March 2017 - aarch64-port-dev

RFR: 8169697: aarch64: vectorized MLA instruction not generated for some test cases
by Ningsheng Jian 04 May '17

04 May '17

Hi, NEON vector MLA instructions can not be generated in some simple multiply-add cases. Bug: https://bugs.openjdk.java.net/browse/JDK-8169697 The following webrev contributed by Yang Zhang (yang.zhang(a)linaro.org) can resolve it. Webrev: http://cr.openjdk.java.net/~njian/8169697/webrev.00/ Please help to review it. To minimize impact, this patch only changes aarch64 specified code. Do you think this is OK for 9? If yes, do we need a FC Extension Request? In the mean time, Yang also has a more generic fix for this issue, which patches share code and could also fix this issue: http://cr.openjdk.java.net/~njian/8169697/webrev.share/ Which do you think is better? Both patches tested with fastdebug jtreg. Thanks, Ningsheng

5 21

aarch64 and arm64 jdk benchmarks result sharing
by Zhongwei Yao 21 Apr '17

21 Apr '17

Hi, We have been running some Java benchmarks on AArch64 servers with different OpenJDK ports: aarch64 and arm64. You can find the results at [1]. And our JDK version: - aarch64 OpenJDK: build from latest JDK9 source with config: "--with-jvm-variants=server --with-debug-level=release" - arm64 OpenJDK: build from latest JDK9 source with config: "--with-jvm-variants=server --with-debug-level=release --wi th-cpu-port=arm64" And from the data, aarch64 OpenJDK is better than arm64 OpenJDK generally, but with some exceptions. So we think it is worth to take a look at following cases for aarch64 OpenJDK: - Dacapo's sunflow, tradebeans and xalan - SPECjvm2008's scimark.lu.large And we also find arm64 is better than aarch64 for byte/short data type vectorization in our JMH cases ([2]), which we are investigating. Feedbacks are welcome. Thanks! [1]: https://docs.google.com/spreadsheets/d/18iklOrbaL67i46XHsPTrdTObWqUJJK8EO4-… [2]: https://git.linaro.org/leg/openjdk/jmh-linaro-org.git/ -- Best regards, Zhongwei

6 26

jdk8: backport of 8177661: AArch64: Incorrect C2 patterns cause system register corruption
by Andrew Dinn 19 Apr '17

19 Apr '17

The patch included inline below is against the latest aarch64-port-dev/jdk8u tree. It backports the upstream jdk9 fix for JDK-8177661 fixing a critical potential crasher bug. It has been tested by running java Hello, javac Hello.java and netbeans. Ok, to push? regards, Andrew Dinn ----------- Senior Principal Software Engineer Red Hat UK Ltd Registered in England and Wales under Company Registration No. 03798903 Directors: Michael Cunningham, Michael ("Mike") O'Neill, Eric Shander ----- 8<-------- 8<-------- 8<-------- 8<-------- 8<-------- 8<--- # HG changeset patch # User adinn # Date 1490695969 14400 # Tue Mar 28 06:12:49 2017 -0400 # Node ID 142389cae1d354241d0268f610fda174d009e596 # Parent daf232a19060305db8af63c118ae3e66f8670468 Correct ad rule output register types from iRegX to iRegXNoSp Backport to jdk8 of JDK9 fix for JDK-8177661 diff -r daf232a19060 -r 142389cae1d3 src/cpu/aarch64/vm/aarch64.ad --- a/src/cpu/aarch64/vm/aarch64.ad Mon Jan 23 21:28:30 2017 +0000 +++ b/src/cpu/aarch64/vm/aarch64.ad Tue Mar 28 06:12:49 2017 -0400 @@ -15152,7 +15152,7 @@ %} instruct string_indexof(iRegP_R1 str1, iRegI_R4 cnt1, iRegP_R3 str2, iRegI_R2 cnt2, - iRegI_R0 result, iRegI tmp1, iRegI tmp2, iRegI tmp3, iRegI tmp4, rFlagsReg cr) + iRegI_R0 result, iRegINoSp tmp1, iRegINoSp tmp2, iRegINoSp tmp3, iRegINoSp tmp4, rFlagsReg cr) %{ match(Set result (StrIndexOf (Binary str1 cnt1) (Binary str2 cnt2))); effect(USE_KILL str1, USE_KILL str2, USE_KILL cnt1, USE_KILL cnt2, @@ -15170,8 +15170,8 @@ %} instruct string_indexof_con(iRegP_R1 str1, iRegI_R4 cnt1, iRegP_R3 str2, - immI_le_4 int_cnt2, iRegI_R0 result, iRegI tmp1, iRegI tmp2, - iRegI tmp3, iRegI tmp4, rFlagsReg cr) + immI_le_4 int_cnt2, iRegI_R0 result, iRegINoSp tmp1, iRegINoSp tmp2, + iRegINoSp tmp3, iRegINoSp tmp4, rFlagsReg cr) %{ match(Set result (StrIndexOf (Binary str1 cnt1) (Binary str2 int_cnt2))); effect(USE_KILL str1, USE_KILL str2, USE_KILL cnt1,

