* [PATCH v3] Add support for duty-cycle for RCG @ 2021-04-25 7:08 Taniya Das 2021-04-25 7:08 ` [PATCH v3] clk: qcom: clk-rcg2: " Taniya Das 2021-04-28 0:19 ` [PATCH v3] " Stephen Boyd 0 siblings, 2 replies; 6+ messages in thread From: Taniya Das @ 2021-04-25 7:08 UTC (permalink / raw) To: Stephen Boyd, Michael Turquette Cc: Rajendra Nayak, linux-arm-msm, linux-soc, linux-clk, linux-kernel, Taniya Das The root clock generators with MND divider has the capability to support change in duty-cycle by updating the 'D'. Add the clock ops which would check all the boundary conditions and enable setting the desired duty-cycle as per the consumer. [v3] * Implement clockops for get_duty_cycle. * Return -EINVAL for Non-MND or HID RCGs. Taniya Das (1): clk: qcom: clk-rcg2: Add support for duty-cycle for RCG drivers/clk/qcom/clk-rcg2.c | 81 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 81 insertions(+) -- Qualcomm INDIA, on behalf of Qualcomm Innovation Center, Inc.is a member of the Code Aurora Forum, hosted by the Linux Foundation. ^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH v3] clk: qcom: clk-rcg2: Add support for duty-cycle for RCG 2021-04-25 7:08 [PATCH v3] Add support for duty-cycle for RCG Taniya Das @ 2021-04-25 7:08 ` Taniya Das 2021-05-26 3:36 ` Stephen Boyd 2021-04-28 0:19 ` [PATCH v3] " Stephen Boyd 1 sibling, 1 reply; 6+ messages in thread From: Taniya Das @ 2021-04-25 7:08 UTC (permalink / raw) To: Stephen Boyd, Michael Turquette Cc: Rajendra Nayak, linux-arm-msm, linux-soc, linux-clk, linux-kernel, Taniya Das The root clock generators with MND divider has the capability to support change in duty-cycle by updating the 'D'. Add the clock ops which would check all the boundary conditions and enable setting the desired duty-cycle as per the consumer. Signed-off-by: Taniya Das <tdas@codeaurora.org> --- drivers/clk/qcom/clk-rcg2.c | 81 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 81 insertions(+) diff --git a/drivers/clk/qcom/clk-rcg2.c b/drivers/clk/qcom/clk-rcg2.c index 05ff3b0..aa03e31 100644 --- a/drivers/clk/qcom/clk-rcg2.c +++ b/drivers/clk/qcom/clk-rcg2.c @@ -357,6 +357,83 @@ static int clk_rcg2_set_floor_rate_and_parent(struct clk_hw *hw, return __clk_rcg2_set_rate(hw, rate, FLOOR); } +static int clk_rcg2_get_duty_cycle(struct clk_hw *hw, struct clk_duty *duty) +{ + struct clk_rcg2 *rcg = to_clk_rcg2(hw); + u32 notn_m_val, n_val, m_val, d_val, not2d_val, mask; + + if (!rcg->mnd_width) { + /* 50 % duty-cycle for Non-MND RCGs */ + duty->num = 1; + duty->den = 2; + return 0; + } + + regmap_read(rcg->clkr.regmap, RCG_D_OFFSET(rcg), ¬2d_val); + regmap_read(rcg->clkr.regmap, RCG_M_OFFSET(rcg), &m_val); + regmap_read(rcg->clkr.regmap, RCG_N_OFFSET(rcg), ¬n_m_val); + + if (!not2d_val && !m_val && !notn_m_val) { + /* 50 % duty-cycle always */ + duty->num = 1; + duty->den = 2; + return 0; + } + + mask = BIT(rcg->mnd_width) - 1; + + d_val = ~(not2d_val) & mask; + d_val = DIV_ROUND_CLOSEST(d_val, 2); + + n_val = (~(notn_m_val) + m_val) & mask; + + duty->num = d_val; + duty->den = n_val; + + return 0; +} + +static int clk_rcg2_set_duty_cycle(struct clk_hw *hw, struct clk_duty *duty) +{ + struct clk_rcg2 *rcg = to_clk_rcg2(hw); + u32 notn_m_val, n_val, m_val, d_val, not2d_val, mask, duty_per; + int ret; + + /* Duty-cycle cannot be modified for non-MND RCGs */ + if (!rcg->mnd_width) + return -EINVAL; + + mask = BIT(rcg->mnd_width) - 1; + + regmap_read(rcg->clkr.regmap, RCG_N_OFFSET(rcg), ¬n_m_val); + regmap_read(rcg->clkr.regmap, RCG_M_OFFSET(rcg), &m_val); + + n_val = (~(notn_m_val) + m_val) & mask; + + duty_per = (duty->num * 100) / duty->den; + + /* Calculate 2d value */ + d_val = DIV_ROUND_CLOSEST(n_val * duty_per * 2, 100); + + /* Check bit widths of 2d. If D is too big reduce duty cycle. */ + if (d_val > mask) + d_val = mask; + + if ((d_val / 2) > (n_val - m_val)) + d_val = (n_val - m_val) * 2; + else if ((d_val / 2) < (m_val / 2)) + d_val = m_val; + + not2d_val = ~d_val & mask; + + ret = regmap_update_bits(rcg->clkr.regmap, RCG_D_OFFSET(rcg), mask, + not2d_val); + if (ret) + return ret; + + return update_config(rcg); +} + const struct clk_ops clk_rcg2_ops = { .is_enabled = clk_rcg2_is_enabled, .get_parent = clk_rcg2_get_parent, @@ -365,6 +442,8 @@ const struct clk_ops clk_rcg2_ops = { .determine_rate = clk_rcg2_determine_rate, .set_rate = clk_rcg2_set_rate, .set_rate_and_parent = clk_rcg2_set_rate_and_parent, + .get_duty_cycle = clk_rcg2_get_duty_cycle, + .set_duty_cycle = clk_rcg2_set_duty_cycle, }; EXPORT_SYMBOL_GPL(clk_rcg2_ops); @@ -376,6 +455,8 @@ const struct clk_ops clk_rcg2_floor_ops = { .determine_rate = clk_rcg2_determine_floor_rate, .set_rate = clk_rcg2_set_floor_rate, .set_rate_and_parent = clk_rcg2_set_floor_rate_and_parent, + .get_duty_cycle = clk_rcg2_get_duty_cycle, + .set_duty_cycle = clk_rcg2_set_duty_cycle, }; EXPORT_SYMBOL_GPL(clk_rcg2_floor_ops); -- Qualcomm INDIA, on behalf of Qualcomm Innovation Center, Inc.is a member of the Code Aurora Forum, hosted by the Linux Foundation. ^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH v3] clk: qcom: clk-rcg2: Add support for duty-cycle for RCG 2021-04-25 7:08 ` [PATCH v3] clk: qcom: clk-rcg2: " Taniya Das @ 2021-05-26 3:36 ` Stephen Boyd 0 siblings, 0 replies; 6+ messages in thread From: Stephen Boyd @ 2021-05-26 3:36 UTC (permalink / raw) To: Michael Turquette, Taniya Das Cc: Rajendra Nayak, linux-arm-msm, linux-soc, linux-clk, linux-kernel, Taniya Das Quoting Taniya Das (2021-04-25 00:08:22) > The root clock generators with MND divider has the capability to support > change in duty-cycle by updating the 'D'. Add the clock ops which would > check all the boundary conditions and enable setting the desired duty-cycle > as per the consumer. > > Signed-off-by: Taniya Das <tdas@codeaurora.org> > --- Applied to clk-next with '_val' removed everywhere as it made it super hard to read. I also don't like the (d / 2) stuff but I can live with it. ---8<--- diff --git a/drivers/clk/qcom/clk-rcg2.c b/drivers/clk/qcom/clk-rcg2.c index aa03e315d891..e1b1b426fae4 100644 --- a/drivers/clk/qcom/clk-rcg2.c +++ b/drivers/clk/qcom/clk-rcg2.c @@ -360,7 +360,7 @@ static int clk_rcg2_set_floor_rate_and_parent(struct clk_hw *hw, static int clk_rcg2_get_duty_cycle(struct clk_hw *hw, struct clk_duty *duty) { struct clk_rcg2 *rcg = to_clk_rcg2(hw); - u32 notn_m_val, n_val, m_val, d_val, not2d_val, mask; + u32 notn_m, n, m, d, not2d, mask; if (!rcg->mnd_width) { /* 50 % duty-cycle for Non-MND RCGs */ @@ -369,11 +369,11 @@ static int clk_rcg2_get_duty_cycle(struct clk_hw *hw, struct clk_duty *duty) return 0; } - regmap_read(rcg->clkr.regmap, RCG_D_OFFSET(rcg), ¬2d_val); - regmap_read(rcg->clkr.regmap, RCG_M_OFFSET(rcg), &m_val); - regmap_read(rcg->clkr.regmap, RCG_N_OFFSET(rcg), ¬n_m_val); + regmap_read(rcg->clkr.regmap, RCG_D_OFFSET(rcg), ¬2d); + regmap_read(rcg->clkr.regmap, RCG_M_OFFSET(rcg), &m); + regmap_read(rcg->clkr.regmap, RCG_N_OFFSET(rcg), ¬n_m); - if (!not2d_val && !m_val && !notn_m_val) { + if (!not2d && !m && !notn_m) { /* 50 % duty-cycle always */ duty->num = 1; duty->den = 2; @@ -382,13 +382,13 @@ static int clk_rcg2_get_duty_cycle(struct clk_hw *hw, struct clk_duty *duty) mask = BIT(rcg->mnd_width) - 1; - d_val = ~(not2d_val) & mask; - d_val = DIV_ROUND_CLOSEST(d_val, 2); + d = ~(not2d) & mask; + d = DIV_ROUND_CLOSEST(d, 2); - n_val = (~(notn_m_val) + m_val) & mask; + n = (~(notn_m) + m) & mask; - duty->num = d_val; - duty->den = n_val; + duty->num = d; + duty->den = n; return 0; } @@ -396,7 +396,7 @@ static int clk_rcg2_get_duty_cycle(struct clk_hw *hw, struct clk_duty *duty) static int clk_rcg2_set_duty_cycle(struct clk_hw *hw, struct clk_duty *duty) { struct clk_rcg2 *rcg = to_clk_rcg2(hw); - u32 notn_m_val, n_val, m_val, d_val, not2d_val, mask, duty_per; + u32 notn_m, n, m, d, not2d, mask, duty_per; int ret; /* Duty-cycle cannot be modified for non-MND RCGs */ @@ -405,29 +405,29 @@ static int clk_rcg2_set_duty_cycle(struct clk_hw *hw, struct clk_duty *duty) mask = BIT(rcg->mnd_width) - 1; - regmap_read(rcg->clkr.regmap, RCG_N_OFFSET(rcg), ¬n_m_val); - regmap_read(rcg->clkr.regmap, RCG_M_OFFSET(rcg), &m_val); + regmap_read(rcg->clkr.regmap, RCG_N_OFFSET(rcg), ¬n_m); + regmap_read(rcg->clkr.regmap, RCG_M_OFFSET(rcg), &m); - n_val = (~(notn_m_val) + m_val) & mask; + n = (~(notn_m) + m) & mask; duty_per = (duty->num * 100) / duty->den; /* Calculate 2d value */ - d_val = DIV_ROUND_CLOSEST(n_val * duty_per * 2, 100); + d = DIV_ROUND_CLOSEST(n * duty_per * 2, 100); /* Check bit widths of 2d. If D is too big reduce duty cycle. */ - if (d_val > mask) - d_val = mask; + if (d > mask) + d = mask; - if ((d_val / 2) > (n_val - m_val)) - d_val = (n_val - m_val) * 2; - else if ((d_val / 2) < (m_val / 2)) - d_val = m_val; + if ((d / 2) > (n - m)) + d = (n - m) * 2; + else if ((d / 2) < (m / 2)) + d = m; - not2d_val = ~d_val & mask; + not2d = ~d & mask; ret = regmap_update_bits(rcg->clkr.regmap, RCG_D_OFFSET(rcg), mask, - not2d_val); + not2d); if (ret) return ret; ^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH v3] Add support for duty-cycle for RCG 2021-04-25 7:08 [PATCH v3] Add support for duty-cycle for RCG Taniya Das 2021-04-25 7:08 ` [PATCH v3] clk: qcom: clk-rcg2: " Taniya Das @ 2021-04-28 0:19 ` Stephen Boyd 2021-04-29 6:55 ` Taniya Das 1 sibling, 1 reply; 6+ messages in thread From: Stephen Boyd @ 2021-04-28 0:19 UTC (permalink / raw) To: Michael Turquette, Taniya Das Cc: Rajendra Nayak, linux-arm-msm, linux-soc, linux-clk, linux-kernel, Taniya Das Quoting Taniya Das (2021-04-25 00:08:21) > The root clock generators with MND divider has the capability to support > change in duty-cycle by updating the 'D'. Add the clock ops which would > check all the boundary conditions and enable setting the desired duty-cycle > as per the consumer. > > [v3] > * Implement clockops for get_duty_cycle. > * Return -EINVAL for Non-MND or HID RCGs. We don't need cover letters for single patches. Please add these details after the dash before the diffstat on the single patch. > > Taniya Das (1): > clk: qcom: clk-rcg2: Add support for duty-cycle for RCG > > drivers/clk/qcom/clk-rcg2.c | 81 +++++++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 