kernel/linux.git/arch/x86/include/asm/stacktrace.h, branch v6.6.132

x86/mm/64: Improve stack overflow warnings

2021-09-21T11:57:43+00:00

Current code has an explicit check for hitting the task stack guard; but overflowing any of the other stacks will get you a non-descript general #DF warning. Improve matters by using get_stack_info_noinstr() to detetrmine if and which stack guard page got hit, enabling a better stack warning. In specific, Michael Wang reported what turned out to be an NMI exception stack overflow, which is now clearly reported as such: [] BUG: NMI stack guard page was hit at 0000000085fd977b (stack is 000000003a55b09e..00000000d8cce1a5) Reported-by: Michael Wang Signed-off-by: Peter Zijlstra (Intel) Tested-by: Michael Wang Link: https://lkml.kernel.org/r/YUTE/NuqnaWbST8n@hirez.programming.kicks-ass.net

x86/dumpstack: Make show_trace_log_lvl() static

2020-11-17T18:05:32+00:00

show_trace_log_lvl() is not used by other compilation units so make it static and remove the declaration from the header file. Signed-off-by: Hui Su Signed-off-by: Borislav Petkov Link: https://lkml.kernel.org/r/20201113133943.GA136221@rlk

x86/dumpstack/64: Add noinstr version of get_stack_info()

2020-09-09T09:33:19+00:00

The get_stack_info() functionality is needed in the entry code for the #VC exception handler. Provide a version of it in the .text.noinstr section which can be called safely from there. Signed-off-by: Joerg Roedel Signed-off-by: Borislav Petkov Link: https://lkml.kernel.org/r/20200907131613.12703-45-joro@8bytes.org

x86: add missing const qualifiers for log_lvl

2020-06-09T16:39:12+00:00

Currently, the log-level of show_stack() depends on a platform realization. It creates situations where the headers are printed with lower log level or higher than the stacktrace (depending on a platform or user). Furthermore, it forces the logic decision from user to an architecture side. In result, some users as sysrq/kdb/etc are doing tricks with temporary rising console_loglevel while printing their messages. And in result it not only may print unwanted messages from other CPUs, but also omit printing at all in the unlucky case where the printk() was deferred. Introducing log-level parameter and KERN_UNSUPPRESSED [1] seems an easier approach than introducing more printk buffers. Also, it will consolidate printings with headers. Keep log_lvl const show_trace_log_lvl() and printk_stack_address() as the new generic show_stack_loglvl() wants to have a proper const qualifier. And gcc rightfully produces warnings in case it's not keept: arch/x86/kernel/dumpstack.c: In function `show_stack': arch/x86/kernel/dumpstack.c:294:37: warning: passing argument 4 of `show_trace_log_lv ' discards `const' qualifier from pointer target type [-Wdiscarded-qualifiers] 294 | show_trace_log_lvl(task, NULL, sp, loglvl); | ^~~~~~ arch/x86/kernel/dumpstack.c:163:32: note: expected `char *' but argument is of type `const char *' 163 | unsigned long *stack, char *log_lvl) | ~~~~~~^~~~~~~ [1]: https://lore.kernel.org/lkml/20190528002412.1625-1-dima@arista.com/T/#u Signed-off-by: Dmitry Safonov Signed-off-by: Andrew Morton Cc: Borislav Petkov Cc: "H. Peter Anvin" Cc: Ingo Molnar Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20200418201944.482088-41-dima@arista.com Signed-off-by: Linus Torvalds

x86/stackframe/32: Provide consistent pt_regs

2019-06-25T08:23:47+00:00

Currently pt_regs on x86_32 has an oddity in that kernel regs (!user_mode(regs)) are short two entries (esp/ss). This means that any code trying to use them (typically: regs->sp) needs to jump through some unfortunate hoops. Change the entry code to fix this up and create a full pt_regs frame. This then simplifies various trampolines in ftrace and kprobes, the stack unwinder, ptrace, kdump and kgdb. Much thanks to Josh for help with the cleanups! Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Josh Poimboeuf Acked-by: Masami Hiramatsu Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Signed-off-by: Ingo Molnar

