[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-block] [PATCH 0/2] Dynamic module loading for block drivers

From: Stefan Hajnoczi
Subject: Re: [Qemu-block] [PATCH 0/2] Dynamic module loading for block drivers
Date: Fri, 17 Jun 2016 10:54:51 +0100
User-agent: Mutt/1.6.1 (2016-04-27)

On Wed, Jun 15, 2016 at 02:40:53PM -0400, Colin Lord wrote:
> 1) Denis Lunev suggested having block_module_load_one return the
> loaded driver to reduce duplicated for loops in many of the functions
> in block.c. I'd be happy to do this but wasn't completely sure how
> error handling would happen in that case since currently the return
> value is an integer error code. Would I switch to using the
> error handling mechanisms provided in util/error.c?

Yes, change "int foo(...)" to "MyObject *foo(..., Error **errp)".  The
Error object allows functions to provide detailed, human-readable error
messages so it can be a win.

If this change would cause a lot of changes you can stop the refactoring
from snowballing using error_setg_errno() to bridge new Error functions
with old int -errno functions:

  MyObject *foo(..., Error **errp)
      /* I don't want propagate Error to all called functions yet, it
       * would snowball.  So just wrap up the errno:
      ret = legacy_function(...);
      if (ret < 0) {
          error_setg_errno(errp, -ret, "legacy_function failed");
          return NULL;

Attachment: signature.asc
Description: PGP signature

reply via email to

[Prev in Thread] Current Thread [Next in Thread]