On Thu, 16 Sep 2021 13:52:24 GMT, Scott Gibbons 
<github.com+6704669+asgibb...@openjdk.org> wrote:

> Change the default code entry alignment to 64 bytes from 32 bytes.  This 
> allows for maintaining proper 64-byte alignment of data within a code 
> segment, which is required by several AVX-512 instructions.
> 
> I ran into this while implementing Base64 encoding and decoding.  Code 
> segments which were allocated with the address mod 32 == 0 but with the 
> address mod 64 != 0 would cause the align() macro to misalign.  This is 
> because the align macro aligns to the size of the code segment and not the 
> offset of the PC.  So align(64) would align the PC to a multiple of 64 bytes 
> from the start of the segment, and not to a pure 64-byte boundary as 
> requested.  Changing the alignment of the segment to 64 bytes fixes the issue.
> 
> I have not seen any measurable difference in either performance or memory 
> usage with the tests I have run.
> 
> See [this 
> ](https://mail.openjdk.java.net/pipermail/hotspot-dev/2021-August/054180.html)
>  article for the discussion thread.

I am back from vacation!

I want to point out that when we generate code for these stubs we don't move 
them in CodeCache (in contrast to compiled methods): 
https://github.com/openjdk/jdk/blob/master/src/hotspot/share/runtime/stubRoutines.cpp#L268

`BufferBlob::create()` allocates specified space (of size `code_size2`, for 
example) directly in CodeCache in `NonNMethod` section:
https://github.com/openjdk/jdk/blob/master/src/hotspot/share/code/codeBlob.cpp#L232
https://github.com/openjdk/jdk/blob/master/src/hotspot/share/code/codeBlob.cpp#L272

Based on that, using `pc()` and new `align64()` should be fine for these few 
cases.

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

PR: https://git.openjdk.java.net/jdk/pull/5547

Reply via email to