Re: [PATCH v1 08/13] arm64: mm: Simplify __flush_tlb_range_limit_excess()

[Ryan Roberts <ryan.roberts@arm.com>]

On 17/12/2025 08:12, Dev Jain wrote:
> 
> On 16/12/25 8:15 pm, Ryan Roberts wrote:
>> From: Will Deacon <will@kernel.org>
>>
>> __flush_tlb_range_limit_excess() is unnecessarily complicated:
>>
>>   - It takes a 'start', 'end' and 'pages' argument, whereas it only
>>     needs 'pages' (which the caller has computed from the other two
>>     arguments!).
>>
>>   - It erroneously compares 'pages' with MAX_TLBI_RANGE_PAGES when
>>     the system doesn't support range-based invalidation but the range to
>>     be invalidated would result in fewer than MAX_DVM_OPS invalidations.
>>
>> Simplify the function so that it no longer takes the 'start' and 'end'
>> arguments and only considers the MAX_TLBI_RANGE_PAGES threshold on
>> systems that implement range-based invalidation.
>>
>> Signed-off-by: Will Deacon <will@kernel.org>
>> Signed-off-by: Ryan Roberts <ryan.roberts@arm.com>
>> ---
>>  arch/arm64/include/asm/tlbflush.h | 20 ++++++--------------
>>  1 file changed, 6 insertions(+), 14 deletions(-)
>>
>> diff --git a/arch/arm64/include/asm/tlbflush.h b/arch/arm64/include/asm/tlbflush.h
>> index 0e1902f66e01..3b72a71feac0 100644
>> --- a/arch/arm64/include/asm/tlbflush.h
>> +++ b/arch/arm64/include/asm/tlbflush.h
>> @@ -527,21 +527,13 @@ static __always_inline void __flush_tlb_range_op(tlbi_op lop, tlbi_op rop,
>>  #define __flush_s2_tlb_range_op(op, start, pages, stride, tlb_level) \
>>  	__flush_tlb_range_op(op, r##op, start, pages, stride, 0, tlb_level, kvm_lpa2_is_enabled())
>>  
>> -static inline bool __flush_tlb_range_limit_excess(unsigned long start,
>> -		unsigned long end, unsigned long pages, unsigned long stride)
>> +static inline bool __flush_tlb_range_limit_excess(unsigned long pages,
>> +						  unsigned long stride)
>>  {
>> -	/*
>> -	 * When the system does not support TLB range based flush
>> -	 * operation, (MAX_DVM_OPS - 1) pages can be handled. But
>> -	 * with TLB range based operation, MAX_TLBI_RANGE_PAGES
>> -	 * pages can be handled.
>> -	 */
>> -	if ((!system_supports_tlb_range() &&
>> -	     (end - start) >= (MAX_DVM_OPS * stride)) ||
>> -	    pages > MAX_TLBI_RANGE_PAGES)
>> +	if (system_supports_tlb_range() && pages > MAX_TLBI_RANGE_PAGES)
>>  		return true;
>>  
>> -	return false;
>> +	return pages >= (MAX_DVM_OPS * stride) >> PAGE_SHIFT;
> 
> The function will return true if tlb range is supported, but
> ((MAX_DVM_OPS * stride) >> PAGE_SHIFT) < pages <= MAX_TLBI_RANGE_PAGES.
> So I think you need to do
> https://lore.kernel.org/all/1b15b4f0-5490-4dac-8344-e716dd189751@arm.com/

I agree with your overall proposal, but I think a few of the details are not
quite correct.

I think the max number of DVM ops that could be issued by a single
__flush_tlb_range() call on a system with tlb-range is 20, not 4 as you suggest;

- 4 for each of the scales
- 1 for the final single page
- 15 to align to a 64K boundary on systems with LPA2 (with 4K page size)

But that doesn't really change your argument.

So proposing to change it to this in next version:

static inline bool __flush_tlb_range_limit_excess(unsigned long pages,
						  unsigned long stride)
{
	/*
	 * Assume that the worst case number of DVM ops required to flush a
	 * given range on a system that supports tlb-range is 20 (4 scales, 1
	 * final page, 15 for alignment on LPA2 systems), which is much smaller
	 * than MAX_DVM_OPS.
	 */
	if (system_supports_tlb_range())
		return pages > MAX_TLBI_RANGE_PAGES;

	return pages >= (MAX_DVM_OPS * stride) >> PAGE_SHIFT;
}

Thanks,
Ryan

> 
>>  }
>>  
>>  static inline void __flush_tlb_range_nosync(struct mm_struct *mm,
>> @@ -555,7 +547,7 @@ static inline void __flush_tlb_range_nosync(struct mm_struct *mm,
>>  	end = round_up(end, stride);
>>  	pages = (end - start) >> PAGE_SHIFT;
>>  
>> -	if (__flush_tlb_range_limit_excess(start, end, pages, stride)) {
>> +	if (__flush_tlb_range_limit_excess(pages, stride)) {
>>  		flush_tlb_mm(mm);
>>  		return;
>>  	}
>> @@ -619,7 +611,7 @@ static inline void flush_tlb_kernel_range(unsigned long start, unsigned long end
>>  	end = round_up(end, stride);
>>  	pages = (end - start) >> PAGE_SHIFT;
>>  
>> -	if (__flush_tlb_range_limit_excess(start, end, pages, stride)) {
>> +	if (__flush_tlb_range_limit_excess(pages, stride)) {
>>  		flush_tlb_all();
>>  		return;
>>  	}