linux-kernel - Re: [PATCH v2] PCI: endpoint: Use DMA channel's 'dev' for dma_map

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <5589ccbc-af34-e5ff-ff2e-f694c4ea696e@ti.com>
Date:   Tue, 4 Jan 2022 14:34:29 +0530
From:   Kishon Vijay Abraham I <kishon@...com>
To:     Lorenzo Pieralisi <lorenzo.pieralisi@....com>
CC:     <linux-pci@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        Rob Herring <robh@...nel.org>,
        Krzysztof Wilczyński <kw@...ux.com>,
        Bjorn Helgaas <bhelgaas@...gle.com>
Subject: Re: [PATCH v2] PCI: endpoint: Use DMA channel's 'dev' for
 dma_map_single()

Hi Lorenzo,

On 16/11/21 7:53 pm, Kishon Vijay Abraham I wrote:
> For the case where the pci-epf-test driver uses DMA for transferring
> data to the root complex device, dma_map_single() is used to map virtual
> address to a physical address (address accessible by DMA controller) and
> provided to the DMAengine API for transferring data. Here instead of
> using the PCIe endpoint controller's 'dev' for dma_map_single(), provide
> DMA channel's 'dev' for dma_map_single() since the data transfer is
> actually done by DMA.
> 
> Signed-off-by: Kishon Vijay Abraham I <kishon@...com>

Can this patch be merged?

Thanks,
Kishon

> ---
> Changes from v1:
> Use dmaengine_get_dma_device() to get dma device from channel
> V1: https://lore.kernel.org/r/20211115044944.31103-1-kishon@ti.com
>  drivers/pci/endpoint/functions/pci-epf-test.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/pci/endpoint/functions/pci-epf-test.c b/drivers/pci/endpoint/functions/pci-epf-test.c
> index 90d84d3bc868..51f5b0b7b225 100644
> --- a/drivers/pci/endpoint/functions/pci-epf-test.c
> +++ b/drivers/pci/endpoint/functions/pci-epf-test.c
> @@ -314,12 +314,12 @@ static int pci_epf_test_read(struct pci_epf_test *epf_test)
>  	u32 crc32;
>  	bool use_dma;
>  	phys_addr_t phys_addr;
> +	struct device *dma_dev;
>  	phys_addr_t dst_phys_addr;
>  	struct timespec64 start, end;
>  	struct pci_epf *epf = epf_test->epf;
>  	struct device *dev = &epf->dev;
>  	struct pci_epc *epc = epf->epc;
> -	struct device *dma_dev = epf->epc->dev.parent;
>  	enum pci_barno test_reg_bar = epf_test->test_reg_bar;
>  	struct pci_epf_test_reg *reg = epf_test->reg[test_reg_bar];
>  
> @@ -353,6 +353,7 @@ static int pci_epf_test_read(struct pci_epf_test *epf_test)
>  			goto err_dma_map;
>  		}
>  
> +		dma_dev = dmaengine_get_dma_device(epf_test->dma_chan);
>  		dst_phys_addr = dma_map_single(dma_dev, buf, reg->size,
>  					       DMA_FROM_DEVICE);
>  		if (dma_mapping_error(dma_dev, dst_phys_addr)) {
> @@ -402,12 +403,12 @@ static int pci_epf_test_write(struct pci_epf_test *epf_test)
>  	void *buf;
>  	bool use_dma;
>  	phys_addr_t phys_addr;
> +	struct device *dma_dev;
>  	phys_addr_t src_phys_addr;
>  	struct timespec64 start, end;
>  	struct pci_epf *epf = epf_test->epf;
>  	struct device *dev = &epf->dev;
>  	struct pci_epc *epc = epf->epc;
> -	struct device *dma_dev = epf->epc->dev.parent;
>  	enum pci_barno test_reg_bar = epf_test->test_reg_bar;
>  	struct pci_epf_test_reg *reg = epf_test->reg[test_reg_bar];
>  
> @@ -444,6 +445,7 @@ static int pci_epf_test_write(struct pci_epf_test *epf_test)
>  			goto err_map_addr;
>  		}
>  
> +		dma_dev = dmaengine_get_dma_device(epf_test->dma_chan);
>  		src_phys_addr = dma_map_single(dma_dev, buf, reg->size,
>  					       DMA_TO_DEVICE);
>  		if (dma_mapping_error(dma_dev, src_phys_addr)) {
>