kernel/linux.git/arch/openrisc/include, branch v6.12.80

openrisc: define arch-specific version of nop()

2026-03-04T12:21:14+00:00

[ Upstream commit 0dfffa5479d6260d04d021f69203b1926f73d889 ] When compiling a driver written for MIPS on OpenRISC that uses the nop() function, it fails due to the following error: drivers/watchdog/pic32-wdt.c: Assembler messages: drivers/watchdog/pic32-wdt.c:125: Error: unrecognized instruction `nop' The driver currently uses the generic version of nop() from include/asm-generic/barrier.h: #ifndef nop #define nop() asm volatile ("nop") #endif Let's fix this on OpenRISC by defining an architecture-specific version of nop(). This was tested by performing an allmodconfig openrisc cross compile on an aarch64 host. Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202601180236.BVy480We-lkp@intel.com/ Signed-off-by: Brian Masney Signed-off-by: Stafford Horne Signed-off-by: Sasha Levin

openrisc: Implement fixmap to fix earlycon

2024-12-05T13:01:27+00:00

[ Upstream commit 1037d186edfc551fa7ba2d4336e74e7575a07a65 ] With commit 53c98e35dcbc ("openrisc: mm: remove unneeded early ioremap code") it was commented that early ioremap was not used in OpenRISC. I acked this but was wrong, earlycon was using it. Earlycon setup now fails with the below trace: Kernel command line: earlycon ------------[ cut here ]------------ WARNING: CPU: 0 PID: 0 at mm/ioremap.c:23 generic_ioremap_prot+0x118/0x130 Modules linked in: CPU: 0 UID: 0 PID: 0 Comm: swapper Not tainted 6.11.0-rc5-00001-gce02fd891c38-dirty #141 Call trace: [<(ptrval)>] dump_stack_lvl+0x7c/0x9c [<(ptrval)>] dump_stack+0x1c/0x2c [<(ptrval)>] __warn+0xb4/0x108 [<(ptrval)>] ? generic_ioremap_prot+0x118/0x130 [<(ptrval)>] warn_slowpath_fmt+0x60/0x98 [<(ptrval)>] generic_ioremap_prot+0x118/0x130 [<(ptrval)>] ioremap_prot+0x20/0x30 [<(ptrval)>] of_setup_earlycon+0xd4/0x2e0 [<(ptrval)>] early_init_dt_scan_chosen_stdout+0x18c/0x1c8 [<(ptrval)>] param_setup_earlycon+0x3c/0x60 [<(ptrval)>] do_early_param+0xb0/0x118 [<(ptrval)>] parse_args+0x184/0x4b8 [<(ptrval)>] ? start_kernel+0x0/0x78c [<(ptrval)>] parse_early_options+0x40/0x50 [<(ptrval)>] ? do_early_param+0x0/0x118 [<(ptrval)>] parse_early_param+0x48/0x68 [<(ptrval)>] ? start_kernel+0x318/0x78c [<(ptrval)>] ? start_kernel+0x0/0x78c ---[ end trace 0000000000000000 ]--- To fix this we could either implement early_ioremap again or implement fixmap. In this patch we choose the later option of implementing basic fixmap support. While fixing this we also remove the old FIX_IOREMAP slots that were used by early ioremap code. That code was also removed by commit 53c98e35dcbc ("openrisc: mm: remove unneeded early ioremap code") but these definitions were not cleaned up. Fixes: 53c98e35dcbc ("openrisc: mm: remove unneeded early ioremap code") Signed-off-by: Stafford Horne Signed-off-by: Sasha Levin

openrisc: convert to generic syscall table

2024-07-10T12:23:38+00:00

The uapi/asm/unistd_32.h and asm/syscall_table_32.h headers can now be generated from scripts/syscall.tbl, which makes this consistent with the other architectures that have their own syscall.tbl. openrisc has one extra system call that gets added to scripts/syscall.tbl. The time32, stat64, rlimit and renameat entries in the syscall_abis_32 line are for system calls that were part of the generic ABI when arch/nios2 got added but are no longer enabled by default for new architectures. Both the user visible side of asm/unistd.h and the internal syscall table in the kernel should have the same effective contents after this. When asm/syscalls.h is included in kernel/fork.c for the purpose of type checking, the redirection macros cause problems. Move these so only the references get redirected. Signed-off-by: Arnd Bergmann

