From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from [140.186.70.92] (port=45330 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1PjbB8-0006T9-Ra for qemu-devel@nongnu.org; Sun, 30 Jan 2011 12:35:11 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1PjbB7-00021F-Ak for qemu-devel@nongnu.org; Sun, 30 Jan 2011 12:35:10 -0500 Received: from mx1.redhat.com ([209.132.183.28]:5161) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1PjbB7-00020w-3k for qemu-devel@nongnu.org; Sun, 30 Jan 2011 12:35:09 -0500 Date: Sun, 30 Jan 2011 19:35:01 +0200 From: Alon Levy Subject: Re: [Qemu-devel] [PATCH 2/7] ccid: add passthru card device Message-ID: <20110130173500.GC7893@playa.tlv.redhat.com> References: <1294735359-4009-1-git-send-email-alevy@redhat.com> <1294735359-4009-3-git-send-email-alevy@redhat.com> <4D3EDB7C.3050604@codemonkey.ws> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4D3EDB7C.3050604@codemonkey.ws> List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Anthony Liguori Cc: qemu-devel@nongnu.org On Tue, Jan 25, 2011 at 08:17:32AM -0600, Anthony Liguori wrote: > On 01/11/2011 02:42 AM, Alon Levy wrote: > >diff --git a/libcacard/vscard_common.h b/libcacard/vscard_common.h > >new file mode 100644 > >index 0000000..9ff1295 > >--- /dev/null > >+++ b/libcacard/vscard_common.h > > This file (and the .c file) need a coding style pass to fixup > comments and the use of _ as a prefix but I want to focus on the > protocol itself. > > First, let's get a written spec into the wiki. I think it's > important that all of our compatibility protocols are documented in > a more formal way such that can be reviewed by a wider audience. http://wiki.qeum.org/Features/Smartcard I'm still working on the rest, but you can review and comment on it. I've done a number of changes from the submitted here. I guess the idea is that iterations on the wiki can be faster? The changes done to the protocol: Removed Reconnect - doesn't scale easily, the same work should be done by whomever is initiating the migration, or via other mechanisms (i.e. spice) Added Flush/FlushComplete - still need to be able to tell client to wrap up the outstanding operations in any way. I'm planning on implementing this using register_savevm_live. Fixes suggested by you - set the enum, removed _ from surrounding #ifdef (btw - why does no one use #pragma once? IIUC it's supported by gcc?) The major issue I haven't tackled yet is the thread removal in ccid-card-emulated.c is that a blocker for integration? Can it be tackled later? Alon [snip]