From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-wm1-f54.google.com (mail-wm1-f54.google.com [209.85.128.54]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 9A85F2AE96 for ; Mon, 24 Jun 2024 15:49:30 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.128.54 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1719244173; cv=none; b=mb6RxRMBvtIn2qF7sZK2Odws7vuUGNhqG1GB77SZiFefmANiy9ouVMrHjlWRezsVuOp/OoaR21eCSQtmo8RlrNCofKdaPNByETlbgKkCEyjLDfnUIWeTIz+HHqz97Yj06G68QWjtRUit35KS6QRN+rLcjKnREJdLzkqOImUgH5M= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1719244173; c=relaxed/simple; bh=a0UJSp2q0uve7X5kQB5gD7q6GeLF7iCBBBnuERdm2ag=; h=Date:From:To:Cc:Subject:Message-ID:References:MIME-Version: Content-Type:Content-Disposition:In-Reply-To; b=VPk4zfZLDxyMixTbYvhr1Fxl880RXn2dKbwsjZtVMLI6/vetdzy6/MYfbNHL0WTMd/FoAIyeh/AWg4w1oYFywGUdn6V50tdEymGL2n8dnpO7OS96BUtytjVDv4dzhI0980SAfJj0thxRL4+y9IVGQSWhJeFHuQgz0R87HEcl1Pw= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=Xz8pFZG6; arc=none smtp.client-ip=209.85.128.54 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="Xz8pFZG6" Received: by mail-wm1-f54.google.com with SMTP id 5b1f17b1804b1-42138eadf64so38263815e9.3 for ; Mon, 24 Jun 2024 08:49:30 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1719244169; x=1719848969; darn=lists.linux.dev; h=in-reply-to:content-disposition:mime-version:references:message-id :subject:cc:to:from:date:from:to:cc:subject:date:message-id:reply-to; bh=Z9kgIYkCaEISawG3z3/PKDsfUfni6CT+YfmQGd+f5vk=; b=Xz8pFZG6yWN10CfQ6/aPhS01TpLShSxVMHcte7hRMYiAOd1y2yMafNndA0pSj7cNOJ 7SXeNmc3wIXa4sZczbHAWFVIYKfj9XMkTDeFO/YBP26D5SpvQwknH7vMHwpKa1fXJ3yJ XmjhuRe1LyW6hEg3MHdg7obnnJ7u/7msDJ6I+Ohz5JAf00VKTiDGN6wHEPaYe8vCjliR cavj64Ez6qozbS0ibHTtd2wQo4MyLuH5OGkaiGR7SBHdxSBnGQeusuh/5EJjLxZ6uof+ LxbU2IHNOsISUWzA7aIYJQAELcgPv15ItIdZ+0RR5Bhw2gjkVfO0jaAJ4Oq5KPRuiz5U J+bw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1719244169; x=1719848969; h=in-reply-to:content-disposition:mime-version:references:message-id :subject:cc:to:from:date:x-gm-message-state:from:to:cc:subject:date :message-id:reply-to; bh=Z9kgIYkCaEISawG3z3/PKDsfUfni6CT+YfmQGd+f5vk=; b=NeKThPRTmXQZRW/IK+1hyGgRf1VQ4cR9MUBRokxl5uRYJnABDQ0yS7/AvZaezXkSqq AmzTOCm6ITdwKJTpZzJrZvw60G94W5CUXUS/Df0D1hIH9fEcFnkQWc5Aph78D9G2/wMu g05GCt2wGdrBKth5pejOYuwVO434eu6cVn9ObK3ZAYORKpk8sJGaacCSNmv2AKSw26kD X450aKx4PBBUMFCBMyobD1e3oSIh5k/4iYLg7+0GXeQxGSQHTr6QpnHCaLaynqbEfaVX BMVtM36IvslJe1meIcKaj0dx3lYABGtgWSkPEZmLdNa5OWiGmQEpG2StpN3JiClE6LVI 5iPQ== X-Forwarded-Encrypted: i=1; AJvYcCXT3DNXFBTqlfgY4fmRiwZ1dCRADaEH02SuhKycBEdA/MTQXGORjOdNo5g4nWhi18Txb3vRUjqwUPzV+n/SR9yaUuD/ X-Gm-Message-State: AOJu0Yy1yjPt09fQTn7rQ0aDGMZdsPb6iBBM796C3Y3bTcDVc2FHYASw Msz5oGjxLcoLgUEcicxkUMdrmjEixHrMH9RYUasmpw4Cftpvu6C8 X-Google-Smtp-Source: AGHT+IF2O0vTFTR2Jv5inHaCYhFCJMD7FD4fHIJxFEqeF2HsybT87gdE0Y0w/WzRojs+ABU4W/YZag== X-Received: by 2002:a05:600c:548b:b0:421:f04d:ebd2 with SMTP id 5b1f17b1804b1-4248cc349afmr33870795e9.22.1719244169043; Mon, 24 Jun 2024 08:49:29 -0700 (PDT) Received: from eichest-laptop ([2a02:168:af72:0:b172:75e:e42d:1a1d]) by smtp.gmail.com with ESMTPSA id 5b1f17b1804b1-4247d0be80dsm177295645e9.19.2024.06.24.08.49.27 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 24 Jun 2024 08:49:28 -0700 (PDT) Date: Mon, 24 Jun 2024 17:49:26 +0200 From: Stefan Eichenberger To: Oleksij Rempel Cc: kernel@pengutronix.de, andi.shyti@kernel.org, shawnguo@kernel.org, s.hauer@pengutronix.de, festevam@gmail.com, jic23@kernel.org, lars@metafoo.de, nuno.sa@analog.com, andriy.shevchenko@linux.intel.com, marcelo.schmitt@analog.com, gnstark@salutedevices.com, francesco.dolcini@toradex.com, wsa+renesas@sang-engineering.com, andrew@lunn.ch, linux-i2c@vger.kernel.org, imx@lists.linux.dev, linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, linux-iio@vger.kernel.org, Stefan Eichenberger Subject: Re: [RFC PATCH] i2c: imx: avoid rescheduling when waiting for bus not busy Message-ID: References: <20240531142437.74831-1-eichest@gmail.com> Precedence: bulk X-Mailing-List: imx@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Hi Oleksij, On Sat, Jun 22, 2024 at 07:02:39AM +0200, Oleksij Rempel wrote: > Hi Stefan, > > On Fri, Jun 21, 2024 at 05:22:46PM +0200, Stefan Eichenberger wrote: > > Hi Andi, Andrew, Wolfram, Oleksij, > > > > After some internal discussion we still have some questions which are > > blocking us from solving the issue. > > > > On Fri, May 31, 2024 at 04:24:37PM +0200, Stefan Eichenberger wrote: > > > From: Stefan Eichenberger > > > > > > On our i.MX8M Mini based module we have an ADS1015 I2C ADC connected to > > > the I2C bus. The ADS1015 I2C ADC will timeout after 25ms when the I2C > > > bus is idle. The imx i2c driver will call schedule when waiting for the > > > bus to become idle after switching to master mode. When the i2c > > > controller switches to master mode it pulls SCL and SDA low, if the > > > ADS1015 I2C ADC sees this for more than 25 ms without seeing SCL > > > clocking, it will timeout and ignore all signals until the next start > > > condition occurs (SCL and SDA low). This can occur when the system load > > > is high and schedule returns after more than 25 ms. > > > > > > This rfc tries to solve the problem by using a udelay for the first 10 > > > ms before calling schedule. This reduces the chance that we will > > > reschedule. However, it is still theoretically possible for the problem > > > to occur. To properly solve the problem, we would also need to disable > > > interrupts during the transfer. > > > > > > After some internal discussion, we see three possible solutions: > > > 1. Use udelay as shown in this rfc and also disable the interrupts > > > during the transfer. This would solve the problem but disable the > > > interrupts. Also, we would have to re-enable the interrupts if the > > > timeout is longer than 1ms (TBD). > > > 2. We use a retry mechanism in the ti-ads1015 driver. When we see a > > > timeout, we try again. > > > 3. We use the suggested solution and accept that there is an edge case > > > where the timeout can happen. > > > > > > There may be a better way to do this, which is why this is an RFC. > > > > > > Signed-off-by: Stefan Eichenberger > > > --- > > > drivers/i2c/busses/i2c-imx.c | 14 +++++++++++--- > > > 1 file changed, 11 insertions(+), 3 deletions(-) > > > > > > diff --git a/drivers/i2c/busses/i2c-imx.c b/drivers/i2c/busses/i2c-imx.c > > > index 3842e527116b7..179f8367490a5 100644 > > > --- a/drivers/i2c/busses/i2c-imx.c > > > +++ b/drivers/i2c/busses/i2c-imx.c > > > @@ -503,10 +503,18 @@ static int i2c_imx_bus_busy(struct imx_i2c_struct *i2c_imx, int for_busy, bool a > > > "<%s> I2C bus is busy\n", __func__); > > > return -ETIMEDOUT; > > > } > > > - if (atomic) > > > + if (atomic) { > > > udelay(100); > > > - else > > > - schedule(); > > > + } else { > > > + /* > > > + * Avoid rescheduling in the first 10 ms to avoid > > > + * timeouts for SMBus like devices > > > + */ > > > + if (time_before(jiffies, orig_jiffies + msecs_to_jiffies(10))) > > > + udelay(10); > > > + else > > > + schedule(); > > > + } > > > } > > > > > > return 0; > > > -- > > > 2.40.1 > > > > If we want to be sure that the ADS1015 I2C ADC will never timeout, we > > would have to add a patch to disable preemption during transmission. > > This would look like this: > > > > @@ -1244,6 +1248,12 @@ static int i2c_imx_xfer_common(struct i2c_adapter *adapter, > > bool is_lastmsg = false; > > struct imx_i2c_struct *i2c_imx = i2c_get_adapdata(adapter); > > > > + /* If we are in SMBus mode we need to do the transfer atomically */ > > + if (i2c_imx->smbus_mode) { > > + preempt_disable(); > > + atomic = true; > > + } > > + > > /* Start I2C transfer */ > > result = i2c_imx_start(i2c_imx, atomic); > > if (result) { > > @@ -1320,6 +1330,9 @@ static int i2c_imx_xfer_common(struct i2c_adapter *adapter, > > if (i2c_imx->slave) > > i2c_imx_slave_init(i2c_imx); > > > > + if (i2c_imx->smbus_mode) > > + preempt_enable(); > > + > > return (result < 0) ? result : num; > > } > > > > However, we are aware that disabling preemption is not a good idea. So > > we were discussing how this is normally handled with SMBus devices? Is > > it just expected that SMBus devices will timeout in rare cases? > > > > For our use case, the problem would be solved if we could get rid of the > > schedule call and replace it with a udelay. It seems that the i.MX8M > > Mini I2C controller needs a few ms to clear the IBB flag. In the > > reference manual, they write: > > > I2C bus busy bit. Indicates the status of the bus. NOTE: When I2C is > > > enabled (I2C_I2CR[IEN] = 1), it continuously polls the bus data (SDA) > > > and clock (SCL) signals to determine a Start or Stop condition. Bus is > > > idle. If a Stop signal is detected, IBB is cleared. Bus is busy. When > > > Start is detected, IBB is set. > > Unfortunately, it is not clear how often they poll. In our tests the > > issue disappeard when we used udelay instead of usleep or schedule for > > the first 10 ms. > > I'm not really happy with this variant. It can be acceptable in some > use cases, but may have not pleasant side effects on other. Since this > is still valid case, I would prefer to have a UAPI to enable polling > mode as optional optimization with a warning that it may affect latency > on other corner of the system. > > > Since we know we don't have a multi-master configuration, another way > > would be to not test for IBB and just start the transfer. We saw that > > other drivers use the multi-master device tree property to determine if > > multi-master is supported. Here another quote from the reference manual: > > > On a multimaster bus system, the busy bus (I2C_I2SR[IBB]) must be > > > tested to determine whether the serial bus is free. > > We assume this means it is not necessary to test for IBB if we know we > > are in a single-master configuration. > > I interpret this part of documentation in the same way, so it will be > great if you can implement this option. Perfect thanks a lot for the feedback. I will try to implement this option and run some tests to make sure it will not break anything. Best regards, Stefan