Ryzen 9 3950X Linux 6.0 Mitigations Benchmark

Zen 2 mitigation benchmarks by Michael Larabel for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2208199-NE-RYZEN939580
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results

Limit displaying results to tests within:

AV1 2 Tests
Web Browsers 1 Tests
Timed Code Compilation 3 Tests
C/C++ Compiler Tests 6 Tests
CPU Massive 16 Tests
Creator Workloads 10 Tests
Database Test Suite 7 Tests
Disk Test Suite 2 Tests
Encoding 3 Tests
Game Development 3 Tests
Go Language Tests 3 Tests
HPC - High Performance Computing 2 Tests
Imaging 4 Tests
Java 2 Tests
Common Kernel Benchmarks 7 Tests
Multi-Core 9 Tests
Node.js + NPM Tests 2 Tests
Productivity 3 Tests
Programmer / Developer System Benchmarks 7 Tests
Python 2 Tests
Scientific Computing 2 Tests
Server 12 Tests
Server CPU Tests 9 Tests
Single-Threaded 6 Tests
Video Encoding 2 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
Default
August 16 2022
  5 Hours, 18 Minutes
mitigations=off
August 17 2022
  5 Hours, 40 Minutes
retbleed=ibpb
August 17 2022
  6 Hours, 42 Minutes
retbleed=ibpb spectre_v2=on
August 18 2022
  6 Hours, 27 Minutes
retbleed=ibpb spectre_v2=on nosmt
August 18 2022
  6 Hours, 2 Minutes
Invert Hiding All Results Option
  6 Hours, 2 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):

Ryzen 9 3950X Linux 6.0 Mitigations Benchmark Zen 2 mitigation benchmarks by Michael Larabel for a future article. Default: Processor: AMD Ryzen 9 3950X 16-Core @ 3.50GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR VII HERO (WI-FI) (3103 BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: Samsung SSD 970 EVO 250GB, Graphics: Sapphire AMD Radeon RX 470 4GB (1260/1750MHz), Audio: AMD Ellesmere HDMI Audio, Monitor: DELL S2409W, Network: Intel I211 + Realtek RTL8822BE 802.11a/b/g/n/ac OS: Ubuntu 22.04, Kernel: 5.19.0-051900daily20220813-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.5 (LLVM 13.0.1 DRM 3.48), Vulkan: 1.3.204, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 1920x1080 mitigations=off: Processor: AMD Ryzen 9 3950X 16-Core @ 3.50GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR VII HERO (WI-FI) (3103 BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: Samsung SSD 970 EVO 250GB, Graphics: Sapphire AMD Radeon RX 470 4GB (1260/1750MHz), Audio: AMD Ellesmere HDMI Audio, Monitor: DELL S2409W, Network: Intel I211 + Realtek RTL8822BE 802.11a/b/g/n/ac OS: Ubuntu 22.04, Kernel: 5.19.0-051900daily20220813-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.5 (LLVM 13.0.1 DRM 3.48), Vulkan: 1.3.204, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 1920x1080 retbleed=ibpb: Processor: AMD Ryzen 9 3950X 16-Core @ 3.50GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR VII HERO (WI-FI) (3103 BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: Samsung SSD 970 EVO 250GB, Graphics: Sapphire AMD Radeon RX 470 4GB (1260/1750MHz), Audio: AMD Ellesmere HDMI Audio, Monitor: DELL S2409W, Network: Intel I211 + Realtek RTL8822BE 802.11a/b/g/n/ac OS: Ubuntu 22.04, Kernel: 5.19.0-051900daily20220813-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.5 (LLVM 13.0.1 DRM 3.48), Vulkan: 1.3.204, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 1920x1080 retbleed=ibpb spectre_v2=on: Processor: AMD Ryzen 9 3950X 16-Core @ 3.50GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR VII HERO (WI-FI) (3103 BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: Samsung SSD 970 EVO 250GB, Graphics: Sapphire AMD Radeon RX 470 4GB (1260/1750MHz), Audio: AMD Ellesmere HDMI Audio, Monitor: DELL S2409W, Network: Intel I211 + Realtek RTL8822BE 802.11a/b/g/n/ac OS: Ubuntu 22.04, Kernel: 5.19.0-051900daily20220813-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.5 (LLVM 13.0.1 DRM 3.48), Vulkan: 1.3.204, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 1920x1080 retbleed=ibpb spectre_v2=on nosmt: Processor: AMD Ryzen 9 3950X 16-Core @ 3.50GHz (16 Cores), Motherboard: ASUS ROG CROSSHAIR VII HERO (WI-FI) (3103 BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: Samsung SSD 970 EVO 250GB, Graphics: Sapphire AMD Radeon RX 470 4GB (1260/1750MHz), Audio: AMD Ellesmere HDMI Audio, Monitor: DELL S2409W, Network: Intel I211 + Realtek RTL8822BE 802.11a/b/g/n/ac OS: Ubuntu 22.04, Kernel: 5.19.0-051900daily20220813-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.5 (LLVM 13.0.1 DRM 3.48), Vulkan: 1.3.204, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 1920x1080 LuaRadio 0.9.1 Test: Five Back to Back FIR Filters MiB/s > Higher Is Better Default ........................... 868.9 |================================== mitigations=off ................... 924.7 |==================================== retbleed=ibpb ..................... 863.6 |================================== retbleed=ibpb spectre_v2=on ....... 838.4 |================================= retbleed=ibpb spectre_v2=on nosmt . 650.2 |========================= Renaissance 0.14 Test: ALS Movie Lens ms < Lower Is Better Default ........................... 13162.7 |========================== mitigations=off ................... 13019.1 |========================== retbleed=ibpb ..................... 16937.8 |================================== retbleed=ibpb spectre_v2=on ....... 16916.4 |================================== retbleed=ibpb spectre_v2=on nosmt . 15340.7 |=============================== WireGuard + Linux Networking Stack Stress Test Seconds < Lower Is Better Default ........................... 214.26 |======================== mitigations=off ................... 200.14 |======================= retbleed=ibpb ..................... 302.49 |=================================== retbleed=ibpb spectre_v2=on ....... 306.37 |=================================== retbleed=ibpb spectre_v2=on nosmt . 302.50 |=================================== Apache Spark 3.3 Row Count: 20000000 - Partitions: 100 - Group By Test Time Seconds < Lower Is Better Default ........................... 