linux-kernel - Re: [PATCH v2 2/2] spi: Add ADI driver for Spreadtrum platform

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 14 Sep 2017 09:50:55 +0800
From:   Baolin Wang <baolin.wang@...aro.org>
To:     Rob Herring <robh@...nel.org>
Cc:     Baolin Wang <baolin.wang@...eadtrum.com>,
        Mark Brown <broonie@...nel.org>,
        Mark Rutland <mark.rutland@....com>, linux-spi@...r.kernel.org,
        devicetree@...r.kernel.org, LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2 2/2] spi: Add ADI driver for Spreadtrum platform

Hi Rob,

On 14 September 2017 at 03:45, Rob Herring <robh@...nel.org> wrote:
> On Fri, Sep 08, 2017 at 04:33:42PM +0800, Baolin Wang wrote:
>> This patch adds ADI driver based on SPI framework for
>> Spreadtrum SC9860 platform.
>>
>> Signed-off-by: Baolin Wang <baolin.wang@...eadtrum.com>
>> ---
>
> [...]
>
>> +++ b/drivers/spi/spi-sprd-adi.c
>> @@ -0,0 +1,419 @@
>> +/*
>> + * Copyright (C) 2017 Spreadtrum Communications Inc.
>> + *
>> + * SPDX-License-Identifier: (GPL-2.0+ OR MIT)
>
> Kernel code should generally not be MIT license.

OK. Will fix in next version.

>
> [...]
>
>> +static int sprd_adi_drain_fifo(struct sprd_adi *sadi)
>> +{
>> +     u32 timeout = ADI_FIFO_DRAIN_TIMEOUT;
>> +     u32 sts;
>> +
>> +     do {
>> +             sts = readl_relaxed(sadi->base + REG_ADI_ARM_FIFO_STS);
>> +             if (sts & BIT_FIFO_EMPTY)
>> +                     break;
>> +
>> +             cpu_relax();
>> +     } while (--timeout);
>
> You can use readl_poll_timeout.

The readl_poll_timeout() function might be sleep, but we can not sleep
when reading/writing data through ADI controller, since the routine is
under hardware spinlock protection. Moreover it is usually very quick
to jump the loop and we no need to sleep here.

>
>> +
>> +     if (timeout == 0) {
>> +             dev_err(sadi->dev, "drain write fifo timeout\n");
>> +             return -EBUSY;
>> +     }
>> +
>> +     return 0;
>> +}
>> +
>> +static int sprd_adi_fifo_is_full(struct sprd_adi *sadi)
>> +{
>> +     return readl_relaxed(sadi->base + REG_ADI_ARM_FIFO_STS) & BIT_FIFO_FULL;
>> +}
>> +
>> +static int sprd_adi_read(struct sprd_adi *sadi, u32 reg_paddr, u32 *read_val)
>> +{
>> +     int read_timeout = ADI_READ_TIMEOUT;
>> +     u32 val, rd_addr;
>> +
>> +     /*
>> +      * Set the physical register address need to read into RD_CMD register,
>> +      * then ADI controller will start to transfer automatically.
>> +      */
>> +     writel_relaxed(reg_paddr, sadi->base + REG_ADI_RD_CMD);
>> +
>> +     /*
>> +      * Wait read operation complete, the BIT_RD_CMD_BUSY will be set
>> +      * simultaneously when writing read command to register, and the
>> +      * BIT_RD_CMD_BUSY will be cleared after the read operation is
>> +      * completed.
>> +      */
>> +     do {
>> +             val = readl_relaxed(sadi->base + REG_ADI_RD_DATA);
>> +             if (!(val & BIT_RD_CMD_BUSY))
>> +                     break;
>> +
>> +             cpu_relax();
>> +     } while (--read_timeout);
>
> readl_poll_timeout

The same reason as above. Thanks for your comments.

-- 
Baolin.wang
Best Regards