bus->read() returns a negative errno on failure, but mt7530_regmap_read() assigns it to a u16, truncating e.g. -ETIMEDOUT into 0xff92, and returns success. The garbage word is then consumed as register data, and read-modify-write cycles write it back to the switch. Check both reads and propagate their errors. The same defect existed in mt7530_mii_read() since the driver was introduced and moved into the regmap backend unchanged.
Unified · Split
| @@ -55,8 +55,15 @@ mt7530_regmap_read(void *context, unsigned int reg, unsigned int *val) · 54 unmodified lines | |||
| 55 | if (ret < 0) | 55 | if (ret < 0) |
| 56 | return ret; | 56 | return ret; |
| 57 | 57 | ||
| 58 | lo = bus->read(bus, priv->mdiodev->addr, r); | 58 | ret = bus->read(bus, priv->mdiodev->addr, r); |
| 59 | hi = bus->read(bus, priv->mdiodev->addr, 0x10); | 59 | if (ret < 0) |
| 60 | return ret; | ||
| 61 | lo = ret; | ||
| 62 | |||
| 63 | ret = bus->read(bus, priv->mdiodev->addr, 0x10); | ||
| 64 | if (ret < 0) | ||
| 65 | return ret; | ||
| 66 | hi = ret; | ||
| 60 | 67 | ||
| 61 | *val = (hi << 16) | (lo & 0xffff); | 68 | *val = (hi << 16) | (lo & 0xffff); |
| 62 | 69 | ||