Alex for next manual#667
Closed
AlexGhiti wants to merge 58 commits into
Closed
Conversation
AlexGhiti
force-pushed
the
alex-for-next_manual
branch
2 times, most recently
from
aa63643 to
43be0d0
Compare
AlexGhiti
force-pushed
the
alex-for-next_manual
branch
2 times, most recently
from
9a0a9ac to
f5ccecb
Compare
__ASSEMBLY__ is only defined by the Makefile of the kernel, so this is not really useful for uapi headers (unless the userspace Makefile defines it, too). Let's switch to __ASSEMBLER__ which gets set automatically by the compiler when compiling assembly code. This is a completely mechanical patch (done with a simple "sed -i" statement). Cc: Paul Walmsley <paul.walmsley@sifive.com> Cc: Palmer Dabbelt <palmer@dabbelt.com> Cc: Albert Ou <aou@eecs.berkeley.edu> Cc: Alexandre Ghiti <alex@ghiti.fr> Cc: linux-riscv@lists.infradead.org Signed-off-by: Thomas Huth <thuth@redhat.com> Link: https://lore.kernel.org/r/20250606070952.498274-2-thuth@redhat.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
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 originally was a completely mechanical patch (done with a simple "sed -i" statement), with some manual fixups during rebasing of the patch later. Cc: Paul Walmsley <paul.walmsley@sifive.com> Cc: Palmer Dabbelt <palmer@dabbelt.com> Cc: Albert Ou <aou@eecs.berkeley.edu> Cc: Alexandre Ghiti <alex@ghiti.fr> Cc: linux-riscv@lists.infradead.org Signed-off-by: Thomas Huth <thuth@redhat.com> Link: https://lore.kernel.org/r/20250606070952.498274-3-thuth@redhat.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
… header files" Thomas Huth <thuth@redhat.com> says: The kernel Makefiles define the __ASSEMBLY__ macro to provide a way to use headers in both, assembly and C source code. However, all the supported versions of the GCC and Clang compilers also define the macro __ASSEMBLER__ automatically already when compiling assembly code, so some kernel headers are using __ASSEMBLER__ instead. With regards to userspace code, this seems also to be constant source of confusion, see for example these links here: https://lore.kernel.org/kvm/20250222014526.2302653-1-seanjc@google.com/ https://stackoverflow.com/questions/28924355/gcc-assembler-preprocessor-not-compatible-with-standard-headers https://forums.raspberrypi.com/viewtopic.php?p=1652944#p1653834 riscv-software-src/opensbi#199 To avoid confusion in the future, it would make sense to standardize on the macro that gets defined by the compiler, so this patch series changes all occurances of __ASSEMBLY__ into __ASSEMBLER__. I split the patches per architecture to ease the review, and I also split the uapi headers from the normal ones in case we decide that uapi needs to be treated differently from the normal headers here. The x86 and parisc patches already got merged via their specific architecture tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=24a295e4ef1ca8 https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=8a141be3233af7 https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=cccaea1d66e94b https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=e2b6a188625a2b * patches from https://lore.kernel.org/r/20250606070952.498274-1-thuth@redhat.com: riscv: Replace __ASSEMBLY__ with __ASSEMBLER__ in non-uapi headers riscv: Replace __ASSEMBLY__ with __ASSEMBLER__ in uapi headers Link: https://lore.kernel.org/r/20250606070952.498274-1-thuth@redhat.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
%.pi.o files are built as prerequisites of other objects. There is no need to use extra-y, which is planned for deprecation. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org> Link: https://lore.kernel.org/r/20250602181023.528550-1-masahiroy@kernel.org Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
