UBUNTU-CVE-2025-37964
In the Linux kernel, the following vulnerability has been resolved: x86/mm: Eliminate window where TLB flushes may be inadvertently skipped tl;dr: There is a window in the mm switching code where the new CR3 is set and the CPU should be getting TLB flushes for the new mm. But should_flush_tlb() has a bug and suppresses the flush. Fix it by widening the window where should_flush_tlb() sends an IPI. Long Version: === History === There were a few things leading up to this. First, updating mm_cpumask() was observed to be too expensive, so it was made lazier. But being lazy caused too many unnecessary IPIs to CPUs due to the now-lazy mm_cpumask(). So code was added to cull mm_cpumask() periodically[2]. But that culling was a bit too aggressive and skipped sending TLB flushes to CPUs that need them. So here we are again. === Problem === The too-aggressive code in should_flush_tlb() strikes in this window: // Turn on IPIs for this CPU/mm combination, but only // if should_flush_tlb() agrees: cpumask_set_cpu(cpu, mm_cpumask(next)); next_tlb_gen = atomic64_read(&next->context.tlb_gen); choose_new_asid(next, next_tlb_gen, &new_asid, &need_flush); load_new_mm_cr3(need_flush); // ^ After 'need_flush' is set to false, IPIs *MUST* // be sent to this CPU and not be ignored. this_cpu_write(cpu_tlbstate.loaded_mm, next); // ^ Not until this point does should_flush_tlb() // become true! should_flush_tlb() will suppress TLB flushes between load_new_mm_cr3() and writing to 'loaded_mm', which is a window where they should not be suppressed. Whoops. === Solution === Thankfully, the fuzzy "just about to write CR3" window is already marked with loaded_mm==LOADED_MM_SWITCHING. Simply checking for that state in should_flush_tlb() is sufficient to ensure that the CPU is targeted with an IPI. This will cause more TLB flush IPIs. But the window is relatively small and I do not expect this to cause any kind of measurable performance impact. Update the comment where LOADED_MM_SWITCHING is written since it grew yet another user. Peter Z also raised a concern that should_flush_tlb() might not observe 'loaded_mm' and 'is_lazy' in the same order that switch_mm_irqs_off() writes them. Add a barrier to ensure that they are observed in the order they are written.
02 / AFFECTED SOFTWARE
Affected packages
11 explicit affected versions
67 explicit affected versions
50 explicit affected versions
6 explicit affected versions
12 explicit affected versions
19 explicit affected versions
6 explicit affected versions
77 explicit affected versions
62 explicit affected versions
62 explicit affected versions
70 explicit affected versions
57 explicit affected versions
10 explicit affected versions
69 explicit affected versions
7 explicit affected versions
67 explicit affected versions
13 explicit affected versions
35 explicit affected versions
29 explicit affected versions
27 explicit affected versions
18 explicit affected versions
27 explicit affected versions
64 explicit affected versions
13 explicit affected versions
4 explicit affected versions
23 explicit affected versions
1 explicit affected versions
21 explicit affected versions
16 explicit affected versions
10 explicit affected versions
8 explicit affected versions
7 explicit affected versions
12 explicit affected versions
3 explicit affected versions
7 explicit affected versions
12 explicit affected versions
14 explicit affected versions
10 explicit affected versions
1 explicit affected versions
7 explicit affected versions
12 explicit affected versions
12 explicit affected versions
23 explicit affected versions
5 explicit affected versions
10 explicit affected versions
37 explicit affected versions
13 explicit affected versions
10 explicit affected versions
3 explicit affected versions
44 explicit affected versions
26 explicit affected versions
13 explicit affected versions
16 explicit affected versions
29 explicit affected versions
13 explicit affected versions
4 explicit affected versions
74 explicit affected versions
69 explicit affected versions
32 explicit affected versions
33 explicit affected versions
23 explicit affected versions
38 explicit affected versions
10 explicit affected versions
11 explicit affected versions
12 explicit affected versions
7 explicit affected versions
12 explicit affected versions
54 explicit affected versions
70 explicit affected versions
10 explicit affected versions
26 explicit affected versions
46 explicit affected versions
3 explicit affected versions
13 explicit affected versions
16 explicit affected versions
71 explicit affected versions
3 explicit affected versions
7 explicit affected versions
51 explicit affected versions
18 explicit affected versions
12 explicit affected versions
59 explicit affected versions
57 explicit affected versions
27 explicit affected versions
44 explicit affected versions
8 explicit affected versions
