On 1/24/25 15:16, Matthew Wilcox wrote: > On Thu, Jan 23, 2025 at 07:56:49PM -0800, Alexei Starovoitov wrote: >> - Considered using __GFP_COMP in try_alloc_pages to simplify >> free_pages_nolock a bit, but then decided to make it work >> for all types of pages, since free_pages_nolock() is used by >> stackdepot and currently it's using non-compound order 2. >> I felt it's best to leave it as-is and make free_pages_nolock() >> support all pages. > > We're trying to eliminate non-use of __GFP_COMP. Because people don't > use __GFP_COMP, there's a security check that we can't turn on. Would > you reconsider this change you made? This means changing stackdepot to use __GFP_COMP. Which would be a good thing on its own. But if you consider if off-topic to your series, I can look at it.
