diff mbox series

can: rockchip_canfd: fix return type of rkcanfd_start_xmit()

Message ID 20240906-rockchip-canfd-wifpts-v1-1-b1398da865b7@kernel.org (mailing list archive)
State New
Headers show
Series can: rockchip_canfd: fix return type of rkcanfd_start_xmit() | expand

Commit Message

Nathan Chancellor Sept. 6, 2024, 8:26 p.m. UTC 
With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
indirect call targets are validated against the expected function
pointer prototype to make sure the call target is valid to help mitigate
ROP attacks. If they are not identical, there is a failure at run time,
which manifests as either a kernel panic or thread getting killed. A
warning in clang aims to catch these at compile time, which reveals:

  drivers/net/can/rockchip/rockchip_canfd-core.c:770:20: error: incompatible function pointer types initializing 'netdev_tx_t (*)(struct sk_buff *, struct net_device *)' (aka 'enum netdev_tx (*)(struct sk_buff *, struct net_device *)') with an expression of type 'int (struct sk_buff *, struct net_device *)' [-Werror,-Wincompatible-function-pointer-types-strict]
    770 |         .ndo_start_xmit = rkcanfd_start_xmit,
        |                           ^~~~~~~~~~~~~~~~~~

->ndo_start_xmit() in 'struct net_device_ops' expects a return type of
'netdev_tx_t', not 'int' (although the types are ABI compatible). Adjust
the return type of rkcanfd_start_xmit() to match the prototype's to
resolve the warning.

Fixes: ff60bfbaf67f ("can: rockchip_canfd: add driver for Rockchip CAN-FD controller")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
---
 drivers/net/can/rockchip/rockchip_canfd-tx.c | 2 +-
 drivers/net/can/rockchip/rockchip_canfd.h    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)


---
base-commit: 52fc70a32573707f70d6b1b5c5fe85cc91457393
change-id: 20240906-rockchip-canfd-wifpts-3a9b76e7c228

Best regards,

Comments

Marc Kleine-Budde Sept. 9, 2024, 7:14 a.m. UTC | #1 
On 06.09.2024 13:26:41, Nathan Chancellor wrote:
> With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
> indirect call targets are validated against the expected function
> pointer prototype to make sure the call target is valid to help mitigate
> ROP attacks. If they are not identical, there is a failure at run time,
> which manifests as either a kernel panic or thread getting killed. A
> warning in clang aims to catch these at compile time, which reveals:
> 
>   drivers/net/can/rockchip/rockchip_canfd-core.c:770:20: error: incompatible function pointer types initializing 'netdev_tx_t (*)(struct sk_buff *, struct net_device *)' (aka 'enum netdev_tx (*)(struct sk_buff *, struct net_device *)') with an expression of type 'int (struct sk_buff *, struct net_device *)' [-Werror,-Wincompatible-function-pointer-types-strict]
>     770 |         .ndo_start_xmit = rkcanfd_start_xmit,
>         |                           ^~~~~~~~~~~~~~~~~~
> 
> ->ndo_start_xmit() in 'struct net_device_ops' expects a return type of
> 'netdev_tx_t', not 'int' (although the types are ABI compatible). Adjust
> the return type of rkcanfd_start_xmit() to match the prototype's to
> resolve the warning.
> 
> Fixes: ff60bfbaf67f ("can: rockchip_canfd: add driver for Rockchip CAN-FD controller")
> Signed-off-by: Nathan Chancellor <nathan@kernel.org>

Applied to linux-can-next.

