- 23 Dec, 2017 7 commits
-
-
Dave Hansen authored
Add the pagetable helper functions do manage the separate user space page tables. [ tglx: Split out from the big combo kaiser patch. Folded Andys simplification and made it out of line as Boris suggested ] Signed-off-by:
Dave Hansen <dave.hansen@linux.intel.com> Signed-off-by:
Thomas Gleixner <tglx@linutronix.de> Cc: Andy Lutomirski <luto@kernel.org> Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: David Laight <David.Laight@aculab.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: Eduardo Valentin <eduval@amazon.com> Cc: Greg KH <gregkh@linuxfoundation.org> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Juergen Gross <jgross@suse.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Will Deacon <will.deacon@arm.com> Cc: aliguori@amazon.com Cc: daniel.gruss@iaik.tugraz.at Cc: hughd@google.com Cc: keescook@google.com Cc: linux-kernel@vger.kernel.org Signed-off-by:
Ingo Molnar <mingo@kernel.org>
-
Borislav Petkov authored
Keep the "nopti" optional for traditional reasons. [ tglx: Don't allow force on when running on XEN PV and made 'on' printout conditional ] Requested-by:
Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
Borislav Petkov <bp@suse.de> Signed-off-by:
-
Thomas Gleixner authored
Add the initial files for kernel page table isolation, with a minimal init function and the boot time detection for this misfeature. Signed-off-by:
-
Dave Hansen authored
PAGE_TABLE_ISOLATION needs to switch to a different CR3 value when it enters the kernel and switch back when it exits. This essentially needs to be done before leaving assembly code. This is extra challenging because the switching context is tricky: the registers that can be clobbered can vary. It is also hard to store things on the stack because there is an established ABI (ptregs) or the stack is entirely unsafe to use. Establish a set of macros that allow changing to the user and kernel CR3 values. Interactions with SWAPGS: Previous versions of the PAGE_TABLE_ISOLATION code relied on having per-CPU scratch space to save/restore a register that can be used for the CR3 MOV. The %GS register is used to index into our per-CPU space, so SWAPGS *had* to be done before the CR3 switch. That scratch space is gone now, but the semantic that SWAPGS must be done before the CR3 MOV is retained. This is good to keep because it is not that hard to do and it allows to do things like add per-CPU debugging information. What this does in the NMI code is worth pointing out. NMIs can interrupt *any* context and they can also be nested with NMIs interrupting other NMIs. The comments below ".Lnmi_from_kernel" explain the format of the stack during this situation. Changing the format of this stack is hard. Instead of storing the old CR3 value on the stack, this depends on the *regular* register save/restore mechanism and then uses %r14 to keep CR3 during the NMI. It is callee-saved and will not be clobbered by the C NMI handlers that get called. [ PeterZ: ESPFIX optimization ] Based-on-code-from: Andy Lutomirski <luto@kernel.org> Signed-off-by:
-
Dave Hansen authored
Global pages stay in the TLB across context switches. Since all contexts share the same kernel mapping, these mappings are marked as global pages so kernel entries in the TLB are not flushed out on a context switch. But, even having these entries in the TLB opens up something that an attacker can use, such as the double-page-fault attack: http://www.ieee-security.org/TC/SP2013/papers/4977a191.pdf That means that even when PAGE_TABLE_ISOLATION switches page tables on return to user space the global pages would stay in the TLB cache. Disable global pages so that kernel TLB entries can be flushed before returning to user space. This way, all accesses to kernel addresses from userspace result in a TLB miss independent of the existence of a kernel mapping. Suppress global pages via the __supported_pte_mask. The user space mappings set PAGE_GLOBAL for the minimal kernel mappings which are required for entry/exit. These mappings are set up manually so the filtering does not take place. [ The __supported_pte_mask simplification was written by Thomas Gleixner. ] Signed-off-by:
-
Thomas Gleixner authored
Many x86 CPUs leak information to user space due to missing isolation of user space and kernel space page tables. There are many well documented ways to exploit that. The upcoming software migitation of isolating the user and kernel space page tables needs a misfeature flag so code can be made runtime conditional. Add the BUG bits which indicates that the CPU is affected and add a feature bit which indicates that the software migitation is enabled. Assume for now that _ALL_ x86 CPUs are affected by this. Exceptions can be made later. Signed-off-by:
-
Thomas Gleixner authored
