Re: [PATCH] Refactor Mantis DMA transfer to deliver 16Kb TS data per interrupt

2011-06-05 Thread Marko Ristola
Hi Mauro and Manu I think that the patch is robust enough to be accepted. Please Mauro accept this patch at appropriate time. Would you Manu, please, read my lengthy detailed description of this patch when you have time? Then you know what it does and why. I provided a descriptive table of the

Re: [PATCH] Refactor Mantis DMA transfer to deliver 16Kb TS data per interrupt

2011-06-01 Thread Mauro Carvalho Chehab
Manu, Em 13-08-2010 10:51, Stefan escreveu: > Hello Marko > > I confirm this patch reduces the amount of interrupts to nearly one third: > > [*] v4l mercurial wo patch: about 600 calls/sec over a 10 seconds interval > [*] v4l mercurial with patch: about 160 calls/sec over a 10 seconds interval >

Re: [PATCH] Refactor Mantis DMA transfer to deliver 16Kb TS data per interrupt

2010-08-13 Thread Stefan
Hello Marko I confirm this patch reduces the amount of interrupts to nearly one third: [*] v4l mercurial wo patch: about 600 calls/sec over a 10 seconds interval [*] v4l mercurial with patch: about 160 calls/sec over a 10 seconds interval Measured using powertop -t 10 Tuning a Twinhan/Azurewave

[PATCH] Refactor Mantis DMA transfer to deliver 16Kb TS data per interrupt

2010-08-07 Thread Marko Ristola
This patch obsoletes patch with broken spaces https://patchwork.kernel.org/patch/107036/ With VDR streaming HDTV into network, generating an interrupt once per 16kb, implemented in this patch, seems to support more robust throughput with HDTV. Fix leaking almost 64kb data from the previous TS af