Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

arch/x86/kernel/module.c:230:49: error: call to undeclared library function 'memcpy' #89

Closed
tklengyel opened this issue Oct 13, 2022 · 13 comments
Closed

arch/x86/kernel/module.c:230:49: error: call to undeclared library function 'memcpy' #89

tklengyel opened this issue Oct 13, 2022 · 13 comments

Comments

@tklengyel
Copy link
Contributor

Trying to compile latest linux git (49da070062390094112b423ba443ea193527b2e4) after KMSAN merge on Ubuntu 22.04 with clang-15 I get the following error:

make LLVM=-15 CC=clang-15 bzImage
...
arch/x86/kernel/module.c:230:49: error: call to undeclared library function 'memcpy' with type 'void *(void *, const void *, unsigned long)'; ISO C99 and later do not support implicit function declarations [-Werror,-Wimplicit-function-declaration]
        void *(*write)(void *, const void *, size_t) = memcpy;
@ramosian-glider
Copy link
Member

Any chance you could share the config you were using to build the kernel? Thanks!

@tklengyel
Copy link
Contributor Author

Here you go kmsan-failure.zip

@ramosian-glider
Copy link
Member

ramosian-glider commented Oct 17, 2022 •
edited
Loading

Ok, this must have something to do with CONFIG_FORTIFY_SOURCE=y, looking...

@ramosian-glider
Copy link
Member

Another finding is that CONFIG_FRAME_WARN makes no sense under KMSAN, and configs like CONFIG_WERROR and CONFIG_KVM_WERROR enforce build-time-errors because of that warning.

@ramosian-glider
Copy link
Member

And another one:

ld: arch/x86/purgatory/purgatory.ro: in function `purgatory':
(.text+0x1a): undefined reference to `__msan_get_context_state'
ld: (.text+0x35): undefined reference to `__msan_poison_alloca'
ld: (.text+0x69): undefined reference to `__msan_metadata_ptr_for_store_4'
ld: (.text+0x126): undefined reference to `__msan_metadata_ptr_for_store_8'
ld: (.text+0x153): undefined reference to `__msan_metadata_ptr_for_load_8'

@tklengyel
Copy link
Contributor Author

FYI the config is largely based off of the default Ubuntu config, just ran through olddefconfig and disabled some signing/lockdown options.

@ramosian-glider
Copy link
Member

FYI the config is largely based off of the default Ubuntu config, just ran through olddefconfig and disabled some signing/lockdown options.

Ok, then it really makes sense for me to make it work. It's nice that it uncovered so many problems at once.

@ramosian-glider
Copy link
Member

Another one:

  MODPOST Module.symvers
ERROR: modpost: "kmsan_copy_page_meta" [fs/nilfs2/nilfs2.ko] undefined!
ERROR: modpost: "kmsan_copy_page_meta" [drivers/gpu/drm/ttm/ttm.ko] undefined!
ERROR: modpost: "kmsan_copy_page_meta" [drivers/block/drbd/drbd.ko] undefined!
make[1]: *** [scripts/Makefile.modpost:126: Module.symvers] Error 1

@nickdesaulniers nickdesaulniers mentioned this issue Oct 20, 2022
Closed
ramosian-glider added a commit that referenced this issue Oct 24, 2022
@ramosian-glider
mm: kmsan: export kmsan_copy_page_meta()
97bfc08 
Certain modules call copy_user_highpage(), which calls
kmsan_copy_page_meta() under KMSAN, so we need to export the latter.

Cc: Andrew Morton <akpm@linux-foundation.org>
Fixes: b073d7f ("mm: kmsan: maintain KMSAN metadata for page operations")
Link: #89
Signed-off-by: Alexander Potapenko <glider@google.com>
ramosian-glider added a commit that referenced this issue Oct 24, 2022
@ramosian-glider
x86/purgatory: disable KMSAN instrumentation
213edc2 
The stand-alone purgatory.ro does not contain the KMSAN runtime,
therefore it can't be built with KMSAN compiler instrumentation.

Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: x86@kernel.org
Link: #89
Signed-off-by: Alexander Potapenko <glider@google.com>
ramosian-glider added a commit that referenced this issue Oct 24, 2022
@ramosian-glider
Kconfig.debug: disable CONFIG_FRAME_WARN for KMSAN by default
b8844c0 
KMSAN adds a lot of instrumentation to the code, which results in
increased stack usage (up to 2048 bytes and more in some cases).
It's hard to predict how big the stack frames can be, so we disable
the warnings for KMSAN instead.

Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: linux-kbuild@vger.kernel.org
Link: #89
Signed-off-by: Alexander Potapenko <glider@google.com>
ramosian-glider added a commit that referenced this issue Oct 24, 2022
@ramosian-glider
x86: fortify: kmsan: fix KMSAN fortify builds
4a3e741 
Ensure that KMSAN builds replace memset/memcpy/memmove calls with the
respective __msan_XXX functions, and that none of the macros are
redefined twice. This should allow building kernel with both
CONFIG_KMSAN and CONFIG_FORTIFY_SOURCE.

Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: linux-mm@kvack.org
Reported-by: Tamas K Lengyel <tamas.lengyel@zentific.com>
Link: #89
Signed-off-by: Alexander Potapenko <glider@google.com>
sj-aws pushed a commit to sjp38/linux that referenced this issue Oct 25, 2022
@ramosian-glider @akpm00
mm: kmsan: export kmsan_copy_page_meta()
dc1156e 
Certain modules call copy_user_highpage(), which calls
kmsan_copy_page_meta() under KMSAN, so we need to export the latter.

Link: https://lkml.kernel.org/r/20221024212144.2852069-1-glider@google.com
Link: google/kmsan#89
Fixes: b073d7f ("mm: kmsan: maintain KMSAN metadata for page operations")
Signed-off-by: Alexander Potapenko <glider@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
sj-aws pushed a commit to sjp38/linux that referenced this issue Oct 25, 2022
@ramosian-glider @akpm00
x86/purgatory: disable KMSAN instrumentation
1346915 
The stand-alone purgatory.ro does not contain the KMSAN runtime, therefore
it can't be built with KMSAN compiler instrumentation.

Link: https://lkml.kernel.org/r/20221024212144.2852069-2-glider@google.com
Link: google/kmsan#89
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
sj-aws pushed a commit to sjp38/linux that referenced this issue Oct 25, 2022
@ramosian-glider @akpm00
Kconfig.debug: disable CONFIG_FRAME_WARN for KMSAN by default
82ff376 
KMSAN adds a lot of instrumentation to the code, which results in
increased stack usage (up to 2048 bytes and more in some cases).  It's
hard to predict how big the stack frames can be, so we disable the
warnings for KMSAN instead.

Link: https://lkml.kernel.org/r/20221024212144.2852069-3-glider@google.com
Link: google/kmsan#89
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
sj-aws pushed a commit to sjp38/linux that referenced this issue Oct 25, 2022
@ramosian-glider @akpm00
x86: fortify: kmsan: fix KMSAN fortify builds
835da14 
Ensure that KMSAN builds replace memset/memcpy/memmove calls with the
respective __msan_XXX functions, and that none of the macros are redefined
twice.  This should allow building kernel with both CONFIG_KMSAN and
CONFIG_FORTIFY_SOURCE.

Link: https://lkml.kernel.org/r/20221024212144.2852069-5-glider@google.com
Link: google/kmsan#89
Signed-off-by: Alexander Potapenko <glider@google.com>
Reported-by: Tamas K Lengyel <tamas.lengyel@zentific.com>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Kees Cook <keescook@chromium.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
intel-lab-lkp pushed a commit to intel-lab-lkp/linux that referenced this issue Oct 26, 2022
@ramosian-glider @akpm00
mm: kmsan: export kmsan_copy_page_meta()
6968499 
Certain modules call copy_user_highpage(), which calls
kmsan_copy_page_meta() under KMSAN, so we need to export the latter.

Link: https://lkml.kernel.org/r/20221024212144.2852069-1-glider@google.com
Link: google/kmsan#89
Fixes: b073d7f ("mm: kmsan: maintain KMSAN metadata for page operations")
Signed-off-by: Alexander Potapenko <glider@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
intel-lab-lkp pushed a commit to intel-lab-lkp/linux that referenced this issue Oct 26, 2022
@ramosian-glider @akpm00
x86/purgatory: disable KMSAN instrumentation
73b01ed 
The stand-alone purgatory.ro does not contain the KMSAN runtime, therefore
it can't be built with KMSAN compiler instrumentation.

Link: https://lkml.kernel.org/r/20221024212144.2852069-2-glider@google.com
Link: google/kmsan#89
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
intel-lab-lkp pushed a commit to intel-lab-lkp/linux that referenced this issue Oct 26, 2022
@ramosian-glider @akpm00
Kconfig.debug: disable CONFIG_FRAME_WARN for KMSAN by default
1814189 
KMSAN adds a lot of instrumentation to the code, which results in
increased stack usage (up to 2048 bytes and more in some cases).  It's
hard to predict how big the stack frames can be, so we disable the
warnings for KMSAN instead.

Link: https://lkml.kernel.org/r/20221024212144.2852069-3-glider@google.com
Link: google/kmsan#89
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
intel-lab-lkp pushed a commit to intel-lab-lkp/linux that referenced this issue Oct 26, 2022
@ramosian-glider @akpm00
x86: fortify: kmsan: fix KMSAN fortify builds
31f4c40 
Ensure that KMSAN builds replace memset/memcpy/memmove calls with the
respective __msan_XXX functions, and that none of the macros are redefined
twice.  This should allow building kernel with both CONFIG_KMSAN and
CONFIG_FORTIFY_SOURCE.

Link: https://lkml.kernel.org/r/20221024212144.2852069-5-glider@google.com
Link: google/kmsan#89
Signed-off-by: Alexander Potapenko <glider@google.com>
Reported-by: Tamas K Lengyel <tamas.lengyel@zentific.com>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Kees Cook <keescook@chromium.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
sj-aws pushed a commit to sjp38/linux that referenced this issue Oct 27, 2022
@ramosian-glider @akpm00
mm: kmsan: export kmsan_copy_page_meta()
80dc6d3 
Certain modules call copy_user_highpage(), which calls
kmsan_copy_page_meta() under KMSAN, so we need to export the latter.

Link: https://lkml.kernel.org/r/20221024212144.2852069-1-glider@google.com
Link: google/kmsan#89
Fixes: b073d7f ("mm: kmsan: maintain KMSAN metadata for page operations")
Signed-off-by: Alexander Potapenko <glider@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
sj-aws pushed a commit to sjp38/linux that referenced this issue Oct 27, 2022
@ramosian-glider @akpm00
x86/purgatory: disable KMSAN instrumentation
26d11e2 
The stand-alone purgatory.ro does not contain the KMSAN runtime, therefore
it can't be built with KMSAN compiler instrumentation.

Link: https://lkml.kernel.org/r/20221024212144.2852069-2-glider@google.com
Link: google/kmsan#89
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
sj-aws pushed a commit to sjp38/linux that referenced this issue Oct 27, 2022
@ramosian-glider @akpm00
Kconfig.debug: disable CONFIG_FRAME_WARN for KMSAN by default
3772ba9 
KMSAN adds a lot of instrumentation to the code, which results in
increased stack usage (up to 2048 bytes and more in some cases).  It's
hard to predict how big the stack frames can be, so we disable the
warnings for KMSAN instead.

Link: https://lkml.kernel.org/r/20221024212144.2852069-3-glider@google.com
Link: google/kmsan#89
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
sj-aws pushed a commit to sjp38/linux that referenced this issue Oct 27, 2022
@ramosian-glider @akpm00
x86: fortify: kmsan: fix KMSAN fortify builds
388bbe1 
Ensure that KMSAN builds replace memset/memcpy/memmove calls with the
respective __msan_XXX functions, and that none of the macros are redefined
twice.  This should allow building kernel with both CONFIG_KMSAN and
CONFIG_FORTIFY_SOURCE.

Link: https://lkml.kernel.org/r/20221024212144.2852069-5-glider@google.com
Link: google/kmsan#89
Signed-off-by: Alexander Potapenko <glider@google.com>
Reported-by: Tamas K Lengyel <tamas.lengyel@zentific.com>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Kees Cook <keescook@chromium.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
ammarfaizi2 pushed a commit to ammarfaizi2/linux-fork that referenced this issue Oct 28, 2022
@ramosian-glider @akpm00
mm: kmsan: export kmsan_copy_page_meta()
f59a3ee 
Certain modules call copy_user_highpage(), which calls
kmsan_copy_page_meta() under KMSAN, so we need to export the latter.