3 8

JTREG, JCStress, SPECjbb2015 and Hadoop/Terasort results for OpenJDK 9 on AArch64
by ci_notify＠linaro.org 31 Mar '17

31 Mar '17

This is a summary of the JTREG test results =========================================== The build and test results are cycled every 15 days. For detailed information on the test output please refer to: http://openjdk.linaro.org/jdk9/openjdk-jtreg-nightly-tests/summary/2017/089… ------------------------------------------------------------------------------- client-release/hotspot ------------------------------------------------------------------------------- Build 0: aarch64/2017/feb/25 pass: 1,386; fail: 5 Build 1: aarch64/2017/feb/28 pass: 1,386; fail: 5 Build 2: aarch64/2017/mar/01 pass: 1,387; fail: 5 Build 3: aarch64/2017/mar/02 pass: 1,387; fail: 5 Build 4: aarch64/2017/mar/03 pass: 1,387; fail: 5 Build 5: aarch64/2017/mar/08 pass: 1,388; fail: 5 Build 6: aarch64/2017/mar/11 pass: 1,388; fail: 5 Build 7: aarch64/2017/mar/13 pass: 1,388; fail: 5 Build 8: aarch64/2017/mar/15 pass: 1,387; fail: 6 Build 9: aarch64/2017/mar/17 pass: 1,390; fail: 6 Build 10: aarch64/2017/mar/19 pass: 1,391; fail: 5 Build 11: aarch64/2017/mar/21 pass: 1,391; fail: 5 Build 12: aarch64/2017/mar/25 pass: 1,393; fail: 4 Build 13: aarch64/2017/mar/27 pass: 1,393; fail: 4 Build 14: aarch64/2017/mar/30 pass: 1,393; fail: 4 ------------------------------------------------------------------------------- client-release/jdk ------------------------------------------------------------------------------- Build 0: aarch64/2017/feb/25 pass: 7,266; fail: 705; error: 20 Build 1: aarch64/2017/feb/28 pass: 7,274; fail: 695; error: 23 Build 2: aarch64/2017/mar/01 pass: 7,264; fail: 711; error: 17 Build 3: aarch64/2017/mar/02 pass: 7,272; fail: 705; error: 15 Build 4: aarch64/2017/mar/03 pass: 7,252; fail: 720; error: 21 Build 5: aarch64/2017/mar/08 pass: 7,251; fail: 720; error: 23 Build 6: aarch64/2017/mar/11 pass: 7,263; fail: 718; error: 18 Build 7: aarch64/2017/mar/13 pass: 7,268; fail: 710; error: 21 Build 8: aarch64/2017/mar/15 pass: 7,283; fail: 693; error: 25 Build 9: aarch64/2017/mar/17 pass: 7,273; fail: 709; error: 20 Build 10: aarch64/2017/mar/19 pass: 7,253; fail: 727; error: 24 Build 11: aarch64/2017/mar/21 pass: 7,294; fail: 689; error: 21 Build 12: aarch64/2017/mar/25 pass: 7,278; fail: 713; error: 23 Build 13: aarch64/2017/mar/27 pass: 7,304; fail: 690; error: 20 Build 14: aarch64/2017/mar/30 pass: 7,302; fail: 691; error: 23 ------------------------------------------------------------------------------- client-release/langtools ------------------------------------------------------------------------------- Build 0: aarch64/2017/feb/25 pass: 3,859; error: 4 Build 1: aarch64/2017/feb/28 pass: 3,861; error: 2 Build 2: aarch64/2017/mar/01 pass: 3,862; error: 2 Build 3: aarch64/2017/mar/02 pass: 3,862; error: 2 Build 4: aarch64/2017/mar/03 pass: 3,863; error: 2 Build 5: aarch64/2017/mar/08 pass: 3,866; error: 1 Build 6: aarch64/2017/mar/11 pass: 3,871; error: 2 Build 7: aarch64/2017/mar/13 pass: 3,872; error: 1 Build 8: aarch64/2017/mar/15 pass: 3,874; error: 4 Build 9: aarch64/2017/mar/17 pass: 3,878; error: 2 Build 10: aarch64/2017/mar/19 pass: 3,878; error: 2 Build 11: aarch64/2017/mar/21 pass: 3,876; error: 5 Build 12: aarch64/2017/mar/25 pass: 3,886; fail: 1 Build 13: aarch64/2017/mar/27 pass: 3,883; error: 4 Build 14: aarch64/2017/mar/30 pass: 3,883; fail: 3; error: 1 ------------------------------------------------------------------------------- server-release/hotspot ------------------------------------------------------------------------------- Build 0: aarch64/2017/feb/25 pass: 1,389; fail: 5 Build 1: aarch64/2017/feb/28 pass: 1,388; fail: 6 Build 2: aarch64/2017/mar/01 pass: 1,390; fail: 5 Build 3: aarch64/2017/mar/02 pass: 1,390; fail: 5 Build 4: aarch64/2017/mar/03 pass: 1,390; fail: 5 Build 5: aarch64/2017/mar/08 pass: 1,391; fail: 6 Build 6: aarch64/2017/mar/11 pass: 1,392; fail: 5 Build 7: aarch64/2017/mar/13 pass: 1,392; fail: 5 Build 8: aarch64/2017/mar/15 pass: 1,392; fail: 5 Build 9: aarch64/2017/mar/17 pass: 1,395; fail: 5 Build 10: aarch64/2017/mar/19 pass: 1,395; fail: 5 Build 11: aarch64/2017/mar/21 pass: 1,394; fail: 6 Build 12: aarch64/2017/mar/25 pass: 1,397; fail: 4 Build 13: aarch64/2017/mar/27 pass: 1,396; fail: 5 Build 14: aarch64/2017/mar/30 pass: 1,397; fail: 4 ------------------------------------------------------------------------------- server-release/jdk ------------------------------------------------------------------------------- Build 0: aarch64/2017/feb/25 pass: 7,279; fail: 694; error: 18 Build 1: aarch64/2017/feb/28 pass: 7,265; fail: 708; error: 19 Build 2: aarch64/2017/mar/01 pass: 7,266; fail: 710; error: 16 Build 3: aarch64/2017/mar/02 pass: 7,284; fail: 689; error: 