IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
Implement the non-faulting kernel access helpers directly instead of
abusing the uaccess routines under set_fs(KERNEL_DS).
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Add new __get_user_nocheck and __put_user_nocheck that switch on the size
and call the actual inline assembly helpers, and move the uaccess enable
/ disable into the actual __get_user and __put_user. This prepares for
natively implementing __get_kernel_nofault and __put_kernel_nofault.
Also don't bother with the deprecated register keyword for the error
return.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
This reverts commit adccfb1a805ea84d2db38eb53032533279bdaa97.
Now that the generic uaccess by mempcy code handles unaligned addresses
the generic code can be used for all RISC-V CPUs.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Put all the set_fs related code under CONFIG_SET_FS so that
asm-generic/uaccess.h can be used for set_fs-less builds.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Add native implementations of __{get,put}_kernel_nofault using
{get,put}_unaligned, just like the {get,put}_user implementations.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Instead of reusing raw_{copy,to}_from_user implement separate handlers
using {get,put}_unaligned. This ensures unaligned access is handled
correctly, and avoid the need for the small constant size optimization
in raw_{copy,to}_from_user.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Define TASK_SIZE_MAX as TASK_SIZE if not otherwise defined.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
This is a dependency for Christoph's removal of set_fs.
* 'base.set_fs' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
powerpc: remove address space overrides using set_fs()
powerpc: use non-set_fs based maccess routines
x86: remove address space overrides using set_fs()
x86: make TASK_SIZE_MAX usable from assembly code
x86: move PAGE_OFFSET, TASK_SIZE & friends to page_{32,64}_types.h
lkdtm: remove set_fs-based tests
test_bitmap: remove user bitmap tests
uaccess: add infrastructure for kernel builds with set_fs()
fs: don't allow splice read/write without explicit ops
fs: don't allow kernel reads and writes without iter ops
sysctl: Convert to iter interfaces
proc: add a read_iter method to proc proc_ops
proc: cleanup the compat vs no compat file ops
proc: remove a level of indentation in proc_get_inode
Extend the current page table dump support in RISC-V to include efi
pages as well.
Here is the output of efi runtime page table mappings.
---[ UEFI runtime start ]---
0x0000000020002000-0x0000000020003000 0x00000000be732000 4K PTE D A . . . W R V
0x0000000020018000-0x0000000020019000 0x00000000be738000 4K PTE D A . . . W R V
0x000000002002c000-0x000000002002d000 0x00000000be73c000 4K PTE D A . . . W R V
0x0000000020031000-0x0000000020032000 0x00000000bff61000 4K PTE D A . . X W R V
---[ UEFI runtime end ]---
Signed-off-by: Atish Patra <atish.patra@wdc.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Add a RISC-V architecture specific stub code that actually copies the
actual kernel image to a valid address and jump to it after boot services
are terminated. Enable UEFI related kernel configs as well for RISC-V.
Signed-off-by: Atish Patra <atish.patra@wdc.com>
Link: https://lore.kernel.org/r/20200421033336.9663-4-atish.patra@wdc.com
[ardb: - move hartid fetch into check_platform_features()
- use image_size not reserve_size
- select ISA_C
- do not use dram_base]
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Linux kernel Image can appear as an EFI application With appropriate
PE/COFF header fields in the beginning of the Image header. An EFI
application loader can directly load a Linux kernel Image and an EFI
stub residing in kernel can boot Linux kernel directly.
Add the necessary PE/COFF header.
Signed-off-by: Atish Patra <atish.patra@wdc.com>
Link: https://lore.kernel.org/r/20200421033336.9663-3-atish.patra@wdc.com
[ardb: - use C prefix for c.li to ensure the expected opcode is emitted
- align all image sections according to PE/COFF section alignment ]
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Anup Patel <anup@brainfault.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Currently, page table setup is done during setup_va_final where fixmap can
be used to create the temporary mappings. The physical frame is allocated
from memblock_alloc_* functions. However, this won't work if page table
mapping needs to be created for a different mm context (i.e. efi mm) at
a later point of time.
Use generic kernel page allocation function & macros for any mapping
after setup_vm_final.