12.80 |=================================== mitigations=off ................... 12.26 |================================= retbleed=ibpb ..................... 13.16 |==================================== retbleed=ibpb spectre_v2=on ....... 13.30 |==================================== retbleed=ibpb spectre_v2=on nosmt . 13.01 |=================================== Apache Spark 3.3 Row Count: 20000000 - Partitions: 100 - Calculate Pi Benchmark Seconds < Lower Is Better Default ........................... 105.70 |================================== mitigations=off ................... 91.58 |============================= retbleed=ibpb ..................... 94.16 |============================== retbleed=ibpb spectre_v2=on ....... 109.19 |=================================== retbleed=ibpb spectre_v2=on nosmt . 103.56 |================================= Apache Spark 3.3 Row Count: 20000000 - Partitions: 500 - Inner Join Test Time Seconds < Lower Is Better Default ........................... 24.30 |==================================== mitigations=off ................... 23.68 |=================================== retbleed=ibpb ..................... 24.52 |==================================== retbleed=ibpb spectre_v2=on ....... 24.05 |=================================== retbleed=ibpb spectre_v2=on nosmt . 21.51 |================================ Apache Spark 3.3 Row Count: 20000000 - Partitions: 500 - Repartition Test Time Seconds < Lower Is Better Default ........................... 21.37 |==================================== mitigations=off ................... 20.96 |=================================== retbleed=ibpb ..................... 21.34 |==================================== retbleed=ibpb spectre_v2=on ....... 21.29 |==================================== retbleed=ibpb spectre_v2=on nosmt . 18.56 |=============================== Apache Spark 3.3 Row Count: 20000000 - Partitions: 500 - Calculate Pi Benchmark Seconds < Lower Is Better Default ........................... 105.14 |================================== mitigations=off ................... 91.20 |============================= retbleed=ibpb ..................... 94.48 |============================== retbleed=ibpb spectre_v2=on ....... 108.93 |=================================== retbleed=ibpb spectre_v2=on nosmt . 103.60 |================================= Memtier_benchmark 1.3 Protocol: Redis Ops/sec > Higher Is Better Default ........................... 2220557.34 |=============================== mitigations=off ................... 2218502.81 |=============================== retbleed=ibpb ..................... 2006224.43 |============================ retbleed=ibpb spectre_v2=on ....... 1988745.57 |============================ retbleed=ibpb spectre_v2=on nosmt . 1937718.00 |=========================== nginx 1.21.1 Concurrent Requests: 1 Requests Per Second > Higher Is Better Default ........................... 49382.88 |============================= mitigations=off ................... 56612.56 |================================= retbleed=ibpb ..................... 20712.40 |============ retbleed=ibpb spectre_v2=on ....... 20664.66 |============ retbleed=ibpb spectre_v2=on nosmt . 29830.25 |================= GROMACS 2022.1 Implementation: MPI CPU - Input: water_GMX50_bare Ns Per Day > Higher Is Better Default ........................... 1.227 |==================================== mitigations=off ................... 1.227 |==================================== retbleed=ibpb ..................... 1.222 |==================================== retbleed=ibpb spectre_v2=on ....... 1.222 |==================================== retbleed=ibpb spectre_v2=on nosmt . 1.232 |==================================== Renaissance 0.14 Test: Genetic Algorithm Using Jenetics + Futures ms < Lower Is Better Default ........................... 2878.9 |=============================== mitigations=off ................... 2865.0 |=============================== retbleed=ibpb ..................... 3179.9 |=================================== retbleed=ibpb spectre_v2=on ....... 3221.1 |=================================== retbleed=ibpb spectre_v2=on nosmt . 1690.7 |================== Stargate Digital Audio Workstation 21.10.9 Sample Rate: 44100 - Buffer Size: 512 Render Ratio > Higher Is Better Default ........................... 5.132632 |================================ mitigations=off ................... 5.259331 |================================= retbleed=ibpb ..................... 4.737688 |============================== retbleed=ibpb spectre_v2=on ....... 4.944057 |=============================== retbleed=ibpb spectre_v2=on nosmt . 5.075336 |================================ Stargate Digital Audio Workstation 21.10.9 Sample Rate: 96000 - Buffer Size: 512 Render Ratio > Higher Is Better Default ........................... 3.883713 |================================= mitigations=off ................... 3.919670 |================================= retbleed=ibpb ..................... 3.520352 |============================== retbleed=ibpb spectre_v2=on ....... 3.182824 |=========================== retbleed=ibpb spectre_v2=on nosmt . 3.626779 |=============================== Apache Cassandra 4.0 Test: Writes Op/s > Higher Is Better Default ........................... 156420 |================================= mitigations=off ................... 166783 |=================================== retbleed=ibpb ..................... 123724 |========================== retbleed=ibpb spectre_v2=on ....... 122019 |========================== retbleed=ibpb spectre_v2=on nosmt . 71705 |=============== Stress-NG 0.14 Test: Futex Bogo Ops/s > Higher Is Better Default ........................... 4960793.44 |=========================== mitigations=off ................... 