mirror of
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
synced 2026-05-03 13:32:07 -04:00
dcf1bf648f
Writes from I2C bus often fail when testing the i2c-designware-slave.c with the slave-eeprom backend. The same writes work correctly when testing with a real 24c02 EEPROM chip. In the tests below an i2c-designware-slave.c instance with the slave-eeprom backend is configured to act as a simulated 24c02 at address 0x65 on an I2C host bus 6. 1. i2cset -y 6 0x65 0x00 0x55 Single byte 0x55 write into address 0x00. No data goes into simulated EEPROM. Debug prints from the i2c_dw_irq_handler_slave(): 0x1 STATUS SLAVE_ACTIVITY=0x0 : RAW_INTR_STAT=0x714 : INTR_STAT=0x204 0x1 STATUS SLAVE_ACTIVITY=0x0 : RAW_INTR_STAT=0x514 : INTR_STAT=0x4 2. i2ctransfer -y 6 w9@0x65 0x00 0xff- Write 8 bytes with decrementing value starting from 0xff at address 0x00 and forward. Only some of the data goes into arbitrary addresses. Content is something like below but varies: 00000000 f9 f8 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| 00000050 00 00 00 00 00 00 ff fe 00 00 00 00 00 00 00 00 |................| 000000f0 00 00 00 00 00 00 00 00 00 00 00 00 00 fc fb fa |................| In this case debug prints were: 0x1 STATUS SLAVE_ACTIVITY=0x1 : RAW_INTR_STAT=0x514 : INTR_STAT=0x4 0x1 STATUS SLAVE_ACTIVITY=0x1 : RAW_INTR_STAT=0x514 : INTR_STAT=0x4 0x1 STATUS SLAVE_ACTIVITY=0x0 : RAW_INTR_STAT=0x714 : INTR_STAT=0x204 0x1 STATUS SLAVE_ACTIVITY=0x0 : RAW_INTR_STAT=0x514 : INTR_STAT=0x4 0x1 STATUS SLAVE_ACTIVITY=0x0 : RAW_INTR_STAT=0x514 : INTR_STAT=0x4 0x1 STATUS SLAVE_ACTIVITY=0x0 : RAW_INTR_STAT=0x514 : INTR_STAT=0x4 0x1 STATUS SLAVE_ACTIVITY=0x0 : RAW_INTR_STAT=0x514 : INTR_STAT=0x4 0x1 STATUS SLAVE_ACTIVITY=0x0 : RAW_INTR_STAT=0x514 : INTR_STAT=0x4 0x1 STATUS SLAVE_ACTIVITY=0x0 : RAW_INTR_STAT=0x514 : INTR_STAT=0x4 0x1 STATUS SLAVE_ACTIVITY=0x0 : RAW_INTR_STAT=0x510 : INTR_STAT=0x0 Both cases show there is more data coming from the receive FIFO still after detecting the STOP condition. This can be seen from interrupt status bits DW_IC_INTR_STOP_DET (0x200) and DW_IC_INTR_RX_FULL (0x4). Perhaps due interrupt latencies the receive FIFO is not read fast enough, STOP detection happens synchronously when it occurs on the I2C bus and the DW_IC_INTR_RX_FULL keeps coming as long as there are more bytes in the receive FIFO. Fix this by reading the receive FIFO completely empty whenever DW_IC_INTR_RX_FULL occurs. Use RFNE, Receive FIFO Not Empty bit in the DW_IC_STATUS register to loop through bytes in the FIFO. While at it do not test the return code from i2c_slave_event() for the I2C_SLAVE_WRITE_RECEIVED since to my understanding this hardware cannot generate NACK to incoming bytes and debug print itself does not have much value. Reported-by: Tian Ye <tianye@sugon.com> Signed-off-by: Jarkko Nikula <jarkko.nikula@linux.intel.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Wolfram Sang <wsa@kernel.org>
Linux kernel
============
There are several guides for kernel developers and users. These guides can
be rendered in a number of formats, like HTML and PDF. Please read
Documentation/admin-guide/README.rst first.
In order to build the documentation, use ``make htmldocs`` or
``make pdfdocs``. The formatted documentation can also be read online at:
https://www.kernel.org/doc/html/latest/
There are various text files in the Documentation/ subdirectory,
several of them using the Restructured Text markup notation.
Please read the Documentation/process/changes.rst file, as it contains the
requirements for building and running the kernel, and information about
the problems which may result by upgrading your kernel.