Thanks,
Marc
Simon Horman Sept. 9, 2024, 8:44 a.m. UTC | #2 
On Fri, Sep 06, 2024 at 01:26:41PM -0700, Nathan Chancellor wrote:
> With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
> indirect call targets are validated against the expected function
> pointer prototype to make sure the call target is valid to help mitigate
> ROP attacks. If they are not identical, there is a failure at run time,
> which manifests as either a kernel panic or thread getting killed. A
> warning in clang aims to catch these at compile time, which reveals:
> 
>   drivers/net/can/rockchip/rockchip_canfd-core.c:770:20: error: incompatible function pointer types initializing 'netdev_tx_t (*)(struct sk_buff *, struct net_device *)' (aka 'enum netdev_tx (*)(struct sk_buff *, struct net_device *)') with an expression of type 'int (struct sk_buff *, struct net_device *)' [-Werror,-Wincompatible-function-pointer-types-strict]
>     770 |         .ndo_start_xmit = rkcanfd_start_xmit,
>         |                           ^~~~~~~~~~~~~~~~~~
> 
> ->ndo_start_xmit() in 'struct net_device_ops' expects a return type of
> 'netdev_tx_t', not 'int' (although the types are ABI compatible). Adjust
> the return type of rkcanfd_start_xmit() to match the prototype's to
> resolve the warning.
> 
> Fixes: ff60bfbaf67f ("can: rockchip_canfd: add driver for Rockchip CAN-FD controller")
> Signed-off-by: Nathan Chancellor <nathan@kernel.org>

Thanks, I was able to reproduce this problem at build time
and that your patch addresses it.

Reviewed-by: Simon Horman <horms@kernel.org>
Marc Kleine-Budde Sept. 9, 2024, 8:57 a.m. UTC | #3 
On 09.09.2024 09:44:48, Simon Horman wrote:
> On Fri, Sep 06, 2024 at 01:26:41PM -0700, Nathan Chancellor wrote:
> > With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
> > indirect call targets are validated against the expected function
> > pointer prototype to make sure the call target is valid to help mitigate
> > ROP attacks. If they are not identical, there is a failure at run time,
> > which manifests as either a kernel panic or thread getting killed. A
> > warning in clang aims to catch these at compile time, which reveals:
> > 
> >   drivers/net/can/rockchip/rockchip_canfd-core.c:770:20: error: incompatible function pointer types initializing 'netdev_tx_t (*)(struct sk_buff *, struct net_device *)' (aka 'enum netdev_tx (*)(struct sk_buff *, struct net_device *)') with an expression of type 'int (struct sk_buff *, struct net_device *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> >     770 |         .ndo_start_xmit = rkcanfd_start_xmit,
> >         |                           ^~~~~~~~~~~~~~~~~~
> > 
> > ->ndo_start_xmit() in 'struct net_device_ops' expects a return type of
> > 'netdev_tx_t', not 'int' (although the types are ABI compatible). Adjust
> > the return type of rkcanfd_start_xmit() to match the prototype's to
> > resolve the warning.
> > 
> > Fixes: ff60bfbaf67f ("can: rockchip_canfd: add driver for Rockchip CAN-FD controller")
> > Signed-off-by: Nathan Chancellor <nathan@kernel.org>
> 
> Thanks, I was able to reproduce this problem at build time
> and that your patch addresses it.

FTR: the default clang in Debian unstable, clang-16.0.6 doesn't support
this. With clang-20 from experimental it works, haven't checked older
versions, though.

regards,
Marc
Simon Horman Sept. 9, 2024, 2:35 p.m. UTC | #4 
On Mon, Sep 09, 2024 at 10:57:06AM +0200, Marc Kleine-Budde wrote:
> On 09.09.2024 09:44:48, Simon Horman wrote:
> > On Fri, Sep 06, 2024 at 01:26:41PM -0700, Nathan Chancellor wrote:
> > > With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
> > > indirect call targets are validated against the expected function
> > > pointer prototype to make sure the call target is valid to help mitigate
> > > ROP attacks. If they are not identical, there is a failure at run time,
> > > which manifests as either a kernel panic or thread getting killed. A
> > > warning in clang aims to catch these at compile time, which reveals:
> > > 
> > >   drivers/net/can/rockchip/rockchip_canfd-core.c:770:20: error: incompatible function pointer types initializing 'netdev_tx_t (*)(struct sk_buff *, struct net_device *)' (aka 'enum netdev_tx (*)(struct sk_buff *, struct net_device *)') with an expression of type 'int (struct sk_buff *, struct net_device *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> > >     770 |         .ndo_start_xmit = rkcanfd_start_xmit,
> > >         |                           ^~~~~~~~~~~~~~~~~~
> > > 
> > > ->ndo_start_xmit() in 'struct net_device_ops' expects a return type of
> > > 'netdev_tx_t', not 'int' (although the types are ABI compatible). Adjust
> > > the return type of rkcanfd_start_xmit() to match the prototype's to
> > > resolve the warning.
> > > 
> > > Fixes: ff60bfbaf67f ("can: rockchip_canfd: add driver for Rockchip CAN-FD controller")
> > > Signed-off-by: Nathan Chancellor <nathan@kernel.org>
> > 
> > Thanks, I was able to reproduce this problem at build time
> > and that your patch addresses it.
> 
> FTR: the default clang in Debian unstable, clang-16.0.6 doesn't support
> this. With clang-20 from experimental it works, haven't checked older
> versions, though.

