[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAMRc=MdRJ_mNM-NteVnBCB=xMwrkSQb+b7SZOWVx0yp7dMqqiw@mail.gmail.com>
Date: Wed, 31 Jan 2024 13:45:04 +0100
From: Bartosz Golaszewski <brgl@...ev.pl>
To: kernel test robot <lkp@...el.com>
Cc: Linus Walleij <linus.walleij@...aro.org>, Kent Gibson <warthog618@...il.com>,
Alex Elder <elder@...aro.org>, Geert Uytterhoeven <geert+renesas@...der.be>,
"Paul E . McKenney" <paulmck@...nel.org>, Andy Shevchenko <andriy.shevchenko@...ux.intel.com>,
Wolfram Sang <wsa-dev@...g-engineering.com>, oe-kbuild-all@...ts.linux.dev,
linux-gpio@...r.kernel.org, linux-kernel@...r.kernel.org,
Bartosz Golaszewski <bartosz.golaszewski@...aro.org>
Subject: Re: [PATCH 01/22] gpio: protect the list of GPIO devices with SRCU
On Wed, Jan 31, 2024 at 10:37 AM kernel test robot <lkp@...el.com> wrote:
>
> Hi Bartosz,
>
> kernel test robot noticed the following build warnings:
>
> [auto build test WARNING on brgl/gpio/for-next]
> [also build test WARNING on linus/master v6.8-rc2 next-20240131]
> [If your patch is applied to the wrong git tree, kindly drop us a note.
> And when submitting patch, we suggest to use '--base' as documented in
> https://git-scm.com/docs/git-format-patch#_base_tree_information]
>
> url: https://github.com/intel-lab-lkp/linux/commits/Bartosz-Golaszewski/gpio-protect-the-list-of-GPIO-devices-with-SRCU/20240130-205537
> base: https://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux.git gpio/for-next
> patch link: https://lore.kernel.org/r/20240130124828.14678-2-brgl%40bgdev.pl
> patch subject: [PATCH 01/22] gpio: protect the list of GPIO devices with SRCU
> config: i386-randconfig-141-20240131 (https://download.01.org/0day-ci/archive/20240131/202401311746.be3dlVTg-lkp@intel.com/config)
> compiler: clang version 17.0.6 (https://github.com/llvm/llvm-project 6009708b4367171ccdbf4b5905cb6a803753fe18)
>
> If you fix the issue in a separate patch/commit (i.e. not just a new version of
> the same patch/commit), kindly add following tags
> | Reported-by: kernel test robot <lkp@...el.com>
> | Closes: https://lore.kernel.org/oe-kbuild-all/202401311746.be3dlVTg-lkp@intel.com/
>
> New smatch warnings:
> drivers/gpio/gpiolib.c:4167 gpiod_find_and_request() error: uninitialized symbol 'ret'.
> drivers/gpio/gpiolib.c:4181 gpiod_find_and_request() error: uninitialized symbol 'desc'.
>
> Old smatch warnings:
> drivers/gpio/gpiolib.c:4184 gpiod_find_and_request() error: uninitialized symbol 'desc'.
>
> vim +/ret +4167 drivers/gpio/gpiolib.c
>
> 0eadd36d9123745 Dmitry Torokhov 2022-09-03 4128
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4129 static struct gpio_desc *gpiod_find_and_request(struct device *consumer,
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4130 struct fwnode_handle *fwnode,
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4131 const char *con_id,
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4132 unsigned int idx,
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4133 enum gpiod_flags flags,
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4134 const char *label,
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4135 bool platform_lookup_allowed)
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4136 {
> ba2dc1cb5491712 Hans de Goede 2022-12-29 4137 unsigned long lookupflags = GPIO_LOOKUP_FLAGS_DEFAULT;
> c122f461ccac0e7 Andy Shevchenko 2023-03-09 4138 struct gpio_desc *desc;
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4139 int ret;
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4140
> 1fe5210a1bfba00 Bartosz Golaszewski 2024-01-30 4141 scoped_guard(srcu, &gpio_devices_srcu) {
> 1fe5210a1bfba00 Bartosz Golaszewski 2024-01-30 4142 desc = gpiod_find_by_fwnode(fwnode, consumer, con_id, idx,
> 1fe5210a1bfba00 Bartosz Golaszewski 2024-01-30 4143 &flags, &lookupflags);
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4144 if (gpiod_not_found(desc) && platform_lookup_allowed) {
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4145 /*
> 1fe5210a1bfba00 Bartosz Golaszewski 2024-01-30 4146 * Either we are not using DT or ACPI, or their lookup
> 1fe5210a1bfba00 Bartosz Golaszewski 2024-01-30 4147 * did not return a result. In that case, use platform
> 1fe5210a1bfba00 Bartosz Golaszewski 2024-01-30 4148 * lookup as a fallback.
