------- Additional Comments From nickc at redhat dot com 2007-12-21 17:13 ------- Hi Jonathan,
Sorry - the revised patch does not work. The problem is that the new instruction patterns include specific numbering for the cache selection, but the assembler does not know which pattern to choose. For example try assembling this test case: .text cpushl %ic,(%a1) cpushl %dc,(%a1) cpushl %bc,(%a1) intouch %a1 and then disassembling the resulting object file. I get: 0: f4e9 cpushl bc,%a1@ 2: f469 cpushl dc,%a1@ 4: f4e9 cpushl bc,%a1@ 6: Address 0x0000000000000008 is out of bounds. I think that you may need to define a new opcode operator, one like 'c' but which only accepts 'ic','dc' or 'bc' and then use this. Cheers Nick PS. Incidentally it would be great if you could include a new gas testsuite test in any future patch which would check to make sure that the syntax is parsed and disassembled correctly.... -- What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |WAITING http://sourceware.org/bugzilla/show_bug.cgi?id=5457 ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is. _______________________________________________ bug-binutils mailing list bug-binutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-binutils