Re: [PATCH 2/2] target/s390x: Fix determination of overflow condition code after subtraction

[David Hildenbrand]

On 23.03.22 17:26, Thomas Huth wrote:
> Reported by Paul Eggert in
> https://lists.gnu.org/archive/html/bug-gnulib/2021-09/msg00050.html
> 
> This program currently prints different results when run with TCG instead
> of running on real s390x hardware:
> 
>  #include <stdio.h>
> 
>  int overflow_32 (int x, int y)
>  {
>    int sum;
>    return __builtin_sub_overflow (x, y, &sum);
>  }
> 
>  int overflow_64 (long long x, long long y)
>  {
>    long sum;
>    return __builtin_sub_overflow (x, y, &sum);
>  }
> 
>  int a1 = 0;
>  int b1 = -2147483648;
>  long long a2 = 0L;
>  long long b2 = -9223372036854775808L;
> 
>  int main ()
>  {
>    {
>      int a = a1;
>      int b = b1;
>      printf ("a = 0x%x, b = 0x%x\n", a, b);
>      printf ("no_overflow = %d\n", ! overflow_32 (a, b));
>    }
>    {
>      long long a = a2;
>      long long b = b2;
>      printf ("a = 0x%llx, b = 0x%llx\n", a, b);
>      printf ("no_overflow = %d\n", ! overflow_64 (a, b));
>    }
>  }
> 
> Resolves: https://gitlab.com/qemu-project/qemu/-/issues/618
> Suggested-by: Bruno Haible <bruno@clisp.org>
> Signed-off-by: Thomas Huth <thuth@redhat.com>
> ---
>  target/s390x/tcg/cc_helper.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/target/s390x/tcg/cc_helper.c b/target/s390x/tcg/cc_helper.c
> index e11cdb745d..b2e8d3d9f5 100644
> --- a/target/s390x/tcg/cc_helper.c
> +++ b/target/s390x/tcg/cc_helper.c
> @@ -151,7 +151,7 @@ static uint32_t cc_calc_add_64(int64_t a1, int64_t a2, 
> int64_t ar)
>  
>  static uint32_t cc_calc_sub_64(int64_t a1, int64_t a2, int64_t ar)
>  {
> -    if ((a1 > 0 && a2 < 0 && ar < 0) || (a1 < 0 && a2 > 0 && ar > 0)) {
> +    if ((a1 >= 0 && a2 < 0 && ar < 0) || (a1 < 0 && a2 > 0 && ar > 0)) {
>          return 3; /* overflow */
>      } else {
>          if (ar < 0) {
> @@ -211,7 +211,7 @@ static uint32_t cc_calc_add_32(int32_t a1, int32_t a2, 
> int32_t ar)
>  
>  static uint32_t cc_calc_sub_32(int32_t a1, int32_t a2, int32_t ar)
>  {
> -    if ((a1 > 0 && a2 < 0 && ar < 0) || (a1 < 0 && a2 > 0 && ar > 0)) {
> +    if ((a1 >= 0 && a2 < 0 && ar < 0) || (a1 < 0 && a2 > 0 && ar > 0)) {
>          return 3; /* overflow */
>      } else {
>          if (ar < 0) {

Again, intuitively I'd check for

a) Subtracting a negative number from a positive one -> Adding two
positive numbers should result in the result being bigger than the first
parameter.

a1 > 0 && a2 < 0 && ar < a1

a) Subtracting a positive number from a negative one -> Adding two
negative numbers should result in something that's smaller than the
first parameter

a1 < 0 && a2 > 0 && ar > a1


But maybe I am missing something :)

-- 
Thanks,

David / dhildenb