| commit | fc16dd344e70b8d1e5cfa16b2987a3afe5ab89d8 | |
| author | Douglas Katzman <dougk@google.com> | |
| Sun, 2 Jul 2017 01:47:31 +0000 (1 21:47 -0400) | ||
| committer | Douglas Katzman <dougk@google.com> | |
| Sun, 2 Jul 2017 01:49:07 +0000 (1 21:49 -0400) | ||
| tree | 7b3d2594eeec64694d5a66530c7d9f794f6d4688 | treesnapshot (tar.gz zip) |
| parent | fc6e1f6d6f4cf27eabd00b0bbec5cb7086d76db3 | commitdiff |
Use similar is_valid_lisp_address() code for everybody
It was apparently the intent - who knows when - to ask the OS whether a
Lisp heap address was valid. None of the current variations do that,
not counting osf1, which had a syscall for it. So move the linux variant
to os-common, deleting the others except win32 which needs this function within
the fault handler and acquires a mutex around the per-thread range tests.
It was apparently the intent - who knows when - to ask the OS whether a
Lisp heap address was valid. None of the current variations do that,
not counting osf1, which had a syscall for it. So move the linux variant
to os-common, deleting the others except win32 which needs this function within
the fault handler and acquires a mutex around the per-thread range tests.