On Thu, Oct 22, 2015 at 07:18:19PM -0400, ira.weiny wrote:
> This follows the rest of the style of the case statement in this function. We
> prefer to leave this as is for a number of reasons.
>
> 1) This is consistent with the coding style elsewhere in this driver.
Don't try to match existing s
On Thu, Oct 22, 2015 at 01:41:58PM +0300, Dan Carpenter wrote:
> On Mon, Oct 19, 2015 at 10:11:29PM -0400, ira.we...@intel.com wrote:
> > + case HFI1_CMD_TID_INVAL_READ:
> > + ret = hfi1_user_exp_rcv_invalid(fp, &tinfo);
> > + if (!ret) {
> > + addr = (unsign
On Mon, Oct 19, 2015 at 10:11:29PM -0400, ira.we...@intel.com wrote:
> + case HFI1_CMD_TID_INVAL_READ:
> + ret = hfi1_user_exp_rcv_invalid(fp, &tinfo);
> + if (!ret) {
> + addr = (unsigned long)cmd.addr +
> + offsetof(struc
From: Mitko Haralanov
Expected receives work by user-space libraries (PSM) calling into the driver
with information about the user's receive buffer and have the driver DMA-map
that buffer and program the HFI to receive data directly into it.
This is an expensive operation as it requires the driv
