Message-ID: <a641d93e-adec-a92b-1b2c-949a79cb4f3a@quicinc.com>
Date:   Tue, 31 May 2022 16:14:48 -0700
From:   Abhinav Kumar <quic_abhinavk@...cinc.com>
To:     Douglas Anderson <dianders@...omium.org>,
        Rob Clark <robdclark@...il.com>,
        Dmitry Baryshkov <dmitry.baryshkov@...aro.org>
CC:     AngeloGioacchino Del Regno 
        <angelogioacchino.delregno@...ainline.org>,
        Bjorn Andersson <bjorn.andersson@...aro.org>,
        Daniel Vetter <daniel@...ll.ch>,
        David Airlie <airlied@...ux.ie>,
        Kalyan Thota <quic_kalyant@...cinc.com>,
        Sean Paul <sean@...rly.run>,
        Stephen Boyd <swboyd@...omium.org>,
        Vinod Polimera <quic_vpolimer@...cinc.com>,
        <dri-devel@...ts.freedesktop.org>,
        <freedreno@...ts.freedesktop.org>, <linux-arm-msm@...r.kernel.org>,
        <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2] drm/msm/dpu: Move min BW request and full BW disable
 back to mdss

Hi Doug

On 5/31/2022 4:01 PM, Douglas Anderson wrote:
> In commit a670ff578f1f ("drm/msm/dpu: always use mdp device to scale
> bandwidth") we fully moved interconnect stuff to the DPU driver. This
> had no change for sc7180 but _did_ have an impact for other SoCs. It
> made them match the sc7180 scheme.
> 
> Unfortunately, the sc7180 scheme seems like it was a bit broken.
> Specifically the interconnect needs to be on for more than just the
> DPU driver's AXI bus. In the very least it also needs to be on for the
> DSI driver's AXI bus. This can be seen fairly easily by doing this on
> a ChromeOS sc7180-trogdor class device:
> 
>    set_power_policy --ac_screen_dim_delay=5 --ac_screen_off_delay=10
>    sleep 10
>    cd /sys/bus/platform/devices/ae94000.dsi/power
>    echo on > control
> 
> When you do that, you'll get a warning splat in the logs about
> "gcc_disp_hf_axi_clk status stuck at 'off'".
> 
> One could argue that perhaps what I have done above is "illegal" and
> that it can't happen naturally in the system because in normal system
> usage the DPU is pretty much always on when DSI is on. That being
> said:
> * In official ChromeOS builds (admittedly a 5.4 kernel with backports)
>    we have seen that splat at bootup.
> * Even though we don't use "autosuspend" for these components, we
>    don't use the "put_sync" variants. Thus plausibly the DSI could stay
>    "runtime enabled" past when the DPU is enabled. Techncially we
>    shouldn't do that if the DPU's suspend ends up yanking our clock.
> 
> Let's change things such that the "bare minimum" request for the
> interconnect happens in the mdss driver again. That means that all of
> the children can assume that the interconnect is on at the minimum
> bandwidth. We'll then let the DPU request the higher amount that it
> wants.
> 
> It should be noted that this isn't as hacky of a solution as it might
> initially appear. Specifically:
> * Since MDSS and DPU individually get their own references to the
>    interconnect then the framework will actually handle aggregating
>    them. The two drivers are _not_ clobbering each other.
> * When the Qualcomm interconnect driver aggregates it takes the max of
>    all the peaks. Thus having MDSS request a peak, as we're doing here,
>    won't actually change the total interconnect bandwidth (it won't be
>    added to the request for the DPU). This perhaps explains why the
>    "average" requested in MDSS was historically 0 since that one
>    _would_ be added in.
> 
> NOTE also that in the downstream ChromeOS 5.4 and 5.15 kernels, we're
> also seeing some RPMH hangs that are addressed by this fix. These
> hangs are showing up in the field and on _some_ devices with enough
> stress testing of suspend/resume. Specifically right at suspend time
> with a stack crawl that looks like this (from chromeos-5.15 tree):
>    rpmh_write_batch+0x19c/0x240
>    qcom_icc_bcm_voter_commit+0x210/0x420
>    qcom_icc_set+0x28/0x38
>    apply_constraints+0x70/0xa4
>    icc_set_bw+0x150/0x24c
>    dpu_runtime_resume+0x50/0x1c4
>    pm_generic_runtime_resume+0x30/0x44
>    __genpd_runtime_resume+0x68/0x7c
>    genpd_runtime_resume+0x12c/0x20c
>    __rpm_callback+0x98/0x138
>    rpm_callback+0x30/0x88
>    rpm_resume+0x370/0x4a0
>    __pm_runtime_resume+0x80/0xb0
>    dpu_kms_enable_commit+0x24/0x30
>    msm_atomic_commit_tail+0x12c/0x630
>    commit_tail+0xac/0x150
>    drm_atomic_helper_commit+0x114/0x11c
>    drm_atomic_commit+0x68/0x78
>    drm_atomic_helper_disable_all+0x158/0x1c8
>    drm_atomic_helper_suspend+0xc0/0x1c0
>    drm_mode_config_helper_suspend+0x2c/0x60
>    msm_pm_prepare+0x2c/0x40
>    pm_generic_prepare+0x30/0x44
>    genpd_prepare+0x80/0xd0
>    device_prepare+0x78/0x17c
>    dpm_prepare+0xb0/0x384
>    dpm_suspend_start+0x34/0xc0
> 
> We don't completely understand all the mechanisms in play, but the
> hang seemed to come and go with random factors. It's not terribly
> surprising that the hang is gone after this patch since the line of
> code that was failing is no longer present in the kernel.
> 
> Fixes: a670ff578f1f ("drm/msm/dpu: always use mdp device to scale bandwidth")
> Fixes: c33b7c0389e1 ("drm/msm/dpu: add support for clk and bw scaling for display")
> Signed-off-by: Douglas Anderson <dianders@...omium.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@...cinc.com>