Link: https://lkml.kernel.org/r/20221024212144.2852069-1-glider@google.com
Link: google/kmsan#89
Fixes: b073d7f ("mm: kmsan: maintain KMSAN metadata for page operations")
Signed-off-by: Alexander Potapenko <glider@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
ammarfaizi2 pushed a commit to ammarfaizi2/linux-fork that referenced this issue Oct 28, 2022
@ramosian-glider @akpm00
x86/purgatory: disable KMSAN instrumentation
42855f5 
The stand-alone purgatory.ro does not contain the KMSAN runtime, therefore
it can't be built with KMSAN compiler instrumentation.

Link: https://lkml.kernel.org/r/20221024212144.2852069-2-glider@google.com
Link: google/kmsan#89
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
ammarfaizi2 pushed a commit to ammarfaizi2/linux-fork that referenced this issue Oct 28, 2022
@ramosian-glider @akpm00
Kconfig.debug: disable CONFIG_FRAME_WARN for KMSAN by default
921757b 
KMSAN adds a lot of instrumentation to the code, which results in
increased stack usage (up to 2048 bytes and more in some cases).  It's
hard to predict how big the stack frames can be, so we disable the
warnings for KMSAN instead.

Link: https://lkml.kernel.org/r/20221024212144.2852069-3-glider@google.com
Link: google/kmsan#89
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
ammarfaizi2 pushed a commit to ammarfaizi2/linux-fork that referenced this issue Oct 28, 2022
@ramosian-glider @akpm00
x86: fortify: kmsan: fix KMSAN fortify builds
78a498c 
Ensure that KMSAN builds replace memset/memcpy/memmove calls with the
respective __msan_XXX functions, and that none of the macros are redefined
twice.  This should allow building kernel with both CONFIG_KMSAN and
CONFIG_FORTIFY_SOURCE.

Link: https://lkml.kernel.org/r/20221024212144.2852069-5-glider@google.com
Link: google/kmsan#89
Signed-off-by: Alexander Potapenko <glider@google.com>
Reported-by: Tamas K Lengyel <tamas.lengyel@zentific.com>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Kees Cook <keescook@chromium.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
ramosian-glider added a commit that referenced this issue Nov 2, 2022
@ramosian-glider
Kconfig.debug: ensure early check for KMSAN in CONFIG_KMSAN_WARN
0de9bec 
As pointed out by Masahiro Yamada, Kconfig picks up the first default
entry which has true 'if' condition. Hence, the previously added check
for KMSAN was never used, because it followed the checks for 64BIT and
!64BIT.

Put KMSAN check before others to ensure it is always applied.

Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: linux-kbuild@vger.kernel.org
Link: #89
Link: https://lore.kernel.org/linux-mm/20221024212144.2852069-3-glider@google.com/
Fixes: 921757b ("Kconfig.debug: disable CONFIG_FRAME_WARN for KMSAN by default")
Signed-off-by: Alexander Potapenko <glider@google.com>
intel-lab-lkp pushed a commit to intel-lab-lkp/linux that referenced this issue Nov 5, 2022
@ramosian-glider @akpm00
Kconfig.debug: ensure early check for KMSAN in CONFIG_KMSAN_WARN
9a969c2 
As pointed out by Masahiro Yamada, Kconfig picks up the first default
entry which has true 'if' condition.  Hence, the previously added check
for KMSAN was never used, because it followed the checks for 64BIT and
!64BIT.

Put KMSAN check before others to ensure it is always applied.

Link: https://lkml.kernel.org/r/20221102110611.1085175-3-glider@google.com
Link: google/kmsan#89
Link: https://lore.kernel.org/linux-mm/20221024212144.2852069-3-glider@google.com/
Fixes: 921757b ("Kconfig.debug: disable CONFIG_FRAME_WARN for KMSAN by default")
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Marco Elver <elver@google.com>
Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
jonhunter pushed a commit to jonhunter/linux that referenced this issue Nov 7, 2022
@ramosian-glider @akpm00
Kconfig.debug: ensure early check for KMSAN in CONFIG_KMSAN_WARN
1838835 
As pointed out by Masahiro Yamada, Kconfig picks up the first default
entry which has true 'if' condition.  Hence, the previously added check
for KMSAN was never used, because it followed the checks for 64BIT and
!64BIT.

Put KMSAN check before others to ensure it is always applied.

