On Thu, 7 Aug 2025 13:54:20 -0600, alex.williamson@xxxxxxxxxx wrote: > On Thu, 7 Aug 2025 12:14:19 +0800 > lizhe.67@xxxxxxxxxxxxx wrote: > > > On Wed, 6 Aug 2025 14:35:15 +0200, david@xxxxxxxxxx wrote: > > > > > On 05.08.25 03:24, Alex Williamson wrote: > > > > Objections were raised to adding this helper to common code with only a > > > > single user and dubious generalism. Pull it back into subsystem code. > > > > > > > > Link: https://lore.kernel.org/all/CAHk-=whhYRMS7Xc9k_JBdrGvp++JLmU0T2xXEgn046hWrj7q8Q@xxxxxxxxxxxxxx/ > > > > Cc: David Hildenbrand <david@xxxxxxxxxx> > > > > Cc: Jason Gunthorpe <jgg@xxxxxxxxxx> > > > > Cc: Li Zhe <lizhe.67@xxxxxxxxxxxxx> > > > > Signed-off-by: Alex Williamson <alex.williamson@xxxxxxxxxx> > > > > --- > > > > > > So, I took the original patch and > > > * moved the code to mm_inline.h (sounds like a better fit) > > > * Tweaked the patch description > > > * Tweaked the documentation and turned it into proper kerneldoc > > > * Made the function return "size_t" as well > > > * Use the page_to_section() trick to avoid nth_page(). > > > > > > Only compile-tested so far. Still running it through some cross compiles. > > > > > > > > > From 36d67849bfdbc184990f21464c53585d35648616 Mon Sep 17 00:00:00 2001 > > > From: Li Zhe <lizhe.67@xxxxxxxxxxxxx> > > > Date: Thu, 10 Jul 2025 16:53:51 +0800 > > > Subject: [PATCH] mm: introduce num_pages_contiguous() > > > > > > Let's add a simple helper for determining the number of contiguous pages > > > that represent contiguous PFNs. > > > > > > In an ideal world, this helper would be simpler or not even required. > > > Unfortunately, on some configs we still have to maintain (SPARSEMEM > > > without VMEMMAP), the memmap is allocated per memory section, and we might > > > run into weird corner cases of false positives when blindly testing for > > > contiguous pages only. > > > > > > One example of such false positives would be a memory section-sized hole > > > that does not have a memmap. The surrounding memory sections might get > > > "struct pages" that are contiguous, but the PFNs are actually not. > > > > > > This helper will, for example, be useful for determining contiguous PFNs > > > in a GUP result, to batch further operations across returned "struct > > > page"s. VFIO will utilize this interface to accelerate the VFIO DMA map > > > process. > > > > > > Implementation based on Linus' suggestions to avoid new usage of > > > nth_page() where avoidable. > > > > > > Suggested-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> > > > Suggested-by: Jason Gunthorpe <jgg@xxxxxxxx> > > > Signed-off-by: Li Zhe <lizhe.67@xxxxxxxxxxxxx> > > > Co-developed-by: David Hildenbrand <david@xxxxxxxxxx> > > > Signed-off-by: David Hildenbrand <david@xxxxxxxxxx> > > > --- > > > include/linux/mm.h | 7 ++++++- > > > include/linux/mm_inline.h | 35 +++++++++++++++++++++++++++++++++++ > > > 2 files changed, 41 insertions(+), 1 deletion(-) > > > > > > diff --git a/include/linux/mm.h b/include/linux/mm.h > > > index fa538feaa8d95..2852bcd792745 100644 > > > --- a/include/linux/mm.h > > > +++ b/include/linux/mm.h > > > @@ -1759,7 +1759,12 @@ static inline unsigned long page_to_section(const struct page *page) > > > { > > > return (page->flags >> SECTIONS_PGSHIFT) & SECTIONS_MASK; > > > } > > > -#endif > > > +#else /* !SECTION_IN_PAGE_FLAGS */ > > > +static inline unsigned long page_to_section(const struct page *page) > > > +{ > > > + return 0; > > > +} > > > +#endif /* SECTION_IN_PAGE_FLAGS */ > > > > > > /** > > > * folio_pfn - Return the Page Frame Number of a folio. > > > diff --git a/include/linux/mm_inline.h b/include/linux/mm_inline.h > > > index 89b518ff097e6..58cb99b69f432 100644 > > > --- a/include/linux/mm_inline.h > > > +++ b/include/linux/mm_inline.h > > > @@ -616,4 +616,39 @@ static inline bool vma_has_recency(struct vm_area_struct *vma) > > > return true; > > > } > > > > > > +/** > > > + * num_pages_contiguous() - determine the number of contiguous pages > > > + * that represent contiguous PFNs > > > + * @pages: an array of page pointers > > > + * @nr_pages: length of the array, at least 1 > > > + * > > > + * Determine the number of contiguous pages that represent contiguous PFNs > > > + * in @pages, starting from the first page. > > > + * > > > + * In kernel configs where contiguous pages might not imply contiguous PFNs > > > + * over memory section boundaries, this function will stop at the memory > > > + * section boundary. > > > + * > > > + * Returns the number of contiguous pages. > > > + */ > > > +static inline size_t num_pages_contiguous(struct page **pages, size_t nr_pages) > > > +{ > > > + struct page *cur_page = pages[0]; > > > + unsigned long section = page_to_section(cur_page); > > > + size_t i; > > > + > > > + for (i = 1; i < nr_pages; i++) { > > > + if (++cur_page != pages[i]) > > > + break; > > > + /* > > > + * In unproblematic kernel configs, page_to_section() == 0 and > > > + * the whole check will get optimized out. > > > + */ > > > + if (page_to_section(cur_page) != section) > > > + break; > > > + } > > > + > > > + return i; > > > +} > > > + > > > #endif > > > > I sincerely appreciate your thoughtful revisions to this patch. The > > code looks great. > > > > Based on this patch, I reran the performance tests for the VFIO > > optimizations. The results show no significant change from the > > previous data. > > > > Since num_pages_contiguous() is now defined in mm_inline.h, the > > second patch "vfio/type1: optimize vfio_pin_pages_remote()" in this > > series needs to include that header to resolve the build error. > > > > diff --git a/drivers/vfio/vfio_iommu_type1.c b/drivers/vfio/vfio_iommu_type1.c > > index 1136d7ac6b59..af98cb94153c 100644 > > --- a/drivers/vfio/vfio_iommu_type1.c > > +++ b/drivers/vfio/vfio_iommu_type1.c > > @@ -37,6 +37,7 @@ > > #include <linux/vfio.h> > > #include <linux/workqueue.h> > > #include <linux/notifier.h> > > +#include <linux/mm_inline.h> > > #include "vfio.h" > > > > #define DRIVER_VERSION "0.2" > > Hi Zhe, > > Once we're all satisfied with the update, please post a full new > series. Since we're restarting with fresh commits, please include the > fixes directly in the original patches. Thanks, Got it. If no more comments come in over the next few days, I'll send out a fresh patchset. Thanks, Zhe
