From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753759AbZELTFc (ORCPT ); Tue, 12 May 2009 15:05:32 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751377AbZELTFP (ORCPT ); Tue, 12 May 2009 15:05:15 -0400 Received: from victor.provo.novell.com ([137.65.250.26]:57996 "EHLO victor.provo.novell.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751128AbZELTFN (ORCPT ); Tue, 12 May 2009 15:05:13 -0400 Message-ID: <4A09C866.1080309@novell.com> Date: Tue, 12 May 2009 15:05:10 -0400 From: Gregory Haskins User-Agent: Thunderbird 2.0.0.21 (Macintosh/20090302) MIME-Version: 1.0 CC: kvm@vger.kernel.org, viro@ZenIV.linux.org.uk, linux-kernel@vger.kernel.org, avi@redhat.com, davidel@xmailserver.org Subject: Re: [KVM PATCH v7 3/3] kvm: add iofd support References: <20090512181134.26131.10023.stgit@dev.haskins.net> <20090512182701.26131.66801.stgit@dev.haskins.net> In-Reply-To: <20090512182701.26131.66801.stgit@dev.haskins.net> X-Enigmail-Version: 0.95.7 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="------------enig043DD2E2AC5D2F7F9749759F" To: unlisted-recipients:; (no To-header on input) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enig043DD2E2AC5D2F7F9749759F Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Gregory Haskins wrote: > iofd is a mechanism to register PIO/MMIO regions to trigger an eventfd > signal when written to. Userspace can register any arbitrary address > with a corresponding eventfd. > =20 Ugg..this patch header sucks, especially given all the talk around how we need to do them better lately :) I will add this text as well for future versions: -------------- Traditional MMIO/PIO exit paths are expensive because they are done within the same context as the VCPU thread and therefore cause a VMX/SVM "heavy-weight" exit, a transition back to userspace, and overhead with the qemu processing of the operation. An eventfd mechanism, on the other hand, allows the VCPU to take a very brief lightweight exit only long enough to trigger the eventfd_signal. This means that clients of the eventfd (supporting both userspace or kernel end-points) can potentially get notified much more efficiently than if we were to register through the traditional mechanism via qemu MMIO/PIO notification= =2E ----------- > Signed-off-by: Gregory Haskins > --- > > include/linux/kvm.h | 12 +++++ > include/linux/kvm_host.h | 2 + > virt/kvm/eventfd.c | 107 ++++++++++++++++++++++++++++++++++++++= ++++++++ > virt/kvm/kvm_main.c | 13 ++++++ > 4 files changed, 134 insertions(+), 0 deletions(-) > > diff --git a/include/linux/kvm.h b/include/linux/kvm.h > index dfc4bcc..99b6e45 100644 > --- a/include/linux/kvm.h > +++ b/include/linux/kvm.h > @@ -292,6 +292,17 @@ struct kvm_guest_debug { > struct kvm_guest_debug_arch arch; > }; > =20 > +#define KVM_IOFD_FLAG_DEASSIGN (1 << 0) > +#define KVM_IOFD_FLAG_PIO (1 << 1) > + > +struct kvm_iofd { > + __u64 addr; > + __u32 len; > + __u32 fd; > + __u32 flags; > + __u8 pad[12]; > +}; > + > #define KVM_TRC_SHIFT 16 > /* > * kvm trace categories > @@ -508,6 +519,7 @@ struct kvm_irqfd { > #define KVM_DEASSIGN_DEV_IRQ _IOW(KVMIO, 0x75, struct kvm_assign= ed_irq) > #define KVM_ASSIGN_IRQFD _IOW(KVMIO, 0x76, struct kvm_irqfd)= > #define KVM_DEASSIGN_IRQFD _IOW(KVMIO, 0x77, __u32) > +#define KVM_IOFD _IOW(KVMIO, 0x78, struct kvm_iofd) > =20 > /* > * ioctls for vcpu fds > diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h > index 1acc528..d53cb70 100644 > --- a/include/linux/kvm_host.h > +++ b/include/linux/kvm_host.h > @@ -529,5 +529,7 @@ static inline void kvm_free_irq_routing(struct kvm = *kvm) {} > int kvm_assign_irqfd(struct kvm *kvm, int fd, int gsi, int flags); > int kvm_deassign_irqfd(struct kvm *kvm, int fd); > void kvm_irqfd_release(struct kvm *kvm); > +int kvm_iofd(struct kvm *kvm, unsigned long addr, size_t len, > + int fd, int flags); > =20 > #endif > diff --git a/virt/kvm/eventfd.c b/virt/kvm/eventfd.c > index 71afd62..8b23317 100644 > --- a/virt/kvm/eventfd.c > +++ b/virt/kvm/eventfd.c > @@ -21,12 +21,16 @@ > */ > =20 > #include > +#include > #include > #include > #include > #include > #include > #include > +#include > + > +#include "iodev.h" > =20 > /* > * -------------------------------------------------------------------= - > @@ -185,3 +189,106 @@ kvm_irqfd_release(struct kvm *kvm) > list_for_each_entry_safe(irqfd, tmp, &kvm->irqfds, list) > irqfd_release(irqfd); > } > + > +/* > + * -------------------------------------------------------------------= - > + * iofd: translate a PIO/MMIO memory write to an eventfd signal. > + * > + * userspace can register a PIO/MMIO address with an eventfd for recie= ving > + * notification when the memory has been touched. > + * -------------------------------------------------------------------= - > + */ > + > +struct _iofd { > + u64 addr; > + size_t length; > + struct file *file; > + struct kvm_io_device dev; > +}; > + > +static int > +iofd_in_range(struct kvm_io_device *this, gpa_t addr, int len, int is_= write) > +{ > + struct _iofd *iofd =3D (struct _iofd *)this->private; > + > + return ((addr >=3D iofd->addr && (addr < iofd->addr + iofd->length)))= ; > +} > + > +/* writes trigger an event */ > +static void > +iofd_write(struct kvm_io_device *this, gpa_t addr, int len, const void= *val) > +{ > + struct _iofd *iofd =3D (struct _iofd *)this->private; > + > + eventfd_signal(iofd->file, 1); > +} > + > +/* reads return all zeros */ > +static void > +iofd_read(struct kvm_io_device *this, gpa_t addr, int len, void *val) > +{ > + memset(val, 0, len); > +} > + > +static void > +iofd_destructor(struct kvm_io_device *this) > +{ > + struct _iofd *iofd =3D (struct _iofd *)this->private; > + > + fput(iofd->file); > + kfree(iofd); > +} > + > +static int > +kvm_assign_iofd(struct kvm *kvm, unsigned long addr, size_t len, > + int fd, int flags) > +{ > + int pio =3D flags & KVM_IOFD_FLAG_PIO; > + struct kvm_io_bus *bus =3D pio ? &kvm->pio_bus : &kvm->mmio_bus; > + struct _iofd *iofd; > + struct file *file; > + > + file =3D eventfd_fget(fd); > + if (IS_ERR(file)) > + return PTR_ERR(file); > + > + iofd =3D kzalloc(sizeof(*iofd), GFP_KERNEL); > + if (!iofd) { > + fput(file); > + return -ENOMEM; > + } > + > + iofd->dev.read =3D iofd_read; > + iofd->dev.write =3D iofd_write; > + iofd->dev.in_range =3D iofd_in_range; > + iofd->dev.destructor =3D iofd_destructor; > + iofd->dev.private =3D iofd; > + > + iofd->addr =3D addr; > + iofd->length =3D len; > + iofd->file =3D file; > + > + kvm_io_bus_register_dev(bus, &iofd->dev); > + > + printk(KERN_DEBUG "registering %s iofd at %lx of size %d\n", > + pio ? "PIO" : "MMIO", addr, (int)len); > + > + return 0; > +} > + > +static int > +kvm_deassign_iofd(struct kvm *kvm, unsigned long addr, size_t len, > + int fd, int flags) > +{ > + /* FIXME: We need an io_bus_unregister() function */ > + return -EINVAL; > +} > + > +int > +kvm_iofd(struct kvm *kvm, unsigned long addr, size_t len, int fd, int = flags) > +{ > + if (flags & KVM_IOFD_FLAG_DEASSIGN) > + return kvm_deassign_iofd(kvm, addr, len, fd, flags); > + > + return kvm_assign_iofd(kvm, addr, len, fd, flags); > +} > diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c > index 7aa9f0a..a443974 100644 > --- a/virt/kvm/kvm_main.c > +++ b/virt/kvm/kvm_main.c > @@ -2228,6 +2228,19 @@ static long kvm_vm_ioctl(struct file *filp, > r =3D kvm_deassign_irqfd(kvm, data); > break; > } > + case KVM_IOFD: { > + struct kvm_iofd entry; > + > + r =3D -EFAULT; > + if (copy_from_user(&entry, argp, sizeof entry)) > + goto out; > + > + r =3D kvm_iofd(kvm, entry.addr, entry.len, entry.fd, > + entry.flags); > + if (r) > + goto out; > + break; > + } > default: > r =3D kvm_arch_vm_ioctl(filp, ioctl, arg); > } > > -- > To unsubscribe from this list: send the line "unsubscribe kvm" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > =20 --------------enig043DD2E2AC5D2F7F9749759F Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.11 (Darwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoJyGYACgkQlOSOBdgZUxmOdwCeOieHWQ5pm5PRcwZLWwfTUqB4 dM8AnjPhrJfvGmBbonxHH/gmx5rozO7E =e6Ou -----END PGP SIGNATURE----- --------------enig043DD2E2AC5D2F7F9749759F--