[PATCH 4/9] nvme-pci: refactor nvme_pci_use_sgls

[Daniel Gomez]

On 10/06/2025 07.06, Christoph Hellwig wrote:
> Move the average segment size into a separate helper, and return a
> tristate to distinguish the case where can use SGL vs where we have to
> use SGLs.  This will allow the simplify the code and make more efficient
> decisions in follow on changes.
> 
> Signed-off-by: Christoph Hellwig <hch at lst.de>
> ---
>  drivers/nvme/host/pci.c | 41 +++++++++++++++++++++++++++--------------
>  1 file changed, 27 insertions(+), 14 deletions(-)
> 
> diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
> index 449a9950b46c..0b85c11d3c96 100644
> --- a/drivers/nvme/host/pci.c
> +++ b/drivers/nvme/host/pci.c

...

> @@ -886,7 +897,9 @@ static blk_status_t nvme_map_data(struct nvme_dev *dev, struct request *req,
>  		goto out_free_sg;
>  	}
>  
> -	if (nvme_pci_use_sgls(dev, req, iod->sgt.nents))
> +	if (use_sgl == SGL_FORCED ||
> +	    (use_sgl == SGL_SUPPORTED &&

FORCED implies SUPPORTED, what about simplifying to:

if (use_sgl >= SGL_SUPPORTED)

or just do:

if (use_sgl)