From be97d0db5f44c0674480cb79ac6f5b0529b84c76 Mon Sep 17 00:00:00 2001 From: Deepak Gupta Date: Wed, 27 Sep 2023 22:47:59 +0000 Subject: riscv: VMAP_STACK overflow detection thread-safe commit 31da94c25aea ("riscv: add VMAP_STACK overflow detection") added support for CONFIG_VMAP_STACK. If overflow is detected, CPU switches to `shadow_stack` temporarily before switching finally to per-cpu `overflow_stack`. If two CPUs/harts are racing and end up in over flowing kernel stack, one or both will end up corrupting each other state because `shadow_stack` is not per-cpu. This patch optimizes per-cpu overflow stack switch by directly picking per-cpu `overflow_stack` and gets rid of `shadow_stack`. Following are the changes in this patch - Defines an asm macro to obtain per-cpu symbols in destination register. - In entry.S, when overflow is detected, per-cpu overflow stack is located using per-cpu asm macro. Computing per-cpu symbol requires a temporary register. x31 is saved away into CSR_SCRATCH (CSR_SCRATCH is anyways zero since we're in kernel). Please see Links for additional relevant disccussion and alternative solution. Tested by `echo EXHAUST_STACK > /sys/kernel/debug/provoke-crash/DIRECT` Kernel crash log below Insufficient stack space to handle exception!/debug/provoke-crash/DIRECT Task stack: [0xff20000010a98000..0xff20000010a9c000] Overflow stack: [0xff600001f7d98370..0xff600001f7d99370] CPU: 1 PID: 205 Comm: bash Not tainted 6.1.0-rc2-00001-g328a1f96f7b9 #34 Hardware name: riscv-virtio,qemu (DT) epc : __memset+0x60/0xfc ra : recursive_loop+0x48/0xc6 [lkdtm] epc : ffffffff808de0e4 ra : ffffffff0163a752 sp : ff20000010a97e80 gp : ffffffff815c0330 tp : ff600000820ea280 t0 : ff20000010a97e88 t1 : 000000000000002e t2 : 3233206874706564 s0 : ff20000010a982b0 s1 : 0000000000000012 a0 : ff20000010a97e88 a1 : 0000000000000000 a2 : 0000000000000400 a3 : ff20000010a98288 a4 : 0000000000000000 a5 : 0000000000000000 a6 : fffffffffffe43f0 a7 : 00007fffffffffff s2 : ff20000010a97e88 s3 : ffffffff01644680 s4 : ff20000010a9be90 s5 : ff600000842ba6c0 s6 : 00aaaaaac29e42b0 s7 : 00fffffff0aa3684 s8 : 00aaaaaac2978040 s9 : 0000000000000065 s10: 00ffffff8a7cad10 s11: 00ffffff8a76a4e0 t3 : ffffffff815dbaf4 t4 : ffffffff815dbaf4 t5 : ffffffff815dbab8 t6 : ff20000010a9bb48 status: 0000000200000120 badaddr: ff20000010a97e88 cause: 000000000000000f Kernel panic - not syncing: Kernel stack overflow CPU: 1 PID: 205 Comm: bash Not tainted 6.1.0-rc2-00001-g328a1f96f7b9 #34 Hardware name: riscv-virtio,qemu (DT) Call Trace: [] dump_backtrace+0x30/0x38 [] show_stack+0x40/0x4c [] dump_stack_lvl+0x44/0x5c [] dump_stack+0x18/0x20 [] panic+0x126/0x2fe [] walk_stackframe+0x0/0xf0 [] recursive_loop+0x48/0xc6 [lkdtm] SMP: stopping secondary CPUs ---[ end Kernel panic - not syncing: Kernel stack overflow ]--- Cc: Guo Ren Cc: Jisheng Zhang Link: https://lore.kernel.org/linux-riscv/Y347B0x4VUNOd6V7@xhacker/T/#t Link: https://lore.kernel.org/lkml/20221124094845.1907443-1-debug@rivosinc.com/ Signed-off-by: Deepak Gupta Co-developed-by: Sami Tolvanen Signed-off-by: Sami Tolvanen Acked-by: Guo Ren Tested-by: Nathan Chancellor Link: https://lore.kernel.org/r/20230927224757.1154247-9-samitolvanen@google.com Signed-off-by: Palmer Dabbelt --- arch/riscv/include/asm/asm-prototypes.h | 1 - arch/riscv/include/asm/asm.h | 22 +++++++++++ arch/riscv/include/asm/thread_info.h | 3 -- arch/riscv/kernel/asm-offsets.c | 1 + arch/riscv/kernel/entry.S | 70 +++++---------------------------- arch/riscv/kernel/traps.c | 36 +---------------- 6 files changed, 34 insertions(+), 99 deletions(-) (limited to 'arch') diff --git a/arch/riscv/include/asm/asm-prototypes.h b/arch/riscv/include/asm/asm-prototypes.h index 61ba8ed43d8f..36b955c762ba 100644 --- a/arch/riscv/include/asm/asm-prototypes.h +++ b/arch/riscv/include/asm/asm-prototypes.h @@ -25,7 +25,6 @@ DECLARE_DO_ERROR_INFO(do_trap_ecall_s); DECLARE_DO_ERROR_INFO(do_trap_ecall_m); DECLARE_DO_ERROR_INFO(do_trap_break); -asmlinkage unsigned long get_overflow_stack(void); asmlinkage