kernel.git/drivers/pinctrl/pinctrl-single.c, branch linux-4.4.y

pinctrl: single: Fix error return code in pcs_parse_bits_in_pinctrl_entry()

2021-09-22T09:41:23Z

[ Upstream commit d789a490d32fdf0465275e3607f8a3bc87d3f3ba ] Fix to return -ENOTSUPP instead of 0 when PCS_HAS_PINCONF is true, which is the same as that returned in pcs_parse_pinconf(). Fixes: 4e7e8017a80e ("pinctrl: pinctrl-single: enhance to configure multiple pins of different modules") Reported-by: Hulk Robot Signed-off-by: Zhen Lei Link: https://lore.kernel.org/r/20210722033930.4034-2-thunder.leizhen@huawei.com Signed-off-by: Linus Walleij Signed-off-by: Sasha Levin

pinctrl-single: fix pcs_parse_pinconf() return value

2020-08-21T08:53:03Z

[ Upstream commit f46fe79ff1b65692a65266a5bec6dbe2bf7fc70f ] This patch causes pcs_parse_pinconf() to return -ENOTSUPP when no pinctrl_map is added. The current behavior is to return 0 when !PCS_HAS_PINCONF or !nconfs. Thus pcs_parse_one_pinctrl_entry() incorrectly assumes that a map was added and sets num_maps = 2. Analysis: ========= The function pcs_parse_one_pinctrl_entry() calls pcs_parse_pinconf() if PCS_HAS_PINCONF is enabled. The function pcs_parse_pinconf() returns 0 to indicate there was no error and num_maps is then set to 2: 980 static int pcs_parse_one_pinctrl_entry(struct pcs_device *pcs, 981 struct device_node *np, 982 struct pinctrl_map **map, 983 unsigned *num_maps, 984 const char **pgnames) 985 { 1053 (*map)->type = PIN_MAP_TYPE_MUX_GROUP; 1054 (*map)->data.mux.group = np->name; 1055 (*map)->data.mux.function = np->name; 1056 1057 if (PCS_HAS_PINCONF && function) { 1058 res = pcs_parse_pinconf(pcs, np, function, map); 1059 if (res) 1060 goto free_pingroups; 1061 *num_maps = 2; 1062 } else { 1063 *num_maps = 1; 1064 } However, pcs_parse_pinconf() will also return 0 if !PCS_HAS_PINCONF or !nconfs. I believe these conditions should indicate that no map was added by returning -ENOTSUPP. Otherwise pcs_parse_one_pinctrl_entry() will set num_maps = 2 even though no maps were successfully added, as it does not reach "m++" on line 940: 895 static int pcs_parse_pinconf(struct pcs_device *pcs, struct device_node *np, 896 struct pcs_function *func, 897 struct pinctrl_map **map) 898 899 { 900 struct pinctrl_map *m = *map; 917 /* If pinconf isn't supported, don't parse properties in below. */ 918 if (!PCS_HAS_PINCONF) 919 return 0; 920 921 /* cacluate how much properties are supported in current node */ 922 for (i = 0; i < ARRAY_SIZE(prop2); i++) { 923 if (of_find_property(np, prop2[i].name, NULL)) 924 nconfs++; 925 } 926 for (i = 0; i < ARRAY_SIZE(prop4); i++) { 927 if (of_find_property(np, prop4[i].name, NULL)) 928 nconfs++; 929 } 930 if (!nconfs) 919 return 0; 932 933 func->conf = devm_kcalloc(pcs->dev, 934 nconfs, sizeof(struct pcs_conf_vals), 935 GFP_KERNEL); 936 if (!func->conf) 937 return -ENOMEM; 938 func->nconfs = nconfs; 939 conf = &(func->conf[0]); 940 m++; This situtation will cause a boot failure [0] on the BeagleBone Black (AM3358) when am33xx_pinmux node in arch/arm/boot/dts/am33xx-l4.dtsi has compatible = "pinconf-single" instead of "pinctrl-single". The patch fixes this issue by returning -ENOSUPP when !PCS_HAS_PINCONF or !nconfs, so that pcs_parse_one_pinctrl_entry() will know that no map was added. Logic is also added to pcs_parse_one_pinctrl_entry() to distinguish between -ENOSUPP and other errors. In the case of -ENOSUPP, num_maps is set to 1 as it is valid for pinconf to be enabled and a given pin group to not any pinconf properties. [0] https://lore.kernel.org/linux-omap/20200529175544.GA3766151@x1/ Fixes: 9dddb4df90d1 ("pinctrl: single: support generic pinconf") Signed-off-by: Drew Fustini Acked-by: Tony Lindgren Link: https://lore.kernel.org/r/20200608125143.GA2789203@x1 Signed-off-by: Linus Walleij Signed-off-by: Sasha Levin

