From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1MDZ4Z-0001UE-Gi for qemu-devel@nongnu.org; Mon, 08 Jun 2009 03:15:11 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1MDZ4U-0001Lo-Vg for qemu-devel@nongnu.org; Mon, 08 Jun 2009 03:15:10 -0400 Received: from [199.232.76.173] (port=42871 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1MDZ4U-0001La-O6 for qemu-devel@nongnu.org; Mon, 08 Jun 2009 03:15:06 -0400 Received: from mx20.gnu.org ([199.232.41.8]:48153) by monty-python.gnu.org with esmtps (TLS-1.0:RSA_AES_256_CBC_SHA1:32) (Exim 4.60) (envelope-from ) id 1MDZ4T-0008WA-T9 for qemu-devel@nongnu.org; Mon, 08 Jun 2009 03:15:06 -0400 Received: from mx2.redhat.com ([66.187.237.31]) by mx20.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1MDZ4S-0000zJ-M8 for qemu-devel@nongnu.org; Mon, 08 Jun 2009 03:15:04 -0400 Message-ID: <4A2CBA70.3050007@redhat.com> Date: Mon, 08 Jun 2009 09:14:56 +0200 From: Gerd Hoffmann MIME-Version: 1.0 Subject: Re: [Qemu-devel] [PATCH] xen: net backend doesn't need linux headers. References: <1243926324-13133-1-git-send-email-kraxel@redhat.com> <28EBEC28-6D03-4321-8629-1F54EA30C634@web.de> <1DB8C28C-ECE6-459C-B9BE-D4747CA4C9CF@web.de> In-Reply-To: Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Blue Swirl Cc: =?UTF-8?B?QW5kcmVhcyBGw6RyYmVy?= , xen-devel@lists.xensource.com, xen-discuss@opensolaris.org, qemu-devel@nongnu.org On 06/07/09 08:56, Blue Swirl wrote: >> - tmpbuf = malloc(PAGE_SIZE); >> + tmpbuf = malloc(XC_PAGE_SIZE); > It also looks like there is no real need for using host or target page > size, so we could use for example MIN(4096, txreq.size). No. XC_PAGE_SIZE is the max buffer size. As the buffer allocated here could be resused for multiple packets you'll have to use that, otherwise you'll overflow the buffer in case the second packet is larger than the first one. >> If then goes on to compile Xen stuff into sparc-softmmu and fails: Hmm, it shouldn't try that in the first place. Guess this is the result of this commit: http://git.savannah.gnu.org/cgit/qemu.git/commit/?id=2567f5796c38238d6f2055f074e347a0338140ce >> xc_gnttab_open >> collect2: ld returned 1 exit status >> Same errors when linking i386-softmmu. For i386-softmmu it should work. > configure should probe for xc_gnttab symbols and disable Xen if these > are not found. Strange. configure checks for xc_interface_open. Both xc_interface_open and xc_gnttab_open are in the same shared library. So it should work in theory, dunno what the problem here is. Hmm, the linker complains about xc_gnttab_* only, not about the other xc_* functions missing. Maybe solaris simply has no support for them? cheers, Gerd