[PATCH v1] drivers:base:Fix unsigned compared with less than zero

[PATCH v1] drivers:base:Fix unsigned compared with less than zero

[Wang Ming]

The return value of the of_count_cache_leaves() is long.
However, the return value is being assigned to an unsigned
long variable 'leaves',so making 'leaves' to long.

silence the warning:
./drivers/base/cacheinfo.c:300:5-11: WARNING: Unsigned expression
compared with zero: leaves > 0

Signed-off-by: Wang Ming <machel@vivo.com>
---
 drivers/base/cacheinfo.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/base/cacheinfo.c b/drivers/base/cacheinfo.c
index cbae8be1f..7280d5cb3 100644
--- a/drivers/base/cacheinfo.c
+++ b/drivers/base/cacheinfo.c
@@ -289,7 +289,8 @@ int init_of_cache_level(unsigned int cpu)
        struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu);
        struct device_node *np = of_cpu_device_node_get(cpu);
        struct device_node *prev = NULL;
-       unsigned int levels = 0, leaves, level;
+       unsigned int levels = 0, level;
+       int leaves;

        if (!of_check_cache_nodes(np)) {
                of_node_put(np);
--
2.25.1


________________________________
本邮件及其附件内容可能含有机密和/或隐私信息，仅供指定个人或机构使用。若您非发件人指定收件人或其代理人，请勿使用、传播、复制或存储此邮件之任何内容或其附件。如您误收本邮件，请即以回复或电话方式通知发件人，并将原始邮件、附件及其所有复本删除。谢谢。
The contents of this message and any attachments may contain confidential and/or privileged information and are intended exclusively for the addressee(s). If you are not the intended recipient of this message or their agent, please note that any use, dissemination, copying, or storage of this message or its attachments is not allowed. If you receive this message in error, please notify the sender by reply the message or phone and delete this message, any attachments and any copies immediately.
Thank you

Re: [PATCH v1] drivers:base:Fix unsigned compared with less than zero

[Conor Dooley]

[-- Attachment #1: Type: text/plain, Size: 2495 bytes --]

Hey,

On Wed, Jun 14, 2023 at 09:35:36PM +0800, Wang Ming wrote:
> The return value of the of_count_cache_leaves() is long.
> However, the return value is being assigned to an unsigned
> long variable 'leaves',so making 'leaves' to long.

But you actually made it an int?

> silence the warning:
> ./drivers/base/cacheinfo.c:300:5-11: WARNING: Unsigned expression
> compared with zero: leaves > 0

I'm dumb, why is comparing an unsigned value with zero a problem in and
of itself? Zero is a valid value for an unsigned type, no?

If you actually look at the function, it only returns positive, non-zero
values anyway, so returning an int is a bit pointless, as are the
checks, no?

> 
> Signed-off-by: Wang Ming <machel@vivo.com>
> ---
>  drivers/base/cacheinfo.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/base/cacheinfo.c b/drivers/base/cacheinfo.c
> index cbae8be1f..7280d5cb3 100644
> --- a/drivers/base/cacheinfo.c
> +++ b/drivers/base/cacheinfo.c
> @@ -289,7 +289,8 @@ int init_of_cache_level(unsigned int cpu)
>         struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu);
>         struct device_node *np = of_cpu_device_node_get(cpu);
>         struct device_node *prev = NULL;
> -       unsigned int levels = 0, leaves, level;
> +       unsigned int levels = 0, level;
> +       int leaves;
> 
>         if (!of_check_cache_nodes(np)) {
>                 of_node_put(np);
> --
> 2.25.1
> 
> 
> ________________________________
> 本邮件及其附件内容可能含有机密和/或隐私信息，仅供指定个人或机构使用。若您非发件人指定收件人或其代理人，请勿使用、传播、复制或存储此邮件之任何内容或其附件。如您误收本邮件，请即以回复或电话方式通知发件人，并将原始邮件、附件及其所有复本删除。谢谢。
> The contents of this message and any attachments may contain confidential and/or privileged information and are intended exclusively for the addressee(s). If you are not the intended recipient of this message or their agent, please note that any use, dissemination, copying, or storage of this message or its attachments is not allowed. If you receive this message in error, please notify the sender by reply the message or phone and delete this message, any attachments and any copies immediately.
^^^
Gotta sort this out chief, I should probably be deleting this, not
replying to it...

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

Re: 回复: [PATCH v1] drivers:base:Fix unsigned compared with less than zero

[Sudeep Holla]

On Thu, Jun 15, 2023 at 07:50:45AM +0100, Conor Dooley wrote:
> Hey,
> 
> For some (probably good) reason, this doesn't appear to have made it to
> the lists.
> 
> On Thu, Jun 15, 2023 at 02:05:18AM +0000, 王明-软件底层技术部 wrote:
> > 发件人: Conor Dooley <conor@kernel.org>
> > > On Wed, Jun 14, 2023 at 09:35:36PM +0800, Wang Ming wrote:
> > > > The return value of the of_count_cache_leaves() is long.
> > > > However, the return value is being assigned to an unsigned long
> > > > variable 'leaves',so making 'leaves' to long.
> > > 
> > > But you actually made it an int?
> > > 
> > > > silence the warning:
> > > > ./drivers/base/cacheinfo.c:300:5-11: WARNING: Unsigned expression
> > > > compared with zero: leaves > 0
> > > 
> > > I'm dumb, why is comparing an unsigned value with zero a problem in
> > > and of itself? Zero is a valid value for an unsigned type, no?
> > > 
> > > If you actually look at the function, it only returns positive,
> > > non-zero values anyway, so returning an int is a bit pointless, as
> > > are the checks, no?
> 
> > Thank you, I see. This doesn't need to be fixed.
> 
> Right, but I was suggesting that you should change the function to not
> return a signed value anymore & remove the check for whether it is
> zero, instead of your patch.
> 
> Cheers,
> Conor.
> 
> +CC Sudeep, Pierre: FYI. Original posting is at
> https://lore.kernel.org/all/20230614133549.3774-1-machel@vivo.com/
> in case threading is broken, as I think the in-reply-to header in what
> I am replying to is corrupted.

Also if the original thread landed, I couldn't have responded because of
the disclaimer 🙁.

-- 
Regards,
Sudeep