* THP handling with driver compound page on fault @ 2021-01-30 0:13 Minchan Kim 2021-01-30 1:00 ` Matthew Wilcox 0 siblings, 1 reply; 7+ messages in thread From: Minchan Kim @ 2021-01-30 0:13 UTC (permalink / raw) To: willy, hughd, kirill; +Cc: linux-mm Hi Folks, A custom driver overrides (vm_operations_struct.fault) and map their compound page(__GFP_COMP) to page table on userprocess on demand. Here problem is page fault occur: __do_fault vma->vm_ops->fault <- driver returns their compould page finish_fault alloc_set_pte PageTransCompound(page) do_set_pmd map the page as THP page Since it's not a THP page, system goes random crash after a while due to conflict of VM THP handling with the driver compound page handling. Even though it happens in old kernel with backporting THP pagecache, I think it's still valid on recent kernel. What's the best way to fix it? Driver don't do that or VM need to deal with it? Thank you. ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: THP handling with driver compound page on fault 2021-01-30 0:13 THP handling with driver compound page on fault Minchan Kim @ 2021-01-30 1:00 ` Matthew Wilcox 2021-01-30 1:22 ` Minchan Kim 0 siblings, 1 reply; 7+ messages in thread From: Matthew Wilcox @ 2021-01-30 1:00 UTC (permalink / raw) To: Minchan Kim; +Cc: hughd, kirill, linux-mm On Fri, Jan 29, 2021 at 04:13:07PM -0800, Minchan Kim wrote: > A custom driver overrides (vm_operations_struct.fault) and map their > compound page(__GFP_COMP) to page table on userprocess on demand. You're looking to backport: commit d01ac3c35214ce362f50cada37cb7bab8c801896 Author: Matthew Wilcox (Oracle) <willy@infradead.org> Date: Thu Oct 15 20:05:26 2020 -0700 mm/memory: remove page fault assumption of compound page size ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: THP handling with driver compound page on fault 2021-01-30 1:00 ` Matthew Wilcox @ 2021-01-30 1:22 ` Minchan Kim 2021-01-30 1:28 ` Matthew Wilcox 0 siblings, 1 reply; 7+ messages in thread From: Minchan Kim @ 2021-01-30 1:22 UTC (permalink / raw) To: Matthew Wilcox; +Cc: hughd, kirill, linux-mm Hi Mattew, On Sat, Jan 30, 2021 at 01:00:02AM +0000, Matthew Wilcox wrote: > On Fri, Jan 29, 2021 at 04:13:07PM -0800, Minchan Kim wrote: > > A custom driver overrides (vm_operations_struct.fault) and map their > > compound page(__GFP_COMP) to page table on userprocess on demand. > > You're looking to backport: > > commit d01ac3c35214ce362f50cada37cb7bab8c801896 > Author: Matthew Wilcox (Oracle) <willy@infradead.org> > Date: Thu Oct 15 20:05:26 2020 -0700 > > mm/memory: remove page fault assumption of compound page size > I guess you meant the check below. if (compound_order(page) != HPAGE_PMD_ORDER) What happens if driver allcoated HPAGE_PMD_ORDER size? ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: THP handling with driver compound page on fault 2021-01-30 1:22 ` Minchan Kim @ 2021-01-30 1:28 ` Matthew Wilcox 2021-01-30 1:49 ` Minchan Kim 0 siblings, 1 reply; 7+ messages in thread From: Matthew Wilcox @ 2021-01-30 1:28 UTC (permalink / raw) To: Minchan Kim; +Cc: hughd, kirill, linux-mm On Fri, Jan 29, 2021 at 05:22:24PM -0800, Minchan Kim wrote: > Hi Mattew, > > On Sat, Jan 30, 2021 at 01:00:02AM +0000, Matthew Wilcox wrote: > > On Fri, Jan 29, 2021 at 04:13:07PM -0800, Minchan Kim wrote: > > > A custom driver overrides (vm_operations_struct.fault) and map their > > > compound page(__GFP_COMP) to page table on userprocess on demand. > > > > You're looking to backport: > > > > commit d01ac3c35214ce362f50cada37cb7bab8c801896 > > Author: Matthew Wilcox (Oracle) <willy@infradead.org> > > Date: Thu Oct 15 20:05:26 2020 -0700 > > > > mm/memory: remove page fault assumption of compound page size > > > > I guess you meant the check below. > > if (compound_order(page) != HPAGE_PMD_ORDER) > > What happens if driver allcoated HPAGE_PMD_ORDER size? ... then it should be fine to map it with a PMD entry? ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: THP handling with driver compound page on fault 2021-01-30 1:28 ` Matthew Wilcox @ 2021-01-30 1:49 ` Minchan Kim 2021-01-30 2:09 ` Matthew Wilcox 0 siblings, 1 reply; 7+ messages in thread From: Minchan Kim @ 2021-01-30 1:49 UTC (permalink / raw) To: Matthew Wilcox; +Cc: hughd, kirill, linux-mm On Sat, Jan 30, 2021 at 01:28:44AM +0000, Matthew Wilcox wrote: > On Fri, Jan 29, 2021 at 05:22:24PM -0800, Minchan Kim wrote: > > Hi Mattew, > > > > On Sat, Jan 30, 2021 at 01:00:02AM +0000, Matthew Wilcox wrote: > > > On Fri, Jan 29, 2021 at 04:13:07PM -0800, Minchan Kim wrote: > > > > A custom driver overrides (vm_operations_struct.fault) and map their > > > > compound page(__GFP_COMP) to page table on userprocess on demand. > > > > > > You're looking to backport: > > > > > > commit d01ac3c35214ce362f50cada37cb7bab8c801896 > > > Author: Matthew Wilcox (Oracle) <willy@infradead.org> > > > Date: Thu Oct 15 20:05:26 2020 -0700 > > > > > > mm/memory: remove page fault assumption of compound page size > > > > > > > I guess you meant the check below. > > > > if (compound_order(page) != HPAGE_PMD_ORDER) > > > > What happens if driver allcoated HPAGE_PMD_ORDER size? > > ... then it should be fine to map it with a PMD entry? I don't think so because following logic assumes it's THP page, for example, page_add_file_rmap, count_vm_event(THP_FILE_MAPPED) in do_set_pmd. Isn't it bug? ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: THP handling with driver compound page on fault 2021-01-30 1:49 ` Minchan Kim @ 2021-01-30 2:09 ` Matthew Wilcox 2021-01-30 2:40 ` Minchan Kim 0 siblings, 1 reply; 7+ messages in thread From: Matthew Wilcox @ 2021-01-30 2:09 UTC (permalink / raw) To: Minchan Kim; +Cc: hughd, kirill, linux-mm On Fri, Jan 29, 2021 at 05:49:11PM -0800, Minchan Kim wrote: > On Sat, Jan 30, 2021 at 01:28:44AM +0000, Matthew Wilcox wrote: > > On Fri, Jan 29, 2021 at 05:22:24PM -0800, Minchan Kim wrote: > > > Hi Mattew, > > > > > > On Sat, Jan 30, 2021 at 01:00:02AM +0000, Matthew Wilcox wrote: > > > > On Fri, Jan 29, 2021 at 04:13:07PM -0800, Minchan Kim wrote: > > > > > A custom driver overrides (vm_operations_struct.fault) and map their > > > > > compound page(__GFP_COMP) to page table on userprocess on demand. > > > > > > > > You're looking to backport: > > > > > > > > commit d01ac3c35214ce362f50cada37cb7bab8c801896 > > > > Author: Matthew Wilcox (Oracle) <willy@infradead.org> > > > > Date: Thu Oct 15 20:05:26 2020 -0700 > > > > > > > > mm/memory: remove page fault assumption of compound page size > > > > > > > > > > I guess you meant the check below. > > > > > > if (compound_order(page) != HPAGE_PMD_ORDER) > > > > > > What happens if driver allcoated HPAGE_PMD_ORDER size? > > > > ... then it should be fine to map it with a PMD entry? > > I don't think so because following logic assumes it's THP page, > for example, page_add_file_rmap, count_vm_event(THP_FILE_MAPPED) > in do_set_pmd. > > Isn't it bug? Clearly nobody's actually tested it, but I think in principle, if a device driver returns a PMD sized page (and it's appropriately aligned), it should be mapped with a PMD entry. ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: THP handling with driver compound page on fault 2021-01-30 2:09 ` Matthew Wilcox @ 2021-01-30 2:40 ` Minchan Kim 0 siblings, 0 replies; 7+ messages in thread From: Minchan Kim @ 2021-01-30 2:40 UTC (permalink / raw) To: Matthew Wilcox; +Cc: hughd, kirill, linux-mm On Sat, Jan 30, 2021 at 02:09:11AM +0000, Matthew Wilcox wrote: > On Fri, Jan 29, 2021 at 05:49:11PM -0800, Minchan Kim wrote: > > On Sat, Jan 30, 2021 at 01:28:44AM +0000, Matthew Wilcox wrote: > > > On Fri, Jan 29, 2021 at 05:22:24PM -0800, Minchan Kim wrote: > > > > Hi Mattew, > > > > > > > > On Sat, Jan 30, 2021 at 01:00:02AM +0000, Matthew Wilcox wrote: > > > > > On Fri, Jan 29, 2021 at 04:13:07PM -0800, Minchan Kim wrote: > > > > > > A custom driver overrides (vm_operations_struct.fault) and map their > > > > > > compound page(__GFP_COMP) to page table on userprocess on demand. > > > > > > > > > > You're looking to backport: > > > > > > > > > > commit d01ac3c35214ce362f50cada37cb7bab8c801896 > > > > > Author: Matthew Wilcox (Oracle) <willy@infradead.org> > > > > > Date: Thu Oct 15 20:05:26 2020 -0700 > > > > > > > > > > mm/memory: remove page fault assumption of compound page size > > > > > > > > > > > > > I guess you meant the check below. > > > > > > > > if (compound_order(page) != HPAGE_PMD_ORDER) > > > > > > > > What happens if driver allcoated HPAGE_PMD_ORDER size? > > > > > > ... then it should be fine to map it with a PMD entry? > > > > I don't think so because following logic assumes it's THP page, > > for example, page_add_file_rmap, count_vm_event(THP_FILE_MAPPED) > > in do_set_pmd. > > > > Isn't it bug? > > Clearly nobody's actually tested it, but I think in principle, if a > device driver returns a PMD sized page (and it's appropriately aligned), > it should be mapped with a PMD entry. Okay, then you mean d01ac3c35214ce362f50cada37cb7bab8c801896 will fix the corruption for every-order driver compound page mapping. I have no idea it's worth to solve THP_FILE_MAPPED misaccounting problem but it doesn't matter for my case, at least. Thanks for the quick answer, Matthew. ^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2021-01-30 2:40 UTC | newest] Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2021-01-30 0:13 THP handling with driver compound page on fault Minchan Kim 2021-01-30 1:00 ` Matthew Wilcox 2021-01-30 1:22 ` Minchan Kim 2021-01-30 1:28 ` Matthew Wilcox 2021-01-30 1:49 ` Minchan Kim 2021-01-30 2:09 ` Matthew Wilcox 2021-01-30 2:40 ` Minchan Kim
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).