Alex Williamson 166fd7d94a vfio: hugepage support for vfio_iommu_type1 ... We currently send all mappings to the iommu in PAGE_SIZE chunks, which prevents the iommu from enabling support for larger page sizes. We still need to pin pages, which means we step through them in PAGE_SIZE chunks, but we can batch up contiguous physical memory chunks to allow the iommu the opportunity to use larger pages. The approach here is a bit different that the one currently used for legacy KVM device assignment. Rather than looking at the vma page size and using that as the maximum size to pass to the iommu, we instead simply look at whether the next page is physically contiguous. This means we might ask the iommu to map a 4MB region, while legacy KVM might limit itself to a maximum of 2MB. Splitting our mapping path also allows us to be smarter about locked memory because we can more easily unwind if the user attempts to exceed the limit. Therefore, rather than assuming that a mapping will result in locked memory, we test each page as it is pinned to determine whether it locks RAM vs an mmap'd MMIO region. This should result in better locking granularity and less locked page fudge factors in userspace. The unmap path uses the same algorithm as legacy KVM. We don't want to track the pfn for each mapping ourselves, but we need the pfn in order to unpin pages. We therefore ask the iommu for the iova to physical address translation, ask it to unpin a page, and see how many pages were actually unpinned. iommus supporting large pages will often return something bigger than a page here, which we know will be physically contiguous and we can unpin a batch of pfns. iommus not supporting large mappings won't see an improvement in batching here as they only unmap a page at a time. With this change, we also make a clarification to the API for mapping and unmapping DMA. We can only guarantee unmaps at the same granularity as used for the original mapping. In other words, unmapping a subregion of a previous mapping is not guaranteed and may result in a larger or smaller unmapping than requested. The size field in the unmapping structure is updated to reflect this. Previously this was unmodified on mapping, always returning the the requested unmap size. This is now updated to return the actual unmap size on success, allowing userspace to appropriately track mappings. Signed-off-by: Alex Williamson <alex.williamson@redhat.com>		2013-06-21 09:38:02 -06:00
..
byteorder
caif	caif: Remove my bouncing email address.	2013-04-23 13:25:51 -04:00
can	can: gw: indicate and count deleted frames due to misconfiguration	2013-01-26 16:59:02 +01:00
dvb	[media] demux.h: Remove duplicated enum	2013-04-08 06:53:15 -03:00
hdlc
hsi
isdn
mmc
netfilter	netfilter: nfnetlink_queue: avoid expensive gso segmentation and checksum fixup	2013-04-29 20:09:07 +02:00
netfilter_arp
netfilter_bridge
netfilter_ipv4
netfilter_ipv6	netfilter: fix struct ip6t_frag field description	2013-04-02 12:25:57 +02:00
nfsd
raid	UAPI: fix endianness conditionals in linux/raid/md_p.h	2013-03-13 15:21:49 -07:00
