This is a quote from Rusty Russell about the FPU from his kernel guide: "The FPU context is not saved; even in user context the FPU state probably won't correspond with the current process: you would mess with some user process' FPU state. If you really want to do this, you would have to explicitly save/restore the full FPU state (and avoid context switches)." On 30 Jul 2002, Kashif Shaikh wrote: > On Tue, 2002-07-30 at 15:19, Tom Bradley wrote: > > > > > > Why Can't I use 'float' inside the kernel? > > > > There are 2 main why you shouldn't use the 'float' or 'double' data > > type inside of the kernel. > > > > - The kernel does not save the FPU registers, therefore the FPU is > > always in an unknown state. > Don't you mean the kernel does not *always* save the FPU registers? I > mean, doesn't the kernel save the FPU regs if a user-space process is > using 'em(less time taken for context-switch then)? > > -- Kashif > > > > - Due to percision differences on different platforms the value of a > > 'float' or 'double' can not be completely trusted to be the same. > > > > > > Tom > > -- Kernelnewbies: Help each other learn about the Linux kernel. Archive: http://mail.nl.linux.org/kernelnewbies/ FAQ: http://kernelnewbies.org/faq/