19 Build 4: aarch64/2017/mar/03 pass: 7,257; fail: 720; error: 16 Build 5: aarch64/2017/mar/08 pass: 7,296; fail: 679; error: 19 Build 6: aarch64/2017/mar/11 pass: 7,261; fail: 722; error: 16 Build 7: aarch64/2017/mar/13 pass: 7,290; fail: 687; error: 22 Build 8: aarch64/2017/mar/15 pass: 7,285; fail: 697; error: 19 Build 9: aarch64/2017/mar/17 pass: 7,278; fail: 703; error: 21 Build 10: aarch64/2017/mar/19 pass: 7,261; fail: 723; error: 20 Build 11: aarch64/2017/mar/21 pass: 7,271; fail: 712; error: 21 Build 12: aarch64/2017/mar/25 pass: 7,284; fail: 708; error: 22 Build 13: aarch64/2017/mar/27 pass: 7,297; fail: 695; error: 22 Build 14: aarch64/2017/mar/30 pass: 7,294; fail: 701; error: 21 ------------------------------------------------------------------------------- server-release/langtools ------------------------------------------------------------------------------- Build 0: aarch64/2017/feb/25 pass: 3,860; error: 3 Build 1: aarch64/2017/feb/28 pass: 3,862; error: 1 Build 2: aarch64/2017/mar/01 pass: 3,860; error: 4 Build 3: aarch64/2017/mar/02 pass: 3,858; fail: 1; error: 5 Build 4: aarch64/2017/mar/03 pass: 3,863; fail: 1; error: 2 Build 5: aarch64/2017/mar/08 pass: 3,867 Build 6: aarch64/2017/mar/11 pass: 3,873 Build 7: aarch64/2017/mar/13 pass: 3,871; error: 2 Build 8: aarch64/2017/mar/15 pass: 3,878 Build 9: aarch64/2017/mar/17 pass: 3,876; error: 4 Build 10: aarch64/2017/mar/19 pass: 3,876; error: 4 Build 11: aarch64/2017/mar/21 pass: 3,879; error: 2 Build 12: aarch64/2017/mar/25 pass: 3,885; error: 2 Build 13: aarch64/2017/mar/27 pass: 3,881; error: 6 Build 14: aarch64/2017/mar/30 pass: 3,879; fail: 2; error: 6 Previous results can be found here: http://openjdk.linaro.org/jdk9/openjdk-jtreg-nightly-tests/index.html SPECjbb2015 composite regression test completed =============================================== This test measures the relative performance of the server compiler running the SPECjbb2015 composite tests and compares the performance against the baseline performance of the server compiler taken on 2016-11-21. In accordance with [1], the SPECjbb2015 tests are run on a system which is not production ready and does not meet all the requirements for publishing compliant results. The numbers below shall be treated as non-compliant (nc) and are for experimental purposes only. Relative performance: Server max-jOPS (nc): 1.02x Relative performance: Server critical-jOPS (nc): 0.89x Details of the test setup and historical results may be found here: http://openjdk.linaro.org/jdk9/SPECjbb2015-results/ [1] http://www.spec.org/fairuse.html#Academic Regression test Hadoop-Terasort completed ========================================= This test measures the performance of the server and client compilers running Hadoop sorting a 1GB file using Terasort and compares the performance against the baseline performance of the Zero interpreter and against the baseline performance of the client and server compilers on 2014-04-01. Relative performance: Zero: 1.0, Client: 72.38, Server: 119.62 Client 72.38 / Client 2014-04-01 (43.00): 1.68x Server 119.62 / Server 2014-04-01 (71.00): 1.68x Details of the test setup and historical results may be found here: http://openjdk.linaro.org/jdk9/hadoop-terasort-benchmark-results/ This is a summary of the jcstress test results ============================================== The build and test results are cycled every 15 days. 2017-02-26 pass rate: 6300/6302, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/056/results/ 2017-03-01 pass rate: 6300/6303, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/059/results/ 2017-03-02 pass rate: 10525/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/060/results/ 2017-03-03 pass rate: 10525/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/061/results/ 2017-03-05 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/062/results/ 2017-03-09 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/067/results/ 2017-03-12 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/070/results/ 2017-03-14 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/072/results/ 2017-03-16 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/074/results/ 2017-03-18 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/076/results/ 2017-03-20 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/078/results/ 2017-03-22 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/080/results/ 2017-03-26 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/084/results/ 2017-03-28 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/086/results/ 2017-03-31 pass rate: 11552/11554, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/089/results/ For detailed information on the test output please refer to: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/

