Reorder and align LRUCacheShard data members to reduce false sharing

[IslamAbdelRahman]

Reorder data members in LRUCacheShard so that the members that we update frequently don't share the same cache line with members that are mostly read-only

[facebook-github-bot]

@IslamAbdelRahman has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

[siying]

Err... Windows doesn't support this...

[siying]

@IslamAbdelRahman can you check whether it is really aligned in GCC too?

[grooverdan]

Also wouldn't you also align the highly accessed members at the end so they don't do false sharing with each other.

For attributes:

For Windows:  __declspec(align(n))
gcc/clang: _attribute__((__aligned__((n))))

[IslamAbdelRahman]

Thanks @grooverdan for the attributes

You mean usage_, lru_usage_ and mutex each have it's own cache line ?

[facebook-github-bot]

@IslamAbdelRahman updated the pull request - view changes - changes since last import

[grooverdan]

Yes marking attributes to their own l1 cache line is what I had in mind as by the description they are heavily written. You can test with perf c2c

[grooverdan]

Looks like you have having trouble with heap allocations. Seems like overloading new with aligned alloc is needed.

Here are some definitions that may help.

https://github.com/MariaDB/server/pull/185/files#diff-05fed5ee53d1f9b413d7468e7699ecb0R21

[siying]

@IslamAbdelRahman still not supported on Windows.

[facebook-github-bot]

@IslamAbdelRahman updated the pull request - view changes - changes since last import

[facebook-github-bot]

@IslamAbdelRahman has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

[IslamAbdelRahman]

I just disabled it for windows, now it pass :D

[siying]

May consider @grooverdan 's suggestion too. But either way, it is good to go.

[siying]

Actually, what we want is to align since elems_, as it is the beginning of frequently changing parameter and from there they are exactly 64 bytes. This is hard to do with the alignment annotation.

[IslamAbdelRahman]

abandoned for the sake of #2620