CCing Sascha On 20/06/2019 11:35, Matthias Brugger wrote: > > > On 13/06/2019 10:14, Pi-Hsun Shih wrote: >> Hi, >> When I tested this patch series (Based on linux 5.2.0-rc2, and with >> various other patch series about MT8183) with lockdep enabled, and I'm >> seeing the following lockdep warning on boot. >> >> By bisecting the commits, the first commit that introduce this warning >> is this patch. The warning also doesn't appear if >> https://lore.kernel.org/patchwork/patch/1086582/ and >> https://lore.kernel.org/patchwork/patch/1086583/ are not applied. >> >> Do anyone have idea on why this is happening, or any suggestion on >> which part I should be digging into to figure this out? Thanks. >> >> [ 4.664194] ====================================================== >> [ 4.670368] WARNING: possible circular locking dependency detected >> [ 4.676545] 5.2.0-rc2-next-20190528-44527-g6c94b6475c04 #20 Tainted: G S >> [ 4.684539] ------------------------------------------------------ >> [ 4.690714] kworker/4:1/51 is trying to acquire lock: >> [ 4.695760] (____ptrval____) (regulator_list_mutex){+.+.}, >> at:regulator_lock_dependent+0xdc/0x6c4 >> [ 4.704732] >> [ 4.704732] but task is already holding lock: >> [ 4.710556] (____ptrval____) (&genpd->mlock/1){+.+.}, >> at:genpd_lock_nested_mtx+0x24/0x30 >> [ 4.718740] >> [ 4.718740] which lock already depends on the new lock. >> [ 4.718740] >> [ 4.726908] >> [ 4.726908] the existing dependency chain (in reverse order) is: >> [ 4.734382] >> [ 4.734382] -> #4 (&genpd->mlock/1){+.+.}: >> [ 4.739963] __mutex_lock_common+0x1a0/0x1fe8 >> [ 4.744836] mutex_lock_nested+0x40/0x50 >> [ 4.749275] genpd_lock_nested_mtx+0x24/0x30 >> [ 4.754063] genpd_add_subdomain+0x150/0x524 >> [ 4.758850] pm_genpd_add_subdomain+0x3c/0x5c >> [ 4.763723] scpsys_probe+0x520/0xe78 >> [ 4.767902] platform_drv_probe+0xf4/0x134 >> [ 4.772517] really_probe+0x214/0x4dc >> [ 4.776696] driver_probe_device+0xcc/0x1d4 >> [ 4.781396] __device_attach_driver+0x10c/0x180 >> [ 4.786442] bus_for_each_drv+0x124/0x184 >> [ 4.790968] __device_attach+0x1c0/0x2d8 >> [ 4.795407] device_initial_probe+0x20/0x2c >> [ 4.800106] bus_probe_device+0x80/0x16c >> [ 4.804546] deferred_probe_work_func+0x120/0x168 >> [ 4.809767] process_one_work+0x858/0x1208 >> [ 4.814379] worker_thread+0x9ec/0xcb8 >> [ 4.818644] kthread+0x2b8/0x2d0 >> [ 4.822391] ret_from_fork+0x10/0x18 >> [ 4.826480] >> [ 4.826480] -> #3 (&genpd->mlock){+.+.}: >> [ 4.831880] __mutex_lock_common+0x1a0/0x1fe8 >> [ 4.836752] mutex_lock_nested+0x40/0x50 >> [ 4.841190] genpd_lock_mtx+0x20/0x2c >> [ 4.845369] genpd_runtime_resume+0x140/0x434 >> [ 4.850241] __rpm_callback+0xb0/0x1e4 >> [ 4.854506] rpm_callback+0x54/0x1a8 >> [ 4.858597] rpm_resume+0xc6c/0x10c4 >> [ 4.862689] __pm_runtime_resume+0xb4/0x124 >> [ 4.867387] device_link_add+0x598/0x8d0 > > For this looks as if you have also patch > [PATCH v2 04/12] memory: mtk-smi: Add device-link between smi-larb and smi-common > from series > [PATCH v2 00/12] Clean up "mediatek,larb" after adding device_link > applied. > My guess is, that we run into this, because we call genpd_runtime_resume before we have finished the subdomain registration