These extensions depends on the F one. Add a validation callback checking for the F extension to be present. Now that extensions are correctly reported using the F/D presence, we can remove the has_fpu() check in hwprobe_isa_ext0(). Signed-off-by: Clément Léger <cleger@rivosinc.com> Reviewed-by: Conor Dooley <conor.dooley@microchip.com> Link: https://lore.kernel.org/r/20250527100001.33284-1-cleger@rivosinc.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Simply fix a typo. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Andrew Jones <ajones@ventanamicro.com> Reviewed-By: Clément Léger <cleger@rivosinc.com> Link: https://lore.kernel.org/r/20250620-dev-alex-insn_duplicate_v5_manual-v5-1-d865dc9ad180@rivosinc.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
RV_X() macro is defined in two different ways which is error prone. So harmonize its first definition and add another macro RV_X_mask() for the second one. Reviewed-by: Andrew Jones <ajones@ventanamicro.com> Reviewed-by: Clément Léger <cleger@rivosinc.com> Link: https://lore.kernel.org/r/20250620-dev-alex-insn_duplicate_v5_manual-v5-2-d865dc9ad180@rivosinc.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
kernel/traps_misaligned.c and kvm/vcpu_insn.c define the same macros to extract information from the instructions. Let's move the definitions into asm/insn.h to avoid this duplication. Reviewed-by: Andrew Jones <ajones@ventanamicro.com> Reviewed-by: Clément Léger <cleger@rivosinc.com> Link: https://lore.kernel.org/r/20250620-dev-alex-insn_duplicate_v5_manual-v5-3-d865dc9ad180@rivosinc.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Alexandre Ghiti <alexghiti@rivosinc.com> says: The instructions parsing macros were duplicated and one of them had different implementations, which is error prone. So let's consolidate those macros in asm/insn.h. * patches from https://lore.kernel.org/r/20250620-dev-alex-insn_duplicate_v5_manual-v5-0-d865dc9ad180@rivosinc.com: riscv: Move all duplicate insn parsing macros into asm/insn.h riscv: Strengthen duplicate and inconsistent definition of RV_X() riscv: Fix typo EXRACT -> EXTRACT Link: https://lore.kernel.org/r/20250620-dev-alex-insn_duplicate_v5_manual-v5-0-d865dc9ad180@rivosinc.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Similar to other instruction-processing macros/functions, branch_rs2_idx should be in insn.h. Move it into insn.h as RV_EXTRACT_RS2_REG. This new name matches the style in insn.h. Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Nam Cao <namcao@linutronix.de> Link: https://lore.kernel.org/r/107d4a6c1818bf169be2407b273a0483e6d55bbb.1747215274.git.namcao@linutronix.de Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Similar to other instruction-processing macros/functions, branch_funct3 should be in insn.h. Move it into insn.h as RV_EXTRACT_FUNCT3. This new name matches the style in insn.h. Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Nam Cao <namcao@linutronix.de> Link: https://lore.kernel.org/r/200c29a26338f19d09963fa02562787e8cfa06f2.1747215274.git.namcao@linutronix.de Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Use RV_EXTRACT_JTYPE_IMM, instead of reimplementing it in simulate_jal(). Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Nam Cao <namcao@linutronix.de> Link: https://lore.kernel.org/r/af502036738d381c6bdb96a236d21bab8c343f74.1747215274.git.namcao@linutronix.de Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Use RV_EXTRACT_RS1_REG instead of reimplementing its code. Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Nam Cao <namcao@linutronix.de> Link: https://lore.kernel.org/r/b441038c991da11a7a48ea7140ab00e3bb119387.1747215274.git.namcao@linutronix.de Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Use RV_EXTRACT_BTYPE_IMM, instead of reimplementing it in simulate_branch(). Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Nam Cao <namcao@linutronix.de> Link: https://lore.kernel.org/r/07dde8313b787261f07ae25240c97c514cd02d11.1747215274.git.namcao@linutronix.de Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Use RVC_EXTRACT_JTYPE_IMM, instead of reimplementing it in simulate_c_j(). Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Nam Cao <namcao@linutronix.de> Link: https://lore.kernel.org/r/24497deaab06d6b12cb84923606ec26f67e25424.1747215274.git.namcao@linutronix.de Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Use RVC_EXTRACT_C2_RS1_REG, instead of reimplementing it in simulate_c_jr_jalr(). Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Nam Cao <namcao@linutronix.de> Link: https://lore.kernel.org/r/d56955cd683411c6d2f63d13c78e0572462a3269.1747215274.git.namcao@linutronix.de Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Use RVC_EXTRACT_BTYPE_IMM, instead of reimplementing it in simulate_c_bnez_beqz(). Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Nam Cao <namcao@linutronix.de> Link: https://lore.kernel.org/r/8a8ed970f279fa5f24c90d840c2130e37bc6d16e.1747215274.git.namcao@linutronix.de Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Use RV_EXTRACT_RD_REG, instead of reimplementing its code. Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Nam Cao <namcao@linutronix.de> Link: https://lore.kernel.org/r/b31e5b41df5839a76103348e54dc034c8a43447a.1747215274.git.namcao@linutronix.de Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Use RV_EXTRACT_UTYPE_IMM, instead of reimplementing it in simulate_auipc(). Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Nam Cao <namcao@linutronix.de> Link: https://lore.kernel.org/r/8f0defce9f1f23f1b44bb9750ed083cfc124213c.1747215274.git.namcao@linutronix.de Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Use RV_EXTRACT_ITYPE_IMM, instead of re-implementing it in simulate_jalr(). Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Nam Cao <namcao@linutronix.de> Link: https://lore.kernel.org/r/8ae34e966c312ae5cf6c09a35ddc290cce942208.1747215274.git.namcao@linutronix.de Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Nam Cao <namcao@linutronix.de> says: Hi, There is some instruction-processing code in kprobes simulate code. These code should be insn.h. In fact, most of them is duplicating insn.h. This series remove the duplicated bits and make use of macros already defined in insn.h. The non-duplicated bits are moved into insn.h. * patches from https://lore.kernel.org/r/cover.1747215274.git.namcao@linutronix.de: riscv: kprobes: Remove duplication of RV_EXTRACT_ITYPE_IMM riscv: kprobes: Remove duplication of RV_EXTRACT_UTYPE_IMM riscv: kprobes: Remove duplication of RV_EXTRACT_RD_REG riscv: kprobes: Remove duplication of RVC_EXTRACT_BTYPE_IMM riscv: kprobes: Remove duplication of RVC_EXTRACT_C2_RS1_REG riscv: kproves: Remove duplication of RVC_EXTRACT_JTYPE_IMM riscv: kprobes: Remove duplication of RV_EXTRACT_BTYPE_IMM riscv: kprobes: Remove duplication of RV_EXTRACT_RS1_REG riscv: kprobes: Remove duplication of RV_EXTRACT_JTYPE_IMM riscv: kprobes: Move branch_funct3 to insn.h riscv: kprobes: Move branch_rs2_idx to insn.h Link: https://lore.kernel.org/r/cover.1747215274.git.namcao@linutronix.de Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Add KUnit test for riscv kprobes, mostly for simulated instructions. The test install kprobes into multiple sample functions, and check that these functions still return the expected magic value. This test can detect some kprobe bugs reported in the past (in Link:). Link: https://lore.kernel.org/linux-riscv/20241119111056.2554419-1-namcao@linutronix.de/ Link: https://lore.kernel.org/stable/c7e463c0-8cad-4f4e-addd-195c06b7b6de@iscas.ac.cn/ Link: https://lore.kernel.org/linux-riscv/20230829182500.61875-1-namcaov@gmail.com/ Signed-off-by: Nam Cao <namcao@linutronix.de> Tested-by: Alexandre Ghiti <alexghiti@rivosinc.com> Link: https://lore.kernel.org/r/20250513151631.3520793-1-namcao@linutronix.de Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
The variable type of offset should be consistent with the relevant interfaces of mmap which described in commit 295f100 ("syscalls: mmap(): use unsigned offset type consistently"). Otherwise, a user input with the top bit set would result in a negative page offset rather than a large one. [ alex: Add missing " in commit log ] Signed-off-by: Jessica Liu <liu.xuemei1@zte.com.cn> Reviewed-by: Nutty Liu <liujingqi@lanxincomputing.com> Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Link: https://lore.kernel.org/r/20250707193411886Kc-TWknP0PER2_sEg-byb@zte.com.cn Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Move vendor errata definitions into errata_list_vendors.h. Signed-off-by: Guo Ren (Alibaba DAMO Academy) <guoren@kernel.org> Tested-by: Han Gao <rabenda.cn@gmail.com> Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Link: https://lore.kernel.org/r/20250713155321.2064856-2-guoren@kernel.org Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
