Received: by 2002:ac0:a5a6:0:0:0:0:0 with SMTP id m35-v6csp6041752imm; Wed, 12 Sep 2018 15:32:06 -0700 (PDT) X-Google-Smtp-Source: ANB0VdYd3W2XlmlKNjcgK2bGDFRZ5V3iBmd/+3wa4wGHCw2rT99vHF2nVxySyGk0j08KauzskDYc X-Received: by 2002:a17:902:b282:: with SMTP id u2-v6mr4266464plr.123.1536791526521; Wed, 12 Sep 2018 15:32:06 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1536791526; cv=none; d=google.com; s=arc-20160816; b=WHWhSxBU9zKYABhfct+ygJ606Dg06SL+OvKYQkk1hwdY8ylyr6rihZZqKU0Atp7rz0 00tNfWk8BJq8+QF5vTuxkMYdgrSSF7bQKLBTSK05z1DD2dxPjvNS5WmN7X/7vIMeDyHC nzl/l15kEBl9YZPfpWisoAsrs01/JZmU7NdluLw3J1jX1lqPK1teqeDaPRxZnKYcGPWA WJJrDuFKBpxVfaaHt/eXf/soEIMxvSy55tpE0daYoDJbztgfmMe3PN1+PQ7ChYBARPVO ytZwePNaK6/FDV61cWF1Fs6k7vHESsmcjK5zTGQjBAjsL7aKvEoe+GNhH7JG1SpSbTjD xLDw== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:sender:content-transfer-encoding :content-language:in-reply-to:mime-version:user-agent:date :message-id:from:references:cc:to:subject; bh=mbDrXy2l2Gat6EbzEWh0uDJUZNLIPZKP8a+4HBNCKeY=; b=zUGGYOHIL1Vp48xxn/DMi47cKViexeOelW1YvZmb8acqdfuCFib51gikMhla/t4jTb FFMhVu8bgCrmRC/+KHxouDxrLr73+CIQX3LcwbSE96/1yoqPneeW9oz+mKJDGXWeAv5W wKR6C2hhYodHMwX9ieNM4QdGsE418uo4Yk8HkCsGAWkCrJc/npfRv7qIlg/NcHwlm5ng /lVnTEG5qSY70Q7OK1+IBfLhJ+p6gOHh2cW1L8VrQwpxTHhe8yHl6lDKqq4G3Yrx1BlT xyP3zeYA2mL94vpkektTbBMkNBhAhKz9RhKWU6ULliUTuwXrCMSJxLaC6nTlpsfqGEvP ReiA== ARC-Authentication-Results: i=1; mx.google.com; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=intel.com Return-Path: Received: from vger.kernel.org (vger.kernel.org. [209.132.180.67]) by mx.google.com with ESMTP id r20-v6si1806170pgk.207.2018.09.12.15.31.50; Wed, 12 Sep 2018 15:32:06 -0700 (PDT) Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67; Authentication-Results: mx.google.com; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=intel.com Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728054AbeIMDhn (ORCPT + 99 others); Wed, 12 Sep 2018 23:37:43 -0400 Received: from mga09.intel.com ([134.134.136.24]:17634 "EHLO mga09.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727739AbeIMDhn (ORCPT ); Wed, 12 Sep 2018 23:37:43 -0400 X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga008.jf.intel.com ([10.7.209.65]) by orsmga102.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 12 Sep 2018 15:31:07 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.53,366,1531810800"; d="scan'208";a="72742415" Received: from yoojae-mobl1.amr.corp.intel.com (HELO [10.241.240.60]) ([10.241.240.60]) by orsmga008.jf.intel.com with ESMTP; 12 Sep 2018 15:31:06 -0700 Subject: Re: [PATCH i2c-next v6] i2c: aspeed: Handle master/slave combined irq events properly To: Guenter Roeck Cc: Joel Stanley , linux-aspeed@lists.ozlabs.org, Vernon Mauery , OpenBMC Maillist , Brendan Higgins , Linux Kernel Mailing List , linux-i2c@vger.kernel.org, jarkko.nikula@linux.intel.com, =?UTF-8?Q?C=c3=a9dric_Le_Goater?