Messages in this thread | | | Date | Tue, 7 Sep 2010 12:21:44 +0900 | Subject | Re: [PATCH] Gigabit Ethernet driver of Topcliff PCH | From | FUJITA Tomonori <> |
| |
On Tue, 7 Sep 2010 10:13:22 +0900 "Masayuki Ohtake" <masa-korg@dsn.okisemi.com> wrote:
> > Probably, this trick doesn't work because zero could be a valid DMA > > address on some architctures. > > > > The flag which shows mapped will be added. > And it is used instead of "if (buffer_info->dma) ".
Sounds fine. That's what other drivers do.
> > > + /*-- Set Buffer infomation --*/ > > > + buffer_info->length = skb->len; > > > + buffer_info->dma = > > > + pci_map_single(adapter->pdev, skb->data, buffer_info->length, > > > + PCI_DMA_TODEVICE); > > > > Needs to handle mapping failure (dma_mapping_errro)? > > Error process will be added.
Nice.
> > Can you use dma_map_* API instead of pci_map_*? Please read: > > > > This DMA of device is not standard. Personal use of Gigabit Ethernet device. > So, I guess that I do not need to use DMA-API.
Sorry, I'm not sure what you mean. If a driver does DMA, it must use the DMA API. Your driver already uses the DMA API (pci_dma_*), however, pci_map_* API is obsolete so please use dma_map_* API instead.
Thanks,
| |