Re: [PATCH 03/22] mm: introduce MMF_KERNEL flag and set it for init_mm

On 05/08/2026 14:07, David Hildenbrand (Arm) wrote:
> On 8/3/26 15:59, Christophe Leroy (CS GROUP) wrote:
>> Le 14/07/2026 à 16:03, Kevin Brodsky a écrit :
>>> mm code often needs to know whether some mm represents a kernel or
>>> user address space. This is currently done by comparing the mm
>>> pointer with &init_mm; besides not being particularly elegant, this
>>> ignores the fact that other mm's (e.g. efi_mm) may also represent
>>> parts of the kernel address space.
>>>
>>> Introduce a new mm flag MMF_KERNEL and set it for init_mm.
>>> Subsequent patches will use this flag to replace comparisons with
>>> &init_mm. No functional change is introduced for now.
>> Did you consider performance impact ? This test is usually done in quite
>> critical memory handling functions.
>>
>> init_mm is known at link time. Before your patch 08/22 there is just a
>> comparison of mm (r3) with a constant (loaded in r10):
>>
>> c0014048 <assert_pte_locked>:
>> c0014048:    3d 40 c1 09     lis     r10,-16119
>> c001404c:    39 4a 03 98     addi    r10,r10,920
>> c0014050:    7c 03 50 00     cmpw    r3,r10
>> c0014054:    4d 82 00 20     beqlr
>> ...
>>
>> After patch 08/22 we have, it first checks that mm is not 0, then it loads the
>> word located at mm+528 then AND it with 0x1. This load might be costly.
>>
>> c0014048 <assert_pte_locked>:
>> c0014048:    2c 03 00 00     cmpwi   r3,0
>> c001404c:    7c 85 23 78     mr      r5,r4
>> c0014050:    41 82 00 10     beq     c0014060 <assert_pte_locked+0x18>
>> c0014054:    81 23 02 10     lwz     r9,528(r3)
>> c0014058:    71 29 00 01     andi.   r9,r9,1
>> c001405c:    4c 82 00 20     bnelr
> Is that a real problem, though?

I doubt it, this is a load from a fixed address and therefore highly
cacheable/predictable. At worst we're keeping an extra cache line
allocated, but I'd expect init_mm to be already in the cache most of the
time.

> What we could do is to just keep the == &init_mm check internally on configs
> where we know that there is only a single such MM context.

Sure, if this provides a clear performance benefit on powerpc or other
architectures/configs that don't have an efi_mm we could do that.

- Kevin