Signed-off-by: Atish Patra <atish.patra@wdc.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Acked-by: Mike Rapoport <rppt@linux.ibm.com>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
UEFI uses early IO or memory mappings for runtime services before
normal ioremap() is usable. Add the necessary fixmap bindings and
pmd mappings for generic ioremap support to work.
Signed-off-by: Atish Patra <atish.patra@wdc.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Reviewed-by: Palmer Dabbelt <palmerdabbelt@google.com>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Currently, RISC-V reserves 1MB of fixmap memory for device tree. However,
it maps only single PMD (2MB) space for fixmap which leaves only < 1MB space
left for other kernel features such as early ioremap which requires fixmap
as well. The fixmap size can be increased by another 2MB but it brings
additional complexity and changes the virtual memory layout as well.
If we require some additional feature requiring fixmap again, it has to be
moved again.
Technically, DT doesn't need a fixmap as the memory occupied by the DT is
only used during boot. That's why, We map device tree in early page table
using two consecutive PGD mappings at lower addresses (< PAGE_OFFSET).
This frees lot of space in fixmap and also makes maximum supported
device tree size supported as PGDIR_SIZE. Thus, init memory section can be used
for the same purpose as well. This simplifies fixmap implementation.
Signed-off-by: Anup Patel <anup.patel@wdc.com>
Signed-off-by: Atish Patra <atish.patra@wdc.com>
Reviewed-by: Palmer Dabbelt <palmerdabbelt@google.com>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
The RISC-V EFI boot and runtime support will be merged for v5.10 via
the RISC-V tree. However, it incorporates some changes that conflict
with other EFI changes that are in flight, so this tag serves as a
shared base that allows those conflicts to be resolved beforehand.
-----BEGIN PGP SIGNATURE-----
iQEzBAABCgAdFiEEnNKg2mrY9zMBdeK7wjcgfpV0+n0FAl9iNdMACgkQwjcgfpV0
+n2trwgAtERh/04Islqjl0QL95wQIK6husJ8y2OGAGHoddXjvD6MJsezQ3twafhw
tr6tT8x6kyuD4NlSMeFcnaCSM2MWnUqSUdQOYQsn+HPUajnbHUKORcgcQ8tp/IQl
ItiImYb4q3dKBx5WJC8ULloiogzxaj3SZlbhBfK7/Gqh4jt2aJOT0tvnmmuyz8VS
wDKLGvVKBKv2adOm7fP/RcqlmcOPEBoYMUxBmcWbYuHfFKCZCGKnKics9ublVdmG
jK2VLHhMpqoW3CqtaS2RL9El29Np4dqbhi6TIdP5+rD/Z2gudjlLpuMlmO6ERC3s
TKDRZ6YTurfOYd1Nk9X3SBj8qGOeAw==
=PxZx
-----END PGP SIGNATURE-----
gpgsig -----BEGIN PGP SIGNATURE-----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=z9Re
-----END PGP SIGNATURE-----
Merge tag 'efi-riscv-shared-for-v5.10' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/efi/efi into for-next
Stable branch for v5.10 shared between the EFI and RISC-V trees
The RISC-V EFI boot and runtime support will be merged for v5.10 via
the RISC-V tree. However, it incorporates some changes that conflict
with other EFI changes that are in flight, so this tag serves as a
shared base that allows those conflicts to be resolved beforehand.
* tag 'efi-riscv-shared-for-v5.10' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/efi/efi:
efi/libstub: arm32: Use low allocation for the uncompressed kernel
efi/libstub: Export efi_low_alloc_above() to other units
efi/libstub: arm32: Base FDT and initrd placement on image address
efi: Rename arm-init to efi-init common for all arch
include: pe.h: Add RISC-V related PE definition
Before commit
d0f9ca9be11f25ef ("ARM: decompressor: run decompressor in place if loaded via UEFI")
we were rather limited in the choice of base address for the uncompressed
kernel, as we were relying on the logic in the decompressor that blindly
rounds down the decompressor execution address to the next multiple of 128
MiB, and decompresses the kernel there. For this reason, we have a lot of
complicated memory region handling code, to ensure that this memory window
is available, even though it could be occupied by reserved regions or
other allocations that may or may not collide with the uncompressed image.
