On Wed, May 11, 2011 at 06:15:58PM +0100, J. Ali Harlow wrote:
> If I try and use comedi_dio_bitfield2() on my shiny new VM110 board, I
> just get EBUSY back permanently. Implementing the insn_bits call seems
> to fix the problem and is good in its own right since I can now read and
> write to all the digital lines at the same time. Tested on a K8055, but
> I don't have access to a K8061.
>
> Ali.
I need a "Signed-off-by:" line as described in the file,
Documentation/SubmittingPatches in order to be able to accept this
patch.
> diff -Naurd linux-staging-2.6.32.8/drivers/staging/comedi/drivers/vmk80xx.c linux-staging-2.6.32.8-bis/drivers/staging/comedi/drivers/vmk80xx.c
The 2.6.32.8 kernel is _very_ old, can you please generate this patch
against the latest development kernel instead so that we can apply it
properly?
> --- linux-staging-2.6.32.8/drivers/staging/comedi/drivers/vmk80xx.c 2010-02-09 12:57:19.000000000 +0000
> +++ linux-staging-2.6.32.8-bis/drivers/staging/comedi/drivers/vmk80xx.c 2011-05-11 14:19:37.551737828 +0100
> @@ -3,6 +3,7 @@
> Velleman USB Board Low-Level Driver
>
> Copyright (C) 2009 Manuel Gebele <forensixs@xxxxxx>, Germany
> + Copyright (C) 2011 J. Ali Harlow <ali@xxxxxxxxxxxxxxx>
I don't think your changes allow you to add your copyright to the file,
based on the guidelines my lawyers have given me. Do you have different
legal advice that contridicts this?
>
> COMEDI - Linux Control and Measurement Device Interface
> Copyright (C) 2000 David A. Schleef <ds@xxxxxxxxxxx>
> @@ -27,7 +28,7 @@
> Description: Velleman USB Board Low-Level Driver
> Devices: K8055/K8061 aka VM110/VM140
> Author: Manuel Gebele <forensixs@xxxxxx>
> -Updated: Sun, 10 May 2009 11:14:59 +0200
> +Updated: Wed, 11 May 2011 12:44:00 +0100
> Status: works
>
> Supports:
> @@ -41,6 +42,7 @@
> /*
> Changelog:
>
> +0.8.81 -4- support handling DIO lines as a block
The "updated" and "changelog" sections don't matter anymore now that the
drivers are in the main kernel tree, so you don't have to do this.
> 0.8.81 -3- code completely rewritten (adjust driver logic)
> 0.8.81 -2- full support for K8061
> 0.8.81 -1- fix some mistaken among others the number of
> @@ -555,7 +557,8 @@
> if ((dir) & DIR_IN) { \
> if (test_bit(TRANS_IN_BUSY, &dev->flags)) \
> return -EBUSY; \
> - } else { /* DIR_OUT */ \
> + } \
> + if ((dir) & DIR_OUT) { \
> if (test_bit(TRANS_OUT_BUSY, &dev->flags)) \
> return -EBUSY; \
> } \
> @@ -683,6 +686,48 @@
> return n;
> }
>
> +static int vmk80xx_di_bits(struct comedi_device *cdev,
> + struct comedi_subdevice *s,
> + struct comedi_insn *insn, unsigned int *data)
> +{
> + struct vmk80xx_usb *dev = cdev->private;
> + unsigned char *rx_buf;
> + int reg;
> + int retval;
> +
> + dbgvm("vmk80xx: %s\n", __func__);
> +
> + rudimentary_check(DIR_IN);
> +
> + down(&dev->limit_sem);
> +
> + rx_buf = dev->usb_rx_buf;
> +
> + if (dev->board.model == VMK8061_MODEL) {
> + reg = VMK8061_DI_REG;
> + dev->usb_tx_buf[0] = VMK8061_CMD_RD_DI;
> + } else
> + reg = VMK8055_DI_REG;
> +
> + retval = vmk80xx_read_packet(dev);
> +
> + if (!retval)
> + {
Please run your patch through the scripts/checkpatch.pl tool and fix the
issues it finds before making us do the same thing and pestering you to
make them :)
Could you make these changes and resend the patch so we can apply it to
the tree?
thanks,
greg k-h
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