Merge branch 'x86-irq-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

2019-05-06T22:56:41+00:00

Pull x86 irq updates from Ingo Molnar: "Here are the main changes in this tree: - Introduce x86-64 IRQ/exception/debug stack guard pages to detect stack overflows immediately and deterministically. - Clean up over a decade worth of cruft accumulated. The outcome of this should be more clear-cut faults/crashes when any of the low level x86 CPU stacks overflow, instead of silent memory corruption and sporadic failures much later on" * 'x86-irq-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (33 commits) x86/irq: Fix outdated comments x86/irq/64: Remove stack overflow debug code x86/irq/64: Remap the IRQ stack with guard pages x86/irq/64: Split the IRQ stack into its own pages x86/irq/64: Init hardirq_stack_ptr during CPU hotplug x86/irq/32: Handle irq stack allocation failure proper x86/irq/32: Invoke irq_ctx_init() from init_IRQ() x86/irq/64: Rename irq_stack_ptr to hardirq_stack_ptr x86/irq/32: Rename hard/softirq_stack to hard/softirq_stack_ptr x86/irq/32: Make irq stack a character array x86/irq/32: Define IRQ_STACK_SIZE x86/dumpstack/64: Speedup in_exception_stack() x86/exceptions: Split debug IST stack x86/exceptions: Enable IST guard pages x86/exceptions: Disconnect IST index and stack order x86/cpu: Remove orig_ist array x86/cpu: Prepare TSS.IST setup for guard pages x86/dumpstack/64: Use cpu_entry_area instead of orig_ist x86/irq/64: Use cpu entry area instead of orig_ist x86/traps: Use cpu_entry_area instead of orig_ist ...

perf/x86: Make perf callchains work without CONFIG_FRAME_POINTER

2019-04-29T06:25:05+00:00

Currently perf callchain doesn't work well with ORC unwinder when sampling from trace point. We'll get useless in kernel callchain like this: perf 6429 [000] 22.498450: kmem:mm_page_alloc: page=0x176a17 pfn=1534487 order=0 migratetype=0 gfp_flags=GFP_KERNEL ffffffffbe23e32e __alloc_pages_nodemask+0x22e (/lib/modules/5.1.0-rc3+/build/vmlinux) 7efdf7f7d3e8 __poll+0x18 (/usr/lib64/libc-2.28.so) 5651468729c1 [unknown] (/usr/bin/perf) 5651467ee82a main+0x69a (/usr/bin/perf) 7efdf7eaf413 __libc_start_main+0xf3 (/usr/lib64/libc-2.28.so) 5541f689495641d7 [unknown] ([unknown]) The root cause is that, for trace point events, it doesn't provide a real snapshot of the hardware registers. Instead perf tries to get required caller's registers and compose a fake register snapshot which suppose to contain enough information for start a unwinding. However without CONFIG_FRAME_POINTER, if failed to get caller's BP as the frame pointer, so current frame pointer is returned instead. We get a invalid register combination which confuse the unwinder, and end the stacktrace early. So in such case just don't try dump BP, and let the unwinder start directly when the register is not a real snapshot. Use SP as the skip mark, unwinder will skip all the frames until it meet the frame of the trace point caller. Tested with frame pointer unwinder and ORC unwinder, this makes perf callchain get the full kernel space stacktrace again like this: perf 6503 [000] 1567.570191: kmem:mm_page_alloc: page=0x16c904 pfn=1493252 order=0 migratetype=0 gfp_flags=GFP_KERNEL ffffffffb523e2ae __alloc_pages_nodemask+0x22e (/lib/modules/5.1.0-rc3+/build/vmlinux) ffffffffb52383bd __get_free_pages+0xd (/lib/modules/5.1.0-rc3+/build/vmlinux) ffffffffb52fd28a __pollwait+0x8a (/lib/modules/5.1.0-rc3+/build/vmlinux) ffffffffb521426f perf_poll+0x2f (/lib/modules/5.1.0-rc3+/build/vmlinux) ffffffffb52fe3e2 do_sys_poll+0x252 (/lib/modules/5.1.0-rc3+/build/vmlinux) ffffffffb52ff027 __x64_sys_poll+0x37 (/lib/modules/5.1.0-rc3+/build/vmlinux) ffffffffb500418b do_syscall_64+0x5b (/lib/modules/5.1.0-rc3+/build/vmlinux) ffffffffb5a0008c entry_SYSCALL_64_after_hwframe+0x44 (/lib/modules/5.1.0-rc3+/build/vmlinux) 7f71e92d03e8 __poll+0x18 (/usr/lib64/libc-2.28.so) 55a22960d9c1 [unknown] (/usr/bin/perf) 55a22958982a main+0x69a (/usr/bin/perf) 7f71e9202413 __libc_start_main+0xf3 (/usr/lib64/libc-2.28.so) 5541f689495641d7 [unknown] ([unknown]) Co-developed-by: Josh Poimboeuf Signed-off-by: Kairui Song Signed-off-by: Peter Zijlstra (Intel) Cc: Alexander Shishkin Cc: Alexei Starovoitov Cc: Arnaldo Carvalho de Melo Cc: Borislav Petkov Cc: Dave Young Cc: Jiri Olsa Cc: Linus Torvalds Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Thomas Gleixner Link: https://lkml.kernel.org/r/20190422162652.15483-1-kasong@redhat.com Signed-off-by: Ingo Molnar