FTR: I checked using 18.1.8 from here [1][2].

[1] https://mirrors.edge.kernel.org/pub/tools/llvm/
[2] https://mirrors.edge.kernel.org/pub/tools/llvm/files/
Nathan Chancellor Sept. 9, 2024, 3:27 p.m. UTC | #5 
Hi Marc,

On Mon, Sep 09, 2024 at 10:57:06AM +0200, Marc Kleine-Budde wrote:
> On 09.09.2024 09:44:48, Simon Horman wrote:
> > On Fri, Sep 06, 2024 at 01:26:41PM -0700, Nathan Chancellor wrote:
> > >   drivers/net/can/rockchip/rockchip_canfd-core.c:770:20: error: incompatible function pointer types initializing 'netdev_tx_t (*)(struct sk_buff *, struct net_device *)' (aka 'enum netdev_tx (*)(struct sk_buff *, struct net_device *)') with an expression of type 'int (struct sk_buff *, struct net_device *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> > >     770 |         .ndo_start_xmit = rkcanfd_start_xmit,
> > >         |                           ^~~~~~~~~~~~~~~~~~
...
> FTR: the default clang in Debian unstable, clang-16.0.6 doesn't support
> this. With clang-20 from experimental it works, haven't checked older
> versions, though.

Hmmm, interesting, the patch that added
-Wincompatible-function-pointer-types-strict was added in LLVM 16, so it
should work for 16.0.6...

https://github.com/llvm/llvm-project/commits/41ce74e6e983f523d44d3a80be5ae778c35df85a

I don't have easy access to Debian at the moment so I can double check
it later. I would like to get this turned on for the whole kernel soon
but there is still one subsystem that has several instances that I have
not been able to workaround at this point so I've just stuck to adding
it via KCFLAGS when testing.

Cheers,
Nathan
Marc Kleine-Budde Sept. 10, 2024, 9:56 a.m. UTC | #6 
On 09.09.2024 15:35:46, Simon Horman wrote:
> On Mon, Sep 09, 2024 at 10:57:06AM +0200, Marc Kleine-Budde wrote:
> > On 09.09.2024 09:44:48, Simon Horman wrote:
> > > On Fri, Sep 06, 2024 at 01:26:41PM -0700, Nathan Chancellor wrote:
> > > > With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
> > > > indirect call targets are validated against the expected function
> > > > pointer prototype to make sure the call target is valid to help mitigate
> > > > ROP attacks. If they are not identical, there is a failure at run time,
> > > > which manifests as either a kernel panic or thread getting killed. A
> > > > warning in clang aims to catch these at compile time, which reveals:
> > > > 
> > > >   drivers/net/can/rockchip/rockchip_canfd-core.c:770:20: error: incompatible function pointer types initializing 'netdev_tx_t (*)(struct sk_buff *, struct net_device *)' (aka 'enum netdev_tx (*)(struct sk_buff *, struct net_device *)') with an expression of type 'int (struct sk_buff *, struct net_device *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> > > >     770 |         .ndo_start_xmit = rkcanfd_start_xmit,
> > > >         |                           ^~~~~~~~~~~~~~~~~~
> > > > 
> > > > ->ndo_start_xmit() in 'struct net_device_ops' expects a return type of
> > > > 'netdev_tx_t', not 'int' (although the types are ABI compatible). Adjust
> > > > the return type of rkcanfd_start_xmit() to match the prototype's to
> > > > resolve the warning.
> > > > 
> > > > Fixes: ff60bfbaf67f ("can: rockchip_canfd: add driver for Rockchip CAN-FD controller")
> > > > Signed-off-by: Nathan Chancellor <nathan@kernel.org>
> > > 
> > > Thanks, I was able to reproduce this problem at build time
> > > and that your patch addresses it.
> > 
> > FTR: the default clang in Debian unstable, clang-16.0.6 doesn't support
> > this. With clang-20 from experimental it works, haven't checked older
> > versions, though.
> 
> FTR: I checked using 18.1.8 from here [1][2].
> 
> [1] https://mirrors.edge.kernel.org/pub/tools/llvm/
> [2] https://mirrors.edge.kernel.org/pub/tools/llvm/files/

