From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stephen Warren Subject: Re: [PATCH 5/6] spi: bcm2835: add polling mode for transfers below 30us Date: Mon, 30 Mar 2015 21:28:38 -0600 Message-ID: <551A1466.2020706@wwwdotorg.org> References: <808DC596-35BB-4A08-8CF4-DFAB6294636E@martin.sperl.org> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Cc: Mark Brown , lee-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org, linux-spi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-rpi-kernel To: Martin Sperl Return-path: In-Reply-To: <808DC596-35BB-4A08-8CF4-DFAB6294636E-TqfNSX0MhmxHKSADF0wUEw@public.gmane.org> Sender: linux-spi-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-ID: On 03/29/2015 08:03 AM, Martin Sperl wrote: > In cases of short transfer times the CPU is spending lots of time > in the interrupt handler and scheduler to reschedule the worker thread. > > Measurements show that we have times where it takes 29.32us to between > the last clock change and the time that the worker-thread is running again > returning from wait_for_completion_timeout(). > > During this time the interrupt-handler is running calling complete() > and then also the scheduler is rescheduling the worker thread. > > This time can vary depending on how much of the code is still in > CPU-caches, when there is a burst of spi transfers the subsequent delays > are in the order of 25us, so the value of 30us seems reasonable. > > With polling the whole transfer of 4 bytes at 10MHz finishes after 6.16us > (CS down to up) with the real transfer (clock running) taking 3.56us. > So the efficiency has much improved and also freeing CPU cycles, reducing > interrupts and context switches. > > Because of the above 30us seems to be a reasonable limit for polling. > diff --git a/drivers/spi/spi-bcm2835.c b/drivers/spi/spi-bcm2835.c > @@ -204,6 +207,37 @@ static int bcm2835_spi_transfer_one(struct spi_master *master, > + spi_used_hz = (cdiv) ? (clk_hz / cdiv) : (clk_hz / 65536); "(cdiv)" can be just "cdiv". > + /* check if we shall run in polling mode */ > + xfer_time_us = tfr->len * 9 * 1000000 / spi_used_hz; Why 9 not 8; presumably thats bits per byte, and IIRC SPI doesn't have anything like I2C's ack bit per byte? > + if (xfer_time_us <= BCM2835_SPI_POLLING_LIMIT_US) { ... > + if ((bs->rx_len) && (time_after(jiffies, timeout))) { There are many extra brackets here too. -- To unsubscribe from this list: send the line "unsubscribe linux-spi" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html