https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97417
--- Comment #54 from jiawei <jiawei at iscas dot ac.cn> ---
Hi Jim.
I had finished the test on the benchmark Coremark-pro.And it shows that the
patch doesn't accidentally increase code size.
This test with the args "XCMD='-c4' certify-all", and the result shows follow:
WORKLOAD RESULTS TABLE(origin gcc-10.2.0 from upstream compiled with -mabi=lp64
-march=rv64imafdc)
MultiCore SingleCore
Workload Name (iter/s) (iter/s)
Scaling
----------------------------------------------- ---------- ----------
----------
cjpeg-rose7-preset 119.05 46.08
2.58
core 1.10 0.28
3.93
linear_alg-mid-100x100-sp 35.19 8.92
3.95
loops-all-mid-10k-sp 1.77 0.46
3.85
nnet_test 1.69 0.51
3.31
parser-125k 43.01 15.62
2.75
radix2-big-64k 257.93 56.84
4.54
sha-test 156.25 57.80
2.70
zip-test 81.63 26.32
3.10
MARK RESULTS TABLE
Mark Name MultiCore SingleCore
Scaling
----------------------------------------------- ---------- ----------
----------
CoreMark-PRO 2669.64 796.33
3.35
WORKLOAD RESULTS TABLE(Add patch one -- Auto-extend Patch)
MultiCore SingleCore
Workload Name (iter/s) (iter/s)
Scaling
----------------------------------------------- ---------- ----------
----------
cjpeg-rose7-preset 131.58 45.87
2.87
core 1.08 0.28
3.86
linear_alg-mid-100x100-sp 35.49 8.75
4.06
loops-all-mid-10k-sp 1.75 0.46
3.80
nnet_test 1.68 0.51
3.29
parser-125k 54.05 15.62
3.46
radix2-big-64k 257.80 65.57
3.93
sha-test 153.85 57.47
2.68
zip-test 76.92 26.32
2.92
MARK RESULTS TABLE
Mark Name MultiCore SingleCore
Scaling
----------------------------------------------- ---------- ----------
----------
CoreMark-PRO 2737.52 806.42
3.39
WORKLOAD RESULTS TABLE(Add all patches -- Auto-extend Patch & untested fix to
use instead of levy's combine.c patch)
MultiCore SingleCore
Workload Name (iter/s) (iter/s)
Scaling
----------------------------------------------- ---------- ----------
----------
cjpeg-rose7-preset 120.48 45.66
2.64
core 1.08 0.28
3.86
linear_alg-mid-100x100-sp 35.61 8.77
4.06
loops-all-mid-10k-sp 1.76 0.46
3.83
nnet_test 1.68 0.51
3.29
parser-125k 46.51 15.62
2.98
radix2-big-64k 257.33 65.18
3.95
sha-test 153.85 57.47
2.68
zip-test 83.33 26.32
3.17
MARK RESULTS TABLE
Mark Name MultiCore SingleCore
Scaling
----------------------------------------------- ---------- ----------
----------
CoreMark-PRO 2691.95 805.68
3.34
The csibe test is still modifying,and will test after it works on
riscv-gnu-toolchain.
