fstat is replaced by statx on the new architecture, so an exception
is added to the checksyscalls script to silence the following build
warning on LoongArch:
CALL scripts/checksyscalls.sh
<stdin>:569:2: warning: #warning syscall fstat not implemented [-Wcpp]
Suggested-by: WANG Xuerui <[email protected]>
Suggested-by: Arnd Bergmann <[email protected]>
Signed-off-by: Tiezhu Yang <[email protected]>
---
Hi Xuerui and Arnd,
Thank you for your reviews and suggestions.
v3:
-- Move the definition to its own block with some explanation
v2:
-- Change scripts/checksyscalls.sh
-- Modify patch subject and commit message
scripts/checksyscalls.sh | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/scripts/checksyscalls.sh b/scripts/checksyscalls.sh
index f33e61a..1e5d2ee 100755
--- a/scripts/checksyscalls.sh
+++ b/scripts/checksyscalls.sh
@@ -114,7 +114,6 @@ cat << EOF
#define __IGNORE_truncate
#define __IGNORE_stat
#define __IGNORE_lstat
-#define __IGNORE_fstat
#define __IGNORE_fcntl
#define __IGNORE_fadvise64
#define __IGNORE_newfstatat
@@ -255,6 +254,9 @@ cat << EOF
/* 64-bit ports never needed these, and new 32-bit ports can use statx */
#define __IGNORE_fstat64
#define __IGNORE_fstatat64
+
+/* Newer ports are not required to provide fstat in favor of statx */
+#define __IGNORE_fstat
EOF
}
--
2.1.0
On 08/30/2022 11:27 AM, Tiezhu Yang wrote:
> fstat is replaced by statx on the new architecture, so an exception
> is added to the checksyscalls script to silence the following build
> warning on LoongArch:
>
> CALL scripts/checksyscalls.sh
> <stdin>:569:2: warning: #warning syscall fstat not implemented [-Wcpp]
Hi all,
The above warning still exists when build the latest loongarch-next.
Do you know which tree this patch will go through?
Could you please pick it up via your tree in this merge window?
https://lore.kernel.org/lkml/[email protected]/
Thanks,
Tiezhu
>
> Suggested-by: WANG Xuerui <[email protected]>
> Suggested-by: Arnd Bergmann <[email protected]>
> Signed-off-by: Tiezhu Yang <[email protected]>
> ---
>
> Hi Xuerui and Arnd,
>
> Thank you for your reviews and suggestions.
>
> v3:
> -- Move the definition to its own block with some explanation
>
> v2:
> -- Change scripts/checksyscalls.sh
> -- Modify patch subject and commit message
>
> scripts/checksyscalls.sh | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/scripts/checksyscalls.sh b/scripts/checksyscalls.sh
> index f33e61a..1e5d2ee 100755
> --- a/scripts/checksyscalls.sh
> +++ b/scripts/checksyscalls.sh
> @@ -114,7 +114,6 @@ cat << EOF
> #define __IGNORE_truncate
> #define __IGNORE_stat
> #define __IGNORE_lstat
> -#define __IGNORE_fstat
> #define __IGNORE_fcntl
> #define __IGNORE_fadvise64
> #define __IGNORE_newfstatat
> @@ -255,6 +254,9 @@ cat << EOF
> /* 64-bit ports never needed these, and new 32-bit ports can use statx */
> #define __IGNORE_fstat64
> #define __IGNORE_fstatat64
> +
> +/* Newer ports are not required to provide fstat in favor of statx */
> +#define __IGNORE_fstat
> EOF
> }
>
>
On Sun, 2022-10-09 at 12:33 +0800, Tiezhu Yang wrote:
>
>
> On 08/30/2022 11:27 AM, Tiezhu Yang wrote:
> > fstat is replaced by statx on the new architecture, so an exception
> > is added to the checksyscalls script to silence the following build
> > warning on LoongArch:
> >
> > CALL scripts/checksyscalls.sh
> > <stdin>:569:2: warning: #warning syscall fstat not implemented [-
> > Wcpp]
>
> Hi all,
>
> The above warning still exists when build the latest loongarch-next.
