On Mon, 26 Jul 2021 20:45:14 GMT, Markus KARG <github.com+1701815+mk...@openjdk.org> wrote:
>> This PR-*draft* is **work in progress** and an invitation to discuss a >> possible solution for issue >> [JDK-8265891](https://bugs.openjdk.java.net/browse/JDK-8265891). It is *not >> yet* intended for a final review. >> >> As proposed in JDK-8265891, this PR provides an implementation for >> `Channels.newInputStream().transferTo()` which provide superior performance >> compared to the current implementation. The changes are: >> * Prevents transfers through the JVM heap as much as possibly by offloading >> to deeper levels via NIO, hence allowing the operating system to optimize >> the transfer. >> * Using more JRE heap in the fallback case when no NIO is possible (still >> only KiBs, hence mostl ynegligible even on SBCs) to better perform on modern >> hardware / fast I/O devides. >> >> Using JMH I have benchmarked both, the original implementation and this >> implementation, and (depending on the used hardware and use case) >> performance change was approx. doubled performance. So this PoC proofs that >> it makes sense to finalize this work and turn it into an actual OpenJDK >> contribution. >> >> I encourage everybody to discuss this draft: >> * Are there valid arguments for *not* doing this change? >> * Is there a *better* way to improve performance of >> `Channels.newInputStream().transferTo()`? >> * How to go on from here: What is missing to get this ready for an actual >> review? > > Markus KARG has refreshed the contents of this pull request, and previous > commits have been removed. The incremental views will show differences > compared to the previous content of the PR. src/java.base/share/classes/sun/nio/ch/ChannelInputStream.java line 179: > 177: for (long n = srcSize - srcPos; bytesWritten < n;) > 178: bytesWritten += src.transferTo(srcPos + bytesWritten, > Long.MAX_VALUE, dest); > 179: return bytesWritten; If `src` is extended at an inconvenient point in time, for example between the return from a call to `src.transferTo()` that makes `bytesWritten < n` false and the evaluation of that terminating condition, then it appears that not all the content of `src` will be transferred to `dest`. I am not however sure that this violates any contract but it could be a behavioral change from the existing implementation. ------------- PR: https://git.openjdk.java.net/jdk/pull/4263