Clean up base/debug/stack_trace.* |
||||||
Issue descriptionbase/debug/stack_trace.* has a mixture of the StackTrace class, for capturing and symbolizing back-traces for use in e.g. CHECK(), and a faster but less reliable stack trace capturer suitable for builds which use frame pointers, for use by memory-infra. We would like to: 1. Either move the fast/unreliable unwinder out of base/debug/stack_trace.* to its own location (e.g. base/debug/fast_stack_trace.*), or preferably find a way to fold it into StackTrace (e.g. try fast unloading if frame-pointers are available and fall-back to slow if we get bogus looking results). 2. Move the platform-specific logic that has accrued in stack_trace.* back out to stack_trace_posix.* etc, where possible.
,
Mar 27 2017
Additional cleanup items: 3. Move the HAVE_TRACE_STACK_FRAME_POINTERS macro definition from base/debug/stack_trace.h out to a GN rule, and set it based on the relevant build args (mainly platform, architecture, whether profiling, or component builds, are enabled).
,
Mar 27 2017
Note that some platforms (e.g. Android 64-bit ARM) always builds with frame pointers enabled, so ideally the GN configuration definition of the macro should be tweaked to account for that as well.
,
Jun 26 2017
,
Jun 27 2017
I'll take this one for now.
,
Jun 28 2018
This issue has been Available for over a year. If it's no longer important or seems unlikely to be fixed, please consider closing it out. If it is important, please re-triage the issue. Sorry for the inconvenience if the bug really should have been left as Available. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Jul 13
|
||||||
►
Sign in to add a comment |
||||||
Comment 1 by w...@chromium.org
, Mar 9 2017Status: Assigned (was: Untriaged)