On Fri, Aug 2, 2019 at 4:39 PM Erik Skultety <eskultet(a)redhat.com> wrote:
On Tue, Jul 09, 2019 at 09:23:22PM +0200, Ilias Stamatis wrote:
> On success update the domain-private data. Consider / and /boot to be
> the only mountpoints avaiable in order to be consistent with the other
> FS-related calls.
>
> Signed-off-by: Ilias Stamatis <stamatis.iliass(a)gmail.com>
> ---
> src/test/test_driver.c | 58 ++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 58 insertions(+)
>
> diff --git a/src/test/test_driver.c b/src/test/test_driver.c
> index af3503c523..8c25c679a5 100644
> --- a/src/test/test_driver.c
> +++ b/src/test/test_driver.c
> @@ -388,6 +388,8 @@ typedef struct _testDomainObjPrivate testDomainObjPrivate;
> typedef testDomainObjPrivate *testDomainObjPrivatePtr;
> struct _testDomainObjPrivate {
> testDriverPtr driver;
> +
> + bool frozen[2]; /* used by file system related calls */
> };
>
>
> @@ -400,6 +402,7 @@ testDomainObjPrivateAlloc(void *opaque)
> return NULL;
>
> priv->driver = opaque;
> + priv->frozen[0] = priv->frozen[1] = false;
>
> return priv;
> }
> @@ -3439,6 +3442,60 @@ static int testDomainUndefine(virDomainPtr domain)
> return testDomainUndefineFlags(domain, 0);
> }
>
> +
> +static int
> +testDomainFSFreeze(virDomainPtr dom,
> + const char **mountpoints,
> + unsigned int nmountpoints,
> + unsigned int flags)
> +{
> + virDomainObjPtr vm;
> + testDomainObjPrivatePtr priv;
> + size_t i;
> + int slash = 0, slash_boot = 0;
One declaration per line please. Also, we should define these explicitly as
booleans, the standard states these would return 0 and 1 respectively.
I have seen in many places in the codebase many variables declared in
the same line so that's what I thought it was okay when the variables
are somehow coupled or related. But okay, apparently the single
declaration per line rule has been added later.
> + int ret = -1;
> +
> + virCheckFlags(0, -1);
> +
> + if (!(vm = testDomObjFromDomain(dom)))
> + goto cleanup;
> +
> + if (virDomainObjCheckActive(vm) < 0)
> + goto cleanup;
> +
> + priv = vm->privateData;
> +
> + if (nmountpoints == 0) {
> + priv->frozen[0] = priv->frozen[1] = true;
> + ret = 2;
Well, this is not always true, e.g. if '/' was frozen before, ret should be 1.
Similarly, if both were frozen prior to calling this API, we should return 0.
Right. I don't remember why but I thought in the past that it made
sense to report that we froze a partition even if it was already
frozen from before. Maybe for simplicity? I can't recall my original
thinking right now.
But thinking about it again, sure, let's make it freeze only what's
not frozen already. I'm sending a new patch with implementing your
feedback.
Thanks!
Ilias
> + } else {
> + for (i = 0; i < nmountpoints; i++) {
> + if (STREQ(mountpoints[i], "/")) {
> + slash = 1;
> + } else if (STREQ(mountpoints[i], "/boot")) {
> + slash_boot = 1;
^here too, if the filesystems were already frozen, we should not account for
them in @ret.
> + } else {
> + virReportError(VIR_ERR_OPERATION_INVALID,
> + _("mount point not found: %s"),
> + mountpoints[i]);
> + goto cleanup;
> + }
> + }
> +
> + if (slash)
> + priv->frozen[0] = true;
> + if (slash_boot)
> + priv->frozen[1] = true;
> +
> + ret = slash + slash_boot;
> + }
We could do ^this or alternatively, we could introduce another iterator
@nfreeze, use it within the loop. We could also declare:
bool freeze[2];
//backup the original values:
memcpy(&freeze, priv->frozen, 2);
for (mountpoints) {
//set the helper array
if (mount[i] == / && !freeze[i]) {
freeze[i] = true;
nfreeze++;
}
else if (mount[i] == /boot && !freeze[i]) {
freeze[i] = true;
nfreeze++;
}
}
ret = nfreeze;
//steal the helper copy
memcpy(priv->frozen, &freeze, 2);
return ret;
Erik