[Top][All Lists]

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

Re: [Qemu-devel] [PATCH v3 2/2] qga/commands-posix: Return per path fstr

From: Justin Ossevoort
Subject: Re: [Qemu-devel] [PATCH v3 2/2] qga/commands-posix: Return per path fstrim result
Date: Fri, 01 May 2015 13:56:09 +0200
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.6.0

On 30-04-15 18:35, Thomas Huth wrote:
On Thu, 30 Apr 2015 16:29:58 +0200
Justin Ossevoort <address@hidden> wrote:

The current guest-fstrim support only returns an error if some
mountpoint was unable to be trimmed, skipping any possible additional
mountpoints. The result of the TRIM operation itself is also discarded.

This change returns a per mountpoint result of the TRIM operation. If an
error occurs on some mountpoints that error is returned and the
guest-fstrim continue with any additional mountpoints.

Signed-off-by: Justin Ossevoort <address@hidden>
  qga/commands-posix.c | 54 ++++++++++++++++++++++++++++++++++++++--------------
  qga/qapi-schema.json | 32 ++++++++++++++++++++++++++++---
  2 files changed, 69 insertions(+), 17 deletions(-)

diff --git a/qga/commands-posix.c b/qga/commands-posix.c
index 4449628..ec0d69e 100644
--- a/qga/commands-posix.c
+++ b/qga/commands-posix.c
@@ -1325,8 +1325,12 @@ static void guest_fsfreeze_cleanup(void)
   * Walk list of mounted file systems in the guest, and trim them.
-void qmp_guest_fstrim(bool has_minimum, int64_t minimum, Error **errp)
+GuestFilesystemTrimResponse *
+qmp_guest_fstrim(bool has_minimum, int64_t minimum, Error **errp)
+    GuestFilesystemTrimResponse *response;
+    GuestFilesystemTrimResultList *list;
+    GuestFilesystemTrimResult *result;
      int ret = 0;
      FsMountList mounts;
      struct FsMount *mount;
@@ -1340,39 +1344,59 @@ void qmp_guest_fstrim(bool has_minimum, int64_t 
minimum, Error **errp)
      build_fs_mount_list(&mounts, &local_err);
      if (local_err) {
          error_propagate(errp, local_err);
-        return;
+        return NULL;

+    response = g_malloc0(sizeof(*response));
      QTAILQ_FOREACH(mount, &mounts, next) {
+        result = g_malloc0(sizeof(*result));
+        result->path = g_strdup(mount->dirname);
+        list = g_malloc0(sizeof(*list));
+        list->value = result;
+        list->next = response->paths;
+        response->paths = list;
          fd = qemu_open(mount->dirname, O_RDONLY);
          if (fd == -1) {
-            error_setg_errno(errp, errno, "failed to open %s", mount->dirname);
-            goto error;
+            result->error = g_strdup_printf("failed to open: %s",
+                                            strerror(errno));
+            result->has_error = true;
+            continue;

          /* We try to cull filesytems we know won't work in advance, but other
           * filesytems may not implement fstrim for less obvious reasons.  
-         * will report EOPNOTSUPP; we simply ignore these errors.  Any other
-         * error means an unexpected error, so return it in those cases.  In
-         * some other cases ENOTTY will be reported (e.g. CD-ROMs).
+         * will report EOPNOTSUPP; while in some other cases ENOTTY will be
+         * reported (e.g. CD-ROMs).
+         * Any other error means an unexpected error.
          r.start = 0;
          r.len = -1;
          r.minlen = has_minimum ? minimum : 0;
          ret = ioctl(fd, FITRIM, &r);
          if (ret == -1) {
-            if (errno != ENOTTY && errno != EOPNOTSUPP) {
-                error_setg_errno(errp, errno, "failed to trim %s",
-                                 mount->dirname);
-                close(fd);
-                goto error;
+            result->has_error = true;
+            if (errno == ENOTTY || errno == EOPNOTSUPP) {
+                result->error = g_strdup("trim not supported");
+            } else {
+                result->error = g_strdup_printf("failed to trim: %s",
+                                                strerror(errno));
+            close(fd);
+            continue;
+        result->has_minimum = true;
+        result->minimum = r.minlen;

I'm not sure, but does this "minimum" result make sense at all? What's
the kernel supposed to return in this field? I had a quick look at some
file system implementations in the kernel, but to me it seems like only
the .len field is updated with a return value.

Without having looked at the actual implementation in the kernel, but based on discussions on the linux-kernel mailinglist, should the minimum reflect the actual minimum size it was able to trim.

I'm not certain if it's the minimum size supported by all parts of the storage stack or if it was the size of the smallest consecutive area trimmed.

It is only filled if something actually got trimmed. But that is still logical for both implementations. The reason I'm returning it, is because it get's updated by the kernel, so it might mean something to someone.

For my use-cases 'trimmed' is meaningful for monitoring, and per file system errors are useful in discovering why some filesystem might not get trimmed (and ensuring all filesystems actually get trimmed). I personally have no use for this 'minimum', so dropping it is fine by me.

+        result->has_trimmed = true;
+        result->trimmed = r.len;

+    return response;
  #endif /* CONFIG_FSTRIM */

I just also had a quick test of this patch and got this behaviour:

{"return": {"paths": [{"minimum": 4096, "path": "/mnt", "trimmed": 2040348672}, {"minimum": 4096, "path": "/mnt2", "trimmed": 2040348672}, 
{"minimum": 0, "path": "/boot", "trimmed": 388968448}, {"minimum": 0, "path": "/", "trimmed": 17699807232}]}}
{"return": {"paths": [{"minimum": 4096, "path": "/mnt", "trimmed": 0}, {"minimum": 4096, "path": "/mnt2", "trimmed": 0}, {"minimum": 0, 
"path": "/boot", "trimmed": 388968448}, {"minimum": 0, "path": "/", "trimmed": 17699799040}]}}
{"return": {"paths": [{"minimum": 4096, "path": "/mnt", "trimmed": 0}, {"minimum": 4096, "path": "/mnt2", "trimmed": 0}, {"minimum": 0, 
"path": "/boot", "trimmed": 388968448}, {"minimum": 0, "path": "/", "trimmed": 17699799040}]}}
{"return": {"paths": [{"minimum": 4096, "path": "/mnt", "trimmed": 0}, {"minimum": 4096, "path": "/mnt2", "trimmed": 0}, {"minimum": 0, 
"path": "/boot", "trimmed": 388968448}, {"minimum": 0, "path": "/", "trimmed": 17699799040}]}}
{"return": {"paths": [{"minimum": 4096, "path": "/mnt", "trimmed": 0}, {"minimum": 4096, "path": "/mnt2", "trimmed": 0}, {"minimum": 0, 
"path": "/boot", "trimmed": 388968448}, {"minimum": 0, "path": "/", "trimmed": 17699799040}]}}

/mnt and /mnt2 got successfully trimmed and consecutive calls then
reported "trimmed: 0". But the values for "/boot" and "/" do not make
sense to me, why does it claim to have always trimmed the same amount
of bytes here? (I only touched the /mnt and /mnt2 file systems before
doing the trim calls, so I wonder why there are bytes trimmed on /
and /boot at all?)

It will probably have trimmed the entire free space map the first time. I don't know if filesystem trim information is persistent or in-memory only. I suspect the response is filesystem / storage stack specific.

Also, I think you need to adjust the stub of qmp_guest_fstrim() in
commands-win32.c, too, so that you don't break the compilation of the
windows target.

Good call, will fix.


Thanks for ther review and tests.



reply via email to

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