warning: symbol 'rfi_flush' was not declared.
warning: symbol 'entry_flush' was not declared.
warning: symbol 'uaccess_flush' was not declared.
We found warnings above in arch/powerpc/kernel/setup_64.c by using
sparse tool.
Define 'entry_flush' and 'uaccess_flush' as static because they are not
referenced outside the file. Include asm/security_features.h in which
'rfi_flush' is declared.
Reported-by: Hulk Robot <[email protected]>
Signed-off-by: He Ying <[email protected]>
---
arch/powerpc/kernel/setup_64.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
index 560ed8b975e7..f92d72a7e7ce 100644
--- a/arch/powerpc/kernel/setup_64.c
+++ b/arch/powerpc/kernel/setup_64.c
@@ -68,6 +68,7 @@
#include <asm/early_ioremap.h>
#include <asm/pgalloc.h>
#include <asm/asm-prototypes.h>
+#include <asm/security_features.h>
#include "setup.h"
@@ -949,8 +950,8 @@ static bool no_rfi_flush;
static bool no_entry_flush;
static bool no_uaccess_flush;
bool rfi_flush;
-bool entry_flush;
-bool uaccess_flush;
+static bool entry_flush;
+static bool uaccess_flush;
DEFINE_STATIC_KEY_FALSE(uaccess_flush_key);
EXPORT_SYMBOL(uaccess_flush_key);
--
2.17.1
Hi He Ying,
Thank you for this patch.
I'm not sure what the precise rules for Fixes are, but I wonder if this
should have:
Fixes: 9a32a7e78bd0 ("powerpc/64s: flush L1D after user accesses")
Fixes: f79643787e0a ("powerpc/64s: flush L1D on kernel entry")
Those are the commits that added the entry_flush and uaccess_flush
symbols. Perhaps one for rfi_flush too but I'm not sure what commit
introduced that.
Kind regards,
Daniel
> warning: symbol 'rfi_flush' was not declared.
> warning: symbol 'entry_flush' was not declared.
> warning: symbol 'uaccess_flush' was not declared.
> We found warnings above in arch/powerpc/kernel/setup_64.c by using
> sparse tool.
>
> Define 'entry_flush' and 'uaccess_flush' as static because they are not
> referenced outside the file. Include asm/security_features.h in which
> 'rfi_flush' is declared.
>
> Reported-by: Hulk Robot <[email protected]>
> Signed-off-by: He Ying <[email protected]>
> ---
> arch/powerpc/kernel/setup_64.c | 5 +++--
> 1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
> index 560ed8b975e7..f92d72a7e7ce 100644
> --- a/arch/powerpc/kernel/setup_64.c
> +++ b/arch/powerpc/kernel/setup_64.c
> @@ -68,6 +68,7 @@
> #include <asm/early_ioremap.h>
> #include <asm/pgalloc.h>
> #include <asm/asm-prototypes.h>
> +#include <asm/security_features.h>
>
> #include "setup.h"
>
> @@ -949,8 +950,8 @@ static bool no_rfi_flush;
> static bool no_entry_flush;
> static bool no_uaccess_flush;
> bool rfi_flush;
> -bool entry_flush;
> -bool uaccess_flush;
> +static bool entry_flush;
> +static bool uaccess_flush;
> DEFINE_STATIC_KEY_FALSE(uaccess_flush_key);
> EXPORT_SYMBOL(uaccess_flush_key);
>
> --
> 2.17.1
Thank you for your reply.
?? 2021/3/17 11:04, Daniel Axtens д??:
> Hi He Ying,
>
> Thank you for this patch.
>
> I'm not sure what the precise rules for Fixes are, but I wonder if this
> should have:
>
> Fixes: 9a32a7e78bd0 ("powerpc/64s: flush L1D after user accesses")
> Fixes: f79643787e0a ("powerpc/64s: flush L1D on kernel entry")
Is that necessary for warning cleanups? I thought 'Fixes' tags are
needed only for
bugfix patches. Can someone tell me whether I am right?
