UbiquitousOS
/
OpenCloudOS-Kernel
mirror of https://gitee.com/OpenCloudOS/OpenCloudOS-Kernel.git
11
0
Fork 0
Code Issues Projects Releases Wiki Activity

ARC: [ASID] get_new_mmu_context() to conditionally allocate new ASID 

ASID allocation changes/1

This patch does 2 things:

(1) get_new_mmu_context() NOW moves mm->ASID to a new value ONLY if it
    was from a prev allocation cycle/generation OR if mm had no ASID
    allocated (vs. before would unconditionally moving to a new ASID)

    Callers desiring unconditional update of ASID, e.g.local_flush_tlb_mm()
    (for parent's address space invalidation at fork) need to first force
    the parent to an unallocated ASID.

(2) get_new_mmu_context() always sets the MMU PID reg with unchanged/new
    ASID value.

The gains are:
- consolidation of all asid alloc logic into get_new_mmu_context()
- avoiding code duplication in switch_mm() for PID reg setting
- Enables future change to fold activate_mm() into switch_mm()

Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
This commit is contained in:
Vineet Gupta 2013-07-24 13:53:45 -07:00
parent 5bd87adf9b
commit 3daa48d1d9
2 changed files with 25 additions and 33 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

45
arch/arc/include/asm/mmu_context.h
View File

@ -69,8 +69,8 @@ extern struct mm_struct *asid_mm_map[NUM_ASID + 1];
extern int asid_cache;
/*
* Assign a new ASID to task. If the task already has an ASID, it is
* relinquished.
* Get a new ASID if task doesn't have a valid one (unalloc or from prev cycle)
* Also set the MMU PID register to existing/updated ASID
*/
static inline void get_new_mmu_context(struct mm_struct *mm)
{
@ -79,6 +79,17 @@ static inline void get_new_mmu_context(struct mm_struct *mm)
local_irq_save(flags);
/*
* Move to new ASID if it was not from current alloc-cycle/generation.
*
* Note: Callers needing new ASID unconditionally, independent of
* generation, e.g. local_flush_tlb_mm() for forking parent,
* first need to destroy the context, setting it to invalid
* value.
*/
if (mm->context.asid <= asid_cache)
goto set_hw;
/*
* Relinquish the currently owned ASID (if any).
* Doing unconditionally saves a cmp-n-branch; for already unused
@ -99,9 +110,9 @@ static inline void get_new_mmu_context(struct mm_struct *mm)
* task with ASID from prev allocation cycle (before ASID roll-over).
*
* This might look wrong - if we are re-using some other task's ASID,
* won't we use it's stale TLB entries too. Actually switch_mm( ) takes
* won't we use it's stale TLB entries too. Actually the algorithm takes
* care of such a case: it ensures that task with ASID from prev alloc
* cycle, when scheduled will refresh it's ASID: see switch_mm( ) below
* cycle, when scheduled will refresh it's ASID
* The stealing scenario described here will only happen if that task
* didn't get a chance to refresh it's ASID - implying stale entries
* won't exist.
@ -114,7 +125,8 @@ static inline void get_new_mmu_context(struct mm_struct *mm)
asid_mm_map[asid_cache] = mm;
mm->context.asid = asid_cache;
write_aux_reg(ARC_REG_PID, asid_cache | MMU_ENABLE);
set_hw:
write_aux_reg(ARC_REG_PID, mm->context.asid | MMU_ENABLE);
local_irq_restore(flags);
}
@ -141,28 +153,7 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
write_aux_reg(ARC_REG_SCRATCH_DATA0, next->pgd);
#endif
/*
* Get a new ASID if task doesn't have a valid one. Possible when
* -task never had an ASID (fresh after fork)
* -it's ASID was stolen - past an ASID roll-over.
* -There's a third obscure scenario (if this task is running for the
* first time afer an ASID rollover), where despite having a valid
* ASID, we force a get for new ASID - see comments at top.
*
* Both the non-alloc scenario and first-use-after-rollover can be
* detected using the single condition below: NO_ASID = 256
* while asid_cache is always a valid ASID value (0-255).
*/
if (next->context.asid > asid_cache) {
get_new_mmu_context(next);
} else {
/*
* XXX: This will never happen given the chks above
* BUG_ON(next->context.asid > MAX_ASID);
*/
write_aux_reg(ARC_REG_PID, next->context.asid | MMU_ENABLE);
}
get_new_mmu_context(next);
}
static inline void destroy_context(struct mm_struct *mm)

13
arch/arc/mm/tlb.c
View File

@ -258,13 +258,14 @@ noinline void local_flush_tlb_mm(struct mm_struct *mm)
return;
/*
* Workaround for Android weirdism:
* A binder VMA could end up in a task such that vma->mm != tsk->mm
* old code would cause h/w - s/w ASID to get out of sync
* - Move to a new ASID, but only if the mm is still wired in
* (Android Binder ended up calling this for vma->mm != tsk->mm,
* causing h/w - s/w ASID to get out of sync)
* - Also get_new_mmu_context() new implementation allocates a new
* ASID only if it is not allocated already - so unallocate first
*/
if (current->mm != mm)
destroy_context(mm);
else
destroy_context(mm);
if (current->mm == mm)
get_new_mmu_context(mm);
}