2015-01-26 20:38 GMT+01:00 Rich Felker <[email protected]>: > On Mon, Jan 26, 2015 at 08:31:08PM +0100, Daniel Cegiełka wrote: >> 2015-01-26 20:23 GMT+01:00 Rich Felker <[email protected]>:
> In particular, one thing I wonder about is whether it requires O(n) > temporary working space in the number of input symbols. This seems to > be a big problem for huge projects. The design I've been working on is > only O(n) in the number of input translation units, not symbols, and > simply uses the input symbol tables in-place with some cheap data > structures to reduce the typical search time down from the O(n) it > would otherwise be by taking this approach. I wonder whether something > like this is feasible with the constraint of using existing > well-documented ELF library APIs though. I can't say, I know this project to poorly. I know it works on linux, because I have it installed. I plan to move it to the musl and gnu make (in the future). Take into account that FreeBSD has some very large projects and they are aware of this (probably). You have to ask ELF Tool Chain's developers :) Daniel > Rich _______________________________________________ Toybox mailing list [email protected] http://lists.landley.net/listinfo.cgi/toybox-landley.net
