Here are a few ways (from www.dbspecialists.com) to check for waits and get stats on flashback:

——————————
Consider performance.

When a database is enabled for flashback, a new background process called RVWR is created. This process writes flashback data to the flashback logs. A new wait event called “flashback buf free by RVWR” shows delays writing to the flashback logs. If this wait event becomes significant then the accumulation of the flashback data is causing delays. There is little that can be done about this except to increase the disk bandwidth to the flash recovery area. Generally, the flash recovery area is a candidate for “slower, cheaper disks” since it generally holds just disk backups. However, this may not be sufficient for the flashback logs.

There is also a new system statistic in v$sysstat called “flashback log writes” that indicates the number of write operations to the flashback logs. In addition, there is a new view called v$flashback_database_stat that records the number of bytes written to the flashback logs, the database files and the redo logs during various intervals. This provides an indication of the relative overhead of the flashback logs and the flashback database feature which uses them.
——————————