Daniel P. Berrangé <[email protected]> writes:
> On Fri, Sep 19, 2025 at 01:30:18PM +0200, Markus Armbruster wrote:
>> Daniel P. Berrangé <[email protected]> writes:
>>
>> > With error_propagate, the stack trace from any error_abort/fatal
>> > usage will start from the error_propagate() call, which is largely
>> > useless. Using ERRP_GUARD ensures the stack trace starts from
>> > the origin that reported the error.
>>
>> Yes.
>>
>> I've been chipping at error_propagate() uses on and off for a while.
>> There are hundreds left.
>
> Are there cases where it is still OK to use error_propagate or should
> we be looking to eliminate all its usage ?
The common use of error_propagate() is to propagate an error received
from a function to the caller. This is better done with ERRP_GUARD().
qapi/error.h:
* Call a function, receive an error from it, and pass it to the caller
* - when the function returns a value that indicates failure, say
* false:
* if (!foo(arg, errp)) {
* handle the error...
* }
* - when it does not, say because it is a void function:
* ERRP_GUARD();
* foo(arg, errp);
* if (*errp) {
* handle the error...
* }
* More on ERRP_GUARD() below.
*
* Code predating ERRP_GUARD() still exists, and looks like this:
* Error *err = NULL;
* foo(arg, &err);
* if (err) {
* handle the error...
* error_propagate(errp, err); // deprecated
* }
We occasionally store errors on the heap, and use error_propagate() to move
them into an @errp argument. qapi/error.h:
* Pass an existing error to the caller:
* error_propagate(errp, err);
* This is rarely needed. When @err is a local variable, use of
* ERRP_GUARD() commonly results in more readable code.
error_propagate() can also be used to accumulate errors. This cannot be
done with ERRP_GUARD(). qapi/error.h:
* Receive and accumulate multiple errors (first one wins):
* Error *err = NULL, *local_err = NULL;
* foo(arg, &err);
* bar(arg, &local_err);
* error_propagate(&err, local_err);
* if (err) {
* handle the error...
* }
Accumulating errors is commonly a bad idea. Note that
g_propagate_error() explicitly prohibits such usage. We deviated from
it.
The *possibility* of intentional error acculumation makes conversions to
ERRP_GUARD() harder. Perhaps we should track down all uses of error
accumulation, then change error_propagate() to prohibit it.
[...]
