kernel_optimize_test/Documentation
Wang YanQing 03f5781be2 bpf, x86_32: add eBPF JIT compiler for ia32
The JIT compiler emits ia32 bit instructions. Currently, It supports eBPF
only. Classic BPF is supported because of the conversion by BPF core.

Almost all instructions from eBPF ISA supported except the following:
BPF_ALU64 | BPF_DIV | BPF_K
BPF_ALU64 | BPF_DIV | BPF_X
BPF_ALU64 | BPF_MOD | BPF_K
BPF_ALU64 | BPF_MOD | BPF_X
BPF_STX | BPF_XADD | BPF_W
BPF_STX | BPF_XADD | BPF_DW

It doesn't support BPF_JMP|BPF_CALL with BPF_PSEUDO_CALL at the moment.

IA32 has few general purpose registers, EAX|EDX|ECX|EBX|ESI|EDI. I use
EAX|EDX|ECX|EBX as temporary registers to simulate instructions in eBPF
ISA, and allocate ESI|EDI to BPF_REG_AX for constant blinding, all others
eBPF registers, R0-R10, are simulated through scratch space on stack.

The reasons behind the hardware registers allocation policy are:
1:MUL need EAX:EDX, shift operation need ECX, so they aren't fit
  for general eBPF 64bit register simulation.
2:We need at least 4 registers to simulate most eBPF ISA operations
  on registers operands instead of on register&memory operands.
3:We need to put BPF_REG_AX on hardware registers, or constant blinding
  will degrade jit performance heavily.

Tested on PC (Intel(R) Core(TM) i5-5200U CPU).
Testing results on i5-5200U:
1) test_bpf: Summary: 349 PASSED, 0 FAILED, [319/341 JIT'ed]
2) test_progs: Summary: 83 PASSED, 0 FAILED.
3) test_lpm: OK
4) test_lru_map: OK
5) test_verifier: Summary: 828 PASSED, 0 FAILED.

Above tests are all done in following two conditions separately:
1:bpf_jit_enable=1 and bpf_jit_harden=0
2:bpf_jit_enable=1 and bpf_jit_harden=2

Below are some numbers for this jit implementation:
Note:
  I run test_progs in kselftest 100 times continuously for every condition,
  the numbers are in format: total/times=avg.
  The numbers that test_bpf reports show almost the same relation.

a:jit_enable=0 and jit_harden=0            b:jit_enable=1 and jit_harden=0
  test_pkt_access:PASS:ipv4:15622/100=156    test_pkt_access:PASS:ipv4:10674/100=106
  test_pkt_access:PASS:ipv6:9130/100=91      test_pkt_access:PASS:ipv6:4855/100=48
  test_xdp:PASS:ipv4:240198/100=2401         test_xdp:PASS:ipv4:138912/100=1389
  test_xdp:PASS:ipv6:137326/100=1373         test_xdp:PASS:ipv6:68542/100=685
  test_l4lb:PASS:ipv4:61100/100=611          test_l4lb:PASS:ipv4:37302/100=373
  test_l4lb:PASS:ipv6:101000/100=1010        test_l4lb:PASS:ipv6:55030/100=550

c:jit_enable=1 and jit_harden=2
  test_pkt_access:PASS:ipv4:10558/100=105
  test_pkt_access:PASS:ipv6:5092/100=50
  test_xdp:PASS:ipv4:131902/100=1319
  test_xdp:PASS:ipv6:77932/100=779
  test_l4lb:PASS:ipv4:38924/100=389
  test_l4lb:PASS:ipv6:57520/100=575

The numbers show we get 30%~50% improvement.

See Documentation/networking/filter.txt for more information.

Changelog:

 Changes v5-v6:
 1:Add do {} while (0) to RETPOLINE_RAX_BPF_JIT for
   consistence reason.
 2:Clean up non-standard comments, reported by Daniel Borkmann.
 3:Fix a memory leak issue, repoted by Daniel Borkmann.

 Changes v4-v5:
 1:Delete is_on_stack, BPF_REG_AX is the only one
   on real hardware registers, so just check with
   it.
 2:Apply commit 1612a981b7 ("bpf, x64: fix JIT emission
   for dead code"), suggested by Daniel Borkmann.