pinctrl: single: Fix missing flush of posted write for a wakeirq

2016-08-10T09:49:27Z

commit 0ac3c0a4025f41748a083bdd4970cb3ede802b15 upstream. With many repeated suspend resume cycles, the pin specific wakeirq may not always work on omaps. This is because the write to enable the pin interrupt may not have reached the device over the interconnect before suspend happens. Let's fix the issue with a flush of posted write with a readback. Reported-by: Nishanth Menon Signed-off-by: Tony Lindgren Signed-off-by: Linus Walleij Signed-off-by: Greg Kroah-Hartman

pinctrl: single: Fix pcs_parse_bits_in_pinctrl_entry to use __ffs than ffs

2016-05-04T21:48:45Z

commit 56b367c0cd67d4c3006738e7dc9dda9273fd2bfe upstream. pcs_parse_bits_in_pinctrl_entry uses ffs which gives bit indices ranging from 1 to MAX. This leads to a corner case where we try to request the pin number = MAX and fails. bit_pos value is being calculted using ffs. pin_num_from_lsb uses bit_pos value. pins array is populated with: pin + pin_num_from_lsb. The above is 1 more than usual bit indices as bit_pos uses ffs to compute first set bit. Hence the last of the pins array is populated with the MAX value and not MAX - 1 which causes error when we call pin_request. mask_pos is rightly calculated as ((pcs->fmask) << (bit_pos - 1)) Consequently val_pos and submask are correct. Hence use __ffs which gives (ffs(x) - 1) as the first bit set. fixes: 4e7e8017a8 ("pinctrl: pinctrl-single: enhance to configure multiple pins of different modules") Signed-off-by: Keerthy Acked-by: Tony Lindgren Signed-off-by: Linus Walleij Signed-off-by: Greg Kroah-Hartman

genirq: Remove irq argument from irq flow handlers

2015-09-16T13:47:51Z

Most interrupt flow handlers do not use the irq argument. Those few which use it can retrieve the irq number from the irq descriptor. Remove the argument. Search and replace was done with coccinelle and some extra helper scripts around it. Thanks to Julia for her help! Signed-off-by: Thomas Gleixner Cc: Julia Lawall Cc: Jiang Liu