void handle_bad_stack(struct pt_regs *regs); asmlinkage void do_page_fault(struct pt_regs *regs); asmlinkage void do_irq(struct pt_regs *regs); diff --git a/arch/riscv/include/asm/asm.h b/arch/riscv/include/asm/asm.h index 114bbadaef41..bfb4c26f113c 100644 --- a/arch/riscv/include/asm/asm.h +++ b/arch/riscv/include/asm/asm.h @@ -82,6 +82,28 @@ .endr .endm +#ifdef CONFIG_SMP +#ifdef CONFIG_32BIT +#define PER_CPU_OFFSET_SHIFT 2 +#else +#define PER_CPU_OFFSET_SHIFT 3 +#endif + +.macro asm_per_cpu dst sym tmp + REG_L \tmp, TASK_TI_CPU_NUM(tp) + slli \tmp, \tmp, PER_CPU_OFFSET_SHIFT + la \dst, __per_cpu_offset + add \dst, \dst, \tmp + REG_L \tmp, 0(\dst) + la \dst, \sym + add \dst, \dst, \tmp +.endm +#else /* CONFIG_SMP */ +.macro asm_per_cpu dst sym tmp + la \dst, \sym +.endm +#endif /* CONFIG_SMP */ + /* save all GPs except x1 ~ x5 */ .macro save_from_x6_to_x31 REG_S x6, PT_T1(sp) diff --git a/arch/riscv/include/asm/thread_info.h b/arch/riscv/include/asm/thread_info.h index 1833beb00489..d18ce0113ca1 100644 --- a/arch/riscv/include/asm/thread_info.h +++ b/arch/riscv/include/asm/thread_info.h @@ -34,9 +34,6 @@ #ifndef __ASSEMBLY__ -extern long shadow_stack[SHADOW_OVERFLOW_STACK_SIZE / sizeof(long)]; -extern unsigned long spin_shadow_stack; - #include #include diff --git a/arch/riscv/kernel/asm-offsets.c b/arch/riscv/kernel/asm-offsets.c index d6a75aac1d27..9f535d5de33f 100644 --- a/arch/riscv/kernel/asm-offsets.c +++ b/arch/riscv/kernel/asm-offsets.c @@ -39,6 +39,7 @@ void asm_offsets(void) OFFSET(TASK_TI_KERNEL_SP, task_struct, thread_info.kernel_sp); OFFSET(TASK_TI_USER_SP, task_struct, thread_info.user_sp); + OFFSET(TASK_TI_CPU_NUM, task_struct, thread_info.cpu); OFFSET(TASK_THREAD_F0, task_struct, thread.fstate.f[0]); OFFSET(TASK_THREAD_F1, task_struct, thread.fstate.f[1]); OFFSET(TASK_THREAD_F2, task_struct, thread.fstate.f[2]); diff --git a/arch/riscv/kernel/entry.S b/arch/riscv/kernel/entry.S index 143a2bb3e697..3d11aa3af105 100644 --- a/arch/riscv/kernel/entry.S +++ b/arch/riscv/kernel/entry.S @@ -10,9 +10,11 @@ #include #include #include +#include #include #include #include +#include SYM_CODE_START(handle_exception) /* @@ -170,67 +172,15 @@ SYM_CODE_END(ret_from_exception) #ifdef CONFIG_VMAP_STACK SYM_CODE_START_LOCAL(handle_kernel_stack_overflow) - /* - * Takes the psuedo-spinlock for the shadow stack, in case multiple - * harts are concurrently overflowing their kernel stacks. We could - * store any value here, but since we're overflowing the kernel stack - * already we only have SP to use as a scratch register. So we just - * swap in the address of the spinlock, as that's definately non-zero. - * - * Pairs with a store_release in handle_bad_stack(). - */ -1: la sp, spin_shadow_stack - REG_AMOSWAP_AQ sp, sp, (sp) - bnez sp, 1b - - la sp, shadow_stack - addi sp, sp, SHADOW_OVERFLOW_STACK_SIZE - - //save caller register to shadow stack - addi sp, sp, -(PT_SIZE_ON_STACK) - REG_S x1, PT_RA(sp) - REG_S x5, PT_T0(sp) - REG_S x6, PT_T1(sp) - REG_S x7, PT_T2(sp) - REG_S x10, PT_A0(sp) - REG_S x11, PT_A1(sp) - REG_S x12, PT_A2(sp) - REG_S x13, PT_A3(sp) - REG_S x14, PT_A4(sp) - REG_S x15, PT_A5(sp) - REG_S x16, PT_A6(sp) - REG_S x17, PT_A7(sp) - REG_S x28, PT_T3(sp) - REG_S x29, PT_T4(sp) - REG_S x30, PT_T5(sp) - REG_S x31, PT_T6(sp) - - la ra, restore_caller_reg - tail get_overflow_stack - -restore_caller_reg: - //save per-cpu overflow stack - REG_S a0, -8(sp) - //restore caller register from shadow_stack - REG_L x1, PT_RA(sp) - REG_L x5, PT_T0(sp) - REG_L x6, PT_T1(sp) - REG_L x7, PT_T2(sp) - REG_L x10, PT_A0(sp) - REG_L x11, PT_A1(sp) - REG_L x12, PT_A2(sp) - REG_L x13, PT_A3(sp) - REG_L x14, PT_A4(sp) - REG_L x15, PT_A5(sp) - REG_L x16, PT_A6(sp) - REG_L x17, PT_A7(sp) - REG_L x28, PT_T3(sp) - REG_L x29, PT_T4(sp) - REG_L x30, PT_T5(sp) - REG_L x31, PT_T6(sp) + /* we reach here from kernel context, sscratch must be 0 */ + csrrw x31, CSR_SCRATCH, x31 + asm_per_cpu sp, overflow_stack, x31 + li x31, OVERFLOW_STACK_SIZE + add sp, sp, x31 + /* zero out x31 again and restore x31 */ + xor x31, x31, x31 + csrrw x31, CSR_SCRATCH, x31 - //load per-cpu overflow stack - REG_L sp, -8(sp) addi sp, sp, -(PT_SIZE_ON_STACK) //save context to overflow stack diff --git a/arch/riscv/kernel/traps.c b/arch/riscv/kernel/traps.c index 19807c4d3805..0063a195deca 100644 --- a/arch/riscv/kernel/traps.c +++ b/arch/riscv/kernel/traps.c @@ -402,48 +402,14 @@ int is_valid_bugaddr(unsigned long pc) #endif /* CONFIG_GENERIC_BUG */ #ifdef CONFIG_VMAP_STACK -/* - * Extra stack space that allows us to provide panic messages when the kernel - * has overflowed its stack. - */ -static DEFINE_PER_CPU(unsigned long [OVERFLOW_STACK_SIZE/sizeof(long)], +DEFINE_PER_CPU(unsigned long [OVERFLOW_STACK_SIZE/sizeof(long)], overflow_stack)__aligned(16); -/* - * A temporary stack for use by handle_kernel_stack_overflow. This is used so - * we can call into C code to get the per-hart overflow stack. Usage of this - * stack must be protected by spin_shadow_stack. - */ -long shadow_stack[SHADOW_OVERFLOW_STACK_SIZE/sizeof(long)] __aligned(16); - -/* - * A pseudo spinlock to protect the shadow stack from being used by multiple - * harts concurrently. This isn't a real spinlock because the lock side must - * be taken without a valid stack and only a single register, it's only taken - * while in the process of panicing anyway so the performance and error - * checking a proper spinlock gives us doesn't matter. - */ -unsigned long spin_shadow_stack; - -asmlinkage unsigned long get_overflow_stack(void) -{ - return (unsigned long)this_cpu_ptr(overflow_stack) + - OVERFLOW_STACK_SIZE; -} asmlinkage void handle_bad_stack(struct pt_regs *regs) { unsigned long tsk_stk = (unsigned long)current->stack; unsigned long ovf_stk = (unsigned long)this_cpu_ptr(overflow_stack); - /* - * We're done with the shadow stack by this point, as we're on the - * overflow stack. Tell any other concurrent overflowing harts that - * they can proceed with panicing by releasing the pseudo-spinlock. - * - * This pairs with an amoswap.aq in handle_kernel_stack_overflow. - */ - smp_store_release(&spin_shadow_stack, 0); - console_verbose(); pr_emerg("Insufficient stack space to handle exception!\n"); -- cgit v1.2.3-58-ga151 From 82982fdd5133fa7e0b2dfaf746d18d6f29922b82 Mon Sep 17 00:00:00 2001 From: Sami Tolvanen Date: Wed, 27 Sep 2023 22:48:00 +0000 Subject: riscv: Deduplicate IRQ stack switching With CONFIG_IRQ_STACKS, we switch to a separate per-CPU IRQ stack before calling handle_riscv_irq or __do_softirq. We currently have duplicate inline assembly snippets for stack switching in both code paths. Now that we can access per-CPU variables in assembly, implement call_on_irq_stack in assembly, and use that instead of redundant inline assembly. Signed-off-by: Sami Tolvanen Tested-by: Nathan Chancellor Reviewed-by: Guo Ren Link: https://lore.kernel.org/r/20230927224757.1154247-10-samitolvanen@google.com Signed-off-by: Palmer Dabbelt --- arch/riscv/include/asm/asm.h | 5 +++++ arch/riscv/include/asm/irq_stack.h | 3 +++ arch/riscv/kernel/asm-offsets.c | 5 +++++ arch/riscv/kernel/entry.S | 30 ++++++++++++++++++++++++++++++ arch/riscv/kernel/irq.c | 35 ++++++++--------------------------- arch/riscv/kernel/traps.c | 32 ++++---------------------------- 6 files changed, 55 insertions(+), 55 deletions(-) (limited to 'arch') diff --git a/arch/riscv/include/asm/asm.h b/arch/riscv/include/asm/asm.h index bfb4c26f113c..8e446be2d57c 100644 --- a/arch/riscv/include/asm/asm.h +++ b/arch/riscv/include/asm/asm.h @@ -104,6 +104,11 @@ .endm #endif /* CONFIG_SMP */ +.macro load_per_cpu dst ptr tmp + asm_per_cpu \dst \ptr \tmp + REG_L \dst, 0(\dst) +.endm + /* save all GPs except x1 ~ x5 */ .macro save_from_x6_to_x31 REG_S x6, PT_T1(sp) diff --git a/arch/riscv/include/asm/irq_stack.h b/arch/riscv/include/asm/irq_stack.h index e4042d297580..6441ded3b0cf 100644 --- a/arch/riscv/include/asm/irq_stack.h +++ b/arch/riscv/include/asm/irq_stack.h @@ -12,6 +12,9 @@ DECLARE_PER_CPU(ulong *, irq_stack_ptr); +asmlinkage void