* Re: [PATCH] tty: gdm724x: use min_t() for size_t varable and a constant
2023-08-16 8:53 [PATCH] tty: gdm724x: use min_t() for size_t varable and a constant Jiri Slaby (SUSE)
@ 2023-08-16 11:45 ` Geert Uytterhoeven
2023-08-16 17:13 ` Nathan Chancellor
2023-08-17 11:37 ` David Laight
2 siblings, 0 replies; 4+ messages in thread
From: Geert Uytterhoeven @ 2023-08-16 11:45 UTC (permalink / raw)
To: Jiri Slaby (SUSE)
Cc: gregkh, linux-serial, linux-kernel, Nathan Chancellor,
kernel test robot
On Wed, 16 Aug 2023, Jiri Slaby (SUSE) wrote:
> My thinking was that ulong is the same as size_t everywhere. No, size_t
> is uint on 32bit. So the below commit introduced a build warning on
> 32bit:
> .../gdm724x/gdm_tty.c:165:24: warning: comparison of distinct pointer types ('typeof (2048UL) *' (aka 'unsigned long *') and 'typeof (remain) *' (aka 'unsigned int *'))
>
> To fix this, partially revert the commit (remove constants' suffixes)
> and switch to min_t() in this case instead.
>
> /me would hope for Z (or alike) suffix for constants.
>
> Signed-off-by: Jiri Slaby (SUSE) <jirislaby@kernel.org>
> Fixes: c3e5c706aefc (tty: gdm724x: convert counts to size_t)
> Reported-by: Nathan Chancellor <nathan@kernel.org>
> Reported-by: kernel test robot <lkp@intel.com>
> Closes: https://lore.kernel.org/oe-kbuild-all/202308151953.rNNnAR2N-lkp@intel.com/
Thanks, this fixes the m68k/allmodconfig build.
Reviewed-by: Geert Uytterhoeven <geert@linux-m68k.org>
> --- a/drivers/staging/gdm724x/gdm_tty.c
> +++ b/drivers/staging/gdm724x/gdm_tty.c
> @@ -17,9 +17,9 @@
> #define GDM_TTY_MAJOR 0
> #define GDM_TTY_MINOR 32
>
> -#define WRITE_SIZE 2048UL
> +#define WRITE_SIZE 2048
>
> -#define MUX_TX_MAX_SIZE 2048UL
> +#define MUX_TX_MAX_SIZE 2048
You probably want to keep the "U" suffix, so at least both parts
of min() are unsigned.
See also "[PATCH next v3 0/5] minmax: Relax type checks in min() and max().".
https://lore.kernel.org/all/01e3e09005e9434b8f558a893a47c053@AcuMS.aculab.com,
>
> static inline bool gdm_tty_ready(struct gdm *gdm)
> {
> @@ -159,7 +159,7 @@ static ssize_t gdm_tty_write(struct tty_struct *tty, const u8 *buf, size_t len)
> return -ENODEV;
>
> while (remain) {
> - size_t sending_len = min(MUX_TX_MAX_SIZE, remain);
> + size_t sending_len = min_t(size_t, MUX_TX_MAX_SIZE, remain);
> gdm->tty_dev->send_func(gdm->tty_dev->priv_dev,
> (void *)(buf + sent_len),
> sending_len,
> --
> 2.41.0
>
>
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] tty: gdm724x: use min_t() for size_t varable and a constant
2023-08-16 8:53 [PATCH] tty: gdm724x: use min_t() for size_t varable and a constant Jiri Slaby (SUSE)
2023-08-16 11:45 ` Geert Uytterhoeven
@ 2023-08-16 17:13 ` Nathan Chancellor
2023-08-17 11:37 ` David Laight
2 siblings, 0 replies; 4+ messages in thread
From: Nathan Chancellor @ 2023-08-16 17:13 UTC (permalink / raw)
To: Jiri Slaby (SUSE); +Cc: gregkh, linux-serial, linux-kernel, kernel test robot
On Wed, Aug 16, 2023 at 10:53:22AM +0200, Jiri Slaby (SUSE) wrote:
> My thinking was that ulong is the same as size_t everywhere. No, size_t
> is uint on 32bit. So the below commit introduced a build warning on
> 32bit:
> .../gdm724x/gdm_tty.c:165:24: warning: comparison of distinct pointer types ('typeof (2048UL) *' (aka 'unsigned long *') and 'typeof (remain) *' (aka 'unsigned int *'))
>
> To fix this, partially revert the commit (remove constants' suffixes)
> and switch to min_t() in this case instead.
