RE: [PATCH v18 07/20] PCI: dwc: endpoint: Add multiple PFs support for dbi2

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi Manivannan,

> From: Manivannan Sadhasivam, Sent: Friday, July 28, 2023 11:35 AM
> 
> On Tue, Jul 25, 2023 at 11:57:34AM +0000, Yoshihiro Shimoda wrote:
> > Hi Manivannan,
> >
> > > From: Manivannan Sadhasivam, Sent: Monday, July 24, 2023 6:25 PM
> > >
> > > On Fri, Jul 21, 2023 at 04:44:39PM +0900, Yoshihiro Shimoda wrote:
> > > > The commit 24ede430fa49 ("PCI: designware-ep: Add multiple PFs support
> > > > for DWC") added .func_conf_select() to get the configuration space of
> > > > different PFs and assumed that the offsets between dbi and dbi2 would
> > > > be the same. However, Renesas R-Car Gen4 PCIe controllers have different
> > > > offsets of function 1: dbi (+0x1000) and dbi2 (+0x800). To get
> > > > the offset for dbi2, add .func_conf_select2() and
> > > > dw_pcie_ep_func_select2().
> > > >
> > >
> > > How about,
> > >
> > > .get_dbi2_offset() and dw_pcie_ep_get_dbi2_offset()?
> >
> > Thank you for your suggestion. I should have shared the following information
> > in the commit log, but dbi2_offset is not depended on the DBI on my environment:
> >
> >  +0x0000 : dbi Function 0
> >  +0x1000 : dbi Function 1
> >  +0x2000 : dbi2 Function 0
> >  +0x2800 : dbi2 Function 1
> >
> > So, on my environment:
> >  - the dbi_base is set to +0x0000..
> >  -- And func_offset of func_no = 1 was 0x1000.
> >  - the dbi_base2 is set to +0x2000.
> >  -- And func_offset2 of function = 1 was 0x0800, not 0x1800.
> >
> > Perhaps, the name of new API should be .func_conf_select_dbi2 instead?
> >                                                         ~~~~~
> 
> "func_conf_select" doesn't look intuitive to me atleast. The idea behind this
> callback is to get the funcion offset based on the supplied function no. So this
> should've been something like, "get_func_offset" and the API should've been
> dw_pcie_ep_get_func_offset().
> 
> Since I do not want you to change the existing naming in this series, I
> suggested to get the next API naming right.

I got it.

> >
> > > This would've been much simpler if dw_pcie_writeX_{dbi/dbi2} APIs accepted the
> > > func_no argument, so that these offset calculations are contained in the API
> > > definitions itself as it should. Then the APIs could just do "func_offset *
> > > func_no" to get DBI base and "(func_offset * func_no) + dbi2_offset" to get DBI2
> > > base, provided these offsets are passed by the vendor drivers.
> >
> > Serge suggested such implementation before [1]
> >
> > [1]
> >
<snip URL>
> >
> 
> Thanks for the link. I missed Serge's suggestion before. But I completely agree
> with him as you can see from my above suggestion. In addition, I also want to
> fix the "func_conf_select" naming as well.
> 
> However, I do not want you to implement the suggestion in this series itself.
> It should be done as a separate cleanup series later.

I got it.

> (I think you both agree to
> that as well).

Yes, I both agree to that.

