linux-toradex.git/fs/proc/kcore.c, branch v6.12

fs/proc/kcore.c: allow translation of physical memory addresses

2024-10-09T19:47:19+00:00

When /proc/kcore is read an attempt to read the first two pages results in
HW-specific page swap on s390 and another (so called prefix) pages are
accessed instead.  That leads to a wrong read.

Allow architecture-specific translation of memory addresses using
kc_xlate_dev_mem_ptr() and kc_unxlate_dev_mem_ptr() callbacks similarily
to /dev/mem xlate_dev_mem_ptr() and unxlate_dev_mem_ptr() callbacks.  That
way an architecture can deal with specific physical memory ranges.

Re-use the existing /dev/mem callback implementation on s390, which
handles the described prefix pages swapping correctly.

For other architectures the default callback is basically NOP.  It is
expected the condition (vaddr == __va(__pa(vaddr))) always holds true for
KCORE_RAM memory type.

Link: https://lkml.kernel.org/r/20240930122119.1651546-1-agordeev@linux.ibm.com
Signed-off-by: Alexander Gordeev 
Suggested-by: Heiko Carstens 
Cc: Vasily Gorbik 
Cc: 
Signed-off-by: Andrew Morton

proc: Fix typo in the comment

2024-09-09T07:51:16+00:00

The deference here confuses me.

Maybe here want to say that because show_fd_locks() does not dereference
the files pointer, using the stale value of the files pointer is safe.

Correctly spelled comments make it easier for the reader to understand
the code.

replace 'deferences' with 'dereferences' in the comment &
replace 'inialized' with 'initialized' in the comment.

Signed-off-by: Yan Zhen 
Link: https://lore.kernel.org/r/20240909063353.2246419-1-yanzhen@vivo.com
Signed-off-by: Christian Brauner

crash: split vmcoreinfo exporting code out from crash_core.c

2024-02-24T01:48:22+00:00

Now move the relevant codes into separate files:
kernel/crash_reserve.c, include/linux/crash_reserve.h.

And add config item CRASH_RESERVE to control its enabling.

And also update the old ifdeffery of CONFIG_CRASH_CORE, including of
 and config item dependency on CRASH_CORE
accordingly.

And also do renaming as follows:
 - arch/xxx/kernel/{crash_core.c => vmcore_info.c}
because they are only related to vmcoreinfo exporting on x86, arm64,
riscv.

And also Remove config item CRASH_CORE, and rely on CONFIG_KEXEC_CORE to
decide if build in crash_core.c.

[yang.lee@linux.alibaba.com: remove duplicated include in vmcore_info.c]
  Link: https://lkml.kernel.org/r/20240126005744.16561-1-yang.lee@linux.alibaba.com
Link: https://lkml.kernel.org/r/20240124051254.67105-3-bhe@redhat.com
Signed-off-by: Baoquan He 
Signed-off-by: Yang Li 
Acked-by: Hari Bathini 
Cc: Al Viro 
Cc: Eric W. Biederman 
Cc: Pingfan Liu 
Cc: Klara Modin 
Cc: Michael Kelley 
Cc: Nathan Chancellor 
Cc: Stephen Rothwell 
Cc: Yang Li 
Signed-off-by: Andrew Morton

proc/kcore: do not try to access unaccepted memory

2023-10-04T17:32:22+00:00

Support for unaccepted memory was added recently, refer commit
dcdfdd40fa82 ("mm: Add support for unaccepted memory"), whereby a virtual
machine may need to accept memory before it can be used.

Do not try to access unaccepted memory because it can cause the guest to
fail.

For /proc/kcore, which is read-only and does not support mmap, this means a
read of unaccepted memory will return zeros.

