On Mon, 2003-01-27 at 13:11, Anthony DeRobertis wrote: > On Mon, 2003-01-27 at 07:10, Philip Blundell wrote: > > > This looks like the familiar problem with mklibs not handling references > > to data symbols. I'll try to find some time to look at that again this > > afternoon. > > The reason (well, the glaring one!) is that the symbol doesn't have UND > or whatever it takes to match the symbol-finding regexp; however, I > don't understand mklibs anywhere near enough yet to know if changing > that is safe, of it it'll lead to Huge Bloat(TM).
Yes, indeed. I think the solution is for mklibs to inspect the relocs rather than the symbol table. I have a modified version that does this, but it seems to have some other problems at the moment. p. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
