Issue metadata
Sign in to add a comment
|
base::Tuple -> std::tuple made content_shell debug binary size increase by 40MB
Reported by
rog...@opera.com,
Mar 10 2016
|
||||||||||||||||||||
Issue description
Version: Cr-Commit-Position: refs/heads/master@{#374878} (50.0.2648.0)
OS: Linux
What steps will reproduce the problem?
(1) Compile content_shell in debug mode
What is the expected output? What do you see instead?
https://codereview.chromium.org/1673563002 had a big impact on binary size of debug builds. See data below for detailed info.
This causes problems when cross compiling for MIPS since it barfs when linking due to that content_message_generator.o has too many sections/symbols or something[1]. The issue goes away if I change from -O0 to -Os in debug build and that is likely also why I don't see any negative effects on release builds from this CL.
[1] relocation truncated to fit: R_MIPS_GOT16 against `no symbol'
Please use labels and text to provide additional information.
Data below collected from local Linux 64bit build using clang.
Before the CL landed:
content_message_generator.o: 37075792 bytes (~35MB), .symtab contains 54202 entries.
content_shell: 2614280384 bytes (~2493MB)
After the CL landed:
content_message_generator.o: 50632696 bytes (~48MB), .symtab contains 74388 entries.
content_shell: 2656176696 bytes (~2533MB)
,
Mar 15 2016
Oh... I didn't aware the debug binary size. CCing tsepez@ as a IPC owner.
I think it's acceptable to split to a separate target.
But, is it enough to split the content_message_generator.{h,cc} to pieces instead of making separate target?
,
Mar 17 2016
Possibly, I tried to just split the .cc but that gave me no improvement. Splitting the .h appeared to be non-trivial so I did not manage to get something working to see what kind of effect it would haves. I might try a bit more later when I have more time to spare.
,
Mar 17 2017
This issue has been available for more than 365 days, and should be re-evaluated. Please re-triage this issue. The Hotlist-Recharge-Cold label is applied for tracking purposes, and should not be removed after re-triaging the issue. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Aug 23
|
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by most...@opera.com
, Mar 10 2016