On Thu, May 26, 2022 at 09:49:03AM +0100, Stefan Hajnoczi wrote: > On Thu, May 19, 2022 at 03:48:33PM +0100, Alberto Faria wrote: > > diff --git a/block/qcow2.c b/block/qcow2.c > > index 5493e6b847..d5a1e8bc43 100644 > > --- a/block/qcow2.c > > +++ b/block/qcow2.c > > @@ -113,7 +113,7 @@ static ssize_t qcow2_crypto_hdr_read_func(QCryptoBlock > > *block, size_t offset, > > error_setg_errno(errp, -ret, "Could not read encryption header"); > > return -1; > > } > > - return ret; > > + return buflen; > > } > > > > > > @@ -174,7 +174,7 @@ static ssize_t qcow2_crypto_hdr_write_func(QCryptoBlock > > *block, size_t offset, > > error_setg_errno(errp, -ret, "Could not read encryption header"); > > return -1; > > } > > - return ret; > > + return buflen; > > } > > I think block/crypto.c's read_func and write_func could be converted to > 0 (success) and -errno (failure) in the same way. Callers don't handle > cases where the return value != buflen or failure.
Yes, the callers expect a full buffer to be written, or for error to be reported. the 'ssize_t' return type should be converted to merely 'int' with 0 success/-1 error. With regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
