linux-kernel - RE: [PATCH v4 char-misc-next] misc: microchip: pci1xxxx: Add OTP/EEPROM driver for the pci1xxxx switch

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]

Message-ID: <PH7PR11MB59588BF687D7D2EDAC36E4899BDF9@PH7PR11MB5958.namprd11.prod.outlook.com>
Date:   Sat, 11 Feb 2023 05:07:43 +0000
From:   <Tharunkumar.Pasumarthi@...rochip.com>
To:     <gregkh@...uxfoundation.org>
CC:     <arnd@...db.de>, <linux-kernel@...r.kernel.org>,
        <linux-gpio@...r.kernel.org>, <UNGLinuxDriver@...rochip.com>
Subject: RE: [PATCH v4 char-misc-next] misc: microchip: pci1xxxx: Add
 OTP/EEPROM driver for the pci1xxxx switch

> From: Greg KH <gregkh@...uxfoundation.org>
> Sent: Thursday, February 9, 2023 2:25 PM
> To: Tharunkumar Pasumarthi - I67821
> <Tharunkumar.Pasumarthi@...rochip.com>
> EXTERNAL EMAIL: Do not click links or open attachments unless you know the
> content is safe
> 
> > +     } while (data & EEPROM_CMD_EPC_BUSY_BIT);
> 
> That's a very busy "sit and spin" loop here, what happens if the read of the
> bit never actually succeeds?  You just locked up the system with no way to
> interrupt it :(
> 
> Please provide some sort of timeout, or way to break out of this.
> 
> > +
> > +     if (data & EEPROM_CMD_EPC_TIMEOUT_BIT) {
> > +             dev_err(&priv->pdev->dev, "EEPROM write timed out\n");
> 
> How can the timeout bit happen if the busy bit was still set?
> 
> And what can userspace do about this if it is reported?

Hi Greg,
If EEPROM_CMD_EPC_BUSY_BIT is set for more than 30ms, it will be cleared automatically by the hardware logic and EEPROM_CMD_EPC_TIMEOUT_BIT bit will be set to indicate the timeout. User space application will inform user about timeout on EEPROM write/read when this error occurs.

> 
> > +             return -EFAULT;
> 
> This return value is ONLY for when we have memory faults from reading
> to/from userspace and the kernel.  It's not a valid return value for a device
> error, sorry.  -EIO maybe?
> 
> You return this error in a number of other places in the driver that shouldn't,
> please fix this up.

Okay.

Thanks,
Tharun Kumar P