x86/exceptions: Disconnect IST index and stack order

2019-04-17T13:01:09+00:00

The entry order of the TSS.IST array and the order of the stack storage/mapping are not required to be the same. With the upcoming split of the debug stack this is going to fall apart as the number of TSS.IST array entries stays the same while the actual stacks are increasing. Make them separate so that code like dumpstack can just utilize the mapping order. The IST index is solely required for the actual TSS.IST array initialization. Signed-off-by: Thomas Gleixner Signed-off-by: Borislav Petkov Cc: Andy Lutomirski Cc: Baoquan He Cc: "Chang S. Bae" Cc: Dominik Brodowski Cc: Dou Liyang Cc: "H. Peter Anvin" Cc: Ingo Molnar Cc: Jann Horn Cc: Josh Poimboeuf Cc: Kees Cook Cc: "Kirill A. Shutemov" Cc: Konrad Rzeszutek Wilk Cc: Nicolai Stange Cc: Peter Zijlstra Cc: Qian Cai Cc: Sean Christopherson Cc: x86-ml Link: https://lkml.kernel.org/r/20190414160145.241588113@linutronix.de

x86/dumpstack: Don't dump kernel memory based on usermode RIP

2018-08-31T15:08:22+00:00

show_opcodes() is used both for dumping kernel instructions and for dumping user instructions. If userspace causes #PF by jumping to a kernel address, show_opcodes() can be reached with regs->ip controlled by the user, pointing to kernel code. Make sure that userspace can't trick us into dumping kernel memory into dmesg. Fixes: 7cccf0725cf7 ("x86/dumpstack: Add a show_ip() function") Signed-off-by: Jann Horn Signed-off-by: Thomas Gleixner Reviewed-by: Kees Cook Reviewed-by: Borislav Petkov Cc: "H. Peter Anvin" Cc: Andy Lutomirski Cc: security@kernel.org Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/20180828154901.112726-1-jannh@google.com

x86/dumpstack: Add a show_ip() function

2018-04-26T14:15:27+00:00

... which shows the Instruction Pointer along with the insn bytes around it. Use it whenever rIP is printed. Drop the rIP < PAGE_OFFSET check since probe_kernel_read() can handle any address properly. Signed-off-by: Borislav Petkov Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Andy Lutomirski Link: https://lkml.kernel.org/r/20180417161124.5294-8-bp@alien8.de