RFR: JDK-8313319: [linux] mmap should use MAP_FIXED_NOREPLACE if available

[Johan Sjölen]

On Sun, 6 Aug 2023 17:09:41 GMT, Thomas Stuefe <stuefe at openjdk.org> wrote:

> os::attempt_reserve_memory_at() is used to map memory at a specific address that we fancy. It is not used to replace an existing mapping (hence we omit MAP_FIXED). 
> 
> But it may result in the kernel creating a mapping at a different address if our wish address is blocked; we then have to unmap again. That mmap-munmap cycle is unnecessary.
> 
> MAP_FIXED_NOREPLACE exists since Linux 4.17 and prevents that unnecessary cycle . We should use that if available.

Hi,

Thanks for this and looks good to me. Two small nits.

src/hotspot/os/linux/os_linux.cpp line 3497:

> 3495:   //
> 3496:   // Backward compatibility: Older kernels will ignore the unknown flag; so mmap will behave
> 3497:   // as in mode (a).

Is this in case of a libc that's more up to date than the kernel?

test/hotspot/gtest/runtime/test_os.cpp line 945:

> 943:     os::release_memory(p1, M);
> 944:   } else {
> 945:     tty->print_cr("Skipped.");

Have you succesfully seen this printed? AFAIK Gtest only prints stderr by default

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

Marked as reviewed by jsjolen (Reviewer).

PR Review: https://git.openjdk.org/jdk/pull/15170#pullrequestreview-1566808133
PR Review Comment: https://git.openjdk.org/jdk/pull/15170#discussion_r1286806000
PR Review Comment: https://git.openjdk.org/jdk/pull/15170#discussion_r1286809194