Hi, please correct me if I'm wrong: - the kernel needs enough ram for the compressed initramfs and the same files decompressed, so it can decompress the files and boot the /init on the initramfs. - if the kernel does not have enough ram, it will kernel panic. my tests were done on a server with 256 mb ram and a 80MB compressed initrd. I removed some packages and now it fits and the kernel no longer segfaults. I know this is maybe not the intendet audience for initramfs, but otherwise it works fine. Regards, Andreas -- ---------------------[ Ciphire Signature ]---------------------- From: aj@ciphirelabs.com signed email body (425 characters) Date: on 22 March 2005 at 14:22:30 UTC To: klibc@zytor.com ---------------------------------------------------------------- : Ciphire has secured this email against identity theft. : Free download at www.ciphire.com. The garbled lines : below are the sender's verifiable digital signature. ---------------------------------------------------------------- 00fAAAAAEAAAAmKkBCqQEAAPkBAAIAAgACACCFlK4A+54we7zwPJsjFBflkYBMj+ KIfUWI8cEb4zJitwEAjuLZDHPnOxRgKfQyXaVwrQqQB78iLphbN7mW+8mUnEEdXX qn5nWzaCtHdOWpZjyKS6aoeBm3yXpEfvZQc1EHSw=------------------[ End Ciphire Signed Message ]----------------
Andreas Jellinghaus wrote:> Hi, > > please correct me if I'm wrong: > - the kernel needs enough ram for the compressed initramfs and the > same files decompressed, so it can decompress the files and boot > the /init on the initramfs. > - if the kernel does not have enough ram, it will kernel panic. > > my tests were done on a server with 256 mb ram and a 80MB > compressed initrd. I removed some packages and now it fits > and the kernel no longer segfaults. I know this is maybe not > the intendet audience for initramfs, but otherwise it works > fine. > > Regards, Andreas >That is correct. It would be possible to reduce the total memory requirement by making the decompressor free the compressed pages as it works, but that is not currently done. -hpa