Today, we simply pass the target address for the decompressed image to the
decompressor directly, and so we can choose a suitable window just by
finding a 16 MiB aligned region, while taking TEXT_OFFSET and the region
for the swapper page tables into account.
So let's get rid of the complicated logic, and instead, use the existing
bottom up allocation routine to allocate a suitable window as low as
possible, and carve out a memory region that has the right properties.
Note that this removes any dependencies on the 'dram_base' argument to
handle_kernel_image(), and so this is removed as well. Given that this
was the only remaining use of dram_base, the code that produces it is
removed entirely as well.
Reviewed-by: Maxim Uvarov <maxim.uvarov@linaro.org>
Tested-by: Maxim Uvarov <maxim.uvarov@linaro.org>
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Permit arm32-stub.c to access efi_low_alloc_above() in a subsequent
patch by giving it external linkage and declaring it in efistub.h.
Reviewed-by: Maxim Uvarov <maxim.uvarov@linaro.org>
Tested-by: Maxim Uvarov <maxim.uvarov@linaro.org>
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
The way we use the base of DRAM in the EFI stub is problematic as it
is ill defined what the base of DRAM actually means. There are some
restrictions on the placement of FDT and initrd which are defined in
terms of dram_base, but given that the placement of the kernel in
memory is what defines these boundaries (as on ARM, this is where the
linear region starts), it is better to use the image address in these
cases, and disregard dram_base altogether.
Reviewed-by: Maxim Uvarov <maxim.uvarov@linaro.org>
Tested-by: Maxim Uvarov <maxim.uvarov@linaro.org>
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
asm/thread_info.h is included more than once, Remove the one that isn't
necessary.
Signed-off-by: Tian Tao <tiantao6@hisilicon.com>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
If the page fault "cause" is EXC_INST_PAGE_FAULT, set the
FAULT_FLAG_INSTRUCTION flag to let handle_mm_fault() and friends know
about it. This has no functional changes because RISC-V uses the default
arch_vma_access_permitted() implementation, which always returns true.
However, dax_pmd_fault(), for example, has a tracepoint that uses
FAULT_FLAG_INSTRUCTION, so we might as well set it.
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
The "inline" keyword is in the wrong place in vmalloc_fault()
declaration:
>> arch/riscv/mm/fault.c:56:1: warning: 'inline' is not at beginning of declaration [-Wold-style-declaration]
56 | static void inline vmalloc_fault(struct pt_regs *regs, int code, unsigned long addr)
| ^~~~~~
Fix that up.
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
There are no standard CSR registers to provide cache information, the
way for RISC-V is to get this information from DT. Currently, AT_L1I_X,
AT_L1D_X and AT_L2_X are present in glibc header, and sysconf syscall
could use them to get information of cache through AUX vector.
The result of 'getconf -a' as follows:
LEVEL1_ICACHE_SIZE 32768
LEVEL1_ICACHE_ASSOC 8
LEVEL1_ICACHE_LINESIZE 64
LEVEL1_DCACHE_SIZE 32768
LEVEL1_DCACHE_ASSOC 8
LEVEL1_DCACHE_LINESIZE 64
LEVEL2_CACHE_SIZE 2097152
LEVEL2_CACHE_ASSOC 32
LEVEL2_CACHE_LINESIZE 64
Signed-off-by: Zong Li <zong.li@sifive.com>
Reviewed-by: Palmer Dabbelt <palmerdabbelt@google.com>
Reviewed-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
AT_VECTOR_SIZE_ARCH should be defined with the maximum number of
NEW_AUX_ENT entries that ARCH_DLINFO can contain, but it wasn't defined
for RISC-V at all even though ARCH_DLINFO will contain one NEW_AUX_ENT
for the VDSO address.
Signed-off-by: Zong Li <zong.li@sifive.com>
Reviewed-by: Palmer Dabbelt <palmerdabbelt@google.com>
Reviewed-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Set cacheinfo.{size,sets,line_size} for each cache node, then we can
get these information from userland through auxiliary vector.