5674415.95 |=============================== retbleed=ibpb ..................... 2102526.60 |=========== retbleed=ibpb spectre_v2=on ....... 1694057.10 |========= retbleed=ibpb spectre_v2=on nosmt . 1044265.16 |====== Hackbench Count: 32 - Type: Process Seconds < Lower Is Better Default ........................... 56.60 |========= mitigations=off ................... 47.47 |======== retbleed=ibpb ..................... 138.09 |====================== retbleed=ibpb spectre_v2=on ....... 147.74 |======================= retbleed=ibpb spectre_v2=on nosmt . 221.42 |=================================== PostMark 1.51 Disk Transaction Performance TPS > Higher Is Better Default ........................... 5906 |============================= mitigations=off ................... 7504 |===================================== retbleed=ibpb ..................... 2578 |============= retbleed=ibpb spectre_v2=on ....... 2501 |============ retbleed=ibpb spectre_v2=on nosmt . 2450 |============ Stargate Digital Audio Workstation 21.10.9 Sample Rate: 96000 - Buffer Size: 1024 Render Ratio > Higher Is Better Default ........................... 4.205922 |================================= mitigations=off ................... 4.262229 |================================= retbleed=ibpb ..................... 3.937179 |============================== retbleed=ibpb spectre_v2=on ....... 3.932585 |============================== retbleed=ibpb spectre_v2=on nosmt . 3.820665 |============================== Selenium Benchmark: Jetstream 2 - Browser: Google Chrome Score > Higher Is Better Default ........................... 172.75 |================================= mitigations=off ................... 177.03 |================================== retbleed=ibpb ..................... 161.33 |=============================== retbleed=ibpb spectre_v2=on ....... 159.72 |=============================== retbleed=ibpb spectre_v2=on nosmt . 182.39 |=================================== InfluxDB 1.8.2 Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 val/sec > Higher Is Better Default ........................... 849704.4 |==================== mitigations=off ................... 834613.9 |==================== retbleed=ibpb ..................... 906678.8 |====================== retbleed=ibpb spectre_v2=on ....... 889855.0 |===================== retbleed=ibpb spectre_v2=on nosmt . 1342020.3 |================================ Stargate Digital Audio Workstation 21.10.9 Sample Rate: 480000 - Buffer Size: 1024 Render Ratio > Higher Is Better Default ........................... 5.463150 |================================= mitigations=off ................... 5.509905 |================================= retbleed=ibpb ..................... 5.265175 |================================ retbleed=ibpb spectre_v2=on ....... 5.285263 |================================ retbleed=ibpb spectre_v2=on nosmt . 5.360599 |================================ SQLite Speedtest 3.30 Timed Time - Size 1,000 Seconds < Lower Is Better Default ........................... 60.35 |================ mitigations=off ................... 55.00 |============== retbleed=ibpb ..................... 127.80 |================================= retbleed=ibpb spectre_v2=on ....... 131.54 |================================== retbleed=ibpb spectre_v2=on nosmt . 134.76 |=================================== KeyDB 6.2.0 Ops/sec > Higher Is Better Default ........................... 555290.25 |============================== mitigations=off ................... 600946.79 |================================ retbleed=ibpb ..................... 301117.97 |================ retbleed=ibpb spectre_v2=on ....... 297634.34 |================ retbleed=ibpb spectre_v2=on nosmt . 337107.71 |================== Apache HTTP Server 2.4.48 Concurrent Requests: 1 Requests Per Second > Higher Is Better Default ........................... 9543.29 |============================== mitigations=off ................... 10608.67 |================================= retbleed=ibpb ..................... 5523.09 |================= retbleed=ibpb spectre_v2=on ....... 5439.32 |================= retbleed=ibpb spectre_v2=on nosmt . 10129.16 |================================ Apache HTTP Server 2.4.48 Concurrent Requests: 1000 Requests Per Second > Higher Is Better Default ........................... 54929.69 |============================== mitigations=off ................... 55587.58 |============================== retbleed=ibpb ..................... 55162.09 |============================== retbleed=ibpb spectre_v2=on ....... 55127.26 |============================== retbleed=ibpb spectre_v2=on nosmt . 60459.88 |================================= Apache HTTP Server 2.4.48 Concurrent Requests: 500 Requests Per Second > Higher Is Better Default ........................... 55168.30 |============================== mitigations=off ................... 56022.25 |============================== retbleed=ibpb ..................... 55131.26 |============================== retbleed=ibpb spectre_v2=on ....... 55356.80 |============================== retbleed=ibpb spectre_v2=on nosmt . 60779.72 |================================= Apache HTTP Server 2.4.48 Concurrent Requests: 200 Requests Per Second > Higher Is Better Default ........................... 52762.31 |============================= mitigations=off ................... 54782.41 |============================== retbleed=ibpb ..................... 51659.41 |============================ retbleed=ibpb spectre_v2=on ....... 51569.13 |============================ retbleed=ibpb spectre_v2=on nosmt . 60458.70 |================================= nginx 1.21.1 Concurrent Requests: 1000 Requests Per Second > Higher Is Better Default ........................... 257129.85 |============================ mitigations=off ................... 298369.31 |================================ retbleed=ibpb ..................... 