This is the mail archive of the
ecos-discuss@sourceware.org
mailing list for the eCos project.
Re: Timing differences before / after starting scheduler
- From: Andrew Lunn <andrew at lunn dot ch>
- To: Stanislav Meduna <stano at meduna dot org>
- Cc: eCos Discussion <ecos-discuss at ecos dot sourceware dot org>
- Date: Sun, 22 Aug 2010 17:03:58 +0200
- Subject: Re: [ECOS] Timing differences before / after starting scheduler
- References: <4C712A17.1070302@gmx.de> <4C713846.7020207@meduna.org>
On Sun, Aug 22, 2010 at 04:46:30PM +0200, Stanislav Meduna wrote:
> On 22.08.2010 15:45, Tim Jaacks wrote:
>
> > I am using eCos on an LPC2378, running at 72MHz. Some driver I have
> > implemented reads out eight double words from a CPU FIFO, using eight
> > equal subsequent lines of C-code.
>
> Which FIFO? The UART FIFO is 16 bytes, why 8 DWORDs? In the
> case you mean 8 bytes (half a buffer), 8 bytes in 17 us means
> 3,75 Mbps. I don't know the LPC2378, but as far as I can see
> the UART is considered a low-speed interface and is connected
> to the APB. Maybe you are stretching the hardware a bit?
I'm guessing it is the SSP serial I/O controller. It has an 8-frame
FIFO for both RX and TX, and frames are 4-bit to 16-bit.
I'd also guess he is using it as a slave, with messages bigger than
8-frames, so has to read from the FIFO fast enough to make space for
the rest of the message. Using the DMA might be the solution...
Andrew
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss