>do you know which peripherals cannot work with 32bit other than the NFC ?
>From top of my head: SPI: UART: PPI: (Only works with 32-bit DMA on BF561/BF54x, on BF561 PPI_CONTROL DMA32 needs to be set, while on BF54x one of the three Peripheral DMA controllers is 32-bit wide supporting 8/16/32-bit by nature) -Michael >-----Original Message----- >From: Mike Frysinger [mailto:[email protected]] >Sent: Wednesday, February 04, 2009 7:19 PM >To: Hennerich, Michael >Cc: [email protected]; [email protected] >Subject: Re: [Linux-kernel-commits] >[6057]trunk/drivers/mtd/nand/bf5xx_nand.c: bug[4833] The DMAs havedifferent >size on BF52x and BF54x > >On Wed, Feb 4, 2009 at 12:26, Hennerich, Michael wrote: >>>On Wed, Feb 4, 2009 at 11:54, Hennerich, Michael wrote: >>>>>where in the HRM do you see it saying "32-bit dma is not supported" ? >>>> >>>> I don't think it's explicitly written in the NFC chapter of the HRM. >>>> But it's a matter of fact that the DMA DAB is 16-bit on BF52x. >>>> >>>> From the PHRM: >>>> >>>> "The 16-bit DMA Access Bus (DAB) connects the DMA controller to the >>>> on-chip peripherals, PPI, SPI, Ethernet MAC, the SPORTs, NFC, >>>> HOSTDP and the UARTs." >>>> >>>> 32-bit DMA won't work! >>> >>>the sport controller supports 32bit dma (and our drivers are using >>>it). same goes for ethernet mac. >> >> Well right. >> >> Some peripherals support 32-bit DMA over an 16-bit DAB (taking 2x >> cycles) , some not. >> >> NFC is one that doesn't support this. >> >> Whole story - Cliff skyped me this morning about the BF52x NFC DMA >> problem. >> In return I called Andrew C. and asked him about the NFC and associated >> software differences between NFC on 54x and 52x. >> >> Straight came the answer BF52x DMA must be WDSIZE_16 !!! > >which is fine, but now i know i need to hassle the docs guys to make >sure this is clearly noted in our HRMs > >do you know which peripherals cannot work with 32bit other than the NFC ? >-mike _______________________________________________ Linux-kernel-commits mailing list [email protected] http://blackfin.uclinux.org/mailman/listinfo/linux-kernel-commits