I was a bit hasty yesterday, clang-20 and W=1 produces these errors:

| include/linux/vmstat.h:517:36: error: arithmetic between different enumeration types ('enum node_stat_item' and 'enum lru_list') [-Werror,-Wenum-enum-conversion]
|   517 |         return node_stat_name(NR_LRU_BASE + lru) + 3; // skip "nr_"
|       |                               ~~~~~~~~~~~ ^ ~~~
| 1 error generated.

However I fail to reproduce the ndo_start_xmit problem. Even with 18.1.8
from kernel.org.


The following command (ARCH is unset, compiling x86 -> x86) produces the
above shown "vmstat.h" problems....

| $ make LLVM=1 LLVM_IAS=1 LLVM_SUFFIX=-20 drivers/net/can/rockchip/  W=1 CONFIG_WERROR=0

... but not the ndo_start_xmit problem.


Am I missing a vital .config option?

| $ grep "CLANG\|CFI" .config
| CONFIG_CC_IS_CLANG=y
| CONFIG_CLANG_VERSION=200000
| CONFIG_CFI_AUTO_DEFAULT=y
| CONFIG_FUNCTION_PADDING_CFI=11
| CONFIG_LTO_CLANG=y
| CONFIG_ARCH_SUPPORTS_LTO_CLANG=y
| CONFIG_ARCH_SUPPORTS_LTO_CLANG_THIN=y
| CONFIG_HAS_LTO_CLANG=y
| CONFIG_LTO_CLANG_THIN=y
| CONFIG_ARCH_SUPPORTS_CFI_CLANG=y
| CONFIG_ARCH_USES_CFI_TRAPS=y
| CONFIG_CFI_CLANG=y
| # CONFIG_CFI_PERMISSIVE is not set


V=1 gives this output:

|   clang-20 -Wp,-MMD,drivers/net/can/rockchip/.rockchip_canfd-core.o.d
|   -nostdinc -Iarch/x86/include -I./arch/x86/include/generated
|   -Iinclude -I./include -Iarch/x86/include/uapi
|   -I./arch/x86/include/generated/uapi -Iinclude/uapi
|   -I./include/generated/uapi -include include/linux/compiler-version.h
|   -include include/linux/kconfig.h -include
|   include/linux/compiler_types.h -D__KERNEL__
|   --target=x86_64-linux-gnu -fintegrated-as
|   -Werror=unknown-warning-option -Werror=ignored-optimization-argument
|   -Werror=option-ignored -Werror=unused-command-line-argument
|   -fmacro-prefix-map== -Wundef -DKBUILD_EXTRA_WARN1 -std=gnu11
|   -fshort-wchar -funsigned-char -fno-common -fno-PIE
|   -fno-strict-aliasing -mno-sse -mno-mmx -mno-sse2 -mno-3dnow -mno-avx
|   -fcf-protection=branch -fno-jump-tables -m64 -falign-loops=1
|   -mno-80387 -mno-fp-ret-in-387 -mstack-alignment=8 -mskip-rax-setup
|   -mtune=generic -mno-red-zone -mcmodel=kernel -Wno-sign-compare
|   -fno-asynchronous-unwind-tables -mretpoline-external-thunk
|   -mindirect-branch-cs-prefix -mfunction-return=thunk-extern
|   -fpatchable-function-entry=11,11 -fno-delete-null-pointer-checks -O2
|   -fstack-protector-strong -fomit-frame-pointer
|   -fno-stack-clash-protection -fno-lto -flto=thin -fsplit-lto-unit
|   -fvisibility=hidden -fsanitize=kcfi -falign-functions=16
|   -fstrict-flex-arrays=3 -fno-strict-overflow -fno-stack-check -Wall
|   -Wundef -Werror=implicit-function-declaration -Werror=implicit-int
|   -Werror=return-type -Werror=strict-prototypes -Wno-format-security
|   -Wno-trigraphs -Wno-frame-address -Wno-address-of-packed-member
|   -Wmissing-declarations -Wmissing-prototypes -Wframe-larger-than=2048
|   -Wno-gnu -Wvla -Wno-pointer-sign -Wcast-function-type
|   -Wimplicit-fallthrough -Werror=date-time
|   -Werror=incompatible-pointer-types -Wenum-conversion -Wextra
|   -Wunused -Wmissing-format-attribute -Wmissing-include-dirs
|   -Wunused-const-variable -Wno-missing-field-initializers
|   -Wno-type-limits -Wno-shift-negative-value -Wno-sign-compare
|   -Wno-unused-parameter -g -gdwarf-5 -DDEBUG
|   -Idrivers/net/can/rockchip -Idrivers/net/can/rockchip -DMODULE
|   -DKBUILD_BASENAME='"rockchip_canfd_core"'
|   -DKBUILD_MODNAME='"rockchip_canfd"'
|   -D__KBUILD_MODNAME=kmod_rockchip_canfd -c -o
|   drivers/net/can/rockchip/rockchip_canfd-core.o
|   drivers/net/can/rockchip/rockchip_canfd-core.c
| In file included from drivers/net/can/rockchip/rockchip_canfd-core.c:25:
| In file included from drivers/net/can/rockchip/rockchip_canfd.h:11:
| In file included from include/linux/can/dev.h:18:
| In file included from include/linux/can/bittiming.h:9:
| In file included from include/linux/netdevice.h:38:
| In file included from include/net/net_namespace.h:43:
| In file included from include/linux/skbuff.h:17:
| In file included from include/linux/bvec.h:10:
| In file included from include/linux/highmem.h:8:
| In file included from include/linux/cacheflush.h:5:
| In file included from arch/x86/include/asm/cacheflush.h:5:
| In file included from include/linux/mm.h:2232:
| include/linux/vmstat.h:517:36: warning: arithmetic between different enumeration types ('enum node_stat_item' and 'enum lru_list') [-Wenum-enum-conversion]
|   517 |         return node_stat_name(NR_LRU_BASE + lru) + 3; // skip "nr_"
|       |                               ~~~~~~~~~~~ ^ ~~~
| 1 warning generated.