81 insertions(+) > > -- > Qualcomm INDIA, on behalf of Qualcomm Innovation Center, Inc.is a member > of the Code Aurora Forum, hosted by the Linux Foundation. > ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v3] Add support for duty-cycle for RCG 2021-04-28 0:19 ` [PATCH v3] " Stephen Boyd @ 2021-04-29 6:55 ` Taniya Das 2021-04-30 2:21 ` Stephen Boyd 0 siblings, 1 reply; 6+ messages in thread From: Taniya Das @ 2021-04-29 6:55 UTC (permalink / raw) To: Stephen Boyd, Michael Turquette Cc: Rajendra Nayak, linux-arm-msm, linux-soc, linux-clk, linux-kernel Thank you for your review. On 4/28/2021 5:49 AM, Stephen Boyd wrote: > Quoting Taniya Das (2021-04-25 00:08:21) >> The root clock generators with MND divider has the capability to support >> change in duty-cycle by updating the 'D'. Add the clock ops which would >> check all the boundary conditions and enable setting the desired duty-cycle >> as per the consumer. >> >> [v3] >> * Implement clockops for get_duty_cycle. >> * Return -EINVAL for Non-MND or HID RCGs. > > We don't need cover letters for single patches. Please add these details > after the dash before the diffstat on the single patch. > Sure Stephen, will take care from next time. Let me know in case I need to re-submit the patch again. >> >> Taniya Das (1): >> clk: qcom: clk-rcg2: Add support for duty-cycle for RCG >> >> drivers/clk/qcom/clk-rcg2.c | 81 +++++++++++++++++++++++++++++++++++++++++++++ >> 1 file changed, 81 insertions(+) >> >> -- >> Qualcomm INDIA, on behalf of Qualcomm Innovation Center, Inc.is a member >> of the Code Aurora Forum, hosted by the Linux Foundation. >> -- QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation. -- ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v3] Add support for duty-cycle for RCG 2021-04-29 6:55 ` Taniya Das @ 2021-04-30 2:21 ` Stephen Boyd 0 siblings, 0 replies; 6+ messages in thread From: Stephen Boyd @ 2021-04-30 2:21 UTC (permalink / raw) To: Michael Turquette, Taniya Das Cc: Rajendra Nayak, linux-arm-msm, linux-soc, linux-clk, linux-kernel Quoting Taniya Das (2021-04-28 23:55:16) > Thank you for your review. > > On 4/28/2021 5:49 AM, Stephen Boyd wrote: > > Quoting Taniya Das (2021-04-25 00:08:21) > >> The root clock generators with MND divider has the capability to support > >> change in duty-cycle by updating the 'D'. Add the clock ops which would > >> check all the boundary conditions and enable setting the desired duty-cycle > >> as per the consumer. > >> > >> [v3] > >> * Implement clockops for get_duty_cycle. > >> * Return -EINVAL for Non-MND or HID RCGs. > > > > We don't need cover letters for single patches. Please add these details > > after the dash before the diffstat on the single patch. > > > > Sure Stephen, will take care from next time. Let me know in case I need > to re-submit the patch again. > Don't think so. Merge window is open so I'll apply it in another week. ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2021-05-26 3:36 UTC | newest] Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2021-04-25 7:08 [PATCH v3] Add support for duty-cycle for RCG Taniya Das 2021-04-25 7:08 ` [PATCH v3] clk: qcom: clk-rcg2: " Taniya Das 2021-05-26 3:36 ` Stephen Boyd 2021-04-28 0:19 ` [PATCH v3] " Stephen Boyd 2021-04-29 6:55 ` Taniya Das 2021-04-30 2:21 ` Stephen Boyd
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).