linux-kernel - Re: [PATCH] powerpc/64s: Rewrite __real

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <e896ced5-f43f-478f-a55c-68a49f0492d9@csgroup.eu>
Date: Sun, 12 Jan 2025 19:26:00 +0100
From: Christophe Leroy <christophe.leroy@...roup.eu>
To: "Ritesh Harjani (IBM)" <ritesh.list@...il.com>,
 Michael Ellerman <mpe@...erman.id.au>, Nicholas Piggin <npiggin@...il.com>,
 Naveen N Rao <naveen@...nel.org>, Madhavan Srinivasan <maddy@...ux.ibm.com>
Cc: linux-kernel@...r.kernel.org, linuxppc-dev@...ts.ozlabs.org,
 kernel test robot <lkp@...el.com>
Subject: Re: [PATCH] powerpc/64s: Rewrite __real_pte() as a static inline



Le 11/01/2025 à 18:03, Ritesh Harjani a écrit :
> Christophe Leroy <christophe.leroy@...roup.eu> writes:
> 
>> Rewrite __real_pte() as a static inline in order to avoid
>> following warning/error when building with 4k page size:
>>
>> 	  CC      arch/powerpc/mm/book3s64/hash_tlb.o
>> 	arch/powerpc/mm/book3s64/hash_tlb.c: In function 'hpte_need_flush':
>> 	arch/powerpc/mm/book3s64/hash_tlb.c:49:16: error: variable 'offset' set but not used [-Werror=unused-but-set-variable]
>> 	   49 |         int i, offset;
>> 	      |                ^~~~~~
>> 	cc1: all warnings being treated as errors
>>
>> Reported-by: kernel test robot <lkp@...el.com>
>> Closes: https://lore.kernel.org/oe-kbuild-all/202501081741.AYFwybsq-lkp@intel.com/
> 
> Great. Why not fix the other warning as well which is reported in above
> link, which is...

Just because .... I overlooked the report and missed the second error.

Sent v2 with both fixes now.


> 
> --
>     arch/powerpc/mm/book3s64/hash_native.c: In function 'native_flush_hash_range':
>>> arch/powerpc/mm/book3s64/hash_native.c:786:29: warning: variable 'index' set but not used [-Wunused-but-set-variable]
>       786 |         unsigned long hash, index, hidx, shift, slot;
> --
> 
> ...similar to how we fixed this warning by making the macro as static
> inline? That means something like this (not tested)?
> 
> -#define __rpte_to_hidx(r,index)        (pte_val(__rpte_to_pte(r)) >> H_PAGE_F_GIX_SHIFT)
> +static inline unsigned long __rpte_to_hidx(real_pte_t r, unsigned long index)
> +{
> +       return pte_val(__rpte_to_pte(r)) >> H_PAGE_F_GIX_SHIFT;
> +}
> 
> 
> -ritesh
> 
> 
>> Fixes: ff31e105464d ("powerpc/mm/hash64: Store the slot information at the right offset for hugetlb")
>> Signed-off-by: Christophe Leroy <christophe.leroy@...roup.eu>
>> ---
>>   arch/powerpc/include/asm/book3s/64/hash-4k.h | 5 ++++-
>>   1 file changed, 4 insertions(+), 1 deletion(-)
>>
>> diff --git a/arch/powerpc/include/asm/book3s/64/hash-4k.h b/arch/powerpc/include/asm/book3s/64/hash-4k.h
>> index c3efacab4b94..a7a68ba9c71b 100644
>> --- a/arch/powerpc/include/asm/book3s/64/hash-4k.h
>> +++ b/arch/powerpc/include/asm/book3s/64/hash-4k.h
>> @@ -77,7 +77,10 @@
>>   /*
>>    * With 4K page size the real_pte machinery is all nops.
>>    */
>> -#define __real_pte(e, p, o)		((real_pte_t){(e)})
>> +static inline real_pte_t __real_pte(pte_t pte, pte_t *ptep, int offset)
>> +{
>> +	return (real_pte_t){pte};
>> +}
>>   #define __rpte_to_pte(r)	((r).pte)
>>   #define __rpte_to_hidx(r,index)	(pte_val(__rpte_to_pte(r)) >> H_PAGE_F_GIX_SHIFT)
>>   
>> -- 
>> 2.47.0