in scpsys_probe. That implies that you also have the scpsys series for mt8183 added to your tree. Anyway it looks to me as if we have to add the subdomains before we call of_genpd_add_provider_onecell(). Regards, Matthias > Regards, > Matthias > >> [ 4.871829] mtk_smi_larb_probe+0x2b0/0x340 >> [ 4.876528] platform_drv_probe+0xf4/0x134 >> [ 4.881141] really_probe+0x214/0x4dc >> [ 4.885320] driver_probe_device+0xcc/0x1d4 >> [ 4.890020] __device_attach_driver+0x10c/0x180 >> [ 4.895066] bus_for_each_drv+0x124/0x184 >> [ 4.899591] __device_attach+0x1c0/0x2d8 >> [ 4.904031] device_initial_probe+0x20/0x2c >> [ 4.908730] bus_probe_device+0x80/0x16c >> [ 4.913169] deferred_probe_work_func+0x120/0x168 >> [ 4.918387] process_one_work+0x858/0x1208 >> [ 4.923000] worker_thread+0x9ec/0xcb8 >> [ 4.927264] kthread+0x2b8/0x2d0 >> [ 4.931009] ret_from_fork+0x10/0x18 >> [ 4.935098] >> [ 4.935098] -> #2 (dpm_list_mtx){+.+.}: >> [ 4.940412] __mutex_lock_common+0x1a0/0x1fe8 >> [ 4.945284] mutex_lock_nested+0x40/0x50 >> [ 4.949722] device_pm_lock+0x1c/0x24 >> [ 4.953900] device_link_add+0x98/0x8d0 >> [ 4.958252] _regulator_get+0x3f0/0x504 >> [ 4.962606] _devm_regulator_get+0x58/0xb8 >> [ 4.967218] devm_regulator_get+0x28/0x34 >> [ 4.971746] pwm_backlight_probe+0x61c/0x1b90 >> [ 4.976617] platform_drv_probe+0xf4/0x134 >> [ 4.981230] really_probe+0x214/0x4dc >> [ 4.985409] driver_probe_device+0xcc/0x1d4 >> [ 4.990108] device_driver_attach+0xe4/0x104 >> [ 4.994894] __driver_attach+0x134/0x14c >> [ 4.999333] bus_for_each_dev+0x120/0x180 >> [ 5.003859] driver_attach+0x48/0x54 >> [ 5.007950] bus_add_driver+0x2ac/0x44c >> [ 5.012303] driver_register+0x160/0x288 >> [ 5.016742] __platform_driver_register+0xcc/0xdc >> [ 5.021964] pwm_backlight_driver_init+0x1c/0x24 >> [ 5.027097] do_one_initcall+0x38c/0x994 >> [ 5.031536] do_initcall_level+0x3a4/0x4b8 >> [ 5.036148] do_basic_setup+0x84/0xa0 >> [ 5.036153] kernel_init_freeable+0x23c/0x324 >> [ 5.036158] kernel_init+0x14/0x110 >> [ 5.036164] ret_from_fork+0x10/0x18 >> [ 5.036166] >> [ 5.036166] -> #1 (device_links_lock){+.+.}: >> [ 5.065905] __mutex_lock_common+0x1a0/0x1fe8 >> [ 5.070777] mutex_lock_nested+0x40/0x50 >> [ 5.075215] device_link_remove+0x40/0xe0 >> [ 5.079740] _regulator_put+0x104/0x2d8 >> [ 5.084093] regulator_put+0x30/0x44 >> [ 5.088184] devm_regulator_release+0x38/0x44 >> [ 5.093056] release_nodes+0x604/0x670 >> [ 5.097320] devres_release_all+0x70/0x8c >> [ 5.101846] really_probe+0x270/0x4dc >> [ 5.106024] driver_probe_device+0xcc/0x1d4 >> [ 5.110724] device_driver_attach+0xe4/0x104 >> [ 5.115510] __driver_attach+0x134/0x14c >> [ 5.119949] bus_for_each_dev+0x120/0x180 >> [ 5.124474] driver_attach+0x48/0x54 >> [ 5.128566] bus_add_driver+0x2ac/0x44c >> [ 5.132919] driver_register+0x160/0x288 >> [ 5.137357] __platform_driver_register+0xcc/0xdc >> [ 5.142576] pwm_backlight_driver_init+0x1c/0x24 >> [ 5.147708] do_one_initcall+0x38c/0x994 >> [ 5.152146] do_initcall_level+0x3a4/0x4b8 >> [ 5.156758] do_basic_setup+0x84/0xa0 >> [ 5.160936] kernel_init_freeable+0x23c/0x324 >> [ 5.165807] kernel_init+0x14/0x110 >> [ 5.169813] ret_from_fork+0x10/0x18 >> [ 5.173901] >> [ 5.173901] -> #0 (regulator_list_mutex){+.