call_on_irq_stack(struct pt_regs *regs, + void (*func)(struct pt_regs *)); + #ifdef CONFIG_VMAP_STACK /* * To ensure that VMAP'd stack overflow detection works correctly, all VMAP'd diff --git a/arch/riscv/kernel/asm-offsets.c b/arch/riscv/kernel/asm-offsets.c index 9f535d5de33f..0af8860f9d68 100644 --- a/arch/riscv/kernel/asm-offsets.c +++ b/arch/riscv/kernel/asm-offsets.c @@ -14,6 +14,7 @@ #include #include #include +#include #include void asm_offsets(void); @@ -480,4 +481,8 @@ void asm_offsets(void) OFFSET(KERNEL_MAP_VIRT_ADDR, kernel_mapping, virt_addr); OFFSET(SBI_HART_BOOT_TASK_PTR_OFFSET, sbi_hart_boot_data, task_ptr); OFFSET(SBI_HART_BOOT_STACK_PTR_OFFSET, sbi_hart_boot_data, stack_ptr); + + DEFINE(STACKFRAME_SIZE_ON_STACK, ALIGN(sizeof(struct stackframe), STACK_ALIGN)); + OFFSET(STACKFRAME_FP, stackframe, fp); + OFFSET(STACKFRAME_RA, stackframe, ra); } diff --git a/arch/riscv/kernel/entry.S b/arch/riscv/kernel/entry.S index 3d11aa3af105..a306562636e4 100644 --- a/arch/riscv/kernel/entry.S +++ b/arch/riscv/kernel/entry.S @@ -218,6 +218,36 @@ SYM_CODE_START(ret_from_fork) tail syscall_exit_to_user_mode SYM_CODE_END(ret_from_fork) +#ifdef CONFIG_IRQ_STACKS +/* + * void call_on_irq_stack(struct pt_regs *regs, + * void (*func)(struct pt_regs *)); + * + * Calls func(regs) using the per-CPU IRQ stack. + */ +SYM_FUNC_START(call_on_irq_stack) + /* Create a frame record to save ra and s0 (fp) */ + addi sp, sp, -STACKFRAME_SIZE_ON_STACK + REG_S ra, STACKFRAME_RA(sp) + REG_S s0, STACKFRAME_FP(sp) + addi s0, sp, STACKFRAME_SIZE_ON_STACK + + /* Switch to the per-CPU IRQ stack and call the handler */ + load_per_cpu t0, irq_stack_ptr, t1 + li t1, IRQ_STACK_SIZE + add sp, t0, t1 + jalr a1 + + /* Switch back to the thread stack and restore ra and s0 */ + addi sp, s0, -STACKFRAME_SIZE_ON_STACK + REG_L ra, STACKFRAME_RA(sp) + REG_L s0, STACKFRAME_FP(sp) + addi sp, sp, STACKFRAME_SIZE_ON_STACK + + ret +SYM_FUNC_END(call_on_irq_stack) +#endif /* CONFIG_IRQ_STACKS */ + /* * Integer register context switch * The callee-saved registers must be saved and restored. diff --git a/arch/riscv/kernel/irq.c b/arch/riscv/kernel/irq.c index a8efa053c4a5..95dafdcbd135 100644 --- a/arch/riscv/kernel/irq.c +++ b/arch/riscv/kernel/irq.c @@ -61,35 +61,16 @@ static void init_irq_stacks(void) #endif /* CONFIG_VMAP_STACK */ #ifdef CONFIG_HAVE_SOFTIRQ_ON_OWN_STACK +static void ___do_softirq(struct pt_regs *regs) +{ + __do_softirq(); +} + void do_softirq_own_stack(void) { -#ifdef CONFIG_IRQ_STACKS - if (on_thread_stack()) { - ulong *sp = per_cpu(irq_stack_ptr, smp_processor_id()) - + IRQ_STACK_SIZE/sizeof(ulong); - __asm__ __volatile( - "addi sp, sp, -"RISCV_SZPTR "\n" - REG_S" ra, (sp) \n" - "addi sp, sp, -"RISCV_SZPTR "\n" - REG_S" s0, (sp) \n" - "addi s0, sp, 2*"RISCV_SZPTR "\n" - "move sp, %[sp] \n" - "call __do_softirq \n" - "addi sp, s0, -2*"RISCV_SZPTR"\n" - REG_L" s0, (sp) \n" - "addi sp, sp, "RISCV_SZPTR "\n" - REG_L" ra, (sp) \n" - "addi sp, sp, "RISCV_SZPTR "\n" - : - : [sp] "r" (sp) - : "a0", "a1", "a2", "a3", "a4", "a5", "a6", "a7", - "t0", "t1", "t2", "t3", "t4", "t5", "t6", -#ifndef CONFIG_FRAME_POINTER - "s0", -#endif - "memory"); - } else -#endif + if (on_thread_stack()) + call_on_irq_stack(NULL, ___do_softirq); + else __do_softirq(); } #endif /* CONFIG_HAVE_SOFTIRQ_ON_OWN_STACK */ diff --git a/arch/riscv/kernel/traps.c b/arch/riscv/kernel/traps.c index 0063a195deca..cda6dcdb0376 100644 --- a/arch/riscv/kernel/traps.c +++ b/arch/riscv/kernel/traps.c @@ -352,34 +352,10 @@ static void noinstr handle_riscv_irq(struct pt_regs *regs) asmlinkage void noinstr do_irq(struct pt_regs *regs) { irqentry_state_t state = irqentry_enter(regs); -#ifdef CONFIG_IRQ_STACKS - if (on_thread_stack()) { - ulong *sp = per_cpu(irq_stack_ptr, smp_processor_id()) - + IRQ_STACK_SIZE/sizeof(ulong); - __asm__ __volatile( - "addi sp, sp, -"RISCV_SZPTR "\n" - REG_S" ra, (sp) \n" - "addi sp, sp, -"RISCV_SZPTR "\n" - REG_S" s0, (sp) \n" - "addi s0, sp, 2*"RISCV_SZPTR "\n" - "move sp, %[sp] \n" - "move a0, %[regs] \n" - "call