1 0

JTREG, JCStress, SPECjbb2015 and Hadoop/Terasort results for OpenJDK 8u on AArch64
by ci_notify＠linaro.org 29 Mar '17

29 Mar '17

This is a summary of the JTREG test results =========================================== The build and test results are cycled every 15 days. For detailed information on the test output please refer to: http://openjdk.linaro.org/jdk8u/openjdk-jtreg-nightly-tests/summary/2017/08… ------------------------------------------------------------------------------- client-release/hotspot ------------------------------------------------------------------------------- Build 0: aarch64/2016/dec/01 pass: 668; fail: 44; error: 6 Build 1: aarch64/2016/dec/21 pass: 668; fail: 44; error: 6 Build 2: aarch64/2017/jan/18 pass: 672; fail: 44; error: 3 Build 3: aarch64/2017/feb/04 pass: 672; fail: 44; error: 3 Build 4: aarch64/2017/feb/24 pass: 672; fail: 44; error: 3 Build 5: aarch64/2017/mar/29 pass: 672; fail: 44; error: 3 ------------------------------------------------------------------------------- client-release/jdk ------------------------------------------------------------------------------- Build 0: aarch64/2016/dec/21 pass: 5,621; fail: 219; error: 45 Build 1: aarch64/2017/jan/18 pass: 5,683; fail: 213; error: 36 Build 2: aarch64/2017/feb/04 pass: 5,667; fail: 234; error: 41 Build 3: aarch64/2017/feb/24 pass: 5,701; fail: 217; error: 27 Build 4: aarch64/2017/mar/29 pass: 5,686; fail: 225; error: 34 ------------------------------------------------------------------------------- client-release/langtools ------------------------------------------------------------------------------- Build 0: aarch64/2016/dec/01 pass: 3,091; error: 16 Build 1: aarch64/2016/dec/21 pass: 3,096; error: 11 Build 2: aarch64/2017/jan/18 pass: 3,098; error: 13 Build 3: aarch64/2017/feb/04 pass: 3,094; error: 17 Build 4: aarch64/2017/feb/24 pass: 3,106; error: 6 Build 5: aarch64/2017/mar/29 pass: 3,105; fail: 2; error: 5 ------------------------------------------------------------------------------- server-release/hotspot ------------------------------------------------------------------------------- Build 0: aarch64/2016/nov/03 pass: 664; fail: 44; error: 6 Build 1: aarch64/2016/nov/08 pass: 664; fail: 44; error: 6 Build 2: aarch64/2016/nov/09 pass: 664; fail: 44; error: 6 Build 3: aarch64/2016/nov/21 pass: 668; fail: 44; error: 6 Build 4: aarch64/2016/dec/01 pass: 669; fail: 43; error: 6 Build 5: aarch64/2016/dec/21 pass: 668; fail: 44; error: 6 Build 6: aarch64/2017/jan/18 pass: 673; fail: 43; error: 3 Build 7: aarch64/2017/feb/04 pass: 673; fail: 43; error: 3 Build 8: aarch64/2017/feb/24 pass: 673; fail: 43; error: 3 Build 9: aarch64/2017/mar/29 pass: 673; fail: 43; error: 3 ------------------------------------------------------------------------------- server-release/jdk ------------------------------------------------------------------------------- Build 0: aarch64/2016/dec/21 pass: 5,618; fail: 226; error: 41 Build 1: aarch64/2017/jan/18 pass: 5,690; fail: 206; error: 36 Build 2: aarch64/2017/feb/04 pass: 5,669; fail: 214; error: 59 Build 3: aarch64/2017/feb/24 pass: 5,701; fail: 221; error: 23 Build 4: aarch64/2017/mar/29 pass: 5,696; fail: 222; error: 27 ------------------------------------------------------------------------------- server-release/langtools ------------------------------------------------------------------------------- Build 0: aarch64/2016/nov/03 