The early version of XuanTie C910 core has a store merge buffer delay problem. The store merge buffer could improve the store queue performance by merging multi-store requests, but when there are not continued store requests, the prior single store request would be waiting in the store queue for a long time. That would cause significant problems for communication between multi-cores. This problem was found on sg2042 & th1520 platforms with the qspinlock lock torture test. So appending a fence w.o could immediately flush the store merge buffer and let other cores see the write result. This will apply the WRITE_ONCE errata to handle the non-standard behavior via appending a fence w.o instruction for WRITE_ONCE(). This problem is only observed on the sg2042 hardware platform by running the lock_torture test program for half an hour. The problem was not found in the user space application, because interrupt can break the livelock. Reviewed-by: Leonardo Bras <leobras@redhat.com> Signed-off-by: Guo Ren (Alibaba DAMO Academy) <guoren@kernel.org> Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Tested-by: Han Gao <rabenda.cn@gmail.com> Link: https://lore.kernel.org/r/20250713155321.2064856-3-guoren@kernel.org Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
These two C files don't reference things defined in simd.h or types.h so remove these redundant #inclusions. Fixes: 6093faa ("raid6: Add RISC-V SIMD syndrome and recovery calculations") Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Chunyan Zhang <zhangchunyan@iscas.ac.cn> Reviewed-by: Nutty Liu <liujingqi@lanxincomputing.com> Link: https://lore.kernel.org/r/20250718072711.3865118-2-zhangchunyan@iscas.ac.cn Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
The implement of cmpxchg() in riscv is based on atomic primitives and has NMI-safe features, so it can be used safely in the in_nmi context. ftrace's ringbuffer relies on NMI-safe cmpxchg() in the NMI context. Currently, in_nmi() is true when riscv kprobe is in trap-based mode, so this config needs to be selected, otherwise kprobetrace will not be available. Signed-off-by: Pu Lehui <pulehui@huawei.com> Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Tested-by: Alexandre Ghiti <alexghiti@rivosinc.com> Link: https://lore.kernel.org/r/20250711090443.1688404-1-pulehui@huaweicloud.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
This selftest tests all the currently emulated instructions (except for the RV32 compressed ones which are left as a future exercise for a RV32 user). For the FPU instructions, all the FPU registers are tested. Signed-off-by: Clément Léger <cleger@rivosinc.com> Tested-by: Alexandre Ghiti <alexghiti@rivosinc.com> Link: https://lore.kernel.org/r/20250711131925.112485-1-cleger@rivosinc.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
guoren@kernel.org <guoren@kernel.org> says: From: "Guo Ren (Alibaba DAMO Academy)" <guoren@kernel.org> The early version of XuanTie C9xx cores has a store merge buffer delay problem. The store merge buffer could improve the store queue performance by merging multi-store requests, but when there are not continued store requests, the prior single store request would be waiting in the store queue for a long time. That would cause significant problems for communication between multi-cores. This problem was found on sg2042 & th1520 platforms with the qspinlock lock torture test. * patches from https://lore.kernel.org/r/20250713155321.2064856-1-guoren@kernel.org: riscv: errata: Add ERRATA_THEAD_WRITE_ONCE fixup riscv: Move vendor errata definitions to new header Link: https://lore.kernel.org/r/20250713155321.2064856-1-guoren@kernel.org Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