>
> Do you know which tree this patch will go through?
>
> Could you please pick it up via your tree in this merge window?
A similar change 3ef6ca4f354c ("checksyscalls: Unconditionally ignore
fstat{,at}64") was merged through RISC-V tree, so it should be OK to
merge this from loongarch-next.
Not sure if we need to get an Ack from someone first though.
--
Xi Ruoyao <[email protected]>
School of Aerospace Science and Technology, Xidian University
Cc KERNEL BUILD maintainers:
Masahiro Yamada <[email protected]>
Michal Marek <[email protected]>
Nick Desaulniers <[email protected]>
On 10/09/2022 10:01 PM, Xi Ruoyao wrote:
> On Sun, 2022-10-09 at 12:33 +0800, Tiezhu Yang wrote:
>>
>>
>> On 08/30/2022 11:27 AM, Tiezhu Yang wrote:
>>> fstat is replaced by statx on the new architecture, so an exception
>>> is added to the checksyscalls script to silence the following build
>>> warning on LoongArch:
>>>
>>> CALL scripts/checksyscalls.sh
>>> <stdin>:569:2: warning: #warning syscall fstat not implemented [-
>>> Wcpp]
>>
>> Hi all,
>>
>> The above warning still exists when build the latest loongarch-next.
>>
>> Do you know which tree this patch will go through?
>>
>> Could you please pick it up via your tree in this merge window?
>
> A similar change 3ef6ca4f354c ("checksyscalls: Unconditionally ignore
> fstat{,at}64") was merged through RISC-V tree, so it should be OK to
> merge this from loongarch-next.
>
> Not sure if we need to get an Ack from someone first though.
>
Since this is a kernel build warning, maybe it is proper through
linux-kbuild.git.
https://lore.kernel.org/lkml/[email protected]/
Thanks,
Tiezhu
Hi Andrew,
On 10/12/2022 02:19 PM, Tiezhu Yang wrote:
>
> Cc KERNEL BUILD maintainers:
>
> Masahiro Yamada <[email protected]>
> Michal Marek <[email protected]>
> Nick Desaulniers <[email protected]>
>
> On 10/09/2022 10:01 PM, Xi Ruoyao wrote:
>> On Sun, 2022-10-09 at 12:33 +0800, Tiezhu Yang wrote:
>>>
>>>
>>> On 08/30/2022 11:27 AM, Tiezhu Yang wrote:
>>>> fstat is replaced by statx on the new architecture, so an exception
>>>> is added to the checksyscalls script to silence the following build
>>>> warning on LoongArch:
>>>>
>>>> CALL scripts/checksyscalls.sh
>>>> <stdin>:569:2: warning: #warning syscall fstat not implemented [-
>>>> Wcpp]
>>>
>>> Hi all,
>>>
>>> The above warning still exists when build the latest loongarch-next.
>>>
>>> Do you know which tree this patch will go through?
>>>
>>> Could you please pick it up via your tree in this merge window?
>>
>> A similar change 3ef6ca4f354c ("checksyscalls: Unconditionally ignore
>> fstat{,at}64") was merged through RISC-V tree, so it should be OK to
>> merge this from loongarch-next.
>>
>> Not sure if we need to get an Ack from someone first though.
>>
>
> Since this is a kernel build warning, maybe it is proper through
> linux-kbuild.git.
>
> https://lore.kernel.org/lkml/[email protected]/
>
>
> Thanks,
> Tiezhu
The build warning still exists when build the latest Linux 6.1-rc3 on
LoongArch, if you are OK with this change, please apply it through your
mm tree, thank you.
"If you cannot find a maintainer for the subsystem you are working on,
Andrew Morton ([email protected]) serves as a maintainer of
last resort."
https://www.kernel.org/doc/html/latest/process/submitting-patches.html
Thanks,
Tiezhu