kernel/linux.git/fs/binfmt_elf.c, branch v6.6.131

binfmt_elf: Move brk for static PIE even if ASLR disabled

2025-05-22T12:12:12+00:00

[ Upstream commit 11854fe263eb1b9a8efa33b0c087add7719ea9b4 ] In commit bbdc6076d2e5 ("binfmt_elf: move brk out of mmap when doing direct loader exec"), the brk was moved out of the mmap region when loading static PIE binaries (ET_DYN without INTERP). The common case for these binaries was testing new ELF loaders, so the brk needed to be away from mmap to avoid colliding with stack, future mmaps (of the loader-loaded binary), etc. But this was only done when ASLR was enabled, in an attempt to minimize changes to memory layouts. After adding support to respect alignment requirements for static PIE binaries in commit 3545deff0ec7 ("binfmt_elf: Honor PT_LOAD alignment for static PIE"), it became possible to have a large gap after the final PT_LOAD segment and the top of the mmap region. This means that future mmap allocations might go after the last PT_LOAD segment (where brk might be if ASLR was disabled) instead of before them (where they traditionally ended up). On arm64, running with ASLR disabled, Ubuntu 22.04's "ldconfig" binary, a static PIE, has alignment requirements that leaves a gap large enough after the last PT_LOAD segment to fit the vdso and vvar, but still leave enough space for the brk (which immediately follows the last PT_LOAD segment) to be allocated by the binary. fffff7f20000-fffff7fde000 r-xp 00000000 fe:02 8110426 /sbin/ldconfig.real fffff7fee000-fffff7ff5000 rw-p 000be000 fe:02 8110426 /sbin/ldconfig.real fffff7ff5000-fffff7ffa000 rw-p 00000000 00:00 0 ***[brk will go here at fffff7ffa000]*** fffff7ffc000-fffff7ffe000 r--p 00000000 00:00 0 [vvar] fffff7ffe000-fffff8000000 r-xp 00000000 00:00 0 [vdso] fffffffdf000-1000000000000 rw-p 00000000 00:00 0 [stack] After commit 0b3bc3354eb9 ("arm64: vdso: Switch to generic storage implementation"), the arm64 vvar grew slightly, and suddenly the brk collided with the allocation. fffff7f20000-fffff7fde000 r-xp 00000000 fe:02 8110426 /sbin/ldconfig.real fffff7fee000-fffff7ff5000 rw-p 000be000 fe:02 8110426 /sbin/ldconfig.real fffff7ff5000-fffff7ffa000 rw-p 00000000 00:00 0 ***[oops, no room any more, vvar is at fffff7ffa000!]*** fffff7ffa000-fffff7ffe000 r--p 00000000 00:00 0 [vvar] fffff7ffe000-fffff8000000 r-xp 00000000 00:00 0 [vdso] fffffffdf000-1000000000000 rw-p 00000000 00:00 0 [stack] The solution is to unconditionally move the brk out of the mmap region for static PIE binaries. Whether ASLR is enabled or not does not change if there may be future mmap allocation collisions with a growing brk region. Update memory layout comments (with kernel-doc headings), consolidate the setting of mm->brk to later (it isn't needed early), move static PIE brk out of mmap unconditionally, and make sure brk(2) knows to base brk position off of mm->start_brk not mm->end_data no matter what the cause of moving it is (via current->brk_randomized). For the CONFIG_COMPAT_BRK case, though, leave the logic unchanged, as we can never safely move the brk. These systems, however, are not using specially aligned static PIE binaries. Reported-by: Ryan Roberts Closes: https://lore.kernel.org/lkml/f93db308-4a0e-4806-9faf-98f890f5a5e6@arm.com/ Fixes: bbdc6076d2e5 ("binfmt_elf: move brk out of mmap when doing direct loader exec") Link: https://lore.kernel.org/r/20250425224502.work.520-kees@kernel.org Reviewed-by: Ryan Roberts Tested-by: Ryan Roberts Signed-off-by: Kees Cook Signed-off-by: Sasha Levin