222660.55 |======================== retbleed=ibpb spectre_v2=on ....... 215234.93 |======================= retbleed=ibpb spectre_v2=on nosmt . 263696.39 |============================ nginx 1.21.1 Concurrent Requests: 500 Requests Per Second > Higher Is Better Default ........................... 257344.77 |============================ mitigations=off ................... 299314.90 |================================ retbleed=ibpb ..................... 221507.79 |======================== retbleed=ibpb spectre_v2=on ....... 215408.91 |======================= retbleed=ibpb spectre_v2=on nosmt . 263560.20 |============================ nginx 1.21.1 Concurrent Requests: 100 Requests Per Second > Higher Is Better Default ........................... 255336.64 |=========================== mitigations=off ................... 299015.72 |================================ retbleed=ibpb ..................... 220224.99 |======================== retbleed=ibpb spectre_v2=on ....... 213755.99 |======================= retbleed=ibpb spectre_v2=on nosmt . 267291.25 |============================= Apache HTTP Server 2.4.48 Concurrent Requests: 100 Requests Per Second > Higher Is Better Default ........................... 56712.95 |================================ mitigations=off ................... 57832.11 |================================= retbleed=ibpb ..................... 56503.13 |================================ retbleed=ibpb spectre_v2=on ....... 55347.49 |================================ retbleed=ibpb spectre_v2=on nosmt . 57332.87 |================================= nginx 1.21.1 Concurrent Requests: 200 Requests Per Second > Higher Is Better Default ........................... 256535.31 |=========================== mitigations=off ................... 301149.26 |================================ retbleed=ibpb ..................... 221738.81 |======================== retbleed=ibpb spectre_v2=on ....... 214677.17 |======================= retbleed=ibpb spectre_v2=on nosmt . 265599.52 |============================ nginx 1.21.1 Concurrent Requests: 20 Requests Per Second > Higher Is Better Default ........................... 260435.54 |============================ mitigations=off ................... 293758.51 |================================ retbleed=ibpb ..................... 225554.08 |========================= retbleed=ibpb spectre_v2=on ....... 218530.10 |======================== retbleed=ibpb spectre_v2=on nosmt . 266600.54 |============================= Apache HTTP Server 2.4.48 Concurrent Requests: 20 Requests Per Second > Higher Is Better Default ........................... 61666.41 |=============================== mitigations=off ................... 65653.49 |================================= retbleed=ibpb ..................... 60591.68 |============================== retbleed=ibpb spectre_v2=on ....... 59088.55 |============================== retbleed=ibpb spectre_v2=on nosmt . 47678.26 |======================== Timed Godot Game Engine Compilation 3.2.3 Time To Compile Seconds < Lower Is Better Default ........................... 74.42 |========================= mitigations=off ................... 72.95 |========================= retbleed=ibpb ..................... 86.86 |============================= retbleed=ibpb spectre_v2=on ....... 88.05 |============================== retbleed=ibpb spectre_v2=on nosmt . 103.35 |=================================== InfluxDB 1.8.2 Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 val/sec > Higher Is Better Default ........................... 1239577.0 |=========================== mitigations=off ................... 1246610.8 |============================ retbleed=ibpb ..................... 1274721.6 |============================ retbleed=ibpb spectre_v2=on ....... 1268118.7 |============================ retbleed=ibpb spectre_v2=on nosmt . 1444954.3 |================================ Node.js V8 Web Tooling Benchmark runs/s > Higher Is Better Default ........................... 11.39 |==================================== mitigations=off ................... 11.07 |=================================== retbleed=ibpb ..................... 10.71 |================================== retbleed=ibpb spectre_v2=on ....... 10.71 |================================== retbleed=ibpb spectre_v2=on nosmt . 10.90 |================================== Renaissance 0.14 Test: In-Memory Database Shootout ms < Lower Is Better Default ........................... 4168.1 |========================== mitigations=off ................... 3905.4 |========================= retbleed=ibpb ..................... 5511.6 |=================================== retbleed=ibpb spectre_v2=on ....... 5575.1 |=================================== retbleed=ibpb spectre_v2=on nosmt . 4071.4 |========================== Stress-NG 0.14 Test: Socket Activity Bogo Ops/s > Higher Is Better Default ........................... 8838.08 |=============================== mitigations=off ................... 9564.50 |================================== retbleed=ibpb ..................... 6557.16 |======================= retbleed=ibpb spectre_v2=on ....... 6538.13 |======================= retbleed=ibpb spectre_v2=on nosmt . 4722.43 |================= Hackbench Count: 8 - Type: Thread Seconds < Lower Is Better Default ........................... 17.96 |=========== mitigations=off ................... 17.42 |=========== retbleed=ibpb ..................... 31.87 |==================== retbleed=ibpb spectre_v2=on ....... 35.87 |======================= retbleed=ibpb spectre_v2=on nosmt . 57.14 |==================================== PyPerformance 1.0.0 Benchmark: python_startup Milliseconds < Lower Is Better Default ........................... 7.44 |=========================== mitigations=off ................... 7.17 |========================== retbleed=ibpb ..................... 10.10 |==================================== retbleed=ibpb spectre_v2=on ....... 