View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0007580 | MMW v4 | Synchronization | public | 2011-03-24 15:35 | 2011-04-07 19:20 |
| Reporter | rusty | Assigned To | |||
| Priority | urgent | Severity | major | Reproducibility | always |
| Status | closed | Resolution | fixed | ||
| Product Version | 4.0 | ||||
| Target Version | 4.0 | Fixed in Version | 4.0 | ||
| Summary | 0007580: MTP Synchronization takes ~5-10s to initiate (regression) | ||||
| Description | In build 1356, whenenever an autosync is initiated to an MTP device, it takes ~5-10 seconds before any sync activity begins (or at least appears to begin as evidenced by the status bar). This leads the user to believe that the sync button isn't working. | ||||
| Tags | No tags attached. | ||||
| Attached Files | |||||
| Fixed in build | 1359 | ||||
|
|
I don't see any difference, sync starts immediately here. Are you sure it's a regression, isn't it a scanning of a quite large device? Maybe a debug log would help to clarify (with indication of when does which action begin). |
|
|
Upon further testing, I think that the bug is actually a UI bug (UI redrawing is holding up the initiation of synchronization), and that it might be related to issue 0007586. Here is a video of the issue: http://www.screencast.com/users/rustymonkey/folders/Jing/media/1f3e7199-848e-4ce7-9d3c-e98f7fed9476 Debug log is attached as well. |
|
|
Follow up: there's definitely a UI bug, but the delay is specific to MTP devices. i.e. when I do the exact same test with an iPod, there's only a ~1s delay. |
|
|
It's apparent from the video that the problem is in showing the in-place device config dialog. The debug log doesn't show any clear reason for the slowdown though. In case we can't find any reason, some debug messages could be added to showing of the dialog. I wonder whether it couldn't be e.g. retrieval of device total&free space? All such actions should be done only after the dialog is shown in order to avoid any such problems. |
|
|
Added useful debug strings in build 1359. My expectation is that this regression is related to Supported formats reading (related to issue 0007595) and I also probably found the problem in the code and fixed in build 1359 althoug on my DB the performance glitch wasn't so critical so there might be something else. |
|
|
This is resolved in 1361. |
