Re: [PATCH] iommu/dma: fix variable 'cookie' set but not used

[Robin Murphy]

On 06/01/2020 3:27 pm, Qian Cai wrote:
> The commit c18647900ec8 ("iommu/dma: Relax locking in
> iommu_dma_prepare_msi()") introduced a compliation warning,
>
> drivers/iommu/dma-iommu.c: In function 'iommu_dma_prepare_msi':
> drivers/iommu/dma-iommu.c:1206:27: warning: variable 'cookie' set but
> not used [-Wunused-but-set-variable]
>    struct iommu_dma_cookie *cookie;
>                             ^~~~~~

Fair enough... I guess this is a W=1 thing? Either way there's certainly 
no harm in cleaning up.

Acked-by: Robin Murphy <robin.murphy@xxxxxxx>

> Fixes: c18647900ec8 ("iommu/dma: Relax locking in iommu_dma_prepare_msi()")
> Signed-off-by: Qian Cai <cai@xxxxxx>
> ---
>   drivers/iommu/dma-iommu.c | 3 ---
>   1 file changed, 3 deletions(-)
>
> diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c
> index c363294b3bb9..a2e96a5fd9a7 100644
> --- a/drivers/iommu/dma-iommu.c
> +++ b/drivers/iommu/dma-iommu.c
> @@ -1203,7 +1203,6 @@ int iommu_dma_prepare_msi(struct msi_desc *desc, phys_addr_t msi_addr)
>   {
>   	struct device *dev = msi_desc_to_dev(desc);
>   	struct iommu_domain *domain = iommu_get_domain_for_dev(dev);
> -	struct iommu_dma_cookie *cookie;
>   	struct iommu_dma_msi_page *msi_page;
>   	static DEFINE_MUTEX(msi_prepare_lock); /* see below */
>   
> @@ -1212,8 +1211,6 @@ int iommu_dma_prepare_msi(struct msi_desc *desc, phys_addr_t msi_addr)
>   		return 0;
>   	}
>   
> -	cookie = domain->iova_cookie;
> -
>   	/*
>   	 * In fact the whole prepare operation should already be serialised by
>   	 * irq_domain_mutex further up the callchain, but that's pretty subtle