Link: https://lkml.kernel.org/r/20221102110611.1085175-3-glider@google.com
Link: google/kmsan#89
Link: https://lore.kernel.org/linux-mm/20221024212144.2852069-3-glider@google.com/
Fixes: 921757b ("Kconfig.debug: disable CONFIG_FRAME_WARN for KMSAN by default")
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Marco Elver <elver@google.com>
Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
ramosian-glider added a commit that referenced this issue Nov 7, 2022
@ramosian-glider
Kconfig.debug: ensure early check for KMSAN in CONFIG_KMSAN_WARN
7877c3a 
As pointed out by Masahiro Yamada, Kconfig picks up the first default
entry which has true 'if' condition. Hence, the previously added check
for KMSAN was never used, because it followed the checks for 64BIT and
!64BIT.

Put KMSAN check before others to ensure it is always applied.

Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: linux-kbuild@vger.kernel.org
Link: #89
Link: https://lore.kernel.org/linux-mm/20221024212144.2852069-3-glider@google.com/
Fixes: 921757b ("Kconfig.debug: disable CONFIG_FRAME_WARN for KMSAN by default")
Signed-off-by: Alexander Potapenko <glider@google.com>
sj-aws pushed a commit to sjp38/linux that referenced this issue Nov 8, 2022
@ramosian-glider @akpm00
Kconfig.debug: ensure early check for KMSAN in CONFIG_KMSAN_WARN
552f0dc 
As pointed out by Masahiro Yamada, Kconfig picks up the first default
entry which has true 'if' condition.  Hence, the previously added check
for KMSAN was never used, because it followed the checks for 64BIT and
!64BIT.

Put KMSAN check before others to ensure it is always applied.

Link: https://lkml.kernel.org/r/20221102110611.1085175-3-glider@google.com
Link: google/kmsan#89
Link: https://lore.kernel.org/linux-mm/20221024212144.2852069-3-glider@google.com/
Fixes: 921757b ("Kconfig.debug: disable CONFIG_FRAME_WARN for KMSAN by default")
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Marco Elver <elver@google.com>
Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
sj-aws pushed a commit to sjp38/linux that referenced this issue Nov 9, 2022
@ramosian-glider @akpm00
Kconfig.debug: ensure early check for KMSAN in CONFIG_KMSAN_WARN
205a478 
As pointed out by Masahiro Yamada, Kconfig picks up the first default
entry which has true 'if' condition.  Hence, the previously added check
for KMSAN was never used, because it followed the checks for 64BIT and
!64BIT.

Put KMSAN check before others to ensure it is always applied.

Link: https://lkml.kernel.org/r/20221102110611.1085175-3-glider@google.com
Link: google/kmsan#89
Link: https://lore.kernel.org/linux-mm/20221024212144.2852069-3-glider@google.com/
Fixes: 921757b ("Kconfig.debug: disable CONFIG_FRAME_WARN for KMSAN by default")
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Marco Elver <elver@google.com>
Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
ammarfaizi2 pushed a commit to ammarfaizi2/linux-block that referenced this issue Nov 9, 2022
@ramosian-glider @akpm00
Kconfig.debug: ensure early check for KMSAN in CONFIG_KMSAN_WARN
ac66998 
As pointed out by Masahiro Yamada, Kconfig picks up the first default
entry which has true 'if' condition.  Hence, the previously added check
for KMSAN was never used, because it followed the checks for 64BIT and
!64BIT.

Put KMSAN check before others to ensure it is always applied.

Link: https://lkml.kernel.org/r/20221102110611.1085175-3-glider@google.com
Link: google/kmsan#89
Link: https://lore.kernel.org/linux-mm/20221024212144.2852069-3-glider@google.com/
Fixes: 921757b ("Kconfig.debug: disable CONFIG_FRAME_WARN for KMSAN by default")
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Marco Elver <elver@google.com>
Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
@ramosian-glider ramosian-glider mentioned this issue Nov 10, 2022
Closed
@ramosian-glider
Copy link
Member

After setting CONFIG_FRAME_WARN=0 (we cannot override what's written in the config for that flag) in the original config, I am able to build the current KMSAN master (which is almost identical to Linux v6.1-rc4).
Yet it is unable to boot, probably there are more config flags that don't paly well with KMSAN.

