Re: [PATCH] mm: Optimize TLB flushes during page reclaim

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 





On 22-Jan-25 4:39 PM, Vinay Banakar wrote:
On Wed, Jan 22, 2025 at 2:59 AM Bharata B Rao <bharata@xxxxxxx> wrote:
and one flush request is issued for the entire list. Where is the PMD
level (512) batching done? Is that implicit elsewhere in the flow?

Yes, shrink_folio_list() operates on one PMD at a time, so
pageout_list will contain at most 512 pages during each call. Please
refer to madvise_cold_or_pageout_pte_range() in mm/madvise.c for more
details.

While that may be true for MADV_PAGEOUT path, does the same assumption hold good for other paths from which shrink_folio_list() gets called?

Regards,
Bharata.




[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux