| 1 |
On Tue, Nov 14, 2017 at 8:42 PM, R0b0t1 <r030t1@×××××.com> wrote: |
| 2 |
> On Tue, Nov 14, 2017 at 11:36 AM, Jorge Almeida <jjalmeida@×××××.com> wrote: |
| 3 |
>> On Fri, Nov 10, 2017 at 12:09 PM, Jorge Almeida <jjalmeida@×××××.com> wrote: |
| 4 |
>> |
| 5 |
>>> http://www.daemonology.net/blog/2014-09-04-how-to-zero-a-buffer.html |
| 6 |
>>> |
| 7 |
>>> |
| 8 |
>>>>> Of course, what would really solve the optimize-into-oblivion problem |
| 9 |
>>>>> is a pragma that when invoked on a particular block of code (maybe |
| 10 |
>>>>> only a function definition) would tell the compiler to do what the |
| 11 |
>>>>> programmer says rather than viewing a function as a kind of black box. |
| 12 |
>>>>> |
| 13 |
>>>> |
| 14 |
>> |
| 15 |
>> It seems a solution exists with gcc: |
| 16 |
>> |
| 17 |
>> https://stackoverflow.com/questions/2219829/how-to-prevent-gcc-optimizing-some-statements-in-c |
| 18 |
>> |
| 19 |
>> The last reply: |
| 20 |
>> |
| 21 |
>> void __attribute__((optimize("O0"))) foo(unsigned char data) { |
| 22 |
>> // unmodifiable compiler code |
| 23 |
>> } |
| 24 |
>> |
| 25 |
>> Any comments, anyone? Yes, it's gcc, but IMO this should be in the |
| 26 |
>> language itself. Am I right to assume this is a poorly known feature |
| 27 |
>> of gcc? |
| 28 |
>> It allows, for example, to replace sensitive data by random bytes, |
| 29 |
>> existing system callls like memset() or getrandom() can be used as |
| 30 |
>> they are, no reimplementation needed. |
| 31 |
>> |
| 32 |
> |
| 33 |
> Very interesting. I imagine the opinion of the standards committee |
| 34 |
> would be that the variability in code generation precludes a standard |
| 35 |
> interface to optimization controls. This might seem unusual, but |
| 36 |
> languages with a very controlling standard (like Java or C#) are a new |
| 37 |
> concept. |
| 38 |
|
| 39 |
Well, we'll have to stick to gcc (or other compilers with the same |
| 40 |
feature). OTOH, boldness doesn't seem to be the comittee's most |
| 41 |
salient feature. |
| 42 |
|
| 43 |
> |
| 44 |
> What I am wondering about is if C code which uses |
| 45 |
> __attribute__((optimize(...))) is against Gentoo package standards and |
| 46 |
> would have to be removed from the Portage tree. |
| 47 |
> |
| 48 |
|
| 49 |
|
| 50 |
You can set your optimization preferences in make.conf, and still an |
| 51 |
ebuild will override them if deemed unsafe. What would be the |
| 52 |
difference? |
| 53 |
|
| 54 |
Cheers |
| 55 |
|
| 56 |
Jorge |
Archives