From: Dan Carpenter <dan.carpenter@oracle.com>
To: Isaac Lleida <isakyllr@openmailbox.org>
Cc: willy@meta-x.org, devel@driverdev.osuosl.org,
isaky <illeida@openaliasbox.org>,
armand.bastien@laposte.net, gregkh@linuxfoundation.org,
linux-kernel@vger.kernel.org, marius.gorski@gmail.com,
domdevlin@free.fr, sudipm.mukherjee@gmail.com
Subject: Re: [PATCH] staging: panel: change struct bits to a bit array
Date: Mon, 16 Mar 2015 12:08:49 +0300 [thread overview]
Message-ID: <20150316090848.GH10964@mwanda> (raw)
In-Reply-To: <1426264923-12392-1-git-send-email-isakyllr@openmailbox.org>
On Fri, Mar 13, 2015 at 05:42:03PM +0100, Isaac Lleida wrote:
> From: isaky <illeida@openaliasbox.org>
Not your name. Leave this out unless you are passing on someone else's
patch.
>
> This path implements a bit array representing the LCD signal states instead of the old "struct bits", which used char to represent a single bit. This will reduce the memory usage.
>
Line wraps the changelog.
> Signed-off-by: Isaac Lleida <illeida@openaliasbox.org>
This should match the email address you used to send the patch.
> ---
> drivers/staging/panel/panel.c | 86 ++++++++++++++++++++++++-------------------
> 1 file changed, 49 insertions(+), 37 deletions(-)
>
> diff --git a/drivers/staging/panel/panel.c b/drivers/staging/panel/panel.c
> index 3339633..7deb092 100644
> --- a/drivers/staging/panel/panel.c
> +++ b/drivers/staging/panel/panel.c
> @@ -310,6 +310,25 @@ static int selected_lcd_type = NOT_SET;
> #define LCD_BITS 6
>
> /*
> + * LCD signal states
> + */
> +#define LCD_BIT_E_MASK 0x1 /* E (data latch on falling edge) */
> +#define LCD_BIT_RS_MASK 0x2 /* RS (0 = cmd, 1 = data) */
> +#define LCD_BIT_RW_MASK 0x4 /* R/W (0 = W, 1 = R) */
> +#define LCD_BIT_BL_MASK 0x8 /* backlight (0 = off, 1 = on) */
> +#define LCD_BIT_CL_MASK 0x10 /* clock (latch on rising edge) */
> +#define LCD_BIT_DA_MASK 0x20 /* data */
> +
> +/*
> + * Bit array operations
> + */
> +#define BIT_ON(b, m) (b |= m)
> +#define BIT_OFF(b, m) (b &= ~m)
> +#define BIT_CHK(b, m) (b & m)
> +#define BIT_MOD(b, m, v) \
> + ((v) ? BIT_ON(b, m) : BIT_OFF(b, m)) \
> +
Ugh. No. These are yuck macros. Just do it directly. Also the
BIT_MOD() has end in a '\' char.
> +/*
> * each bit can be either connected to a DATA or CTRL port
> */
> #define LCD_PORT_C 0
> @@ -653,15 +672,8 @@ static const char nexcom_keypad_profile[][4][9] = {
>
> static const char (*keypad_profile)[4][9] = old_keypad_profile;
>
> -/* FIXME: this should be converted to a bit array containing signals states */
> -static struct {
> - unsigned char e; /* parallel LCD E (data latch on falling edge) */
> - unsigned char rs; /* parallel LCD RS (0 = cmd, 1 = data) */
> - unsigned char rw; /* parallel LCD R/W (0 = W, 1 = R) */
> - unsigned char bl; /* parallel LCD backlight (0 = off, 1 = on) */
> - unsigned char cl; /* serial LCD clock (latch on rising edge) */
> - unsigned char da; /* serial LCD data */
> -} bits;
> +/* Bit array containing signals states */
> +static char bits;
I don't think this is what the comment means. I think it means that
we do something like:
static struct {
unsigned long e:1;
unsigned long rs:1;
...
} bits;
1) This space saving are very small. On 64 bit systems then we don't
save any space at all because everything is aligned at 8 bytes.
2) The name "e" is not self documenting. None of them are.
Christophe had a bunch more comments on this patch so I'm not going to
read any further.
regards,
dan carpenter
next prev parent reply other threads:[~2015-03-16 9:09 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-03-13 16:42 [PATCH] staging: panel: change struct bits to a bit array Isaac Lleida
2015-03-16 9:08 ` Dan Carpenter [this message]
-- strict thread matches above, loose matches on Subject: below --
2015-03-13 16:48 Isaac Lleida
2015-03-13 18:20 ` Christophe JAILLET
2015-03-13 20:10 ` Isaac Lleida
2015-03-13 21:44 ` Marion & Christophe JAILLET
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20150316090848.GH10964@mwanda \
--to=dan.carpenter@oracle.com \
--cc=armand.bastien@laposte.net \
--cc=devel@driverdev.osuosl.org \
--cc=domdevlin@free.fr \
--cc=gregkh@linuxfoundation.org \
--cc=illeida@openaliasbox.org \
--cc=isakyllr@openmailbox.org \
--cc=linux-kernel@vger.kernel.org \
--cc=marius.gorski@gmail.com \
--cc=sudipm.mukherjee@gmail.com \
--cc=willy@meta-x.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.