linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Thinh Nguyen <Thinh.Nguyen@synopsys.com>
To: Sean Anderson <sean.anderson@seco.com>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	"linux-usb@vger.kernel.org" <linux-usb@vger.kernel.org>
Cc: Felipe Balbi <balbi@kernel.org>,
	Thinh Nguyen <Thinh.Nguyen@synopsys.com>,
	Balaji Prakash J <bjagadee@codeaurora.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Robert Hancock <robert.hancock@calian.com>,
	Baruch Siach <baruch@tkos.co.il>
Subject: Re: [PATCH v2 4/7] usb: dwc3: Program GFLADJ
Date: Mon, 24 Jan 2022 22:46:59 +0000	[thread overview]
Message-ID: <4696c5a4-5921-f7cb-196c-5ad956e696f9@synopsys.com> (raw)
In-Reply-To: <20220119002438.106079-5-sean.anderson@seco.com>

Sean Anderson wrote:
> GUCTL.REFCLKPER can only account for clock frequencies with integer
> periods. To address this, program REFCLK_FLADJ with the relative error
> caused by period truncation. The formula given in the register reference
> has been rearranged to allow calculation based on rate (instead of
> period), and to allow for fixed-point arithmetic.
> 
> Additionally, calculate a value for 240MHZDECR. This configures a
> simulated 240Mhz clock using a counter with one fractional bit (PLS1).
> 
> This register is programmed only for versions >= 2.50a, since this is
> the check also used by commit db2be4e9e30c ("usb: dwc3: Add frame length
> adjustment quirk").
> 
> Signed-off-by: Sean Anderson <sean.anderson@seco.com>
> ---
> 
> Changes in v2:
> - Also program GFLADJ.240MHZDECR
> - Don't program GFLADJ if the version is < 2.50a
> 
>  drivers/usb/dwc3/core.c | 37 +++++++++++++++++++++++++++++++++++--
>  drivers/usb/dwc3/core.h |  3 +++
>  2 files changed, 38 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
> index 5214daceda86..883e119377f0 100644
> --- a/drivers/usb/dwc3/core.c
> +++ b/drivers/usb/dwc3/core.c
> @@ -348,7 +348,7 @@ static void dwc3_frame_length_adjustment(struct dwc3 *dwc)
>  static void dwc3_ref_clk_period(struct dwc3 *dwc)
>  {
>  	u32 reg;
> -	unsigned long rate, period;
> +	unsigned long decr, fladj, rate, period;

Minor style nit: Felipe prefers to keep the declaration on separate
lines, let's keep it consistent with the rest in this driver.

>  
>  	if (dwc->ref_clk) {
>  		rate = clk_get_rate(dwc->ref_clk);
> @@ -357,6 +357,7 @@ static void dwc3_ref_clk_period(struct dwc3 *dwc)
>  		period = NSEC_PER_SEC / rate;
>  	} else if (dwc->ref_clk_per) {
>  		period = dwc->ref_clk_per;
> +		rate = NSEC_PER_SEC / period;
>  	} else {
>  		return;
>  	}
> @@ -365,9 +366,41 @@ static void dwc3_ref_clk_period(struct dwc3 *dwc)
>  	reg &= ~DWC3_GUCTL_REFCLKPER_MASK;
>  	reg |=  FIELD_PREP(DWC3_GUCTL_REFCLKPER_MASK, period);
>  	dwc3_writel(dwc->regs, DWC3_GUCTL, reg);
> +
> +	if (DWC3_VER_IS_PRIOR(DWC3, 250A))
> +		return;
> +
> +	/*
> +	 * The calculation below is
> +	 *
> +	 * 125000 * (NSEC_PER_SEC / (rate * period) - 1)
> +	 *
> +	 * but rearranged for fixed-point arithmetic.
> +	 *
> +	 * Note that rate * period ~= NSEC_PER_SECOND, minus the number of
> +	 * nanoseconds of error caused by the truncation which happened during
> +	 * the division when calculating rate or period (whichever one was
> +	 * derived from the other). We first calculate the relative error, then
> +	 * scale it to units of 0.08%.
> +	 */
> +	fladj = div64_u64(125000ULL * NSEC_PER_SEC, (u64)rate * period);

Can we rearrange the math so we don't have to operate on different data
type and deal with conversion/truncation?

> +	fladj -= 125000;
> +
> +	/*
> +	 * The documented 240MHz constant is scaled by 2 to get PLS1 as well.
> +	 */
> +	decr = 480000000 / rate;
> +
> +	reg = dwc3_readl(dwc->regs, DWC3_GFLADJ);
> +	reg &= ~DWC3_GFLADJ_REFCLK_FLADJ_MASK
> +	    &  ~DWC3_GFLADJ_240MHZDECR
> +	    &  ~DWC3_GFLADJ_240MHZDECR_PLS1;
> +	reg |= FIELD_PREP(DWC3_GFLADJ_REFCLK_FLADJ_MASK, fladj)
> +	    |  FIELD_PREP(DWC3_GFLADJ_240MHZDECR, decr >> 1)
> +	    |  FIELD_PREP(DWC3_GFLADJ_240MHZDECR_PLS1, decr & 1);

Does this pass checkpatch?

> +	dwc3_writel(dwc->regs, DWC3_GFLADJ, reg);
>  }
>  
> -
>  /**
>   * dwc3_free_one_event_buffer - Frees one event buffer
>   * @dwc: Pointer to our controller context structure
> diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h
> index 45cfa7d9f27a..eb9c1efced05 100644
> --- a/drivers/usb/dwc3/core.h
> +++ b/drivers/usb/dwc3/core.h
> @@ -388,6 +388,9 @@
>  /* Global Frame Length Adjustment Register */
>  #define DWC3_GFLADJ_30MHZ_SDBND_SEL		BIT(7)
>  #define DWC3_GFLADJ_30MHZ_MASK			0x3f
> +#define DWC3_GFLADJ_REFCLK_FLADJ_MASK		GENMASK(21, 8)
> +#define DWC3_GFLADJ_240MHZDECR			GENMASK(30, 24)
> +#define DWC3_GFLADJ_240MHZDECR_PLS1		BIT(31)
>  
>  /* Global User Control Register*/
>  #define DWC3_GUCTL_REFCLKPER_MASK		0xffc00000
Thanks,
Thinh

  parent reply	other threads:[~2022-01-24 23:37 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-19  0:24 [PATCH v2 0/7] usb: dwc3: Calculate REFCLKPER et. al. from reference clock Sean Anderson
2022-01-19  0:24 ` [PATCH v2 1/7] dt-bindings: usb: dwc3: Deprecate snps,ref-clock-period-ns Sean Anderson
2022-01-19  0:24 ` [PATCH v2 2/7] usb: dwc3: Get clocks individually Sean Anderson
2022-01-20 16:52   ` Robert Hancock
2022-01-19  0:24 ` [PATCH v2 3/7] usb: dwc3: Calculate REFCLKPER based on reference clock Sean Anderson
2022-01-20 16:53   ` Robert Hancock
2022-01-19  0:24 ` [PATCH v2 4/7] usb: dwc3: Program GFLADJ Sean Anderson
2022-01-20 16:55   ` Robert Hancock
2022-01-24 22:46   ` Thinh Nguyen [this message]
2022-01-24 23:06     ` Sean Anderson
2022-01-25  2:11       ` Thinh Nguyen
2022-01-25  6:17         ` Felipe Balbi
2022-01-25 20:02           ` Thinh Nguyen
2022-01-25 16:22         ` Sean Anderson
2022-01-25 19:36           ` Thinh Nguyen
2022-01-26 10:56             ` Felipe Balbi
2022-01-26 10:53           ` Felipe Balbi
2022-01-27 16:45             ` Sean Anderson
2022-01-19  0:24 ` [PATCH v2 5/7] usb: dwc3: Add snps,ref-clock-frequency-hz property for ACPI Sean Anderson
2022-01-24 22:44   ` Thinh Nguyen
2022-01-24 23:07     ` Sean Anderson
2022-01-24 23:14       ` Sean Anderson
2022-01-25  2:21         ` Thinh Nguyen
2022-01-19  0:24 ` [PATCH v2 6/7] arm64: dts: zynqmp: Move USB clocks to dwc3 node Sean Anderson
2022-01-20 16:56   ` Robert Hancock
2022-01-19  0:24 ` [PATCH v2 7/7] arm64: dts: ipq6018: Use reference clock to set dwc3 period Sean Anderson
2022-01-19 18:14 ` [PATCH v2 0/7] usb: dwc3: Calculate REFCLKPER et. al. from reference clock Baruch Siach
2022-01-19 18:23   ` Sean Anderson
2022-01-20  5:23   ` Kathiravan T
2022-01-20 10:29     ` Baruch Siach
2022-01-24 15:11       ` Kathiravan Thirumoorthy
2022-01-24 23:01         ` Thinh Nguyen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4696c5a4-5921-f7cb-196c-5ad956e696f9@synopsys.com \
    --to=thinh.nguyen@synopsys.com \
    --cc=balbi@kernel.org \
    --cc=baruch@tkos.co.il \
    --cc=bjagadee@codeaurora.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-usb@vger.kernel.org \
    --cc=robert.hancock@calian.com \
    --cc=sean.anderson@seco.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).