Soheil Hassas Yeganeh b75eba76d3 tcp: send in-queue bytes in cmsg upon read ... Applications with many concurrent connections, high variance in receive queue length and tight memory bounds cannot allocate worst-case buffer size to drain sockets. Knowing the size of receive queue length, applications can optimize how they allocate buffers to read from the socket. The number of bytes pending on the socket is directly available through ioctl(FIONREAD/SIOCINQ) and can be approximated using getsockopt(MEMINFO) (rmem_alloc includes skb overheads in addition to application data). But, both of these options add an extra syscall per recvmsg. Moreover, ioctl(FIONREAD/SIOCINQ) takes the socket lock. Add the TCP_INQ socket option to TCP. When this socket option is set, recvmsg() relays the number of bytes available on the socket for reading to the application via the TCP_CM_INQ control message. Calculate the number of bytes after releasing the socket lock to include the processed backlog, if any. To avoid an extra branch in the hot path of recvmsg() for this new control message, move all cmsg processing inside an existing branch for processing receive timestamps. Since the socket lock is not held when calculating the size of receive queue, TCP_INQ is a hint. For example, it can overestimate the queue size by one byte, if FIN is received. With this method, applications can start reading from the socket using a small buffer, and then use larger buffers based on the remaining data when needed. V3 change-log: As suggested by David Miller, added loads with barrier to check whether we have multiple threads calling recvmsg in parallel. When that happens we lock the socket to calculate inq. V4 change-log: Removed inline from a static function. Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com> Signed-off-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: Willem de Bruijn <willemb@google.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Neal Cardwell <ncardwell@google.com> Suggested-by: David Miller <davem@davemloft.net> Signed-off-by: David S. Miller <davem@davemloft.net>		2018-05-01 18:56:29 -04:00
..
android
byteorder
caif
can
cifs
dvb
genwqe
hdlc
hsi
iio
isdn
mmc
netfilter	netfilter: ctnetlink: synproxy support	2018-03-20 14:39:31 +01:00
netfilter_arp
netfilter_bridge	netfilter: ebtables: Add support for specifying match revision	2018-03-30 11:03:39 +02:00
netfilter_ipv4
netfilter_ipv6
nfsd
raid
sched
spi
sunrpc
tc_act
tc_ematch
usb	ALSA: usb: initial USB Audio Device Class 3.0 support	2018-03-21 11:46:33 +01:00
wimax
a.out.h
acct.h
adb.h
adfs_fs.h
affs_hardblocks.h
agpgart.h
aio_abi.h
am437x-vpfe.h
apm_bios.h
arcfb.h
arm_sdei.h
aspeed-lpc-ctrl.h
atalk.h
atm_eni.h
atm_he.h
atm_idt77105.h
atm_nicstar.h
atm_tcp.h
atm_zatm.h
atm.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
auto_dev-ioctl.h
auto_fs4.h
auto_fs.h
auxvec.h
ax25.h
b1lli.h
batadv_packet.h
batman_adv.h	batman-adv: add multicast flags netlink support	2018-03-14 10:15:34 +01:00
baycom.h
bcache.h
bcm933xx_hcs.h
bfs_fs.h
binfmts.h
blkpg.h
blktrace_api.h	blktrace: fix comment in blktrace_api.h	2018-03-30 14:16:24 -06:00
blkzoned.h
bpf_common.h
bpf_perf_event.h	bpf: add support to read sample address in bpf program	2018-03-08 02:22:34 +01:00
bpf.h	bpf: add documentation for eBPF helpers (65-66)	2018-04-27 00:21:59 +02:00
bpqether.h
bsg.h
bt-bmc.h
btf.h	bpf: btf: Clean up btf.h in uapi	2018-04-23 11:32:01 +02:00
btrfs_tree.h
btrfs.h
can.h
capability.h
capi.h
cciss_defs.h
cciss_ioctl.h
cdrom.h
cec-funcs.h
cec.h