= , Linux ARM , James Feist References: <1f34fe8c-69ef-5f2d-25dc-d5f6037cc558@linux.intel.com> <20180911204107.GA26017@roeck-us.net> <20180911233302.GA18799@roeck-us.net> <5698ca34-14c9-8d05-c4e6-5acf85ff9d14@linux.intel.com> <20180912013449.GA12612@roeck-us.net> <7fd98646-fb5a-be4d-ce37-84b74e0fa8b3@linux.intel.com> <20180912195844.GA6893@roeck-us.net> <20180912203059.GA18201@roeck-us.net> From: Jae Hyun Yoo Message-ID: <6a99dc09-b06d-88ee-2843-3bd88f883dc2@linux.intel.com> Date: Wed, 12 Sep 2018 15:31:06 -0700 User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:52.0) Gecko/20100101 Thunderbird/52.9.1 MIME-Version: 1.0 In-Reply-To: <20180912203059.GA18201@roeck-us.net> Content-Type: text/plain; charset=utf-8; format=flowed Content-Language: en-US Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 9/12/2018 1:30 PM, Guenter Roeck wrote: > On Wed, Sep 12, 2018 at 01:10:45PM -0700, Jae Hyun Yoo wrote: >> On 9/12/2018 12:58 PM, Guenter Roeck wrote: >>> On Wed, Sep 12, 2018 at 09:54:51AM -0700, Jae Hyun Yoo wrote: >>>> On 9/11/2018 6:34 PM, Guenter Roeck wrote: >>>>> On Tue, Sep 11, 2018 at 04:58:44PM -0700, Jae Hyun Yoo wrote: >>>>>> On 9/11/2018 4:33 PM, Guenter Roeck wrote: >>>>>>> Looking into the patch, clearing the interrupt status at the end of an >>>>>>> interrupt handler is always suspicious and tends to result in race >>>>>>> conditions (because additional interrupts may have arrived while handling >>>>>>> the existing interrupts, or because interrupt handling itself may trigger >>>>>>> another interrupt). With that in mind, the following patch fixes the >>>>>>> problem for me. >>>>>>> >>>>>>> Guenter >>>>>>> >>>>>>> --- >>>>>>> >>>>>>> diff --git a/drivers/i2c/busses/i2c-aspeed.c b/drivers/i2c/busses/i2c-aspeed.c >>>>>>> index c258c4d9a4c0..c488e6950b7c 100644 >>>>>>> --- a/drivers/i2c/busses/i2c-aspeed.c >>>>>>> +++ b/drivers/i2c/busses/i2c-aspeed.c >>>>>>> @@ -552,6 +552,8 @@ static irqreturn_t aspeed_i2c_bus_irq(int irq, void *dev_id) >>>>>>> spin_lock(&bus->lock); >>>>>>> irq_received = readl(bus->base + ASPEED_I2C_INTR_STS_REG); >>>>>>> + /* Ack all interrupt bits. */ >>>>>>> + writel(irq_received, bus->base + ASPEED_I2C_INTR_STS_REG); >>>>>>> irq_remaining = irq_received; >>>>>>> #if IS_ENABLED(CONFIG_I2C_SLAVE) >>>>>>> @@ -584,8 +586,6 @@ static irqreturn_t aspeed_i2c_bus_irq(int irq, void *dev_id) >>>>>>> "irq handled != irq. expected 0x%08x, but was 0x%08x\n", >>>>>>> irq_received, irq_handled); >>>>>>> - /* Ack all interrupt bits. */ >>>>>>> - writel(irq_received, bus->base + ASPEED_I2C_INTR_STS_REG); >>>>>>> spin_unlock(&bus->lock); >>>>>>> return irq_remaining ? IRQ_NONE : IRQ_HANDLED; >>>>>>> } >>>>>>> >>>>>> >>>>>> My intention of putting the code at the end of interrupt handler was, >>>>>> to reduce possibility of combined irq calls which is explained in this >>>>>> patch. But YES, I agree with you. It could make a potential race >>>>> >>>>> Hmm, yes, but that doesn't explain why it would make sense to acknowledge >>>>> the interrupt late. The interrupt ack only means "I am going to handle these >>>>> interrupts". If additional interrupts arrive while the interrupt handler >>>>> is active, those will have to be acknowledged separately. >>>>> >>>>> Sure, there is a risk that an interrupt arrives while the handler is >>>>> running, and that it is handled but not acknowledged. That can happen >>>>> with pretty much all interrupt handlers, and there are mitigations to >>>>> limit the impact (for example, read the interrupt status register in >>>>> a loop until no more interrupts are pending). But acknowledging >>>>> an interrupt that was possibly not handled is always bad idea. >>>> >>>> Well, that's generally right but not always. Sometimes that depends on >>>> hardware and Aspeed I2C is the case. >>>> >>>> This is a description from Aspeed AST2500 datasheet: >>>> I2CD10 Interrupt Status Register >>>> bit 2 Receive Done Interrupt status >>>> S/W needs to clear this status bit to allow next data receiving. >>>> >>>> It means, driver should hold this bit to prevent transition of hardware >>>> state machine until the driver handles received data, so the bit should >>>> be cleared at the end of interrupt handler. >>>> >>> That makes sense. Does that apply to the other status bits as well ? >>> Reason for asking is that the current code actually gets stuck >>> in transmit, not receive. >>> >> Only bit 2 has that description in datasheet. Is slave config enabled >> for QEMU build? Does that get stuck in master sending or slave >> receiving? >> > qemu does not support slave mode. Linux gets stuck in master tx. > > I played with the code on both sides. I had to make changes in both > the linux kernel and in qemu to get the code to work again. > See attached. > > Guenter > > --- > Linux: > > diff --git a/drivers/i2c/busses/i2c-aspeed.c b/drivers/i2c/busses/i2c-aspeed.c > index c258c4d9a4c0..3d518e09369f 100644 > --- a/drivers/i2c/busses/i2c-aspeed.c > +++ b/drivers/i2c/busses/i2c-aspeed.c > @@ -552,6 +552,9 @@ static irqreturn_t aspeed_i2c_bus_irq(int irq, void *dev_id) > > spin_lock(&bus->lock); > irq_received = readl(bus->base + ASPEED_I2C_INTR_STS_REG); > + /* Ack all interrupts except for Rx done */ > + writel(irq_received & ~ASPEED_I2CD_INTR_RX_DONE, > + bus->base + ASPEED_I2C_INTR_STS_REG); > irq_remaining = irq_received; > > #if IS_ENABLED(CONFIG_I2C_SLAVE) > @@ -584,8 +587,10 @@ static irqreturn_t aspeed_i2c_bus_irq(int irq, void *dev_id) > "irq handled != irq. expected 0x%08x, but was 0x%08x\n", > irq_received, irq_handled); > > - /* Ack all interrupt bits. */ > - writel(irq_received, bus->base + ASPEED_I2C_INTR_STS_REG); > + /* Ack Rx done */ > + if (irq_received & ASPEED_I2CD_INTR_RX_DONE) > + writel(ASPEED_I2CD_INTR_RX_DONE, > + bus->base + ASPEED_I2C_INTR_STS_REG); > spin_unlock(&bus->lock); > return irq_remaining ? IRQ_NONE : IRQ_HANDLED; > } > > --- > qemu: > > diff --git a/hw/i2c/aspeed_i2c.c b/hw/i2c/aspeed_i2c.c > index c762c73..0d4aa08 100644 > --- a/hw/i2c/aspeed_i2c.c > +++ b/hw/i2c/aspeed_i2c.c > @@ -180,6 +180,33 @@ static uint8_t aspeed_i2c_get_state(AspeedI2CBus *bus) > return (bus->cmd >> I2CD_TX_STATE_SHIFT) & I2CD_TX_STATE_MASK; > } > > +static void aspeed_i2c_handle_rx_cmd(AspeedI2CBus *bus) > +{ > + int ret; > + > + if (!