traceviewer ETW importer breaks if Paging Executive on
Reported by
ron...@gmail.com,
Apr 18 2016
|
||||||||||
Issue description
UserAgent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/50.0.2661.75 Safari/537.36
Steps to reproduce the problem:
1. Open a web page.
2. Open a new tab with chrome://tracing and start a trace (I tried Rendering, JS with Rendering and manual settings (including all categories)).
3. Go back to the "traced" web page for few seconds, and then stop the tracing.
What is the expected behavior?
Getting a trace.
What went wrong?
Getting an error:
While importing:
Error: Cannot determine pointer size of the system trace.
at Error (native)
at Object.EtwImporter.importEvents (chrome://tracing/tracing.js:4033:7)
at Object.runImportEventsOnOneImporter (chrome://tracing/tracing.js:877:385)
at Object.timedNamedFunction (chrome://tracing/tracing.js:313:310)
at Object.Task.run (chrome://tracing/tracing.js:1048:13)
at runAnother (chrome://tracing/tracing.js:1060:136)
at runTask (chrome://tracing/tracing.js:1023:57)
at processIdleWork (chrome://tracing/tracing.js:1029:116)
at chrome://tracing/tracing.js:1017:81
Did this work before? N/A
Chrome version: 50.0.2661.75 Channel: stable
OS Version: 6.1 (Windows 7, Windows Server 2008 R2)
Flash Version: Shockwave Flash 21.0 r0
,
Apr 18 2016
Marking the above issue as Feature request and Untriage as per comment #1. Dev Team will take a final call on the above request. Thank you!
,
Apr 29 2016
=> sullivan to triage
,
Apr 29 2016
+brucedawson, georgesak: can you help triage?
,
Apr 29 2016
Hey, nice blog post. Very well written. Genius. That was my first (of many) ETW blog post. On 64-bit Windows the metadata needed for stack walking is in pageable memory by default which means that it cannot be referenced from kernel mode which means that stack walking doesn't work, unless DisablePagingExecutive is set. The exact failure modes of not having DisablePagingExecutive set have never been well explained but it is expected that some useful aspects of ETW tracing will fail, and certainly stack walking from kernel mode is on the list. So, we either need to handle not getting stacks, or not try getting stacks if DisablePagingExecutive is not set, or ??? BTW, the reason that DisablePagingExecutive is not set by default is that it wastes a little bit of memory. I think it's typically only about 20 MB, but on low-memory machines that can be important. On our machines the amount should never be significant.
,
Jun 1 2016
Moving this nonessential bug to the next milestone. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Jul 13 2016
This issue has been moved once and is lower than Pri-1. Removing the milestone. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Feb 10 2017
,
Feb 10 2017
,
Nov 2 2017
,
Jul 10
|
||||||||||
►
Sign in to add a comment |
||||||||||
Comment 1 by ron...@gmail.com
, Apr 18 2016