We will test this out even on RB3/RB5 to make sure it boots up fine and 
give Tested-by.

Thanks

Abhinav

> ---
> 
> Changes in v2:
> - Don't set bandwidth in init.
> 
>   drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c |  8 ----
>   drivers/gpu/drm/msm/msm_mdss.c          | 57 +++++++++++++++++++++++++
>   2 files changed, 57 insertions(+), 8 deletions(-)
> 
> diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
> index 2b9d931474e0..3025184053e0 100644
> --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
> +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
> @@ -49,8 +49,6 @@
>   #define DPU_DEBUGFS_DIR "msm_dpu"
>   #define DPU_DEBUGFS_HWMASKNAME "hw_log_mask"
>   
> -#define MIN_IB_BW	400000000ULL /* Min ib vote 400MB */
> -
>   static int dpu_kms_hw_init(struct msm_kms *kms);
>   static void _dpu_kms_mmu_destroy(struct dpu_kms *dpu_kms);
>   
> @@ -1303,15 +1301,9 @@ static int __maybe_unused dpu_runtime_resume(struct device *dev)
>   	struct dpu_kms *dpu_kms = to_dpu_kms(priv->kms);
>   	struct drm_encoder *encoder;
>   	struct drm_device *ddev;
> -	int i;
>   
>   	ddev = dpu_kms->dev;
>   
> -	WARN_ON(!(dpu_kms->num_paths));
> -	/* Min vote of BW is required before turning on AXI clk */
> -	for (i = 0; i < dpu_kms->num_paths; i++)
> -		icc_set_bw(dpu_kms->path[i], 0, Bps_to_icc(MIN_IB_BW));
> -
>   	rc = clk_bulk_prepare_enable(dpu_kms->num_clocks, dpu_kms->clocks);
>   	if (rc) {
>   		DPU_ERROR("clock enable failed rc:%d\n", rc);
> diff --git a/drivers/gpu/drm/msm/msm_mdss.c b/drivers/gpu/drm/msm/msm_mdss.c
> index 0454a571adf7..e13c5c12b775 100644
> --- a/drivers/gpu/drm/msm/msm_mdss.c
> +++ b/drivers/gpu/drm/msm/msm_mdss.c
> @@ -5,6 +5,7 @@
>   
>   #include <linux/clk.h>
>   #include <linux/delay.h>
> +#include <linux/interconnect.h>
>   #include <linux/irq.h>
>   #include <linux/irqchip.h>
>   #include <linux/irqdesc.h>
> @@ -25,6 +26,8 @@
>   #define UBWC_CTRL_2			0x150
>   #define UBWC_PREDICTION_MODE		0x154
>   
> +#define MIN_IB_BW	400000000UL /* Min ib vote 400MB */
> +
>   struct msm_mdss {
>   	struct device *dev;
>   
> @@ -36,8 +39,47 @@ struct msm_mdss {
>   		unsigned long enabled_mask;
>   		struct irq_domain *domain;
>   	} irq_controller;
> +	struct icc_path *path[2];
> +	u32 num_paths;
>   };
>   
> +static int msm_mdss_parse_data_bus_icc_path(struct device *dev,