pass: 3,092; error: 15 Build 1: aarch64/2016/nov/08 pass: 3,092; error: 15 Build 2: aarch64/2016/nov/09 pass: 3,091; error: 16 Build 3: aarch64/2016/nov/21 pass: 3,095; error: 12 Build 4: aarch64/2016/dec/01 pass: 3,095; error: 12 Build 5: aarch64/2016/dec/21 pass: 3,092; error: 15 Build 6: aarch64/2017/jan/18 pass: 3,102; error: 9 Build 7: aarch64/2017/feb/04 pass: 3,099; error: 12 Build 8: aarch64/2017/feb/24 pass: 3,109; error: 3 Build 9: aarch64/2017/mar/29 pass: 3,104; fail: 2; error: 6 Previous results can be found here: http://openjdk.linaro.org/jdk8u/openjdk-jtreg-nightly-tests/index.html SPECjbb2015 composite regression test completed =============================================== This test measures the relative performance of the server compiler running the SPECjbb2015 composite tests and compares the performance against the baseline performance of the server compiler taken on 2016-11-21. In accordance with [1], the SPECjbb2015 tests are run on a system which is not production ready and does not meet all the requirements for publishing compliant results. The numbers below shall be treated as non-compliant (nc) and are for experimental purposes only. Relative performance: Server max-jOPS (nc): 0.97x Relative performance: Server critical-jOPS (nc): 0.97x Details of the test setup and historical results may be found here: http://openjdk.linaro.org/jdk8u/SPECjbb2015-results/ [1] http://www.spec.org/fairuse.html#Academic Regression test Hadoop-Terasort completed ========================================= This test measures the performance of the server and client compilers running Hadoop sorting a 1GB file using Terasort and compares the performance against the baseline performance of the Zero interpreter and against the baseline performance of the client and server compilers on 2014-04-01. Relative performance: Zero: 1.0, Client: 57.61, Server: 108.58 Client 57.61 / Client 2014-04-01 (43.00): 1.34x Server 108.58 / Server 2014-04-01 (71.00): 1.53x Details of the test setup and historical results may be found here: http://openjdk.linaro.org/jdk8u/hadoop-terasort-benchmark-results/ This is a summary of the jcstress test results ============================================== The build and test results are cycled every 15 days. 2016-11-03 pass rate: 5140/5140, results: http://openjdk.linaro.org/jdk8u/jcstress-nightly-runs/2016/308/results/ 2016-11-21 pass rate: 5140/5140, results: http://openjdk.linaro.org/jdk8u/jcstress-nightly-runs/2016/326/results/ 2016-12-01 pass rate: 5140/5140, results: http://openjdk.linaro.org/jdk8u/jcstress-nightly-runs/2016/336/results/ 2016-12-22 pass rate: 5140/5140, results: http://openjdk.linaro.org/jdk8u/jcstress-nightly-runs/2016/356/results/ 2017-01-18 pass rate: 5140/5140, results: http://openjdk.linaro.org/jdk8u/jcstress-nightly-runs/2017/018/results/ 2017-02-06 pass rate: 5140/5140, results: http://openjdk.linaro.org/jdk8u/jcstress-nightly-runs/2017/035/results/ 2017-02-25 pass rate: 5176/5176, results: http://openjdk.linaro.org/jdk8u/jcstress-nightly-runs/2017/055/results/ 2017-03-29 pass rate: 8484/8485, results: http://openjdk.linaro.org/jdk8u/jcstress-nightly-runs/2017/088/results/ For detailed information on the test output please refer to: http://openjdk.linaro.org/jdk8u/jcstress-nightly-runs/

1 0

JTREG, JCStress, SPECjbb2015 and Hadoop/Terasort results for OpenJDK 9 on AArch64
by ci_notify＠linaro.org 29 Mar '17

