[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] No change when set_processor affinity() is used

From: Nemanja Savic
Subject: Re: [Discuss-gnuradio] No change when set_processor affinity() is used
Date: Tue, 12 May 2015 19:07:02 +0200

You are right, cause I was looking in a process manager and only one core was 100% busy, but since the middle filter had 37k taps it might be that the other two were starving a little bit.
BTW, in the shown flograph, I would like to have signal in upper and down paths synchronized, so I introduced delay of (N-1)/2, where N is the number of taps in the middle filter. However signals in the files are not synchronized. What could cause that?

On Tue, May 12, 2015 at 7:01 PM, Marcus D. Leech <address@hidden> wrote:
On 05/12/2015 12:52 PM, Nemanja Savic wrote:

You are probably right, cause that file doesn't even exist. I looked at default gnuradio-core.conf in conf.d.
I suppose that something like that should be written in gnuradi-core.conf, but really nothing.
Where can I find which option should be added?
How are you determining that this is only occupying a single core?

Are you running on a VM?   If so, how many cores does your VM simulate?

On Tue, May 12, 2015 at 6:33 PM, Marcus D. Leech <address@hidden> wrote:
On 05/12/2015 12:25 PM, Nemanja Savic wrote:
Hi all guys,

I have a flowgraph where I have three parallel paths for filtering signal stored in a file. Here the picture of my flowgraph:

Inline image 1

I use gnuradio When I run the script it uses only one processor, and since I have 4 cores, I would like to run every of the paths on another processor. For every filter I do set_processr_affinity[some number between 0 and ]. When I run the script nothing changes, it still runs on a single core. Am I missing something with set_processor_affinity method?


Nemanja Savić

Discuss-gnuradio mailing list
My guess is that your ~/.gnuradio/config.conf   specifies to use the single-threaded scheduler.

Nemanja Savić

Nemanja Savić

reply via email to

[Prev in Thread] Current Thread [Next in Thread]