Hello,
this is two patches (one for OCL and one for RTT) to add real-time output
string stream to the OCL::Logging. With that, you could do something like:
my_rt_log_cat.getRTStream(log4cpp::Priority::WARN) << "Here is the variable
content: " << myVar;
The patch has been generated for toolchain-2.2. It would be great if this
could be merged for the next minor release.
Philippe Hamelin
| Attachment | Size |
|---|---|
| 0001-OCL-Add-real-time-output-string-stream.patch | 6.67 KB |
| 0001-RTT-Add-real-time-output-string-stream.patch | 1 KB |
[Patch] Add real-time logging output string stream
On Jan 26, 2011, at 11:09 , Philippe Hamelin wrote:
> Hello,
>
> this is two patches (one for OCL and one for RTT) to add real-time output string stream to the OCL::Logging. With that, you could do something like:
>
> my_rt_log_cat.getRTStream(log4cpp::Priority::WARN) << "Here is the variable content: " << myVar;
>
> The patch has been generated for toolchain-2.2. It would be great if this could be merged for the next minor release.
>
> Philippe Hamelin
>
[Patch] Add real-time logging output string stream
2011/3/9 S Roderick <kiwi [dot] net [..] ...>
> On Jan 26, 2011, at 11:09 , Philippe Hamelin wrote:
>
> Hello,
>
> this is two patches (one for OCL and one for RTT) to add real-time output
> string stream to the OCL::Logging. With that, you could do something like:
>
> my_rt_log_cat.getRTStream(log4cpp::Priority::WARN) << "Here is the variable
> content: " << myVar;
>
> The patch has been generated for toolchain-2.2. It would be great if this
> could be merged for the next minor release.
>
> Philippe Hamelin
>
> <0001-OCL-Add-real-time-output-string-stream.patch>
> <0001-RTT-Add-real-time-output-string-stream.patch>--
>
>
> FYI preliminary testing shows repeated minor page faults using this
> approach, which does not occur with the boost::iostreams version. I know
> that major page faults are non-realtime, but I need to research some to
> check on minor page faults. Does anyone else know?
>
> According to [1] minor page faults "cause relatively small latencies"
>
> Tested on Ubuntu Linux 10.04 w/ 2.6.31-11-rt
> S
>
> [1] https://rt.wiki.kernel.org/index.php/HOWTO:_Build_an_RT-application
>
The only thing I really tested concerning real-time is to check that there's
no system call during a log (using strace). I don't know much about page
faults. Please excuse my ignorance, but how do you know that there's page
faults?
Philippe
[Patch] Add real-time logging output string stream
On Mar 9, 2011, at 08:30 , Philippe Hamelin wrote:
> 2011/3/9 S Roderick <kiwi [dot] net [..] ...>
> On Jan 26, 2011, at 11:09 , Philippe Hamelin wrote:
>
>> Hello,
>>
>> this is two patches (one for OCL and one for RTT) to add real-time output string stream to the OCL::Logging. With that, you could do something like:
>>
>> my_rt_log_cat.getRTStream(log4cpp::Priority::WARN) << "Here is the variable content: " << myVar;
>>
>> The patch has been generated for toolchain-2.2. It would be great if this could be merged for the next minor release.
>>
>> Philippe Hamelin
>>
[Patch] Add real-time logging output string stream
On Wednesday 26 January 2011 17:09:29 Philippe Hamelin wrote:
> Hello,
>
> this is two patches (one for OCL and one for RTT) to add real-time output
> string stream to the OCL::Logging. With that, you could do something like:
>
> my_rt_log_cat.getRTStream(log4cpp::Priority::WARN) << "Here is the variable
> content: " << myVar;
>
> The patch has been generated for toolchain-2.2. It would be great if this
> could be merged for the next minor release.
Seems not a problem to me. The patch includes a big part /*...*/ commented
out. Should it be removed from the patch or did you forget to implement them
in the header ?
Peter
[Patch] Add real-time logging output string stream
2011/1/27 Peter Soetens <peter [..] ...>
> On Wednesday 26 January 2011 17:09:29 Philippe Hamelin wrote:
> > Hello,
> >
> > this is two patches (one for OCL and one for RTT) to add real-time output
> > string stream to the OCL::Logging. With that, you could do something
> like:
> >
> > my_rt_log_cat.getRTStream(log4cpp::Priority::WARN) << "Here is the
> variable
> > content: " << myVar;
> >
> > The patch has been generated for toolchain-2.2. It would be great if this
> > could be merged for the next minor release.
>
> Seems not a problem to me. The patch includes a big part /*...*/ commented
> out. Should it be removed from the patch or did you forget to implement
> them
> in the header ?
>
>
You're my best code reviewer Peter... that's right this is dead code.. sorry
:-)