29 Mar '17

This is a summary of the JTREG test results =========================================== The build and test results are cycled every 15 days. For detailed information on the test output please refer to: http://openjdk.linaro.org/jdk9/openjdk-jtreg-nightly-tests/summary/2017/086… ------------------------------------------------------------------------------- client-release/hotspot ------------------------------------------------------------------------------- Build 0: aarch64/2017/feb/22 pass: 1,378; fail: 4 Build 1: aarch64/2017/feb/25 pass: 1,386; fail: 5 Build 2: aarch64/2017/feb/28 pass: 1,386; fail: 5 Build 3: aarch64/2017/mar/01 pass: 1,387; fail: 5 Build 4: aarch64/2017/mar/02 pass: 1,387; fail: 5 Build 5: aarch64/2017/mar/03 pass: 1,387; fail: 5 Build 6: aarch64/2017/mar/08 pass: 1,388; fail: 5 Build 7: aarch64/2017/mar/11 pass: 1,388; fail: 5 Build 8: aarch64/2017/mar/13 pass: 1,388; fail: 5 Build 9: aarch64/2017/mar/15 pass: 1,387; fail: 6 Build 10: aarch64/2017/mar/17 pass: 1,390; fail: 6 Build 11: aarch64/2017/mar/19 pass: 1,391; fail: 5 Build 12: aarch64/2017/mar/21 pass: 1,391; fail: 5 Build 13: aarch64/2017/mar/25 pass: 1,393; fail: 4 Build 14: aarch64/2017/mar/27 pass: 1,393; fail: 4 ------------------------------------------------------------------------------- client-release/jdk ------------------------------------------------------------------------------- Build 0: aarch64/2017/feb/22 pass: 7,226; fail: 714; error: 19 Build 1: aarch64/2017/feb/25 pass: 7,266; fail: 705; error: 20 Build 2: aarch64/2017/feb/28 pass: 7,274; fail: 695; error: 23 Build 3: aarch64/2017/mar/01 pass: 7,264; fail: 711; error: 17 Build 4: aarch64/2017/mar/02 pass: 7,272; fail: 705; error: 15 Build 5: aarch64/2017/mar/03 pass: 7,252; fail: 720; error: 21 Build 6: aarch64/2017/mar/08 pass: 7,251; fail: 720; error: 23 Build 7: aarch64/2017/mar/11 pass: 7,263; fail: 718; error: 18 Build 8: aarch64/2017/mar/13 pass: 7,268; fail: 710; error: 21 Build 9: aarch64/2017/mar/15 pass: 7,283; fail: 693; error: 25 Build 10: aarch64/2017/mar/17 pass: 7,273; fail: 709; error: 20 Build 11: aarch64/2017/mar/19 pass: 7,253; fail: 727; error: 24 Build 12: aarch64/2017/mar/21 pass: 7,294; fail: 689; error: 21 Build 13: aarch64/2017/mar/25 pass: 7,278; fail: 713; error: 23 Build 14: aarch64/2017/mar/27 pass: 7,304; fail: 690; error: 20 ------------------------------------------------------------------------------- client-release/langtools ------------------------------------------------------------------------------- Build 0: aarch64/2017/feb/22 pass: 3,859; error: 2 Build 1: aarch64/2017/feb/25 pass: 3,859; error: 4 Build 2: aarch64/2017/feb/28 pass: 3,861; error: 2 Build 3: aarch64/2017/mar/01 pass: 3,862; error: 2 Build 4: aarch64/2017/mar/02 pass: 3,862; error: 2 Build 5: aarch64/2017/mar/03 pass: 3,863; error: 2 Build 6: aarch64/2017/mar/08 pass: 3,866; error: 1 Build 7: aarch64/2017/mar/11 pass: 3,871; error: 2 Build 8: aarch64/2017/mar/13 pass: 3,872; error: 1 Build 9: aarch64/2017/mar/15 pass: 3,874; error: 4 Build 10: aarch64/2017/mar/17 pass: 3,878; error: 2 Build 11: aarch64/2017/mar/19 pass: 3,878; error: 2 Build 12: aarch64/2017/mar/21 pass: 3,876; error: 5 Build 13: aarch64/2017/mar/25 pass: 3,886; fail: 1 Build 14: aarch64/2017/mar/27 pass: 3,883; error: 4 ------------------------------------------------------------------------------- server-release/hotspot ------------------------------------------------------------------------------- Build 0: aarch64/2017/feb/22 pass: 1,378; fail: 6; error: 1 Build 1: aarch64/2017/feb/25 pass: 1,389; fail: 5 Build 2: aarch64/2017/feb/28 pass: 1,388; fail: 6 Build 3: aarch64/2017/mar/01 pass: 1,390; fail: 5 Build 4: aarch64/2017/mar/02 pass: 1,390; fail: 5 Build 5: aarch64/2017/mar/03 pass: 1,390; fail: 5 Build 6: aarch64/2017/mar/08 pass: 1,391; fail: 6 Build 7: aarch64/2017/mar/11 pass: 1,392; fail: 5 Build 8: aarch64/2017/mar/13 pass: 1,392; fail: 5 Build 9: aarch64/2017/mar/15 pass: 1,392; fail: 5 Build 10: aarch64/2017/mar/17 pass: 1,395; fail: 5 Build 11: aarch64/2017/mar/19 pass: 1,395; fail: 5 Build 12: aarch64/2017/mar/21 pass: 1,394; fail: 6 Build 13: aarch64/2017/mar/25 pass: 1,397; fail: 4 Build 14: aarch64/2017/mar/27 pass: 1,396; fail: 5 ------------------------------------------------------------------------------- server-release/jdk ------------------------------------------------------------------------------- Build 0: aarch64/2017/feb/22 pass: 7,235; fail: 707; error: 17 Build 1: aarch64/2017/feb/25 pass: 7,279; fail: 694; error: 18 Build 2: aarch64/2017/feb/28 pass: 7,265; fail: 708; error: 19 Build 3: aarch64/2017/mar/01 pass: 7,266; fail: 710; error: 16 Build 4: aarch64/2017/mar/02 pass: 7,284; fail: 689; error: 19 Build 