Link: https://lkml.kernel.org/r/20230911112114.91323-3-adrian.hunter@intel.com
Signed-off-by: Adrian Hunter 
Reviewed-by: David Hildenbrand 
Cc: Ard Biesheuvel 
Cc: Baoquan He 
Cc: Borislav Petkov (AMD) 
Cc: Dave Hansen 
Cc: Dave Young 
Cc: Kirill A. Shutemov 
Cc: Lorenzo Stoakes 
Cc: Mike Rapoport 
Cc: Tom Lendacky 
Cc: Vivek Goyal 
Cc: Vlastimil Babka 
Signed-off-by: Andrew Morton

fs/proc/kcore: reinstate bounce buffer for KCORE_TEXT regions

2023-08-04T20:03:42+00:00

Some architectures do not populate the entire range categorised by
KCORE_TEXT, so we must ensure that the kernel address we read from is
valid.

Unfortunately there is no solution currently available to do so with a
purely iterator solution so reinstate the bounce buffer in this instance
so we can use copy_from_kernel_nofault() in order to avoid page faults
when regions are unmapped.

This change partly reverts commit 2e1c0170771e ("fs/proc/kcore: avoid
bounce buffer for ktext data"), reinstating the bounce buffer, but adapts
the code to continue to use an iterator.

[lstoakes@gmail.com: correct comment to be strictly correct about reasoning]
  Link: https://lkml.kernel.org/r/525a3f14-74fa-4c22-9fca-9dab4de8a0c3@lucifer.local
Link: https://lkml.kernel.org/r/20230731215021.70911-1-lstoakes@gmail.com
Fixes: 2e1c0170771e ("fs/proc/kcore: avoid bounce buffer for ktext data")
Signed-off-by: Lorenzo Stoakes 
Reported-by: Jiri Olsa 
Closes: https://lore.kernel.org/all/ZHc2fm+9daF6cgCE@krava
Tested-by: Jiri Olsa 
Tested-by: Will Deacon 
Cc: Alexander Viro 
Cc: Ard Biesheuvel 
Cc: Baoquan He 
Cc: Catalin Marinas 
Cc: David Hildenbrand 
Cc: Jens Axboe 
Cc: Kefeng Wang 
Cc: Liu Shixin 
Cc: Matthew Wilcox 
Cc: Mike Galbraith 
Cc: Thorsten Leemhuis 
Cc: Uladzislau Rezki (Sony) 
Cc: 
Signed-off-by: Andrew Morton

Merge tag 'asm-generic-6.5' of git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic

2023-07-06T17:06:04+00:00

Pull asm-generic updates from Arnd Bergmann:
 "These are cleanups for architecture specific header files:

   - the comments in include/linux/syscalls.h have gone out of sync and
     are really pointless, so these get removed

   - The asm/bitsperlong.h header no longer needs to be architecture
     specific on modern compilers, so use a generic version for newer
     architectures that use new enough userspace compilers

   - A cleanup for virt_to_pfn/virt_to_bus to have proper type checking,
     forcing the use of pointers"

* tag 'asm-generic-6.5' of git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic:
  syscalls: Remove file path comments from headers
  tools arch: Remove uapi bitsperlong.h of hexagon and microblaze
  asm-generic: Unify uapi bitsperlong.h for arm64, riscv and loongarch
  m68k/mm: Make pfn accessors static inlines
  arm64: memory: Make virt_to_pfn() a static inline
  ARM: mm: Make virt_to_pfn() a static inline
  asm-generic/page.h: Make pfn accessors static inlines
  xen/netback: Pass (void *) to virt_to_page()
  netfs: Pass a pointer to virt_to_page()
  cifs: Pass a pointer to virt_to_page() in cifsglob
  cifs: Pass a pointer to virt_to_page()
  riscv: mm: init: Pass a pointer to virt_to_page()
  ARC: init: Pass a pointer to virt_to_pfn() in init
  m68k: Pass a pointer to virt_to_pfn() virt_to_page()
  fs/proc/kcore.c: Pass a pointer to virt_addr_valid()

procfs: replace all non-returning strlcpy with strscpy

2023-06-10T00:44:13+00:00

strlcpy() reads the entire source buffer first.  This read may exceed the
destination size limit.  This is both inefficient and can lead to linear
read overflows if a source string is not NUL-terminated [1].  In an effort
to remove strlcpy() completely [2], replace strlcpy() here with strscpy().
No return values were used, so direct replacement is safe.