@ramosian-glider
Copy link
Member

ramosian-glider commented Nov 15, 2022 •
edited
Loading

@tklengyel please note that according to your config you are using Clang 14, which does not support all the required features:

    1 #
    2 # Automatically generated file; DO NOT EDIT.
    3 # Linux/x86 6.0.0 Kernel Configuration
    4 #
    5 CONFIG_CC_VERSION_TEXT="Ubuntu clang version 14.0.0-1ubuntu1"
    6 CONFIG_GCC_VERSION=0
    7 CONFIG_CC_IS_CLANG=y
    8 CONFIG_CLANG_VERSION=140000
    9 CONFIG_AS_IS_LLVM=y
   10 CONFIG_AS_VERSION=140000
...
10502 CONFIG_HAVE_ARCH_KMSAN=y
10503 CONFIG_HAVE_KMSAN_COMPILER=y
10504 CONFIG_KMSAN=y
10505 # end of Memory Debugging

-- it does not have CONFIG_HAVE_KMSAN_PARAM_RETVAL=y or CONFIG_KMSAN_CHECK_PARAM_RETVAL=y

This shouldn't prevent KMSAN from booting though.

@tklengyel
Copy link
Contributor Author

tklengyel commented Nov 15, 2022 •
edited
Loading

@ramosian-glider Thanks, have already moved on to clang-15, this was just the backup of the config I tried first. Very first thing I tried was to move to clang-15 to see if that resolved the compile issue (it did not) ;)

@ramosian-glider
Copy link
Member

@tklengyel can you please try to disable the following configs and see if that helps?

  • CONFIG_XEN_PV=n
  • CONFIG_X86_MEM_ENCRYPT=n
  • CONFIG_HARDENED_USERCOPY=n
  • CONFIG_FRAME_WARN=0

ramosian-glider added a commit that referenced this issue Nov 18, 2022
@ramosian-glider
x86: suppress KMSAN reports in arch_within_stack_frames()
6b3059a 
arch_within_stack_frames() performs stack walking and may confuse
KMSAN by stepping on stale shadow values. To prevent false positive
reports, disable KMSAN checks in this function.

This fixes KMSAN's interoperability with CONFIG_HARDENED_USERCOPY.

Link: #89
Link: https://lore.kernel.org/lkml/Y3b9AAEKp2Vr3e6O@sol.localdomain/
Cc: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Alexander Potapenko <glider@google.com>
ashok-raj pushed a commit to ashok-raj/linux-ucode that referenced this issue Nov 20, 2022
@ramosian-glider @ashok-raj
mm: kmsan: export kmsan_copy_page_meta()
2a2633f 
Certain modules call copy_user_highpage(), which calls
kmsan_copy_page_meta() under KMSAN, so we need to export the latter.

Link: https://lkml.kernel.org/r/20221024212144.2852069-1-glider@google.com
Link: google/kmsan#89
Fixes: b073d7f ("mm: kmsan: maintain KMSAN metadata for page operations")
Signed-off-by: Alexander Potapenko <glider@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
ashok-raj pushed a commit to ashok-raj/linux-ucode that referenced this issue Nov 20, 2022
@ramosian-glider @ashok-raj
x86/purgatory: disable KMSAN instrumentation
9075b5b 
The stand-alone purgatory.ro does not contain the KMSAN runtime, therefore
it can't be built with KMSAN compiler instrumentation.

Link: https://lkml.kernel.org/r/20221024212144.2852069-2-glider@google.com
Link: google/kmsan#89
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
ashok-raj pushed a commit to ashok-raj/linux-ucode that referenced this issue Nov 20, 2022
@ramosian-glider @ashok-raj
Kconfig.debug: disable CONFIG_FRAME_WARN for KMSAN by default
b560c50 
KMSAN adds a lot of instrumentation to the code, which results in
increased stack usage (up to 2048 bytes and more in some cases).  It's
hard to predict how big the stack frames can be, so we disable the
warnings for KMSAN instead.

Link: https://lkml.kernel.org/r/20221024212144.2852069-3-glider@google.com
Link: google/kmsan#89
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
ashok-raj pushed a commit to ashok-raj/linux-ucode that referenced this issue Nov 20, 2022
@ramosian-glider @ashok-raj
x86: fortify: kmsan: fix KMSAN fortify builds
7726c3c 
Ensure that KMSAN builds replace memset/memcpy/memmove calls with the
respective __msan_XXX functions, and that none of the macros are redefined
twice.  This should allow building kernel with both CONFIG_KMSAN and
CONFIG_FORTIFY_SOURCE.

