RFR: 8275202: C2: optimize out more redundant conditions

Tue Nov 28 11:15:06 UTC 2023

On Mon, 27 Nov 2023 14:23:47 GMT, Roberto Castañeda Lozano <rcastanedalo at openjdk.org> wrote:

> except for a potential regression in SPECjvm2008-Compress which I am trying to confirm

I did a second run of this benchmark with more repetitions (20, compared to 5 in the original run) and the regression went down from around 4% to 2%. I guess it might be due to some unfortunate interaction and does not deserve much attention. [Here](https://github.com/openjdk/jdk/files/13487199/SPECjvm2008-Compress.log) is the standard output of running the benchmark with `-Xbatch -XX:+PrintCompilation -XX:+PrintLoopConditionalPropagation` for whatever it's worth.

-------------

PR Comment: https://git.openjdk.org/jdk/pull/14586#issuecomment-1829608314