The loop which populates the CPU entry area PMDs can wrap around on 32bit machines when the number of CPUs is small. It worked wonderful for NR_CPUS=64 for whatever reason and the moron who wrote that code did not bother to test it with !SMP. Check for the wraparound to fix it. Fixes: 92a0f81d ("x86/cpu_entry_area: Move it out of the fixmap") Reported-by:
kernel test robot <fengguang.wu@intel.com> Signed-off-by:
Borislav Petkov <bp@alien8.de>
-
- 22 Dec, 2017 23 commits
-
-
Thomas Gleixner authored
init_espfix_bsp() needs to be invoked before the page table isolation initialization. Move it into mm_init() which is the place where pti_init() will be added. While at it get rid of the #ifdeffery and provide proper stub functions. Signed-off-by:
-
Thomas Gleixner authored
Put the cpu_entry_area into a separate P4D entry. The fixmap gets too big and 0-day already hit a case where the fixmap PTEs were cleared by cleanup_highmap(). Aside of that the fixmap API is a pain as it's all backwards. Signed-off-by:
-
Thomas Gleixner authored
Separate the cpu_entry_area code out of cpu/common.c and the fixmap. Signed-off-by:
-
Peter Zijlstra authored
Unclutter tlbflush.h a little. Signed-off-by:
Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Andy Lutomirski <luto@kernel.org> Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: David Laight <David.Laight@aculab.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: Eduardo Valentin <eduval@amazon.com> Cc: Greg KH <gregkh@linuxfoundation.org> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Juergen Gross <jgross@suse.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Will Deacon <will.deacon@arm.com> Cc: aliguori@amazon.com Cc: daniel.gruss@iaik.tugraz.at Cc: hughd@google.com Cc: keescook@google.com Cc: linux-mm@kvack.org Signed-off-by:
-
Dave Hansen authored
There are effectively two ASID types: 1. The one stored in the mmu_context that goes from 0..5 2. The one programmed into the hardware that goes from 1..6 This consolidates the locations where converting between the two (by doing a +1) to a single place which gives us a nice place to comment. PAGE_TABLE_ISOLATION will also need to, given an ASID, know which hardware ASID to flush for the userspace mapping. Signed-off-by:
-
Dave Hansen authored
First, it's nice to remove the magic numbers. Second, PAGE_TABLE_ISOLATION is going to consume half of the available ASID space. The space is currently unused, but add a comment to spell out this new restriction. Signed-off-by:
-
Dave Hansen authored
For flushing the TLB, the ASID which has been programmed into the hardware must be known. That differs from what is in 'cpu_tlbstate'. Add functions to transform the 'cpu_tlbstate' values into to the one programmed into the hardware (CR3). It's not easy to include mmu_context.h into tlbflush.h, so just move the CR3 building over to tlbflush.h. Signed-off-by:
-
Peter Zijlstra authored
Per popular request.. Signed-off-by:
-
Peter Zijlstra authored
atomic64_inc_return() already implies smp_mb() before and after. Signed-off-by:
-
Peter Zijlstra authored
__flush_tlb_single() is for user mappings, __flush_tlb_one() for kernel mappings. Signed-off-by:
-
Peter Zijlstra authored
Commit: ec400dde ("x86/microcode_intel_early.c: Early update ucode on Intel's CPU") ... grubbed into tlbflush internals without coherent explanation. Since it says its a precaution and the SDM doesn't mention anything like this, take it out back. Signed-off-by:
-
Peter Zijlstra authored
Since uv_flush_tlb_others() implements flush_tlb_others() which is about flushing user mappings, we should use __flush_tlb_single(), which too is about flushing user mappings. Signed-off-by:
Andrew Banman <abanman@hpe.com> Cc: Andy Lutomirski <luto@kernel.org> Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: David Laight <David.Laight@aculab.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: Eduardo Valentin <eduval@amazon.com> Cc: Greg KH <gregkh@linuxfoundation.org> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Juergen Gross <jgross@suse.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Mike Travis <mike.travis@hpe.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Will Deacon <will.deacon@arm.com> Cc: aliguori@amazon.com Cc: daniel.gruss@iaik.tugraz.at Cc: hughd@google.com Cc: keescook@google.com Cc: linux-mm@kvack.org Signed-off-by:
-
Dave Hansen authored
If the kernel oopses while on the trampoline stack, it will print "<SYSENTER>" even if SYSENTER is not involved. That is rather confusing. The "SYSENTER" stack is used for a lot more than SYSENTER now. Give it a better string to display in stack dumps, and rename the kernel code to match. Also move the 32-bit code over to the new naming even though it still uses the entry stack only for SYSENTER. Signed-off-by:
-
Peter Zijlstra authored
Signed-off-by:
-
Andy Lutomirski authored
The old docs had the vsyscall range wrong and were missing the fixmap. Fix both. There used to be 8 MB reserved for future vsyscalls, but that's long gone. Signed-off-by:
Andy Lutomirski <luto@kernel.org> Signed-off-by:
-
Thomas Gleixner authored
