> Then why isn't there a call like pci_unmap_single()
<<< linux/include/asm-i386/pci.h:
/* Unmap a single streaming mode DMA translation. The dma_addr and size
* must match what was provided for in a previous pci_map_single call.
All
* other usages are undefined.
*
* After this call, reads by the cpu to the buffer are guarenteed to see
* whatever the device wrote there.
*/
static inline void pci_unmap_single(struct pci_dev *hwdev, dma_addr_t
dma_addr,
size_t size, int direction)
<<<<
Please try to store both the dma address and the virtual address in your
driver and use pci_map_single().
I'm aware of one case where that's not possible: if your hardware builds
a linked list of finished commands, and that list is in DMA addresses,
not virtual addresses.
I think both sym53c8xx and one of the USB controllers (usb-ohci.c?) have
that problem, they use a hash table for the reverse dma address->linear
address mapping.
I hope that answers your question.
--
Manfred
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Oct 07 2001 - 21:00:29 EST