[1] https://www.kernel.org/doc/html/latest/process/deprecated.html#strlcpy
[2] https://github.com/KSPP/linux/issues/89

Link: https://lkml.kernel.org/r/20230510212457.3491385-1-azeemshaikh38@gmail.com
Signed-off-by: Azeem Shaikh 
Reviewed-by: Kees Cook 
Reviewed-by: Alexey Dobriyan 
Cc: Baoquan He 
Cc: David Hildenbrand 
Cc: Kefeng Wang 
Cc: Liu Shixin 
Cc: Lorenzo Stoakes 
Signed-off-by: Andrew Morton

fs/proc/kcore.c: Pass a pointer to virt_addr_valid()

2023-05-29T09:27:07+00:00

The virt_addr_valid() should be passed a pointer, the current
code passing a long unsigned int is just exploiting the
unintentional polymorphism of these calls being implemented
as preprocessor macros.

Signed-off-by: Linus Walleij

mm: vmalloc: convert vread() to vread_iter()

2023-04-06T02:42:57+00:00

Having previously laid the foundation for converting vread() to an
iterator function, pull the trigger and do so.

This patch attempts to provide minimal refactoring and to reflect the
existing logic as best we can, for example we continue to zero portions of
memory not read, as before.

Overall, there should be no functional difference other than a performance
improvement in /proc/kcore access to vmalloc regions.

Now we have eliminated the need for a bounce buffer in read_kcore_iter(),
we dispense with it, and try to write to user memory optimistically but
with faults disabled via copy_page_to_iter_nofault().  We already have
preemption disabled by holding a spin lock.  We continue faulting in until
the operation is complete.

Additionally, we must account for the fact that at any point a copy may
fail (most likely due to a fault not being able to occur), we exit
indicating fewer bytes retrieved than expected.

[sfr@canb.auug.org.au: fix sparc64 warning]
  Link: https://lkml.kernel.org/r/20230320144721.663280c3@canb.auug.org.au
[lstoakes@gmail.com: redo Stephen's sparc build fix]
  Link: https://lkml.kernel.org/r/8506cbc667c39205e65a323f750ff9c11a463798.1679566220.git.lstoakes@gmail.com
[akpm@linux-foundation.org: unbreak uio.h includes]
Link: https://lkml.kernel.org/r/941f88bc5ab928e6656e1e2593b91bf0f8c81e1b.1679511146.git.lstoakes@gmail.com
Signed-off-by: Lorenzo Stoakes 
Signed-off-by: Stephen Rothwell 
Reviewed-by: Baoquan He 
Cc: Alexander Viro 
Cc: David Hildenbrand 
Cc: Jens Axboe 
Cc: Jiri Olsa 
Cc: Liu Shixin 
Cc: Matthew Wilcox (Oracle) 
Cc: Uladzislau Rezki (Sony) 
Signed-off-by: Andrew Morton

fs/proc/kcore: convert read_kcore() to read_kcore_iter()

2023-04-06T02:42:56+00:00

For the time being we still use a bounce buffer for vread(), however in
the next patch we will convert this to interact directly with the iterator
and eliminate the bounce buffer altogether.

Link: https://lkml.kernel.org/r/ebe12c8d70eebd71f487d80095605f3ad0d1489c.1679511146.git.lstoakes@gmail.com
Signed-off-by: Lorenzo Stoakes 
Reviewed-by: David Hildenbrand 
Reviewed-by: Baoquan He 
Cc: Alexander Viro 
Cc: Jens Axboe 
Cc: Jiri Olsa 
Cc: Liu Shixin 
Cc: Matthew Wilcox (Oracle) 
Cc: Uladzislau Rezki (Sony) 
Signed-off-by: Andrew Morton