Am 06.07.2015 um 05:23 schrieb Jeffrey Walton: >> ... >> ------ Build started: Project: cryptlib, Configuration: DLL-Import Release >> x64 ------ >> Build started 7/3/2015 7:07:03 PM. >> InitializeBuildStatus: >> Touching "x64\cryptlib\DLL-Import Release\cryptlib.unsuccessfulbuild". >> CustomBuild: >> All outputs are up-to-date. >> ClCompile: >> All outputs are up-to-date. >> hrtimer.cpp >> hrtimer.cpp(98): warning C4273: >> 'CryptoPP::ThreadUserTimer::GetCurrentTimerValue' : inconsistent dll linkage >> ...\cryptopp\hrtimer.h(46) : see previous definition of >> 'GetCurrentTimerValue' >> hrtimer.cpp(128): warning C4273: 'CryptoPP::ThreadUserTimer::TicksPerSecond' >> : inconsistent dll linkage >> ...\cryptopp\hrtimer.h(47) : see previous definition of >> 'TicksPerSecond' >> >> This looks like it was caused by your edit 26 days ago, you added >> CRYPTOPP_DLL to the class declaration. >> I'm not sure if this actually breaks something, we'd have to try if we can >> use these two function in a DLL-build (which I actually never use) >> > Ah, OK. I'm working on sorting this out now. > > The issue here is the Crypto++ DLL *is not* a DLL version of the > static library. The DLL is a purpose-built DLL used for FIPS. That's > why the DLL is missing so many things - it has to provide a security > boundary. > > I'm thinking part of the solution is to have 2 DLLs in the solution. > One is for non-FIPS (i.e., behaves like users expect - it mirrors the > static library), and the other is to have FIPS DLL (i.e., provides the > security boundary and validated cryptography). The FIPS validated one would only include FIPS validated algorithms and the other would include everything? Sounds good, although it may not be trivial to do this. Personally I'd still not use the DLL, as static libs are still better, but there may be users having a different opinion than me. As a bonus we could include it in our installer (the latest FIPS validated DLL (?), the current "FIPS"-DLL, the current Full-DLL and the current static Lib, making it 200MB+ :) ) >> ------ Build started: Project: cryptest, Configuration: Release Win32 ------ >> Build started 7/3/2015 7:09:02 PM. >> InitializeBuildStatus: >> Touching "Win32\cryptest\Release\cryptest.unsuccessfulbuild". >> ClCompile: >> All outputs are up-to-date. >> All outputs are up-to-date. >> LINK : fatal error C1905: Front end and back end not compatible (must target >> same processor). >> >> Did you link against the 64-bit library here? > Well, I did not do any thing other than a fresh checkout, Visual > Studio 2008 -> 2010 conversion, and then Build -> Batch Build -> All. Strange... This shouldn't happen... I've just downloaded the current sources from GitHub and ran it through VS2010 Professional and it built perfectly (with some warnings we have to fix...) -> No trivial repro :(
BR JPM > > We are also going to address the gaps in the engineering process in a > separate thread (its already been started with the question on CI). > > Jeff > -- -- You received this message because you are subscribed to the "Crypto++ Users" Google Group. To unsubscribe, send an email to [email protected]. More information about Crypto++ and this group is available at http://www.cryptopp.com. --- You received this message because you are subscribed to the Google Groups "Crypto++ Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
smime.p7s
Description: S/MIME Cryptographic Signature