5: aarch64/2017/mar/03 pass: 7,257; fail: 720; error: 16 Build 6: aarch64/2017/mar/08 pass: 7,296; fail: 679; error: 19 Build 7: aarch64/2017/mar/11 pass: 7,261; fail: 722; error: 16 Build 8: aarch64/2017/mar/13 pass: 7,290; fail: 687; error: 22 Build 9: aarch64/2017/mar/15 pass: 7,285; fail: 697; error: 19 Build 10: aarch64/2017/mar/17 pass: 7,278; fail: 703; error: 21 Build 11: aarch64/2017/mar/19 pass: 7,261; fail: 723; error: 20 Build 12: aarch64/2017/mar/21 pass: 7,271; fail: 712; error: 21 Build 13: aarch64/2017/mar/25 pass: 7,284; fail: 708; error: 22 Build 14: aarch64/2017/mar/27 pass: 7,297; fail: 695; error: 22 ------------------------------------------------------------------------------- server-release/langtools ------------------------------------------------------------------------------- Build 0: aarch64/2017/feb/22 pass: 3,858; error: 3 Build 1: aarch64/2017/feb/25 pass: 3,860; error: 3 Build 2: aarch64/2017/feb/28 pass: 3,862; error: 1 Build 3: aarch64/2017/mar/01 pass: 3,860; error: 4 Build 4: aarch64/2017/mar/02 pass: 3,858; fail: 1; error: 5 Build 5: aarch64/2017/mar/03 pass: 3,863; fail: 1; error: 2 Build 6: aarch64/2017/mar/08 pass: 3,867 Build 7: aarch64/2017/mar/11 pass: 3,873 Build 8: aarch64/2017/mar/13 pass: 3,871; error: 2 Build 9: aarch64/2017/mar/15 pass: 3,878 Build 10: aarch64/2017/mar/17 pass: 3,876; error: 4 Build 11: aarch64/2017/mar/19 pass: 3,876; error: 4 Build 12: aarch64/2017/mar/21 pass: 3,879; error: 2 Build 13: aarch64/2017/mar/25 pass: 3,885; error: 2 Build 14: aarch64/2017/mar/27 pass: 3,881; error: 6 Previous results can be found here: http://openjdk.linaro.org/jdk9/openjdk-jtreg-nightly-tests/index.html SPECjbb2015 composite regression test completed =============================================== This test measures the relative performance of the server compiler running the SPECjbb2015 composite tests and compares the performance against the baseline performance of the server compiler taken on 2016-11-21. In accordance with [1], the SPECjbb2015 tests are run on a system which is not production ready and does not meet all the requirements for publishing compliant results. The numbers below shall be treated as non-compliant (nc) and are for experimental purposes only. Relative performance: Server max-jOPS (nc): 1.02x Relative performance: Server critical-jOPS (nc): 0.88x Details of the test setup and historical results may be found here: http://openjdk.linaro.org/jdk9/SPECjbb2015-results/ [1] http://www.spec.org/fairuse.html#Academic Regression test Hadoop-Terasort completed ========================================= This test measures the performance of the server and client compilers running Hadoop sorting a 1GB file using Terasort and compares the performance against the baseline performance of the Zero interpreter and against the baseline performance of the client and server compilers on 2014-04-01. Relative performance: Zero: 1.0, Client: 71.29, Server: 114.76 Client 71.29 / Client 2014-04-01 (43.00): 1.66x Server 114.76 / Server 2014-04-01 (71.00): 1.62x Details of the test setup and historical results may be found here: http://openjdk.linaro.org/jdk9/hadoop-terasort-benchmark-results/ This is a summary of the jcstress test results ============================================== The build and test results are cycled every 15 days. 2017-02-23 pass rate: 6050/6050, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/053/results/ 2017-02-26 pass rate: 6300/6302, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/056/results/ 2017-03-01 pass rate: 6300/6303, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/059/results/ 2017-03-02 pass rate: 10525/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/060/results/ 2017-03-03 pass rate: 10525/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/061/results/ 2017-03-05 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/062/results/ 2017-03-09 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/067/results/ 2017-03-12 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/070/results/ 2017-03-14 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/072/results/ 2017-03-16 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/074/results/ 2017-03-18 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/076/results/ 2017-03-20 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/078/results/ 2017-03-22 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/080/results/ 2017-03-26 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/084/results/ 2017-03-28 pass rate: 10524/11284, results: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/2017/086/results/ For detailed information on the test output please refer to: http://openjdk.linaro.org/jdk9/jcstress-nightly-runs/