…port" Chunyan Zhang <zhangchunyan@iscas.ac.cn> says: The 1st patch is a cleanup; Patch 2/4 is an optimization that takes Palmer's suggestion; The last two patches add raid6test support and make the raid6 RVV code buildable on user space. * patches from https://lore.kernel.org/r/20250718072711.3865118-1-zhangchunyan@iscas.ac.cn: raid6: test: Add support for RISC-V raid6: riscv: Allow code to be compiled in userspace raid6: riscv: Prevent compiler with vector support to build already vectorized code raid6: riscv: replace one load with a move to speed up the caculation raid6: riscv: Clean up unused header file inclusion Link: https://lore.kernel.org/r/20250718072711.3865118-1-zhangchunyan@iscas.ac.cn Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Implement endianness swap macros for RISC-V. Use the rev8 instruction when Zbb is available. Otherwise, rely on the default mask-and-shift implementation. Acked-by: Palmer Dabbelt <palmer@rivosinc.com> Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Tested-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Ignacio Encinas <ignacio@iencinas.com> Link: https://lore.kernel.org/r/20250723-riscv-swab-v6-1-fc11e9a2efc9@iencinas.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Junhui Liu <junhui.liu@pigmoral.tech> says:
This patch series improves RISC-V kernel compatibility and robustness by
refining how the SATP mode is determined during early boot. Some RISC-V
implementations, such as the Anlogic DR1V90 FPSoC with a UX900 RISC-V
core designed by Nuclei, which I am currently attempting to run the
mainline kernel on [1], may hang when attempting to write an unsupported
SATP mode.
To address this, the patch determines the SATP mode limit by taking the
minimum of the value specified by the kernel command line option
("noXlvl", e.g., "no4lvl" or "no5lvl") and the "mmu-type" property in
the device tree for the first enabled CPU. If only one is specified,
that value will be used as the limit.
- If the resulting SATP mode limit is sv48 or higher, the kernel will
probe SATP modes from this limit down to lower modes until a supported
mode is found.
- If the limit is sv39, the kernel will directly use sv39 without
probing lower modes.
* patches from https://lore.kernel.org/r/20250722-satp-from-fdt-v1-0-5ba22218fa5f@pigmoral.tech:
riscv: mm: Use mmu-type from FDT to limit SATP mode
riscv: mm: Return intended SATP mode for noXlvl options
Link: https://lore.kernel.org/r/20250722-satp-from-fdt-v1-0-5ba22218fa5f@pigmoral.tech
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
The type of the value to write should be determined by the size of the destination, not by the value itself, which may be a constant. This aligns the behavior with x86_64, where __typeof__(*(__gu_ptr)) is used to infer the correct type. This fixes an issue in put_cmsg, which was only writing 4 out of 8 bytes to the cmsg_len field, causing the glibc tst-socket-timestamp test to fail. Fixes: ca1a66c ("riscv: uaccess: do not do misaligned accesses in get/put_user()") Cc: <stable@vger.kernel.org> Signed-off-by: Aurelien Jarno <aurelien@aurel32.net> Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Link: https://lore.kernel.org/r/20250724220853.1969954-1-aurelien@aurel32.net Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
AlexGhiti
force-pushed
the
alex-for-next_manual
branch
from
f5ccecb to
4d04628
Compare
REG_L is wrong, because thread_info.cpu is 32-bit, not xlen-bit wide. The struct currently has a hole after cpu, so little endian accesses seemed fine. Fixes: 503638e ("riscv: Stop emitting preventive sfence.vma for new vmalloc mappings") Cc: stable@vger.kernel.org Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Radim Krčmář <rkrcmar@ventanamicro.com> Link: https://lore.kernel.org/r/20250725165410.2896641-4-rkrcmar@ventanamicro.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
REG_L is wrong, because thread_info.cpu is 32-bit, not xlen-bit wide. The struct currently has a hole after cpu, so little endian accesses seemed fine. Fixes: be97d0d ("riscv: VMAP_STACK overflow detection thread-safe") Cc: stable@vger.kernel.org Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Radim Krčmář <rkrcmar@ventanamicro.com> Link: https://lore.kernel.org/r/20250725165410.2896641-5-rkrcmar@ventanamicro.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
