fashion. - use batched clearing via clear_pages() wherever exposed. The first is useful because it allows us to make much better use of hardware prefetchers. The second, enables advertising the real extent to the processor. Where specific instructions support it (ex. string instructions on x86; "mops" on arm64 etc), a processor can optimize based on this because, instead of seeing a sequence of 8-byte stores, or a sequence of 4KB pages, it sees a larger unit being operated on. For instance, AMD Zen uarchs (for extents larger than LLC-size) switch to a mode where they start eliding cacheline allocation. This is helpful not just because it results in higher bandwidth, but also because now the cache is not evicting useful cachelines and replacing them with zeroes. Demand faulting a 64GB region shows performance improvement: $ perf bench mem mmap -p $pg-sz -f demand -s 64GB -l 5 baseline +series (GBps +- %stdev) (GBps +- %stdev) pg-sz=2MB 11.76 +- 1.10% 25.34 +- 1.18% [*] +115.47% preempt=* pg-sz=1GB 24.85 +- 2.41% 39.22 +- 2.32% + 57.82% preempt=none|voluntary pg-sz=1GB (similar) 52.73 +- 0.20% [#] +112.19% preempt=full|lazy [*] This improvement is because switching to sequential clearing allows the hardware prefetchers to do a much better job. [#] For pg-sz=1GB a large part of the improvement is because of the cacheline elision mentioned above. preempt=full|lazy improves upon that because, not needing explicit invocations of cond_resched() to ensure reasonable preemption latency, it can clear the full extent as a single unit. In comparison the maximum extent used for preempt=none|voluntary is PROCESS_PAGES_NON_PREEMPT_BATCH (32MB). When provided the full extent the processor forgoes allocating cachelines on this path almost entirely. (The hope is that eventually, in the fullness of time, the lazy preemption model will be able to do the same job that none or voluntary models are used for, allowing us to do away with cond_resched().) Raghavendra also tested previous version of the series on AMD Genoa and sees similar improvement [1] with preempt=lazy. $ perf bench mem map -p $page-size -f populate -s 64GB -l 10 base patched change pg-sz=2MB 12.731939 GB/sec 26.304263 GB/sec 106.6% pg-sz=1GB 26.232423 GB/sec 61.174836 GB/sec 133.2% This patch (of 8): Let's drop all variants that effectively map to clear_page() and provide it in a generic variant instead. We'll use the macro clear_user_page to indicate whether an architecture provides it's own variant. Also, clear_user_page() is only called from the generic variant of clear_user_highpage(), so define it only if the architecture does not provide a clear_user_highpage(). And, for simplicity define it in linux/highmem.h. Note that for parisc, clear_page() and clear_user_page() map to clear_page_asm(), so we can just get rid of the custom clear_user_page() implementation. There is a clear_user_page_asm() function on parisc, that seems to be unused. Not sure what's up with that. Link: https://lkml.kernel.org/r/20260107072009.1615991-1-ankur.a.arora@oracle.com Link: https://lkml.kernel.org/r/20260107072009.1615991-2-ankur.a.arora@oracle.com Signed-off-by: David Hildenbrand Co-developed-by: Ankur Arora Signed-off-by: Ankur Arora Cc: Andy Lutomirski Cc: Ankur Arora Cc: "Borislav Petkov (AMD)" Cc: Boris Ostrovsky Cc: David Hildenbrand Cc: "H. Peter Anvin" Cc: Ingo Molnar Cc: Konrad Rzessutek Wilk Cc: Lance Yang Cc: "Liam R. Howlett" Cc: Li Zhe Cc: Lorenzo Stoakes Cc: Mateusz Guzik Cc: Matthew Wilcox (Oracle) Cc: Michal Hocko Cc: Mike Rapoport Cc: Peter Zijlstra Cc: Raghavendra K T Cc: Suren Baghdasaryan Cc: Thomas Gleixner Cc: Vlastimil Babka Signed-off-by: Andrew Morton

microblaze: Replace ASSEMBLY with ASSEMBLER in non-uapi headers

2025-09-12T11:12:42+00:00

While the GCC and Clang compilers already define __ASSEMBLER__ automatically when compiling assembly code, __ASSEMBLY__ is a macro that only gets defined by the Makefiles in the kernel. This can be very confusing when switching between userspace and kernelspace coding, or when dealing with uapi headers that rather should use __ASSEMBLER__ instead. So let's standardize on the __ASSEMBLER__ macro that is provided by the compilers now. This is a completely mechanical patch (done with a simple "sed -i" statement). Cc: Michal Simek Signed-off-by: Thomas Huth Link: https://lore.kernel.org/r/20250314071013.1575167-19-thuth@redhat.com Signed-off-by: Michal Simek