10.10 |==================================== retbleed=ibpb spectre_v2=on nosmt . 9.91 |=================================== Compile Bench 0.6 Test: Initial Create MB/s > Higher Is Better Default ........................... 514.56 |============================== mitigations=off ................... 606.28 |=================================== retbleed=ibpb ..................... 240.60 |============== retbleed=ibpb spectre_v2=on ....... 242.55 |============== retbleed=ibpb spectre_v2=on nosmt . 235.39 |============== Timed Linux Kernel Compilation 5.18 Build: defconfig Seconds < Lower Is Better Default ........................... 53.67 |======================== mitigations=off ................... 52.38 |======================= retbleed=ibpb ..................... 66.26 |============================= retbleed=ibpb spectre_v2=on ....... 67.04 |============================= retbleed=ibpb spectre_v2=on nosmt . 81.85 |==================================== Hackbench Count: 16 - Type: Thread Seconds < Lower Is Better Default ........................... 25.61 |======== mitigations=off ................... 22.37 |======= retbleed=ibpb ..................... 63.89 |==================== retbleed=ibpb spectre_v2=on ....... 75.32 |======================== retbleed=ibpb spectre_v2=on nosmt . 111.91 |=================================== RawTherapee Total Benchmark Time Seconds < Lower Is Better Default ........................... 54.41 |================================ mitigations=off ................... 53.10 |=============================== retbleed=ibpb ..................... 58.84 |================================== retbleed=ibpb spectre_v2=on ....... 59.15 |=================================== retbleed=ibpb spectre_v2=on nosmt . 61.56 |==================================== Hackbench Count: 16 - Type: Process Seconds < Lower Is Better Default ........................... 25.04 |======== mitigations=off ................... 20.88 |======= retbleed=ibpb ..................... 63.95 |===================== retbleed=ibpb spectre_v2=on ....... 69.19 |====================== retbleed=ibpb spectre_v2=on nosmt . 108.64 |=================================== Stargate Digital Audio Workstation 21.10.9 Sample Rate: 44100 - Buffer Size: 1024 Render Ratio > Higher Is Better Default ........................... 5.615871 |================================= mitigations=off ................... 5.647994 |================================= retbleed=ibpb ..................... 5.324162 |=============================== retbleed=ibpb spectre_v2=on ....... 5.214571 |============================== retbleed=ibpb spectre_v2=on nosmt . 5.518860 |================================ Selenium Benchmark: ARES-6 - Browser: Google Chrome ms < Lower Is Better Default ........................... 15.89 |================================== mitigations=off ................... 15.58 |================================= retbleed=ibpb ..................... 16.79 |==================================== retbleed=ibpb spectre_v2=on ....... 16.44 |=================================== retbleed=ibpb spectre_v2=on nosmt . 14.11 |============================== ASTC Encoder 4.0 Preset: Exhaustive MT/s > Higher Is Better Default ........................... 1.0449 |=================================== mitigations=off ................... 1.0451 |=================================== retbleed=ibpb ..................... 1.0448 |=================================== retbleed=ibpb spectre_v2=on ....... 1.0429 |=================================== retbleed=ibpb spectre_v2=on nosmt . 0.8887 |============================== Selenium Benchmark: Speedometer - Browser: Google Chrome Runs Per Minute > Higher Is Better Default ........................... 212 |====================================== mitigations=off ................... 211 |====================================== retbleed=ibpb ..................... 187 |================================== retbleed=ibpb spectre_v2=on ....... 187 |================================== retbleed=ibpb spectre_v2=on nosmt . 194 |=================================== Selenium Benchmark: Octane - Browser: Google Chrome Geometric Mean > Higher Is Better Default ........................... 58308 |=================================== mitigations=off ................... 59134 |==================================== retbleed=ibpb ..................... 56488 |================================== retbleed=ibpb spectre_v2=on ....... 56525 |================================== retbleed=ibpb spectre_v2=on nosmt . 56246 |================================== DaCapo Benchmark 9.12-MR1 Java Test: Tradesoap msec < Lower Is Better Default ........................... 3504 |================================== mitigations=off ................... 3394 |================================= retbleed=ibpb ..................... 3834 |===================================== retbleed=ibpb spectre_v2=on ....... 3840 |===================================== retbleed=ibpb spectre_v2=on nosmt . 3144 |============================== LibreOffice Test: 20 Documents To PDF Seconds < Lower Is Better Default ........................... 6.915 |=============================== mitigations=off ................... 6.707 |============================== retbleed=ibpb ..................... 7.570 |================================== retbleed=ibpb spectre_v2=on ....... 7.525 |================================== retbleed=ibpb spectre_v2=on nosmt . 7.924 |==================================== PyPerformance 1.0.0 Benchmark: django_template Milliseconds < Lower Is Better Default ........................... 38.0 |===================================== mitigations=off ................... 38.0 |===================================== retbleed=ibpb ..................... 37.0 |==================================== retbleed=ibpb spectre_v2=on ....... 37.1 |==================================== retbleed=ibpb spectre_v2=on nosmt . 