The offsets of TASK_TI_CPU and TASK_TI_CPU_NUM are identical, and TASK_TI_CPU is a better name for thread_info.cpu. Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Radim Krčmář <rkrcmar@ventanamicro.com> Link: https://lore.kernel.org/r/20250725165410.2896641-6-rkrcmar@ventanamicro.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
On rv64, preempt_count and cpu were both 32-bit followed by 64-bit, so placing one in the hole saves 8 bytes in the struct. Tested-by: Alexandre Ghiti <alexghiti@rivosinc.com> Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Signed-off-by: Radim Krčmář <rkrcmar@ventanamicro.com> Link: https://lore.kernel.org/r/20250725165410.2896641-7-rkrcmar@ventanamicro.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Radim Krčmář <rkrcmar@ventanamicro.com> says: I noticed that we give thread_info.cpu two different names in the assembly, load it as a wider type, and store it suboptimally. * patches from https://lore.kernel.org/r/20250725165410.2896641-3-rkrcmar@ventanamicro.com: riscv: pack rv64 thread_info better riscv: use TASK_TI_CPU instead of TASK_TI_CPU_NUM riscv: use lw when reading int cpu in asm_per_cpu riscv: use lw when reading int cpu in new_vmalloc_check Link: https://lore.kernel.org/r/20250725165410.2896641-3-rkrcmar@ventanamicro.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
The xmipsexectl extension is described in the MIPS RV64 P8700/P8700-F Multiprocessing System Programmer’s Guide linked at [1]. Link: https://mips.com/wp-content/uploads/2025/06/P8700_Programmers_Reference_Manual_Rev1.84_5-31-2025.pdf Signed-off-by: Aleksa Paunovic <aleksa.paunovic@htecgroup.com> Acked-by: Conor Dooley <conor.dooley@microchip.com> Link: https://lore.kernel.org/r/20250724-p8700-pause-v5-1-a6cbbe1c3412@htecgroup.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Add support for MIPS vendor extensions. Add support for the xmipsexectl vendor extension. Signed-off-by: Aleksa Paunovic <aleksa.paunovic@htecgroup.com> Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Link: https://lore.kernel.org/r/20250724-p8700-pause-v5-2-a6cbbe1c3412@htecgroup.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Add xmipsexectl instruction opcodes. This includes the MIPS.PAUSE, MIPS.EHB, and MIPS.IHB instructions. Signed-off-by: Aleksa Paunovic <aleksa.paunovic@htecgroup.com> Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Link: https://lore.kernel.org/r/20250724-p8700-pause-v5-3-a6cbbe1c3412@htecgroup.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Add a new hwprobe key "RISCV_HWPROBE_KEY_VENDOR_EXT_MIPS_0" which allows userspace to probe for the new xmipsexectl vendor extension. Signed-off-by: Aleksa Paunovic <aleksa.paunovic@htecgroup.com> Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Link: https://lore.kernel.org/r/20250724-p8700-pause-v5-4-a6cbbe1c3412@htecgroup.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Document support for MIPS vendor extensions using the key "RISCV_HWPROBE_KEY_VENDOR_EXT_MIPS_0" and xmipsexectl vendor extension using the key "RISCV_HWPROBE_VENDOR_EXT_XMIPSEXECTL". Signed-off-by: Aleksa Paunovic <aleksa.paunovic@htecgroup.com> Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Link: https://lore.kernel.org/r/20250724-p8700-pause-v5-5-a6cbbe1c3412@htecgroup.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Use the hwprobe syscall to decide which PAUSE instruction to execute in userspace code. Signed-off-by: Aleksa Paunovic <aleksa.paunovic@htecgroup.com> Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Link: https://lore.kernel.org/r/20250724-p8700-pause-v5-6-a6cbbe1c3412@htecgroup.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Add ERRATA_MIPS and ERRATA_MIPS_P8700_PAUSE_OPCODE configs. Handle errata for the MIPS PAUSE instruction. Signed-off-by: Djordje Todorovic <djordje.todorovic@htecgroup.com> Signed-off-by: Aleksandar Rikalo <arikalo@gmail.com> Signed-off-by: Raj Vishwanathan4 <rvishwanathan@mips.com> Signed-off-by: Aleksa Paunovic <aleksa.paunovic@htecgroup.com> Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Link: https://lore.kernel.org/r/20250724-p8700-pause-v5-7-a6cbbe1c3412@htecgroup.