Mr. Bouissou: Perhaps you will feel compelled to publish your RPM generating scripts. If you do, you may find people uploading RPMs that they built out of work they had to do anyway in using the crpytoapi stuff. I am still curious, how Microsoft is able to distribute their operating system with crypto APIs in it, and Linux is not! This is crazy! What are we doing wrong? Very Respectfully, Stuart Blake Tener, IT3, USNR-R, N3GWG Beverly Hills, California VTU 1904G (Volunteer Training Unit) stuart@xxxxxxxxxxx west coast: (310)-358-0202 P.O. Box 16043, Beverly Hills, CA 90209-2043 east coast: (215)-338-6005 P.O. Box 45859, Philadelphia, PA 19149-5859 Telecopier: (419)-715-6073 fax to email gateway via www.efax.com (it's free!) JOIN THE US NAVY RESERVE, SERVE YOUR COUNTRY, AND BENEFIT FROM IT ALL. Friday, September 28, 2001 12:55 AM -----Original Message----- From: Michel Bouissou [mailto:michel@xxxxxxxxxxxx] Sent: Wednesday, September 26, 2001 2:54 PM To: stuart@xxxxxxxxxxx; linux-crypto@xxxxxxxxxxxx Subject: Re: RPM for LOOP-AES? / RPMs for new MDK stuff Le Mercredi 26 Septembre 2001 19:24, vous avez écrit : > > Do you plan to place the RPMs on the sourceforge site? I'll be glad if the project admin for cryptoapi wants to mirror my RPMs on the project official page, yes. But that's up to him ;-) > Any possibility of seeing an RPM for LOOP-AES (source/binary)? Well, I don't know the loop-aes stuff well, but I've taken a quick look at it. As far as I understand, what it does is just a subset of what cryptoapi does, so I don't feel the urge of RPMing it. But I may miss some point. According to the README of loop-aes, it necessitates to recompile the kernel just to remove kernel original loop device support. That would mean either to create again a number of kernel binary RPM packages just for changing this very little configuration option (and I'm reluctant to do so...), or rely on loop-aes users to recompile their kernel by themselves... And I feel that if the users feel comfortable enough to recompile the kernel by themselves, they don't really need a RPM for loop-aes, they can built its binary by themselves as well... The matter is always with the kernel. For cryptoapi, I have built 100+ megabytes of RPMs including different flavours of the kernel with SRPMS, just for a little patch which source is about 2 KB... And the cryptoapi binary RPM is only about 200 KB, and losetup + mount patched binary RPMs are about 100 KB total. That makes 300 KB of real meat, but a total of 100+ MB of RPMs because the kernel has to be there as well :-((( I really hope the iv-mode-sector patch will integrate the vanilla kernel soon!!! Building the complete set of kernel RPMs took about 5-6 hours of compilation on my machine, and used more than 500 MB of temporary space on my disk... Now you may understand why I'm reluctant to do it every couple of days ;-) Best regards. -- michel@xxxxxxxxxxxx OpenPGP DH/DSS ID 0x5C2BEE8F Si vous préférez que votre e-mail privé reste... privé. Utilisez GnuPG: http://www.geocities.com/openpgp Linux-crypto: cryptography in and on the Linux system Archive: http://mail.nl.linux.org/linux-crypto/
