riscv: __asm_copy_to-from_user: Optimize unaligned memory access and pipeline stall

[mcd500]

This patch will reduce cpu usage dramatically in kernel space especially
for user apps which use sys-call with large buffer size, such as network
applications. The main reason behind this is that every unaligned memory
access will raise exceptions and switch between s-mode and m-mode causing
large overhead.

First copy in bytes until reaches the first word aligned boundary in
destination memory address. This is the preparation before the bulk
aligned word copy.

The destination address is aligned now, but oftentimes the source address
is not in an aligned boundary. To reduce the unaligned memory access, it
reads the data from source in aligned boundaries, which will cause the
data to have an offset, and then combines the data in the next iteration
by fixing offset with shifting before writing to destination. The majority
of the improving copy speed comes from this shift copy.

In the lucky situation that the both source and destination address are on
the aligned boundary, perform load and store with register size to copy the
data. Without the unrolling, it will reduce the speed since the next store
instruction for the same register using from the load will stall the
pipeline.

At last, copying the remainder in one byte at a time.

Signed-off-by: Akira Tsukamoto akira.tsukamoto@gmail.com

[mcd500]

This is the PR reflecting the discussion at
http://lists.infradead.org/pipermail/linux-riscv/2021-June/007165.html

[mcd500]

The cover letter.

Optimizing copy_to_user and copy_from_user.

I rewrote the functions but heavily influenced by Garry's memcpy
function [1]. It must be written in assembler to handle page faults
manually inside the function unlike other memcpy functions.

This patch will reduce cpu usage dramatically in kernel space especially
for applications which use sys-call with large buffer size, such as network
applications. The main reason behind this is that every unaligned memory
access will raise exceptions and switch between s-mode and m-mode causing
large overhead.

The motivation to create the patch was to improve network performance on
beaglev beta board. By observing with perf, the memcpy and
__asm_copy_to_user had heavy cpu usage and the network speed was limited
at around 680Mbps on 1Gbps lan. Matteo is creating the patches to improve
memcpy in C and this patch is meant to be used with them.

Typical network applications use system calls with a large buffer on
send/recv() and sendto/recvfrom() for the optimization.
 
The bench result, when patching only copy_user. The memcpy is without
Matteo's patches but listing the both since they are the top two largest
overhead.

All results are from the same base kernel, same rootfs and same BeagleV
beta board.

Results of iperf3 have speedup on UDP with the copy_user patch alone.

--- UDP send ---
306 Mbits/sec      362 Mbits/sec
305 Mbits/sec      362 Mbits/sec

--- UDP recv ---
772 Mbits/sec      787 Mbits/sec
773 Mbits/sec      784 Mbits/sec

Comparison by "perf top -Ue task-clock" while running iperf3.

--- TCP recv ---
 * Before
  40.40%  [kernel]  [k] memcpy
  33.09%  [kernel]  [k] __asm_copy_to_user
 * With patch
  50.35%  [kernel]  [k] memcpy
  13.76%  [kernel]  [k] __asm_copy_to_user

--- TCP send ---
 * Before
  19.96%  [kernel]  [k] memcpy
   9.84%  [kernel]  [k] __asm_copy_to_user
 * With patch
  14.27%  [kernel]  [k] memcpy
   7.37%  [kernel]  [k] __asm_copy_to_user

--- UDP recv ---
 * Before
  44.45%  [kernel]  [k] memcpy
  31.04%  [kernel]  [k] __asm_copy_to_user
 * With patch
  55.62%  [kernel]  [k] memcpy
  11.22%  [kernel]  [k] __asm_copy_to_user

--- UDP send ---
 * Before
  25.18%  [kernel]  [k] memcpy
  22.50%  [kernel]  [k] __asm_copy_to_user
 * With patch
  28.90%  [kernel]  [k] memcpy
   9.49%  [kernel]  [k] __asm_copy_to_user

---
v2 -> v3:
- Merged all patches

v1 -> v2:
- Added shift copy
- Separated patches for readability of changes in assembler
- Using perf results

[1] https://lkml.org/lkml/2021/2/16/778

Akira Tsukamoto (1):
  riscv: __asm_copy_to-from_user: Optimize unaligned memory access and
    pipeline stall

 arch/riscv/lib/uaccess.S | 181 +++++++++++++++++++++++++++++++--------
 1 file changed, 146 insertions(+), 35 deletions(-)

-- 
2.17.1

[mcd500]

The Matteo's memcpy is merged in beaglev branch, so only this patch is required from the initial network performance improvement #17.

[esmil]

I picked it from riscv/for-next, thanks.

[mcd500]

@esmil Thanks!