 Changes v3-v4:
 1:Fix changelog in commit.
   I install llvm-6.0, then test_progs willn't report errors.
   I submit another patch:
   "bpf: fix misaligned access for BPF_PROG_TYPE_PERF_EVENT program type on x86_32 platform"
   to fix another problem, after that patch, test_verifier willn't report errors too.
 2:Fix clear r0[1] twice unnecessarily in *BPF_IND|BPF_ABS* simulation.

 Changes v2-v3:
 1:Move BPF_REG_AX to real hardware registers for performance reason.
 3:Using bpf_load_pointer instead of bpf_jit32.S, suggested by Daniel Borkmann.
 4:Delete partial codes in 1c2a088a66, suggested by Daniel Borkmann.
 5:Some bug fixes and comments improvement.

 Changes v1-v2:
 1:Fix bug in emit_ia32_neg64.
 2:Fix bug in emit_ia32_arsh_r64.
 3:Delete filename in top level comment, suggested by Thomas Gleixner.
 4:Delete unnecessary boiler plate text, suggested by Thomas Gleixner.
 5:Rewrite some words in changelog.
 6:CodingSytle improvement and a little more comments.

Signed-off-by: Wang YanQing <udknight@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-05-03 18:15:25 +02:00
..
ABI RTC for 4.17 2018-04-10 10:22:27 -07:00
accelerators ocxl: Document the OCXL_IOCTL_GET_METADATA IOCTL 2018-03-02 13:02:15 +11:00
accounting
acpi ACPI / LPIT: Add Low Power Idle Table (LPIT) support 2017-10-11 15:38:10 +02:00
admin-guide ARM: 2018-04-09 11:42:31 -07:00
aoe
arm MTD changes: 2018-04-06 12:15:41 -07:00
arm64 ARM: 2018-04-09 11:42:31 -07:00
auxdisplay
backlight
block block, bfq: move debug blkio stats behind CONFIG_DEBUG_BLK_CGROUP 2017-11-14 20:13:33 -07:00
blockdev
bpf bpf: Document sockmap '-target bpf' requirement for PROG_TYPE_SK_MSG 2018-04-23 23:42:21 +02:00
bus-devices
cdrom Documentation/cdrom: fix German sharp s in LaTex 2018-03-08 19:35:29 -07:00
cgroup-v1 page cache: use xa_lock 2018-04-11 10:28:39 -07:00
cma
connector
console
core-api textsearch: fix kernel-doc warnings and add kernel-api section 2018-04-16 18:53:13 -04:00
cpu-freq cpufreq: Drop cpufreq_table_validate_and_show() 2018-04-10 08:40:45 +02:00
cpuidle cpuidle: Add definition of residency to sysfs documentation 2018-04-09 13:44:37 +02:00
crypto crypto: doc - clarify hash callbacks state machine 2018-03-31 01:33:02 +08:00
dev-tools There's been a fair amount of activity in Documentation/ this time around: 2018-04-03 13:35:51 -07:00
device-mapper dm verity: add 'check_at_most_once' option to only validate hashes once 2018-04-03 15:04:29 -04:00
devicetree Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2018-04-21 16:32:48 -04:00
doc-guide doc-guide: kernel-doc: add comment about formatting verification 2018-02-23 08:06:15 -07:00
driver-api MTD changes: 2018-04-06 12:15:41 -07:00
driver-model serdev: Introduce devm_serdev_device_open() 2018-01-08 10:08:34 +00:00
early-userspace
EDID
extcon
fault-injection Documentation: nvme: Documentation for nvme fault injection 2018-03-26 08:53:43 -06:00
fb documentation: fb: update list of available compiled-in fonts 2017-11-08 03:39:52 -07:00
features arch: remove obsolete architecture ports 2018-04-02 20:20:12 -07:00
filesystems ipconfig: Write NTP server IPs to /proc/net/ipconfig/ntp_servers 2018-04-24 13:40:42 -04:00
firmware_class
fmc
fpga fpga: mgr: separate getting/locking FPGA manager 2017-11-28 16:30:37 +01:00
gpio Documentation: gpio: Move drivers-on-gpio.txt to driver-api 2018-03-23 04:22:29 +01:00
gpu Linux 4.16-rc7 2018-03-28 14:30:41 +10:00
hid Documentation: fix input related doc refs 2017-10-12 11:14:06 -06:00
