[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <aEdroxwd/TOyxxIr@MiWiFi-R3L-srv>
Date: Tue, 10 Jun 2025 07:17:55 +0800
From: Baoquan He <bhe@...hat.com>
To: Alexander Graf <graf@...zon.com>,
Andrew Morton <akpm@...ux-foundation.org>
Cc: kexec@...ts.infradead.org, linux-kernel@...r.kernel.org,
Pasha Tatashin <pasha.tatashin@...een.com>,
nh-open-source@...zon.com,
Zhongkun He <hezhongkun.hzk@...edance.com>, Bcc: bhe@...hat.com;
Subject: Re: [PATCH v4] kexec: Enable CMA based contiguous allocation
On 06/09/25 at 09:03am, Alexander Graf wrote:
>
> On 21.05.25 17:29, Alexander Graf wrote:
......
> > Using CMA allocations has two advantages:
> >
> > 1) Faster by 4-50 ms per 100 MiB. There is no more need to copy in the
> > hot phase.
> > 2) More robust. Even if by accident some page is still in use for DMA,
> > the new kernel image will be safe from that access because it resides
> > in a memory region that is considered allocated in the old kernel and
> > has a chance to reinitialize that component.
> >
> > Signed-off-by: Alexander Graf <graf@...zon.com>
>
>
> Ping? Not seeing this patch in Linus' master tree :)
I saw Andrew had added this to his mm-nonmem-unstable, it could be
dropped later because of new version posting.
I would like to ACK this patch, let's see how it's going. One thing is
if IOMMU is on, the devices of no .shutdown method could still be on
flight and look up the IO page table and corrupt kernel, that's another
story.
Acked-by: Baoquan He <bhe@...hat.com>
Powered by blists - more mailing lists