> -----Original Message----- > From: Oscar Salvador [mailto:osalvador@xxxxxxx] > Sent: Wednesday, November 18, 2020 8:45 AM > To: Muchun Song <songmuchun@xxxxxxxxxxxxx> > Cc: Song Bao Hua (Barry Song) <song.bao.hua@xxxxxxxxxxxxx>; > corbet@xxxxxxx; mike.kravetz@xxxxxxxxxx; tglx@xxxxxxxxxxxxx; > mingo@xxxxxxxxxx; bp@xxxxxxxxx; x86@xxxxxxxxxx; hpa@xxxxxxxxx; > dave.hansen@xxxxxxxxxxxxxxx; luto@xxxxxxxxxx; peterz@xxxxxxxxxxxxx; > viro@xxxxxxxxxxxxxxxxxx; akpm@xxxxxxxxxxxxxxxxxxxx; paulmck@xxxxxxxxxx; > mchehab+huawei@xxxxxxxxxx; pawan.kumar.gupta@xxxxxxxxxxxxxxx; > rdunlap@xxxxxxxxxxxxx; oneukum@xxxxxxxx; anshuman.khandual@xxxxxxx; > jroedel@xxxxxxx; almasrymina@xxxxxxxxxx; rientjes@xxxxxxxxxx; > willy@xxxxxxxxxxxxx; mhocko@xxxxxxxx; duanxiongchun@xxxxxxxxxxxxx; > linux-doc@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; linux-mm@xxxxxxxxx; > linux-fsdevel@xxxxxxxxxxxxxxx > Subject: Re: [External] RE: [PATCH v4 00/21] Free some vmemmap pages of > hugetlb page > > On 2020-11-17 17:29, Muchun Song wrote: > > Now for the 2MB HugrTLB page, we only free 6 vmemmap pages. > > But your words woke me up. Maybe we really can free 7 vmemmap > > pages. In this case, we can see 8 of the 512 struct page structures > > has beed set PG_head flag. If we can adjust compound_head() > > slightly and make compound_head() return the real head struct > > page when the parameter is the tail struct page but with PG_head > > flag set. I will start an investigation and a test. > > I would not overcomplicate things at this stage, but rather keep it > simple as the code is already tricky enough(without counting the LOC > thatvit adds). > We can always build on top later on in order to improve things. Yep. I am not expecting freeing tail page to be done at this stage. This could be something in the todo list after the first patchset is solid. > > -- > Oscar Salvador > SUSE L3 Thanks Barry