3 2

hg: aarch64-port/jdk8u-shenandoah/hotspot: Correct ad rule output register types from iRegX to iRegXNoSp
by adinn＠redhat.com 28 Mar '17

28 Mar '17

Changeset: e22d85bf039e Author: adinn Date: 2017-03-28 06:12 -0400 URL: http://hg.openjdk.java.net/aarch64-port/jdk8u-shenandoah/hotspot/rev/e22d85… Correct ad rule output register types from iRegX to iRegXNoSp Backport to jdk8 of JDK9 fix for JDK-8177661 ! src/cpu/aarch64/vm/aarch64.ad

1 0

Notations and reliability
by Andrew Haley 28 Mar '17

28 Mar '17

We've come a cropper a couple of times because of poorly-chosen notations. Several times we've had a confusion between iRegI and iRegINoSp, which causes very obscure crashes. Also, a little while ago I found several bugs (all mine) caused by getting mixed up between e.g. strw and str. I now believe that I made a mistake when I assumed that Xwords (i.e. 64 bits) were in some way the "default" size on AArch64 and omitted the X suffix on them. In hindsight, it is much better for every operation to have a size associated with it, so the programmer would be forced to choose which one to use, and state it explicitly. This is how the "standard" assembly language works, and it is better than what we do. [Interestingly, the Oracle-authored AArch64 port made exactly the same (poor, IMO) choice. Perhaps programmers' brains are isomorphic.] With regard to register classes, GCC uses "GENERAL_REGS" as the name of the register class that can be allocated as a general-purpose operands, with "ALL_REGS" for everything. It would be better we did something similar. Andrew.

2 2

hg: aarch64-port/jdk8u/hotspot: Correct ad rule output register types from iRegX to iRegXNoSp
by adinn＠redhat.com 28 Mar '17

28 Mar '17

Changeset: 142389cae1d3 Author: adinn Date: 2017-03-28 06:12 -0400 URL: http://hg.openjdk.java.net/aarch64-port/jdk8u/hotspot/rev/142389cae1d3 Correct ad rule output register types from iRegX to iRegXNoSp Backport to jdk8 of JDK9 fix for JDK-8177661 ! src/cpu/aarch64/vm/aarch64.ad

1 0

SIGSEGV in aarch64 OpenJDK8u
by Ningsheng Jian 27 Mar '17

27 Mar '17

Hi, We got a SIGSEGV while building Apache Ambari and Apache BigTop using OpenJDK8u on AArch64 platform. The segfault happens while running Apache FindBugs. I have extracted the test packages and shared them at (with hs_err log): http://people.linaro.org/~ningsheng.jian/test/findbugs-test.tar.gz Just run test.sh to reproduce this issue. However, the test is large and I am not able to reduce it. This issue only happens on AArch64 OpenJDK8u, i.e. not on JDK9 or Oracle JDK8u. After some bisect work, I found that the failure exists since this patch: http://hg.openjdk.java.net/aarch64-port/jdk8u/hotspot/rev/f9b6277551dc Reverting this patch makes the test passed. However the patch looks good to me (though it is OpenJDK8u only patch) and I don't think it is the direct cause for the issue. Could you please help to look at it and provide some hints or even a fix? log: ----- # A fatal error has been detected by the Java Runtime Environment: # # SIGSEGV (0xb) at pc=0x0000ffff95425f48, pid=8074, tid=0x0000ffff9c9d5200 # # JRE version: OpenJDK Runtime Environment (8.0_111-b14) (build 1.8.0_111-8u111-b14-2ubuntu0.16.04.2-b14) # Java VM: OpenJDK 64-Bit Server VM (25.111-b14 mixed mode linux-aarch64 compressed oops) # Problematic frame: # J 1399 C2 java.util.HashMap.putVal(ILjava/lang/Object;Ljava/lang/Object;ZZ)Ljava/lang/Object; (300 bytes) @ 0x0000ffff95425f48 [0x0000ffff95425e00+0x148] # # Core dump written. Default location: /mnt/share/homes/ninjia02/Work/findbugs-test/core or core.8074 # # An error report file with more information is saved as: # /mnt/share/homes/ninjia02/Work/findbugs-test/hs_err_pid8074.log # # If you would like to submit a bug report, please visit: # http://bugreport.java.com/bugreport/crash.jsp # ----- Thanks, Ningsheng

2 4