In the Linux kernel, the following vulnerability has been resolved:
can: ucan: fix out of bound read in strscpy() source
Commit 7fdaf8966aae ("can: ucan: use strscpy() to instead of strncpy()") unintentionally introduced a one byte out of bound read on strscpy()'s source argument (which is kind of ironic knowing that strscpy() is meant to be a more secure alternative :)).
Let's consider below buffers:
dest[len + 1]; /* will be NUL terminated / src[len]; / may not be NUL terminated */
When doing:
strncpy(dest, src, len); dest[len] = '\0';
strncpy() will read up to len bytes from src.
On the other hand:
strscpy(dest, src, len + 1);
will read up to len + 1 bytes from src, that is to say, an out of bound read of one byte will occur on src if it is not NUL terminated. Note that the src[len] byte is never copied, but strscpy() still needs to read it to check whether a truncation occurred or not.
This exact pattern happened in ucan.
The root cause is that the source is not NUL terminated. Instead of doing a copy in a local buffer, directly NUL terminate it as soon as usbcontrolmsg() returns. With this, the local firmware_str[] variable can be removed.
On top of this do a couple refactors:
ucanctlpayload->raw is only used for the firmware string, so rename it to ucanctlpayload->fw_str and change its type from u8 to char.
ucandevicerequestin() is only used to retrieve the firmware string, so rename it to ucangetfwstr() and refactor it to make it directly handle all the string termination logic.