>
> Those are the commits that added the entry_flush and uaccess_flush
> symbols. Perhaps one for rfi_flush too but I'm not sure what commit
> introduced that.
>
> Kind regards,
> Daniel
>
>> warning: symbol 'rfi_flush' was not declared.
>> warning: symbol 'entry_flush' was not declared.
>> warning: symbol 'uaccess_flush' was not declared.
>> We found warnings above in arch/powerpc/kernel/setup_64.c by using
>> sparse tool.
>>
>> Define 'entry_flush' and 'uaccess_flush' as static because they are not
>> referenced outside the file. Include asm/security_features.h in which
>> 'rfi_flush' is declared.
>>
>> Reported-by: Hulk Robot <[email protected]>
>> Signed-off-by: He Ying <[email protected]>
>> ---
>> arch/powerpc/kernel/setup_64.c | 5 +++--
>> 1 file changed, 3 insertions(+), 2 deletions(-)
>>
>> diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
>> index 560ed8b975e7..f92d72a7e7ce 100644
>> --- a/arch/powerpc/kernel/setup_64.c
>> +++ b/arch/powerpc/kernel/setup_64.c
>> @@ -68,6 +68,7 @@
>> #include <asm/early_ioremap.h>
>> #include <asm/pgalloc.h>
>> #include <asm/asm-prototypes.h>
>> +#include <asm/security_features.h>
>>
>> #include "setup.h"
>>
>> @@ -949,8 +950,8 @@ static bool no_rfi_flush;
>> static bool no_entry_flush;
>> static bool no_uaccess_flush;
>> bool rfi_flush;
>> -bool entry_flush;
>> -bool uaccess_flush;
>> +static bool entry_flush;
>> +static bool uaccess_flush;
>> DEFINE_STATIC_KEY_FALSE(uaccess_flush_key);
>> EXPORT_SYMBOL(uaccess_flush_key);
>>
>> --
>> 2.17.1
> .
"heying (H)" <[email protected]> writes:
> Thank you for your reply.
>
>
> 在 2021/3/17 11:04, Daniel Axtens 写道:
>> Hi He Ying,
>>
>> Thank you for this patch.
>>
>> I'm not sure what the precise rules for Fixes are, but I wonder if this
>> should have:
>>
>> Fixes: 9a32a7e78bd0 ("powerpc/64s: flush L1D after user accesses")
>> Fixes: f79643787e0a ("powerpc/64s: flush L1D on kernel entry")
>
> Is that necessary for warning cleanups? I thought 'Fixes' tags are
> needed only for
>
> bugfix patches. Can someone tell me whether I am right?
Yeah, I'm not sure either. Hopefully mpe will let us know.
Kind regards,
Daniel
>
>>
>> Those are the commits that added the entry_flush and uaccess_flush
>> symbols. Perhaps one for rfi_flush too but I'm not sure what commit
>> introduced that.
>>
>> Kind regards,
>> Daniel
>>
>>> warning: symbol 'rfi_flush' was not declared.
>>> warning: symbol 'entry_flush' was not declared.
>>> warning: symbol 'uaccess_flush' was not declared.
>>> We found warnings above in arch/powerpc/kernel/setup_64.c by using
>>> sparse tool.
>>>
>>> Define 'entry_flush' and 'uaccess_flush' as static because they are not
>>> referenced outside the file. Include asm/security_features.h in which
>>> 'rfi_flush' is declared.