38.4 |===================================== Timed Mesa Compilation 21.0 Time To Compile Seconds < Lower Is Better Default ........................... 33.41 |================================ mitigations=off ................... 33.26 |================================ retbleed=ibpb ..................... 36.32 |=================================== retbleed=ibpb spectre_v2=on ....... 36.41 |=================================== retbleed=ibpb spectre_v2=on nosmt . 37.19 |==================================== Selenium Benchmark: PSPDFKit WASM - Browser: Google Chrome Score < Lower Is Better Default ........................... 2891 |================================= mitigations=off ................... 2852 |================================ retbleed=ibpb ..................... 3268 |===================================== retbleed=ibpb spectre_v2=on ....... 3267 |===================================== retbleed=ibpb spectre_v2=on nosmt . 2960 |================================== Stress-NG 0.14 Test: MMAP Bogo Ops/s > Higher Is Better Default ........................... 287.61 |================================== mitigations=off ................... 294.75 |=================================== retbleed=ibpb ..................... 249.86 |============================== retbleed=ibpb spectre_v2=on ....... 230.11 |=========================== retbleed=ibpb spectre_v2=on nosmt . 254.15 |============================== ASTC Encoder 4.0 Preset: Thorough MT/s > Higher Is Better Default ........................... 9.5416 |=================================== mitigations=off ................... 9.5490 |=================================== retbleed=ibpb ..................... 9.5506 |=================================== retbleed=ibpb spectre_v2=on ....... 9.5182 |=================================== retbleed=ibpb spectre_v2=on nosmt . 8.3786 |=============================== Stress-NG 0.14 Test: Malloc Bogo Ops/s > Higher Is Better Default ........................... 23110108.89 |======================== mitigations=off ................... 28438903.40 |============================== retbleed=ibpb ..................... 6167954.31 |======= retbleed=ibpb spectre_v2=on ....... 6039417.12 |====== retbleed=ibpb spectre_v2=on nosmt . 4213740.42 |==== Google SynthMark 20201109 Test: VoiceMark_100 Voices > Higher Is Better Default ........................... 790.98 |================================== mitigations=off ................... 820.38 |=================================== retbleed=ibpb ..................... 782.18 |================================= retbleed=ibpb spectre_v2=on ....... 816.40 |=================================== retbleed=ibpb spectre_v2=on nosmt . 761.09 |================================ Stress-NG 0.14 Test: SENDFILE Bogo Ops/s > Higher Is Better Default ........................... 293235.64 |======================= mitigations=off ................... 412629.17 |================================ retbleed=ibpb ..................... 305358.03 |======================== retbleed=ibpb spectre_v2=on ....... 301936.36 |======================= retbleed=ibpb spectre_v2=on nosmt . 214958.36 |================= Stress-NG 0.14 Test: System V Message Passing Bogo Ops/s > Higher Is Better Default ........................... 9616996.63 |============================== mitigations=off ................... 9836938.88 |=============================== retbleed=ibpb ..................... 6092876.52 |=================== retbleed=ibpb spectre_v2=on ....... 5819643.79 |================== retbleed=ibpb spectre_v2=on nosmt . 4221595.08 |============= PyPerformance 1.0.0 Benchmark: regex_compile Milliseconds < Lower Is Better Default ........................... 151 |===================================== mitigations=off ................... 150 |===================================== retbleed=ibpb ..................... 148 |==================================== retbleed=ibpb spectre_v2=on ....... 147 |==================================== retbleed=ibpb spectre_v2=on nosmt . 155 |====================================== Inkscape Operation: SVG Files To PNG Seconds < Lower Is Better Default ........................... 24.46 |=========================== mitigations=off ................... 23.74 |=========================== retbleed=ibpb ..................... 32.04 |==================================== retbleed=ibpb spectre_v2=on ....... 32.16 |==================================== retbleed=ibpb spectre_v2=on nosmt . 32.16 |==================================== DaCapo Benchmark 9.12-MR1 Java Test: H2 msec < Lower Is Better Default ........................... 3644 |=================================== mitigations=off ................... 3534 |================================= retbleed=ibpb ..................... 3849 |==================================== retbleed=ibpb spectre_v2=on ....... 3907 |===================================== retbleed=ibpb spectre_v2=on nosmt . 2596 |========================= Redis 6.0.9 Test: SET Requests Per Second > Higher Is Better Default ........................... 1853189.96 |============================== mitigations=off ................... 1939841.82 |=============================== retbleed=ibpb ..................... 1374092.29 |====================== retbleed=ibpb spectre_v2=on ....... 1364607.38 |====================== retbleed=ibpb spectre_v2=on nosmt . 1327456.50 |===================== Redis 6.0.9 Test: GET Requests Per Second > Higher Is Better Default ........................... 2587396.87 |=============================== mitigations=off ................... 2495320.75 |============================== retbleed=ibpb ..................... 1627014.00 |=================== retbleed=ibpb spectre_v2=on ....... 1620511.60 |=================== retbleed=ibpb spectre_v2=on nosmt . 1573874.50 |=================== Hackbench Count: 8 - Type: Process Seconds < Lower Is Better Default ........................... 