[PATCH] riscv: trace: fix snapshot deadlock with sbi ecall

[PATCH] riscv: trace: fix snapshot deadlock with sbi ecall

[Martin Kaiser]

If sbi_ecall.c's functions are traceable,

echo "__sbi_ecall:snapshot" > /sys/kernel/tracing/set_ftrace_filter

may get the kernel into a deadlock.

(Functions in sbi_ecall.c are excluded from tracing if
CONFIG_RISCV_ALTERNATIVE_EARLY is set.)

__sbi_ecall triggers a snapshot of the ringbuffer. The snapshot code
raises an IPI interrupt, which results in another call to __sbi_ecall
and another snapshot...

All it takes to get into this endless loop is one initial __sbi_ecall.
On RISC-V systems without SSTC extension, the clock events in
timer-riscv.c issue periodic sbi ecalls, making the problem easy to
trigger.

Always exclude the sbi_ecall.c functions from tracing to fix the
potential deadlock.

sbi ecalls can easiliy be logged via trace events, excluding ecall
functions from function tracing is not a big limitation.

Signed-off-by: Martin Kaiser <martin@kaiser.cx>
---
 arch/riscv/kernel/Makefile | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/arch/riscv/kernel/Makefile b/arch/riscv/kernel/Makefile
index 9026400cba10..cabb99cadfb6 100644
--- a/arch/riscv/kernel/Makefile
+++ b/arch/riscv/kernel/Makefile
@@ -3,12 +3,6 @@
 # Makefile for the RISC-V Linux kernel
 #
 
-ifdef CONFIG_FTRACE
-CFLAGS_REMOVE_ftrace.o	= $(CC_FLAGS_FTRACE)
-CFLAGS_REMOVE_patch.o	= $(CC_FLAGS_FTRACE)
-CFLAGS_REMOVE_sbi.o	= $(CC_FLAGS_FTRACE)
-CFLAGS_REMOVE_return_address.o	= $(CC_FLAGS_FTRACE)
-endif
 CFLAGS_syscall_table.o	+= $(call cc-disable-warning, override-init)
 CFLAGS_compat_syscall_table.o += $(call cc-disable-warning, override-init)
 
@@ -24,7 +18,6 @@ CFLAGS_sbi_ecall.o := -mcmodel=medany
 ifdef CONFIG_FTRACE
 CFLAGS_REMOVE_alternative.o = $(CC_FLAGS_FTRACE)
 CFLAGS_REMOVE_cpufeature.o = $(CC_FLAGS_FTRACE)
-CFLAGS_REMOVE_sbi_ecall.o = $(CC_FLAGS_FTRACE)
 endif
 ifdef CONFIG_RELOCATABLE
 CFLAGS_alternative.o += -fno-pie
@@ -43,6 +36,14 @@ CFLAGS_sbi_ecall.o += -D__NO_FORTIFY
 endif
 endif
 
+ifdef CONFIG_FTRACE
+CFLAGS_REMOVE_ftrace.o	= $(CC_FLAGS_FTRACE)
+CFLAGS_REMOVE_patch.o	= $(CC_FLAGS_FTRACE)
+CFLAGS_REMOVE_sbi.o	= $(CC_FLAGS_FTRACE)
+CFLAGS_REMOVE_return_address.o	= $(CC_FLAGS_FTRACE)
+CFLAGS_REMOVE_sbi_ecall.o = $(CC_FLAGS_FTRACE)
+endif
+
 always-$(KBUILD_BUILTIN) += vmlinux.lds
 
 obj-y	+= head.o
-- 
2.43.7

Re: [PATCH] riscv: trace: fix snapshot deadlock with sbi ecall

[Paul Walmsley]

On Tue, 23 Dec 2025, Martin Kaiser wrote:

> If sbi_ecall.c's functions are traceable,
> 
> echo "__sbi_ecall:snapshot" > /sys/kernel/tracing/set_ftrace_filter
> 
> may get the kernel into a deadlock.
> 
> (Functions in sbi_ecall.c are excluded from tracing if
> CONFIG_RISCV_ALTERNATIVE_EARLY is set.)
> 
> __sbi_ecall triggers a snapshot of the ringbuffer. The snapshot code
> raises an IPI interrupt, which results in another call to __sbi_ecall
> and another snapshot...
> 
> All it takes to get into this endless loop is one initial __sbi_ecall.
> On RISC-V systems without SSTC extension, the clock events in
> timer-riscv.c issue periodic sbi ecalls, making the problem easy to
> trigger.
> 
> Always exclude the sbi_ecall.c functions from tracing to fix the
> potential deadlock.
> 
> sbi ecalls can easiliy be logged via trace events, excluding ecall
> functions from function tracing is not a big limitation.
> 
> Signed-off-by: Martin Kaiser <martin@kaiser.cx>

Thanks, queued for v6.19-rc.

- Paul