regards,
Marc
Nathan Chancellor Sept. 10, 2024, 7:05 p.m. UTC | #7 
Hi Marc,

On Tue, Sep 10, 2024 at 11:56:56AM +0200, Marc Kleine-Budde wrote:
> On 09.09.2024 15:35:46, Simon Horman wrote:
> > On Mon, Sep 09, 2024 at 10:57:06AM +0200, Marc Kleine-Budde wrote:
> > > On 09.09.2024 09:44:48, Simon Horman wrote:
> > > > On Fri, Sep 06, 2024 at 01:26:41PM -0700, Nathan Chancellor wrote:
> > > > > With clang's kernel control flow integrity (kCFI, CONFIG_CFI_CLANG),
> > > > > indirect call targets are validated against the expected function
> > > > > pointer prototype to make sure the call target is valid to help mitigate
> > > > > ROP attacks. If they are not identical, there is a failure at run time,
> > > > > which manifests as either a kernel panic or thread getting killed. A
> > > > > warning in clang aims to catch these at compile time, which reveals:
> > > > > 
> > > > >   drivers/net/can/rockchip/rockchip_canfd-core.c:770:20: error: incompatible function pointer types initializing 'netdev_tx_t (*)(struct sk_buff *, struct net_device *)' (aka 'enum netdev_tx (*)(struct sk_buff *, struct net_device *)') with an expression of type 'int (struct sk_buff *, struct net_device *)' [-Werror,-Wincompatible-function-pointer-types-strict]
> > > > >     770 |         .ndo_start_xmit = rkcanfd_start_xmit,
> > > > >         |                           ^~~~~~~~~~~~~~~~~~
> > > > > 
> > > > > ->ndo_start_xmit() in 'struct net_device_ops' expects a return type of
> > > > > 'netdev_tx_t', not 'int' (although the types are ABI compatible). Adjust
> > > > > the return type of rkcanfd_start_xmit() to match the prototype's to
> > > > > resolve the warning.
> > > > > 
> > > > > Fixes: ff60bfbaf67f ("can: rockchip_canfd: add driver for Rockchip CAN-FD controller")
> > > > > Signed-off-by: Nathan Chancellor <nathan@kernel.org>
> > > > 
> > > > Thanks, I was able to reproduce this problem at build time
> > > > and that your patch addresses it.
> > > 
> > > FTR: the default clang in Debian unstable, clang-16.0.6 doesn't support
> > > this. With clang-20 from experimental it works, haven't checked older
> > > versions, though.
> > 
> > FTR: I checked using 18.1.8 from here [1][2].
> > 
> > [1] https://mirrors.edge.kernel.org/pub/tools/llvm/
> > [2] https://mirrors.edge.kernel.org/pub/tools/llvm/files/
> 
> I was a bit hasty yesterday, clang-20 and W=1 produces these errors:
> 
> | include/linux/vmstat.h:517:36: error: arithmetic between different enumeration types ('enum node_stat_item' and 'enum lru_list') [-Werror,-Wenum-enum-conversion]
> |   517 |         return node_stat_name(NR_LRU_BASE + lru) + 3; // skip "nr_"
> |       |                               ~~~~~~~~~~~ ^ ~~~
> | 1 error generated.

Unfortunately, this is a completely tangential issue. You can see some
backstory behind it in commit 75b5ab134bb5 ("kbuild: Move
-Wenum-{compare-conditional,enum-conversion} into W=1"). To be honest, I
should consider moving that to W=2...

> However I fail to reproduce the ndo_start_xmit problem. Even with 18.1.8
> from kernel.org.
> 
> 
> The following command (ARCH is unset, compiling x86 -> x86) produces the
> above shown "vmstat.h" problems....
> 
> | $ make LLVM=1 LLVM_IAS=1 LLVM_SUFFIX=-20 drivers/net/can/rockchip/  W=1 CONFIG_WERROR=0

FYI, you could shorten this to just:

  $ make LLVM=-20 drivers/net/can/rockchip/ W=1 CONFIG_WERROR=0

As LLVM_SUFFIX will be set through LLVM and LLVM_IAS has defaulted to 1
since 5.15.

Does CONFIG_WERROR=0 work? It seems like it is still present above.

> ... but not the ndo_start_xmit problem.
> 
> 
> Am I missing a vital .config option?

No, I might not have made it clear in this commit message but this
warning is not on by default. I am looking to turn it on at some point
so I keep up with the warnings that it produces but there is one
subsystem that has several instances and I am unsure of how to solve
them to the maintainer's satisfaction. You can test it by adding

  KCFLAGS=-Wincompatible-function-pointer-types-strict

to your make command above and it should reproduce.

Cheers,
Nathan
Marc Kleine-Budde Sept. 11, 2024, 8:18 a.m. UTC | #8 
On 10.09.2024 12:05:25, Nathan Chancellor wrote:
> > I was a bit hasty yesterday, clang-20 and W=1 produces these errors:
> > 
> > | include/linux/vmstat.h:517:36: error: arithmetic between different enumeration types ('enum node_stat_item' and 'enum lru_list') [-Werror,-Wenum-enum-conversion]
> > |   517 |         return node_stat_name(NR_LRU_BASE + lru) + 3; // skip "nr_"
> > |       |                               ~~~~~~~~~~~ ^ ~~~
> > | 1 error generated.
> 
> Unfortunately, this is a completely tangential issue. You can see some
> backstory behind it in commit 75b5ab134bb5 ("kbuild: Move
> -Wenum-{compare-conditional,enum-conversion} into W=1"). To be honest, I
> should consider moving that to W=2...