+.}: >> [ 5.179910] lock_acquire+0x350/0x4d4 >> [ 5.184088] __mutex_lock_common+0x1a0/0x1fe8 >> [ 5.184095] mutex_lock_nested+0x40/0x50 >> [ 5.197475] regulator_lock_dependent+0xdc/0x6c4 >> [ 5.197482] regulator_disable+0xa0/0x138 >> [ 5.197487] scpsys_power_off+0x38c/0x4bc >> [ 5.197495] genpd_power_off+0x3d8/0x6a0 >> [ 5.209399] genpd_power_off+0x530/0x6a0 >> [ 5.209406] genpd_power_off_work_fn+0x74/0xc0 >> [ 5.209411] process_one_work+0x858/0x1208 >> [ 5.209419] worker_thread+0x9ec/0xcb8 >> [ 5.219067] kthread+0x2b8/0x2d0 >> [ 5.219073] ret_from_fork+0x10/0x18 >> [ 5.219077] >> [ 5.219077] other info that might help us debug this: >> [ 5.219077] >> [ 5.219080] Chain exists of: >> [ 5.219080] regulator_list_mutex --> &genpd->mlock --> &genpd->mlock/1 >> [ 5.219080] >> [ 5.228039] Possible unsafe locking scenario: >> [ 5.228039] >> [ 5.228042] CPU0 CPU1 >> [ 5.228046] ---- ---- >> [ 5.228048] lock(&genpd->mlock/1); >> [ 5.228058] lock(&genpd->mlock); >> [ 5.311647] lock(&genpd->mlock/1); >> [ 5.317736] lock(regulator_list_mutex); >> [ 5.321742] >> [ 5.321742] *** DEADLOCK *** >> [ 5.321742] >> [ 5.327655] 4 locks held by kworker/4:1/51: >> [ 5.331831] #0: (____ptrval____) ((wq_completion)pm){+.+.}, >> at:process_one_work+0x57c/0x1208 >> [ 5.340444] #1: (____ptrval____) >> ((work_completion)(&genpd->power_off_work)){+.+.}, >> at:process_one_work+0x5b8/0x1208 >> [ 5.351139] #2: (____ptrval____) (&genpd->mlock){+.+.}, >> at:genpd_lock_mtx+0x20/0x2c >> [ 5.358970] #3: (____ptrval____) (&genpd->mlock/1){+.+.}, >> at:genpd_lock_nested_mtx+0x24/0x30 >> [ 5.367584] >> [ 5.367584] stack backtrace: >> [ 5.371939] CPU: 4 PID: 51 Comm: kworker/4:1 Tainted: G S >> 5.2.0-rc2-next-20190528-44527-g6c94b6475c04 #20 >> [ 5.382809] Workqueue: pm genpd_power_off_work_fn >> [ 5.382816] Call trace: >> [ 5.382822] dump_backtrace+0x0/0x2c0 >> [ 5.382830] show_stack+0x20/0x2c >> [ 5.409174] dump_stack+0x10c/0x17c >> [ 5.412659] print_circular_bug+0x42c/0x4d0 >> [ 5.416838] __lock_acquire+0x4c88/0x5484 >> [ 5.420843] lock_acquire+0x350/0x4d4 >> [ 5.424500] __mutex_lock_common+0x1a0/0x1fe8 >> [ 5.428851] mutex_lock_nested+0x40/0x50 >> [ 5.432770] regulator_lock_dependent+0xdc/0x6c4 >> [ 5.437383] regulator_disable+0xa0/0x138 >> [ 5.441389] scpsys_power_off+0x38c/0x4bc >> [ 5.445393] genpd_power_off+0x3d8/0x6a0 >> [ 5.449310] genpd_power_off+0x530/0x6a0 >> [ 5.453229] genpd_power_off_work_fn+0x74/0xc0 >> [ 5.457667] process_one_work+0x858/0x1208 >> [ 5.461758] worker_thread+0x9ec/0xcb8 >> [ 5.465503] kthread+0x2b8/0x2d0 >> [ 5.468727] ret_from_fork+0x10/0x18 >> >> On Mon, Jun 10, 2019 at 8:21 PM Yong Wu <yong.wu@xxxxxxxxxxxx> wrote: >>> ... >> >> >> On Mon, Jun 10, 2019 at 8:21 PM Yong Wu <yong.wu@xxxxxxxxxxxx> wrote: >> >>> There are 2 mmu cells in a M4U HW. we could adjust some larbs entering >>> mmu0 or mmu1 to balance the bandwidth via the smi-common register >>> SMI_BUS_SEL(0x220)(Each larb occupy 2 bits). >>> >>> In mt8183, For better performance, we switch larb1/2/5/7 to enter >>> mmu1 while the others still keep enter mmu0. >>> >>> In mt8173 and mt2712, we don't get the performance issue, >>> Keep its default