spi
sunrpc
tc_act
tc_ematch
usb	USB: fix an incorrect table index in comment	2013-04-08 09:08:34 -07:00
wimax
a.out.h
acct.h	UAPI: fix endianness conditionals in linux/acct.h	2013-03-13 15:21:48 -07:00
adb.h
adfs_fs.h
affs_hardblocks.h
agpgart.h
aio_abi.h	UAPI: fix endianness conditionals in linux/aio_abi.h	2013-03-13 15:21:48 -07:00
apm_bios.h
arcfb.h
atalk.h
atm.h
atm_eni.h
atm_he.h
atm_idt77105.h
atm_nicstar.h
atm_tcp.h
atm_zatm.h
atmapi.h
atmarp.h
atmbr2684.h
atmclip.h
atmdev.h
atmioc.h
atmlec.h
atmmpc.h
atmppp.h
atmsap.h
atmsvc.h
audit.h	audit: Make testing for a valid loginuid explicit.	2013-05-07 22:27:15 -04:00
auto_fs.h	unbreak automounter support on 64-bit kernel with 32-bit userspace (v2)	2013-02-08 20:42:18 +01:00
auto_fs4.h
auxvec.h	powerpc: Add HWCAP2 aux entry	2013-04-26 16:08:16 +10:00
ax25.h
b1lli.h
baycom.h
bfs_fs.h
binfmts.h
blkpg.h
blktrace_api.h
bpqether.h
bsg.h
btrfs.h	Btrfs: rescan for qgroups	2013-05-06 15:55:19 -04:00
can.h
capability.h
capi.h
cciss_defs.h
cciss_ioctl.h
cdrom.h	libata: identify and init ZPODD devices	2013-01-21 15:40:35 -05:00
cgroupstats.h
chio.h
cm4000_cs.h
cn_proc.h	connector: Added coredumping event to the process connector	2013-03-20 13:23:21 -04:00
coda.h
coda_psdev.h
coff.h
connector.h	Drivers: hv: Add a new driver to support host initiated backup	2013-03-15 12:12:36 -07:00
const.h
cramfs_fs.h
cuda.h
cyclades.h
cycx_cfm.h
dcbnl.h
dccp.h
dlm.h
dlm_device.h
dlm_netlink.h
dlm_plock.h
dlmconstants.h
dm-ioctl.h	dm ioctl: allow message to return data	2013-03-01 22:45:49 +00:00
dm-log-userspace.h
dn.h
dqblk_xfs.h
edd.h
efs_fs_sb.h
elf-em.h
elf-fdpic.h
elf.h	metag: ptrace: Implement NT_METAG_TLS	2013-03-27 14:37:47 +00:00
elfcore.h
errno.h
errqueue.h
ethtool.h
eventpoll.h
fadvise.h
falloc.h
fanotify.h
fb.h
fcntl.h
fd.h
fdreg.h
fib_rules.h
fiemap.h
filter.h	filter: add ANC_PAY_OFFSET instruction for loading payload start offset	2013-03-20 13:15:45 -04:00
firewire-cdev.h
firewire-constants.h
flat.h
fs.h	mm: make snapshotting pages for stable writes a per-bio operation	2013-04-29 15:54:33 -07:00
fsl_hypervisor.h
fuse.h	fuse: add flag to turn on async direct IO	2013-05-01 14:37:21 +02:00
futex.h
gameport.h
gen_stats.h
genetlink.h
gfs2_ondisk.h
gigaset_dev.h
hdlc.h
hdlcdrv.h
hdreg.h
hid.h
hiddev.h
hidraw.h
hpet.h
hw_breakpoint.h
hysdn_if.h
i2c-dev.h
i2c.h
i2o-dev.h
i8k.h
icmp.h
icmpv6.h
if.h
if_addr.h
if_addrlabel.h
if_alg.h
if_arcnet.h
if_arp.h
if_bonding.h
if_bridge.h	bridge: use __u16 in if_bridge.h	2013-02-14 00:54:17 -05:00
if_cablemodem.h	if_cablemodem.h: Add parenthesis around ioctl macros	2013-05-08 13:13:30 -07:00
if_eql.h
if_ether.h	net: add ETH_P_802_3_MIN	2013-03-28 01:20:42 -04:00
if_fc.h
if_fddi.h
if_frad.h
if_hippi.h
if_infiniband.h
if_link.h	vxlan: allow choosing destination port per vxlan	2013-04-29 11:53:12 -04:00
if_ltalk.h
if_packet.h	packet: if hw/sw ts enabled in rx/tx ring, report which ts we got	2013-04-25 01:22:22 -04:00
if_phonet.h
if_plip.h
if_ppp.h
if_pppol2tp.h
if_pppox.h
if_slip.h
if_team.h
if_tun.h
if_tunnel.h