13.25 |========= mitigations=off ................... 11.08 |======== retbleed=ibpb ..................... 31.18 |===================== retbleed=ibpb spectre_v2=on ....... 33.38 |======================= retbleed=ibpb spectre_v2=on nosmt . 52.78 |==================================== LuaJIT 2.1-git Test: Composite Mflops > Higher Is Better Default ........................... 1654.91 |================================= mitigations=off ................... 1726.41 |================================== retbleed=ibpb ..................... 1706.81 |================================== retbleed=ibpb spectre_v2=on ....... 1655.74 |================================= retbleed=ibpb spectre_v2=on nosmt . 1640.69 |================================ ebizzy 0.3 Records/s > Higher Is Better Default ........................... 1413624 |================================== mitigations=off ................... 1409922 |================================== retbleed=ibpb ..................... 1355964 |================================= retbleed=ibpb spectre_v2=on ....... 1369472 |================================= retbleed=ibpb spectre_v2=on nosmt . 1272519 |=============================== PyBench 2018-02-16 Total For Average Test Times Milliseconds < Lower Is Better Default ........................... 953 |===================================== mitigations=off ................... 930 |==================================== retbleed=ibpb ..................... 948 |===================================== retbleed=ibpb spectre_v2=on ....... 917 |==================================== retbleed=ibpb spectre_v2=on nosmt . 977 |====================================== Selenium Benchmark: Kraken - Browser: Google Chrome ms < Lower Is Better Default ........................... 601.1 |================================== mitigations=off ................... 603.9 |================================== retbleed=ibpb ..................... 630.2 |==================================== retbleed=ibpb spectre_v2=on ....... 623.7 |==================================== retbleed=ibpb spectre_v2=on nosmt . 585.0 |================================= Node.js Express HTTP Load Test Requests Per Second > Higher Is Better Default ........................... 6863 |==================================== mitigations=off ................... 7064 |===================================== retbleed=ibpb ..................... 4559 |======================== retbleed=ibpb spectre_v2=on ....... 4532 |======================== retbleed=ibpb spectre_v2=on nosmt . 4543 |======================== GNU Octave Benchmark 6.4.0 Seconds < Lower Is Better Default ........................... 6.970 |============================ mitigations=off ................... 6.730 |=========================== retbleed=ibpb ..................... 8.992 |==================================== retbleed=ibpb spectre_v2=on ....... 9.044 |==================================== retbleed=ibpb spectre_v2=on nosmt . 8.962 |==================================== DaCapo Benchmark 9.12-MR1 Java Test: Tradebeans msec < Lower Is Better Default ........................... 4387 |=============================== mitigations=off ................... 4037 |============================ retbleed=ibpb ..................... 5261 |===================================== retbleed=ibpb spectre_v2=on ....... 5320 |===================================== retbleed=ibpb spectre_v2=on nosmt . 2859 |==================== Selenium Benchmark: WASM collisionDetection - Browser: Google Chrome ms < Lower Is Better Default ........................... 273.73 |================================== mitigations=off ................... 275.62 |================================== retbleed=ibpb ..................... 274.16 |================================== retbleed=ibpb spectre_v2=on ....... 269.21 |================================= retbleed=ibpb spectre_v2=on nosmt . 285.07 |=================================== Primesieve 8.0 Length: 1e12 Seconds < Lower Is Better Default ........................... 10.73 |================================== mitigations=off ................... 10.72 |================================== retbleed=ibpb ..................... 10.74 |================================== retbleed=ibpb spectre_v2=on ....... 10.82 |================================== retbleed=ibpb spectre_v2=on nosmt . 11.35 |==================================== libavif avifenc 0.10 Encoder Speed: 6, Lossless Seconds < Lower Is Better Default ........................... 9.857 |============================= mitigations=off ................... 9.599 |============================ retbleed=ibpb ..................... 10.716 |=============================== retbleed=ibpb spectre_v2=on ....... 10.885 |================================ retbleed=ibpb spectre_v2=on nosmt . 12.031 |=================================== libavif avifenc 0.10 Encoder Speed: 10, Lossless Seconds < Lower Is Better Default ........................... 5.716 |============================== mitigations=off ................... 5.528 |============================= retbleed=ibpb ..................... 6.577 |================================== retbleed=ibpb spectre_v2=on ....... 6.705 |=================================== retbleed=ibpb spectre_v2=on nosmt . 6.932 |==================================== ASTC Encoder 4.0 Preset: Medium MT/s > Higher Is Better Default ........................... 79.60 |==================================== mitigations=off ................... 79.80 |==================================== retbleed=ibpb ..................... 79.69 |==================================== retbleed=ibpb spectre_v2=on ....... 79.47 |==================================== retbleed=ibpb spectre_v2=on nosmt . 69.23 |=============================== SVT-AV1 1.2 Encoder Mode: Preset 10 - Input: Bosphorus 4K Frames Per Second > Higher Is Better Default ........................... 