Hello Lothar Rubusch,
Commit e05ce444e9e5 ("crypto: atmel-sha204a - add reading from otp
zone") from May 3, 2024 (linux-next), leads to the following Smatch
static checker warning:
drivers/crypto/atmel-sha204a.c:109 atmel_sha204a_otp_read()
warn: should this return really be negated?
drivers/crypto/atmel-sha204a.c
94 static int atmel_sha204a_otp_read(struct i2c_client *client, u16 addr, u8 *otp)
95 {
96 struct atmel_i2c_cmd cmd;
97 int ret = -1;
ret is set to -EPERM here.
98
99 if (atmel_i2c_init_read_otp_cmd(&cmd, addr) < 0) {
100 dev_err(&client->dev, "failed, invalid otp address %04X\n",
101 addr);
102 return ret;
It would be more readable to return a literal.
103 }
104
105 ret = atmel_i2c_send_receive(client, &cmd);
106
107 if (cmd.data[0] == 0xff) {
108 dev_err(&client->dev, "failed, device not ready\n");
--> 109 return -ret;
atmel_i2c_send_receive() either returns zero on success or a negative
kernel error code or a positive error code from error_list[]. It really
doesn't work to mix them like this where -EPERM and negative 0x01 could
be confused.
110 }
111
112 memcpy(otp, cmd.data+1, 4);
113
114 return ret;
115 }
regards,
dan carpenter