hwmon hwmon: (lm92) Add max6635 to lm92_id[] 2018-03-22 09:33:24 -07:00
i2c i2c: i801: Add missing documentation entries for Braswell and Kaby Lake 2018-02-21 09:17:20 +01:00
ia64 ia64: doc: tweak whitespace for 'console=' parameter 2018-03-05 14:41:38 -08:00
ide
iio
infiniband Documentation/ABI: update infiniband sysfs interfaces 2018-02-23 08:18:33 -07:00
input Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input 2018-04-05 13:21:57 -07:00
ioctl arch: remove tile port 2018-03-16 10:56:03 +01:00
isdn Documentation/isdn: check and fix dead links ... 2018-03-26 12:31:13 -04:00
kbuild Kconfig updates for v4.17 2018-04-03 16:28:01 -07:00
kdump
kernel-hacking Documentation: Fix misconversion of #if 2018-01-17 16:45:01 -07:00
laptops Documentation: fix admin-guide doc refs 2017-10-12 11:13:28 -06:00
leds Documentation: leds: Update 00-INDEX file 2017-10-23 20:17:03 +02:00
lightnvm
livepatch livepatch: Allow to call a custom callback when freeing shadow variables 2018-04-17 13:42:48 +02:00
locking Linux 4.16-rc2 2018-02-21 09:57:55 +01:00
m68k
maintainer docs: Add an intro note to the maintainers handbook 2017-12-11 14:46:10 -07:00
md raid5-ppl: PPL support for disks with write-back cache enabled 2018-01-15 14:29:42 -08:00
media media: extended-controls.rst: transmitter -> receiver 2018-04-04 08:05:20 -04:00
memory-devices
mic
mips Documentation: mips: Update AU1xxx_IDE Kconfig dependencies 2018-02-01 12:45:35 -07:00
misc-devices
mmc
mtd mtd: spi-nor: add an API to restore the status of SPI flash chip 2017-12-13 00:36:00 +01:00
namespaces
netlabel
networking bpf, doc: Update bpf_jit_enable limitation for CONFIG_BPF_JIT_ALWAYS_ON 2018-04-27 12:08:13 +02:00
nfc
nios2
nvdimm
nvmem
openrisc Documentation: openrisc: Updates to README 2017-10-30 21:37:53 +09:00
parisc
PCI PCI: Update location of pci.ids file 2018-02-22 15:00:43 -06:00
pcmcia
perf drivers/bus: Move Arm CCN PMU driver 2018-03-06 17:26:15 +01:00
phy
platform
power PM / hibernate: Make passing hibernate offsets more friendly 2018-03-30 12:01:20 +02:00
powerpc
pps
process Merge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2018-04-15 16:12:35 -07:00
pti
ptp ptp: Fix documentation to match code. 2018-03-26 12:13:21 -04:00
rapidio Documentation: rapidio: move sysfs interface to ABI 2018-02-23 08:25:45 -07:00
RCU Merge branches 'cond_resched.2017.12.04a', 'dyntick.2017.11.28a', 'fixes.2017.12.11a', 'srbd.2017.12.05a' and 'torture.2017.12.11a' into HEAD 2017-12-11 09:21:58 -08:00
s390 vfio-ccw: update documentation 2018-03-01 17:32:14 +01:00
scheduler sched/deadline: Fix the description of runtime accounting in the documentation 2017-11-16 09:00:35 +01:00
scsi scsi: documentation: Obsolete documentation references 2018-03-21 18:34:20 -04:00
security selinux: Update SELinux SCTP documentation 2018-03-20 16:26:15 -04:00
serial
sh
sound sound updates for 4.15-rc1 2017-11-14 18:01:46 -08:00
sparc sparc64: Add support for ADI (Application Data Integrity) 2018-03-18 07:38:48 -07:00
sphinx Documentation/sphinx: Fix Directive import error 2018-03-09 10:46:14 -07:00
sphinx-static
spi
sysctl bpf, x86_32: add eBPF JIT compiler for ia32 2018-05-03 18:15:25 +02:00
target
thermal thermal: Add cooling device's statistics in sysfs 2018-04-02 21:49:01 +08:00
timers sched/isolation: Eliminate NO_HZ_FULL_ALL 2018-02-15 15:40:37 -08:00
trace New features: 2018-04-10 11:27:30 -07:00
translations Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/pmladek/printk 2018-02-01 13:36:15 -08:00
