RFR (M) CR 8050147: StoreLoad barrier interferes with stack usages
vladimir.kozlov at oracle.com
Wed Sep 3 23:30:05 UTC 2014
On 9/3/14 11:07 AM, David Chase wrote:
>> Any opinions?
> Not a capital-R reviewer.
> Looks good to me, with one question that is not directly relevant to the patch.
> Is there any particular reason why we have two copies of “bang_size_in_bytes”
> in compile.cpp and c1_LIRAssembler.cpp that appear to be (probably,
> not certainly) the same?
Only name is the same but they execute totally different code, C1 and C2
specific, and they are used only locally in C1 and C2. There is no
reason to move it into shared code, and it will be difficult.
More information about the hotspot-compiler-dev