The LDT is inherited across fork() or exec(), but that makes no sense at all because exec() is supposed to start the process clean. The reason why this happens is that init_new_context_ldt() is called from init_new_context() which obviously needs to be called for both fork() and exec(). It would be surprising if anything relies on that behaviour, so it seems to be safe to remove that misfeature. Split the context initialization into two parts. Clear the LDT pointer and initialize the mutex from the general context init and move the LDT duplication to arch_dup_mmap() which is only called on fork(). Signed-off-by:
-
Peter Zijlstra authored
The LDT is duplicated on fork() and on exec(), which is wrong as exec() should start from a clean state, i.e. without LDT. To fix this the LDT duplication code will be moved into arch_dup_mmap() which is only called for fork(). This introduces a locking problem. arch_dup_mmap() holds mmap_sem of the parent process, but the LDT duplication code needs to acquire mm->context.lock to access the LDT data safely, which is the reverse lock order of write_ldt() where mmap_sem nests into context.lock. Solve this by introducing a new rw semaphore which serializes the read/write_ldt() syscall operations and use context.lock to protect the actual installment of the LDT descriptor. So context.lock stabilizes mm->context.ldt and can nest inside of the new semaphore or mmap_sem. Signed-off-by:
-
Thomas Gleixner authored
In order to sanitize the LDT initialization on x86 arch_dup_mmap() must be allowed to fail. Fix up all instances. Signed-off-by:
-
Andy Lutomirski authored
If something goes wrong with pagetable setup, vsyscall=native will accidentally fall back to emulation. Make it warn and fail so that we notice. Signed-off-by:
-
Andy Lutomirski authored
The kernel is very erratic as to which pagetables have _PAGE_USER set. The vsyscall page gets lucky: it seems that all of the relevant pagetables are among the apparently arbitrary ones that set _PAGE_USER. Rather than relying on chance, just explicitly set _PAGE_USER. This will let us clean up pagetable setup to stop setting _PAGE_USER. The added code can also be reused by pagetable isolation to manage the _PAGE_USER bit in the usermode tables. [ tglx: Folded paravirt fix from Juergen Gross ] Signed-off-by:
-
Thomas Gleixner authored
The address hints are a trainwreck. The array entry numbers have to kept magically in sync with the actual hints, which is doomed as some of the array members are initialized at runtime via the entry numbers. Designated initializers have been around before this code was implemented.... Use the entry numbers to populate the address hints array and add the missing bits and pieces. Split 32 and 64 bit for readability sake. Signed-off-by:
-
Thomas Gleixner authored
The check for a present page in printk_prot(): if (!pgprot_val(prot)) { /* Not present */ is bogus. If a PTE is set to PAGE_NONE then the pgprot_val is not zero and the entry is decoded in bogus ways, e.g. as RX GLB. That is confusing when analyzing mapping correctness. Check for the present bit to make an informed decision. Signed-off-by: -
Thomas Gleixner authored
The recent cpu_entry_area changes fail to compile on 32-bit when BIGSMP=y and NR_CPUS=512, because the fixmap area becomes too big. Limit the number of CPUs with BIGSMP to 64, which is already way to big for 32-bit, but it's at least a working limitation. We performed a quick survey of 32-bit-only machines that might be affected by this change negatively, but found none. Signed-off-by:
-
- 17 Dec, 2017 10 commits
-
-
Thomas Gleixner authored
There is currently no way to force CPU bug bits like CPU feature bits. That makes it impossible to set a bug bit once at boot and have it stick for all upcoming CPUs. Extend the force set/clear arrays to handle bug bits as well. Signed-off-by:
-
Thomas Gleixner authored
There is no generic way to test whether a kernel is running on a specific hypervisor. But that's required to prevent the upcoming user address space separation feature in certain guest modes. Make the hypervisor type enum unconditionally available and provide a helper function which allows to test for a specific type. Signed-off-by:
Juergen Gross <jgross@suse.com> Cc: Andy Lutomirski <luto@kernel.org> Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Borislav Petkov <bpetkov@suse.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: David Laight <David.Laight@aculab.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: Eduardo Valentin <eduval@amazon.com> Cc: Greg KH <gregkh@linuxfoundation.org> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Rik van Riel <riel@redhat.com> Cc: Will Deacon <will.deacon@arm.com> Cc: aliguori@amazon.com Cc: daniel.gruss@iaik.tugraz.at Cc: hughd@google.com Cc: keescook@google.com Link: https://lkml.kernel.org/r/20171204150606.912938129@linutronix.deSigned-off-by:
-
Thomas Gleixner authored