>>>
>>> Reported-by: Hulk Robot <[email protected]>
>>> Signed-off-by: He Ying <[email protected]>
>>> ---
>>> arch/powerpc/kernel/setup_64.c | 5 +++--
>>> 1 file changed, 3 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
>>> index 560ed8b975e7..f92d72a7e7ce 100644
>>> --- a/arch/powerpc/kernel/setup_64.c
>>> +++ b/arch/powerpc/kernel/setup_64.c
>>> @@ -68,6 +68,7 @@
>>> #include <asm/early_ioremap.h>
>>> #include <asm/pgalloc.h>
>>> #include <asm/asm-prototypes.h>
>>> +#include <asm/security_features.h>
>>>
>>> #include "setup.h"
>>>
>>> @@ -949,8 +950,8 @@ static bool no_rfi_flush;
>>> static bool no_entry_flush;
>>> static bool no_uaccess_flush;
>>> bool rfi_flush;
>>> -bool entry_flush;
>>> -bool uaccess_flush;
>>> +static bool entry_flush;
>>> +static bool uaccess_flush;
>>> DEFINE_STATIC_KEY_FALSE(uaccess_flush_key);
>>> EXPORT_SYMBOL(uaccess_flush_key);
>>>
>>> --
>>> 2.17.1
>> .
Daniel Axtens <[email protected]> writes:
> "heying (H)" <[email protected]> writes:
>
>> Thank you for your reply.
>>
>> 在 2021/3/17 11:04, Daniel Axtens 写道:
>>> Hi He Ying,
>>>
>>> Thank you for this patch.
>>>
>>> I'm not sure what the precise rules for Fixes are, but I wonder if this
>>> should have:
>>>
>>> Fixes: 9a32a7e78bd0 ("powerpc/64s: flush L1D after user accesses")
>>> Fixes: f79643787e0a ("powerpc/64s: flush L1D on kernel entry")
>>
>> Is that necessary for warning cleanups? I thought 'Fixes' tags are
>> needed only for
>>
>> bugfix patches. Can someone tell me whether I am right?
>
> Yeah, I'm not sure either. Hopefully mpe will let us know.
It's not necessary to add a Fixes tag for a patch like this, but you can
add one if you think it's important that the fix gets backported.
I don't think the cleanups in this case are that important, so I
wouldn't bother with a Fixes tag.
cheers
在 2021/3/17 19:57, Michael Ellerman 写道:
> Daniel Axtens <[email protected]> writes:
>> "heying (H)" <[email protected]> writes:
>>
>>> Thank you for your reply.
>>>
>>> 在 2021/3/17 11:04, Daniel Axtens 写道:
>>>> Hi He Ying,
>>>>
>>>> Thank you for this patch.
>>>>
>>>> I'm not sure what the precise rules for Fixes are, but I wonder if this
>>>> should have:
>>>>
>>>> Fixes: 9a32a7e78bd0 ("powerpc/64s: flush L1D after user accesses")
>>>> Fixes: f79643787e0a ("powerpc/64s: flush L1D on kernel entry")
>>> Is that necessary for warning cleanups? I thought 'Fixes' tags are
>>> needed only for
>>>
>>> bugfix patches. Can someone tell me whether I am right?
>> Yeah, I'm not sure either. Hopefully mpe will let us know.
> It's not necessary to add a Fixes tag for a patch like this, but you can
> add one if you think it's important that the fix gets backported.
>
> I don't think the cleanups in this case are that important, so I
> wouldn't bother with a Fixes tag.
Okay. That's a good explanation to me. Thanks.
On Tue, 16 Mar 2021 00:11:48 -0400, He Ying wrote:
> warning: symbol 'rfi_flush' was not declared.
> warning: symbol 'entry_flush' was not declared.
> warning: symbol 'uaccess_flush' was not declared.
> We found warnings above in arch/powerpc/kernel/setup_64.c by using
> sparse tool.
>
> Define 'entry_flush' and 'uaccess_flush' as static because they are not
> referenced outside the file. Include asm/security_features.h in which
> 'rfi_flush' is declared.
Applied to powerpc/next.
[1/1] powerpc: arch/powerpc/kernel/setup_64.c - cleanup warnings
https://git.kernel.org/powerpc/c/d2313da4ff56bd631a3afe7a17992ed5bd0e04a6
cheers