linux-kernel - Re: [PATCH v6 3/4] iommu: debug-pagealloc: Track IOMMU pages

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAFgf54ovfnTHmMuZGk73SEEKsP3k-_exR1wqUE4W9tLYKv_iDw@mail.gmail.com>
Date: Tue, 13 Jan 2026 10:49:28 +0000
From: Mostafa Saleh <smostafa@...gle.com>
To: "David Hildenbrand (Red Hat)" <david@...nel.org>
Cc: Jason Gunthorpe <jgg@...pe.ca>, linux-mm@...ck.org, iommu@...ts.linux.dev, 
	linux-kernel@...r.kernel.org, linux-doc@...r.kernel.org, corbet@....net, 
	joro@...tes.org, will@...nel.org, robin.murphy@....com, 
	akpm@...ux-foundation.org, vbabka@...e.cz, surenb@...gle.com, mhocko@...e.com, 
	jackmanb@...gle.com, hannes@...xchg.org, ziy@...dia.com, 
	lorenzo.stoakes@...cle.com, Liam.Howlett@...cle.com, rppt@...nel.org, 
	xiaqinxin@...wei.com, baolu.lu@...ux.intel.com, rdunlap@...radead.org, 
	Samiullah Khawaja <skhawaja@...gle.com>
Subject: Re: [PATCH v6 3/4] iommu: debug-pagealloc: Track IOMMU pages

On Tue, Jan 13, 2026 at 10:31 AM David Hildenbrand (Red Hat)
<david@...nel.org> wrote:
>
> On 1/13/26 11:22, Mostafa Saleh wrote:
> > On Mon, Jan 12, 2026 at 7:12 PM David Hildenbrand (Red Hat)
> > <david@...nel.org> wrote:
> >>
> >> On 1/12/26 15:58, Mostafa Saleh wrote:
> >>> On Mon, Jan 12, 2026 at 1:52 PM Jason Gunthorpe <jgg@...pe.ca> wrote:
> >>>>
> >>>> On Mon, Jan 12, 2026 at 01:43:41PM +0000, Mostafa Saleh wrote:
> >>>>> But I don’t see why not. from the documentation:
> >>>>> /**
> >>>>>    * pfn_valid - check if there is a valid memory map entry for a PFN
> >>>>>    * @pfn: the page frame number to check
> >>>>>    *
> >>>>>    * Check if there is a valid memory map entry aka struct page for the @pfn.
> >>>>>    * Note, that availability of the memory map entry does not imply that
> >>>>>    * there is actual usable memory at that @pfn. The struct page may
> >>>>>    * represent a hole or an unusable page frame.
> >>>>> …
> >>>>>
> >>>>> That means that struct page exists, which is all what we need here.
> >>>>
> >>>> A struct page that has never been initialize shouldn't ever be read. I
> >>>> don't know how that relates to page_ext, but are you really sure that
> >>>> is all you need?
> >>>>
> >>>
> >>> AFAIU, if pfn_valid() returns true, it means the struct page is valid,
> >>> and lookup_page_ext() will check that a valid page_ext exists for this
> >>> entry.
> >>
> >> Not always. Offline memory blocks have a memory map but no page ext. We
> >> allocate the page ext at memory onlining time.
> >>
> >> Also, I'm not sure about ZONE_DEVICE memory, very likely we never
> >> allocate a page_ext for them?
> >>
> >> I'd assume both cases are not relevant for your use case, though.
> >>
> >
> >  From my understanding, in that case, page_ext_get() will return NULL.
> >
> > So, as long as struct page exists, page_ext_get won't misbehave.
> >
> > I am not sure about offline memory, but MMIO can be used. We use
> > pfn_valid() before getting the struct page that we pass to page_ext.
> > Would that be OK?
> It's tricky. If you look at lookup_page_ext(), it relies on extracting
> the pfn+nid from the "struct page".
>
> If the "struct page" is uninitialized (e.g., offline memory) that cannot
> possibly work, as it could just give you random garbage.
>
> (note that there are two implementations of lookup_page_ext(), both
> extracting the PFN but only one extracting the NID).
>

I see, thanks for the clarification.

I see that pfn_to_online_page() exits, which seems to handle online
memory and ZONE_DEVICE. Would that be a suitable alternative?

Would you have a problem if we added a new function in page_ext
"page_ext_from_phys()" as Jason suggested?

Thanks,
Mostafa


Thanks,
Mostafa

> --
> Cheers
>
> David
>