)]}' { "commit": "a8edd994792a6426d3f28ef4c85b6bb3f0d7ec05", "tree": "5ca2e3917be14a1f2c1039dd3e9a7305308929f3", "parents": [ "bd04f65511791860276f2f3f982133f7be007448" ], "author": { "name": "kmvijay", "email": "kiran.m.vijay@ibm.com", "time": "Wed Jun 18 10:57:46 2025 +0000" }, "committer": { "name": "Gopher Robot", "email": "gobot@golang.org", "time": "Thu Jul 24 10:09:39 2025 -0700" }, "message": "runtime: improvement in memclr for s390x\n\nThe unrolled loop for sizes \u003e\u003d 4KB is further optimized.\nOffsets are computed and included in the XC instruction directly.\nThis reduces code size and instructions, improves performance.\n\ngoos: linux\ngoarch: s390x\npkg: runtime\n | Orig_Memclr_for_benchstat_2.log | MM_Memclr_for_benchstat_No_VSTL_3.log |\n | sec/op | sec/op vs base |\nMemclr/5 1.925n ± 0% 1.925n ± 0% ~ (p\u003d0.211 n\u003d10)\nMemclr/16 2.604n ± 13% 2.633n ± 11% ~ (p\u003d0.912 n\u003d10)\nMemclr/64 3.598n ± 2% 3.520n ± 5% ~ (p\u003d0.190 n\u003d10)\nMemclr/256 3.571n ± 12% 3.538n ± 11% ~ (p\u003d0.739 n\u003d10)\nMemclr/4096 15.15n ± 0% 15.14n ± 0% ~ (p\u003d0.204 n\u003d10)\nMemclr/65536 226.3n ± 0% 224.9n ± 0% -0.62% (p\u003d0.000 n\u003d10)\nMemclr/1M 12.77µ ± 0% 12.60µ ± 0% -1.35% (p\u003d0.000 n\u003d10)\nMemclr/4M 51.07µ ± 0% 50.37µ ± 0% -1.38% (p\u003d0.000 n\u003d10)\nMemclr/8M 102.1µ ± 0% 100.7µ ± 0% -1.36% (p\u003d0.000 n\u003d10)\nMemclr/16M 204.4µ ± 0% 201.6µ ± 0% -1.35% (p\u003d0.000 n\u003d10)\nMemclr/64M 965.4µ ± 0% 935.3µ ± 0% -3.12% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/0_5 2.671n ± 6% 2.618n ± 0% ~ (p\u003d0.194 n\u003d10)\nMemclrUnaligned/0_16 3.143n ± 6% 2.955n ± 8% ~ (p\u003d0.089 n\u003d10)\nMemclrUnaligned/0_64 3.622n ± 3% 3.571n ± 2% ~ (p\u003d0.304 n\u003d10)\nMemclrUnaligned/0_256 3.712n ± 8% 3.653n ± 5% ~ (p\u003d0.754 n\u003d10)\nMemclrUnaligned/0_4096 15.14n ± 0% 15.14n ± 0% ~ (p\u003d1.000 n\u003d10) ¹\nMemclrUnaligned/0_65536 231.9n ± 0% 225.2n ± 0% -2.91% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/1_5 2.620n ± 8% 2.620n ± 0% ~ (p\u003d0.866 n\u003d10)\nMemclrUnaligned/1_16 3.103n ± 7% 2.933n ± 9% ~ (p\u003d0.052 n\u003d10)\nMemclrUnaligned/1_64 3.576n ± 3% 3.568n ± 3% ~ (p\u003d0.748 n\u003d10)\nMemclrUnaligned/1_256 3.744n ± 9% 3.709n ± 10% ~ (p\u003d0.853 n\u003d10)\nMemclrUnaligned/1_4096 26.23n ± 0% 26.23n ± 0% ~ (p\u003d1.000 n\u003d10) ¹\nMemclrUnaligned/1_65536 401.1n ± 0% 399.5n ± 0% -0.40% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/4_5 2.620n ± 6% 2.623n ± 0% ~ (p\u003d0.985 n\u003d10)\nMemclrUnaligned/4_16 3.095n ± 7% 3.005n ± 9% ~ (p\u003d0.247 n\u003d10)\nMemclrUnaligned/4_64 3.586n ± 1% 3.578n ± 3% ~ (p\u003d1.000 n\u003d10)\nMemclrUnaligned/4_256 3.843n ± 5% 3.742n ± 10% ~ (p\u003d0.971 n\u003d10)\nMemclrUnaligned/4_4096 26.23n ± 0% 26.23n ± 0% ~ (p\u003d1.000 n\u003d10)\nMemclrUnaligned/4_65536 401.1n ± 0% 399.5n ± 0% -0.41% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/7_5 2.634n ± 6% 2.644n ± 4% ~ (p\u003d0.896 n\u003d10)\nMemclrUnaligned/7_16 3.119n ± 7% 3.044n ± 9% ~ (p\u003d0.529 n\u003d10)\nMemclrUnaligned/7_64 3.568n ± 1% 3.585n ± 3% ~ (p\u003d0.499 n\u003d10)\nMemclrUnaligned/7_256 3.741n ± 9% 3.629n ± 6% ~ (p\u003d0.853 n\u003d10)\nMemclrUnaligned/7_4096 26.23n ± 0% 26.23n ± 0% ~ (p\u003d1.000 n\u003d10) ¹\nMemclrUnaligned/7_65536 401.1n ± 0% 399.4n ± 0% -0.42% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/0_1M 12.82µ ± 0% 12.60µ ± 0% -1.70% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/0_4M 51.28µ ± 0% 50.37µ ± 0% -1.77% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/0_8M 102.5µ ± 0% 100.8µ ± 0% -1.75% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/0_16M 205.1µ ± 0% 201.7µ ± 0% -1.62% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/0_64M 