diff mbox series

net/cxgbe: fix build with clang 3.4.2

Message ID 20200318125236.652514-1-ferruh.yigit@intel.com (mailing list archive)
State Accepted, archived
Delegated to: Ferruh Yigit
Headers
Series net/cxgbe: fix build with clang 3.4.2 |

Checks

Context Check Description
ci/checkpatch success coding style OK
ci/Intel-compilation fail apply issues

Commit Message

Ferruh Yigit March 18, 2020, 12:52 p.m. UTC 
  Build error:
.../drivers/net/cxgbe/cxgbe_flow.c:315:2:
    error: implicit truncation from 'int' to bitfield changes value
    from -1 to 7 [-Werror,-Wbitfield-constant-conversion]
        CXGBE_FILL_FS(adap->pf, ~0, pf);
        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.../drivers/net/cxgbe/cxgbe_flow.c:25:2:
    note: expanded from macro 'CXGBE_FILL_FS'
        __CXGBE_FILL_FS(v, m, fs, elem, e)
        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.../drivers/net/cxgbe/cxgbe_flow.c:15:18: note:
    expanded from macro '__CXGBE_FILL_FS'
        (fs)->mask.elem = (__m); \
                        ^ ~~~~~
Commit in fixes line changes 'PF_BITWIDTH' to '3', which makes 'fs->pf'
bitfield size '3', changing '~0' to '0x7' to prevent truncation in
assignment.

Fixes: dcd456abe46f ("net/cxgbe: support flow API for matching all packets on PF")

Reported-by: Raslan Darawsheh <rasland@mellanox.com>
Reported-by: Xueming Zhang <xuemingx.zhang@intel.com>
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
---
Cc: kaara.satwik@chelsio.com
---
 drivers/net/cxgbe/cxgbe_flow.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Ferruh Yigit March 18, 2020, 1:09 p.m. UTC | #1 
On 3/18/2020 12:52 PM, Ferruh Yigit wrote:
> Build error:
> .../drivers/net/cxgbe/cxgbe_flow.c:315:2:
>     error: implicit truncation from 'int' to bitfield changes value
>     from -1 to 7 [-Werror,-Wbitfield-constant-conversion]
>         CXGBE_FILL_FS(adap->pf, ~0, pf);
>         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> .../drivers/net/cxgbe/cxgbe_flow.c:25:2:
>     note: expanded from macro 'CXGBE_FILL_FS'
>         __CXGBE_FILL_FS(v, m, fs, elem, e)
>         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> .../drivers/net/cxgbe/cxgbe_flow.c:15:18: note:
>     expanded from macro '__CXGBE_FILL_FS'
>         (fs)->mask.elem = (__m); \
>                         ^ ~~~~~
> Commit in fixes line changes 'PF_BITWIDTH' to '3', which makes 'fs->pf'
> bitfield size '3', changing '~0' to '0x7' to prevent truncation in
> assignment.
> 
> Fixes: dcd456abe46f ("net/cxgbe: support flow API for matching all packets on PF")
> 
> Reported-by: Raslan Darawsheh <rasland@mellanox.com>
> Reported-by: Xueming Zhang <xuemingx.zhang@intel.com>
> Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>


Hi Raslan, Xueming,

Can you please confirm the above patch on top of next-net fixes the build issue?

Thanks,
ferruh
Rahul Lakkireddy March 18, 2020, 1:48 p.m. UTC | #2 
On Wednesday, March 03/18/20, 2020 at 12:52:36 +0000, Ferruh Yigit wrote:
> Build error:
> .../drivers/net/cxgbe/cxgbe_flow.c:315:2:
>     error: implicit truncation from 'int' to bitfield changes value
>     from -1 to 7 [-Werror,-Wbitfield-constant-conversion]
>         CXGBE_FILL_FS(adap->pf, ~0, pf);
>         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> .../drivers/net/cxgbe/cxgbe_flow.c:25:2:
>     note: expanded from macro 'CXGBE_FILL_FS'
>         __CXGBE_FILL_FS(v, m, fs, elem, e)
>         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> .../drivers/net/cxgbe/cxgbe_flow.c:15:18: note:
>     expanded from macro '__CXGBE_FILL_FS'
>         (fs)->mask.elem = (__m); \
>                         ^ ~~~~~
> Commit in fixes line changes 'PF_BITWIDTH' to '3', which makes 'fs->pf'
> bitfield size '3', changing '~0' to '0x7' to prevent truncation in
> assignment.
> 
> Fixes: dcd456abe46f ("net/cxgbe: support flow API for matching all packets on PF")
> 
> Reported-by: Raslan Darawsheh <rasland@mellanox.com>
> Reported-by: Xueming Zhang <xuemingx.zhang@intel.com>
> Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>

I've tested CLANG compilation with 9.0.1 without this patch,
but couldn't hit the above mentioned error.

# clang --version
clang version 9.0.1 (Fedora 9.0.1-2.fc31)

However, after grabbing CLANG 3.4.2, I could see the error.

# clang --version
clang version 3.4.2 (tags/RELEASE_34/dot2-final)

This patch fixes it and the fix is correct too.

Thanks for fixing it.

Acked-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>

> ---
> Cc: kaara.satwik@chelsio.com
> ---
>  drivers/net/cxgbe/cxgbe_flow.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/net/cxgbe/cxgbe_flow.c b/drivers/net/cxgbe/cxgbe_flow.c
> index 13fd78aaf..a46515d3b 100644
> --- a/drivers/net/cxgbe/cxgbe_flow.c
> +++ b/drivers/net/cxgbe/cxgbe_flow.c
> @@ -312,7 +312,7 @@ ch_rte_parsetype_pf(const void *dmask __rte_unused,
>  
>  	CXGBE_FILL_FS(1, 1, pfvf_vld);
>  
> -	CXGBE_FILL_FS(adap->pf, ~0, pf);
> +	CXGBE_FILL_FS(adap->pf, 0x7, pf);
>  	return 0;
>  }
>  
> -- 
> 2.25.1
>
Raslan Darawsheh March 18, 2020, 2:46 p.m. UTC | #3 
Hi,
> -----Original Message-----
> From: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
> Sent: Wednesday, March 18, 2020 3:48 PM
> To: Ferruh Yigit <ferruh.yigit@intel.com>
> Cc: Karra Satwik <kaara.satwik@chelsio.com>; dev@dpdk.org; Raslan
> Darawsheh <rasland@mellanox.com>; Xueming Zhang
> <xuemingx.zhang@intel.com>
> Subject: Re: [PATCH] net/cxgbe: fix build with clang 3.4.2
> 
> On Wednesday, March 03/18/20, 2020 at 12:52:36 +0000, Ferruh Yigit wrote:
> > Build error:
> > .../drivers/net/cxgbe/cxgbe_flow.c:315:2:
> >     error: implicit truncation from 'int' to bitfield changes value
> >     from -1 to 7 [-Werror,-Wbitfield-constant-conversion]
> >         CXGBE_FILL_FS(adap->pf, ~0, pf);
> >         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> > .../drivers/net/cxgbe/cxgbe_flow.c:25:2:
> >     note: expanded from macro 'CXGBE_FILL_FS'
> >         __CXGBE_FILL_FS(v, m, fs, elem, e)
> >         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> > .../drivers/net/cxgbe/cxgbe_flow.c:15:18: note:
> >     expanded from macro '__CXGBE_FILL_FS'
> >         (fs)->mask.elem = (__m); \
> >                         ^ ~~~~~
> > Commit in fixes line changes 'PF_BITWIDTH' to '3', which makes 'fs->pf'
> > bitfield size '3', changing '~0' to '0x7' to prevent truncation in
> > assignment.
> >
> > Fixes: dcd456abe46f ("net/cxgbe: support flow API for matching all packets
> on PF")
> >
> > Reported-by: Raslan Darawsheh <rasland@mellanox.com>
> > Reported-by: Xueming Zhang <xuemingx.zhang@intel.com>
> > Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
> 
> I've tested CLANG compilation with 9.0.1 without this patch,
> but couldn't hit the above mentioned error.
> 
> # clang --version
> clang version 9.0.1 (Fedora 9.0.1-2.fc31)
> 
> However, after grabbing CLANG 3.4.2, I could see the error.
> 
> # clang --version
> clang version 3.4.2 (tags/RELEASE_34/dot2-final)
> 
> This patch fixes it and the fix is correct too.
> 
> Thanks for fixing it.
> 
Working just fine for me as well, thanks for the update
Tested-by:  Raslan Darawsheh <rasland@mellanox.com>

> Acked-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
> 
> > ---
> > Cc: kaara.satwik@chelsio.com
> > ---
> >  drivers/net/cxgbe/cxgbe_flow.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/drivers/net/cxgbe/cxgbe_flow.c
> b/drivers/net/cxgbe/cxgbe_flow.c
> > index 13fd78aaf..a46515d3b 100644
> > --- a/drivers/net/cxgbe/cxgbe_flow.c
> > +++ b/drivers/net/cxgbe/cxgbe_flow.c
> > @@ -312,7 +312,7 @@ ch_rte_parsetype_pf(const void *dmask
> __rte_unused,
> >
> >  	CXGBE_FILL_FS(1, 1, pfvf_vld);
> >
> > -	CXGBE_FILL_FS(adap->pf, ~0, pf);
> > +	CXGBE_FILL_FS(adap->pf, 0x7, pf);
> >  	return 0;
> >  }
> >
> > --
> > 2.25.1
> >
Ferruh Yigit March 18, 2020, 2:48 p.m. UTC | #4 
On 3/18/2020 1:48 PM, Rahul Lakkireddy wrote:
> On Wednesday, March 03/18/20, 2020 at 12:52:36 +0000, Ferruh Yigit wrote:
>> Build error:
>> .../drivers/net/cxgbe/cxgbe_flow.c:315:2:
>>     error: implicit truncation from 'int' to bitfield changes value
>>     from -1 to 7 [-Werror,-Wbitfield-constant-conversion]
>>         CXGBE_FILL_FS(adap->pf, ~0, pf);
>>         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> .../drivers/net/cxgbe/cxgbe_flow.c:25:2:
>>     note: expanded from macro 'CXGBE_FILL_FS'
>>         __CXGBE_FILL_FS(v, m, fs, elem, e)
>>         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> .../drivers/net/cxgbe/cxgbe_flow.c:15:18: note:
>>     expanded from macro '__CXGBE_FILL_FS'
>>         (fs)->mask.elem = (__m); \
>>                         ^ ~~~~~
>> Commit in fixes line changes 'PF_BITWIDTH' to '3', which makes 'fs->pf'
>> bitfield size '3', changing '~0' to '0x7' to prevent truncation in
>> assignment.
>>
>> Fixes: dcd456abe46f ("net/cxgbe: support flow API for matching all packets on PF")
>>
>> Reported-by: Raslan Darawsheh <rasland@mellanox.com>
>> Reported-by: Xueming Zhang <xuemingx.zhang@intel.com>
>> Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
> 
> I've tested CLANG compilation with 9.0.1 without this patch,
> but couldn't hit the above mentioned error.
> 
> # clang --version
> clang version 9.0.1 (Fedora 9.0.1-2.fc31)
> 
> However, after grabbing CLANG 3.4.2, I could see the error.
> 
> # clang --version
> clang version 3.4.2 (tags/RELEASE_34/dot2-final)
> 
> This patch fixes it and the fix is correct too.
> 
> Thanks for fixing it.
> 
> Acked-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>

Squashed into relevant commit in next-net, thanks.
Zhang, XuemingX March 19, 2020, 7:31 a.m. UTC | #5 
Many thanks Ferruh.

In dpdk-next-net and dpdk-next-net-intel, the patch verified.


>-----Original Message-----
>From: Yigit, Ferruh
>Sent: Wednesday, March 18, 2020 9:10 PM
>To: Raslan Darawsheh <rasland@mellanox.com>; Zhang, XuemingX
><xuemingx.zhang@intel.com>
>Cc: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>; Karra Satwik
><kaara.satwik@chelsio.com>; dev@dpdk.org
>Subject: Re: [PATCH] net/cxgbe: fix build with clang 3.4.2
>
>On 3/18/2020 12:52 PM, Ferruh Yigit wrote:
>> Build error:
>> .../drivers/net/cxgbe/cxgbe_flow.c:315:2:
>>     error: implicit truncation from 'int' to bitfield changes
>value
>>     from -1 to 7 [-Werror,-Wbitfield-constant-conversion]
>>         CXGBE_FILL_FS(adap->pf, ~0, pf);
>>         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> .../drivers/net/cxgbe/cxgbe_flow.c:25:2:
>>     note: expanded from macro 'CXGBE_FILL_FS'
>>         __CXGBE_FILL_FS(v, m, fs, elem, e)
>>         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> .../drivers/net/cxgbe/cxgbe_flow.c:15:18: note:
>>     expanded from macro '__CXGBE_FILL_FS'
>>         (fs)->mask.elem = (__m); \
>>                         ^ ~~~~~
>> Commit in fixes line changes 'PF_BITWIDTH' to '3', which makes
>'fs->pf'
>> bitfield size '3', changing '~0' to '0x7' to prevent truncation in
>> assignment.
>>
>> Fixes: dcd456abe46f ("net/cxgbe: support flow API for matching all
>> packets on PF")
>>
>> Reported-by: Raslan Darawsheh <rasland@mellanox.com>
>> Reported-by: Xueming Zhang <xuemingx.zhang@intel.com>
>> Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
>
>
>Hi Raslan, Xueming,
>
>Can you please confirm the above patch on top of next-net fixes the
>build issue?
>
>Thanks,
>ferruh
diff mbox series

Patch

diff --git a/drivers/net/cxgbe/cxgbe_flow.c b/drivers/net/cxgbe/cxgbe_flow.c
index 13fd78aaf..a46515d3b 100644
--- a/drivers/net/cxgbe/cxgbe_flow.c
+++ b/drivers/net/cxgbe/cxgbe_flow.c
@@ -312,7 +312,7 @@  ch_rte_parsetype_pf(const void *dmask __rte_unused,
 
 	CXGBE_FILL_FS(1, 1, pfvf_vld);
 
-	CXGBE_FILL_FS(adap->pf, ~0, pf);
+	CXGBE_FILL_FS(adap->pf, 0x7, pf);
 	return 0;
 }