From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Korsgaard Date: Fri, 25 Oct 2019 21:23:18 +0200 Subject: [Buildroot] [PATCH] package/jitterentropy-library: fix build failure due to missing -fPIC In-Reply-To: <2f610d45-59a6-da96-b32e-ed1ff8c2dfbd@mind.be> (Arnout Vandecappelle's message of "Fri, 25 Oct 2019 20:52:53 +0200") References: <20191023110605.23836-1-giulio.benetti@benettiengineering.com> <831e0090-8e52-9c2c-6d06-4557d21fde79@mind.be> <8736fgkdas.fsf@dell.be.48ers.dk> <2f610d45-59a6-da96-b32e-ed1ff8c2dfbd@mind.be> Message-ID: <87d0ekiq8p.fsf@dell.be.48ers.dk> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net >>>>> "Arnout" == Arnout Vandecappelle writes: Hi, >> > Hm, but maybe there are indeed good reasons to override the CFLAGS, because the >> > fstack-protector-all requires support from the toolchain. On the other hand, we >> > probably *do* want to keep the -O0 and -fwrapv flags... >> >> The fstack-protector-* options should be handled by BR2_SSP_* and >> hardcoded in the package. > If with "hardcoded in the package" you mean "set by the package's .mk file", > then I disagree, since it is passed by the toolchain wrapper. Sorry, there was a NOT missing, E.G. NOT hardcoded in .mk. >> Why would we want to use -O0? > The package description is "Hardware RNG based on CPU timing jitter." I expect > that compiler optimisations mess with CPU timing jitter. Ah yes, possibly. -- Bye, Peter Korsgaard