[EXT] Re: [PATCH] Revert "phy: cadence-torrent: Do not configure SERDES if it's already configured"

[Li Chen]

Hi, Vinod

> -----Original Message-----
> From: Vinod Koul [mailto:vkoul at kernel.org]
> Sent: Thursday, December 23, 2021 8:45 PM
> To: Li Chen
> Cc: Kishon Vijay Abraham I; Philipp Zabel; Swapnil Jakhade; linux-
> phy at lists.infradead.org; linux-kernel at vger.kernel.org; Dan Carpenter
> Subject: Re: [EXT] Re: [PATCH] Revert "phy: cadence-torrent: Do not configure
> SERDES if it's already configured"
> 
> On 23-12-21, 06:27, Li Chen wrote:
> > Hi, Vinod
> >
> > > -----Original Message-----
> > > From: Vinod Koul [mailto:vkoul at kernel.org]
> > > Sent: Thursday, December 23, 2021 1:27 PM
> > > To: Li Chen
> > > Cc: Kishon Vijay Abraham I; Philipp Zabel; Swapnil Jakhade; linux-
> > > phy at lists.infradead.org; linux-kernel at vger.kernel.org; Dan Carpenter
> > > Subject: [EXT] Re: [PATCH] Revert "phy: cadence-torrent: Do not configure
> > > SERDES if it's already configured"
> > >
> > > On 26-11-21, 05:06, Li Chen wrote:
> > > > This reverts commit
> > > > b69d39f68419("phy: cadence-torrent: Do not configure SERDES if it's already
> > > configured")
> > >
> > > space between commit id and open brace...
> > >
> > > >
> > > > our soc will hang on any regmap field read before reset.
> > >
> > > okay, in this case the right fix would be to keep track of reset in SW
> > > and still skip reset if it is already configured?
> > >
> >
> > I should be grateful if you would give me more details of reset in SW.
> 
> Store the reset state in a driver variable reset and use that for
> finding already_configured rather than reading a hw value

Sorry for my late reply. From commit d44b4bf49e0b ("phy: cadence-torrent: Group reset APIs and clock APIs "):
> No functional change intended. Group reset APIs and clock APIs in
preparation for adding support to skip configuration if the SERDES
is already configured by bootloader.

So, the already configured detection is to detect configured by bootload and I cannot know its state without reading the hw value from linux driver, so I don't think storing variable reset in sw inside linux driver is the right way.

> 
> >
> > > >
> > > > Signed-off-by: Li Chen <lchen at ambarella.com>
> > > > ---
> > > >  drivers/phy/cadence/phy-cadence-torrent.c | 31 +++++++----------------
> > > >  1 file changed, 9 insertions(+), 22 deletions(-)
> > > >
> > > > diff --git a/drivers/phy/cadence/phy-cadence-torrent.c
> > > b/drivers/phy/cadence/phy-cadence-torrent.c
> > > > index 415ace64adc5c..e57e0b1523aff 100644
> > > > --- a/drivers/phy/cadence/phy-cadence-torrent.c
> > > > +++ b/drivers/phy/cadence/phy-cadence-torrent.c
> > > > @@ -2031,11 +2031,6 @@ static int cdns_torrent_noop_phy_on(struct phy
> > > *phy)
> > > >  	return 0;
> > > >  }
> > > >
> > > > -static const struct phy_ops noop_ops = {
> > > > -	.power_on	= cdns_torrent_noop_phy_on,
> > > > -	.owner		= THIS_MODULE,
> > > > -};
> > > > -
> > > >  static
> > > >  int cdns_torrent_phy_configure_multilink(struct cdns_torrent_phy
> *cdns_phy)
> > > >  {
> > > > @@ -2282,7 +2277,6 @@ static int cdns_torrent_phy_probe(struct
> > > platform_device *pdev)
> > > >  	struct device_node *child;
> > > >  	int ret, subnodes, node = 0, i;
> > > >  	u32 total_num_lanes = 0;
> > > > -	int already_configured;
> > > >  	u8 init_dp_regmap = 0;
> > > >  	u32 phy_type;
> > > >
> > > > @@ -2321,20 +2315,16 @@ static int cdns_torrent_phy_probe(struct
> > > platform_device *pdev)
> > > >  	if (ret)
> > > >  		return ret;
> > > >
> > > > -	regmap_field_read(cdns_phy->phy_pma_cmn_ctrl_1,
> > > &already_configured);
> > > > -
> > > > -	if (!already_configured) {
> > > > -		ret = cdns_torrent_reset(cdns_phy);
> > > > -		if (ret)
> > > > -			goto clk_cleanup;
> > > > +	ret = cdns_torrent_reset(cdns_phy);
> > > > +	if (ret)
> > > > +		goto clk_cleanup;
> > > >
> > > > -		ret = cdns_torrent_clk(cdns_phy);
> > > > -		if (ret)
> > > > -			goto clk_cleanup;
> > > > +	ret = cdns_torrent_clk(cdns_phy);
> > > > +	if (ret)
> > > > +		goto clk_cleanup;
> > > >
> > > >  		/* Enable APB */
> > > > -		reset_control_deassert(cdns_phy->apb_rst);
> > > > -	}
> > > > +	reset_control_deassert(cdns_phy->apb_rst);
> > > >
> > > >  	for_each_available_child_of_node(dev->of_node, child) {
> > > >  		struct phy *gphy;
> > > > @@ -2404,10 +2394,7 @@ static int cdns_torrent_phy_probe(struct
> > > platform_device *pdev)
> > > >  		of_property_read_u32(child, "cdns,ssc-mode",
> > > >  				     &cdns_phy->phys[node].ssc_mode);
> > > >
> > > > -		if (!already_configured)
> > > > -			gphy = devm_phy_create(dev, child,
> > > &cdns_torrent_phy_ops);
> > > > -		else
> > > > -			gphy = devm_phy_create(dev, child, &noop_ops);
> > > > +		gphy = devm_phy_create(dev, child, &cdns_torrent_phy_ops);
> > > >  		if (IS_ERR(gphy)) {
> > > >  			ret = PTR_ERR(gphy);
> > > >  			goto put_child;
> > > > @@ -2490,7 +2477,7 @@ static int cdns_torrent_phy_probe(struct
> > > platform_device *pdev)
> > > >  		goto put_lnk_rst;
> > > >  	}
> > > >
> > > > -	if (cdns_phy->nsubnodes > 1 && !already_configured) {
> > > > +	if (cdns_phy->nsubnodes > 1) {
> > > >  		ret = cdns_torrent_phy_configure_multilink(cdns_phy);
> > > >  		if (ret)
> > > >  			goto put_lnk_rst;
> > > > --
> > > > 2.33.1
> > > >
> > > >
> > > >
> > >
> **************************************************************
> > > ********
> > > > This email and attachments contain Ambarella Proprietary and/or
> Confidential
> > > Information and is intended solely for the use of the individual(s) to whom it
> is
> > > addressed. Any unauthorized review, use, disclosure, distribute, copy, or
> print is
> > > prohibited. If you are not an intended recipient, please contact the sender by
> > > reply email and destroy all copies of the original message. Thank you.
> 
> You do realize this is not OK in public emails... Pls fix
> 
> > >
> > > Aha!!!
> > >
> > > Okay destroyed this now!
> > >
> > > --
> > > ~Vinod
> > >
> > >
> ##############################################################
> > > ########
> > > This EXTERNAL email has been scanned by Proofpoint Email Protect service.
> >
> > Regards,
> > Li
> 
> --
> ~Vinod