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Aleksa Paunovic <aleksa.paunovic@htecgroup.com> says: This patch series adds support for the xmipsexectl vendor extension. A new hardware probe key has also been added to allow userspace to probe for MIPS vendor extensions. Additionally, since the standard Zihintpause PAUSE instruction encoding is not supported on some MIPS CPUs, an errata was implemented for replacing this instruction with the xmipsexectl MIPS.PAUSE alternative encoding. * patches from https://lore.kernel.org/r/20250724-p8700-pause-v5-0-a6cbbe1c3412@htecgroup.com: riscv: errata: Fix the PAUSE Opcode for MIPS P8700 riscv: Add tools support for xmipsexectl riscv: hwprobe: Document MIPS xmipsexectl vendor extension riscv: hwprobe: Add MIPS vendor extension probing riscv: Add xmipsexectl instructions riscv: Add xmipsexectl as a vendor extension dt-bindings: riscv: Add xmipsexectl ISA extension description Link: https://lore.kernel.org/r/20250724-p8700-pause-v5-0-a6cbbe1c3412@htecgroup.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
The BGRT table is used to display a vendor logo during the boot process. Add the code for parsing it. Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com> Reviewed-by: Sunil V L <sunilvl@ventanamicro.com> Link: https://lore.kernel.org/r/20250729131535.522205-2-heinrich.schuchardt@canonical.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
The BGRT table is used to display a vendor logo in the boot process. There is no good reason why RISC-V should not support it. Remove the architecture constraint. Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com> Reviewed-by: Sunil V L <sunilvl@ventanamicro.com> Link: https://lore.kernel.org/r/20250729131535.522205-3-heinrich.schuchardt@canonical.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Heinrich Schuchardt <heinrich.schuchardt@canonical.com> says: The BGRT table is used to display a vendor logo in the boot process. There is no good reason why RISC-V should not support it. Remove the architecture constraint. * patches from https://lore.kernel.org/r/20250729131535.522205-1-heinrich.schuchardt@canonical.com: ACPI: support BGRT table on RISC-V RISC-V: ACPI: enable parsing the BGRT table Link: https://lore.kernel.org/r/20250729131535.522205-1-heinrich.schuchardt@canonical.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
…abled As described in commit 1845d38 ("riscv: cacheinfo: Add back init_cache_level() function"), when CONFIG_SMP is undefined, the cache hierarchy detection needs to be performed through the init_cache_level(), whereas when CONFIG_SMP is defined, this detection is handled during the init_cpu_topology() process. Furthermore, while commit 66381d3 ("RISC-V: Select ACPI PPTT drivers") enables cache information retrieval through the ACPI PPTT table, the init_of_cache_level() called within init_cache_level() cannot support cache hierarchy detection through ACPI PPTT. Therefore, when CONFIG_SMP is undefined, we directly invoke the fetch_cache_info function to initialize the cache levels. Signed-off-by: Jessica Liu <liu.xuemei1@zte.com.cn> Link: https://lore.kernel.org/r/20250725104425229ArF_W-F7CmscLJKs0yoiQ@zte.com.cn Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
AlexGhiti
force-pushed
the
alex-for-next_manual
branch
from
4d04628 to
fd7f4ca
Compare
Compared with IO attributes, NC attributes can improve performance, specifically in these aspects: Relaxed Order, Gathering, Supports Read Speculation, Supports Unaligned Access. Signed-off-by: Yunhui Cui <cuiyunhui@bytedance.com> Signed-off-by: Qingfang Deng <qingfang.deng@siflower.com.cn> Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com> Link: https://lore.kernel.org/r/20250722091504.45974-2-cuiyunhui@bytedance.com Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
AlexGhiti
force-pushed
the
alex-for-next_manual
branch
from
fd7f4ca to
b5e60b2
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
17 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.