Re: CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP is broken, was Re: [RFC PATCH 0/6] Deep talk about folio vmap

2025-04-07 Thread Muchun Song
> On Apr 7, 2025, at 16:59, Christoph Hellwig wrote: > > On Mon, Apr 07, 2025 at 02:43:20PM +0800, Muchun Song wrote: >> By the way, in case you truly struggle to comprehend the fundamental >> aspects of HVO, I would like to summarize for you the user-visible >> behaviors in comparison to the

Re: CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP is broken, was Re: [RFC PATCH 0/6] Deep talk about folio vmap

2025-04-07 Thread Christoph Hellwig
On Mon, Apr 07, 2025 at 02:43:20PM +0800, Muchun Song wrote: > By the way, in case you truly struggle to comprehend the fundamental > aspects of HVO, I would like to summarize for you the user-visible > behaviors in comparison to the situation where HVO is disabled. > > HVO StatusTail

Re: CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP is broken, was Re: [RFC PATCH 0/6] Deep talk about folio vmap

2025-04-07 Thread Huan Yang
在 2025/4/7 15:22, Muchun Song 写道: On Apr 7, 2025, at 15:09, Huan Yang wrote: 在 2025/4/7 14:43, Muchun Song 写道: On Apr 7, 2025, at 11:37, Muchun Song wrote: On Apr 7, 2025, at 11:21, Huan Yang wrote: 在 2025/4/7 10:57, Muchun Song 写道: On Apr 7, 2025, at 09:59, Huan Yang wrote: 在

Re: CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP is broken, was Re: [RFC PATCH 0/6] Deep talk about folio vmap

2025-04-07 Thread Muchun Song
> On Apr 7, 2025, at 15:09, Huan Yang wrote: > > > 在 2025/4/7 14:43, Muchun Song 写道: >> >>> On Apr 7, 2025, at 11:37, Muchun Song wrote: >>> >>> >>> On Apr 7, 2025, at 11:21, Huan Yang wrote: 在 2025/4/7 10:57, Muchun Song 写道: >> On Apr 7, 2025, at 09:59, Huan Y

Re: CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP is broken, was Re: [RFC PATCH 0/6] Deep talk about folio vmap

2025-04-07 Thread Huan Yang
在 2025/4/7 14:43, Muchun Song 写道: On Apr 7, 2025, at 11:37, Muchun Song wrote: On Apr 7, 2025, at 11:21, Huan Yang wrote: 在 2025/4/7 10:57, Muchun Song 写道: On Apr 7, 2025, at 09:59, Huan Yang wrote: 在 2025/4/4 18:07, Muchun Song 写道: On Apr 4, 2025, at 17:38, Muchun Song wrote:

Re: CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP is broken, was Re: [RFC PATCH 0/6] Deep talk about folio vmap

2025-04-06 Thread Muchun Song
> On Apr 7, 2025, at 11:37, Muchun Song wrote: > > > >> On Apr 7, 2025, at 11:21, Huan Yang wrote: >> >> >> 在 2025/4/7 10:57, Muchun Song 写道: >>> On Apr 7, 2025, at 09:59, Huan Yang wrote: 在 2025/4/4 18:07, Muchun Song 写道: >> On Apr 4, 2025, at 17:38, Muchun So

Re: CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP is broken, was Re: [RFC PATCH 0/6] Deep talk about folio vmap

2025-04-06 Thread Muchun Song
> On Apr 7, 2025, at 11:21, Huan Yang wrote: > > > 在 2025/4/7 10:57, Muchun Song 写道: >> >>> On Apr 7, 2025, at 09:59, Huan Yang wrote: >>> >>> >>> 在 2025/4/4 18:07, Muchun Song 写道: > On Apr 4, 2025, at 17:38, Muchun Song wrote: > > > >> On Apr 4, 2025, at 17:01, Ch

Re: CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP is broken, was Re: [RFC PATCH 0/6] Deep talk about folio vmap

2025-04-06 Thread Huan Yang
在 2025/4/7 10:57, Muchun Song 写道: On Apr 7, 2025, at 09:59, Huan Yang wrote: 在 2025/4/4 18:07, Muchun Song 写道: On Apr 4, 2025, at 17:38, Muchun Song wrote: On Apr 4, 2025, at 17:01, Christoph Hellwig wrote: After the btrfs compressed bio discussion I think the hugetlb changes that

Re: CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP is broken, was Re: [RFC PATCH 0/6] Deep talk about folio vmap

2025-04-06 Thread Muchun Song
> On Apr 7, 2025, at 09:59, Huan Yang wrote: > > > 在 2025/4/4 18:07, Muchun Song 写道: >> >>> On Apr 4, 2025, at 17:38, Muchun Song wrote: >>> >>> >>> On Apr 4, 2025, at 17:01, Christoph Hellwig wrote: After the btrfs compressed bio discussion I think the hugetlb changes t

Re: CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP is broken, was Re: [RFC PATCH 0/6] Deep talk about folio vmap

2025-04-06 Thread Huan Yang
在 2025/4/4 18:07, Muchun Song 写道: On Apr 4, 2025, at 17:38, Muchun Song wrote: On Apr 4, 2025, at 17:01, Christoph Hellwig wrote: After the btrfs compressed bio discussion I think the hugetlb changes that skip the tail pages are fundamentally unsafe in the current kernel. That is beca

Re: CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP is broken, was Re: [RFC PATCH 0/6] Deep talk about folio vmap

2025-04-04 Thread Muchun Song
> On Apr 4, 2025, at 17:38, Muchun Song wrote: > > > >> On Apr 4, 2025, at 17:01, Christoph Hellwig wrote: >> >> After the btrfs compressed bio discussion I think the hugetlb changes that >> skip the tail pages are fundamentally unsafe in the current kernel. >> >> That is because the bio_

Re: CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP is broken, was Re: [RFC PATCH 0/6] Deep talk about folio vmap

2025-04-04 Thread Muchun Song
> On Apr 4, 2025, at 17:01, Christoph Hellwig wrote: > > After the btrfs compressed bio discussion I think the hugetlb changes that > skip the tail pages are fundamentally unsafe in the current kernel. > > That is because the bio_vec representation assumes tail pages do exist, so > as soon as

CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP is broken, was Re: [RFC PATCH 0/6] Deep talk about folio vmap

2025-04-04 Thread Christoph Hellwig
After the btrfs compressed bio discussion I think the hugetlb changes that skip the tail pages are fundamentally unsafe in the current kernel. That is because the bio_vec representation assumes tail pages do exist, so as soon as you are doing direct I/O that generates a bvec starting beyond the pr

Re: [RFC PATCH 0/6] Deep talk about folio vmap

2025-03-28 Thread Vishal Moola (Oracle)
On Thu, Mar 27, 2025 at 05:28:27PM +0800, Huan Yang wrote: > Bingbu reported an issue in [1] that udmabuf vmap failed and in [2], we > discussed the scenario of folio vmap due to the misuse of vmap_pfn > in udmabuf. > > We reached the conclusion that vmap_pfn prohibits the use of page-based > PFNs