native_flush_tlb_single() will be changed with the upcoming PAGE_TABLE_ISOLATION feature. This requires to have more code in there than INVLPG. Remove the paravirt patching for it. Signed-off-by:
Josh Poimboeuf <jpoimboe@redhat.com> Reviewed-by:
-
Andy Lutomirski authored
The TSS is a fairly juicy target for exploits, and, now that the TSS is in the cpu_entry_area, it's no longer protected by kASLR. Make it read-only on x86_64. On x86_32, it can't be RO because it's written by the CPU during task switches, and we use a task gate for double faults. I'd also be nervous about errata if we tried to make it RO even on configurations without double fault handling. [ tglx: AMD confirmed that there is no problem on 64-bit with TSS RO. So it's probably safe to assume that it's a non issue, though Intel might have been creative in that area. Still waiting for confirmation. ] Signed-off-by:
Borislav Petkov <bpetkov@suse.de> Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: David Laight <David.Laight@aculab.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: Eduardo Valentin <eduval@amazon.com> Cc: Greg KH <gregkh@linuxfoundation.org> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Juergen Gross <jgross@suse.com> Cc: Kees Cook <keescook@chromium.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Rik van Riel <riel@redhat.com> Cc: Will Deacon <will.deacon@arm.com> Cc: aliguori@amazon.com Cc: daniel.gruss@iaik.tugraz.at Cc: hughd@google.com Cc: keescook@google.com Link: https://lkml.kernel.org/r/20171204150606.733700132@linutronix.deSigned-off-by:
-
Andy Lutomirski authored
The existing code was a mess, mainly because C arrays are nasty. Turn SYSENTER_stack into a struct, add a helper to find it, and do all the obvious cleanups this enables. Signed-off-by:
-
Andy Lutomirski authored
Now that the SYSENTER stack has a guard page, there's no need for a canary to detect overflow after the fact. Signed-off-by:
-
Andy Lutomirski authored
The IST stacks are needed when an IST exception occurs and are accessed before any kernel code at all runs. Move them into struct cpu_entry_area. The IST stacks are unlike the rest of cpu_entry_area: they're used even for entries from kernel mode. This means that they should be set up before we load the final IDT. Move cpu_entry_area setup to trap_init() for the boot CPU and set it up for all possible CPUs at once in native_smp_prepare_cpus(). Signed-off-by:
-
Andy Lutomirski authored
Handling SYSCALL is tricky: the SYSCALL handler is entered with every single register (except FLAGS), including RSP, live. It somehow needs to set RSP to point to a valid stack, which means it needs to save the user RSP somewhere and find its own stack pointer. The canonical way to do this is with SWAPGS, which lets us access percpu data using the %gs prefix. With PAGE_TABLE_ISOLATION-like pagetable switching, this is problematic. Without a scratch register, switching CR3 is impossible, so %gs-based percpu memory would need to be mapped in the user pagetables. Doing that without information leaks is difficult or impossible. Instead, use a different sneaky trick. Map a copy of the first part of the SYSCALL asm at a different address for each CPU. Now RIP varies depending on the CPU, so we can use RIP-relative memory access to access percpu memory. By putting the relevant information (one scratch slot and the stack address) at a constant offset relative to RIP, we can make SYSCALL work without relying on %gs. A nice thing about this approach is that we can easily switch it on and off if we want pagetable switching to be configurable. The compat variant of SYSCALL doesn't have this problem in the first place -- there are plenty of scratch registers, since we don't care about preserving r8-r15. This patch therefore doesn't touch SYSCALL32 at all. This patch actually seems to be a small speedup. With this patch, SYSCALL touches an extra cache line and an extra virtual page, but the pipeline no longer stalls waiting for SWAPGS. It seems that, at least in a tight loop, the latter outweights the former. Thanks to David Laight for an optimization tip. Signed-off-by:
-
Andy Lutomirski authored
By itself, this is useless. It gives us the ability to run some final code before exit that cannnot run on the kernel stack. This could include a CR3 switch a la PAGE_TABLE_ISOLATION or some kernel stack erasing, for example. (Or even weird things like *changing* which kernel stack gets used as an ASLR-strengthening mechanism.) The SYSRET32 path is not covered yet. It could be in the future or we could just ignore it and force the slow path if needed. Signed-off-by:
-
Andy Lutomirski authored
Historically, IDT entries from usermode have always gone directly to the running task's kernel stack. Rearrange it so that we enter on a per-CPU trampoline stack and then manually switch to the task's stack. This touches a couple of extra cachelines, but it gives us a chance to run some code before we touch the kernel stack. The asm isn't exactly beautiful, but I think that fully refactoring it can wait. Signed-off-by:
-
