| Message ID | 20190822094516.55130-1-hsinyi@chromium.org (mailing list archive) |
|---|---|
| State | Mainlined |
| Commit | abf4923e97c3abbbd1e59f0e13c7c214c93c6aaa |
| Headers | show |
| Series | [v2] i2c: mediatek: disable zero-length transfers for mt8183 | expand |
On Thu, 2019-08-22 at 17:45 +0800, Hsin-Yi Wang wrote: > When doing i2cdetect quick write mode, we would get transfer > error ENOMEM, and i2cdetect shows there's no device at the address. > Quoting from mt8183 datasheet, the number of transfers to be > transferred in one transaction should be set to bigger than 1, > so we should forbid zero-length transfer and update functionality. > > Incorrect return: > localhost ~ # i2cdetect -q -y 0 > 0 1 2 3 4 5 6 7 8 9 a b c d e f > 00: -- -- -- -- -- -- -- -- -- -- -- -- -- > 10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > 20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > 30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > 40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > 50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > 60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > 70: -- -- -- -- -- -- -- -- > > After this patch: > localhost ~ # i2cdetect -q -y 0 > Error: Can't use SMBus Quick Write command on this bus > > localhost ~ # i2cdetect -y 0 > Warning: Can't use SMBus Quick Write command, will skip some addresses > 0 1 2 3 4 5 6 7 8 9 a b c d e f > 00: > 10: > 20: > 30: -- -- -- -- -- -- -- -- > 40: > 50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > 60: > 70: > > Reported-by: Alexandru M Stan <amstan@chromium.org> > Signed-off-by: Hsin-Yi Wang <hsinyi@chromium.org> > --- > Change from v1: > * restore the order of algo and quirks > --- > drivers/i2c/busses/i2c-mt65xx.c | 11 ++++++++++- > 1 file changed, 10 insertions(+), 1 deletion(-) > > diff --git a/drivers/i2c/busses/i2c-mt65xx.c b/drivers/i2c/busses/i2c-mt65xx.c > index 252edb433fdf..29eae1bf4f86 100644 > --- a/drivers/i2c/busses/i2c-mt65xx.c > +++ b/drivers/i2c/busses/i2c-mt65xx.c > @@ -234,6 +234,10 @@ static const struct i2c_adapter_quirks mt7622_i2c_quirks = { > .max_num_msgs = 255, > }; > > +static const struct i2c_adapter_quirks mt8183_i2c_quirks = { > + .flags = I2C_AQ_NO_ZERO_LEN, > +}; > + > static const struct mtk_i2c_compatible mt2712_compat = { > .regs = mt_i2c_regs_v1, > .pmic_i2c = 0, > @@ -298,6 +302,7 @@ static const struct mtk_i2c_compatible mt8173_compat = { > }; > > static const struct mtk_i2c_compatible mt8183_compat = { > + .quirks = &mt8183_i2c_quirks, > .regs = mt_i2c_regs_v2, > .pmic_i2c = 0, > .dcm = 0, > @@ -870,7 +875,11 @@ static irqreturn_t mtk_i2c_irq(int irqno, void *dev_id) > > static u32 mtk_i2c_functionality(struct i2c_adapter *adap) > { > - return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL; > + if (adap->quirks->flags & I2C_AQ_NO_ZERO_LEN) > + return I2C_FUNC_I2C | > + (I2C_FUNC_SMBUS_EMUL & ~I2C_FUNC_SMBUS_QUICK); > + else > + return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL; It can be removed? > } > > static const struct i2c_algorithm mtk_i2c_algorithm = {
On Fri, Aug 23, 2019 at 4:09 PM Qii Wang <qii.wang@mediatek.com> wrote: > > > > static u32 mtk_i2c_functionality(struct i2c_adapter *adap) > > { > > - return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL; > > + if (adap->quirks->flags & I2C_AQ_NO_ZERO_LEN) > > + return I2C_FUNC_I2C | > > + (I2C_FUNC_SMBUS_EMUL & ~I2C_FUNC_SMBUS_QUICK); > > + else > > + return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL; > > It can be removed? See previous discussion: https://patchwork.kernel.org/patch/10814391/#22484435 but not all SoC's quirks has I2C_AQ_NO_ZERO_LEN.