Merge tag 'pinctrl-v4.3-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl

2015-09-04T17:22:09Z

Pull pin control updates from Linus Walleij: "This is the bulk of pin control changes for the v4.3 development cycle. Like with GPIO it's a lot of stuff. If my subsystems are any sign of the overall tempo of the kernel v4.3 will be a gigantic diff. [ It looks like 4.3 is calmer than 4.2 in most other subsystems, but we'll see - Linus ] Core changes: - It is possible configure groups in debugfs. - Consolidation of chained IRQ handler install/remove replacing all call sites where irq_set_handler_data() and irq_set_chained_handler() were done in succession with a combined call to irq_set_chained_handler_and_data(). This series was created by Thomas Gleixner after the problem was observed by Russell King. - Tglx also made another series of patches switching __irq_set_handler_locked() for irq_set_handler_locked() which is way cleaner. - Tglx also wrote a good bunch of patches to make use of irq_desc_get_xxx() accessors and avoid looking up irq_descs from IRQ numbers. The goal is to get rid of the irq number from the handlers in the IRQ flow which is nice. Driver feature enhancements: - Power management support for the SiRF SoC Atlas 7. - Power down support for the Qualcomm driver. - Intel Cherryview and Baytrail: switch drivers to use raw spinlocks in IRQ handlers to play nice with the realtime patch set. - Rework and new modes handling for Qualcomm SPMI-MPP. - Pinconf power source config for SH PFC. New drivers and subdrivers: - A new driver for Conexant Digicolor CX92755. - A new driver for UniPhier PH1-LD4, PH1-Pro4, PH1-sLD8, PH1-Pro5, ProXtream2 and PH1-LD6b SoC pin control support. - Reverse-egineered the S/PDIF settings for the Allwinner sun4i driver. - Support for Qualcomm Technologies QDF2xxx ARM64 SoCs - A new Freescale i.mx6ul subdriver. Cleanup: - Remove platform data support in a number of SH PFC subdrivers" * tag 'pinctrl-v4.3-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl: (95 commits) pinctrl: at91: fix null pointer dereference pinctrl: mediatek: Implement wake handler and suspend resume pinctrl: mediatek: Fix multiple registration issue. pinctrl: sh-pfc: r8a7794: add USB pin groups pinctrl: at91: Use generic irq_{request,release}_resources() pinctrl: cherryview: Use raw_spinlock for locking pinctrl: baytrail: Use raw_spinlock for locking pinctrl: imx6ul: Remove .owner field pinctrl: zynq: Fix typos in smc0_nand_grp and smc0_nor_grp pinctrl: sh-pfc: Implement pinconf power-source param for voltage switching clk: rockchip: add pclk_pd_pmu to the list of rk3288 critical clocks pinctrl: sun4i: add spdif to pin description. pinctrl: atlas7: clear ugly branch statements for pull and drivestrength pinctrl: baytrail: Serialize all register access pinctrl: baytrail: Drop FSF mailing address pinctrl: rockchip: only enable gpio clock when it setting pinctrl/mediatek: fix spelling mistake in dev_err error message pinctrl: cherryview: Serialize all register access pinctrl: UniPhier: PH1-Pro5: add I2C ch6 pin-mux setting pinctrl: nomadik: reflect current input value ...

pinctrl: kill off set_irq_flags usage

2015-07-28T10:15:19Z

set_irq_flags is ARM specific with custom flags which have genirq equivalents. Convert drivers to use the genirq interfaces directly, so we can kill off set_irq_flags. The translation of flags is as follows: IRQF_VALID -> !IRQ_NOREQUEST IRQF_PROBE -> !IRQ_NOPROBE IRQF_NOAUTOEN -> IRQ_NOAUTOEN For IRQs managed by an irqdomain, the irqdomain core code handles clearing and setting IRQ_NOREQUEST already, so there is no need to do this in .map() functions and we can simply remove the set_irq_flags calls. Some users also modify IRQ_NOPROBE and this has been maintained although it is not clear that is really needed. There appears to be a great deal of blind copy and paste of this code. Signed-off-by: Rob Herring Cc: Stephen Warren Cc: Lee Jones Cc: Matthias Brugger Cc: Tomasz Figa Cc: Thomas Abraham Cc: Kukjin Kim Cc: Krzysztof Kozlowski Cc: linux-gpio@vger.kernel.org Cc: linux-rpi-kernel@lists.infradead.org Cc: linux-arm-kernel@lists.infradead.org Cc: linux-mediatek@lists.infradead.org Cc: linux-samsung-soc@vger.kernel.org Signed-off-by: Linus Walleij