handle_riscv_irq \n" - "addi sp, s0, -2*"RISCV_SZPTR"\n" - REG_L" s0, (sp) \n" - "addi sp, sp, "RISCV_SZPTR "\n" - REG_L" ra, (sp) \n" - "addi sp, sp, "RISCV_SZPTR "\n" - : - : [sp] "r" (sp), [regs] "r" (regs) - : "a0", "a1", "a2", "a3", "a4", "a5", "a6", "a7", - "t0", "t1", "t2", "t3", "t4", "t5", "t6", -#ifndef CONFIG_FRAME_POINTER - "s0", -#endif - "memory"); - } else -#endif + + if (IS_ENABLED(CONFIG_IRQ_STACKS) && on_thread_stack()) + call_on_irq_stack(regs, handle_riscv_irq); + else handle_riscv_irq(regs); irqentry_exit(regs, state); -- cgit v1.2.3-58-ga151 From e609b4f4252a2ad2454736078693571b9fbff019 Mon Sep 17 00:00:00 2001 From: Sami Tolvanen Date: Wed, 27 Sep 2023 22:48:01 +0000 Subject: riscv: Move global pointer loading to a macro In Clang 17, -fsanitize=shadow-call-stack uses the newly declared platform register gp for storing shadow call stack pointers. As this is obviously incompatible with gp relaxation, in preparation for CONFIG_SHADOW_CALL_STACK support, move global pointer loading to a single macro, which we can cleanly disable when SCS is used instead. Link: https://reviews.llvm.org/rGaa1d2693c256 Link: https://github.com/riscv-non-isa/riscv-elf-psabi-doc/commit/a484e843e6eeb51f0cb7b8819e50da6d2444d769 Signed-off-by: Sami Tolvanen Tested-by: Nathan Chancellor Link: https://lore.kernel.org/r/20230927224757.1154247-11-samitolvanen@google.com Signed-off-by: Palmer Dabbelt --- arch/riscv/include/asm/asm.h | 8 ++++++++ arch/riscv/kernel/entry.S | 6 ++---- arch/riscv/kernel/head.S | 15 +++------------ arch/riscv/kernel/suspend_entry.S | 5 +---- 4 files changed, 14 insertions(+), 20 deletions(-) (limited to 'arch') diff --git a/arch/riscv/include/asm/asm.h b/arch/riscv/include/asm/asm.h index 8e446be2d57c..f34dd1a526a1 100644 --- a/arch/riscv/include/asm/asm.h +++ b/arch/riscv/include/asm/asm.h @@ -109,6 +109,14 @@ REG_L \dst, 0(\dst) .endm +/* load __global_pointer to gp */ +.macro load_global_pointer +.option push +.option norelax + la gp, __global_pointer$ +.option pop +.endm + /* save all GPs except x1 ~ x5 */ .macro save_from_x6_to_x31 REG_S x6, PT_T1(sp) diff --git a/arch/riscv/kernel/entry.S b/arch/riscv/kernel/entry.S index a306562636e4..6215dcf2e83b 100644 --- a/arch/riscv/kernel/entry.S +++ b/arch/riscv/kernel/entry.S @@ -75,10 +75,8 @@ _save_context: csrw CSR_SCRATCH, x0 /* Load the global pointer */ -.option push -.option norelax - la gp, __global_pointer$ -.option pop + load_global_pointer + move a0, sp /* pt_regs */ la ra, ret_from_exception diff --git a/arch/riscv/kernel/head.S b/arch/riscv/kernel/head.S index 3710ea5d160f..a0484d95d8fb 100644 --- a/arch/riscv/kernel/head.S +++ b/arch/riscv/kernel/head.S @@ -110,10 +110,7 @@ relocate_enable_mmu: csrw CSR_TVEC, a0 /* Reload the global pointer */ -.option push -.option norelax - la gp, __global_pointer$ -.option pop + load_global_pointer /* * Switch to kernel page tables. A full fence is necessary in order to @@ -134,10 +131,7 @@ secondary_start_sbi: csrw CSR_IP, zero /* Load the global pointer */ - .option push - .option norelax - la gp, __global_pointer$ - .option pop + load_global_pointer /* * Disable FPU & VECTOR to detect illegal usage of @@ -228,10 +222,7 @@ pmp_done: #endif /* CONFIG_RISCV_M_MODE */ /* Load the global pointer */ -.option push -.option norelax - la gp, __global_pointer$ -.option pop + load_global_pointer /* * Disable FPU & VECTOR to detect illegal usage of diff --git a/arch/riscv/kernel/suspend_entry.S b/arch/riscv/kernel/suspend_entry.S index f7960c7c5f9e..d5cf8b575777 100644 --- a/arch/riscv/kernel/suspend_entry.S +++ b/arch/riscv/kernel/suspend_entry.S @@ -61,10 +61,7 @@ END(__cpu_suspend_enter) SYM_TYPED_FUNC_START(__cpu_resume_enter) /* Load the global pointer */ - .option push - .option norelax - la gp, __global_pointer$ - .option pop + load_global_pointer #ifdef CONFIG_MMU /* Save A0 and A1 */ -- cgit v1.2.3-58-ga151 From d1584d791a297aa8ed93503382a682a6ecfc4218 Mon Sep 17 00:00:00 2001 From: Sami Tolvanen Date: Wed, 27 Sep 2023 22:48:02 +0000 