value(0x0), that means all the larbs enter mmu0. >>> >>> Note: smi gen1(mt2701/mt7623) don't have this bus_sel. >>> >>> And, the base of smi-common is completely different with smi_ao_base >>> of gen1, thus I add new variable for that. >>> >>> CC: Matthias Brugger <matthias.bgg@xxxxxxxxx> >>> Signed-off-by: Yong Wu <yong.wu@xxxxxxxxxxxx> >>> Reviewed-by: Evan Green <evgreen@xxxxxxxxxxxx> >>> --- >>> drivers/memory/mtk-smi.c | 22 ++++++++++++++++++++-- >>> 1 file changed, 20 insertions(+), 2 deletions(-) >>> >>> diff --git a/drivers/memory/mtk-smi.c b/drivers/memory/mtk-smi.c >>> index 9790801..08cf40d 100644 >>> --- a/drivers/memory/mtk-smi.c >>> +++ b/drivers/memory/mtk-smi.c >>> @@ -49,6 +49,12 @@ >>> #define SMI_LARB_NONSEC_CON(id) (0x380 + ((id) * 4)) >>> #define F_MMU_EN BIT(0) >>> >>> +/* SMI COMMON */ >>> +#define SMI_BUS_SEL 0x220 >>> +#define SMI_BUS_LARB_SHIFT(larbid) ((larbid) << 1) >>> +/* All are MMU0 defaultly. Only specialize mmu1 here. */ >>> +#define F_MMU1_LARB(larbid) (0x1 << SMI_BUS_LARB_SHIFT(larbid)) >>> + >>> enum mtk_smi_gen { >>> MTK_SMI_GEN1, >>> MTK_SMI_GEN2 >>> @@ -57,6 +63,7 @@ enum mtk_smi_gen { >>> struct mtk_smi_common_plat { >>> enum mtk_smi_gen gen; >>> bool has_gals; >>> + u32 bus_sel; /* Balance some larbs to enter mmu0 or >>> mmu1 */ >>> }; >>> >>> struct mtk_smi_larb_gen { >>> @@ -72,8 +79,8 @@ struct mtk_smi { >>> struct clk *clk_apb, *clk_smi; >>> struct clk *clk_gals0, *clk_gals1; >>> struct clk *clk_async; /*only needed by >>> mt2701*/ >>> - void __iomem *smi_ao_base; >>> - >>> + void __iomem *smi_ao_base; /* only for gen1 */ >>> + void __iomem *base; /* only for gen2 */ >>> const struct mtk_smi_common_plat *plat; >>> }; >>> >>> @@ -410,6 +417,8 @@ static int __maybe_unused mtk_smi_larb_suspend(struct >>> device *dev) >>> static const struct mtk_smi_common_plat mtk_smi_common_mt8183 = { >>> .gen = MTK_SMI_GEN2, >>> .has_gals = true, >>> + .bus_sel = F_MMU1_LARB(1) | F_MMU1_LARB(2) | F_MMU1_LARB(5) | >>> + F_MMU1_LARB(7), >>> }; >>> >>> static const struct of_device_id mtk_smi_common_of_ids[] = { >>> @@ -482,6 +491,11 @@ static int mtk_smi_common_probe(struct >>> platform_device *pdev) >>> ret = clk_prepare_enable(common->clk_async); >>> if (ret) >>> return ret; >>> + } else { >>> + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); >>> + common->base = devm_ioremap_resource(dev, res); >>> + if (IS_ERR(common->base)) >>> + return PTR_ERR(common->base); >>> } >>> pm_runtime_enable(dev); >>> platform_set_drvdata(pdev, common); >>> @@ -497,6 +511,7 @@ static int mtk_smi_common_remove(struct >>> platform_device *pdev) >>> static int __maybe_unused mtk_smi_common_resume(struct device *dev) >>> { >>> struct mtk_smi *common = dev_get_drvdata(dev); >>> + u32 bus_sel = common->plat->bus_sel; >>> int ret; >>> >>> ret = mtk_smi_clk_enable(common); >>> @@ -504,6 +519,9 @@ static int __maybe_unused mtk_smi_common_resume(struct >>> device *dev) >>> dev_err(common->dev, "Failed to enable clock(%d).\n", ret); >>> return ret; >>> } >>> + >>> + if (common->plat->gen == MTK_SMI_GEN2 && bus_sel) >>> + writel(bus_sel, common->base + SMI_BUS_SEL); >>> return 0; >>> } >>> >>> -- >>> 1.9.1 >>> >>> >>
