1 |
On Tue, 2011-07-05 at 13:02 +0000, gentoo-amd64+help@l.g.o |
2 |
wrote: |
3 |
> On Thu, 30 Jun 2011 21:36:38 -0700 |
4 |
> Mark Knecht <markknecht@×××××.com> wrote: |
5 |
> |
6 |
> > |
7 |
> > I think it's completely appropriate for this list. This |
8 |
> distro expects |
9 |
> > that we put CFLAG options in make.conf so I need to hear |
10 |
> about this |
11 |
> > stuff even if I don't have to background to completely |
12 |
> understand |
13 |
> > what's really causing the problem. |
14 |
> > |
15 |
> |
16 |
> In this case, or in the case of any program where |
17 |
> "-fno-strict-aliasing" |
18 |
> could make a difference, the maintainer of the program would |
19 |
> include |
20 |
> the option in the ebuild. The user would not have to worry |
21 |
> too much |
22 |
> about it. |
23 |
> |
24 |
> But yes, it is always good to know about the compiler flags. |
25 |
> |
26 |
> To see exactly what compile flags are being used in your |
27 |
> programs, here |
28 |
> is a neat method I picked up from somewhere. Just open a |
29 |
> terminal and |
30 |
> enter the following command: |
31 |
> |
32 |
> echo 'int main(){return 0;}' > test.c && gcc -v -Q $CFLAGS |
33 |
> test.c -o test && rm test.c test |
34 |
> |
35 |
> In place of $CFLAGS just substitute any option of interest. |
36 |
> There will |
37 |
> be a flood of output, but just scroll back a few lines to find |
38 |
> the "options |
39 |
> passed:" and "options enabled:" sections. |
40 |
> |
41 |
> For example, using "-O2" for $CFLAGS indicates that |
42 |
> "-fstrict-aliasing" is |
43 |
> used, but it is not used with "-O1." |
44 |
> |
45 |
> It also shows that with "-O2" the option "-mno-sse4" is used, |
46 |
> and so if |
47 |
> you want to use SSE4 for certain programs (e.g. video, audio) |
48 |
> you will |
49 |
> need to specifically enable it. |
50 |
> |
51 |
> There may be an even slicker way to reveal the flags, but this |
52 |
> is the |
53 |
> only way I know. |
54 |
> |
55 |
> Frank Peters |
56 |
> |
57 |
|
58 |
An example of the issue discussed |
59 |
https://bugs.gentoo.org/show_bug.cgi?id=347818 |
60 |
|
61 |
see in particular comment #3 they are understaffed and have a ton to do |
62 |
but in the norm don't care to here it. |
63 |
|
64 |
https://bugs.gentoo.org/show_bug.cgi?id=339485 |
65 |
|
66 |
Is a discussion/flame about the report upstream qa messages. |
67 |
Help me out here guys and weigh in. (dons flame suit) |
68 |
|
69 |
Further on the conversations wandered-to in the topic 'optimization' |
70 |
in the kernel config menu under the heading 'General Setup' lies |
71 |
[*] Optimize trace point call sites |
72 |
|
73 |
CONFIG_JUMP_LABEL: |
74 |
|
75 |
If it is detected that the compiler has support for "asm goto", the |
76 |
kernel will compile trace point locations with just a nop instruction. |
77 |
When trace points are enabled, the nop will be converted to a jump to |
78 |
the trace function. This technique lowers overhead and stress on the |
79 |
branch prediction of the processor. |
80 |
|
81 |
I've had this checked for a good deal of time and see no issues because |
82 |
of it. |
83 |
|
84 |
David J Cozatt |
85 |
aka user99 |