Date: Sun, 2 Nov 2003 15:05:47 -0800 (PST) From: Juli Mallett <jmallett@FreeBSD.org> To: Perforce Change Reviews <perforce@freebsd.org> Subject: PERFORCE change 41142 for review Message-ID: <200311022305.hA2N5lUm042514@repoman.freebsd.org>
index | next in thread | raw e-mail
http://perforce.freebsd.org/chv.cgi?CH=41142 Change 41142 by jmallett@jmallett_dalek on 2003/11/02 15:04:49 If we run across a 32-bit TLB refill, then we're confused, but better to go right into trap than to call panic. Affected files ... .. //depot/projects/mips/sys/mips/mips/exception.S#6 edit Differences ... ==== //depot/projects/mips/sys/mips/mips/exception.S#6 (text+ko) ==== @@ -141,15 +141,27 @@ LEAF(TLBMissVector) .set noat - dla a0, 1f - dla k0, panic - dmtc0 k0, MIPS_COP_0_EXC_PC + dsubu sp, sp, TF_SIZE + dla k0, 1f + j exception_save_registers + move k1, sp +1: + /* + * No turning back, and nothing we can do. Just call into + * trap and let it tell the user lovely things about how bad + * their cache has been, + */ + mfc0 a1, MIPS_COP_0_CAUSE + dmfc0 a2, MIPS_COP_0_BAD_VADDR + jal trap + move a0, k1 + + jal exception_restore_registers + move a0, sp + daddu sp, sp, TF_SIZE eret .set at VEND(TLBMissVector) - .data -1: .asciiz "32-bit TLB miss!?\n" - .text /* * XXX kernel only. For now that makes sense.help
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200311022305.hA2N5lUm042514>