> +					    struct msm_mdss *msm_mdss)
> +{
> +	struct icc_path *path0 = of_icc_get(dev, "mdp0-mem");
> +	struct icc_path *path1 = of_icc_get(dev, "mdp1-mem");
> +
> +	if (IS_ERR_OR_NULL(path0))
> +		return PTR_ERR_OR_ZERO(path0);
> +
> +	msm_mdss->path[0] = path0;
> +	msm_mdss->num_paths = 1;
> +
> +	if (!IS_ERR_OR_NULL(path1)) {
> +		msm_mdss->path[1] = path1;
> +		msm_mdss->num_paths++;
> +	}
> +
> +	return 0;
> +}
> +
> +static void msm_mdss_put_icc_path(void *data)
> +{
> +	struct msm_mdss *msm_mdss = data;
> +	int i;
> +
> +	for (i = 0; i < msm_mdss->num_paths; i++)
> +		icc_put(msm_mdss->path[i]);
> +}
> +
> +static void msm_mdss_icc_request_bw(struct msm_mdss *msm_mdss, unsigned long bw)
> +{
> +	int i;
> +
> +	for (i = 0; i < msm_mdss->num_paths; i++)
> +		icc_set_bw(msm_mdss->path[i], 0, Bps_to_icc(bw));
> +}
> +
>   static void msm_mdss_irq(struct irq_desc *desc)
>   {
>   	struct msm_mdss *msm_mdss = irq_desc_get_handler_data(desc);
> @@ -136,6 +178,13 @@ static int msm_mdss_enable(struct msm_mdss *msm_mdss)
>   {
>   	int ret;
>   
> +	/*
> +	 * Several components have AXI clocks that can only be turned on if
> +	 * the interconnect is enabled (non-zero bandwidth). Let's make sure
> +	 * that the interconnects are at least at a minimum amount.
> +	 */
> +	msm_mdss_icc_request_bw(msm_mdss, MIN_IB_BW);
> +
>   	ret = clk_bulk_prepare_enable(msm_mdss->num_clocks, msm_mdss->clocks);
>   	if (ret) {
>   		dev_err(msm_mdss->dev, "clock enable failed, ret:%d\n", ret);
> @@ -178,6 +227,7 @@ static int msm_mdss_enable(struct msm_mdss *msm_mdss)
>   static int msm_mdss_disable(struct msm_mdss *msm_mdss)
>   {
>   	clk_bulk_disable_unprepare(msm_mdss->num_clocks, msm_mdss->clocks);
> +	msm_mdss_icc_request_bw(msm_mdss, 0);
>   
>   	return 0;
>   }
> @@ -271,6 +321,13 @@ static struct msm_mdss *msm_mdss_init(struct platform_device *pdev, bool is_mdp5
>   
>   	dev_dbg(&pdev->dev, "mapped mdss address space @%pK\n", msm_mdss->mmio);
>   
> +	ret = msm_mdss_parse_data_bus_icc_path(&pdev->dev, msm_mdss);
> +	if (ret)
> +		return ERR_PTR(ret);
> +	ret = devm_add_action_or_reset(&pdev->dev, msm_mdss_put_icc_path, msm_mdss);
> +	if (ret)
> +		return ERR_PTR(ret);
> +
>   	if (is_mdp5)
>   		ret = mdp5_mdss_parse_clock(pdev, &msm_mdss->clocks);
>   	else

Powered by blists - more mailing lists