[U-Boot] [PATCH] arm: socfpga: set the fpga global bit to disable HPS to FPGA signals

[Chin Liang See]

On Thu, 2016-01-07 at 16:33 +0100, Marek Vasut wrote:
> On Thursday, January 07, 2016 at 04:23:09 PM, Dinh Nguyen wrote:
> > On 01/07/2016 09:03 AM, Dinh Nguyen wrote:
> > > On 01/06/2016 09:20 PM, Marek Vasut wrote:
> > > > On Thursday, January 07, 2016 at 03:50:00 AM, Dinh Nguyen
> > > > wrote:
> > > > > On 01/06/2016 08:21 PM, Marek Vasut wrote:
> > > > > > On Wednesday, January 06, 2016 at 08:48:43 PM,
> > > > > > dinguyen at opensource.altera.com
> > > > > > 
> > > > > > wrote:
> > > > > > > From: Dinh Nguyen <dinh.linux at gmail.com>
> > > > > > > 
> > > > > > > We should be setting the FPGA Interface Group global bit
> > > > > > > that will
> > > > > > > correctly disable all interfaces between the FPGA and
> > > > > > > HPS.
> > > > > > > 
> > > > > > > Signed-off-by: Dinh Nguyen <
> > > > > > > dinguyen at opensource.altera.com>
> > > > > > > ---
> > > > > > > 
> > > > > > >  drivers/fpga/socfpga.c | 2 +-
> > > > > > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > > > > > > 
> > > > > > > diff --git a/drivers/fpga/socfpga.c
> > > > > > > b/drivers/fpga/socfpga.c
> > > > > > > index 4448250..431e159 100644
> > > > > > > --- a/drivers/fpga/socfpga.c
> > > > > > > +++ b/drivers/fpga/socfpga.c
> > > > > > > @@ -269,7 +269,7 @@ int socfpga_load(Altera_desc *desc,
> > > > > > > const void
> > > > > > > *rbf_data, size_t rbf_size) /* Prior programming the
> > > > > > > FPGA, all
> > > > > > > bridges need to be shut off */
> > > > > > > 
> > > > > > >  	/* Disable all signals from hps peripheral
> > > > > > > controller to fpga */
> > > > > > > 
> > > > > > > -	writel(0, &sysmgr_regs->fpgaintfgrp_module);
> > > > > > > +	writel(0, &sysmgr_regs->fpgaintfgrp_gbl);
> > > > > > > 
> > > > > > >  	/* Disable all signals from FPGA to HPS SDRAM */
> > > > > > >  
> > > > > > >  #define SDR_CTRLGRP_FPGAPORTRST_ADDRESS	0x5080
> > > > > > 
> > > > > > Looks fine,
> > > > > > 
> > > > > > Acked-by: Marek Vasut <marex at denx.de>
> > > > > > 
> > > > > > btw Shouldn't you also fix the same thing in
> > > > > > drivers/fpga/socfpga.c ?
> > > > > > Even
> > > > > 
> > > > > I'm confused by what you mean here. The patch is for
> > > > > drivers/fpga/socfpga.c.
> > > > 
> > > > Oops, sorry, there is another one in arch/arm/mach
> > > > -socfpga/misc.c . I
> > > > mispasted the wrong filename.
> > > 
> > > Ah yes, then we should probably add a function to enable/disable
> > > the
> > > interfaces.
> > 
> > But I wonder if I wanted to enable the individual interfaces, i.e.
> > the
> > separate EMACs that is in fpgaintfgrp_module, would I need to
> > enable the
> > fpgaintfgrp_gbl bit?
> 
> Aren't you the one working for Altera? You're the one who should know
> the best 
> :-) But I think this gbl bit is like the top-level killswitch, so you
> need to
> enable this one and only then you can fine-tune config of the
> subblocks (like
> the EMACs etc).

We would need to enable the global bit to have the individual interface
works. Each interface enable signal is a result of AND between
indvidual enable register and global enable register.

Thanks
Chin Liang

> 
> Best regards,
> Marek Vasut