From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from [140.186.70.92] (port=32801 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1OHvTX-0004Fp-Is for qemu-devel@nongnu.org; Fri, 28 May 2010 05:03:32 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69) (envelope-from ) id 1OHvTW-0005FD-FO for qemu-devel@nongnu.org; Fri, 28 May 2010 05:03:31 -0400 Received: from mail-fx0-f45.google.com ([209.85.161.45]:37293) by eggs.gnu.org with esmtp (Exim 4.69) (envelope-from ) id 1OHvTW-0005Ey-A6 for qemu-devel@nongnu.org; Fri, 28 May 2010 05:03:30 -0400 Received: by fxm17 with SMTP id 17so779344fxm.4 for ; Fri, 28 May 2010 02:03:29 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <4BFF8223.5040701@siriusit.co.uk> References: <20100525121143.19347.27207.stgit@skyserv> <4BFEA489.9050307@siriusit.co.uk> <4BFF8223.5040701@siriusit.co.uk> Date: Fri, 28 May 2010 13:03:28 +0400 Message-ID: From: Igor Kovalenko Content-Type: text/plain; charset=ISO-8859-1 Subject: [Qemu-devel] Re: [OpenBIOS] [PATCH 0/3] sparc64 cleanups v1 List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Mark Cave-Ayland Cc: The OpenBIOS Mailinglist , qemu-devel On Fri, May 28, 2010 at 12:43 PM, Mark Cave-Ayland wrote: > Igor Kovalenko wrote: > >> With many debian iso images I consistently get scrolling blanks after >> the following line on qemu video console: >> >> io sched cfq registered (default) >> >> Please share your qemu command line, and installer prompt input if any. > > Yeah, I do too. I originally thought that the Debian kernel was broken, but > if you leave it long enough then it does proceed to the end. My guess is > that there is a bug in the OpenBIOS console which is obscuring the output. I allowed it to scroll and now I can confirm it would crash with tl=5. Last insn is ldda with ASI=0x24 (Nucleus quad LDD 128 bit atomic) IN: 0x0000000000424d18: ldda [ %g1 ] (36), %g4 0x0000000000424d1c: cmp %g4, %g6 0x0000000000424d20: bne,pn %xcc, 0x4076c0 0x0000000000424d24: mov 2, %g3 qemu: fatal: Trap 0x0068 while trap level (5) >= MAXTL (5), Error state pc: 0000000000424d18 npc: 0000000000424d1c General Registers: %g0-3: 0000000000000000 0000000008000000 0000000000004000 0000000000000002 %g4-7: 00000000000003ff 0000000000000001 0000000000000020 0000000000004000 Well, not sure how it worked before. Related code in helper_ld_asi() states: case 0x24: // Nucleus quad LDD 128 bit atomic case 0x2c: // Nucleus quad LDD 128 bit atomic LE // Only ldda allowed raise_exception(TT_ILL_INSN); return 0; HelenOS also uses quad ldd when configured with TSB for memory management. I guess that atomic quad ldd deserves to be implemented. I'll see if it is not hard. -- Kind regards, Igor V. Kovalenko