Best regards,
Yoshihiro Shimoda

> - Mani
> 
> > > It can be done in a separate cleanup series later.
> > >
> > > > Notes that dw_pcie_ep_func_select2() will call .func_conf_select()
> > >
> > > s/Notes/Note
> >
> > I'll fix it.
> >
> > > > if .func_conf_select2() doesn't exist for backward compatibility.
> > > >
> > > > Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@xxxxxxxxxxx>
> > > > ---
> > > >  .../pci/controller/dwc/pcie-designware-ep.c   | 32 ++++++++++++++-----
> > > >  drivers/pci/controller/dwc/pcie-designware.h  |  3 +-
> > > >  2 files changed, 26 insertions(+), 9 deletions(-)
> > > >
> > > > diff --git a/drivers/pci/controller/dwc/pcie-designware-ep.c b/drivers/pci/controller/dwc/pcie-designware-ep.c
> > > > index 1d24ebf9686f..bd57516d5313 100644
> > > > --- a/drivers/pci/controller/dwc/pcie-designware-ep.c
> > > > +++ b/drivers/pci/controller/dwc/pcie-designware-ep.c
> > > > @@ -54,21 +54,35 @@ static unsigned int dw_pcie_ep_func_select(struct dw_pcie_ep *ep, u8 func_no)
> > > >  	return func_offset;
> > > >  }
> > > >
> > > > +static unsigned int dw_pcie_ep_func_select2(struct dw_pcie_ep *ep, u8 func_no)
> > > > +{
> > > > +	unsigned int func_offset = 0;
> > > > +
> > > > +	if (ep->ops->func_conf_select2)
> > > > +		func_offset = ep->ops->func_conf_select2(ep, func_no);
> > > > +	else if (ep->ops->func_conf_select)	/* for backward compatibility */
> > > > +		func_offset = ep->ops->func_conf_select(ep, func_no);
> > > > +
> > > > +	return func_offset;
> > > > +}
> > > > +
> > > >  static void __dw_pcie_ep_reset_bar(struct dw_pcie *pci, u8 func_no,
> > > >  				   enum pci_barno bar, int flags)
> > > >  {
> > > > -	u32 reg;
> > > > -	unsigned int func_offset = 0;
> > > > +	u32 reg, reg_dbi2;
> > > > +	unsigned int func_offset, func_offset_dbi2;
> > >
> > > Please maitain reverse Xmas tree order.
> >
> > I got it.
> >
> > Best regards,
> > Yoshihiro Shimoda
> >
> > > - Mani
> > >
> > > >  	struct dw_pcie_ep *ep = &pci->ep;
> > > >
> > > >  	func_offset = dw_pcie_ep_func_select(ep, func_no);
> > > > +	func_offset_dbi2 = dw_pcie_ep_func_select2(ep, func_no);
> > > >
> > > >  	reg = func_offset + PCI_BASE_ADDRESS_0 + (4 * bar);
> > > > +	reg_dbi2 = func_offset_dbi2 + PCI_BASE_ADDRESS_0 + (4 * bar);
> > > >  	dw_pcie_dbi_ro_wr_en(pci);
> > > > -	dw_pcie_writel_dbi2(pci, reg, 0x0);
> > > > +	dw_pcie_writel_dbi2(pci, reg_dbi2, 0x0);
> > > >  	dw_pcie_writel_dbi(pci, reg, 0x0);
> > > >  	if (flags & PCI_BASE_ADDRESS_MEM_TYPE_64) {
> > > > -		dw_pcie_writel_dbi2(pci, reg + 4, 0x0);
> > > > +		dw_pcie_writel_dbi2(pci, reg_dbi2 + 4, 0x0);
> > > >  		dw_pcie_writel_dbi(pci, reg + 4, 0x0);
> > > >  	}
> > > >  	dw_pcie_dbi_ro_wr_dis(pci);
> > > > @@ -232,13 +246,15 @@ static int dw_pcie_ep_set_bar(struct pci_epc *epc, u8 func_no, u8 vfunc_no,
> > > >  	enum pci_barno bar = epf_bar->barno;
> > > >  	size_t size = epf_bar->size;
> > > >  	int flags = epf_bar->flags;
> > > > -	unsigned int func_offset = 0;
> > > > +	unsigned int func_offset, func_offset_dbi2;
> > > >  	int ret, type;
> > > > -	u32 reg;
> > > > +	u32 reg, reg_dbi2;
> > > >
> > > >  	func_offset = dw_pcie_ep_func_select(ep, func_no);
> > > > +	func_offset_dbi2 = dw_pcie_ep_func_select2(ep, func_no);
> > > >
> > > >  	reg = PCI_BASE_ADDRESS_0 + (4 * bar) + func_offset;
> > > > +	reg_dbi2 = PCI_BASE_ADDRESS_0 + (4 * bar) + func_offset_dbi2;
> > > >
> > > >  	if (!(flags & PCI_BASE_ADDRESS_SPACE))
> > > >  		type = PCIE_ATU_TYPE_MEM;
> > > > @@ -254,11 +270,11 @@ static int dw_pcie_ep_set_bar(struct pci_epc *epc, u8 func_no, u8 vfunc_no,
> > > >
> > > >  	dw_pcie_dbi_ro_wr_en(pci);
> > > >
> > > > -	dw_pcie_writel_dbi2(pci, reg, lower_32_bits(size - 1));
> > > > +	dw_pcie_writel_dbi2(pci, reg_dbi2, lower_32_bits(size - 1));
> > > >  	dw_pcie_writel_dbi(pci, reg, flags);
> > > >
> > > >  	if (flags & PCI_BASE_ADDRESS_MEM_TYPE_64) {
> > > > -		dw_pcie_writel_dbi2(pci, reg + 4, upper_32_bits(size - 1));
> > > > +		dw_pcie_writel_dbi2(pci, reg_dbi2 + 4, upper_32_bits(size - 1));
> > > >  		dw_pcie_writel_dbi(pci, reg + 4, 0);
> > > >  	}
> > > >
> > > > diff --git a/drivers/pci/controller/dwc/pcie-designware.h b/drivers/pci/controller/dwc/pcie-designware.h
> > > > index 812c221b3f7c..94bc20f5f600 100644
> > > > --- a/drivers/pci/controller/dwc/pcie-designware.h
> > > > +++ b/drivers/pci/controller/dwc/pcie-designware.h
> > > > @@ -340,9 +340,10 @@ struct dw_pcie_ep_ops {
> > > >  	 * access for different platform, if different func have different
> > > >  	 * offset, return the offset of func. if use write a register way
> > > >  	 * return a 0, and implement code in callback function of platform
> > > > -	 * driver.
> > > > +	 * driver. The func_conf_select2 is for dbi2.
> > > >  	 */
> > > >  	unsigned int (*func_conf_select)(struct dw_pcie_ep *ep, u8 func_no);
> > > > +	unsigned int (*func_conf_select2)(struct dw_pcie_ep *ep, u8 func_no);
> > > >  };
> > > >
> > > >  struct dw_pcie_ep_func {
> > > > --
> > > > 2.25.1
> > > >
> > >
> > > --
> > > மணிவண்ணன் சதாசிவம்
> 
> --
> மணிவண்ணன் சதாசிவம்




[Index of Archives]     [DMA Engine]     [Linux Coverity]     [Linux USB]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [Greybus]

  Powered by Linux