From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756353AbYEEATY (ORCPT ); Sun, 4 May 2008 20:19:24 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753360AbYEEATK (ORCPT ); Sun, 4 May 2008 20:19:10 -0400 Received: from ozlabs.org ([203.10.76.45]:59322 "EHLO ozlabs.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753277AbYEEATJ (ORCPT ); Sun, 4 May 2008 20:19:09 -0400 From: Rusty Russell To: Adrian Bunk Subject: Re: s390 kvm_virtio.c build error Date: Mon, 5 May 2008 10:19:01 +1000 User-Agent: KMail/1.9.9 Cc: Christian Borntraeger , Martin Schwidefsky , Carsten Otte , Avi Kivity , kvm-devel@lists.sourceforge.net, linux-s390@vger.kernel.org, linux-kernel@vger.kernel.org References: <20080503174716.GM5838@cs181133002.pp.htv.fi> In-Reply-To: <20080503174716.GM5838@cs181133002.pp.htv.fi> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200805051019.02289.rusty@rustcorp.com.au> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sunday 04 May 2008 03:47:17 Adrian Bunk wrote: > Commit c45a6816c19dee67b8f725e6646d428901a6dc24 > (virtio: explicit advertisement of driver features) > and commit e976a2b997fc4ad70ccc53acfe62811c4aaec851 > (s390: KVM guest: virtio device support, and kvm hypercalls) > don't like each other: Yep, I broke s390. This was kind of expected, but I didn't want to try to fix it as I am unable to test. It would look something like this: virtio: Attempt to fix kvm_virtio after feature management changes Very similar to lguest code: set and get feature are now separate callbacks. Signed-off-by: Rusty Russell diff -r 219d6c116996 drivers/s390/kvm/kvm_virtio.c --- a/drivers/s390/kvm/kvm_virtio.c Mon May 05 10:03:16 2008 +1000 +++ b/drivers/s390/kvm/kvm_virtio.c Mon May 05 10:17:25 2008 +1000 @@ -78,27 +78,34 @@ static unsigned desc_size(const struct k + desc->config_len; } -/* - * This tests (and acknowleges) a feature bit. - */ -static bool kvm_feature(struct virtio_device *vdev, unsigned fbit) +/* This gets the device's feature bits. */ +static u32 kvm_get_features(struct virtio_device *vdev) { + unsigned int i; + u32 features = 0; struct kvm_device_desc *desc = to_kvmdev(vdev)->desc; - u8 *features; + u8 *in_features = kvm_vq_features(desc); - if (fbit / 8 > desc->feature_len) - return false; + /* We do this the slow but generic way. */ + for (i = 0; i < min(desc->feature_len * 8, 32); i++) + if (in_features[i / 8] & (1 << (i % 8))) + features |= (1 << i); - features = kvm_vq_features(desc); - if (!(features[fbit / 8] & (1 << (fbit % 8)))) - return false; + return features; +} - /* - * We set the matching bit in the other half of the bitmap to tell the - * Host we want to use this feature. - */ - features[desc->feature_len + fbit / 8] |= (1 << (fbit % 8)); - return true; +static void kvm_set_features(struct virtio_device *vdev, u32 features) +{ + unsigned int i; + struct kvm_device_desc *desc = to_kvmdev(vdev)->desc; + /* Second half of bitmap is features we accept. */ + u8 *out_features = kvm_vq_features(desc) + desc->feature_len; + + memset(out_features, 0, desc->feature_len); + for (i = 0; i < min(desc->feature_len * 8, 32); i++) { + if (features & (1 << i)) + out_features[i / 8] |= (1 << (i % 8)); + } } /* @@ -221,7 +228,8 @@ static void kvm_del_vq(struct virtqueue * The config ops structure as defined by virtio config */ static struct virtio_config_ops kvm_vq_configspace_ops = { - .feature = kvm_feature, + .get_features = kvm_get_features, + .set_features = kvm_set_features, .get = kvm_get, .set = kvm_set, .get_status = kvm_get_status,