From: Jes Sorensen <Jes.Sorensen@redhat.com>
To: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Joe Perches <joe@perches.com>, Kalle Valo <kvalo@codeaurora.org>,
	linux-wireless@vger.kernel.org,
	Greg KH <gregkh@linuxfoundation.org>
Subject: Re: [PATCH v1 1/1] rtl8xxxu: tuse %*ph to dump buffers
References: <1464274339-103386-1-git-send-email-andriy.shevchenko@linux.intel.com>
	<1464278521.16344.35.camel@perches.com>
	<1464281830.27624.27.camel@linux.intel.com>
Date: Thu, 26 May 2016 16:16:49 -0400
In-Reply-To: <1464281830.27624.27.camel@linux.intel.com> (Andy Shevchenko's
	message of "Thu, 26 May 2016 19:57:10 +0300")
Message-ID: <wrfjlh2wbmmm.fsf@redhat.com> (sfid-20160526_221655_526537_B64C5CFB)
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Sender: linux-wireless-owner@vger.kernel.org

Andy Shevchenko <andriy.shevchenko@linux.intel.com> writes:
> On Thu, 2016-05-26 at 09:02 -0700, Joe Perches wrote:
>> On Thu, 2016-05-26 at 17:52 +0300, Andy Shevchenko wrote:
>> > Use %*ph specifier to dump small buffers in hex format instead of
>> > doing this
>> > byte-by-byte.
>> []
>> > diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c
>> > b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c
>> []
>> > @@ -413,13 +413,8 @@ static int rtl8192cu_parse_efuse(struct
>> > rtl8xxxu_priv *priv)
>> >  		dev_info(&priv->udev->dev,
>> >  			 "%s: dumping efuse (0x%02zx bytes):\n",
>> >  			 __func__, sizeof(struct rtl8192cu_efuse));
>> > -		for (i = 0; i < sizeof(struct rtl8192cu_efuse); i
>> > += 8) {
>> > -			dev_info(&priv->udev->dev, "%02x: "
>> > -				 "%02x %02x %02x %02x %02x %02x
>> > %02x %02x\n", i,
>> > -				 raw[i], raw[i + 1], raw[i + 2],
>> > -				 raw[i + 3], raw[i + 4], raw[i +
>> > 5],
>> > -				 raw[i + 6], raw[i + 7]);
>> > -		}
>> > +		for (i = 0; i < sizeof(struct rtl8192cu_efuse); i
>> > += 8)
>> > +			dev_info(&priv->udev->dev, "%02x: %8ph\n",
>> > i, &raw[i]);
>> 
>> print_hex_dump_bytes?
>
> I know, perhaps I had to mention that it is not too beneficial here. 
>
>> or maybe add a dev_print_hex_dump variant?
>
> I don't know how much users we have to introduce a such. This is
> actually a main reason I avoid to use print_hex_dump() above. Btw, the
> above might be collapsed into one function instead of 3 times of copy-n-
> paste.

I think a dev_print_hex_dump() would be useful here. When I first read
your patch I was a bit hmmmm but then I realized why you had stuck with
the %8ph rather than print_hex_dump().

I'll apply it shortly.

Jes