binfmt_elf: Honor PT_LOAD alignment for static PIE

2025-05-22T12:12:11+00:00

[ Upstream commit 3545deff0ec7a37de7ed9632e262598582b140e9 ] The p_align values in PT_LOAD were ignored for static PIE executables (i.e. ET_DYN without PT_INTERP). This is because there is no way to request a non-fixed mmap region with a specific alignment. ET_DYN with PT_INTERP uses a separate base address (ELF_ET_DYN_BASE) and binfmt_elf performs the ASLR itself, which means it can also apply alignment. For the mmap region, the address selection happens deep within the vm_mmap() implementation (when the requested address is 0). The earlier attempt to implement this: commit 9630f0d60fec ("fs/binfmt_elf: use PT_LOAD p_align values for static PIE") commit 925346c129da ("fs/binfmt_elf: fix PT_LOAD p_align values for loaders") did not take into account the different base address origins, and were eventually reverted: aeb7923733d1 ("revert "fs/binfmt_elf: use PT_LOAD p_align values for static PIE"") In order to get the correct alignment from an mmap base, binfmt_elf must perform a 0-address load first, then tear down the mapping and perform alignment on the resulting address. Since this is slightly more overhead, only do this when it is needed (i.e. the alignment is not the default ELF alignment). This does, however, have the benefit of being able to use MAP_FIXED_NOREPLACE, to avoid potential collisions. With this fixed, enable the static PIE self tests again. Reported-by: H.J. Lu Closes: https://bugzilla.kernel.org/show_bug.cgi?id=215275 Link: https://lore.kernel.org/r/20240508173149.677910-3-keescook@chromium.org Signed-off-by: Kees Cook Stable-dep-of: 11854fe263eb ("binfmt_elf: Move brk for static PIE even if ASLR disabled") Signed-off-by: Sasha Levin

binfmt_elf: Calculate total_size earlier

2025-05-22T12:12:11+00:00

[ Upstream commit 2d4cf7b190bbfadd4986bf5c34da17c1a88adf8e ] In preparation to support PT_LOAD with large p_align values on non-PT_INTERP ET_DYN executables (i.e. "static pie"), we'll need to use the total_size details earlier. Move this separately now to make the next patch more readable. As total_size and load_bias are currently calculated separately, this has no behavioral impact. Link: https://lore.kernel.org/r/20240508173149.677910-2-keescook@chromium.org Signed-off-by: Kees Cook Stable-dep-of: 11854fe263eb ("binfmt_elf: Move brk for static PIE even if ASLR disabled") Signed-off-by: Sasha Levin

binfmt_elf: Leave a gap between .bss and brk

2025-05-22T12:12:11+00:00

[ Upstream commit 2a5eb9995528441447d33838727f6ec1caf08139 ] Currently the brk starts its randomization immediately after .bss, which means there is a chance that when the random offset is 0, linear overflows from .bss can reach into the brk area. Leave at least a single page gap between .bss and brk (when it has not already been explicitly relocated into the mmap range). Reported-by: Closes: https://lore.kernel.org/linux-hardening/CA+2EKTVLvc8hDZc+2Yhwmus=dzOUG5E4gV7ayCbu0MPJTZzWkw@mail.gmail.com/ Link: https://lore.kernel.org/r/20240217062545.1631668-2-keescook@chromium.org Signed-off-by: Kees Cook Stable-dep-of: 11854fe263eb ("binfmt_elf: Move brk for static PIE even if ASLR disabled") Signed-off-by: Sasha Levin

binfmt_elf: elf_bss no longer used by load_elf_binary()

2025-05-22T12:12:11+00:00

[ Upstream commit 8ed2ef21ff564cf4a25c098ace510ee6513c9836 ] With the BSS handled generically via the new filesz/memsz mismatch handling logic in elf_load(), elf_bss no longer needs to be tracked. Drop the variable. Cc: Eric Biederman Cc: Alexander Viro Cc: Christian Brauner Cc: linux-fsdevel@vger.kernel.org Cc: linux-mm@kvack.org Suggested-by: Eric Biederman Tested-by: Pedro Falcato Signed-off-by: Sebastian Ott Link: https://lore.kernel.org/r/20230929032435.2391507-2-keescook@chromium.org Signed-off-by: Kees Cook Stable-dep-of: 11854fe263eb ("binfmt_elf: Move brk for static PIE even if ASLR disabled") Signed-off-by: Sasha Levin

