RFR 8214262: SunEC native code does not compile with debug on

[Jamil Nimeh]

Hi Max,

The changes look fine.  I did notice in the files that there is a header 
block comment that has a last modified date from the original code.  
Should those be updated to 2018?

--Jamil

On 11/23/2018 6:31 AM, Weijun Wang wrote:
> I was curious in some SunEC internals and turned EC_DEBUG on, and 
> found out the library cannot be built. It looks like back when we 
> switched to warnings-as-errors some codes are not updated. Here is the 
> patch:
>
> *diff --git a/src/jdk.crypto.ec/share/native/libsunec/impl/ec.c 
> b/src/jdk.crypto.ec/share/native/libsunec/impl/ec.c*
> *--- a/src/jdk.crypto.ec/share/native/libsunec/impl/ec.c*
> *+++ b/src/jdk.crypto.ec/share/native/libsunec/impl/ec.c*
> @@ -109,16 +109,16 @@
>   printf("\n");
>
>
>       if (k1 != NULL) {
> -               mp_tohex(k1, mpstr);
> +               mp_tohex((mp_int*)k1, mpstr);
>               printf("ec_points_mul: scalar k1: %s\n", mpstr);
> -               mp_todecimal(k1, mpstr);
> +               mp_todecimal((mp_int*)k1, mpstr);
>               printf("ec_points_mul: scalar k1: %s (dec)\n", mpstr);
>       }
>
>
>       if (k2 != NULL) {
> -               mp_tohex(k2, mpstr);
> +               mp_tohex((mp_int*)k2, mpstr);
>               printf("ec_points_mul: scalar k2: %s\n", mpstr);
> -               mp_todecimal(k2, mpstr);
> +               mp_todecimal((mp_int*)k2, mpstr);
>               printf("ec_points_mul: scalar k2: %s (dec)\n", mpstr);
>       }
>
>
> *diff --git a/src/jdk.crypto.ec/share/native/libsunec/impl/ecdecode.c 
> b/src/jdk.crypto.ec/share/native/libsunec/impl/ecdecode.c*
> *--- a/src/jdk.crypto.ec/share/native/libsunec/impl/ecdecode.c*
> *+++ b/src/jdk.crypto.ec/share/native/libsunec/impl/ecdecode.c*
> @@ -55,6 +55,9 @@
>  #include "ecl-curve.h"
>  #include "ecc_impl.h"
>
>
> +#if EC_DEBUG
> +#include <stdio.h>
> +#endif
>  #define MAX_ECKEY_LEN           72
>  #define SEC_ASN1_OBJECT_ID      0x06
>
> In the first file, k1 and k2 are defined as const mp_int * but 
> mp_tohex accepts a mp_int* as its 1st argument.
>
> In the second file, there is a printf call in a EC_DEBUG block.
>
> Noreg-cleanup.
>
> Thanks
> Max
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/security-dev/attachments/20181124/b49f557e/attachment.htm>