cgroupstats.h
chio.h
cm4000_cs.h
cn_proc.h	connector: add parent pid and tgid to coredump and exit events	2018-05-01 14:25:37 -04:00
coda_psdev.h
coda.h
coff.h
connector.h
const.h	linux/const.h: refactor _BITUL and _BITULL a bit	2018-04-11 10:28:38 -07:00
coresight-stm.h
cramfs_fs.h
cryptouser.h
cuda.h
cyclades.h
cycx_cfm.h
dcbnl.h
dccp.h
devlink.h
dlm_device.h
dlm_netlink.h
dlm_plock.h
dlm.h
dlmconstants.h
dm-ioctl.h	dm: hold DM table for duration of ioctl rather than use blkdev_get	2018-04-04 12:12:38 -04:00
dm-log-userspace.h
dma-buf.h
dn.h
dqblk_xfs.h
edd.h
efs_fs_sb.h
elf-em.h
elf-fdpic.h
elf.h
elfcore.h
errno.h
errqueue.h
erspan.h
ethtool.h	ethtool: Add support for configuring PFC stall prevention in ethtool	2018-03-26 13:46:46 -07:00
eventpoll.h
fadvise.h
falloc.h
fanotify.h
fb.h
fcntl.h
fd.h
fdreg.h
fib_rules.h
fiemap.h
filter.h
firewire-cdev.h
firewire-constants.h
flat.h
fou.h
fs.h
fsl_hypervisor.h
fsmap.h
fuse.h
futex.h
gameport.h
gen_stats.h
genetlink.h
gfs2_ondisk.h
gigaset_dev.h
gpio.h
gsmmux.h
gtp.h
hash_info.h
hdlc.h
hdlcdrv.h
hdreg.h
hid.h
hiddev.h
hidraw.h
hpet.h
hsr_netlink.h
hw_breakpoint.h
hyperv.h
hysdn_if.h
i2c-dev.h
i2c.h
i2o-dev.h
i8k.h
icmp.h
icmpv6.h
if_addr.h
if_addrlabel.h
if_alg.h
if_arcnet.h
if_arp.h
if_bonding.h
if_bridge.h
if_cablemodem.h
if_eql.h
if_ether.h	Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net	2018-03-23 11:31:58 -04:00
if_fc.h
if_fddi.h
if_frad.h
if_hippi.h
if_infiniband.h
if_link.h	vxlan: add ttl inherit support	2018-04-17 13:53:13 -04:00
if_ltalk.h
if_macsec.h
if_packet.h
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
if_x25.h
if.h
ife.h
igmp.h
ila.h
in6.h
in_route.h
in.h
inet_diag.h
inotify.h
input-event-codes.h
input.h
ioctl.h
ip6_tunnel.h
ip_vs.h
ip.h
ipc.h
ipmi_bmc.h
ipmi_msgdefs.h
ipmi.h
ipsec.h
ipv6_route.h
ipv6.h
ipx.h
irqnr.h
isdn_divertif.h
isdn_ppp.h
isdn.h
isdnif.h
iso_fs.h
ivtv.h
ivtvfb.h
jffs2.h
joystick.h
Kbuild
kcm.h
kcmp.h
kcov.h
kd.h
kdev_t.h
kernel-page-flags.h
kernel.h
kernelcapi.h
kexec.h
keyboard.h
keyctl.h
kfd_ioctl.h	drm/amdkfd: Add ioctls for GPUVM memory management	2018-03-15 17:27:51 -04:00
kvm_para.h
kvm.h	KVM: X86: Provide a capability to disable MWAIT intercepts	2018-03-16 22:03:51 +01:00
l2tp.h
libc-compat.h
lightnvm.h
limits.h
lirc.h	media updates for v4.17-rc1	2018-04-03 17:16:59 -07:00
llc.h
loop.h
lp.h
lwtunnel.h
magic.h
major.h
map_to_7segment.h
matroxfb.h
max2175.h
mdio.h
media-bus-format.h
media.h
mei.h
membarrier.h
memfd.h
mempolicy.h
meye.h
mic_common.h
mic_ioctl.h
mii.h
minix_fs.h
mman.h
mmtimer.h
module.h
mpls_iptunnel.h
mpls.h
mqueue.h
mroute6.h
mroute.h
msdos_fs.h	block: Move SECTOR_SIZE and SECTOR_SHIFT definitions into <linux/blkdev.h>	2018-03-17 14:45:23 -06:00
msg.h	ipc/msg: introduce msgctl(MSG_STAT_ANY)	2018-04-11 10:28:37 -07:00
mtio.h
n_r3964.h
nbd-netlink.h
nbd.h
ncp_fs.h
ncp_mount.h
ncp_no.h
ncp.h
ncsi.h	net/ncsi: Add generic netlink family	2018-03-05 10:43:37 -05:00
ndctl.h
neighbour.h
net_dropmon.h
net_namespace.h
net_tstamp.h
net.h
netconf.h
netdevice.h
netfilter_arp.h
netfilter_bridge.h
netfilter_decnet.h
netfilter_ipv4.h
netfilter_ipv6.h
netfilter.h
netlink_diag.h
netlink.h
netrom.h
nfc.h
nfs2.h
nfs3.h
nfs4_mount.h
nfs4.h
nfs_fs.h
nfs_idmap.h
nfs_mount.h
nfs.h
nfsacl.h
nilfs2_api.h
nilfs2_ondisk.h
nl80211.h	nl80211: Add CONTROL_PORT_OVER_NL80211 attribute	2018-03-29 13:45:04 +02:00
nsfs.h
nubus.h
nvme_ioctl.h
nvram.h
omap3isp.h
omapfb.h
oom.h
openvswitch.h
packet_diag.h
param.h
parport.h
patchkey.h
pci_regs.h	PCI: Add two more values for PCIe Max_Read_Request_Size	2018-04-16 18:55:04 -04:00
