linux-kernel - Re: [PATCH v3] staging: gpib: change return type of t1

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] [day] [month] [year] [list]

Message-ID: <2025032004-nerd-germless-a1ec@gregkh>
Date: Thu, 20 Mar 2025 07:08:57 -0700
From: Greg KH <gregkh@...uxfoundation.org>
To: Rodrigo Gobbi <rodrigo.gobbi.7@...il.com>
Cc: dan.carpenter@...aro.org, dpenkler@...il.com,
	~lkcamp/patches@...ts.sr.ht, linux-staging@...ts.linux.dev,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH v3] staging: gpib: change return type of t1_delay
 function to report errors

On Mon, Feb 24, 2025 at 10:44:05PM -0300, Rodrigo Gobbi wrote:
> The current code returns "unsigned int" and it doesn't handle errors
> correctly if it happens during ioctl call for t1 delay configuration.
> 
> The ni_usb_t1_delay(), from NI, is the only function returning -1
> at this point. The caller, t1_delay_ioctl(), doesn't check for errors
> and sets board->t1_nano_sec to -1 and returns success.
> The board->t1_nano_sec value is also used in ni_usb_setup_t1_delay()
> besides the ioctl call and a value of -1 is treated as being above 1100ns.
> It may or may not have a noticeable effect, but it's obviously not right
> considering the content of ni_usb_setup_t1_delay().
> 
> Typical delays are in the 200-2000 range, but definitely not more
> than INT_MAX so we can fix this code by changing the return type to int
> and adding a check for errors. While we're at it, lets change the error
> code in ni_usb_t1_delay() from -1 and instead propagate the error from
> ni_usb_write_registers().
> 
> Fixes: 4e127de14fa7 ("staging: gpib: Add National Instruments USB GPIB driver")
> Signed-off-by: Rodrigo Gobbi <rodrigo.gobbi.7@...il.com>

As other recently submitted patches caused this not to apply, I fixed it
up by hand and got it added to my tree so you wouldn't have to rebase it
again.

thanks!

greg k-h