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4149 */
> 1fe5210a1bfba00 Bartosz Golaszewski 2024-01-30 4150 dev_dbg(consumer,
> 1fe5210a1bfba00 Bartosz Golaszewski 2024-01-30 4151 "using lookup tables for GPIO lookup\n");
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4152 desc = gpiod_find(consumer, con_id, idx, &lookupflags);
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4153 }
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4154
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4155 if (IS_ERR(desc)) {
> 1fe5210a1bfba00 Bartosz Golaszewski 2024-01-30 4156 dev_dbg(consumer, "No GPIO consumer %s found\n",
> 1fe5210a1bfba00 Bartosz Golaszewski 2024-01-30 4157 con_id);
> 0eadd36d9123745 Dmitry Torokhov 2022-09-03 4158 return desc;
> 0eadd36d9123745 Dmitry Torokhov 2022-09-03 4159 }
> 0eadd36d9123745 Dmitry Torokhov 2022-09-03 4160
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4161 /*
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4162 * If a connection label was passed use that, else attempt to use
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4163 * the device name as label
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4164 */
> 0eadd36d9123745 Dmitry Torokhov 2022-09-03 4165 ret = gpiod_request(desc, label);
> 1fe5210a1bfba00 Bartosz Golaszewski 2024-01-30 4166 }
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 @4167 if (ret) {
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4168 if (!(ret == -EBUSY && flags & GPIOD_FLAGS_BIT_NONEXCLUSIVE))
> 0eadd36d9123745 Dmitry Torokhov 2022-09-03 4169 return ERR_PTR(ret);
> 0eadd36d9123745 Dmitry Torokhov 2022-09-03 4170
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4171 /*
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4172 * This happens when there are several consumers for
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4173 * the same GPIO line: we just return here without
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4174 * further initialization. It is a bit of a hack.
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4175 * This is necessary to support fixed regulators.
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4176 *
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4177 * FIXME: Make this more sane and safe.
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4178 */
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4179 dev_info(consumer,
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4180 "nonexclusive access to GPIO for %s\n", con_id);
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 @4181 return desc;
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4182 }
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4183
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4184 ret = gpiod_configure_flags(desc, con_id, lookupflags, flags);
> 0eadd36d9123745 Dmitry Torokhov 2022-09-03 4185 if (ret < 0) {
> 8eb1f71e7acca4f Dmitry Torokhov 2022-11-11 4186 dev_dbg(consumer, "setup of GPIO %s failed\n", con_id);
> 0eadd36d9123745 Dmitry Torokhov 2022-09-03 4187 gpiod_put(desc);
> 0eadd36d9123745 Dmitry Torokhov 2022-09-03 4188 return ERR_PTR(ret);
> 0eadd36d9123745 Dmitry Torokhov 2022-09-03 4189 }
> 0eadd36d9123745 Dmitry Torokhov 2022-09-03 4190
> 9ce4ed5b4db1363 Bartosz Golaszewski 2023-08-21 4191 gpiod_line_state_notify(desc, GPIOLINE_CHANGED_REQUESTED);
> 0eadd36d9123745 Dmitry Torokhov 2022-09-03 4192
> 0eadd36d9123745 Dmitry Torokhov 2022-09-03 4193 return desc;
> 0eadd36d9123745 Dmitry Torokhov 2022-09-03 4194 }
> 0eadd36d9123745 Dmitry Torokhov 2022-09-03 4195
>
> --
> 0-DAY CI Kernel Test Service
> https://github.com/intel/lkp-tests/wiki
This is a false-positive coming from the fact the scoped_guard() is
implemented as a for loop. I will initialize the variables anyway to
make smatch happy.
Bart
Powered by blists - more mailing lists