(bus->cmd & (I2CD_M_RX_CMD | I2CD_M_S_RX_CMD_LAST))) { > + return; > + } > + if (bus->intr_status & I2CD_INTR_RX_DONE) { > + return; > + } > + > + aspeed_i2c_set_state(bus, I2CD_MRXD); > + ret = i2c_recv(bus->bus); > + if (ret < 0) { > + qemu_log_mask(LOG_GUEST_ERROR, "%s: read failed\n", __func__); > + ret = 0xff; > + } else { > + bus->intr_status |= I2CD_INTR_RX_DONE; > + } > + bus->buf = (ret & I2CD_BYTE_BUF_RX_MASK) << I2CD_BYTE_BUF_RX_SHIFT; > + if (bus->cmd & I2CD_M_S_RX_CMD_LAST) { > + i2c_nack(bus->bus); > + } > + bus->cmd &= ~(I2CD_M_RX_CMD | I2CD_M_S_RX_CMD_LAST); > + aspeed_i2c_set_state(bus, I2CD_MACTIVE); > +} > + > /* > * The state machine needs some refinement. It is only used to track > * invalid STOP commands for the moment. > @@ -188,7 +215,7 @@ static void aspeed_i2c_bus_handle_cmd(AspeedI2CBus *bus, uint64_t value) > { > bus->cmd &= ~0xFFFF; > bus->cmd |= value & 0xFFFF; > - bus->intr_status = 0; > + bus->intr_status &= I2CD_INTR_RX_DONE; > > if (bus->cmd & I2CD_M_START_CMD) { > uint8_t state = aspeed_i2c_get_state(bus) & I2CD_MACTIVE ? > @@ -227,22 +254,7 @@ static void aspeed_i2c_bus_handle_cmd(AspeedI2CBus *bus, uint64_t value) > } > > if (bus->cmd & (I2CD_M_RX_CMD | I2CD_M_S_RX_CMD_LAST)) { > - int ret; > - > - aspeed_i2c_set_state(bus, I2CD_MRXD); > - ret = i2c_recv(bus->bus); > - if (ret < 0) { > - qemu_log_mask(LOG_GUEST_ERROR, "%s: read failed\n", __func__); > - ret = 0xff; > - } else { > - bus->intr_status |= I2CD_INTR_RX_DONE; > - } > - bus->buf = (ret & I2CD_BYTE_BUF_RX_MASK) << I2CD_BYTE_BUF_RX_SHIFT; > - if (bus->cmd & I2CD_M_S_RX_CMD_LAST) { > - i2c_nack(bus->bus); > - } > - bus->cmd &= ~(I2CD_M_RX_CMD | I2CD_M_S_RX_CMD_LAST); > - aspeed_i2c_set_state(bus, I2CD_MACTIVE); > + aspeed_i2c_handle_rx_cmd(bus); > } > > if (bus->cmd & I2CD_M_STOP_CMD) { > @@ -263,6 +275,7 @@ static void aspeed_i2c_bus_write(void *opaque, hwaddr offset, > uint64_t value, unsigned size) > { > AspeedI2CBus *bus = opaque; > + int status; > > switch (offset) { > case I2CD_FUN_CTRL_REG: > @@ -283,9 +296,16 @@ static void aspeed_i2c_bus_write(void *opaque, hwaddr offset, > bus->intr_ctrl = value & 0x7FFF; > break; > case I2CD_INTR_STS_REG: > + status = bus->intr_status; > bus->intr_status &= ~(value & 0x7FFF); > - bus->controller->intr_status &= ~(1 << bus->id); > - qemu_irq_lower(bus->controller->irq); > + if (!bus->intr_status) { > + bus->controller->intr_status &= ~(1 << bus->id); > + qemu_irq_lower(bus->controller->irq); > + } > + if ((status & I2CD_INTR_RX_DONE) && !(bus->intr_status & I2CD_INTR_RX_DONE)) { > + aspeed_i2c_handle_rx_cmd(bus); > + aspeed_i2c_bus_raise_interrupt(bus); > + } > break; > case I2CD_DEV_ADDR_REG: > qemu_log_mask(LOG_UNIMP, "%s: slave mode not implemented\n", > Nice fix! LGTM. I've tested the new patch and checked that it works well on both low and high bus speed environments. Thanks a lot! Can you please submit this patch? Thanks, Jae