Merge tag 'asm-generic-3.13' of git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic

2024-11-20T23:13:02+00:00

Pull asm-generic updates from Arnd Bergmann: "These are a number of unrelated cleanups, generally simplifying the architecture specific header files: - A series from Al Viro simplifies asm/vga.h, after it turns out that most of it can be generalized. - A series from Julian Vetter adds a common version of memcpy_{to,from}io() and memset_io() and changes most architectures to use that instead of their own implementation - A series from Niklas Schnelle concludes his work to make PC style inb()/outb() optional - Nicolas Pitre contributes improvements for the generic do_div() helper - Christoph Hellwig adds a generic version of page_to_phys() and phys_to_page(), replacing the slightly different architecture specific definitions. - Uwe Kleine-Koenig has a minor cleanup for ioctl definitions" * tag 'asm-generic-3.13' of git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic: (24 commits) empty include/asm-generic/vga.h sparc: get rid of asm/vga.h asm/vga.h: don't bother with scr_mem{cpy,move}v() unless we need to vt_buffer.h: get rid of dead code in default scr_...() instances tty: serial: export serial_8250_warn_need_ioport lib/iomem_copy: fix kerneldoc format style hexagon: simplify asm/io.h for !HAS_IOPORT loongarch: Use new fallback IO memcpy/memset csky: Use new fallback IO memcpy/memset arm64: Use new fallback IO memcpy/memset New implementation for IO memcpy and IO memset watchdog: Add HAS_IOPORT dependency for SBC8360 and SBC7240 __arch_xprod64(): make __always_inline when optimizing for performance ARM: div64: improve __arch_xprod_64() asm-generic/div64: optimize/simplify __div64_const32() lib/math/test_div64: add some edge cases relevant to __div64_const32() asm-generic: add an optional pfn_valid check to page_to_phys asm-generic: provide generic page_to_phys and phys_to_page implementations asm-generic/io.h: Remove I/O port accessors for HAS_IOPORT=n tty: serial: handle HAS_IOPORT dependencies ...

asm-generic: provide generic page_to_phys and phys_to_page implementations

2024-10-28T21:44:28+00:00

page_to_phys is duplicated by all architectures, and from some strange reason placed in where it doesn't fit at all. phys_to_page is only provided by a few architectures despite having a lot of open coded users. Provide generic versions in to make these helpers more easily usable. Note with this patch powerpc loses the CONFIG_DEBUG_VIRTUAL pfn_valid check. It will be added back in a generic version later. Signed-off-by: Christoph Hellwig Signed-off-by: Arnd Bergmann

vdso: Introduce vdso/page.h

2024-10-15T22:13:04+00:00

The VDSO implementation includes headers from outside of the vdso/ namespace. Introduce vdso/page.h to make sure that the generic library uses only the allowed namespace. Signed-off-by: Vincenzo Frascino Signed-off-by: Thomas Gleixner Reviewed-by: Arnd Bergmann Acked-by: Geert Uytterhoeven # m68k Link: https://lore.kernel.org/all/20241014151340.1639555-3-vincenzo.frascino@arm.com

arch: define CONFIG_PAGE_SIZE_*KB on all architectures

2024-03-06T18:29:09+00:00

Most architectures only support a single hardcoded page size. In order to ensure that each one of these sets the corresponding Kconfig symbols, change over the PAGE_SHIFT definition to the common one and allow only the hardware page size to be selected. Acked-by: Guo Ren Acked-by: Heiko Carstens Acked-by: Stafford Horne Acked-by: Johannes Berg Acked-by: Geert Uytterhoeven Reviewed-by: Thomas Gleixner Signed-off-by: Arnd Bergmann

microblaze: Make virt_to_pfn() a static inline

2023-08-23T07:35:22+00:00

Making virt_to_pfn() a static inline taking a strongly typed (const void *) makes the contract of a passing a pointer of that type to the function explicit and exposes any misuse of the macro virt_to_pfn() acting polymorphic and accepting many types such as (void *), (unitptr_t) or (unsigned long) as arguments without warnings. Move the function down in the file so __pa() exists in our scope, and it compiles. This in turn requires moving __pa() as it depends on __virt_to_phys() that was below. (Lazy macro evaluation conflicts with strict function ordering.) Make a symmetric change to pfn_to_virt() so we have type checking both ways. Due to this the file being included into some assembly files, some further inclusion guards are needed to make sure assembly keeps compiling. Signed-off-by: Linus Walleij Link: https://lore.kernel.org/r/20230808-virt-to-phys-microblaze-v1-1-e6df710fe0a1@linaro.org Signed-off-by: Michal Simek

