Skip to content
  1. Sep 23, 2005
  3. Sep 21, 2005
  5. Sep 15, 2005
  7. Sep 14, 2005
    • Hugh Dickins's avatar
      [PATCH] error path in setup_arg_pages() misses vm_unacct_memory() · 2fd4ef85
      Hugh Dickins authored 
      

Pavel Emelianov and Kirill Korotaev observe that fs and arch users of
security_vm_enough_memory tend to forget to vm_unacct_memory when a
failure occurs further down (typically in setup_arg_pages variants).

These are all users of insert_vm_struct, and that reservation will only
be unaccounted on exit if the vma is marked VM_ACCOUNT: which in some
cases it is (hidden inside VM_STACK_FLAGS) and in some cases it isn't.

So x86_64 32-bit and ppc64 vDSO ELFs have been leaking memory into
Committed_AS each time they're run.  But don't add VM_ACCOUNT to them,
it's inappropriate to reserve against the very unlikely case that gdb
be used to COW a vDSO page - we ought to do something about that in
do_wp_page, but there are yet other inconsistencies to be resolved.

The safe and economical way to fix this is to let insert_vm_struct do
the security_vm_enough_memory check when it finds VM_ACCOUNT is set.

And the MIPS irix_brk has been calling security_vm_enough_memory before
calling do_brk which repeats it, doubly accounting and so also leaking.
Remove that, and all the fs and arch calls to security_vm_enough_memory:
give it a less misleading name later on.

Signed-off-by: default avatarHugh Dickins <hugh@veritas.com>
Signed-Off-By: default avatarKirill Korotaev <dev@sw.ru>
Signed-off-by: default avatarLinus Torvalds <torvalds@osdl.org>
      2fd4ef85
  9. Sep 13, 2005
    • Paul Mackerras's avatar
      [PATCH] ppc64: Make eeh_init function again · 0160f53e
      Paul Mackerras authored 
      

My patch "Separate pci bits out of struct device_node" (commit
1635317f) had the unfortunate
side-effect that it stopped eeh_init() from working correctly.

It needs the pointers set up by find_and_init_phbs(), but it was being
called just before find_and_init_phbs().  That meant that we didn't
enable EEH (pSeries PCI error recovery) on any devices, and that meant
that on POWER5 systems, the hypervisor wouldn't let us enable memory or
I/O space access to any devices, and their drivers got somewhat
confused.

This fixes it by moving the eeh_init call after find_and_init_phbs.
Tested on a POWER5 partition.

Signed-of-by: default avatarPaul Mackerras <paulus@samba.org>
Signed-of-by: default avatarLinus Torvalds <torvalds@osdl.org>
      0160f53e
  11. Sep 12, 2005
  13. Sep 10, 2005
  15. Sep 09, 2005
  17. Sep 08, 2005
  19. Sep 07, 2005
  21. Sep 06, 2005