Re: [PATCH v4 2/4] sysemu/os-win32: fix setjmp/longjmp on windows-arm64

[Pierrick Bouvier <pierrick.bouvier@linaro.org>]

On 2/21/23 23:27, Philippe Mathieu-Daudé wrote:
> On 21/2/23 16:30, Pierrick Bouvier wrote:
>> Windows implementation of setjmp/longjmp is done in
>> C:/WINDOWS/system32/ucrtbase.dll. Alas, on arm64, it seems to *always*
>> perform stack unwinding, which crashes from generated code.
>>
>> By using alternative implementation built in mingw, we avoid doing stack
>> unwinding and this fixes crash when calling longjmp.
>>
>> Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
>> Acked-by: Richard Henderson <richard.henderson@linaro.org>
>> ---
>>    include/sysemu/os-win32.h | 28 ++++++++++++++++++++++++----
>>    meson.build               | 21 +++++++++++++++++++++
>>    2 files changed, 45 insertions(+), 4 deletions(-)
> 
> 
>> -#if defined(_WIN64)
>> -/* On w64, setjmp is implemented by _setjmp which needs a second parameter.
>> +#if defined(__aarch64__)
>> +/*
>> + * On windows-arm64, setjmp is available in only one variant, and longjmp always
>> + * does stack unwinding. This crash with generated code.
>> + * Thus, we use another implementation of setjmp (not windows one), coming from
>> + * mingw, which never performs stack unwinding.
>> + */
>> +#undef setjmp
>> +#undef longjmp
>> +/*
>> + * These functions are not declared in setjmp.h because __aarch64__ defines
>> + * setjmp to _setjmpex instead. However, they are still defined in libmingwex.a,
>> + * which gets linked automatically.
>> + */
>> +extern int __mingw_setjmp(jmp_buf);
>> +extern void __attribute__((noreturn)) __mingw_longjmp(jmp_buf, int);
>> +#define setjmp(env) __mingw_setjmp(env)
>> +#define longjmp(env, val) __mingw_longjmp(env, val)
>> +#elif defined(_WIN64)
>> +/*
>> + * On windows-x64, setjmp is implemented by _setjmp which needs a second parameter.
>>     * If this parameter is NULL, longjump does no stack unwinding.
>>     * That is what we need for QEMU. Passing the value of register rsp (default)
>> - * lets longjmp try a stack unwinding which will crash with generated code. */
>> + * lets longjmp try a stack unwinding which will crash with generated code.
>> + */
>>    # undef setjmp
>>    # define setjmp(env) _setjmp(env, NULL)
>> -#endif
>> +#endif /* __aarch64__ */
> 
> This comment is confusing, the previous if ladder is about i86. Maybe
> better not add any comment?

If I am not mistaken, before we had:

#if x64
define setjmp as _setjmp(env, 0)
#endif
// nothing done for x86

and now we have:

#if aarch64
define setjmp as __mingw_setjmp
define longjmp as __mingw_longjmp
#elif x64
define setjmp as _setjmp(env, 0)
#endif
// nothing done for x86

Maybe the patch format is confusing, or I missed what you pointed.

Pierrick

> 
> Otherwise,
> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
>