On Thu, Oct 29, 2015 at 3:47 PM, Nicolin Chen nicoleotsuka@gmail.com wrote:
On Thu, Oct 29, 2015 at 03:23:41PM -0700, Caleb Crome wrote:
I saw your problem in the other reply. And I suggested you to let DMA work first before SSI gets enabled. As SDMA in that case would transfer one burst length (16 if you applied my patch I sent you) and pause before SSI gets enabled. Then SSI would have enough data to send out without any startup issue.
Ah ha, you are exactly right. The root cause is that TE and SSIE are enabled at the same regmap write, with no opportunity for delay between the SSIE and TE. DMA can only get going if SSIE is enabled, and the only place SSIE gets enabled is exactly the same line that TE gets enabled.
A little difference between your point and mine is that you think DMA request only starts when SSIE and TDMAE both get set while I only think about TDMAE. It's hard to say which one is correct as it depends on the design of IP wrapper but you can fairly test it with your change below: Mask both TE with SSIE and set them after the delay. If it doesn't work, yours is the correct one.
Ah, that's one thing that's very clear in the FSL datasheet: the FIFOs are ZEROED if SSIE is 0. This means that even if the DMA were trying to dump data in before SSIE is enabled, the data would go to bit heaven.
The docs for TE say, "The normal transmit enable sequence is to write data to the STX register(s) and then set the TE bit." (page 5145 of IMX6SDLRM.pdf)
So in the DMA + fifo case the words, "write data to the STX register(s)" imply that it's actually DMA writing to FIFOs, which then write the STX register. So, the sequence must be: enable SSIE & TDMAE to allow DMA to write to the fifo, then later enable TE, right?
I've looked over your emails and I don't see the patch that shows a
You may need to open an offline email that I sent you with patches in its attachment. I can see it via Gmail anyway.
Will do. Thanks.
pause between SSIE enable and TE enable. (I do see the dual-fifo example -- thank you! I'll give that a try -- it may further reduce stress on the system).
I'm sure dual FIFO will get better performance. But the example I gave you doesn't set RX parameters so well. You may need to fine tune it later.
Is adding the udelay the best way to put a delay between SSIE and TE enable? Are there any other mechanisms for that?
Having a delay is much safer for you but surely it's not a common practice that's best all other platforms such as two-channel cases and those who needs performance.
I encourage you to try to follow one of patches I gave you that sets TDMAE/RDMAE at the beginning of the trigger(). Surely you may change it to TDMAE | SSIE after you find out that SSIE is indeed required. If you are still having trouble, adding a delay would be nice for you but it may be hard for me to ack it if you want to merge it in the driver.
I now I see your patch! Okay, I'll give that a go, but it's still just a race condition between the regmap_update_bits with TDMAE (your patch) verses the regmap_update_bits from fsl_ssi_config. You're just hoping that a DMA write happens between TDMAE and the end of fsl_ssi_config where TE is enabled.
Now I think I get it though. We do TMDAE + SSIEN like your patch, then a short while loop on SFCSR.TFCNT0. After the first word gets written to the fifo, TFCNT0 should go > 0, and then we can release TE.
There may be a better status register to wait on but TFCNT0 seems like it will do the trick.
What do you think of that solution? Any better register to wait on? Would that be acceptable to merge into the driver?
Thanks, -Caleb
Nicolin