pinctrl: single: ensure pcs irq will not be forced threaded

2015-07-20T09:01:52Z

The PSC IRQ is requested using request_irq() API and as result it can be forced to be threaded IRQ in RT-Kernel if PCS_QUIRK_HAS_SHARED_IRQ is enabled for pinctrl domain. As result, following 'possible irq lock inversion dependency' report can be seen: ========================================================= [ INFO: possible irq lock inversion dependency detected ] 3.14.43-rt42-00360-g96ff499-dirty #24 Not tainted --------------------------------------------------------- irq/369-pinctrl/927 just changed the state of lock: (&pcs->lock){+.....}, at: [] pcs_irq_handle+0x48/0x9c but this lock was taken by another, HARDIRQ-safe lock in the past: (&irq_desc_lock_class){-.....} and interrupts could create inverse lock ordering between them. other info that might help us debug this: Possible interrupt unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&pcs->lock); local_irq_disable(); lock(&irq_desc_lock_class); lock(&pcs->lock); lock(&irq_desc_lock_class); *** DEADLOCK *** no locks held by irq/369-pinctrl/927. the shortest dependencies between 2nd lock and 1st lock: -> (&irq_desc_lock_class){-.....} ops: 58724 { IN-HARDIRQ-W at: [] lock_acquire+0x9c/0x158 [] _raw_spin_lock+0x48/0x58 [] handle_fasteoi_irq+0x24/0x15c [] generic_handle_irq+0x3c/0x4c [] handle_IRQ+0x50/0xa0 [] gic_handle_irq+0x3c/0x6c [] __irq_svc+0x44/0x8c [] arch_cpu_idle+0x40/0x4c [] cpu_startup_entry+0x270/0x2e0 [] rest_init+0xd4/0xe4 [] start_kernel+0x3d0/0x3dc [<80008084>] 0x80008084 INITIAL USE at: [] lock_acquire+0x9c/0x158 [] _raw_spin_lock_irqsave+0x54/0x68 [] __irq_get_desc_lock+0x64/0xa4 [] irq_set_chip+0x30/0x78 [] irq_set_chip_and_handler_name+0x24/0x3c [] gic_irq_domain_map+0x48/0xb4 [] irq_domain_associate+0x84/0x1d4 [] irq_create_mapping+0x80/0x11c [] irq_create_of_mapping+0x80/0x120 [] irq_of_parse_and_map+0x34/0x3c [] omap_dm_timer_init_one+0x90/0x30c [] omap5_realtime_timer_init+0x8c/0x48c [] time_init+0x28/0x38 [] start_kernel+0x240/0x3dc [<80008084>] 0x80008084 } ... key at: [] irq_desc_lock_class+0x0/0x8 ... acquired at: [] _raw_spin_lock+0x48/0x58 [] pcs_irq_unmask+0x58/0xa0 [] irq_enable+0x38/0x48 [] irq_startup+0x78/0x7c [] __setup_irq+0x4a8/0x4f4 [] request_threaded_irq+0xb8/0x138 [] omap_8250_startup+0x4c/0x148 [] serial8250_startup+0x24/0x30 [] uart_startup.part.9+0x5c/0x1b4 [] uart_open+0xf4/0x16c [] tty_open+0x170/0x61c [] chrdev_open+0xbc/0x1b4 [] do_dentry_open+0x1e8/0x2bc [] finish_open+0x44/0x5c [] do_last.isra.47+0x710/0xca0 [] path_openat+0xc4/0x640 [] do_filp_open+0x3c/0x98 [] do_sys_open+0x114/0x1d8 [] SyS_open+0x28/0x2c [] kernel_init_freeable+0x168/0x1e4 [] kernel_init+0x1c/0xf8 [] ret_from_fork+0x14/0x20 -> (&pcs->lock){+.....} ops: 65 { HARDIRQ-ON-W at: [] lock_acquire+0x9c/0x158 [] _raw_spin_lock+0x48/0x58 [] pcs_irq_handle+0x48/0x9c [] pcs_irq_handler+0x1c/0x28 [] irq_forced_thread_fn+0x30/0x74 [] irq_thread+0x158/0x1c4 [] kthread+0xd4/0xe8 [] ret_from_fork+0x14/0x20 INITIAL USE at: [] lock_acquire+0x9c/0x158 [] _raw_spin_lock_irqsave+0x54/0x68 [] pcs_enable+0x7c/0xe8 [] pinmux_enable_setting+0x178/0x220 [] pinctrl_select_state+0x110/0x194 [] pinctrl_bind_pins+0x7c/0x108 [] driver_probe_device+0x70/0x254 [] __driver_attach+0x9c/0xa0 [] bus_for_each_dev+0x78/0xac [] driver_attach+0x2c/0x30 [] bus_add_driver+0x15c/0x204 [] driver_register+0x88/0x108 [] __platform_driver_register+0x64/0x6c [] omap_hsmmc_driver_init+0x1c/0x20 [] do_one_initcall+0x110/0x170 [] kernel_init_freeable+0x140/0x1e4 [] kernel_init+0x1c/0xf8 [] ret_from_fork+0x14/0x20 } ... key at: [] __key.18572+0x0/0x8 ... acquired at: [] mark_lock+0x388/0x76c [] __lock_acquire+0x6d0/0x1f98 [] lock_acquire+0x9c/0x158 [] _raw_spin_lock+0x48/0x58 [] pcs_irq_handle+0x48/0x9c [] pcs_irq_handler+0x1c/0x28 [] irq_forced_thread_fn+0x30/0x74 [] irq_thread+0x158/0x1c4 [] kthread+0xd4/0xe8 [] ret_from_fork+0x14/0x20 stack backtrace: CPU: 1 PID: 927 Comm: irq/369-pinctrl Not tainted 3.14.43-rt42-00360-g96ff499-dirty #24 [] (unwind_backtrace) from [] (show_stack+0x20/0x24) [] (show_stack) from [] (dump_stack+0x84/0xd0) [] (dump_stack) from [] (print_irq_inversion_bug+0x1d0/0x21c) [] (print_irq_inversion_bug) from [] (check_usage_backwards+0xb4/0x11c) [] (check_usage_backwards) from [] (mark_lock+0x388/0x76c) [] (mark_lock) from [] (__lock_acquire+0x6d0/0x1f98) [] (__lock_acquire) from [] (lock_acquire+0x9c/0x158) [] (lock_acquire) from [] (_raw_spin_lock+0x48/0x58) [] (_raw_spin_lock) from [] (pcs_irq_handle+0x48/0x9c) [] (pcs_irq_handle) from [] (pcs_irq_handler+0x1c/0x28) [] (pcs_irq_handler) from [] (irq_forced_thread_fn+0x30/0x74) [] (irq_forced_thread_fn) from [] (irq_thread+0x158/0x1c4) [] (irq_thread) from [] (kthread+0xd4/0xe8) [] (kthread) from [] (ret_from_fork+0x14/0x20) To fix it use IRQF_NO_THREAD to ensure that pcs irq will not be forced threaded. Cc: Tony Lindgren Cc: Sebastian Andrzej Siewior Signed-off-by: Grygorii Strashko Acked-by: Tony Lindgren Signed-off-by: Linus Walleij

pinctrl: Use irq_desc_get_xxx() to avoid redundant lookup of irq_desc

2015-07-17T19:56:20Z

Use irq_desc_get_xxx() to avoid redundant lookup of irq_desc while we already have a pointer to corresponding irq_desc. Signed-off-by: Jiang Liu Cc: Linus Walleij Cc: linux-gpio@vger.kernel.org Signed-off-by: Thomas Gleixner

pinctrl: Consolidate chained IRQ handler install/remove

2015-07-17T19:56:15Z

Chained irq handlers usually set up handler data as well. We now have a function to set both under irq_desc->lock. Replace the two calls with one. Search and conversion was done with coccinelle. Reported-by: Russell King Signed-off-by: Thomas Gleixner Cc: Julia Lawall Cc: Linus Walleij Cc: linux-gpio@vger.kernel.org