On Fri, Aug 27, 2021 at 09:05:21PM +0800, Li, Zhijian wrote:

on 2021/8/27 20:10, Jason Gunthorpe wrote:

quoted

On Fri, Aug 27, 2021 at 08:15:40AM +0000, lizhijian@fujitsu.com wrote:

quoted

i looked over the change-log of hmm_vma_handle_pte(), and found that before
4055062 ("mm/hmm: add missing call to hmm_pte_need_fault in HMM_PFN_SPECIAL handling")

hmm_vma_handle_pte() will not check pte_special(pte) if pte_devmap(pte) is true.

when we reached
"if (pte_special(pte) && !is_zero_pfn(pte_pfn(pte))) {"
the pte have already presented and its pte's flag already fulfilled the request flags.


My question is that
Per https://01.org/blogs/dave/2020/linux-consumption-x86-page-table-bits,
pte_devmap(pte) and pte_special(pte) could be both true in fsdax user case, right ?

How? what code creates that?

I see:

insert_pfn():
	/* Ok, finally just insert the thing.. */
	if (pfn_t_devmap(pfn))
		entry = pte_mkdevmap(pfn_t_pte(pfn, prot));
	else
		entry = pte_mkspecial(pfn_t_pte(pfn, prot));

So what code path ends up setting both bits?

 pte_mkdevmap() will set both _PAGE_SPECIAL | PAGE_DEVMAP

 395 static inline pte_t pte_mkdevmap(pte_t pte)
 396 {
 397         return pte_set_flags(pte, _PAGE_SPECIAL|_PAGE_DEVMAP);
 398 }

below is a calltrace example

[  400.728559] Call Trace:
[  400.731595] dump_stack+0x6d/0x8b
[  400.735536] insert_pfn+0x16c/0x180
[  400.739596] __vm_insert_mixed+0x84/0xc0
[  400.744144] dax_iomap_pte_fault+0x845/0x870
[  400.749089] ext4_dax_huge_fault+0x171/0x1e0
[  400.754096] __do_fault+0x31/0xe0
[  400.758090]  ? pmd_devmap_trans_unstable+0x37/0x90
[  400.763541] handle_mm_fault+0x11b1/0x1680
[  400.768260] exc_page_fault+0x2f4/0x570
[  400.772788]  ? asm_exc_page_fault+0x8/0x30
[  400.777539]  asm_exc_page_fault+0x1e/0x30


So is my previous change reasonable ?

Yes, can you send a proper patch and include the mm mailing list?

Jason