On Fri, 2019-08-23 at 16:13 +0800, Hsin-Yi Wang wrote: > On Fri, Aug 23, 2019 at 4:09 PM Qii Wang <qii.wang@mediatek.com> wrote: > > > > > > > static u32 mtk_i2c_functionality(struct i2c_adapter *adap) > > > { > > > - return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL; > > > + if (adap->quirks->flags & I2C_AQ_NO_ZERO_LEN) > > > + return I2C_FUNC_I2C | > > > + (I2C_FUNC_SMBUS_EMUL & ~I2C_FUNC_SMBUS_QUICK); > > > + else > > > + return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL; > > > > It can be removed? > See previous discussion: https://patchwork.kernel.org/patch/10814391/#22484435 > but not all SoC's quirks has I2C_AQ_NO_ZERO_LEN. ok, it looks good for me, thanks. Reviewed-by: Qii Wang <qii.wang@mediatek.com>
On Thu, Aug 22, 2019 at 05:45:17PM +0800, Hsin-Yi Wang wrote: > When doing i2cdetect quick write mode, we would get transfer > error ENOMEM, and i2cdetect shows there's no device at the address. > Quoting from mt8183 datasheet, the number of transfers to be > transferred in one transaction should be set to bigger than 1, > so we should forbid zero-length transfer and update functionality. > > Incorrect return: > localhost ~ # i2cdetect -q -y 0 > 0 1 2 3 4 5 6 7 8 9 a b c d e f > 00: -- -- -- -- -- -- -- -- -- -- -- -- -- > 10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > 20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > 30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > 40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > 50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > 60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > 70: -- -- -- -- -- -- -- -- > > After this patch: > localhost ~ # i2cdetect -q -y 0 > Error: Can't use SMBus Quick Write command on this bus > > localhost ~ # i2cdetect -y 0 > Warning: Can't use SMBus Quick Write command, will skip some addresses > 0 1 2 3 4 5 6 7 8 9 a b c d e f > 00: > 10: > 20: > 30: -- -- -- -- -- -- -- -- > 40: > 50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > 60: > 70: Thanks for providing these test results. Much appreciated. To shorten the commit log a little, I removed them before applying, though. > > Reported-by: Alexandru M Stan <amstan@chromium.org> > Signed-off-by: Hsin-Yi Wang <hsinyi@chromium.org> Applied to for-next, thanks!
diff --git a/drivers/i2c/busses/i2c-mt65xx.c b/drivers/i2c/busses/i2c-mt65xx.c index 252edb433fdf..29eae1bf4f86 100644 --- a/drivers/i2c/busses/i2c-mt65xx.c +++ b/drivers/i2c/busses/i2c-mt65xx.c @@ -234,6 +234,10 @@ static const struct i2c_adapter_quirks mt7622_i2c_quirks = { .max_num_msgs = 255, }; +static const struct i2c_adapter_quirks mt8183_i2c_quirks = { + .flags = I2C_AQ_NO_ZERO_LEN, +}; + static const struct mtk_i2c_compatible mt2712_compat = { .regs = mt_i2c_regs_v1, .pmic_i2c = 0, @@ -298,6 +302,7 @@ static const struct mtk_i2c_compatible mt8173_compat = { }; static const struct mtk_i2c_compatible mt8183_compat = { + .quirks = &mt8183_i2c_quirks, .regs = mt_i2c_regs_v2, .pmic_i2c = 0, .dcm = 0, @@ -870,7 +875,11 @@ static irqreturn_t mtk_i2c_irq(int irqno, void *dev_id) static u32 mtk_i2c_functionality(struct i2c_adapter *adap) { - return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL; + if (adap->quirks->flags & I2C_AQ_NO_ZERO_LEN) + return I2C_FUNC_I2C | + (I2C_FUNC_SMBUS_EMUL & ~I2C_FUNC_SMBUS_QUICK); + else + return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL; } static const struct i2c_algorithm mtk_i2c_algorithm = {
When doing i2cdetect quick write mode, we would get transfer error ENOMEM, and i2cdetect shows there's no device at the address. Quoting from mt8183 datasheet, the number of transfers to be transferred in one transaction should be set to bigger than 1, so we should forbid zero-length transfer and update functionality. Incorrect return: localhost ~ # i2cdetect -q -y 0 0 1 2 3 4 5 6 7 8 9 a b c d e f 00: -- -- -- -- -- -- -- -- -- -- -- -- -- 10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 70: -- -- -- -- -- -- -- -- After this patch: localhost ~ # i2cdetect -q -y 0 Error: Can't use SMBus Quick Write command on this bus localhost ~ # i2cdetect -y 0 Warning: Can't use SMBus Quick Write command, will skip some addresses 0 1 2 3 4 5 6 7 8 9 a b c d e f 00: 10: 20: 30: -- -- -- -- -- -- -- -- 40: 50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 60: 70: Reported-by: Alexandru M Stan <amstan@chromium.org> Signed-off-by: Hsin-Yi Wang <hsinyi@chromium.org> --- Change from v1: * restore the order of algo and quirks --- drivers/i2c/busses/i2c-mt65xx.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-)