Add __noreturn annotation to fix the warning:
net/core/skbuff.o: warning: objtool: skb_push+0x3c: skb_panic() is missing a __noreturn annotation
net/core/skbuff.o: warning: objtool: skb_put+0x4e: skb_panic() is missing a __noreturn annotation
Signed-off-by: Miaohe Lin <[email protected]>
---
net/core/skbuff.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 6724a84ebb09..12b525aa4783 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -181,8 +181,8 @@ EXPORT_SYMBOL_GPL(drop_reasons_unregister_subsys);
* Keep out of line to prevent kernel bloat.
* __builtin_return_address is not used because it is not always reliable.
*/
-static void skb_panic(struct sk_buff *skb, unsigned int sz, void *addr,
- const char msg[])
+static void __noreturn skb_panic(struct sk_buff *skb, unsigned int sz, void *addr,
+ const char msg[])
{
pr_emerg("%s: text:%px len:%d put:%d head:%px data:%px tail:%#lx end:%#lx dev:%s\n",
msg, addr, skb->len, sz, skb->head, skb->data,
--
2.27.0
From: Miaohe Lin <[email protected]>
Date: Sat, 27 May 2023 19:04:09 +0800
> Add __noreturn annotation to fix the warning:
> net/core/skbuff.o: warning: objtool: skb_push+0x3c: skb_panic() is missing a __noreturn annotation
> net/core/skbuff.o: warning: objtool: skb_put+0x4e: skb_panic() is missing a __noreturn annotation
What arch are you using ?
IIUC, BUG() should have an annotation for objtool, for
example, __builtin_unreachable() for x86.
Maybe the arch is missing such an annotation ?
Also I'm curious why objtool complains about only skb_push(),
there should be more non-inline functions that has BUG().
Thanks,
Kuniyuki
>
> Signed-off-by: Miaohe Lin <[email protected]>
> ---
> net/core/skbuff.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/net/core/skbuff.c b/net/core/skbuff.c
> index 6724a84ebb09..12b525aa4783 100644
> --- a/net/core/skbuff.c
> +++ b/net/core/skbuff.c
> @@ -181,8 +181,8 @@ EXPORT_SYMBOL_GPL(drop_reasons_unregister_subsys);
> * Keep out of line to prevent kernel bloat.
> * __builtin_return_address is not used because it is not always reliable.
> */
> -static void skb_panic(struct sk_buff *skb, unsigned int sz, void *addr,
> - const char msg[])
> +static void __noreturn skb_panic(struct sk_buff *skb, unsigned int sz, void *addr,
> + const char msg[])
> {
> pr_emerg("%s: text:%px len:%d put:%d head:%px data:%px tail:%#lx end:%#lx dev:%s\n",
> msg, addr, skb->len, sz, skb->head, skb->data,
> --
> 2.27.0
On 2023/5/27 12:00, Kuniyuki Iwashima wrote:
> From: Miaohe Lin <[email protected]>
> Date: Sat, 27 May 2023 19:04:09 +0800
>> Add __noreturn annotation to fix the warning:
>> net/core/skbuff.o: warning: objtool: skb_push+0x3c: skb_panic() is missing a __noreturn annotation
>> net/core/skbuff.o: warning: objtool: skb_put+0x4e: skb_panic() is missing a __noreturn annotation
>
> What arch are you using ?
>
> IIUC, BUG() should have an annotation for objtool, for
> example, __builtin_unreachable() for x86.
>
> Maybe the arch is missing such an annotation ?
>
> Also I'm curious why objtool complains about only skb_push(),
> there should be more non-inline functions that has BUG().
I'm not really sure. I was just compiling the kernel using my old gcc (GCC) 7.3.0 and saw this warning.
Maybe this warning is just false positive.
Thanks.
>
> Thanks,
> Kuniyuki
>
>
>>
>> Signed-off-by: Miaohe Lin <[email protected]>
>> ---
>> net/core/skbuff.c | 4 ++--
>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/net/core/skbuff.c b/net/core/skbuff.c
>> index 6724a84ebb09..12b525aa4783 100644
>> --- a/net/core/skbuff.c
>> +++ b/net/core/skbuff.c
>> @@ -181,8 +181,8 @@ EXPORT_SYMBOL_GPL(drop_reasons_unregister_subsys);
>> * Keep out of line to prevent kernel bloat.
>> * __builtin_return_address is not used because it is not always reliable.
>> */
>> -static void skb_panic(struct sk_buff *skb, unsigned int sz, void *addr,
>> - const char msg[])
>> +static void __noreturn skb_panic(struct sk_buff *skb, unsigned int sz, void *addr,
>> + const char msg[])
>> {
>> pr_emerg("%s: text:%px len:%d put:%d head:%px data:%px tail:%#lx end:%#lx dev:%s\n",
>> msg, addr, skb->len, sz, skb->head, skb->data,
>> --
>> 2.27.0
> .
>
On Fri, 2023-05-26 at 21:00 -0700, Kuniyuki Iwashima wrote:
> From: Miaohe Lin <[email protected]>
> Date: Sat, 27 May 2023 19:04:09 +0800
> > Add __noreturn annotation to fix the warning:
> > net/core/skbuff.o: warning: objtool: skb_push+0x3c: skb_panic() is missing a __noreturn annotation
> > net/core/skbuff.o: warning: objtool: skb_put+0x4e: skb_panic() is missing a __noreturn annotation
>
> What arch are you using ?
>
> IIUC, BUG() should have an annotation for objtool, for
> example, __builtin_unreachable() for x86.
>
> Maybe the arch is missing such an annotation ?
>
> Also I'm curious why objtool complains about only skb_push(),
> there should be more non-inline functions that has BUG().
AFAICS, the BUG() macro implementation should already carry the
__noreturn annotation, via panic() or other arch-specific way.
This looks like the old toolchain not being able to
successfully/correctly propagate the annotation???
I think we can drop this patch.
Cheers,
Paolo
On 2023/5/30 18:28, Paolo Abeni wrote:
> On Fri, 2023-05-26 at 21:00 -0700, Kuniyuki Iwashima wrote:
>> From: Miaohe Lin <[email protected]>
>> Date: Sat, 27 May 2023 19:04:09 +0800
>>> Add __noreturn annotation to fix the warning:
>>> net/core/skbuff.o: warning: objtool: skb_push+0x3c: skb_panic() is missing a __noreturn annotation
>>> net/core/skbuff.o: warning: objtool: skb_put+0x4e: skb_panic() is missing a __noreturn annotation
>>
>> What arch are you using ?
>>
>> IIUC, BUG() should have an annotation for objtool, for
>> example, __builtin_unreachable() for x86.
>>
>> Maybe the arch is missing such an annotation ?
>>
>> Also I'm curious why objtool complains about only skb_push(),
>> there should be more non-inline functions that has BUG().
>
> AFAICS, the BUG() macro implementation should already carry the
> __noreturn annotation, via panic() or other arch-specific way.
>
> This looks like the old toolchain not being able to
> successfully/correctly propagate the annotation???
I tend to agree with you. Will drop it. Thanks for your comment.
>
> I think we can drop this patch.
>
> Cheers,
>
> Paolo
>
> .
>