Thanks for the background info. I wanted to point out that at least
clang-20 with W=1 finds _something_, though not that what I wanted to
reproduce :)

> > However I fail to reproduce the ndo_start_xmit problem. Even with 18.1.8
> > from kernel.org.
> > 
> > 
> > The following command (ARCH is unset, compiling x86 -> x86) produces the
> > above shown "vmstat.h" problems....
> > 
> > | $ make LLVM=1 LLVM_IAS=1 LLVM_SUFFIX=-20 drivers/net/can/rockchip/  W=1 CONFIG_WERROR=0
> 
> FYI, you could shorten this to just:
> 
>   $ make LLVM=-20 drivers/net/can/rockchip/ W=1 CONFIG_WERROR=0
> 
> As LLVM_SUFFIX will be set through LLVM and LLVM_IAS has defaulted to 1
> since 5.15.

Thanks, scripts updated :)

> Does CONFIG_WERROR=0 work? It seems like it is still present above.

Yes, it works.

> > ... but not the ndo_start_xmit problem.
> > 
> > Am I missing a vital .config option?
> 
> No, I might not have made it clear in this commit message but this
> warning is not on by default. I am looking to turn it on at some point
> so I keep up with the warnings that it produces but there is one
> subsystem that has several instances and I am unsure of how to solve
> them to the maintainer's satisfaction. You can test it by adding
> 
>   KCFLAGS=-Wincompatible-function-pointer-types-strict
> 
> to your make command above and it should reproduce.

That was the missing link. Can reproduce now, thanks!

regards,
Marc
diff mbox series

Patch

diff --git a/drivers/net/can/rockchip/rockchip_canfd-tx.c b/drivers/net/can/rockchip/rockchip_canfd-tx.c
index f954f38b955f6dc26b856e4c2f40e20256a7f114..865a15e033a9e50f44af342eec43d4ba10b40fa7 100644
--- a/drivers/net/can/rockchip/rockchip_canfd-tx.c
+++ b/drivers/net/can/rockchip/rockchip_canfd-tx.c
@@ -63,7 +63,7 @@  void rkcanfd_xmit_retry(struct rkcanfd_priv *priv)
 	rkcanfd_start_xmit_write_cmd(priv, reg_cmd);
 }
 
-int rkcanfd_start_xmit(struct sk_buff *skb, struct net_device *ndev)
+netdev_tx_t rkcanfd_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 {
 	struct rkcanfd_priv *priv = netdev_priv(ndev);
 	u32 reg_frameinfo, reg_id, reg_cmd;
diff --git a/drivers/net/can/rockchip/rockchip_canfd.h b/drivers/net/can/rockchip/rockchip_canfd.h
index 3efd7f174e14c2b67b1feef897bc34336f578226..93131c7d7f54d036727356ca9a2d36dcf24d702b 100644
--- a/drivers/net/can/rockchip/rockchip_canfd.h
+++ b/drivers/net/can/rockchip/rockchip_canfd.h
@@ -546,7 +546,7 @@  void rkcanfd_timestamp_stop_sync(struct rkcanfd_priv *priv);
 
 unsigned int rkcanfd_get_effective_tx_free(const struct rkcanfd_priv *priv);
 void rkcanfd_xmit_retry(struct rkcanfd_priv *priv);
-int rkcanfd_start_xmit(struct sk_buff *skb, struct net_device *ndev);
+netdev_tx_t rkcanfd_start_xmit(struct sk_buff *skb, struct net_device *ndev);
 void rkcanfd_handle_tx_done_one(struct rkcanfd_priv *priv, const u32 ts,
 				unsigned int *frame_len_p);