clone3: drop __ARCH_WANT_SYS_CLONE3 macro

2024-07-10T12:23:38+00:00

When clone3() was introduced, it was not obvious how each architecture deals with setting up the stack and keeping the register contents in a fork()-like system call, so this was left for the architecture maintainers to implement, with __ARCH_WANT_SYS_CLONE3 defined by those that already implement it. Five years later, we still have a few architectures left that are missing clone3(), and the macro keeps getting in the way as it's fundamentally different from all the other __ARCH_WANT_SYS_* macros that are meant to provide backwards-compatibility with applications using older syscalls that are no longer provided by default. Address this by reversing the polarity of the macro, adding an __ARCH_BROKEN_SYS_CLONE3 macro to all architectures that don't already provide the syscall, and remove __ARCH_WANT_SYS_CLONE3 from all the other ones. Acked-by: Geert Uytterhoeven Signed-off-by: Arnd Bergmann

openrisc: Move FPU state out of pt_regs

2024-04-15T14:20:39+00:00

My original, naive, FPU support patch had the FPCSR register stored during both the *mode switch* and *context switch*. This is wasteful. Also, the original patches did not save the FPU state when handling signals during the system call fast path. We fix this by moving the FPCSR state to thread_struct in task_struct. We also introduce new helper functions save_fpu and restore_fpu which can be used to sync the FPU with thread_struct. These functions are now called when needed: - Setting up and restoring sigcontext when handling signals - Before and after __switch_to during context switches - When handling FPU exceptions - When reading and writing FPU register sets In the future we can further optimize this by doing lazy FPU save and restore. For example, FPU sync is not needed when switching to and from kernel threads (x86 does this). FPU save and restore does not need to be done two times if we have both rescheduling and signal work to do. However, since OpenRISC FPU state is a single register, I leave these optimizations for future consideration. Signed-off-by: Stafford Horne

openrisc: Define openrisc relocation types

2024-04-15T14:20:39+00:00

This defines the current OpenRISC relocation types using the current R_OR1K_* naming conventions. The old R_OR32_* definitions are left for backwards compatibility. Note, the R_OR32_VTENTRY and R_OR32_VTINHERIT macros were defined with the wrong values the have always been 7 and 8 respectively, not 8 and 7. They are not used for module loading and I have updated them to use the correct values. Signed-off-by: Stafford Horne

Merge tag 'for-linus' of https://github.com/openrisc/linux

2024-03-14T22:53:10+00:00

Pull OpenRISC updates from Stafford Horne: "Just a few cleanups and updates that were sent in: - Replace asm/fixmap.h with asm-generic version - Fix to move memblock setup up before it's used during init" * tag 'for-linus' of https://github.com/openrisc/linux: openrisc: Use asm-generic's version of fix_to_virt() & virt_to_fix() openrisc: Call setup_memory() earlier in the init sequence

openrisc: Use asm-generic's version of fix_to_virt() & virt_to_fix()

2024-03-10T08:55:46+00:00

Openrisc's implementation of fix_to_virt() & virt_to_fix() share same functionality with ones of asm generic. Plus, generic version of fix_to_virt() can trap invalid index at compile time. Thus, Replace the arch-specific implementations with asm generic's ones. Signed-off-by: Dawei Li Signed-off-by: Stafford Horne

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

mm: Remove broken pfn_to_virt() on arch csky/hexagon/openrisc

2024-03-05T17:04:20+00:00

Remove the broken pfn_to_virt() on architectures csky/hexagon/openrisc. The pfn_to_virt() on those architectures takes PFN instead of PA as the input to macro __va(), with PAGE_SHIFT applying to the converted VA, which is not right, especially when there's a non-zero offset in __va(). [1] The broken pfn_to_virt() was noticed when checking how page_to_virt() is unfolded on x86. It turns out that the one in linux/mm.h instead of in asm-generic/page.h is compiled for x86. However, page_to_virt() in asm-generic/page.h is found out expanding to pfn_to_virt() with a bug described above. The pfn_to_virt() is found out not right as well on architectures csky/hexagon/openrisc. Since there's no single caller on csky/hexagon/openrisc to pfn_to_virt() and there are functions doing similar things as pfn_to_virt() -- - pfn_to_kaddr() in asm/page.h for csky, - page_to_virt() in asm/page.h for hexagon, and - page_to_virt() in linux/mm.h for openrisc, just delete the pfn_to_virt() on those architectures. The pfn_to_virt() in asm-generic/page.h is not touched in this patch as it's referenced by page_to_virt() in that header while the whole header is going to be removed as a whole due to no one including it. Link:https://lore.kernel.org/all/20240131055159.2506-1-yan.y.zhao@intel.com [1] Cc: Linus Walleij Suggested-by: Arnd Bergmann Signed-off-by: Yan Zhao Signed-off-by: Arnd Bergmann