| 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
| ||
|
Message-ID: <87a7254bxd.fsf@mpe.ellerman.id.au>
Date: Mon, 18 May 2020 23:20:30 +1000
From: Michael Ellerman <mpe@...erman.id.au>
To: Nathan Chancellor <natechancellor@...il.com>,
kbuild test robot <lkp@...el.com>
Cc: kbuild-all@...ts.01.org, clang-built-linux@...glegroups.com,
linux-kernel@...r.kernel.org, linuxppc-dev@...ts.ozlabs.org
Subject: Remove WALNUT hacks in serio/i8042 (was Re: ld.lld: error: undefined symbol: kb_cs)
[ Cc += linuxppc-dev ]
Nathan Chancellor <natechancellor@...il.com> writes:
> On Thu, May 14, 2020 at 08:13:48AM +0800, kbuild test robot wrote:
>> tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
>> head: 24085f70a6e1b0cb647ec92623284641d8270637
>> commit: 5990cdee689c6885b27c6d969a3d58b09002b0bc lib/mpi: Fix building for powerpc with clang
>
> I am certain that this patch did nothing to cause this... Maybe exposed
> it but not the root cause.
>
>> date: 3 weeks ago
>> config: powerpc-randconfig-r034-20200514 (attached as .config)
>> compiler: clang version 11.0.0 (https://github.com/llvm/llvm-project 310d32cb80a611e6384a921e85607fea05841f26)
>> reproduce:
>> wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
>> chmod +x ~/bin/make.cross
>> # install powerpc cross compiling tool for clang build
>> # apt-get install binutils-powerpc-linux-gnu
>> git checkout 5990cdee689c6885b27c6d969a3d58b09002b0bc
>> # save the attached .config to linux build tree
>> COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross ARCH=powerpc
>>
>> If you fix the issue, kindly add following tag as appropriate
>> Reported-by: kbuild test robot <lkp@...el.com>
>>
>> All errors (new ones prefixed by >>):
>>
>> >> ld.lld: error: undefined symbol: kb_cs
>> >>> referenced by i8042-ppcio.h:28 (drivers/input/serio/i8042-ppcio.h:28)
>> >>> input/serio/i8042.o:(__i8042_command) in archive drivers/built-in.a
>> >>> referenced by i8042-ppcio.h:28 (drivers/input/serio/i8042-ppcio.h:28)
>> >>> input/serio/i8042.o:(__i8042_command) in archive drivers/built-in.a
>> >>> referenced by i8042-ppcio.h:28 (drivers/input/serio/i8042-ppcio.h:28)
>> >>> input/serio/i8042.o:(__i8042_command) in archive drivers/built-in.a
>> >>> referenced 45 more times
>> --
>> >> ld.lld: error: undefined symbol: kb_data
>> >>> referenced by i8042.c:309 (drivers/input/serio/i8042.c:309)
>> >>> input/serio/i8042.o:(__i8042_command) in archive drivers/built-in.a
>> >>> referenced by i8042-ppcio.h:33 (drivers/input/serio/i8042-ppcio.h:33)
>> >>> input/serio/i8042.o:(__i8042_command) in archive drivers/built-in.a
>> >>> referenced by i8042.c:319 (drivers/input/serio/i8042.c:319)
>> >>> input/serio/i8042.o:(__i8042_command) in archive drivers/built-in.a
>> >>> referenced 15 more times
>
> kb_cs and kb_data are declared as extern void pointers when
> CONFIG_WALNUT is set, which this config does. However, it looks like
> the definitions of these variables were removed in
> commit 917f0af9e5a9 ("powerpc: Remove arch/ppc and include/asm-ppc"),
> way back in 2.6.27-rc1... So presumably, this has been broken for 12
> years and nobody noticed?
Hmm, nice. 12 years is a pretty typical time frame for finding bugs in
the powerpc tree ... o_O
Our ppc40x_defconfig has CONFIG_SERIO=m, but explicitly disables
CONFIG_SERIO_I8042. And walnut_defconfig disables CONFIG_SERIO.
So I guess it has a PS2 controller but most folks didn't use it?