78.58 |=================================== mitigations=off ................... 80.48 |==================================== retbleed=ibpb ..................... 76.92 |================================== retbleed=ibpb spectre_v2=on ....... 77.23 |=================================== retbleed=ibpb spectre_v2=on nosmt . 76.18 |================================== libavif avifenc 0.10 Encoder Speed: 6 Seconds < Lower Is Better Default ........................... 6.658 |============================= mitigations=off ................... 6.399 |============================ retbleed=ibpb ..................... 7.132 |=============================== retbleed=ibpb spectre_v2=on ....... 7.294 |================================ retbleed=ibpb spectre_v2=on nosmt . 8.158 |==================================== SVT-AV1 1.2 Encoder Mode: Preset 12 - Input: Bosphorus 4K Frames Per Second > Higher Is Better Default ........................... 106.99 |=================================== mitigations=off ................... 108.33 |=================================== retbleed=ibpb ..................... 103.60 |================================= retbleed=ibpb spectre_v2=on ....... 102.22 |================================= retbleed=ibpb spectre_v2=on nosmt . 101.99 |================================= Selenium Benchmark: WASM imageConvolute - Browser: Google Chrome ms < Lower Is Better Default ........................... 25.31 |=================================== mitigations=off ................... 24.84 |================================== retbleed=ibpb ..................... 25.75 |=================================== retbleed=ibpb spectre_v2=on ....... 25.54 |=================================== retbleed=ibpb spectre_v2=on nosmt . 26.31 |==================================== SVT-AV1 1.2 Encoder Mode: Preset 12 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better Default ........................... 335.17 |=============================== mitigations=off ................... 343.98 |=============================== retbleed=ibpb ..................... 311.78 |============================ retbleed=ibpb spectre_v2=on ....... 317.13 |============================= retbleed=ibpb spectre_v2=on nosmt . 384.32 |=================================== Darktable 3.8.1 Test: Masskrug - Acceleration: CPU-only Seconds < Lower Is Better Default ........................... 3.693 |================================= mitigations=off ................... 3.612 |================================ retbleed=ibpb ..................... 3.966 |=================================== retbleed=ibpb spectre_v2=on ....... 3.987 |=================================== retbleed=ibpb spectre_v2=on nosmt . 4.061 |==================================== Darktable 3.8.1 Test: Server Room - Acceleration: CPU-only Seconds < Lower Is Better Default ........................... 3.334 |=================================== mitigations=off ................... 3.300 |================================== retbleed=ibpb ..................... 3.422 |==================================== retbleed=ibpb spectre_v2=on ....... 3.448 |==================================== retbleed=ibpb spectre_v2=on nosmt . 3.207 |================================= Darktable 3.8.1 Test: Boat - Acceleration: CPU-only Seconds < Lower Is Better Default ........................... 3.142 |=================================== mitigations=off ................... 3.087 |================================== retbleed=ibpb ..................... 3.260 |==================================== retbleed=ibpb spectre_v2=on ....... 3.270 |==================================== retbleed=ibpb spectre_v2=on nosmt . 3.119 |================================== ctx_clock Context Switch Time Clocks < Lower Is Better Default ........................... 245 |====== mitigations=off ................... 175 |==== retbleed=ibpb ..................... 1435 |================================== retbleed=ibpb spectre_v2=on ....... 1458 |================================== retbleed=ibpb spectre_v2=on nosmt . 1575 |===================================== SVT-AV1 1.2 Encoder Mode: Preset 10 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better Default ........................... 251.82 |================================== mitigations=off ................... 260.32 |=================================== retbleed=ibpb ..................... 238.88 |================================ retbleed=ibpb spectre_v2=on ....... 240.80 |================================ retbleed=ibpb spectre_v2=on nosmt . 247.72 |================================= Darktable 3.8.1 Test: Server Rack - Acceleration: CPU-only Seconds < Lower Is Better Default ........................... 0.225 |================================== mitigations=off ................... 0.223 |================================= retbleed=ibpb ..................... 0.229 |================================== retbleed=ibpb spectre_v2=on ....... 0.230 |================================== retbleed=ibpb spectre_v2=on nosmt . 0.241 |==================================== Compile Bench 0.6 Test: Read Compiled Tree MB/s > Higher Is Better Default ........................... 3463.27 |=============================== mitigations=off ................... 3796.97 |================================== retbleed=ibpb ..................... 1284.64 |============ retbleed=ibpb spectre_v2=on ....... 1300.95 |============ retbleed=ibpb spectre_v2=on nosmt . 1250.47 |=========== Geometric Mean Of All Test Results Result Composite - Ryzen 9 3950X Linux 6.0 Mitigations Benchmark Geometric Mean > Higher Is Better Default ........................... 136.13 |================================= mitigations=off ................... 143.87 |=================================== retbleed=ibpb ..................... 109.14 |=========================== retbleed=ibpb spectre_v2=on ....... 107.19 |========================== retbleed=ibpb spectre_v2=on nosmt . 107.17 |==========================