binfmt_elf: Support segments with 0 filesz and misaligned starts

2025-05-22T12:12:11+00:00

[ Upstream commit 585a018627b4d7ed37387211f667916840b5c5ea ] Implement a helper elf_load() that wraps elf_map() and performs all of the necessary work to ensure that when "memsz > filesz" the bytes described by "memsz > filesz" are zeroed. An outstanding issue is if the first segment has filesz 0, and has a randomized location. But that is the same as today. In this change I replaced an open coded padzero() that did not clear all of the way to the end of the page, with padzero() that does. I also stopped checking the return of padzero() as there is at least one known case where testing for failure is the wrong thing to do. It looks like binfmt_elf_fdpic may have the proper set of tests for when error handling can be safely completed. I found a couple of commits in the old history https://git.kernel.org/pub/scm/linux/kernel/git/tglx/history.git, that look very interesting in understanding this code. commit 39b56d902bf3 ("[PATCH] binfmt_elf: clearing bss may fail") commit c6e2227e4a3e ("[SPARC64]: Missing user access return value checks in fs/binfmt_elf.c and fs/compat.c") commit 5bf3be033f50 ("v2.4.10.1 -> v2.4.10.2") Looking at commit 39b56d902bf3 ("[PATCH] binfmt_elf: clearing bss may fail"): > commit 39b56d902bf35241e7cba6cc30b828ed937175ad > Author: Pavel Machek > Date: Wed Feb 9 22:40:30 2005 -0800 > > [PATCH] binfmt_elf: clearing bss may fail > > So we discover that Borland's Kylix application builder emits weird elf > files which describe a non-writeable bss segment. > > So remove the clear_user() check at the place where we zero out the bss. I > don't _think_ there are any security implications here (plus we've never > checked that clear_user() return value, so whoops if it is a problem). > > Signed-off-by: Pavel Machek > Signed-off-by: Andrew Morton > Signed-off-by: Linus Torvalds It seems pretty clear that binfmt_elf_fdpic with skipping clear_user() for non-writable segments and otherwise calling clear_user(), aka padzero(), and checking it's return code is the right thing to do. I just skipped the error checking as that avoids breaking things. And notably, it looks like Borland's Kylix died in 2005 so it might be safe to just consider read-only segments with memsz > filesz an error. Reported-by: Sebastian Ott Reported-by: Thomas Weißschuh Closes: https://lkml.kernel.org/r/20230914-bss-alloc-v1-1-78de67d2c6dd@weissschuh.net Signed-off-by: "Eric W. Biederman" Link: https://lore.kernel.org/r/87sf71f123.fsf@email.froward.int.ebiederm.org Tested-by: Pedro Falcato Signed-off-by: Sebastian Ott Link: https://lore.kernel.org/r/20230929032435.2391507-1-keescook@chromium.org Signed-off-by: Kees Cook Stable-dep-of: 11854fe263eb ("binfmt_elf: Move brk for static PIE even if ASLR disabled") Signed-off-by: Sasha Levin

ELF: fix kernel.randomize_va_space double read

2024-09-12T09:11:29+00:00

[ Upstream commit 2a97388a807b6ab5538aa8f8537b2463c6988bd2 ] ELF loader uses "randomize_va_space" twice. It is sysctl and can change at any moment, so 2 loads could see 2 different values in theory with unpredictable consequences. Issue exactly one load for consistent value across one exec. Signed-off-by: Alexey Dobriyan Link: https://lore.kernel.org/r/3329905c-7eb8-400a-8f0a-d87cff979b5b@p183 Signed-off-by: Kees Cook Signed-off-by: Sasha Levin

Merge branch 'expand-stack'