Subject: riscv: Implement Shadow Call Stack Implement CONFIG_SHADOW_CALL_STACK for RISC-V. When enabled, the compiler injects instructions to all non-leaf C functions to store the return address to the shadow stack and unconditionally load it again before returning, which makes it harder to corrupt the return address through a stack overflow, for example. The active shadow call stack pointer is stored in the gp register, which makes SCS incompatible with gp relaxation. Use --no-relax-gp to ensure gp relaxation is disabled and disable global pointer loading. Add SCS pointers to struct thread_info, implement SCS initialization, and task switching Signed-off-by: Sami Tolvanen Tested-by: Nathan Chancellor Link: https://lore.kernel.org/r/20230927224757.1154247-12-samitolvanen@google.com Signed-off-by: Palmer Dabbelt --- arch/riscv/Kconfig | 6 +++++ arch/riscv/Makefile | 4 +++ arch/riscv/include/asm/asm.h | 6 +++++ arch/riscv/include/asm/scs.h | 47 ++++++++++++++++++++++++++++++++++++ arch/riscv/include/asm/thread_info.h | 13 ++++++++++ arch/riscv/kernel/asm-offsets.c | 3 +++ arch/riscv/kernel/entry.S | 11 +++++++++ arch/riscv/kernel/head.S | 4 +++ arch/riscv/kernel/vdso/Makefile | 2 +- arch/riscv/purgatory/Makefile | 4 +++ 10 files changed, 99 insertions(+), 1 deletion(-) create mode 100644 arch/riscv/include/asm/scs.h (limited to 'arch') diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig index d607ab0f7c6d..2f14ccb5fb01 100644 --- a/arch/riscv/Kconfig +++ b/arch/riscv/Kconfig @@ -48,6 +48,7 @@ config RISCV select ARCH_SUPPORTS_HUGETLBFS if MMU select ARCH_SUPPORTS_PAGE_TABLE_CHECK if MMU select ARCH_SUPPORTS_PER_VMA_LOCK if MMU + select ARCH_SUPPORTS_SHADOW_CALL_STACK if HAVE_SHADOW_CALL_STACK select ARCH_USE_MEMTEST select ARCH_USE_QUEUED_RWLOCKS select ARCH_USES_CFI_TRAPS if CFI_CLANG @@ -174,6 +175,11 @@ config GCC_SUPPORTS_DYNAMIC_FTRACE def_bool CC_IS_GCC depends on $(cc-option,-fpatchable-function-entry=8) +config HAVE_SHADOW_CALL_STACK + def_bool $(cc-option,-fsanitize=shadow-call-stack) + # https://github.com/riscv-non-isa/riscv-elf-psabi-doc/commit/a484e843e6eeb51f0cb7b8819e50da6d2444d769 + depends on $(ld-option,--no-relax-gp) + config ARCH_MMAP_RND_BITS_MIN default 18 if 64BIT default 8 diff --git a/arch/riscv/Makefile b/arch/riscv/Makefile index 1329e060c548..304b94446507 100644 --- a/arch/riscv/Makefile +++ b/arch/riscv/Makefile @@ -55,6 +55,10 @@ endif endif endif +ifeq ($(CONFIG_SHADOW_CALL_STACK),y) + KBUILD_LDFLAGS += --no-relax-gp +endif + # ISA string setting riscv-march-$(CONFIG_ARCH_RV32I) := rv32ima riscv-march-$(CONFIG_ARCH_RV64I) := rv64ima diff --git a/arch/riscv/include/asm/asm.h b/arch/riscv/include/asm/asm.h index f34dd1a526a1..b0487b39e674 100644 --- a/arch/riscv/include/asm/asm.h +++ b/arch/riscv/include/asm/asm.h @@ -109,6 +109,11 @@ REG_L \dst, 0(\dst) .endm +#ifdef CONFIG_SHADOW_CALL_STACK +/* gp is used as the shadow call stack pointer instead */ +.macro load_global_pointer +.endm +#else /* load __global_pointer to gp */ .macro load_global_pointer .option push @@ -116,6 +121,7 @@ la gp, __global_pointer$ .option pop .endm +#endif /* CONFIG_SHADOW_CALL_STACK */ /* save all GPs except x1 ~ x5 */ .macro save_from_x6_to_x31 diff --git a/arch/riscv/include/asm/scs.h b/arch/riscv/include/asm/scs.h new file mode 100644 index 000000000000..94726ea773e3 --- /dev/null +++ b/arch/riscv/include/asm/scs.h @@ -0,0 +1,47 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _ASM_SCS_H +#define _ASM_SCS_H + +#ifdef __ASSEMBLY__ +#include + +#ifdef CONFIG_SHADOW_CALL_STACK + +/* Load init_shadow_call_stack to gp. */ +.macro scs_load_init_stack + la gp, init_shadow_call_stack + XIP_FIXUP_OFFSET gp +.endm + +/* Load task_scs_sp(current) to gp. */ +.macro scs_load_current + REG_L gp, TASK_TI_SCS_SP(tp) +.endm + +/* Load task_scs_sp(current) to gp, but only if tp has changed. */ +.macro scs_load_current_if_task_changed prev + beq \prev, tp, _skip_scs + scs_load_current +_skip_scs: +.endm + +/* Save gp to task_scs_sp(current). */ +.macro