Remove one more use of STREQLEN with strlen as its argument.

In the unlikely case the iSCSI session path exists, but does not
contain an entry starting with "target", we would silently use
an initialized value.

Rewrite the function to correctly report errors.

Do not call it with a magic constant matching the length
of the pattern.

Many moons ago, commit id '8d7800a5' adjusted the format of the output
to add a space on the HEADER and the DATA... the docs weren't updated to
reflect that... This makes that adjustment.

On PPC the legacy passthrough is not supported and only
VFIO is supported. So, the checks at places to confirm if the
host is passthrough capable checks only legacy, fix it. This
is seen at only one place now.
Signed-off-by: NShivaprasad G Bhat <sbhat@linux.vnet.ibm.com>

Incorrect conflict resolution in my commit e81de04c broke this.

Prefer virDirOpen.

Remove all the remaining usage of opendir.

A helper function that does not report any errors.

Use it instead of opendir everywhere we need to check for ENOENT.

Just like virDirOpen, but it returns 0 without reporting an error
on ENOENT.

Switch from opendir to virDirOpen everywhere we need to report an error.

A helper that calls opendir and reports an error if it fails.

After commit e808d3f2 cbdata is always available here, so the
check is pointless

It may cause unwanted behaviour (of course, is there any wanted one for
that case?) so we should rather disable the possibility of doing so.

Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1320893Signed-off-by: NMartin Kletzander <mkletzan@redhat.com>

Put the 'continue' statement on a separate line.

Unfortunately, we can't just call qemuDomainMachineIsPSeries()
here, because we don't have a virDomainDef instance; that said,
the open-coded check should match said function as closely as
possible.

This new function checks for both the architecture and the
machine type, so we can use it instead of writing the same
checks over and over again.

Remove all external architecture checks that have been
made redundant by this change.

Since we're already passing the full virDomainDef, it doesn't
make sense to also pass def->os.arch and def->os.machine as
separate arguments.

Add SASL at the end to make the format obvious.

Add SASL at the end to make the format obvious.

Rename them to virNetSocket{Local,Remote}AddrStringSASL
to make their format more obvious.

Make it more obvious that these are in the SASL format.

Use it in virNetServerClientGetInfo to switch back to using
the URI-format (separated by ':') instead of the SASL format
(separated by ';').

Also use it in the error message reported by virNetServerAddClient.

It will return the socket address and port in a URI-like
format: [::1]:1234
Add a test case to virnetsockettest.

This partially reverts commit 9b45c9f0.

It changed the default format of socket address from the one SASL
requires, but did not adjust all the callers.

It also removed the test coverage for it.

Revert most of the changes except the virSocketAddrFormatFull support
for URI-formatted strings.

This fixes https://bugzilla.redhat.com/show_bug.cgi?id=1345743 while
reverting the format used by virt-admin's client-info command from
the URI one to the SASL one.

https://bugzilla.redhat.com/show_bug.cgi?id=1345743

'.' and '..' are skipped by virDirRead already.

This fixes the disappearance of domains and networks starting with a
dot.

https://bugzilla.redhat.com/show_bug.cgi?id=1333248

The device names are unlikely to start with a dot.
'.' and '..' are already skipped by virDirRead.

The directories we iterate over are unlikely to contain any entries
starting with a dot, other than '.' and '..' which is already skipped
by virDirRead.

'.' and '..' are now skipped by virDirRead, there's no need to mention
them in the comment.

It skips those directory entries.

All of the callers either skip these explicitly, skip all entries
starting with a dot or match the entry name against stricter patterns.

Introduce a helper that only calls closedir if DIR* is non-NULL
and sets it to NULL afterwards.

The only possible error is EBADFD.
Since we only use the directory stream returned by opendir,
this should never happen.

The virUSBDeviceFind* callers do not check errno after calling
this function.

In cases where we expect parse failure of the test input file the
testsuite can't differentiate if the parser failed when parsing or when
opening the file. Add a call to virFileExists and error out on missing
input files.

Missing output files are partially expected when regenerating test
output.