d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Use vm_page_hold() instead of vm_page_wire(). Otherwise, a multithreaded

Browse Source 
application could cause a wired page to be freed.  In general,
vm_page_hold() should be preferred for ephemeral kernel mappings of pages
borrowed from a user-level address space.  (vm_page_wire() should really be
reserved for indefinite duration pinning by the "owner" of the page.)

Discussed with:	silby
Submitted by:	tegge
This commit is contained in:
Alan Cox 2003-08-11 00:17:44 +00:00
parent e059b0f016
commit f9999c67be
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
sys/kern/sys_pipe.c
View File

@ -696,7 +696,7 @@ pipe_build_write_buffer(wpipe, uio)
vm_page_lock_queues();
for (j = 0; j < i; j++) {
vm_page_unwire(wpipe->pipe_map.ms[j], 1);
vm_page_unhold(wpipe->pipe_map.ms[j]);
atomic_subtract_int(&amountpipekvawired,
PAGE_SIZE);
}
@ -706,7 +706,7 @@ pipe_build_write_buffer(wpipe, uio)
m = PHYS_TO_VM_PAGE(paddr);
vm_page_lock_queues();
vm_page_wire(m);
vm_page_hold(m);
atomic_add_int(&amountpipekvawired, PAGE_SIZE);
vm_page_unlock_queues();
wpipe->pipe_map.ms[i] = m;
@ -775,7 +775,7 @@ pipe_destroy_write_buffer(wpipe)
}
vm_page_lock_queues();
for (i = 0; i < wpipe->pipe_map.npages; i++) {
vm_page_unwire(wpipe->pipe_map.ms[i], 1);
vm_page_unhold(wpipe->pipe_map.ms[i]);
atomic_subtract_int(&amountpipekvawired, PAGE_SIZE);
}
vm_page_unlock_queues();