2023-06-29T03:35:21+00:00

This modifies our user mode stack expansion code to always take the mmap_lock for writing before modifying the VM layout. It's actually something we always technically should have done, but because we didn't strictly need it, we were being lazy ("opportunistic" sounds so much better, doesn't it?) about things, and had this hack in place where we would extend the stack vma in-place without doing the proper locking. And it worked fine. We just needed to change vm_start (or, in the case of grow-up stacks, vm_end) and together with some special ad-hoc locking using the anon_vma lock and the mm->page_table_lock, it all was fairly straightforward. That is, it was all fine until Ruihan Li pointed out that now that the vma layout uses the maple tree code, we *really* don't just change vm_start and vm_end any more, and the locking really is broken. Oops. It's not actually all _that_ horrible to fix this once and for all, and do proper locking, but it's a bit painful. We have basically three different cases of stack expansion, and they all work just a bit differently: - the common and obvious case is the page fault handling. It's actually fairly simple and straightforward, except for the fact that we have something like 24 different versions of it, and you end up in a maze of twisty little passages, all alike. - the simplest case is the execve() code that creates a new stack. There are no real locking concerns because it's all in a private new VM that hasn't been exposed to anybody, but lockdep still can end up unhappy if you get it wrong. - and finally, we have GUP and page pinning, which shouldn't really be expanding the stack in the first place, but in addition to execve() we also use it for ptrace(). And debuggers do want to possibly access memory under the stack pointer and thus need to be able to expand the stack as a special case. None of these cases are exactly complicated, but the page fault case in particular is just repeated slightly differently many many times. And ia64 in particular has a fairly complicated situation where you can have both a regular grow-down stack _and_ a special grow-up stack for the register backing store. So to make this slightly more manageable, the bulk of this series is to first create a helper function for the most common page fault case, and convert all the straightforward architectures to it. Thus the new 'lock_mm_and_find_vma()' helper function, which ends up being used by x86, arm, powerpc, mips, riscv, alpha, arc, csky, hexagon, loongarch, nios2, sh, sparc32, and xtensa. So we not only convert more than half the architectures, we now have more shared code and avoid some of those twisty little passages. And largely due to this common helper function, the full diffstat of this series ends up deleting more lines than it adds. That still leaves eight architectures (ia64, m68k, microblaze, openrisc, parisc, s390, sparc64 and um) that end up doing 'expand_stack()' manually because they are doing something slightly different from the normal pattern. Along with the couple of special cases in execve() and GUP. So there's a couple of patches that first create 'locked' helper versions of the stack expansion functions, so that there's a obvious path forward in the conversion. The execve() case is then actually pretty simple, and is a nice cleanup from our old "grow-up stackls are special, because at execve time even they grow down". The #ifdef CONFIG_STACK_GROWSUP in that code just goes away, because it's just more straightforward to write out the stack expansion there manually, instead od having get_user_pages_remote() do it for us in some situations but not others and have to worry about locking rules for GUP. And the final step is then to just convert the remaining odd cases to a new world order where 'expand_stack()' is called with the mmap_lock held for reading, but where it might drop it and upgrade it to a write, only to return with it held for reading (in the success case) or with it completely dropped (in the failure case). In the process, we remove all the stack expansion from GUP (where dropping the lock wouldn't be ok without special rules anyway), and add it in manually to __access_remote_vm() for ptrace(). Thanks to Adrian Glaubitz and Frank Scheiner who tested the ia64 cases. Everything else here felt pretty straightforward, but the ia64 rules for stack expansion are really quite odd and very different from everything else. Also thanks to Vegard Nossum who caught me getting one of those odd conditions entirely the wrong way around. Anyway, I think I want to actually move all the stack expansion code to a whole new file of its own, rather than have it split up between mm/mmap.c and mm/memory.c, but since this will have to be backported to the initial maple tree vma introduction anyway, I tried to keep the patches _fairly_ minimal. Also, while I don't think it's valid to expand the stack from GUP, the final patch in here is a "warn if some crazy GUP user wants to try to expand the stack" patch. That one will be reverted before the final release, but it's left to catch any odd cases during the merge window and release candidates. Reported-by: Ruihan Li * branch 'expand-stack': gup: add warning if some caller would seem to want stack expansion mm: always expand the stack with the mmap write lock held execve: expand new process stack manually ahead of time mm: make find_extend_vma() fail if write lock not held powerpc/mm: convert coprocessor fault to lock_mm_and_find_vma() mm/fault: convert remaining simple cases to lock_mm_and_find_vma() arm/mm: Convert to using lock_mm_and_find_vma() riscv/mm: Convert to using lock_mm_and_find_vma() mips/mm: Convert to using lock_mm_and_find_vma() powerpc/mm: Convert to using lock_mm_and_find_vma() arm64/mm: Convert to using lock_mm_and_find_vma() mm: make the page fault mmap locking killable mm: introduce new 'lock_mm_and_find_vma()' page fault helper

mm: always expand the stack with the mmap write lock held

2023-06-27T16:41:30+00:00

This finishes the job of always holding the mmap write lock when extending the user stack vma, and removes the 'write_locked' argument from the vm helper functions again. For some cases, we just avoid expanding the stack at all: drivers and page pinning really shouldn't be extending any stacks. Let's see if any strange users really wanted that. It's worth noting that architectures that weren't converted to the new lock_mm_and_find_vma() helper function are left using the legacy "expand_stack()" function, but it has been changed to drop the mmap_lock and take it for writing while expanding the vma. This makes it fairly straightforward to convert the remaining architectures. As a result of dropping and re-taking the lock, the calling conventions for this function have also changed, since the old vma may no longer be valid. So it will now return the new vma if successful, and NULL - and the lock dropped - if the area could not be extended. Tested-by: Vegard Nossum Tested-by: John Paul Adrian Glaubitz # ia64 Tested-by: Frank Scheiner # ia64 Signed-off-by: Linus Torvalds

mm: make find_extend_vma() fail if write lock not held

2023-06-24T21:13:54+00:00

Make calls to extend_vma() and find_extend_vma() fail if the write lock is required. To avoid making this a flag-day event, this still allows the old read-locking case for the trivial situations, and passes in a flag to say "is it write-locked". That way write-lockers can say "yes, I'm being careful", and legacy users will continue to work in all the common cases until they have been fully converted to the new world order. Co-Developed-by: Matthew Wilcox (Oracle) Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: Liam R. Howlett Signed-off-by: Linus Torvalds