On Thu, May 14, 2020 at 05:39:53PM +0200, Jeremie Courreges-Anglas wrote:
> > Sounds good. Probably will prevent a future phantom failure if py-M2Crypto
> > gets junked during build.
> Adding m2crypto to RUN_DEPENDS won't prevent a failure at build
> time.
You're right. I plead "trying to recove
On Thu, May 14 2020, Kurt Mosiejczuk wrote:
> On Thu, May 14, 2020 at 03:22:27PM +0200, Theo Buehler wrote:
>> This an optional runtime dependency that speeds up the some of the
>> tlsfuzzer tests quite a bit (by a factor of 4-6 on my laptop) simply by
>> replacing the native python crypto impleme
On Thu, May 14, 2020 at 03:22:27PM +0200, Theo Buehler wrote:
> This an optional runtime dependency that speeds up the some of the
> tlsfuzzer tests quite a bit (by a factor of 4-6 on my laptop) simply by
> replacing the native python crypto implementations by faster versions.
> This is purely con
On Thu, May 14 2020, Theo Buehler wrote:
> This an optional runtime dependency that speeds up the some of the
> tlsfuzzer tests quite a bit (by a factor of 4-6 on my laptop) simply by
> replacing the native python crypto implementations by faster versions.
>
> This is purely convenience. I can al
