Simon Glass | fdd72be | 2019-09-25 08:55:48 -0600 | [diff] [blame] | 1 | .. SPDX-License-Identifier: GPL-2.0+ |
| 2 | .. sectionauthor:: Simon Glass <sjg@chromium.org> |
| 3 | |
| 4 | Debugging driver model |
| 5 | ====================== |
| 6 | |
| 7 | This document aims to provide help when you cannot work out why driver model is |
| 8 | not doing what you expect. |
| 9 | |
| 10 | |
| 11 | Useful techniques in general |
| 12 | ---------------------------- |
| 13 | |
| 14 | Here are some useful debugging features generally. |
| 15 | |
| 16 | - If you are writing a new feature, consider doing it in sandbox instead of |
| 17 | on your board. Sandbox has no limits, allows easy debugging (e.g. gdb) and |
| 18 | you can write emulators for most common devices. |
| 19 | - Put '#define DEBUG' at the top of a file, to activate all the debug() and |
| 20 | log_debug() statements in that file. |
| 21 | - Where logging is used, change the logging level, e.g. in SPL with |
| 22 | CONFIG_SPL_LOG_MAX_LEVEL=7 (which is LOGL_DEBUG) and |
| 23 | CONFIG_LOG_DEFAULT_LEVEL=7 |
| 24 | - Where logging of return values is implemented with log_msg_ret(), set |
| 25 | CONFIG_LOG_ERROR_RETURN=y to see exactly where the error is happening |
| 26 | - Make sure you have a debug UART enabled - see CONFIG_DEBUG_UART. With this |
| 27 | you can get serial output (printf(), etc.) before the serial driver is |
| 28 | running. |
| 29 | - Use a JTAG emulator to set breakpoints and single-step through code |
| 30 | |
| 31 | Not that most of these increase code/data size somewhat when enabled. |
| 32 | |
| 33 | |
| 34 | Failure to locate a device |
| 35 | -------------------------- |
| 36 | |
| 37 | Let's say you have uclass_first_device_err() and it is not finding anything. |
| 38 | |
| 39 | If it is returning an error, then that gives you a clue. Look up linux/errno.h |
| 40 | to see errors. Common ones are: |
| 41 | |
| 42 | - -ENOMEM which indicates that memory is short. If it happens in SPL or |
| 43 | before relocation in U-Boot, check CONFIG_SPL_SYS_MALLOC_F_LEN and |
| 44 | CONFIG_SYS_MALLOC_F_LEN as they may need to be larger. Add '#define DEBUG' |
| 45 | at the very top of malloc_simple.c to get an idea of where your memory is |
| 46 | going. |
| 47 | - -EINVAL which typically indicates that something was missing or wrong in |
| 48 | the device tree node. Check that everything is correct and look at the |
Simon Glass | d1998a9 | 2020-12-03 16:55:21 -0700 | [diff] [blame] | 49 | of_to_plat() method in the driver. |
Simon Glass | fdd72be | 2019-09-25 08:55:48 -0600 | [diff] [blame] | 50 | |
| 51 | If there is no error, you should check if the device is actually bound. Call |
Dylan Corrales | 63507f9 | 2023-10-19 10:02:30 -0400 | [diff] [blame] | 52 | dm_dump_tree() just before you locate the device to make sure it exists. |
Simon Glass | fdd72be | 2019-09-25 08:55:48 -0600 | [diff] [blame] | 53 | |
| 54 | If it does not exist, check your device tree compatible strings match up with |
| 55 | what the driver expects (in the struct udevice_id array). |
| 56 | |
| 57 | If you are using of-platdata (e.g. CONFIG_SPL_OF_PLATDATA), check that the |
| 58 | driver name is the same as the first compatible string in the device tree (with |
| 59 | invalid-variable characters converted to underscore). |
| 60 | |
Fabio Estevam | 1d6a0c9 | 2020-01-29 14:45:33 -0300 | [diff] [blame] | 61 | If you are really stuck, putting '#define LOG_DEBUG' at the top of |
| 62 | drivers/core/lists.c should show you what is going on. |