Re: [Qemu-devel] [PULL 0/6] linux-user patches for 2.4 softfreeze

[Laurent Vivier <laurent@vivier.eu>]

Le 15/06/2015 17:14, Peter Maydell a écrit :
> On 15 June 2015 at 13:20,  <riku.voipio@linaro.org> wrote:
>> From: Riku Voipio <riku.voipio@linaro.org>
>>
>> The following changes since commit 0a2df857a7038c75379cc575de5d4be4c0ac629e:
>>
>>   Merge remote-tracking branch 'remotes/stefanha/tags/net-pull-request' into staging (2015-06-12 15:39:05 +0100)
>>
>> are available in the git repository at:
>>
>>   git://git.linaro.org/people/riku.voipio/qemu.git tags/pull-linux-user-20150615
>>
>> for you to fetch changes up to d2897da1f1e97d684f80ff62d473c31b79bc643a:
>>
>>   linux-user: fix the breakpoint inheritance in spawned threads (2015-06-15 11:36:59 +0300)
>>
>> ----------------------------------------------------------------
>> linux-user patches for 2.4 softfreeze
>>
>> ----------------------------------------------------------------
> 
> I get a lot of build errors with clang:
> 
> /home/petmay01/linaro/qemu-for-merges/linux-user/syscall.c:3610:10:
> error: overflow
>       converting case value to switch condition type (3241737481 to
> 18446744072656321801)
>       [-Werror,-Wswitch]
>     case DM_TABLE_LOAD:
>          ^
> /usr/include/linux/dm-ioctl.h:259:26: note: expanded from macro 'DM_TABLE_LOAD'
> #define DM_TABLE_LOAD    _IOWR(DM_IOCTL, DM_TABLE_LOAD_CMD, struct dm_ioctl)
>                          ^
> /usr/include/asm-generic/ioctl.h:77:29: note: expanded from macro '_IOWR'
> #define _IOWR(type,nr,size)
> _IOC(_IOC_READ|_IOC_WRITE,(type),(nr),(_IOC_TYPECHE...
>                                 ^
> /usr/include/asm-generic/ioctl.h:66:2: note: expanded from macro '_IOC'
>         (((dir)  << _IOC_DIRSHIFT) | \
>         ^
> /home/petmay01/linaro/qemu-for-merges/linux-user/syscall.c:3606:10:
> error: overflow
>       converting case value to switch condition type (3241737486 to
> 18446744072656321806)
>       [-Werror,-Wswitch]
>     case DM_TARGET_MSG:
>          ^
> (etc etc for all the cases until clang gives up because it's emitted
> too many errors).
> 
> Guessing this is the result of the ioctl patch?

Yes, I guess too.

Could drop this patch out ? I will rework it.

Laurent