Messages in this thread Patch in this message | | | Subject | [PATCH 21 of 24] xen: allow set_pte_at on init_mm to be lockless | Date | Wed, 02 Apr 2008 10:54:10 -0700 | From | Jeremy Fitzhardinge <> |
| |
The usual pagetable locking protocol doesn't seem to apply to updates to init_mm, so don't rely on preemption being disabled in xen_set_pte_at on init_mm.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> --- arch/x86/xen/mmu.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-)
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -156,6 +156,10 @@ void xen_set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pteval) { + /* updates to init_mm may be done without lock */ + if (mm == &init_mm) + preempt_disable(); + if (mm == current->mm || mm == &init_mm) { if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU) { struct multicall_space mcs; @@ -163,12 +167,16 @@ MULTI_update_va_mapping(mcs.mc, addr, pteval, 0); xen_mc_issue(PARAVIRT_LAZY_MMU); - return; + goto out; } else if (HYPERVISOR_update_va_mapping(addr, pteval, 0) == 0) - return; + goto out; } xen_set_pte(ptep, pteval); + +out: + if (mm == &init_mm) + preempt_enable(); } pteval_t xen_pte_val(pte_t pte)
| |