usb Documentation updates for 4.16. New stuff includes refcount_t 2018-01-31 19:25:25 -08:00
userspace-api
virtual KVM: trivial documentation cleanups 2018-03-28 22:47:06 +02:00
vm page cache: use xa_lock 2018-04-11 10:28:39 -07:00
w1 Documentation updates for 4.16. New stuff includes refcount_t 2018-01-31 19:25:25 -08:00
watchdog watchdog: remove bfin_wdt driver 2018-03-26 15:57:04 +02:00
wimax
x86 Merge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2018-04-15 16:12:35 -07:00
xtensa xtensa: add support for KASAN 2017-12-16 22:37:12 -08:00
.gitignore
00-INDEX CRIS: Drop support for the CRIS port 2018-03-16 10:56:05 +01:00
atomic_bitops.txt locking/atomic/bitops: Document and clarify ordering semantics for failed test_and_{}_bit() 2018-02-13 14:55:53 +01:00
atomic_t.txt
bcache.txt
bt8xxgpio.txt
btmrvl.txt
bus-virt-phys-mapping.txt
cachetlb.txt
cgroup-v2.txt cgroup, docs: document the root cgroup behavior of cpu and io controllers 2018-01-16 08:07:09 -08:00
Changes
circular-buffers.txt doc: De-emphasize smp_read_barrier_depends 2017-12-05 11:57:53 -08:00
clearing-warn-once.txt kernel debug: support resetting WARN*_ONCE 2017-11-17 16:10:00 -08:00
clk.txt Documentation: clk: enable lock is not held for clk_is_enabled API 2018-03-16 15:44:43 -07:00
CodingStyle
conf.py docs: Remove "could not extract kernel version" warning 2017-12-11 15:20:04 -07:00
cpu-load.txt
cputopology.txt
crc32.txt
dcdbas.txt
debugging-modules.txt
debugging-via-ohci1394.txt
dell_rbu.txt
digsig.txt
DMA-API-HOWTO.txt
DMA-API.txt
DMA-attributes.txt
DMA-ISA-LPC.txt
docutils.conf
dontdiff
efi-stub.txt
eisa.txt
flexible-arrays.txt
futex-requeue-pi.txt
gcc-plugins.txt
highuid.txt
hw_random.txt
hwspinlock.txt
index.rst Documentation: add Linux tracing to Sphinx TOC tree 2018-03-07 10:22:53 -07:00
intel_txt.txt
Intel-IOMMU.txt
io_ordering.txt
io-mapping.txt
iostats.txt
IPMI.txt
IRQ-affinity.txt
IRQ-domain.txt irqdomain: Kill CONFIG_IRQ_DOMAIN_DEBUG 2018-01-24 12:32:58 +01:00
IRQ.txt
irqflags-tracing.txt
isa.txt
isapnp.txt
kernel-per-CPU-kthreads.txt
kobject.txt
kprobes.txt kprobes/docs: Remove jprobes related documents 2017-10-20 11:02:55 +02:00
kref.txt
ldm.txt
lockup-watchdogs.txt
logo.gif
logo.txt
lsm.txt
lzo.txt
mailbox.txt
Makefile Documentation: add script and build target to check for broken file references 2017-10-12 11:07:42 -06:00
memory-barriers.txt locking/memory-barriers: De-emphasize smp_read_barrier_depends() some more 2018-03-10 10:22:22 +01:00
memory-hotplug.txt
men-chameleon-bus.txt
nommu-mmap.txt
ntb.txt
numastat.txt
padata.txt
parport-lowlevel.txt
percpu-rw-semaphore.txt
phy.txt
pi-futex.txt Documentation: fix locking rt-mutex doc refs 2017-10-19 12:56:44 -06:00
pnp.txt
preempt-locking.txt
pwm.txt
rbtree.txt
remoteproc.txt
rfkill.txt
robust-futex-ABI.txt
robust-futexes.txt
rpmsg.txt
rtc.txt Documentation: rtc: move iotcl interface documentation to ABI 2018-01-12 00:20:41 +01:00
SAK.txt
sgi-ioc4.txt
siphash.txt
SM501.txt
smsc_ece1099.txt
speculation.txt Documentation: Document array_index_nospec 2018-01-30 21:54:28 +01:00
static-keys.txt
SubmittingPatches
svga.txt documentation/svga.txt: update outdated file 2017-11-20 10:45:50 -07:00
switchtec.txt NTB: switchtec_ntb: Update switchtec documentation with notes for NTB 2017-11-18 20:37:13 -05:00
sync_file.txt
tee.txt
this_cpu_ops.txt
unaligned-memory-access.txt
vfio-mediated-device.txt
vfio.txt
video-output.txt
xillybus.txt
xz.txt
zorro.txt