Re: [PATCH V7] i2c: tegra: remove BUG, BUG_ON

[Bitan Biswas]

On 6/17/19 12:28 PM, Dmitry Osipenko wrote:
> 17.06.2019 21:41, Bitan Biswas ÐÐÑÐÑ:
> > On 6/17/19 5:13 AM, Dmitry Osipenko wrote:
> > > 17.06.2019 8:09, Bitan Biswas ÐÐÑÐÑ:
> > > > Remove BUG, BUG_ON as it makes system usable:
> > > >  Â - Remove redundant BUG_ON calls or replace with WARN_ON_ONCE
> > > >  ÂÂÂ as needed.
> > > >  Â - Remove BUG() and mask Rx interrupt similar as Tx
> > > >  ÂÂÂ for message fully sent case.
> > > >  Â - Add caller error handling and WARN_ON_ONCE check for non-zero
> > > >  ÂÂÂ rx_fifo_avail in tegra_i2c_empty_rx_fifo() after all processing.
> > >
> > > The commit message should describe motivation of the change and not the change itself,
> > > unless it's some additional information which is required for better understanding of
> > > the code.
> > >
> > > In yours case it could be something like that:
> > >
> > >  ÂÂÂÂ The usage of BUG() macro is generally discouraged in kernel, unless
> > >  ÂÂÂÂ it's a problem that results in a physical damage or loss of data.
> > >  ÂÂÂÂ This patch removes unnecessary BUG() macros and replaces the rest
> > >  ÂÂÂÂ with a warnings.
> > I shall update as per above comments.
> >
> > > > Signed-off-by: Bitan Biswas <bbiswas@xxxxxxxxxx>
> > > > ---
> > > >  Â drivers/i2c/busses/i2c-tegra.c | 45 ++++++++++++++++++++++++++++++++++--------
> > > >  Â 1 file changed, 37 insertions(+), 8 deletions(-)
> > > >
> > > > diff --git a/drivers/i2c/busses/i2c-tegra.c b/drivers/i2c/busses/i2c-tegra.c
> > > > index 4dfb4c1..b155b61 100644
> > > > --- a/drivers/i2c/busses/i2c-tegra.c
> > > > +++ b/drivers/i2c/busses/i2c-tegra.c
> > > > @@ -73,6 +73,7 @@
> > > >  Â #define I2C_ERR_NO_ACKÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ BIT(0)
> > > >  Â #define I2C_ERR_ARBITRATION_LOSTÂÂÂÂÂÂÂ BIT(1)
> > > >  Â #define I2C_ERR_UNKNOWN_INTERRUPTÂÂÂÂÂÂÂ BIT(2)
> > > > +#define I2C_ERR_RX_BUFFER_OVERFLOWÂÂÂÂÂÂÂ BIT(3)
> > > >  Â Â #define PACKET_HEADER0_HEADER_SIZE_SHIFTÂÂÂ 28
> > > >  Â #define PACKET_HEADER0_PACKET_ID_SHIFTÂÂÂÂÂÂÂ 16
> > > > @@ -515,7 +516,11 @@ static int tegra_i2c_empty_rx_fifo(struct tegra_i2c_dev *i2c_dev)
> > > >  ÂÂÂÂÂÂ * prevent overwriting past the end of buf
> > > >  ÂÂÂÂÂÂ */
> > > >  ÂÂÂÂÂ if (rx_fifo_avail > 0 && buf_remaining > 0) {
> > > > -ÂÂÂÂÂÂÂ BUG_ON(buf_remaining > 3);
> > > > +ÂÂÂÂÂÂÂ /*
> > > > +ÂÂÂÂÂÂÂÂ * buf_remaining > 3 check not needed as rx_fifo_avail == 0
> > > > +ÂÂÂÂÂÂÂÂ * when (words_to_transfer was > rx_fifo_avail) earlier
> > > > +ÂÂÂÂÂÂÂÂ * in this function.
> > > > +ÂÂÂÂÂÂÂÂ */
> > > >  ÂÂÂÂÂÂÂÂÂ val = i2c_readl(i2c_dev, I2C_RX_FIFO);
> > > >  ÂÂÂÂÂÂÂÂÂ val = cpu_to_le32(val);
> > > >  ÂÂÂÂÂÂÂÂÂ memcpy(buf, &val, buf_remaining);
> > > > @@ -523,7 +528,15 @@ static int tegra_i2c_empty_rx_fifo(struct tegra_i2c_dev *i2c_dev)
> > > >  ÂÂÂÂÂÂÂÂÂ rx_fifo_avail--;
> > > >  ÂÂÂÂÂ }
> > > >  Â -ÂÂÂ BUG_ON(rx_fifo_avail > 0 && buf_remaining > 0);
> > > > +ÂÂÂ if ((!(i2c_dev->msg_buf_remaining)) &&
> > >
> > > The RX FIFO shall be drained completely no matter what.
> > >
> > > Hence why the "i2c_dev->msg_buf_remaining" checking is needed here?
> > I moved the part of below condition in Patch V6 to function tegra_i2c_empty_rx_fifo:
> >
> > > > +ÂÂÂÂÂÂÂÂÂÂÂ err_val = tegra_i2c_empty_rx_fifo(i2c_dev);
> > > > +ÂÂÂÂÂÂÂÂÂÂÂ if ((!(i2c_dev->msg_buf_remaining)) &&
> >
> > > Let's move this check into tegra_i2c_empty_rx_fifo() and return -EINVAL for that case.
> > > This will make code to look cleaner.
> >
> > Is above condition not needed?
>
> Let's put it at the very beginning. This may give a bit more information about the
> problem by knowing if the offending overflow happens after or during of the buffer's
> fill up.
>
> static int tegra_i2c_empty_rx_fifo(struct tegra_i2c_dev *i2c_dev)
> {
> 	u32 val;
> 	int rx_fifo_avail;
> 	u8 *buf = i2c_dev->msg_buf;
> 	size_t buf_remaining = i2c_dev->msg_buf_remaining;
> 	int words_to_transfer;
>
> 	if (WARN_ON(!i2c_dev->msg_buf_remaining))
> 		return -EINVAL;
> ...
OK

> In general, the original logic should be preserved during of refactoring. In this case
> we are keeping the original check and then also making it a bit more informative.
I feel the msg_buf_remaining check was not there in original code. The 
corresponding line was probably the following checking for error when 
(buf_remaining > 0) after all work in the function tegra_i2c_empty_rx_fifo()

https://elixir.bootlin.com/linux/v5.2-rc5/source/drivers/i2c/busses/i2c-tegra.c#L536



> > > Secondly, in the future please don't add parens where they are not needed. In this
> > > case parens around !i2c_dev->msg_buf_remaining are not needed at all.
> > I shall look out for similar unnecessary parentheses and update the patch.
>
> Yes, please clean up all the occurrences in the code if there are any. And please do
> it in a separate patch.
I reviewed the source for unnecessary parentheses and do not find any. 
Hence not planning to push any patch. Let me know if I missed a case.

-Thanks,
 Bitan