Opened 9 years ago
Closed 6 years ago
#12770 closed bug (fixed)
[Patch] file_systems: fix gcc6 build
Reported by: | mt | Owned by: | nobody |
---|---|---|---|
Priority: | normal | Milestone: | Unscheduled |
Component: | File Systems | Version: | R1/Development |
Keywords: | Cc: | ||
Blocked By: | Blocking: | ||
Platform: | All |
Description
In ObjectTracker::AddTrackable() and ObjectTracker::RemoveTrackable(), gcc6 warms '-Werror=nonnull-compare', but I don't know if this can be NULL here. So I suppress warnings with Jamfiles.
Attachments (2)
Change History (6)
by , 9 years ago
Attachment: | 0022-file_systems-fix-gcc6-build.patch added |
---|
comment:1 by , 9 years ago
patch: | 0 → 1 |
---|
by , 9 years ago
Attachment: | file_systems_buildlog.txt added |
---|
comment:2 by , 8 years ago
patch: | 1 → 0 |
---|
comment:3 by , 8 years ago
comment:4 by , 6 years ago
Resolution: | → fixed |
---|---|
Status: | new → closed |
This was fixed by disabling null-pointer checks in the kernel.
Note:
See TracTickets
for help on using tickets.
The warning does indicate a real problem: these functions are checking the "this" pointer against NULL, which is not allowed by the C++ standard.
If this was just an attempt at a "safety check" and not meant to be used, it should be removed. If calling this function with a NULL this pointer is intended behavior, then it should be refactored to not need that (for example, it could be made a static method explicitly taking the object in a parameter).