[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200225102809.GA2591@amd>
Date: Tue, 25 Feb 2020 11:28:09 +0100
From: Pavel Machek <pavel@...x.de>
To: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc: linux-kernel@...r.kernel.org, stable@...r.kernel.org,
Aditya Pakki <pakki001@....edu>,
Kalle Valo <kvalo@...eaurora.org>,
Sasha Levin <sashal@...nel.org>
Subject: Re: [PATCH 4.19 079/191] orinoco: avoid assertion in case of NULL
pointer
Hi!
> From: Aditya Pakki <pakki001@....edu>
>
> [ Upstream commit c705f9fc6a1736dcf6ec01f8206707c108dca824 ]
>
> In ezusb_init, if upriv is NULL, the code crashes. However, the caller
> in ezusb_probe can handle the error and print the failure message.
> The patch replaces the BUG_ON call to error return.
The caller already checked that upriv is not NULL, AFAICT.
priv = alloc_orinocodev(sizeof(*upriv), &udev->dev,
ezusb_hard_reset, NULL);
if (!priv) {
err("Couldn't allocate orinocodev");
retval = -ENOMEM;
goto exit;
}
I don't see this as an improvement.
Best regards,
Pavel
> +++ b/drivers/net/wireless/intersil/orinoco/orinoco_usb.c
> @@ -1364,7 +1364,8 @@ static int ezusb_init(struct hermes *hw)
> int retval;
>
> BUG_ON(in_interrupt());
> - BUG_ON(!upriv);
> + if (!upriv)
> + return -EINVAL;
>
> upriv->reply_count = 0;
> /* Write the MAGIC number on the simulated registers to keep
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
Download attachment "signature.asc" of type "application/pgp-signature" (182 bytes)
Powered by blists - more mailing lists