if_vlan.h	net/8021q: Implement Multiple VLAN Registration Protocol (MVRP)	2013-02-10 20:37:22 -05:00
if_x25.h
igmp.h
in.h
in6.h	mcast: define and use MRT[6]_MAX in ip[6]_mroute_opt()	2013-01-21 13:55:14 -05:00
in_route.h
inet_diag.h
inotify.h
input.h	Linux 3.9-rc3	2013-03-17 19:40:50 -07:00
ioctl.h
ip.h
ip6_tunnel.h
ip_vs.h	ipvs: fix sparse warnings for some parameters	2013-04-23 11:43:05 +09:00
ipc.h
ipmi.h	ipmi: remove superfluous kernel/userspace explanation	2013-02-27 19:10:21 -08:00
ipmi_msgdefs.h
ipsec.h
ipv6.h	ipv6: Store Router Alert option in IP6CB directly.	2013-01-13 20:17:14 -05:00
ipv6_route.h
ipx.h
irda.h
irqnr.h
isdn.h
isdn_divertif.h
isdn_ppp.h
isdnif.h
iso_fs.h
ivtv.h
ivtvfb.h
ixjuser.h
jffs2.h
joystick.h
Kbuild	Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/jesse/openvswitch	2013-04-17 13:30:32 -04:00
kd.h
kdev_t.h
kernel-page-flags.h
kernel.h
kernelcapi.h
kexec.h
keyboard.h
keyctl.h
kvm.h	kvm: Add definition of KVM_REG_MIPS	2013-06-11 11:06:34 +03:00
kvm_para.h
l2tp.h
limits.h
llc.h
loop.h
lp.h
magic.h	hostfs: move HOSTFS_SUPER_MAGIC to <linux/magic.h>	2013-05-04 15:48:44 -04:00
major.h
map_to_7segment.h
matroxfb.h
mdio.h
media.h	[media] media: add support for decoder as one of media entity types	2013-03-21 14:05:31 -03:00
mei.h
mempolicy.h
meye.h	[media] meye: convert to the control framework	2013-02-05 18:23:47 -02:00
mii.h
minix_fs.h
mman.h
mmtimer.h
module.h
mqueue.h
mroute.h	mcast: add multicast proxy support (IPv4 and IPv6)	2013-01-21 13:55:14 -05:00
mroute6.h	mcast: add multicast proxy support (IPv4 and IPv6)	2013-01-21 13:55:14 -05:00
msdos_fs.h	fat: mark fs as dirty on mount and clean on umount	2013-02-27 19:10:11 -08:00
msg.h	ipc: introduce message queue copy feature	2013-01-04 16:11:45 -08:00
mtio.h
n_r3964.h
nbd.h	nbd: support FLUSH requests	2013-02-27 19:10:22 -08:00
ncp.h
ncp_fs.h
ncp_mount.h
ncp_no.h
neighbour.h	vxlan: generalize forwarding tables	2013-03-17 12:23:46 -04:00
net.h
net_dropmon.h
net_tstamp.h
netconf.h
netdevice.h
netfilter.h
netfilter_arp.h
netfilter_bridge.h
netfilter_decnet.h
netfilter_ipv4.h
netfilter_ipv6.h
netlink.h	netlink: mmaped netlink: ring setup	2013-04-19 14:57:57 -04:00
netlink_diag.h	netlink: add RX/TX-ring support to netlink diag	2013-04-19 14:57:58 -04:00
netrom.h
nfc.h	NFC: llcp: Add support in getsockopt for RW, LTO, and MIU remote parameters	2013-04-11 16:28:59 +02:00
nfs.h
nfs2.h
nfs3.h
nfs4.h
nfs4_mount.h
nfs_fs.h
nfs_idmap.h
nfs_mount.h
nfsacl.h
nl80211.h	cfg80211: introduce critical protocol indication from user-space	2013-04-22 15:48:00 +02:00
nubus.h
nvram.h
omap3isp.h
omapfb.h
oom.h
openvswitch.h	openvswitch: Expose <linux/openvswitch.h> to userspace	2013-03-29 17:54:33 -07:00
packet_diag.h	sock_diag: allow to dump bpf filters	2013-04-29 13:21:30 -04:00
param.h
parport.h
patchkey.h
pci.h
pci_regs.h	PCI: Clean up MSI/MSI-X capability #defines	2013-04-23 09:50:30 -06:00
perf_event.h	perf: Fix comments in PERF_MEM_LVL bitmask	2013-04-08 13:54:24 +02:00
personality.h
pfkeyv2.h
pg.h
phantom.h
phonet.h
pkt_cls.h
pkt_sched.h	htb: add HTB_DIRECT_QLEN attribute	2013-03-06 15:40:53 -05:00
pktcdvd.h