scs_save_current + REG_S gp, TASK_TI_SCS_SP(tp) +.endm + +#else /* CONFIG_SHADOW_CALL_STACK */ + +.macro scs_load_init_stack +.endm +.macro scs_load_current +.endm +.macro scs_load_current_if_task_changed prev +.endm +.macro scs_save_current +.endm + +#endif /* CONFIG_SHADOW_CALL_STACK */ +#endif /* __ASSEMBLY__ */ + +#endif /* _ASM_SCS_H */ diff --git a/arch/riscv/include/asm/thread_info.h b/arch/riscv/include/asm/thread_info.h index d18ce0113ca1..574779900bfb 100644 --- a/arch/riscv/include/asm/thread_info.h +++ b/arch/riscv/include/asm/thread_info.h @@ -57,8 +57,20 @@ struct thread_info { long user_sp; /* User stack pointer */ int cpu; unsigned long syscall_work; /* SYSCALL_WORK_ flags */ +#ifdef CONFIG_SHADOW_CALL_STACK + void *scs_base; + void *scs_sp; +#endif }; +#ifdef CONFIG_SHADOW_CALL_STACK +#define INIT_SCS \ + .scs_base = init_shadow_call_stack, \ + .scs_sp = init_shadow_call_stack, +#else +#define INIT_SCS +#endif + /* * macros/functions for gaining access to the thread information structure * @@ -68,6 +80,7 @@ struct thread_info { { \ .flags = 0, \ .preempt_count = INIT_PREEMPT_COUNT, \ + INIT_SCS \ } void arch_release_task_struct(struct task_struct *tsk); diff --git a/arch/riscv/kernel/asm-offsets.c b/arch/riscv/kernel/asm-offsets.c index 0af8860f9d68..a03129f40c46 100644 --- a/arch/riscv/kernel/asm-offsets.c +++ b/arch/riscv/kernel/asm-offsets.c @@ -39,6 +39,9 @@ void asm_offsets(void) OFFSET(TASK_TI_PREEMPT_COUNT, task_struct, thread_info.preempt_count); OFFSET(TASK_TI_KERNEL_SP, task_struct, thread_info.kernel_sp); OFFSET(TASK_TI_USER_SP, task_struct, thread_info.user_sp); +#ifdef CONFIG_SHADOW_CALL_STACK + OFFSET(TASK_TI_SCS_SP, task_struct, thread_info.scs_sp); +#endif OFFSET(TASK_TI_CPU_NUM, task_struct, thread_info.cpu); OFFSET(TASK_THREAD_F0, task_struct, thread.fstate.f[0]); diff --git a/arch/riscv/kernel/entry.S b/arch/riscv/kernel/entry.S index 6215dcf2e83b..52793193a763 100644 --- a/arch/riscv/kernel/entry.S +++ b/arch/riscv/kernel/entry.S @@ -9,6 +9,7 @@ #include #include +#include #include #include #include @@ -77,6 +78,9 @@ _save_context: /* Load the global pointer */ load_global_pointer + /* Load the kernel shadow call stack pointer if coming from userspace */ + scs_load_current_if_task_changed s5 + move a0, sp /* pt_regs */ la ra, ret_from_exception @@ -123,6 +127,9 @@ SYM_CODE_START_NOALIGN(ret_from_exception) addi s0, sp, PT_SIZE_ON_STACK REG_S s0, TASK_TI_KERNEL_SP(tp) + /* Save the kernel shadow call stack pointer */ + scs_save_current + /* * Save TP into the scratch register , so we can find the kernel data * structures again. @@ -275,6 +282,8 @@ SYM_FUNC_START(__switch_to) REG_S s9, TASK_THREAD_S9_RA(a3) REG_S s10, TASK_THREAD_S10_RA(a3) REG_S s11, TASK_THREAD_S11_RA(a3) + /* Save the kernel shadow call stack pointer */ + scs_save_current /* Restore context from next->thread */ REG_L ra, TASK_THREAD_RA_RA(a4) REG_L sp, TASK_THREAD_SP_RA(a4) @@ -292,6 +301,8 @@ SYM_FUNC_START(__switch_to) REG_L s11, TASK_THREAD_S11_RA(a4) /* The offset of thread_info in task_struct is zero. */ move tp, a1 + /* Switch to the next shadow call stack */ + scs_load_current ret SYM_FUNC_END(__switch_to) diff --git a/arch/riscv/kernel/head.S b/arch/riscv/kernel/head.S index a0484d95d8fb..18f97ec0f7ed 100644 --- a/arch/riscv/kernel/head.S +++ b/arch/riscv/kernel/head.S @@ -14,6 +14,7 @@ #include #include #include +#include #include #include "efi-header.S" @@ -153,6 +154,7 @@ secondary_start_sbi: XIP_FIXUP_OFFSET a3 add a3, a3, a1 REG_L sp, (a3) + scs_load_current .Lsecondary_start_common: @@ -289,6 +291,7 @@ clear_bss_done: la sp, init_thread_union + THREAD_SIZE XIP_FIXUP_OFFSET sp addi sp, sp, -PT_SIZE_ON_STACK + scs_load_init_stack #ifdef CONFIG_BUILTIN_DTB la a0, __dtb_start XIP_FIXUP_OFFSET a0 @@ -307,6 +310,7 @@ clear_bss_done: la tp, init_task la sp, init_thread_union + THREAD_SIZE addi sp, sp, -PT_SIZE_ON_STACK + scs_load_current #ifdef CONFIG_KASAN call kasan_early_init diff --git a/arch/riscv/kernel/vdso/Makefile b/arch/riscv/kernel/vdso/Makefile