|
Message-ID: <20150922003214.GA12554@openwall.com> Date: Tue, 22 Sep 2015 03:32:14 +0300 From: Solar Designer <solar@...nwall.com> To: john-dev@...ts.openwall.com Subject: Re: Agnieszka's GSoC formats On Tue, Sep 22, 2015 at 01:57:04AM +0200, magnum wrote: > The bots also spell out some problems for you: > > opencl_yescrypt_fmt_plug.c:546:9: warning: comparison of constant > 2251799813685247 with expression of type 'uint32_t' (aka > 'unsigned int') > is always false [-Wtautological-constant-out-of-range-compare] > if (p > SIZE_MAX / Sbytes) { > ~ ^ ~~~~~~~~~~~~~~~~~ > 1 warning generated. > yescrypt_fmt_plug.c:344:9: warning: comparison of constant > 2251799813685247 with > expression of type 'uint32_t' (aka 'unsigned int') is always false > [-Wtautological-constant-out-of-range-compare] > if (p > SIZE_MAX / Sbytes) { > ~ ^ ~~~~~~~~~~~~~~~~~ > 1 warning generated. > In file included from yescrypt-best_plug.c:2: > ./yescrypt-simd.c:1093:9: warning: comparison of constant > 2251799813685247 with > expression of type 'uint32_t' (aka 'unsigned int') is always false > [-Wtautological-constant-out-of-range-compare] > if (p > SIZE_MAX / Sbytes) { > ~ ^ ~~~~~~~~~~~~~~~~~ > 1 warning generated. These are certainly just warnings, not indicative of any real bugs. The lines in question originate from my yescrypt tree, and they are in there in case the compile-time settings are different or/and we're building for a 32-bit platform. It's possible to silence the warnings with some #if's around these checks, but this would actually introduce a risk of failing to perform the checks in a build where they are needed. I'll see what I can do about this in a future version of yescrypt. Meanwhile, feel free to #if them in jumbo somehow, even though in this usage yescrypt is fairly likely to face untrusted salt strings. Alexander
Powered by blists - more mailing lists
Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.