[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <aW-REdQ9BAKA-GC9@stanley.mountain>
Date: Tue, 20 Jan 2026 17:28:33 +0300
From: Dan Carpenter <dan.carpenter@...aro.org>
To: Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
Cc: Minu Jin <s9430939@...er.com>, gregkh@...uxfoundation.org,
abrahamadekunle50@...il.com, milospuric856@...il.com,
zxcv2569763104@...il.com, linux-staging@...ts.linux.dev,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] staging: rtl8723bs: fix unchecked return value of
skb_copy_bits
On Tue, Jan 20, 2026 at 04:03:37PM +0200, Andy Shevchenko wrote:
> On Tue, Jan 20, 2026 at 04:59:29PM +0300, Dan Carpenter wrote:
> > On Tue, Jan 20, 2026 at 10:22:41PM +0900, Minu Jin wrote:
>
> ...
>
> > > + signed int ret;
> >
> > Don't put a blank line in the middle of the declaration block. Just do
> > "int ret;". Everyone knows "int" is signed. Don't follow the local
> > style when the local style is wrong. Also only use s32 when it's part
> > of a networking or hardware spec. Just use int. Do people imagine that
> > we'll change int to be unsigned?
>
> Not in a far past we changed char to be unsigned :-)
>
> *Yes I know that it's a bit different case.
Heh.
char was always unsigned on s390 and the s390 devs were really militant
about avoiding declaring variables as "unsigned char" so they'd write
all their arch/s390/ code to look like:
if (char_variable == 255) {
I guess they were annoyed at the rest of the world who declared their
variables as "char" when it should have been "signed char". They stuck
to their guns and defeated the rest of us in the end. It's some kind of
life lesson or potentially a message of hope? :P
regards,
dan carpenter
Powered by blists - more mailing lists