Link: https://lkml.kernel.org/r/20221024212144.2852069-5-glider@google.com
Link: google/kmsan#89
Signed-off-by: Alexander Potapenko <glider@google.com>
Reported-by: Tamas K Lengyel <tamas.lengyel@zentific.com>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Kees Cook <keescook@chromium.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
ashok-raj pushed a commit to ashok-raj/linux-ucode that referenced this issue Nov 20, 2022
@ramosian-glider @ashok-raj
Kconfig.debug: ensure early check for KMSAN in CONFIG_KMSAN_WARN
7934710 
As pointed out by Masahiro Yamada, Kconfig picks up the first default
entry which has true 'if' condition.  Hence, the previously added check
for KMSAN was never used, because it followed the checks for 64BIT and
!64BIT.

Put KMSAN check before others to ensure it is always applied.

Link: https://lkml.kernel.org/r/20221102110611.1085175-3-glider@google.com
Link: google/kmsan#89
Link: https://lore.kernel.org/linux-mm/20221024212144.2852069-3-glider@google.com/
Fixes: 921757b ("Kconfig.debug: disable CONFIG_FRAME_WARN for KMSAN by default")
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Marco Elver <elver@google.com>
Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
ramosian-glider added a commit that referenced this issue Nov 21, 2022
@ramosian-glider
x86: suppress KMSAN reports in arch_within_stack_frames()
217acf2 
arch_within_stack_frames() performs stack walking and may confuse
KMSAN by stepping on stale shadow values. To prevent false positive
reports, disable KMSAN checks in this function.

This fixes KMSAN's interoperability with CONFIG_HARDENED_USERCOPY.

Link: #89
Link: https://lore.kernel.org/lkml/Y3b9AAEKp2Vr3e6O@sol.localdomain/
Cc: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Alexander Potapenko <glider@google.com>
ramosian-glider added a commit that referenced this issue Nov 28, 2022
@ramosian-glider
x86: suppress KMSAN reports in arch_within_stack_frames()
ec83f00 
arch_within_stack_frames() performs stack walking and may confuse
KMSAN by stepping on stale shadow values. To prevent false positive
reports, disable KMSAN checks in this function.

This fixes KMSAN's interoperability with CONFIG_HARDENED_USERCOPY.

Link: #89
Link: https://lore.kernel.org/lkml/Y3b9AAEKp2Vr3e6O@sol.localdomain/
Cc: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Alexander Potapenko <glider@google.com>
ramosian-glider added a commit that referenced this issue Dec 5, 2022
@ramosian-glider
x86: suppress KMSAN reports in arch_within_stack_frames()
8c23000 
arch_within_stack_frames() performs stack walking and may confuse
KMSAN by stepping on stale shadow values. To prevent false positive
reports, disable KMSAN checks in this function.

This fixes KMSAN's interoperability with CONFIG_HARDENED_USERCOPY.

Link: #89
Link: https://lore.kernel.org/lkml/Y3b9AAEKp2Vr3e6O@sol.localdomain/
Cc: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Alexander Potapenko <glider@google.com>
zhijianli88 pushed a commit to zhijianli88/linux that referenced this issue Dec 6, 2022
@ramosian-glider @zhijianli88
Kconfig.debug: ensure early check for KMSAN in CONFIG_KMSAN_WARN
e6e51cd 
As pointed out by Masahiro Yamada, Kconfig picks up the first default
entry which has true 'if' condition.  Hence, the previously added check
for KMSAN was never used, because it followed the checks for 64BIT and
!64BIT.

Put KMSAN check before others to ensure it is always applied.

Link: https://lkml.kernel.org/r/20221102110611.1085175-3-glider@google.com
Link: google/kmsan#89
Link: https://lore.kernel.org/linux-mm/20221024212144.2852069-3-glider@google.com/
Fixes: 921757b ("Kconfig.debug: disable CONFIG_FRAME_WARN for KMSAN by default")
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Marco Elver <elver@google.com>
Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
ramosian-glider added a commit that referenced this issue Dec 12, 2022
@ramosian-glider
x86: suppress KMSAN reports in arch_within_stack_frames()
c977aed 
arch_within_stack_frames() performs stack walking and may confuse
KMSAN by stepping on stale shadow values. To prevent false positive
reports, disable KMSAN checks in this function.

