From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752859AbbIOPIB (ORCPT ); Tue, 15 Sep 2015 11:08:01 -0400 Received: from mx1.redhat.com ([209.132.183.28]:60168 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752417AbbIOPIA (ORCPT ); Tue, 15 Sep 2015 11:08:00 -0400 Subject: Re: [PATCH V6 6/6] kvm: add fast mmio capabilitiy To: Jason Wang , gleb@kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org References: <1442299319-5253-1-git-send-email-jasowang@redhat.com> <1442299319-5253-7-git-send-email-jasowang@redhat.com> Cc: mst@redhat.com, cornelia.huck@de.ibm.com From: Paolo Bonzini Message-ID: <55F8344B.50300@redhat.com> Date: Tue, 15 Sep 2015 17:07:55 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.2.0 MIME-Version: 1.0 In-Reply-To: <1442299319-5253-7-git-send-email-jasowang@redhat.com> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 15/09/2015 08:41, Jason Wang wrote: > +With KVM_CAP_FAST_MMIO, a zero length mmio eventfd is allowed for > +kernel to ignore the length of guest write and get a possible faster > +response. Note the speedup may only work on some specific > +architectures and setups. Otherwise, it's as fast as wildcard mmio > +eventfd. I don't really like tying the capability to MMIO, especially since zero length ioeventfd is already accepted for virtio-ccw. What about the following? diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt index 7a3cb48a644d..247944071cc8 100644 --- a/Documentation/virtual/kvm/api.txt +++ b/Documentation/virtual/kvm/api.txt @@ -1627,11 +1627,10 @@ to the registered address is equal to datamatch in struct kvm_ioeventfd. For virtio-ccw devices, addr contains the subchannel id and datamatch the virtqueue index. -With KVM_CAP_FAST_MMIO, a zero length mmio eventfd is allowed for -kernel to ignore the length of guest write and get a possible faster -response. Note the speedup may only work on some specific -architectures and setups. Otherwise, it's as fast as wildcard mmio -eventfd. +With KVM_CAP_IOEVENTFD_ANY_LENGTH, a zero length ioeventfd is allowed, and +the kernel will ignore the length of guest write and get a faster vmexit. +The speedup may only apply to specific architectures, but the ioeventfd will +work anyway. 4.60 KVM_DIRTY_TLB diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index b4f6aeaf94a6..03f3618612aa 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -830,7 +830,7 @@ struct kvm_ppc_smmu_info { #define KVM_CAP_GUEST_DEBUG_HW_BPS 119 #define KVM_CAP_GUEST_DEBUG_HW_WPS 120 #define KVM_CAP_SPLIT_IRQCHIP 121 -#define KVM_CAP_FAST_MMIO 122 +#define KVM_CAP_IOEVENTFD_ANY_LENGTH 122 #ifdef KVM_CAP_IRQ_ROUTING diff --git a/virt/kvm/eventfd.c b/virt/kvm/eventfd.c index 79db45336e3a..1dc8c45d2270 100644 --- a/virt/kvm/eventfd.c +++ b/virt/kvm/eventfd.c @@ -914,9 +914,7 @@ kvm_assign_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args) return -EINVAL; /* ioeventfd with no length can't be combined with DATAMATCH */ - if (!args->len && - args->flags & (KVM_IOEVENTFD_FLAG_PIO | - KVM_IOEVENTFD_FLAG_DATAMATCH)) + if (!args->len && (args->flags & KVM_IOEVENTFD_FLAG_DATAMATCH)) return -EINVAL; ret = kvm_assign_ioeventfd_idx(kvm, bus_idx, args); diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 0780d970d087..0b48aadedcee 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -2717,7 +2717,7 @@ static long kvm_vm_ioctl_check_extension_generic(struct kvm *kvm, long arg) case KVM_CAP_IRQFD: case KVM_CAP_IRQFD_RESAMPLE: #endif - case KVM_CAP_FAST_MMIO: + case KVM_CAP_IOEVENTFD_ANY_LENGTH: case KVM_CAP_CHECK_EXTENSION_VM: return 1; #ifdef CONFIG_HAVE_KVM_IRQ_ROUTING Paolo