JDK-8187601 : Unrolling more when SLP auto-vectorization failed
- Type: Enhancement
- Component: hotspot
- Sub-Component: compiler
- Affected Version: 10
- Priority: P4
- Status: Resolved
- Resolution: Fixed
- OS: linux
- CPU: generic
- Submitted: 2017-09-18
- Updated: 2025-09-19
- Resolved: 2017-10-05
The Version table provides details related to the release that this issue/RFE will be addressed.
Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.
To download the current JDK release, click here.
Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.
To download the current JDK release, click here.
| JDK 10 |
|---|
10 b31Fixed  |
Related Reports
|
Relates :
|
|
|
Relates :
|
Description
For below Java code, which can't be auto-vectorized:
public static void accessArrayConstants(int[] array) {
for (int j = 0; j < 1024; j++) {
array[0]++;
array[1]++;
}
}
Current C2 unrolls the loop to 8 times on X86 and 4 times on AArch64, which is limited by architecture's vector size. Get better performance, we should unroll more until reaching the unroll limitation.
Comments
|