> Probably means there is a good amount of dead
> code that can be removed, or we could just avoid ever selecting this
> driver with CONFIG_WALNUT (if it does not actually work without the
> special handling from i8042-ppcio.h) while removing the dead code.
I'm inclined to let the driver be selected, rather than carry a wart for
it. I doubt anyone will ever turn it on explicitly, and if they do they
can have fun debugging it.
> diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
> index 924c541a9260..59b2f655e39e 100644
> --- a/arch/powerpc/Kconfig
> +++ b/arch/powerpc/Kconfig
> @@ -138,7 +138,7 @@ config PPC
> select ARCH_HAVE_NMI_SAFE_CMPXCHG
> select ARCH_KEEP_MEMBLOCK
> select ARCH_MIGHT_HAVE_PC_PARPORT
> - select ARCH_MIGHT_HAVE_PC_SERIO
> + select ARCH_MIGHT_HAVE_PC_SERIO if !WALNUT
> select ARCH_OPTIONAL_KERNEL_RWX if ARCH_HAS_STRICT_KERNEL_RWX
> select ARCH_SUPPORTS_ATOMIC_RMW
> select ARCH_USE_BUILTIN_BSWAP
So I'd drop this hunk.
> diff --git a/drivers/input/serio/i8042-ppcio.h b/drivers/input/serio/i8042-ppcio.h
> deleted file mode 100644
> index 391f94d9e47d..000000000000
> --- a/drivers/input/serio/i8042-ppcio.h
> +++ /dev/null
> @@ -1,57 +0,0 @@
> -/* SPDX-License-Identifier: GPL-2.0-only */
> -#ifndef _I8042_PPCIO_H
> -#define _I8042_PPCIO_H
> -
> -
> -#if defined(CONFIG_WALNUT)
> -
> -#define I8042_KBD_IRQ 25
> -#define I8042_AUX_IRQ 26
> -
> -#define I8042_KBD_PHYS_DESC "walnutps2/serio0"
> -#define I8042_AUX_PHYS_DESC "walnutps2/serio1"
> -#define I8042_MUX_PHYS_DESC "walnutps2/serio%d"
> -
> -extern void *kb_cs;
> -extern void *kb_data;
> -
> -#define I8042_COMMAND_REG (*(int *)kb_cs)
> -#define I8042_DATA_REG (*(int *)kb_data)
> -
> -static inline int i8042_read_data(void)
> -{
> - return readb(kb_data);
> -}
> -
> -static inline int i8042_read_status(void)
> -{
> - return readb(kb_cs);
> -}
> -
> -static inline void i8042_write_data(int val)
> -{
> - writeb(val, kb_data);
> -}
> -
> -static inline void i8042_write_command(int val)
> -{
> - writeb(val, kb_cs);
> -}
> -
> -static inline int i8042_platform_init(void)
> -{
> - i8042_reset = I8042_RESET_ALWAYS;
> - return 0;
> -}
> -
> -static inline void i8042_platform_exit(void)
> -{
> -}
> -
> -#else
> -
> -#include "i8042-io.h"
> -
> -#endif
> -
> -#endif /* _I8042_PPCIO_H */
> diff --git a/drivers/input/serio/i8042.h b/drivers/input/serio/i8042.h
> index 38dc27ad3c18..eb376700dfff 100644
> --- a/drivers/input/serio/i8042.h
> +++ b/drivers/input/serio/i8042.h
> @@ -17,8 +17,6 @@
> #include "i8042-ip22io.h"
> #elif defined(CONFIG_SNI_RM)
> #include "i8042-snirm.h"
> -#elif defined(CONFIG_PPC)
> -#include "i8042-ppcio.h"
> #elif defined(CONFIG_SPARC)
> #include "i8042-sparcio.h"
> #elif defined(CONFIG_X86) || defined(CONFIG_IA64)
#elif defined(CONFIG_UNICORE32)
#include "i8042-unicore32io.h"
#else
#include "i8042-io.h"
#endif
The rest of the context is ^
Which makes it clear that your removal of the entire file results in no
change when WALNUT=n.
I'm not sure if anyone really maintains that code, but this is all PPC
really so I think we could merge it.
Do you want to send a proper SOB'ed patch?
cheers
Powered by blists - more mailing lists