microblaze: move the ARCH_{DMA,SLAB}_MINALIGN definitions to asm/cache.h

2023-06-19T23:19:24+00:00

The microblaze architecture defines ARCH_DMA_MINALIGN in asm/page.h. Move it to asm/cache.h to allow a generic ARCH_DMA_MINALIGN definition in linux/cache.h without redefine errors/warnings. While at it, also move ARCH_SLAB_MINALIGN to asm/cache.h for consistency. Link: https://lkml.kernel.org/r/20230613155245.1228274-3-catalin.marinas@arm.com Signed-off-by: Catalin Marinas Cc: Michal Simek Cc: Christophe Leroy Cc: John Paul Adrian Glaubitz Cc: kernel test robot Cc: Michael Ellerman Cc: Nicholas Piggin Cc: Rich Felker Cc: Vlastimil Babka Cc: Yoshinori Sato Signed-off-by: Andrew Morton

mm, arch: add generic implementation of pfn_valid() for FLATMEM

2023-02-10T00:51:41+00:00

Every architecture that supports FLATMEM memory model defines its own version of pfn_valid() that essentially compares a pfn to max_mapnr. Use mips/powerpc version implemented as static inline as a generic implementation of pfn_valid() and drop its per-architecture definitions. [rppt@kernel.org: fix the generic pfn_valid()] Link: https://lkml.kernel.org/r/Y9lg7R1Yd931C+y5@kernel.org Link: https://lkml.kernel.org/r/20230129124235.209895-5-rppt@kernel.org Signed-off-by: Mike Rapoport (IBM) Acked-by: Arnd Bergmann Acked-by: Guo Ren [csky] Acked-by: Huacai Chen [LoongArch] Acked-by: Stafford Horne [OpenRISC] Acked-by: Michael Ellerman [powerpc] Reviewed-by: David Hildenbrand Tested-by: Conor Dooley Cc: Brian Cain Cc: "David S. Miller" Cc: Dinh Nguyen Cc: Geert Uytterhoeven Cc: Greg Ungerer Cc: Helge Deller Cc: Huacai Chen Cc: Matt Turner Cc: Max Filippov Cc: Michal Simek Cc: Palmer Dabbelt Cc: Richard Weinberger Cc: Rich Felker Cc: Russell King Cc: Thomas Bogendoerfer Cc: Vineet Gupta Cc: WANG Xuerui Cc: Yoshinori Sato Signed-off-by: Andrew Morton

mm/page_alloc: always initialize memory map for the holes

2021-09-03T16:58:15+00:00

Patch series "mm: ensure consistency of memory map poisoning". Currently memory map allocation for FLATMEM case does not poison the struct pages regardless of CONFIG_PAGE_POISON setting. This happens because allocation of the memory map for FLATMEM and SPARSMEM use different memblock functions and those that are used for SPARSMEM case (namely memblock_alloc_try_nid_raw() and memblock_alloc_exact_nid_raw()) implicitly poison the allocated memory. Another side effect of this implicit poisoning is that early setup code that uses the same functions to allocate memory burns cycles for the memory poisoning even if it was not intended. These patches introduce memmap_alloc() wrapper that ensure that the memory map allocation is consistent for different memory models. This patch (of 4): Currently memory map for the holes is initialized only when SPARSEMEM memory model is used. Yet, even with FLATMEM there could be holes in the physical memory layout that have memory map entries. For instance, the memory reserved using e820 API on i386 or "reserved-memory" nodes in device tree would not appear in memblock.memory and hence the struct pages for such holes will be skipped during memory map initialization. These struct pages will be zeroed because the memory map for FLATMEM systems is allocated with memblock_alloc_node() that clears the allocated memory. While zeroed struct pages do not cause immediate problems, the correct behaviour is to initialize every page using __init_single_page(). Besides, enabling page poison for FLATMEM case will trigger PF_POISONED_CHECK() unless the memory map is properly initialized. Make sure init_unavailable_range() is called for both SPARSEMEM and FLATMEM so that struct pages representing memory holes would appear as PG_Reserved with any memory layout. [rppt@kernel.org: fix microblaze] Link: https://lkml.kernel.org/r/YQWW3RCE4eWBuMu/@kernel.org Link: https://lkml.kernel.org/r/20210714123739.16493-1-rppt@kernel.org Link: https://lkml.kernel.org/r/20210714123739.16493-2-rppt@kernel.org Signed-off-by: Mike Rapoport Acked-by: David Hildenbrand Tested-by: Guenter Roeck Cc: Michal Simek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds