Getting rid of "uint" and friends

[RFC 1/3] ARM: Use normal types

Posted by Juan Quintela 2 years, 9 months ago

Someone has a good reason why this is not a good idea?

Signed-off-by: Juan Quintela <quintela@redhat.com>
---
 bsd-user/arm/target_arch_reg.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/bsd-user/arm/target_arch_reg.h b/bsd-user/arm/target_arch_reg.h
index 070fa24da1..5f1eea4291 100644
--- a/bsd-user/arm/target_arch_reg.h
+++ b/bsd-user/arm/target_arch_reg.h
@@ -32,7 +32,7 @@ typedef struct target_reg {
 typedef struct target_fp_reg {
     uint32_t        fp_exponent;
     uint32_t        fp_mantissa_hi;
-    u_int32_t       fp_mantissa_lo;
+    uint32_t       fp_mantissa_lo;
 } target_fp_reg_t;
 
 typedef struct target_fpreg {
-- 
2.40.1

Re: [RFC 1/3] ARM: Use normal types

Posted by Warner Losh 2 years, 9 months ago

On Wed, May 10, 2023 at 8:39 AM Juan Quintela <quintela@redhat.com> wrote:

> Someone has a good reason why this is not a good idea?
>
> Signed-off-by: Juan Quintela <quintela@redhat.com>
>

Reviewed by:  Warner Losh <imp@bsdimp.com>

This has been that way the bsd-user sources were reorganized in 2015. I can
find
no good reason in the FreeBSD sources to do this (we've been transitioning
from
the pre-standardized BSD convention of u_intXX_t -> uintXX_t for 25 years
now
it seems). I don't see any old or ancient usage as far back as I looked why
they'd
be different. Up through FreeBSD 12.x, this was u_int32_t (for all of
them), but
they switched to __uint32_t in FreeBSD 13 to avoid namespace pollution.

tl;dr: change good, all should match.

> ---
>  bsd-user/arm/target_arch_reg.h | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/bsd-user/arm/target_arch_reg.h
> b/bsd-user/arm/target_arch_reg.h
> index 070fa24da1..5f1eea4291 100644
> --- a/bsd-user/arm/target_arch_reg.h
> +++ b/bsd-user/arm/target_arch_reg.h
> @@ -32,7 +32,7 @@ typedef struct target_reg {
>  typedef struct target_fp_reg {
>      uint32_t        fp_exponent;
>      uint32_t        fp_mantissa_hi;
> -    u_int32_t       fp_mantissa_lo;
> +    uint32_t       fp_mantissa_lo;
>  } target_fp_reg_t;
>
>  typedef struct target_fpreg {
> --
> 2.40.1
>
>

Re: [RFC 1/3] ARM: Use normal types

Posted by Warner Losh 2 years, 9 months ago

On Mon, May 15, 2023 at 10:44 AM Warner Losh <imp@bsdimp.com> wrote:

>
>
> On Wed, May 10, 2023 at 8:39 AM Juan Quintela <quintela@redhat.com> wrote:
>
>> Someone has a good reason why this is not a good idea?
>>
>> Signed-off-by: Juan Quintela <quintela@redhat.com>
>>
>
> Reviewed by:  Warner Losh <imp@bsdimp.com>
>
> This has been that way the bsd-user sources were reorganized in 2015. I
> can find
> no good reason in the FreeBSD sources to do this (we've been transitioning
> from
> the pre-standardized BSD convention of u_intXX_t -> uintXX_t for 25 years
> now
> it seems). I don't see any old or ancient usage as far back as I looked
> why they'd
> be different. Up through FreeBSD 12.x, this was u_int32_t (for all of
> them), but
> they switched to __uint32_t in FreeBSD 13 to avoid namespace pollution.
>
> tl;dr: change good, all should match.
>

Though a better commit message would be good. With that, I'll queue it to
my branch.

Warner


> ---
>>  bsd-user/arm/target_arch_reg.h | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/bsd-user/arm/target_arch_reg.h
>> b/bsd-user/arm/target_arch_reg.h
>> index 070fa24da1..5f1eea4291 100644
>> --- a/bsd-user/arm/target_arch_reg.h
>> +++ b/bsd-user/arm/target_arch_reg.h
>> @@ -32,7 +32,7 @@ typedef struct target_reg {
>>  typedef struct target_fp_reg {
>>      uint32_t        fp_exponent;
>>      uint32_t        fp_mantissa_hi;
>> -    u_int32_t       fp_mantissa_lo;
>> +    uint32_t       fp_mantissa_lo;
>>  } target_fp_reg_t;
>>
>>  typedef struct target_fpreg {
>> --
>> 2.40.1
>>
>>

Re: [RFC 1/3] ARM: Use normal types

Posted by Juan Quintela 2 years, 9 months ago

Warner Losh <imp@bsdimp.com> wrote:
> On Mon, May 15, 2023 at 10:44 AM Warner Losh <imp@bsdimp.com> wrote:
>
>>
>>
>> On Wed, May 10, 2023 at 8:39 AM Juan Quintela <quintela@redhat.com> wrote:
>>
>>> Someone has a good reason why this is not a good idea?
>>>
>>> Signed-off-by: Juan Quintela <quintela@redhat.com>
>>>
>>
>> Reviewed by:  Warner Losh <imp@bsdimp.com>
>>
>> This has been that way the bsd-user sources were reorganized in 2015. I
>> can find
>> no good reason in the FreeBSD sources to do this (we've been transitioning
>> from
>> the pre-standardized BSD convention of u_intXX_t -> uintXX_t for 25 years
>> now
>> it seems). I don't see any old or ancient usage as far back as I looked
>> why they'd
>> be different. Up through FreeBSD 12.x, this was u_int32_t (for all of
>> them), but
>> they switched to __uint32_t in FreeBSD 13 to avoid namespace pollution.
>>
>> tl;dr: change good, all should match.
>>
>
> Though a better commit message would be good. With that, I'll queue it to
> my branch.

I think your bit of history would be good O:-)

Later, Juan.

Re: [RFC 1/3] ARM: Use normal types

Posted by Richard Henderson 2 years, 9 months ago

On 5/10/23 15:39, Juan Quintela wrote:
> Someone has a good reason why this is not a good idea?
> 
> Signed-off-by: Juan Quintela<quintela@redhat.com>
> ---
>   bsd-user/arm/target_arch_reg.h | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>


r~