34 explicit affected versions
5 explicit affected versions
8 explicit affected versions
10 explicit affected versions
7 explicit affected versions
13 explicit affected versions
13 explicit affected versions
7 explicit affected versions
8 explicit affected versions
10 explicit affected versions
12 explicit affected versions
1 explicit affected versions
17 explicit affected versions
44 explicit affected versions
16 explicit affected versions
3 explicit affected versions
13 explicit affected versions
26 explicit affected versions
45 explicit affected versions
37 explicit affected versions
14 explicit affected versions
38 explicit affected versions
9 explicit affected versions
11 explicit affected versions
70 explicit affected versions
7 explicit affected versions
4 explicit affected versions
59 explicit affected versions
1 explicit affected versions
68 explicit affected versions
58 explicit affected versions
58 explicit affected versions
13 explicit affected versions
43 explicit affected versions
10 explicit affected versions
8 explicit affected versions
14 explicit affected versions
4 explicit affected versions
78 explicit affected versions
11 explicit affected versions
1 explicit affected versions
59 explicit affected versions
03 / CONNECTIONS
Connected vulnerabilities
04 / EVIDENCE
Source records
In the Linux kernel, the following vulnerability has been resolved: x86/mm: Eliminate window where TLB flushes may be inadvertently skipped tl;dr: There is a window in the mm switching code where the new CR3 is set and the CPU should be getting TLB flushes for the new mm. But should_flush_tlb() has a bug and suppresses the flush. Fix it by widening the window where should_flush_tlb() sends an IPI. Long Version: === History === There were a few things leading up to this. First, updating mm_cpumask() was observed to be too expensive, so it was made lazier. But being lazy caused too many unnecessary IPIs to CPUs due to the now-lazy mm_cpumask(). So code was added to cull mm_cpumask() periodically[2]. But that culling was a bit too aggressive and skipped sending TLB flushes to CPUs that need them. So here we are again. === Problem === The too-aggressive code in should_flush_tlb() strikes in this window: // Turn on IPIs for this CPU/mm combination, but only // if should_flush_tlb() agrees: cpumask_set_cpu(cpu, mm_cpumask(next)); next_tlb_gen = atomic64_read(&next->context.tlb_gen); choose_new_asid(next, next_tlb_gen, &new_asid, &need_flush); load_new_mm_cr3(need_flush); // ^ After 'need_flush' is set to false, IPIs *MUST* // be sent to this CPU and not be ignored. this_cpu_write(cpu_tlbstate.loaded_mm, next); // ^ Not until this point does should_flush_tlb() // become true! should_flush_tlb() will suppress TLB flushes between load_new_mm_cr3() and writing to 'loaded_mm', which is a window where they should not be suppressed. Whoops. === Solution === Thankfully, the fuzzy "just about to write CR3" window is already marked with loaded_mm==LOADED_MM_SWITCHING. Simply checking for that state in should_flush_tlb() is sufficient to ensure that the CPU is targeted with an IPI. This will cause more TLB flush IPIs. But the window is relatively small and I do not expect this to cause any kind of measurable performance impact. Update the comment where LOADED_MM_SWITCHING is written since it grew yet another user. Peter Z also raised a concern that should_flush_tlb() might not observe 'loaded_mm' and 'is_lazy' in the same order that switch_mm_irqs_off() writes them. Add a barrier to ensure that they are observed in the order they are written.
05 / REFERENCES
Further evidence
- https://git.kernel.org/linus/fea4e317f9e7e1f449ce90dedc27a2d2a95bee5a
- https://git.kernel.org/stable/c/02ad4ce144bd27f71f583f667fdf3b3ba0753477
- https://git.kernel.org/stable/c/12f703811af043d32b1c8a30001b2fa04d5cd0ac
- https://git.kernel.org/stable/c/399ec9ca8fc4999e676ff89a90184ec40031cf59
- https://git.kernel.org/stable/c/d41072906abec8bb8e01ed16afefbaa558908c89
- https://git.kernel.org/stable/c/d87392094f96e162fa5fa5a8640d70cc0952806f
- https://git.kernel.org/stable/c/fea4e317f9e7e1f449ce90dedc27a2d2a95bee5a
- https://ubuntu.com/security/CVE-2025-37964
- https://ubuntu.com/security/notices/USN-7654-1
- https://ubuntu.com/security/notices/USN-7654-2
- https://ubuntu.com/security/notices/USN-7654-3
- https://ubuntu.com/security/notices/USN-7654-4
- https://ubuntu.com/security/notices/USN-7654-5
- https://ubuntu.com/security/notices/USN-7655-1
- https://ubuntu.com/security/notices/USN-7686-1
- https://ubuntu.com/security/notices/USN-7699-1
- https://ubuntu.com/security/notices/USN-7699-2
- https://ubuntu.com/security/notices/USN-7711-1
- https://ubuntu.com/security/notices/USN-7712-1
- https://ubuntu.com/security/notices/USN-7712-2
- https://ubuntu.com/security/notices/USN-7721-1
- https://www.cve.org/CVERecord?id=CVE-2025-37964