On 12/09/21 at 02:10pm, David Hildenbrand wrote:
..
> >>> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> >>> index c5952749ad40..ac0ea42a4e5f 100644
> >>> --- a/mm/page_alloc.c
> >>> +++ b/mm/page_alloc.c
> >>> @@ -9459,4 +9459,15 @@ bool take_page_off_buddy(struct page *page)
> >>>
On 09.12.21 14:02, Baoquan He wrote:
> On 12/07/21 at 12:23pm, David Hildenbrand wrote:
>> On 07.12.21 04:07, Baoquan He wrote:
>>> In some places of the current kernel, it assumes that dma zone must have
>>> managed pages if CONFIG_ZONE_DMA is enabled. While this is not always true.
>>> E.g in
On 12/07/21 at 12:23pm, David Hildenbrand wrote:
> On 07.12.21 04:07, Baoquan He wrote:
> > In some places of the current kernel, it assumes that dma zone must have
> > managed pages if CONFIG_ZONE_DMA is enabled. While this is not always true.
> > E.g in kdump kernel of x86_64, only low 1M is
On 07.12.21 04:07, Baoquan He wrote:
> In some places of the current kernel, it assumes that dma zone must have
> managed pages if CONFIG_ZONE_DMA is enabled. While this is not always true.
> E.g in kdump kernel of x86_64, only low 1M is presented and locked down
> at very early stage of boot, so
On 12/6/21 9:07 PM, Baoquan He wrote:
In some places of the current kernel, it assumes that dma zone must have
managed pages if CONFIG_ZONE_DMA is enabled. While this is not always true.
E.g in kdump kernel of x86_64, only low 1M is presented and locked down
at very early stage of boot, so that
In some places of the current kernel, it assumes that dma zone must have
managed pages if CONFIG_ZONE_DMA is enabled. While this is not always true.
E.g in kdump kernel of x86_64, only low 1M is presented and locked down
at very early stage of boot, so that there's no managed pages at all in
DMA
