linux-kernel - Re: [PATCH v2 1/2] Drivers: hv: util: Don't force error code to ENODEV in util

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 for Android: free password hash cracker in your pocket

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

Message-ID: <20241208173049.GA14100@linuxonhyperv3.guj3yctzbm1etfxqx2vob5hsef.xx.internal.cloudapp.net>
Date: Sun, 8 Dec 2024 09:30:49 -0800
From: Saurabh Singh Sengar <ssengar@...ux.microsoft.com>
To: mhklinux@...look.com
Cc: kys@...rosoft.com, haiyangz@...rosoft.com, wei.liu@...nel.org,
	decui@...rosoft.com, gregkh@...uxfoundation.org,
	vkuznets@...hat.com, linux-hyperv@...r.kernel.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 1/2] Drivers: hv: util: Don't force error code to
 ENODEV in util_probe()

On Wed, Nov 06, 2024 at 07:42:46AM -0800, mhkelley58@...il.com wrote:
> From: Michael Kelley <mhklinux@...look.com>
> 
> If the util_init function call in util_probe() returns an error code,
> util_probe() always return ENODEV, and the error code from the util_init
> function is lost. The error message output in the caller, vmbus_probe(),
> doesn't show the real error code.
> 
> Fix this by just returning the error code from the util_init function.
> There doesn't seem to be a reason to force ENODEV, as other errors
> such as ENOMEM can already be returned from util_probe(). And the
> code in call_driver_probe() implies that ENODEV should mean that a
> matching driver wasn't found, which is not the case here.
> 
> Suggested-by: Dexuan Cui <decui@...rosoft.com>
> Signed-off-by: Michael Kelley <mhklinux@...look.com>
> ---
> Changes in v2: None. This is the first version of Patch 1 of this series.
> The "v2" is due to changes to Patch 2 of the series.
> 
>  drivers/hv/hv_util.c | 4 +---
>  1 file changed, 1 insertion(+), 3 deletions(-)
> 
> diff --git a/drivers/hv/hv_util.c b/drivers/hv/hv_util.c
> index c4f525325790..370722220134 100644
> --- a/drivers/hv/hv_util.c
> +++ b/drivers/hv/hv_util.c
> @@ -590,10 +590,8 @@ static int util_probe(struct hv_device *dev,
>  	srv->channel = dev->channel;
>  	if (srv->util_init) {
>  		ret = srv->util_init(srv);
> -		if (ret) {
> -			ret = -ENODEV;
> +		if (ret)
>  			goto error1;
> -		}

After reviewing V2 of this series, I couldn’t find any scenario where
'util_init' in any driver returns a value other than 0. In such cases,
could we consider making all these functions 'void' ?

After this ee can remove the check for util_int return type.

- Saurabh