index 6b1dba11bf6d..48c362c0cb3d 100644 --- a/arch/riscv/kernel/vdso/Makefile +++ b/arch/riscv/kernel/vdso/Makefile @@ -36,7 +36,7 @@ CPPFLAGS_vdso.lds += -DHAS_VGETTIMEOFDAY endif # Disable -pg to prevent insert call site -CFLAGS_REMOVE_vgettimeofday.o = $(CC_FLAGS_FTRACE) +CFLAGS_REMOVE_vgettimeofday.o = $(CC_FLAGS_FTRACE) $(CC_FLAGS_SCS) # Disable profiling and instrumentation for VDSO code GCOV_PROFILE := n diff --git a/arch/riscv/purgatory/Makefile b/arch/riscv/purgatory/Makefile index 9e6476719abb..6a3c16bd5ca3 100644 --- a/arch/riscv/purgatory/Makefile +++ b/arch/riscv/purgatory/Makefile @@ -81,6 +81,10 @@ ifdef CONFIG_CFI_CLANG PURGATORY_CFLAGS_REMOVE += $(CC_FLAGS_CFI) endif +ifdef CONFIG_SHADOW_CALL_STACK +PURGATORY_CFLAGS_REMOVE += $(CC_FLAGS_SCS) +endif + CFLAGS_REMOVE_purgatory.o += $(PURGATORY_CFLAGS_REMOVE) CFLAGS_purgatory.o += $(PURGATORY_CFLAGS) -- cgit v1.2.3-58-ga151 From c40fef858d002fb027033c572ac8bdf8756a2c6b Mon Sep 17 00:00:00 2001 From: Sami Tolvanen Date: Wed, 27 Sep 2023 22:48:03 +0000 Subject: riscv: Use separate IRQ shadow call stacks When both CONFIG_IRQ_STACKS and SCS are enabled, also use a separate per-CPU shadow call stack. Signed-off-by: Sami Tolvanen Tested-by: Nathan Chancellor Link: https://lore.kernel.org/r/20230927224757.1154247-13-samitolvanen@google.com Signed-off-by: Palmer Dabbelt --- arch/riscv/include/asm/scs.h | 7 +++++++ arch/riscv/kernel/entry.S | 7 +++++++ arch/riscv/kernel/irq.c | 21 +++++++++++++++++++++ 3 files changed, 35 insertions(+) (limited to 'arch') diff --git a/arch/riscv/include/asm/scs.h b/arch/riscv/include/asm/scs.h index 94726ea773e3..0e45db78b24b 100644 --- a/arch/riscv/include/asm/scs.h +++ b/arch/riscv/include/asm/scs.h @@ -13,6 +13,11 @@ XIP_FIXUP_OFFSET gp .endm +/* Load the per-CPU IRQ shadow call stack to gp. */ +.macro scs_load_irq_stack tmp + load_per_cpu gp, irq_shadow_call_stack_ptr, \tmp +.endm + /* Load task_scs_sp(current) to gp. */ .macro scs_load_current REG_L gp, TASK_TI_SCS_SP(tp) @@ -34,6 +39,8 @@ _skip_scs: .macro scs_load_init_stack .endm +.macro scs_load_irq_stack tmp +.endm .macro scs_load_current .endm .macro scs_load_current_if_task_changed prev diff --git a/arch/riscv/kernel/entry.S b/arch/riscv/kernel/entry.S index 52793193a763..3a0db310325a 100644 --- a/arch/riscv/kernel/entry.S +++ b/arch/riscv/kernel/entry.S @@ -237,12 +237,19 @@ SYM_FUNC_START(call_on_irq_stack) REG_S s0, STACKFRAME_FP(sp) addi s0, sp, STACKFRAME_SIZE_ON_STACK + /* Switch to the per-CPU shadow call stack */ + scs_save_current + scs_load_irq_stack t0 + /* Switch to the per-CPU IRQ stack and call the handler */ load_per_cpu t0, irq_stack_ptr, t1 li t1, IRQ_STACK_SIZE add sp, t0, t1 jalr a1 + /* Switch back to the thread shadow call stack */ + scs_load_current + /* Switch back to the thread stack and restore ra and s0 */ addi sp, s0, -STACKFRAME_SIZE_ON_STACK REG_L ra, STACKFRAME_RA(sp) diff --git a/arch/riscv/kernel/irq.c b/arch/riscv/kernel/irq.c index 95dafdcbd135..7bfea97ee7e7 100644 --- a/arch/riscv/kernel/irq.c +++ b/arch/riscv/kernel/irq.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include @@ -34,6 +35,24 @@ EXPORT_SYMBOL_GPL(riscv_get_intc_hwnode); #ifdef CONFIG_IRQ_STACKS #include +DECLARE_PER_CPU(ulong *, irq_shadow_call_stack_ptr); + +#ifdef CONFIG_SHADOW_CALL_STACK +DEFINE_PER_CPU(ulong *, irq_shadow_call_stack_ptr); +#endif + +static void init_irq_scs(void) +{ + int cpu; + + if (!scs_is_enabled()) + return; + + for_each_possible_cpu(cpu) + per_cpu(irq_shadow_call_stack_ptr, cpu) = + scs_alloc(cpu_to_node(cpu)); +} + DEFINE_PER_CPU(ulong *, irq_stack_ptr); #ifdef CONFIG_VMAP_STACK @@ -76,6 +95,7 @@ void do_softirq_own_stack(void) #endif /* CONFIG_HAVE_SOFTIRQ_ON_OWN_STACK */ #else +static void init_irq_scs(void) {} static void init_irq_stacks(void) {} #endif /* CONFIG_IRQ_STACKS */ @@ -87,6 +107,7 @@ int arch_show_interrupts(struct seq_file *p, int prec) void __init init_IRQ(void) { + init_irq_scs(); init_irq_stacks(); irqchip_init(); if (!handle_arch_irq) -- cgit v1.2.3-58-ga151