RFR: 8303215: Make thread stacks not use huge pages

Poonam Bajaj poonam at openjdk.org
Wed May 24 18:51:05 UTC 2023


On Tue, 23 May 2023 18:01:51 GMT, Poonam Bajaj <poonam at openjdk.org> wrote:

> When a system has Transparent Huge Pages (THP) enabled (/sys/kernel/mm/transparent_hugepage/enabled is set to 'always'), thread stacks can have significantly more resident set size (RSS) than they actually require. This occurs when the stack size is 2MB or larger, which makes the memory range of the stack more likely to be aligned on a Large Page Size boundary (2MB on most systems). This in turn makes the stack eligible to be backed by transparent huge pages resulting in more memory consumption than it would otherwise when standard small pages are used. This issue is more apparent on AArch64 platforms where the default stack size is 2MB. 
> 
> 
> Example mapping from smaps illustrating this issue:
> fffced200000-fffced204000 ---p 00000000 00:00 0 
> Size:                16 kB     # guard pages
> KernelPageSize:       4 kB
> MMUPageSize:          4 kB
> ...
> fffced204000-fffced400000 rw-p 00000000 00:00 0 
> Size:              2032 kB    # stack space
> KernelPageSize:       4 kB
> MMUPageSize:          4 kB
> Rss:               2032 kB    # entire stack resident in memory
> 
> 
> This fix addresses this issue with the following two main changes:
> 
> 1. Change the default stack size to 2040KB, which is 2 pages less than 2MB. This ensures that stacks don't get 2MB aligned. And why 2 pages less than 2MB, because for non-JavaThreads, glibc adds an additional guard page to the total stack size. To keep it simple and to keep the default stack size value for  all options - ThreadStackSize, CompilerThreadStackSize, and VMThreadStackSize same, we use the default value as 2040K.
> 
> Example mapping for a JavaThread:
> 
> ffff6e913000-ffff6e917000 ---p 00000000 00:00 0 
> Size:                 16 kB
> KernelPageSize:        4 kB
> MMUPageSize:           4 kB
> ...
> ffff6e917000-ffff6eb11000 rw-p 00000000 00:00 0 
> Size:               2024 kB
> KernelPageSize:        4 kB
> MMUPageSize:           4 kB
> Rss:                  92 kB
> 
> Example Mapping for a non-JavaThread (WatcherThread):
> 
> ffff6eb11000-ffff6eb12000 ---p 00000000 00:00 0 
> Size:                  4 kB
> KernelPageSize:        4 kB
> MMUPageSize:           4 kB
> ...
> ffff6eb12000-ffff6ed10000 rw-p 00000000 00:00 0 
> Size:               2040 kB
> KernelPageSize:        4 kB
> MMUPageSize:           4 kB
> Rss:                  12 kB
> 
> 
> 2. If the requested stack size is greater than or equal to the default large page size (2MB on most systems) and can be large-page aligned, then add an additional page to the stack size. This reduces t...

For the 64K page size, the default huge page size is usually 512M. The aarch64 machine with 64K page size that I am testing on has:


bash-4.4$ cat /proc/meminfo | grep Hugepagesize
Hugepagesize:     524288 kB


And on that, the current set of changes behave correctly when I run with -Xss512m. Snippet of pmap output:


0000fff8b7ff0000     192       0       0 -----   [ anon ]  # guard pages
0000fff8b8020000  524160     192     192 rw---   [ anon ]  # usable stack area
0000fff8d8000000   65472   65472   65472 rw---   [ anon ]
0000fff8dbff0000      64       0       0 -----   [ anon ]
0000fff8dfff0000     192       0       0 -----   [ anon ]
0000fff8e0020000  524160     192     192 rw---   [ anon ]

an additional 64k page gets added to the stack size to disturb the large page alignment.

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

PR Comment: https://git.openjdk.org/jdk/pull/14105#issuecomment-1561765080


More information about the hotspot-runtime-dev mailing list