gcry_mpi_sub_ui result is positive when it should be negative

[Jussi Kivilinna]

Hello,

On 30.11.2021 15.21, Guido Vranken via Gcrypt-devel wrote:
> In the program below, the result of the computation -5 - 2 should be -7, but it is 7.
> 
> Compare with the behavior of gcry_mpi_add_ui and gcry_mpi_mul_ui (commented), which do produce a negative result, as they should.
> 
> This was tested on the latest repository checkout.

Thanks for report.

This looks to go all the way back to code from 1997, commit "initial checkin" (4b5e71ca4e84e61e595dec19e1c7cab0c0a73f24).

-Jussi

> 
> #include <gcrypt.h>
> 
> #define CF_CHECK_EQ(expr, res) if ( (expr) != (res) ) { goto end; }
> 
> int main(void)
> {
>      gcry_mpi_t A;
>      gcry_mpi_t res;
>      gcry_error_t err;
>      char *buf;
> 
>      CF_CHECK_EQ(err = gcry_mpi_scan(&A, GCRYMPI_FMT_HEX, "-5", 0, NULL), 0);
>      CF_CHECK_EQ(err = gcry_mpi_scan(&res, GCRYMPI_FMT_HEX, "0", 0, NULL), 0);
>      gcry_mpi_sub_ui(res, A, 2);
>      //gcry_mpi_add_ui(res, A, 2);
>      //gcry_mpi_mul_ui(res, A, 2);
>      CF_CHECK_EQ(err = gcry_mpi_aprint(GCRYMPI_FMT_HEX, (unsigned char**)&buf, NULL, res), 0);
>      printf("%s\n", buf);
> end:
> 
>      return 0;
> }
> 
> _______________________________________________
> Gcrypt-devel mailing list
> Gcrypt-devel at gnupg.org
> http://lists.gnupg.org/mailman/listinfo/gcrypt-devel
>