Re: [PATCH v13 09/35] gpu: host1x: Add runtime PM and OPP support
From: Ulf Hansson <hidden>
Date: 2021-10-01 13:25:22
Also in:
dri-devel, linux-clk, linux-mmc, linux-pm, linux-pwm, linux-staging, linux-tegra, linux-usb, lkml
On Mon, 27 Sept 2021 at 00:42, Dmitry Osipenko [off-list ref] wrote:
Add runtime PM and OPP support to the Host1x driver. For the starter we will keep host1x always-on because dynamic power management require a major refactoring of the driver code since lot's of code paths are missing the RPM handling and we're going to remove some of these paths in the future. Tested-by: Peter Geis <redacted> # Ouya T30 Tested-by: Paul Fertser <fercerpav@gmail.com> # PAZ00 T20 Tested-by: Nicolas Chauvet <redacted> # PAZ00 T20 and TK1 T124 Tested-by: Matt Merhar <redacted> # Ouya T30 Signed-off-by: Dmitry Osipenko <digetx@gmail.com> ---
[...]
quoted hunk ↗ jump to hunk
--- a/drivers/gpu/host1x/dev.c +++ b/drivers/gpu/host1x/dev.c@@ -6,14 +6,18 @@ */ #include <linux/clk.h> +#include <linux/delay.h> #include <linux/dma-mapping.h> #include <linux/io.h> #include <linux/list.h> #include <linux/module.h> #include <linux/of_device.h> #include <linux/of.h> +#include <linux/pm_runtime.h> #include <linux/slab.h> +#include <soc/tegra/common.h> + #define CREATE_TRACE_POINTS #include <trace/events/host1x.h> #undef CREATE_TRACE_POINTS@@ -190,6 +194,9 @@ static void host1x_setup_sid_table(struct host1x *host) const struct host1x_info *info = host->info; unsigned int i; + if (!info->has_hypervisor) + return; + for (i = 0; i < info->num_sid_entries; i++) { const struct host1x_sid_entry *entry = &info->sid_table[i];@@ -347,6 +354,27 @@ static void host1x_iommu_exit(struct host1x *host) } } +static int host1x_get_resets(struct host1x *host) +{ + int err; + + host->resets[0].id = "mc"; + host->resets[1].id = "host1x"; + host->nresets = ARRAY_SIZE(host->resets); + + err = devm_reset_control_bulk_get_optional_exclusive_released( + host->dev, host->nresets, host->resets); + if (err) { + dev_err(host->dev, "failed to get reset: %d\n", err); + return err; + } + + if (WARN_ON(!host->resets[1].rstc)) + return -ENOENT; + + return 0; +} + static int host1x_probe(struct platform_device *pdev) { struct host1x *host;@@ -423,12 +451,9 @@ static int host1x_probe(struct platform_device *pdev) return err; } - host->rst = devm_reset_control_get(&pdev->dev, "host1x"); - if (IS_ERR(host->rst)) { - err = PTR_ERR(host->rst); - dev_err(&pdev->dev, "failed to get reset: %d\n", err); + err = host1x_get_resets(host); + if (err) return err; - } err = host1x_iommu_init(host); if (err < 0) {@@ -443,22 +468,10 @@ static int host1x_probe(struct platform_device *pdev) goto iommu_exit; } - err = clk_prepare_enable(host->clk); - if (err < 0) { - dev_err(&pdev->dev, "failed to enable clock\n"); - goto free_channels; - } - - err = reset_control_deassert(host->rst); - if (err < 0) { - dev_err(&pdev->dev, "failed to deassert reset: %d\n", err); - goto unprepare_disable; - } - err = host1x_syncpt_init(host); if (err) { dev_err(&pdev->dev, "failed to initialize syncpts\n"); - goto reset_assert; + goto free_channels; } err = host1x_intr_init(host, syncpt_irq);@@ -467,10 +480,18 @@ static int host1x_probe(struct platform_device *pdev) goto deinit_syncpt; } - host1x_debug_init(host); + pm_runtime_enable(&pdev->dev); + + err = devm_tegra_core_dev_init_opp_table_common(&pdev->dev); + if (err) + goto pm_disable; - if (host->info->has_hypervisor) - host1x_setup_sid_table(host); + /* the driver's code isn't ready yet for the dynamic RPM */ + err = pm_runtime_resume_and_get(&pdev->dev); + if (err) + goto pm_disable; + + host1x_debug_init(host); err = host1x_register(host); if (err < 0)@@ -486,13 +507,14 @@ static int host1x_probe(struct platform_device *pdev) host1x_unregister(host); deinit_debugfs: host1x_debug_deinit(host); + + pm_runtime_put(&pdev->dev);
pm_runtime_put() is asynchronous, so it may not actually succeed to trigger the ->runtime_suspend() callback to be invoked. Thus, this could end up that we leave clocks prepared/enabled when ->probe() fails, for example. I guess pm_runtime_put_sync_suspend() is slightly better. Another option is to call pm_runtime_force_suspend(), but then you must skip the call pm_runtime_disable() afterwards, as that has already been done inside that function.
quoted hunk ↗ jump to hunk
+pm_disable: + pm_runtime_disable(&pdev->dev); + host1x_intr_deinit(host); deinit_syncpt: host1x_syncpt_deinit(host); -reset_assert: - reset_control_assert(host->rst); -unprepare_disable: - clk_disable_unprepare(host->clk); free_channels: host1x_channel_list_free(&host->channel_list); iommu_exit:@@ -507,19 +529,94 @@ static int host1x_remove(struct platform_device *pdev) host1x_unregister(host); host1x_debug_deinit(host); + + pm_runtime_put(&pdev->dev);
Similar comment as in ->probe().
+ pm_runtime_disable(&pdev->dev);
+
host1x_intr_deinit(host);
host1x_syncpt_deinit(host);
- reset_control_assert(host->rst);
- clk_disable_unprepare(host->clk);
host1x_iommu_exit(host);
return 0;
}
+
+ host1x_setup_sid_table(host);
+ host1x_syncpt_restore(host);
+ host1x_intr_start(host);
+
+ return 0;
+
+disable_clk:
+ clk_disable_unprepare(host->clk);
+release_reset:
+ reset_control_bulk_release(host->nresets, host->resets);
+
+ return err;
+}
+
+static const struct dev_pm_ops host1x_pm = {
+ SET_RUNTIME_PM_OPS(host1x_runtime_suspend, host1x_runtime_resume,
+ NULL)
+ /* TODO: add system suspend-resume once driver will be ready for that */
+};[...] Kind regards Uffe