DPDK patches and discussions
 help / color / mirror / Atom feed
From: Ferruh Yigit <ferruh.yigit@intel.com>
To: "Zhao1, Wei" <wei.zhao1@intel.com>,
	"Wu, Jingjing" <jingjing.wu@intel.com>,
	"Lu, Wenzhuo" <wenzhuo.lu@intel.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>, "stable@dpdk.org" <stable@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH] net/avf/base: fix shifting 32 bits signed variable 31 times
Date: Mon, 5 Nov 2018 14:30:47 +0000	[thread overview]
Message-ID: <07fea5ad-c01f-763a-c4b7-0f4948fb421c@intel.com> (raw)
In-Reply-To: <A2573D2ACFCADC41BB3BE09C6DE313CA07E67DC5@PGSMSX103.gar.corp.intel.com>

On 10/31/2018 3:34 AM, Zhao1, Wei wrote:
> 
> 
>> -----Original Message-----
>> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Ferruh Yigit
>> Sent: Sunday, October 28, 2018 11:52 AM
>> To: Wu, Jingjing <jingjing.wu@intel.com>; Lu, Wenzhuo
>> <wenzhuo.lu@intel.com>
>> Cc: dev@dpdk.org; Yigit, Ferruh <ferruh.yigit@intel.com>; stable@dpdk.org
>> Subject: [dpdk-dev] [PATCH] net/avf/base: fix shifting 32 bits signed variable
>> 31 times
>>
>> Fixes: e5b2a9e957e7 ("net/avf/base: add base code for avf PMD")
>> Cc: stable@dpdk.org
>>
>> Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
>> ---
>>  drivers/net/avf/base/avf_register.h | 4 ++--
>>  1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/net/avf/base/avf_register.h
>> b/drivers/net/avf/base/avf_register.h
>> index ba5a9f3fa..adb989583 100644
>> --- a/drivers/net/avf/base/avf_register.h
>> +++ b/drivers/net/avf/base/avf_register.h
>> @@ -76,7 +76,7 @@ POSSIBILITY OF SUCH DAMAGE.
>>  #define AVF_ARQLEN1_ARQCRIT_SHIFT   30
>>  #define AVF_ARQLEN1_ARQCRIT_MASK    AVF_MASK(0x1,
>> AVF_ARQLEN1_ARQCRIT_SHIFT)
>>  #define AVF_ARQLEN1_ARQENABLE_SHIFT 31
>> -#define AVF_ARQLEN1_ARQENABLE_MASK  AVF_MASK(0x1,
>> AVF_ARQLEN1_ARQENABLE_SHIFT)
>> +#define AVF_ARQLEN1_ARQENABLE_MASK  AVF_MASK(0x1U,
>> +AVF_ARQLEN1_ARQENABLE_SHIFT)
>>  #define AVF_ARQT1            0x00007000 /* Reset: EMPR */
>>  #define AVF_ARQT1_ARQT_SHIFT 0
>>  #define AVF_ARQT1_ARQT_MASK  AVF_MASK(0x3FF,
>> AVF_ARQT1_ARQT_SHIFT) @@ -99,7 +99,7 @@ POSSIBILITY OF SUCH
>> DAMAGE.
>>  #define AVF_ATQLEN1_ATQCRIT_SHIFT   30
>>  #define AVF_ATQLEN1_ATQCRIT_MASK    AVF_MASK(0x1,
>> AVF_ATQLEN1_ATQCRIT_SHIFT)
>>  #define AVF_ATQLEN1_ATQENABLE_SHIFT 31
>> -#define AVF_ATQLEN1_ATQENABLE_MASK  AVF_MASK(0x1,
>> AVF_ATQLEN1_ATQENABLE_SHIFT)
>> +#define AVF_ATQLEN1_ATQENABLE_MASK  AVF_MASK(0x1U,
>> +AVF_ATQLEN1_ATQENABLE_SHIFT)
>>  #define AVF_ATQT1            0x00008400 /* Reset: EMPR */
>>  #define AVF_ATQT1_ATQT_SHIFT 0
>>  #define AVF_ATQT1_ATQT_MASK  AVF_MASK(0x3FF,
>> AVF_ATQT1_ATQT_SHIFT)
>> --
>> 2.17.2
> 
> I have test this code change with a small function to observe the difference.
> Reviewed-by: Wei Zhao <wei.zhao1@intel.com>
> 
> #define AVF_ARQLEN1_ARQENABLE_SHIFT 31
> #define AVF_MASK(mask, shift) (mask << shift)
> #define AVF_ARQLEN1_ARQENABLE_MASK1  AVF_MASK(0x1, AVF_ARQLEN1_ARQENABLE_SHIFT)
> #define AVF_ARQLEN1_ARQENABLE_MASK2  AVF_MASK(0x1U, AVF_ARQLEN1_ARQENABLE_SHIFT)
> 
> void main(int argc,char **argv)
> {
>         signed int a=1;
> 
>         printf("Hello Linux\n");
>         printf("%x,=%x\n",AVF_ARQLEN1_ARQENABLE_MASK1,AVF_ARQLEN1_ARQENABLE_MASK2);
>         printf("a=%x\n",a<<31);
> 
> }

I guess you are tying to say there is no difference, output of your code is:
Hello Linux
80000000,=80000000
a=80000000


Let's have a little addition to your program,
 #include <stdio.h>

 #define AVF_ARQLEN1_ARQENABLE_SHIFT 31
 #define AVF_MASK(mask, shift) (mask << shift)
 #define AVF_ARQLEN1_ARQENABLE_MASK1  AVF_MASK(0x1, AVF_ARQLEN1_ARQENABLE_SHIFT)
 #define AVF_ARQLEN1_ARQENABLE_MASK2  AVF_MASK(0x1U, AVF_ARQLEN1_ARQENABLE_SHIFT)
 #define AVF_MASK2(mask, shift) (mask >> shift)
 #define AVF_ARQLEN1_ARQENABLE_MASK11  AVF_MASK2(AVF_ARQLEN1_ARQENABLE_MASK1, 30)
 #define AVF_ARQLEN1_ARQENABLE_MASK22  AVF_MASK2(AVF_ARQLEN1_ARQENABLE_MASK2, 30)

 void main(int argc,char **argv)
 {
         signed int a=1;

         printf("Hello Linux\n");
         printf("%x,=%x\n", AVF_ARQLEN1_ARQENABLE_MASK1,
AVF_ARQLEN1_ARQENABLE_MASK2);
         printf("%x,=%x\n", AVF_ARQLEN1_ARQENABLE_MASK11,
AVF_ARQLEN1_ARQENABLE_MASK22);
         printf("a=%x\n", a<<31);

 }


Will results same for second line? Both does ((1 << 31) >> 30).


'1' is promoted to signed int by default and cppcheck complains that shifting
signed variable 31 times is undefined behaviour, and here intention is not
really have the 1 as signed integer, so better to fix it.

Thanks,
ferruh

  reply	other threads:[~2018-11-05 14:30 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-10-28  3:51 Ferruh Yigit
2018-10-31  3:34 ` Zhao1, Wei
2018-11-05 14:30   ` Ferruh Yigit [this message]
2018-11-07  2:42     ` Zhao1, Wei
2018-11-05 15:03   ` Ferruh Yigit

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=07fea5ad-c01f-763a-c4b7-0f4948fb421c@intel.com \
    --to=ferruh.yigit@intel.com \
    --cc=dev@dpdk.org \
    --cc=jingjing.wu@intel.com \
    --cc=stable@dpdk.org \
    --cc=wei.zhao1@intel.com \
    --cc=wenzhuo.lu@intel.com \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).