Christophe Leroy <christophe.le...@csgroup.eu> writes:
> Le 09/02/2024 à 08:59, Naveen N Rao a écrit :
>> Michael reported that we are seeing ftrace bug on bootup when KASAN is
>> enabled, and if we are using -fpatchable-function-entry:
>>
...
>> diff --git a/arch/powerpc/include/asm/sections.h
>> b/arch/powerpc/include/asm/sections.h
>> index ea26665f82cf..d389dcecdb0b 100644
>> --- a/arch/powerpc/include/asm/sections.h
>> +++ b/arch/powerpc/include/asm/sections.h
>> @@ -14,6 +14,7 @@ typedef struct func_desc func_desc_t;
>>
>> extern char __head_end[];
>> extern char __srwx_boundary[];
>> +extern char _sexittext[], _eexittext[];
>
> Should we try to at least use the same symbols as others, or best try to
> move this into include/asm-generic/sections.h, just like inittext ?
>
> $ git grep exittext
> arch/arm64/include/asm/sections.h:extern char __exittext_begin[],
> __exittext_end[];
> arch/arm64/kernel/patching.c: addr >= (unsigned
> long)__exittext_begin &&
> arch/arm64/kernel/patching.c: addr < (unsigned
> long)__exittext_end;
> arch/arm64/kernel/vmlinux.lds.S: __exittext_begin = .;
> arch/arm64/kernel/vmlinux.lds.S: __exittext_end = .;
> arch/riscv/include/asm/sections.h:extern char __exittext_begin[],
> __exittext_end[];
> arch/riscv/kernel/patch.c:static inline bool
> is_kernel_exittext(uintptr_t addr)
> arch/riscv/kernel/patch.c: addr >=
> (uintptr_t)__exittext_begin &&
> arch/riscv/kernel/patch.c: addr < (uintptr_t)__exittext_end;
> arch/riscv/kernel/patch.c: if (core_kernel_text(uintaddr) ||
> is_kernel_exittext(uintaddr))
> arch/riscv/kernel/vmlinux-xip.lds.S: __exittext_begin = .;
> arch/riscv/kernel/vmlinux-xip.lds.S: __exittext_end = .;
> arch/riscv/kernel/vmlinux.lds.S: __exittext_begin = .;
> arch/riscv/kernel/vmlinux.lds.S: __exittext_end = .;
I'll change it to use __exittext_begin/end.
>> diff --git a/arch/powerpc/kernel/trace/ftrace.c
>> b/arch/powerpc/kernel/trace/ftrace.c
>> index 82010629cf88..b5efd8d7bc01 100644
>> --- a/arch/powerpc/kernel/trace/ftrace.c
>> +++ b/arch/powerpc/kernel/trace/ftrace.c
>> @@ -27,10 +27,22 @@
>> #include <asm/ftrace.h>
>> #include <asm/syscall.h>
>> #include <asm/inst.h>
>> +#include <asm/sections.h>
>>
>> #define NUM_FTRACE_TRAMPS 2
>> static unsigned long ftrace_tramps[NUM_FTRACE_TRAMPS];
>>
>> +unsigned long ftrace_call_adjust(unsigned long addr)
>> +{
>> + if (addr >= (unsigned long)_sexittext && addr < (unsigned
>> long)_eexittext)
>> + return 0;
>
> Then arm64 has a function called is_exit_text() and riscv has
> is_kernel_exittext(). Can we refactor ?
I'd like to get the fix in and backported, so I'll take it as-is but
with the section names changed to match the other arches.
We can do further refactoring on top.
cheers
