We don't have a real use-case for this. Understanding the consequences,
we are questioning the patch in downstream itself.
Please ignore this patch for now.
On 12/12/18 1:36 PM, Christoph Hellwig wrote:
scatterlist elements longer than 4GB sound odd. Please submit it
in a series with your
Hi Krishna,
Thank you for the patch! Yet something to improve:
[auto build test ERROR on linus/master]
[also build test ERROR on v4.20-rc6 next-20181214]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system]
url:
Hi Krishna,
Thank you for the patch! Yet something to improve:
[auto build test ERROR on linus/master]
[also build test ERROR on v4.20-rc6 next-20181214]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system]
url:
scatterlist elements longer than 4GB sound odd. Please submit it
in a series with your actual user so that we can help figuring out
if it really makes sense or if there is a better way to solve your
problem.
As is this patch will massively increase the memory usage for all users
of struct
On Wed, 12 Dec 2018 at 07:25, Ashish Mhetre wrote:
>
> From: Krishna Reddy
>
> In the cases where greater than 4GB allocations are required, current
> definition of scatterlist doesn't support it. For example, Tegra devices
> have more than 4GB of system memory available. So they are expected to
On 12/12/18 12:19 PM, Sagi Grimberg wrote:
struct nvme_sgl_desc {
__le64 addr;
- __le32 length;
+ __le64 length;
__u8 rsvd[3];
__u8 type;
};
Isn't this a device or protocol defined datastructure? You can't just
change it like this.
You're
struct nvme_sgl_desc {
__le64 addr;
- __le32 length;
+ __le64 length;
__u8rsvd[3];
__u8type;
};
Isn't this a device or protocol defined datastructure? You can't just
change it like this.
You're correct, we can't...
[Replied before seeing
struct nvme_sgl_desc {
__le64 addr;
- __le32 length;
+ __le64 length;
__u8rsvd[3];
__u8type;
};
in what world changing a wire protocol for this make sense?
please get rid of this hunk, NVMe will never cross the 32 bit sg element
size.
From: Ashish Mhetre
Date: Wed, 12 Dec 2018 11:54:13 +0530
> diff --git a/include/linux/nvme.h b/include/linux/nvme.h
> index 68e91ef..0a07a29 100644
> --- a/include/linux/nvme.h
> +++ b/include/linux/nvme.h
> @@ -587,7 +587,7 @@ enum {
>
> struct nvme_sgl_desc {
> __le64 addr;
> -
From: Krishna Reddy
In the cases where greater than 4GB allocations are required, current
definition of scatterlist doesn't support it. For example, Tegra devices
have more than 4GB of system memory available. So they are expected to
support larger allocation requests.
This patch updates the
10 matches
Mail list logo