Signed-off-by: Zong Li <zong.li@sifive.com>
Reviewed-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Move the access error check into a access_error() function to simplify
the control flow in do_page_fault().
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Let's handle the translation of EXC_STORE_PAGE_FAULT to FAULT_FLAG_WRITE
once before looking up the VMA. This makes it easier to extract access
error logic in the next patch.
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Simplify the mm_fault_error() handling function by eliminating the
unnecessary gotos.
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
This patch moves the fault error handling to mm_fault_error() function
and converts gotos to calls to the new function.
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Move fault error handling after retry logic. This simplifies the code
flow and makes it easier to move fault error handling to its own
function.
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
This patch moves the vmalloc fault handling in do_page_fault() to
vmalloc_fault() function and converts gotos to calls to the new
function.
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
This patch moves the bad area handling in do_page_fault() to bad_area()
function and converts gotos to calls to the new function.
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
This patch moves the no context handling in do_page_fault() to
no_context() function and converts gotos to calls to the new function.
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Let's combine the two retry logic if statements in do_page_fault() to
simplify the code.
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
arm-init is responsible for setting up efi runtime and doesn't actually
do any ARM specific stuff. RISC-V can use the same source code as it is.
Rename it to efi-init so that RISC-V can use it.
Signed-off-by: Atish Patra <atish.patra@wdc.com>
Link: https://lore.kernel.org/r/20200819222425.30721-8-atish.patra@wdc.com
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Stop providing the possibility to override the address space using
set_fs() now that there is no need for that any more.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Provide __get_kernel_nofault and __put_kernel_nofault routines to
implement the maccess routines without messing with set_fs and without
opening up access to user space.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Stop providing the possibility to override the address space using
set_fs() now that there is no need for that any more. To properly
handle the TASK_SIZE_MAX checking for 4 vs 5-level page tables on
x86 a new alternative is introduced, which just like the one in
entry_64.S has to use the hardcoded virtual address bits to escape
the fact that TASK_SIZE_MAX isn't actually a constant when 5-level
page tables are enabled.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
For 64-bit the only thing missing was a strategic _AC, and for 32-bit we
need to use __PAGE_OFFSET instead of PAGE_OFFSET in the TASK_SIZE
definition to escape the explicit unsigned long cast. This just works
because __PAGE_OFFSET is defined using _AC itself and thus never needs
the cast anyway.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
At least for 64-bit this moves them closer to some of the defines
they are based on, and it prepares for using the TASK_SIZE_MAX
definition from assembly.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Once we can't manipulate the address limit, we also can't test what
happens when the manipulation is abused.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
We can't run the tests for userspace bitmap parsing if set_fs() doesn't
exist, and it is about to go away for x86, powerpc with other major
architectures to follow.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Add a CONFIG_SET_FS option that is selected by architecturess that
implement set_fs, which is all of them initially. If the option is not
set stubs for routines related to overriding the address space are
provided so that architectures can start to opt out of providing set_fs.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
default_file_splice_write is the last piece of generic code that uses
set_fs to make the uaccess routines operate on kernel pointers. It
implements a "fallback loop" for splicing from files that do not actually
provide a proper splice_read method. The usual file systems and other
high bandwidth instances all provide a ->splice_read, so this just removes
support for various device drivers and procfs/debugfs files. If splice
support for any of those turns out to be important it can be added back
by switching them to the iter ops and using generic_file_splice_read.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Don't allow calling ->read or ->write with set_fs as a preparation for
killing off set_fs. All the instances that we use kernel_read/write on
are using the iter ops already.
If a file has both the regular ->read/->write methods and the iter
variants those could have different semantics for messed up enough
drivers. Also fails the kernel access to them in that case.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Using the read_iter/write_iter interfaces allows for in-kernel users
to set sysctls without using set_fs(). Also, the buffer is a string,
so give it the real type of 'char *', not void *.
[AV: Christoph's fixup folded in]
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
This will allow proc files to implement iter read semantics.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Instead of providing a special no-compat version provide a special
compat version for operations with ->compat_ioctl.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