pci.h
pcitest.h
perf_event.h	perf/core: Store context switch out type in PERF_RECORD_SWITCH[_CPU_WIDE]	2018-04-17 09:47:39 -03:00
personality.h
pfkeyv2.h
pg.h
phantom.h
phonet.h
pkt_cls.h	net/sched: cls_flower: Add support to handle first frag as match field	2018-03-08 12:16:29 -05:00
pkt_sched.h
pktcdvd.h
pmu.h
poll.h
posix_acl_xattr.h
posix_acl.h
posix_types.h
ppdev.h
ppp_defs.h
ppp-comp.h
ppp-ioctl.h
pps.h
pr.h
prctl.h
psample.h
psci.h
psp-sev.h
ptp_clock.h
ptrace.h
qemu_fw_cfg.h	fw_cfg: write vmcoreinfo details	2018-03-20 03:17:41 +02:00
qnx4_fs.h
qnxtypes.h
qrtr.h
quota.h
radeonfb.h
random.h	random: add new ioctl RNDRESEEDCRNG	2018-04-14 11:59:31 -04:00
raw.h
rds.h
reboot.h
reiserfs_fs.h
reiserfs_xattr.h
resource.h
rfkill.h
rio_cm_cdev.h
rio_mport_cdev.h
romfs_fs.h
rose.h
route.h
rpmsg.h
rtc.h
rtnetlink.h
rxrpc.h
scc.h
sched.h
scif_ioctl.h
screen_info.h
sctp.h	selinux/stable-4.17 PR 20180403	2018-04-06 15:39:26 -07:00
sdla.h
seccomp.h
securebits.h
sed-opal.h
seg6_genl.h
seg6_hmac.h
seg6_iptunnel.h
seg6_local.h
seg6.h
selinux_netlink.h
sem.h	ipc/sem: introduce semctl(SEM_STAT_ANY)	2018-04-11 10:28:37 -07:00
serial_core.h	serial: 8250: Add Nuvoton NPCM UART	2018-03-09 11:01:19 -08:00
serial_reg.h
serial.h
serio.h
shm.h	ipc/shm: introduce shmctl(SHM_STAT_ANY)	2018-04-11 10:28:37 -07:00
signal.h
signalfd.h
smc_diag.h
smc.h
smiapp.h
snmp.h	tcp: export packets delivery info	2018-04-19 13:05:16 -04:00
sock_diag.h
socket.h
sockios.h
sonet.h
sonypi.h
sound.h
soundcard.h
stat.h
stddef.h
stm.h	stm class: Make dummy's master/channel ranges configurable	2018-03-28 18:47:18 +03:00
string.h
suspend_ioctls.h
swab.h
switchtec_ioctl.h
sync_file.h
synclink.h
sysctl.h
sysinfo.h
target_core_user.h
taskstats.h
tcp_metrics.h
tcp.h	tcp: send in-queue bytes in cmsg upon read	2018-05-01 18:56:29 -04:00
tee.h
termios.h
thermal.h
time.h	y2038: Introduce struct __kernel_old_timeval	2018-03-19 15:23:03 +01:00
timerfd.h
times.h
timex.h
tiocl.h
tipc_config.h	tipc: set default MTU for UDP media	2018-04-20 11:04:05 -04:00
tipc_netlink.h	tipc: implement configuration of UDP media MTU	2018-04-20 11:04:05 -04:00
tipc_sockets_diag.h	tipc: implement socket diagnostics for AF_TIPC	2018-03-22 14:43:35 -04:00
tipc.h	tipc: introduce ioctl for fetching node identity	2018-04-27 11:05:41 -04:00
tls.h	tls: RX path for ktls	2018-03-23 12:25:54 -04:00
toshiba.h
tty_flags.h
tty.h
types.h
udf_fs_i.h
udp.h	udp: generate gso with UDP_SEGMENT	2018-04-26 15:08:04 -04:00
uhid.h
uinput.h
uio.h
uleds.h
ultrasound.h
un.h
unistd.h
unix_diag.h
usbdevice_fs.h
usbip.h
userfaultfd.h
userio.h
utime.h
utsname.h
uuid.h
uvcvideo.h
v4l2-common.h
v4l2-controls.h	media: v4l2: Add v4l2 control IDs for HEVC encoder	2018-03-22 06:32:15 -04:00
v4l2-dv-timings.h
v4l2-mediabus.h
v4l2-subdev.h
vbox_err.h
vbox_vmmdev_types.h
vboxguest.h
veth.h
vfio_ccw.h
vfio.h	vfio/pci: Add ioeventfd support	2018-03-26 13:22:58 -06:00
vhost.h
videodev2.h	media: videodev2.h: Add v4l2 definition for HEVC	2018-03-22 06:26:05 -04:00
virtio_9p.h
virtio_balloon.h	virtio_balloon: export hugetlb page allocation counts	2018-04-10 21:23:55 +03:00
virtio_blk.h
virtio_config.h
virtio_console.h
virtio_crypto.h
virtio_gpu.h
virtio_ids.h
virtio_input.h
virtio_mmio.h
virtio_net.h
virtio_pci.h
virtio_ring.h
virtio_rng.h
virtio_scsi.h
virtio_types.h
virtio_vsock.h
vm_sockets_diag.h
vm_sockets.h
vsockmon.h
vt.h
vtpm_proxy.h
wait.h
wanrouter.h
watchdog.h
wimax.h
wireless.h
wmi.h
x25.h
xattr.h
xfrm.h
xilinx-v4l2-controls.h
zorro_ids.h
zorro.h