multiple levels of page table is supported, the new default
value would be the PMD size, similar to huge page.
Signed-off-by: Zhu Lingshan
Reported-and-tested-by: Li Jingxiang
---
include/drm/ttm/ttm_bo.h | 4
1 file changed, 4 insertions(+)
diff --git a/include/drm/ttm/ttm_bo.h b/include
Please ignore this patch, V2 will be send out soon
Thanks
Zhu Lingshan
On 5/30/2024 12:02 PM, Huang Rui wrote:
> On Thu, May 30, 2024 at 11:41:04AM +0800, Zhu, Lingshan wrote:
>> ttm page fault handler ttm_bo_vm_fault_reserved() maps
>> TTM_BO_VM_NUM_PREFAULT more pages beforehan
multiple levels of page table is supported, the new default
value would be the PMD size, similar to huge page.
Signed-off-by: Zhu Lingshan
---
include/drm/ttm/ttm_bo.h | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/include/drm/ttm/ttm_bo.h b/include/drm/ttm/ttm_bo.h
index
al driver for it? A more or less useless
sample driver doesn't really qualify for our normal kernel requirements
that infrastructure should have a real user.
Hello Christoph,
I am working on a real hardware driver for it, it's called IFC, I have
posted RFC V1 and will post RFC V2 s
On 10/29/2019 6:42 PM, Jason Wang wrote:
On 2019/10/29 下午3:42, Zhu Lingshan wrote:
+ void (*set_status)(struct mdev_device *mdev, u8 status);
Hi Jason
Is it possible to make set_status() return an u8 or bool, because this
may fail in real hardware. Without a returned code, I am not sure
;
+ void (*set_status)(struct mdev_device *mdev, u8 status);
Hi Jason
Is it possible to make set_status() return an u8 or bool, because this
may fail in real hardware. Without a returned code, I am not sure
whether it is a good idea to set the status | NEED_RESET when fail.
Th
