[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <HE1PR0401MB233186A0CDA481B775B9B92888A29@HE1PR0401MB2331.eurprd04.prod.outlook.com>
Date: Tue, 14 Feb 2023 21:50:51 +0000
From: Frank Li <frank.li@....com>
To: Sean Anderson <sean.anderson@...o.com>,
"M.H. Lian" <minghuan.lian@....com>,
Mingkai Hu <mingkai.hu@....com>, Roy Zang <roy.zang@....com>,
Lorenzo Pieralisi <lpieralisi@...nel.org>,
Rob Herring <robh@...nel.org>,
Krzysztof WilczyĆski <kw@...ux.com>,
Bjorn Helgaas <bhelgaas@...gle.com>,
"open list:PCI DRIVER FOR FREESCALE LAYERSCAPE"
<linuxppc-dev@...ts.ozlabs.org>,
"open list:PCI DRIVER FOR FREESCALE LAYERSCAPE"
<linux-pci@...r.kernel.org>,
"moderated list:PCI DRIVER FOR FREESCALE LAYERSCAPE"
<linux-arm-kernel@...ts.infradead.org>,
open list <linux-kernel@...r.kernel.org>
CC: "imx@...ts.linux.dev" <imx@...ts.linux.dev>
Subject: RE: [EXT] Re: [PATCH 1/1] PCI: layerscape: Set 64-bit DMA mask
> >
> > + /* set 64-bit DMA mask and coherent DMA mask */
> > + if (dma_set_mask_and_coherent(dev, DMA_BIT_MASK(64)))
> > + dev_warn(dev, "Failed to set 64-bit DMA mask.\n");
>
> Isn't this mandatory? Why not dev_err_probe and return the error?
I don't think it is mandatory. If failure, dma will use swiotlb.
Just an additional memcpy involved.
Frank Li
>
> --Sean
>
> > platform_set_drvdata(pdev, pcie);
> >
> > ret = dw_pcie_ep_init(&pci->ep);
Powered by blists - more mailing lists