This fixes KMSAN's interoperability with CONFIG_HARDENED_USERCOPY.

Link: #89
Link: https://lore.kernel.org/lkml/Y3b9AAEKp2Vr3e6O@sol.localdomain/
Cc: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Alexander Potapenko <glider@google.com>
ramosian-glider added a commit that referenced this issue Jan 23, 2023
@ramosian-glider
x86: suppress KMSAN reports in arch_within_stack_frames()
442097d 
arch_within_stack_frames() performs stack walking and may confuse
KMSAN by stepping on stale shadow values. To prevent false positive
reports, disable KMSAN checks in this function.

This fixes KMSAN's interoperability with CONFIG_HARDENED_USERCOPY.

Link: #89
Link: https://lore.kernel.org/lkml/Y3b9AAEKp2Vr3e6O@sol.localdomain/
Cc: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Alexander Potapenko <glider@google.com>
ammarfaizi2 pushed a commit to ammarfaizi2/linux-block that referenced this issue Jan 27, 2023
@ramosian-glider @hansendc
x86: Suppress KMSAN reports in arch_within_stack_frames()
ce3ba2a 
arch_within_stack_frames() performs stack walking and may confuse
KMSAN by stepping on stale shadow values. To prevent false positive
reports, disable KMSAN checks in this function.

This fixes KMSAN's interoperability with CONFIG_HARDENED_USERCOPY.

Signed-off-by: Alexander Potapenko <glider@google.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Tested-by: Eric Biggers <ebiggers@google.com>
Link: google/kmsan#89
Link: https://lore.kernel.org/lkml/Y3b9AAEKp2Vr3e6O@sol.localdomain/
Link: https://lore.kernel.org/all/20221118172305.3321253-1-glider%40google.com
ramosian-glider added a commit that referenced this issue Feb 1, 2023
@ramosian-glider
x86: suppress KMSAN reports in arch_within_stack_frames()
b98005b 
arch_within_stack_frames() performs stack walking and may confuse
KMSAN by stepping on stale shadow values. To prevent false positive
reports, disable KMSAN checks in this function.

This fixes KMSAN's interoperability with CONFIG_HARDENED_USERCOPY.

Link: #89
Link: https://lore.kernel.org/lkml/Y3b9AAEKp2Vr3e6O@sol.localdomain/
Cc: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Alexander Potapenko <glider@google.com>
ramosian-glider added a commit that referenced this issue Feb 7, 2023
@ramosian-glider
x86: suppress KMSAN reports in arch_within_stack_frames()
41112c0 
arch_within_stack_frames() performs stack walking and may confuse
KMSAN by stepping on stale shadow values. To prevent false positive
reports, disable KMSAN checks in this function.

This fixes KMSAN's interoperability with CONFIG_HARDENED_USERCOPY.

Link: #89
Link: https://lore.kernel.org/lkml/Y3b9AAEKp2Vr3e6O@sol.localdomain/
Cc: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Alexander Potapenko <glider@google.com>
ramosian-glider added a commit that referenced this issue Feb 13, 2023
@ramosian-glider
x86: suppress KMSAN reports in arch_within_stack_frames()
6a8bdc5 
arch_within_stack_frames() performs stack walking and may confuse
KMSAN by stepping on stale shadow values. To prevent false positive
reports, disable KMSAN checks in this function.

This fixes KMSAN's interoperability with CONFIG_HARDENED_USERCOPY.

Link: #89
Link: https://lore.kernel.org/lkml/Y3b9AAEKp2Vr3e6O@sol.localdomain/
Cc: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Alexander Potapenko <glider@google.com>
ramosian-glider added a commit that referenced this issue Feb 20, 2023
@ramosian-glider
x86: suppress KMSAN reports in arch_within_stack_frames()
48fab7e 
arch_within_stack_frames() performs stack walking and may confuse
KMSAN by stepping on stale shadow values. To prevent false positive
reports, disable KMSAN checks in this function.

This fixes KMSAN's interoperability with CONFIG_HARDENED_USERCOPY.

Link: #89
Link: https://lore.kernel.org/lkml/Y3b9AAEKp2Vr3e6O@sol.localdomain/
Cc: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Alexander Potapenko <glider@google.com>
@tklengyel
Copy link
Contributor Author

I just checked with the latest tip of linux (9d64600) and I no longer have this issue popping up.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@ramosian-glider @tklengyel