965.2µ ± 0% 934.7µ ± 0% -3.16% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/1_1M 16.02µ ± 0% 15.81µ ± 0% -1.34% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/1_4M 64.03µ ± 0% 63.20µ ± 0% -1.29% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/1_8M 128.0µ ± 0% 126.4µ ± 0% -1.27% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/1_16M 256.3µ ± 0% 253.2µ ± 0% -1.21% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/1_64M 1.210m ± 0% 1.187m ± 0% -1.88% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/4_1M 16.03µ ± 0% 15.81µ ± 0% -1.37% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/4_4M 64.04µ ± 0% 63.20µ ± 0% -1.31% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/4_8M 128.0µ ± 0% 126.4µ ± 0% -1.27% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/4_16M 256.1µ ± 0% 253.0µ ± 0% -1.20% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/4_64M 1.210m ± 0% 1.188m ± 0% -1.81% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/7_1M 16.02µ ± 0% 15.81µ ± 0% -1.32% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/7_4M 64.06µ ± 0% 63.21µ ± 0% -1.34% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/7_8M 128.1µ ± 0% 126.4µ ± 0% -1.29% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/7_16M 256.2µ ± 0% 253.2µ ± 0% -1.18% (p\u003d0.000 n\u003d10)\nMemclrUnaligned/7_64M 1.210m ± 0% 1.188m ± 0% -1.82% (p\u003d0.000 n\u003d10)\nMemclrRange/1K_2K 841.1n ± 1% 879.0n ± 3% +4.51% (p\u003d0.002 n\u003d10)\nMemclrRange/2K_8K 1.435µ ± 2% 1.415µ ± 0% -1.39% (p\u003d0.000 n\u003d10)\nMemclrRange/4K_16K 1.241µ ± 0% 1.209µ ± 0% -2.58% (p\u003d0.000 n\u003d10)\nMemclrRange/160K_228K 19.83µ ± 0% 19.59µ ± 0% -1.22% (p\u003d0.000 n\u003d10)\nMemclrKnownSize1 1.732n ± 0% 1.732n ± 0% ~ (p\u003d0.474 n\u003d10)\nMemclrKnownSize2 1.925n ± 3% 1.925n ± 1% ~ (p\u003d0.929 n\u003d10)\nMemclrKnownSize4 1.732n ± 0% 1.732n ± 0% ~ (p\u003d1.000 n\u003d10) ¹\nMemclrKnownSize8 1.732n ± 0% 1.732n ± 0% ~ (p\u003d1.000 n\u003d10)\nMemclrKnownSize16 2.413n ± 9% 2.681n ± 14% +11.10% (p\u003d0.004 n\u003d10)\nMemclrKnownSize32 3.284n ± 4% 3.328n ± 2% ~ (p\u003d0.671 n\u003d10)\nMemclrKnownSize64 4.893n ± 1% 4.882n ± 1% ~ (p\u003d0.591 n\u003d10)\nMemclrKnownSize112 5.623n ± 2% 5.596n ± 2% -0.48% (p\u003d0.027 n\u003d10)\nMemclrKnownSize128 5.612n ± 1% 5.599n ± 0% ~ (p\u003d0.066 n\u003d10)\nMemclrKnownSize192 7.128n ± 1% 7.337n ± 2% +2.93% (p\u003d0.000 n\u003d10)\nMemclrKnownSize248 6.740n ± 1% 6.829n ± 3% +1.33% (p\u003d0.005 n\u003d10)\nMemclrKnownSize256 3.657n ± 8% 3.512n ± 14% ~ (p\u003d0.436 n\u003d10)\nMemclrKnownSize512 3.624n ± 3% 3.982n ± 9% +9.88% (p\u003d0.017 n\u003d10)\nMemclrKnownSize1024 4.662n ± 0% 4.680n ± 0% +0.39% (p\u003d0.000 n\u003d10)\nMemclrKnownSize4096 15.14n ± 0% 15.15n ± 0% +0.07% (p\u003d0.000 n\u003d10)\nMemclrKnownSize512KiB 6.388µ ± 0% 6.309µ ± 0% -1.24% (p\u003d0.000 n\u003d10)\ngeomean 268.9n 266.9n -0.75%\n¹ all samples are equal\n\nChange-Id: I2911866fb82777311ec4219600fb48c85f7bf862\nReviewed-on: https://go-review.googlesource.com/c/go/+/682595\nReviewed-by: Keith Randall \u003ckhr@google.com\u003e\nReviewed-by: Keith Randall \u003ckhr@golang.org\u003e\nAuto-Submit: Michael Knyszek \u003cmknyszek@google.com\u003e\nAuto-Submit: Keith Randall \u003ckhr@golang.org\u003e\nLUCI-TryBot-Result: Go LUCI \u003cgolang-scoped@luci-project-accounts.iam.gserviceaccount.com\u003e\nReviewed-by: Michael Knyszek \u003cmknyszek@google.com\u003e\n", "tree_diff": [ { "type": "modify", "old_id": "656e96998c93e32ac5937d0d41777880a303baeb", "old_mode": 33188, "old_path": "src/runtime/memclr_s390x.s", "new_id": "392057565e8cb390d9b5346985604160a47e2039", "new_mode": 33188, "new_path": "src/runtime/memclr_s390x.s" } ] }