On 12/07/2014 05:23 PM, Petr Machata wrote:
Josh Stone <jistone(a)redhat.com> writes:
> I'll see if I can grab that old kernel debuginfo to do a more direct
> comparison.
You could grab the old code and compare that. If you're still in the
80's, it's the data, not the code.
Ok, well my former optimizations went into 0.158. Here are some quick
runs of "varlocs -k >/dev/null" on kernel-3.17.4-301.fc21.x86_64:
0.157 96.00user 0.12system 1:36.08elapsed
0.158 70.22user 0.12system 1:10.32elapsed
0.159 71.40user 0.11system 1:11.48elapsed
0.160 73.24user 0.12system 1:13.33elapsed
master 77.65user 0.10system 1:17.72elapsed
patched 77.87user 0.10system 1:17.93elapsed
So each release has ceded a bit of performance, but this particular
patch isn't much of an issue. Correctness trumps performance anyway.
For fun, here are the counts from dwgrep -e 'entry attribute form':
3.11.9-300.fc20.x86_64/vmlinux:
2 DW_FORM_block2
252 DW_FORM_block1
24975 DW_FORM_data4
182227 DW_FORM_data8
259701 DW_FORM_sdata
330195 DW_FORM_addr
387491 DW_FORM_sec_offset
537586 DW_FORM_exprloc
558135 DW_FORM_string
1397719 DW_FORM_flag_present
3353170 DW_FORM_data2
5220907 DW_FORM_strp
9263689 DW_FORM_ref4
13543053 DW_FORM_data1
3.17.4-301.fc21.x86_64/vmlinux:
2 DW_FORM_block2
256 DW_FORM_block1
26740 DW_FORM_data4
112315 DW_FORM_exprloc
197526 DW_FORM_data8
213497 DW_FORM_sec_offset
340039 DW_FORM_sdata
358462 DW_FORM_addr
641163 DW_FORM_string
1578298 DW_FORM_flag_present
3857598 DW_FORM_data2
6023993 DW_FORM_strp
10463444 DW_FORM_ref4
15532042 DW_FORM_data1
Also .text size grew from 10082549 to 11513127.
So it's all bigger, but roughly proportional.