>
> /me would hope for Z (or alike) suffix for constants.
>
> Signed-off-by: Jiri Slaby (SUSE) <jirislaby@kernel.org>
> Fixes: c3e5c706aefc (tty: gdm724x: convert counts to size_t)
> Reported-by: Nathan Chancellor <nathan@kernel.org>
> Reported-by: kernel test robot <lkp@intel.com>
> Closes: https://lore.kernel.org/oe-kbuild-all/202308151953.rNNnAR2N-lkp@intel.com/
Tested-by: Nathan Chancellor <nathan@kernel.org> # build
> ---
> drivers/staging/gdm724x/gdm_tty.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/staging/gdm724x/gdm_tty.c b/drivers/staging/gdm724x/gdm_tty.c
> index 67d9bf41e836..32b2e817ff04 100644
> --- a/drivers/staging/gdm724x/gdm_tty.c
> +++ b/drivers/staging/gdm724x/gdm_tty.c
> @@ -17,9 +17,9 @@
> #define GDM_TTY_MAJOR 0
> #define GDM_TTY_MINOR 32
>
> -#define WRITE_SIZE 2048UL
> +#define WRITE_SIZE 2048
>
> -#define MUX_TX_MAX_SIZE 2048UL
> +#define MUX_TX_MAX_SIZE 2048
>
> static inline bool gdm_tty_ready(struct gdm *gdm)
> {
> @@ -159,7 +159,7 @@ static ssize_t gdm_tty_write(struct tty_struct *tty, const u8 *buf, size_t len)
> return -ENODEV;
>
> while (remain) {
> - size_t sending_len = min(MUX_TX_MAX_SIZE, remain);
> + size_t sending_len = min_t(size_t, MUX_TX_MAX_SIZE, remain);
> gdm->tty_dev->send_func(gdm->tty_dev->priv_dev,
> (void *)(buf + sent_len),
> sending_len,
> --
> 2.41.0
>
^ permalink raw reply [flat|nested] 4+ messages in thread* RE: [PATCH] tty: gdm724x: use min_t() for size_t varable and a constant
2023-08-16 8:53 [PATCH] tty: gdm724x: use min_t() for size_t varable and a constant Jiri Slaby (SUSE)
2023-08-16 11:45 ` Geert Uytterhoeven
2023-08-16 17:13 ` Nathan Chancellor
@ 2023-08-17 11:37 ` David Laight
2 siblings, 0 replies; 4+ messages in thread
From: David Laight @ 2023-08-17 11:37 UTC (permalink / raw)
To: 'Jiri Slaby (SUSE)', gregkh@linuxfoundation.org
Cc: linux-serial@vger.kernel.org, linux-kernel@vger.kernel.org,
Nathan Chancellor, kernel test robot
From: Jiri Slaby (SUSE)
> Sent: Wednesday, August 16, 2023 9:53 AM
>
> My thinking was that ulong is the same as size_t everywhere. No, size_t
> is uint on 32bit. So the below commit introduced a build warning on
> 32bit:
> .../gdm724x/gdm_tty.c:165:24: warning: comparison of distinct pointer types ('typeof (2048UL) *' (aka
> 'unsigned long *') and 'typeof (remain) *' (aka 'unsigned int *'))
>
> To fix this, partially revert the commit (remove constants' suffixes)
> and switch to min_t() in this case instead.
>
....
> - size_t sending_len = min(MUX_TX_MAX_SIZE, remain);
> + size_t sending_len = min_t(size_t, MUX_TX_MAX_SIZE, remain);
It would be slightly safer to use:
min(remain, (size_t)MAX_TX_MAX_SIZE);
since that maintains the type check.
(It is also nicer to put the constant second.)
David
-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)
^ permalink raw reply [flat|nested] 4+ messages in thread