Message ID | 20220616040924.1022607-1-anshuman.khandual@arm.com (mailing list archive) |
---|---|
Headers | show |
Series | mm/mmap: Drop __SXXX/__PXXX macros from across platforms | expand |
Le 16/06/2022 à 06:09, Anshuman Khandual a écrit : > __SXXX/__PXXX macros is an unnecessary abstraction layer in creating the > generic protection_map[] array which is used for vm_get_page_prot(). This > abstraction layer can be avoided, if the platforms just define the array > protection_map[] for all possible vm_flags access permission combinations. > > This series drops __SXXX/__PXXX macros from across platforms in the tree. > First it makes protection_map[] array private (static) on platforms which > enable ARCH_HAS_VM_GET_PAGE_PROT, later moves protection_map[] array into > arch for all remaining platforms (!ARCH_HAS_VM_GET_PAGE_PROT), dropping > the generic one. In the process __SXXX/__PXXX macros become redundant and > thus get dropped off completely. I understand that the diff stat is large > here, but please do suggest if there is a better way. This series applies > on v5.19-rc1 and has been build tested for multiple platforms. Maybe this patch could be split with one patch per architecture. All you have to do for that is to guard the generic protection_map declaration with #ifdef __S000 , then the architectures can be migrated one by one. > > The CC list for this series has been reduced to just minimum, until there > is some initial agreement. Agreement with who if people don't know this series exists ? I think you should keep the architecture lists in copy allthough you don't include individual maintainers/reviewers for now. > > - Anshuman > > Changes in V3: > > - Fix build issues on powerpc and riscv > > Changes in V2: I guess V2 was only sent to linux-mm as well ? Too bad. > > https://lore.kernel.org/all/20220613053354.553579-1-anshuman.khandual@arm.com/ > > - Add 'const' identifier to protection_map[] on powerpc > - Dropped #ifndef CONFIG_ARCH_HAS_VM_GET_PAGE_PROT check from sparc 32 > - Dropped protection_map[] init from sparc 64 > - Dropped all new platform changes subscribing ARCH_HAS_VM_GET_PAGE_PROT > - Added a second patch which moves generic protection_map[] array into > all remaining platforms (!ARCH_HAS_VM_GET_PAGE_PROT) > > Changes in V1: > > https://lore.kernel.org/linux-mm/20220603101411.488970-1-anshuman.khandual@arm.com/ > > Cc: Andrew Morton <akpm@linux-foundation.org> > Cc: linux-mm@kvack.org > Cc: linux-kernel@vger.kernel.org > > Anshuman Khandual (2): > mm/mmap: Restrict generic protection_map[] array visibility > mm/mmap: Drop generic protection_map[] array > > arch/alpha/include/asm/pgtable.h | 17 ------- > arch/alpha/mm/init.c | 21 +++++++++ > arch/arc/include/asm/pgtable-bits-arcv2.h | 18 -------- > arch/arc/mm/mmap.c | 19 ++++++++ > arch/arm/include/asm/pgtable.h | 17 ------- > arch/arm/lib/uaccess_with_memcpy.c | 2 +- > arch/arm/mm/mmu.c | 19 ++++++++ > arch/arm64/include/asm/pgtable-prot.h | 18 -------- > arch/arm64/mm/mmap.c | 21 +++++++++ > arch/csky/include/asm/pgtable.h | 18 -------- > arch/csky/mm/init.c | 19 ++++++++ > arch/hexagon/include/asm/pgtable.h | 27 ------------ > arch/hexagon/mm/init.c | 41 +++++++++++++++++ > arch/ia64/include/asm/pgtable.h | 18 -------- > arch/ia64/mm/init.c | 27 +++++++++++- > arch/loongarch/include/asm/pgtable-bits.h | 19 -------- > arch/loongarch/mm/cache.c | 45 +++++++++++++++++++ > arch/m68k/include/asm/mcf_pgtable.h | 54 ----------------------- > arch/m68k/include/asm/motorola_pgtable.h | 22 --------- > arch/m68k/include/asm/sun3_pgtable.h | 17 ------- > arch/m68k/mm/mcfmmu.c | 54 +++++++++++++++++++++++ > arch/m68k/mm/motorola.c | 19 ++++++++ > arch/m68k/mm/sun3mmu.c | 19 ++++++++ > arch/microblaze/include/asm/pgtable.h | 17 ------- > arch/microblaze/mm/init.c | 19 ++++++++ > arch/mips/include/asm/pgtable.h | 22 --------- > arch/mips/mm/cache.c | 2 + > arch/nios2/include/asm/pgtable.h | 16 ------- > arch/nios2/mm/init.c | 19 ++++++++ > arch/openrisc/include/asm/pgtable.h | 18 -------- > arch/openrisc/mm/init.c | 19 ++++++++ > arch/parisc/include/asm/pgtable.h | 18 -------- > arch/parisc/mm/init.c | 19 ++++++++ > arch/powerpc/include/asm/pgtable.h | 18 -------- > arch/powerpc/mm/book3s64/pgtable.c | 6 +++ > arch/powerpc/mm/pgtable.c | 20 +++++++++ > arch/riscv/include/asm/pgtable.h | 20 --------- > arch/riscv/mm/init.c | 19 ++++++++ > arch/s390/include/asm/pgtable.h | 17 ------- > arch/s390/mm/mmap.c | 19 ++++++++ > arch/sh/include/asm/pgtable.h | 17 ------- > arch/sh/mm/mmap.c | 19 ++++++++ > arch/sparc/include/asm/pgtable_32.h | 19 -------- > arch/sparc/include/asm/pgtable_64.h | 19 -------- > arch/sparc/mm/init_32.c | 19 ++++++++ > arch/sparc/mm/init_64.c | 3 ++ > arch/um/include/asm/pgtable.h | 17 ------- > arch/um/kernel/mem.c | 19 ++++++++ > arch/x86/include/asm/pgtable_types.h | 19 -------- > arch/x86/mm/pgprot.c | 19 ++++++++ > arch/x86/um/mem_32.c | 2 +- > arch/xtensa/include/asm/pgtable.h | 18 -------- > arch/xtensa/mm/init.c | 19 ++++++++ > include/linux/mm.h | 2 + > mm/mmap.c | 19 -------- > 55 files changed, 547 insertions(+), 522 deletions(-) >
On Thu, Jun 16, 2022 at 05:22:23AM +0000, Christophe Leroy wrote: > I think you should keep the architecture lists in copy allthough you > don't include individual maintainers/reviewers for now. That's what I tend to do for big global changes.
On 6/16/22 10:52, Christophe Leroy wrote: > > > Le 16/06/2022 à 06:09, Anshuman Khandual a écrit : >> __SXXX/__PXXX macros is an unnecessary abstraction layer in creating the >> generic protection_map[] array which is used for vm_get_page_prot(). This >> abstraction layer can be avoided, if the platforms just define the array >> protection_map[] for all possible vm_flags access permission combinations. >> >> This series drops __SXXX/__PXXX macros from across platforms in the tree. >> First it makes protection_map[] array private (static) on platforms which >> enable ARCH_HAS_VM_GET_PAGE_PROT, later moves protection_map[] array into >> arch for all remaining platforms (!ARCH_HAS_VM_GET_PAGE_PROT), dropping >> the generic one. In the process __SXXX/__PXXX macros become redundant and >> thus get dropped off completely. I understand that the diff stat is large >> here, but please do suggest if there is a better way. This series applies >> on v5.19-rc1 and has been build tested for multiple platforms. > > Maybe this patch could be split with one patch per architecture. All you > have to do for that is to guard the generic protection_map declaration > with #ifdef __S000 , then the architectures can be migrated one by one. > >> >> The CC list for this series has been reduced to just minimum, until there >> is some initial agreement. > > Agreement with who if people don't know this series exists ? > > I think you should keep the architecture lists in copy allthough you > don't include individual maintainers/reviewers for now. Sure, will do. > >> >> - Anshuman >> >> Changes in V3: >> >> - Fix build issues on powerpc and riscv >> >> Changes in V2: > > I guess V2 was only sent to linux-mm as well ? Too bad. I was in a dilemma, whether to first arrive at something more acceptable or just engage all stake holders from the beginning. I understand your concern and hence will copy architecture mailing lists from next time onward.