Max Vozeler wrote: > I've been thinking about this a bit. Do you think this could - in > principle - be solved by changes to the build system? > > a) Adding a "ciphers" build target for building (and installing) > just the cipher modules, skipping loop.$(VM1)o, be a workable > approach? I've hacked together a proof-of-concept patch to that > effect; Note that it is quite surely not working as-is, it is > merely an illustration and meant as basis for discussion. Should work ok, but I prefer EXTRA_CIPHERS=y make parameter. > b) Alternatively, although not clean, maybe the extra unneeded > loop.$(VM1)o could still be built and installed; If I see it > correctly, it should do no harm then. I admit that this is not > a nice solution, but it might(?) be reasonable if patch+ciphers > is only seldom used. Just thinking aloud.. What do you think if loop-AES package is modified to: 1) Include additional cipher module sources. 2) Only build additional cipher modules if EXTRA_CIPHERS=y make command line parameter is present. 3) Makefile does not build loop module if BUILD_LOOP=n make command line parameter is present. 4) USE_KBUILD=y is made default for all 2.6 kernels. -- Jari Ruusu 1024R/3A220F51 5B 4B F9 BB D3 3F 52 E9 DB 1D EB E3 24 0E A9 DD - Linux-crypto: cryptography in and on the Linux system Archive: http://mail.nl.linux.org/linux-crypto/