pmu.h
poll.h
posix_types.h
ppdev.h
ppp-comp.h
ppp-ioctl.h
ppp_defs.h
pps.h
prctl.h
ptp_clock.h
ptrace.h	ptrace: add ability to retrieve signals without removing from a queue (v4)	2013-04-30 17:04:05 -07:00
qnx4_fs.h
qnxtypes.h
quota.h
radeonfb.h
random.h
raw.h
rds.h
reboot.h
reiserfs_fs.h
reiserfs_xattr.h
resource.h
rfkill.h	rfkill: Add NFC to the list of supported radios	2013-04-12 16:54:38 +02:00
romfs_fs.h
rose.h
route.h
rtc.h
rtnetlink.h	bridge: Dump vlan information from a bridge port	2013-02-13 19:41:46 -05:00
scc.h
sched.h
screen_info.h
sctp.h	net: sctp: introduce uapi header for sctp	2013-04-09 13:19:39 -04:00
sdla.h
seccomp.h
securebits.h
selinux_netlink.h
sem.h
serial.h
serial_core.h	tty/serial: Add support for Altera serial port	2013-03-12 08:53:23 -07:00
serial_reg.h
serio.h
shm.h
signal.h	unify SS_ONSTACK/SS_DISABLE definitions	2012-12-19 18:07:39 -05:00
signalfd.h
snmp.h	net: Add MIB counters for checksum errors	2013-04-29 15:14:03 -04:00
sock_diag.h
socket.h
sockios.h
som.h
sonet.h
sonypi.h
sound.h
soundcard.h
stat.h
stddef.h
string.h
suspend_ioctls.h
swab.h
synclink.h
sysctl.h
sysinfo.h
taskstats.h
tcp.h	tcp: Remove TCPCT	2013-03-17 14:35:13 -04:00
tcp_metrics.h
telephony.h
termios.h
time.h	timekeeping: Add CLOCK_TAI clockid	2013-03-22 16:19:59 -07:00
times.h
timex.h
tiocl.h
tipc.h
tipc_config.h
toshiba.h
tty.h
tty_flags.h
types.h
udf_fs_i.h
udp.h
uhid.h	HID: uhid: use __packed__ for uhid_feature_answer_req	2013-01-03 10:38:24 +01:00
uinput.h
uio.h
ultrasound.h
un.h
unistd.h
unix_diag.h	net: fix *_DIAG_MAX constants	2013-03-21 12:36:33 -04:00
usbdevice_fs.h
utime.h
utsname.h
uuid.h
uvcvideo.h
v4l2-common.h
v4l2-controls.h	[media] V4L: Add MATRIX option to V4L2_CID_EXPOSURE_METERING control	2013-03-31 10:55:19 -03:00
v4l2-dv-timings.h	[media] v4l2-dv-timings.h: add 480i59.94 and 576i50 CEA-861-E timings	2013-04-14 19:56:36 -03:00
v4l2-mediabus.h	[media] soc_camera: Add RGB666 & RGB888 formats	2013-04-04 19:40:08 -03:00
v4l2-subdev.h
veth.h
vfio.h	vfio: hugepage support for vfio_iommu_type1	2013-06-21 09:38:02 -06:00
vhost.h	tcm_vhost: header split up	2013-05-02 13:40:15 +03:00
videodev2.h	[media] videodev2.h: Remove the unused old V4L1 buffer types	2013-04-25 10:29:14 -03:00
virtio_9p.h
virtio_balloon.h	virtio: do not export "u16" and "u64" to userspace	2013-04-02 16:42:58 +10:30
virtio_blk.h
virtio_config.h
virtio_console.h	virtio_console: fix uapi header	2013-05-20 10:25:28 +09:30
virtio_ids.h	caif_virtio: Introduce caif over virtio	2013-03-20 14:06:06 +10:30
virtio_net.h	virtio: don't expose u16 in userspace api	2013-05-05 16:58:01 -04:00
virtio_pci.h
virtio_ring.h
virtio_rng.h
vm_sockets.h	VSOCK: Split vm_sockets.h into kernel/uapi	2013-03-08 12:24:48 -05:00
vt.h
wait.h
wanrouter.h	wanrouter: delete now orphaned header content, files/drivers	2013-01-31 19:56:35 -05:00
watchdog.h
wimax.h
wireless.h
x25.h
xattr.h	hfsplus: add osx.* prefix for handling namespace of Mac OS X extended attributes	2013-02-27 19:10:10 -08:00
xfrm.h	xfrm: allow to avoid copying DSCP during encapsulation	2013-03-06 07:02:45 +01:00