Re: [PATCH v5 9/9] x86/nmi: Include NMI-source information in tracepoint and debug prints
From: Sohil Mehta <hidden>
Date: 2025-05-08 00:02:52
Also in:
kvm, linux-edac, linux-perf-users, linux-pm, lkml
On 5/7/2025 2:48 PM, Steven Rostedt wrote:
On Tue, 6 May 2025 18:21:45 -0700 Sohil Mehta [off-list ref] wrote:quoted
diff --git a/include/trace/events/nmi.h b/include/trace/events/nmi.h index 18e0411398ba..6e4a1ff70a44 100644 --- a/include/trace/events/nmi.h +++ b/include/trace/events/nmi.h@@ -10,29 +10,32 @@ TRACE_EVENT(nmi_handler, - TP_PROTO(void *handler, s64 delta_ns, int handled), + TP_PROTO(void *handler, s64 delta_ns, int handled, unsigned long source_bitmap),Even though x86 is currently the only architecture using the nmi tracepoint, this "source_bitmap" makes it become very x86 specific.
Sure. Will move it into x86.
This file should be moved into arch/x86/include/asm/trace/ And that would require adding to the Makefile: CFLAGS_nmi.o := -I $(src)/../include/asm/trace
Thank you for the detailed instructions. It makes it a lot easier.
quoted hunk ↗ jump to hunk
diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c index 183e3e717326..b9ece0b63ca7 100644 --- a/arch/x86/kernel/nmi.c +++ b/arch/x86/kernel/nmi.c@@ -202,7 +202,7 @@ static int nmi_handle(unsigned int type, struct pt_regs *regs) thishandled = a->handler(type, regs); handled += thishandled; delta = sched_clock() - delta; - trace_nmi_handler(a->handler, (int)delta, thishandled); + trace_nmi_handler(a->handler, (int)delta, thishandled, source_bitmap); nmi_check_duration(a, delta); }
Also, I just realized that the source_bitmap information might be incorrect for !NMI_LOCAL. With the new changes suggested by PeterZ in patch 5, the source_bitmap would be significantly different from the original one received from fred_event_data(). I need to figure out a solution to print it accurately in all situations. Sohil