On Thu, Nov 23, 2023 at 8:12 AM <andrey.konovalov@xxxxxxxxx> wrote: > > From: Andrey Konovalov <andreyknvl@xxxxxxxxxx> > > When both KASAN and slub_debug are enabled, when a free object is being > prepared in setup_object, slub_debug poisons the object data before KASAN > initializes its per-object metadata. > > Right now, in setup_object, KASAN only initializes the alloc metadata, > which is always stored outside of the object. slub_debug is aware of > this and it skips poisoning and checking that memory area. > > However, with the following patch in this series, KASAN also starts > initializing its free medata in setup_object. As this metadata might be > stored within the object, this initialization might overwrite the > slub_debug poisoning. This leads to slub_debug reports. > > Thus, skip checking slub_debug poisoning of the object data area that > overlaps with the in-object KASAN free metadata. > > Also make slub_debug poisoning of tail kmalloc redzones more precise when > KASAN is enabled: slub_debug can still poison and check the tail kmalloc > allocation area that comes after the KASAN free metadata. > > Signed-off-by: Andrey Konovalov <andreyknvl@xxxxxxxxxx> Thank you for looking at this quickly! Unfortunately the problem isn't fixed yet with the patch. I applied this on top of linux-next and built a kernel with the same config, it is still stuck at boot. [dmesg] [ 0.000000] Linux version 6.7.0-rc2-next-20231122-00001-gfc1613c2f6f3 (hyeyoo@localhost.localdomain) (gcc (GCC) 11.33 [ 0.000000] Command line: console=ttyS0 root=/dev/sda1 nokaslr [ 0.000000] RIP: 0010:setup_arch (arch/x86/kernel/setup.c:443 arch/x86/kernel/setup.c:665 arch/x86/kernel/setup.c:81 [ 0.000000] Code: b6 0a 08 00 48 89 c5 48 85 c0 0f 84 58 13 00 00 48 c1 e8 03 48 83 05 4e a9 66 00 01 80 3c 18 00 0f3 Code starting with the faulting instruction =========================================== 0: b6 0a mov $0xa,%dh 2: 08 00 or %al,(%rax) 4: 48 89 c5 mov %rax,%rbp 7: 48 85 c0 test %rax,%rax a: 0f 84 58 13 00 00 je 0x1368 10: 48 c1 e8 03 shr $0x3,%rax 14: 48 83 05 4e a9 66 00 addq $0x1,0x66a94e(%rip) # 0x66a96a 1b: 01 1c: 80 3c 18 00 cmpb $0x0,(%rax,%rbx,1) 20: f3 repz [ 0.000000] RSP: 0000:ffffffff86207e00 EFLAGS: 00010046 ORIG_RAX: 0000000000000009 [ 0.000000] RAX: 1fffffffffe40069 RBX: dffffc0000000000 RCX: 1ffffffff1230a30 [ 0.000000] RDX: 0000000000000000 RSI: 0107d62120059000 RDI: ffffffff89185180 [ 0.000000] RBP: ffffffffff200348 R08: 8000000000000163 R09: 1ffffffff1230a28 [ 0.000000] R10: ffffffff89194150 R11: 0000000000000000 R12: 0000000000000010 [ 0.000000] R13: ffffffffff200354 R14: 0107d62120058348 R15: 0107d62120058348 [ 0.000000] FS: 0000000000000000(0000) GS:ffffffff88f75000(0000) knlGS:0000000000000000 [ 0.000000] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 0.000000] CR2: ffffffffff200348 CR3: 0000000009128000 CR4: 00000000000000b0 [ 0.000000] Call Trace: [ 0.000000] <TASK> [ 0.000000] ? show_regs (arch/x86/kernel/dumpstack.c:478) [ 0.000000] ? early_fixup_exception (arch/x86/mm/extable.c:364) [ 0.000000] ? do_early_exception (arch/x86/kernel/head64.c:423) [ 0.000000] ? early_idt_handler_common (arch/x86/kernel/head_64.S:555) [ 0.000000] ? setup_arch (arch/x86/kernel/setup.c:443 arch/x86/kernel/setup.c:665 arch/x86/kernel/setup.c:814) [ 0.000000] ? __pfx_setup_arch (arch/x86/kernel/setup.c:728) [ 0.000000] ? vprintk_default (kernel/printk/printk.c:2318) [ 0.000000] ? vprintk (kernel/printk/printk_safe.c:45) [ 0.000000] ? _printk (kernel/printk/printk.c:2328) [ 0.000000] ? __pfx__printk (kernel/printk/printk.c:2323) [ 0.000000] ? init_cgroup_root (kernel/cgroup/cgroup.c:2054) [ 0.000000] ? cgroup_init_early (kernel/cgroup/cgroup.c:6077 (discriminator 13)) [ 0.000000] ? start_kernel (init/main.c:897 (discriminator 3)) [ 0.000000] ? x86_64_start_reservations (arch/x86/kernel/head64.c:543) [ 0.000000] ? x86_64_start_kernel (arch/x86/kernel/head64.c:536) [ 0.000000] ? secondary_startup_64_no_verify (arch/x86/kernel/head_64.S:432) [ 0.000000] </TASK>
