public inbox for kvm@vger.kernel.org
 help / color / mirror / Atom feed
From: "Nicholas Piggin" <npiggin@gmail.com>
To: "Thomas Huth" <thuth@redhat.com>
Cc: "Laurent Vivier" <lvivier@redhat.com>,
	"Andrew Jones" <andrew.jones@linux.dev>,
	"Paolo Bonzini" <pbonzini@redhat.com>,
	"Joel Stanley" <joel@jms.id.au>, <linuxppc-dev@lists.ozlabs.org>,
	<kvm@vger.kernel.org>
Subject: Re: [kvm-unit-tests PATCH 04/32] powerpc: interrupt stack backtracing
Date: Tue, 05 Mar 2024 12:12:39 +1000	[thread overview]
Message-ID: <CZLGB4BCREUU.SZ1AQ0LTGNKB@wheely> (raw)
In-Reply-To: <9d1166ed-e24f-4257-a441-080f577d3dde@redhat.com>

On Fri Mar 1, 2024 at 7:53 PM AEST, Thomas Huth wrote:
> On 26/02/2024 11.11, Nicholas Piggin wrote:
> > Add support for backtracing across interrupt stacks, and
> > add interrupt frame backtrace for unhandled interrupts.
> > 
> > Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> > ---
> >   lib/powerpc/processor.c |  4 ++-
> >   lib/ppc64/asm/stack.h   |  3 +++
> >   lib/ppc64/stack.c       | 55 +++++++++++++++++++++++++++++++++++++++++
> >   powerpc/Makefile.ppc64  |  1 +
> >   powerpc/cstart64.S      |  7 ++++--
> >   5 files changed, 67 insertions(+), 3 deletions(-)
> >   create mode 100644 lib/ppc64/stack.c
> > 
> > diff --git a/lib/powerpc/processor.c b/lib/powerpc/processor.c
> > index ad0d95666..114584024 100644
> > --- a/lib/powerpc/processor.c
> > +++ b/lib/powerpc/processor.c
> > @@ -51,7 +51,9 @@ void do_handle_exception(struct pt_regs *regs)
> >   		return;
> >   	}
> >   
> > -	printf("unhandled cpu exception %#lx at NIA:0x%016lx MSR:0x%016lx\n", regs->trap, regs->nip, regs->msr);
> > +	printf("Unhandled cpu exception %#lx at NIA:0x%016lx MSR:0x%016lx\n",
> > +			regs->trap, regs->nip, regs->msr);
> > +	dump_frame_stack((void *)regs->nip, (void *)regs->gpr[1]);
> >   	abort();
> >   }
> >   
> > diff --git a/lib/ppc64/asm/stack.h b/lib/ppc64/asm/stack.h
> > index 9734bbb8f..94fd1021c 100644
> > --- a/lib/ppc64/asm/stack.h
> > +++ b/lib/ppc64/asm/stack.h
> > @@ -5,4 +5,7 @@
> >   #error Do not directly include <asm/stack.h>. Just use <stack.h>.
> >   #endif
> >   
> > +#define HAVE_ARCH_BACKTRACE
> > +#define HAVE_ARCH_BACKTRACE_FRAME
> > +
> >   #endif
> > diff --git a/lib/ppc64/stack.c b/lib/ppc64/stack.c
> > new file mode 100644
> > index 000000000..fcb7fa860
> > --- /dev/null
> > +++ b/lib/ppc64/stack.c
> > @@ -0,0 +1,55 @@
> > +#include <libcflat.h>
> > +#include <asm/ptrace.h>
> > +#include <stack.h>
> > +
> > +extern char exception_stack_marker[];
> > +
> > +int backtrace_frame(const void *frame, const void **return_addrs, int max_depth)
> > +{
> > +	static int walking;
> > +	int depth = 0;
> > +	const unsigned long *bp = (unsigned long *)frame;
> > +	void *return_addr;
> > +
> > +	asm volatile("" ::: "lr"); /* Force it to save LR */
> > +
> > +	if (walking) {
> > +		printf("RECURSIVE STACK WALK!!!\n");
> > +		return 0;
> > +	}
> > +	walking = 1;
> > +
> > +	bp = (unsigned long *)bp[0];
> > +	return_addr = (void *)bp[2];
> > +
> > +	for (depth = 0; bp && depth < max_depth; depth++) {
> > +		return_addrs[depth] = return_addr;
> > +		if (return_addrs[depth] == 0)
> > +			break;
> > +		if (return_addrs[depth] == exception_stack_marker) {
> > +			struct pt_regs *regs;
> > +
> > +			regs = (void *)bp + STACK_FRAME_OVERHEAD;
> > +			bp = (unsigned long *)bp[0];
> > +			/* Represent interrupt frame with vector number */
> > +			return_addr = (void *)regs->trap;
> > +			if (depth + 1 < max_depth) {
> > +				depth++;
> > +				return_addrs[depth] = return_addr;
> > +				return_addr = (void *)regs->nip;
> > +			}
> > +		} else {
> > +			bp = (unsigned long *)bp[0];
> > +			return_addr = (void *)bp[2];
> > +		}
> > +	}
> > +
> > +	walking = 0;
> > +	return depth;
> > +}
> > +
> > +int backtrace(const void **return_addrs, int max_depth)
> > +{
> > +	return backtrace_frame(__builtin_frame_address(0), return_addrs,
> > +			       max_depth);
> > +}
> > diff --git a/powerpc/Makefile.ppc64 b/powerpc/Makefile.ppc64
> > index b0ed2b104..eb682c226 100644
> > --- a/powerpc/Makefile.ppc64
> > +++ b/powerpc/Makefile.ppc64
> > @@ -17,6 +17,7 @@ cstart.o = $(TEST_DIR)/cstart64.o
> >   reloc.o  = $(TEST_DIR)/reloc64.o
> >   
> >   OBJDIRS += lib/ppc64
> > +cflatobjs += lib/ppc64/stack.o
> >   
> >   # ppc64 specific tests
> >   tests = $(TEST_DIR)/spapr_vpa.elf
> > diff --git a/powerpc/cstart64.S b/powerpc/cstart64.S
> > index 14ab0c6c8..278af84a6 100644
> > --- a/powerpc/cstart64.S
> > +++ b/powerpc/cstart64.S
> > @@ -188,6 +188,7 @@ call_handler:
> >   	.endr
> >   	mfsprg1	r0
> >   	std	r0,GPR1(r1)
> > +	std	r0,0(r1)
> >   
> >   	/* lr, xer, ccr */
> >   
> > @@ -206,12 +207,12 @@ call_handler:
> >   	subi	r31, r31, 0b - start_text
> >   	ld	r2, (p_toc_text - start_text)(r31)
> >   
> > -	/* FIXME: build stack frame */
> > -
> >   	/* call generic handler */
> >   
> >   	addi	r3,r1,STACK_FRAME_OVERHEAD
> >   	bl	do_handle_exception
> > +	.global exception_stack_marker
> > +exception_stack_marker:
> >   
> >   	/* restore context */
> >   
> > @@ -321,6 +322,7 @@ handler_trampoline:
> >   	/* nip and msr */
> >   	mfsrr0	r0
> >   	std	r0, _NIP(r1)
> > +	std	r0, INT_FRAME_SIZE+16(r1)
>
> So if I got that right, INT_FRAME_SIZE+16 points to the stack frame of the 
> function that was running before the interrupt handler? Is it such a good 
> idea to change that here?

No, we switch to exception stack and don't support recursing interrupts
on stack at the moment, so this goes into the initial frame.

> Please add a comment here explaining this line. 
> Thanks!

Yes, good idea.

Thanks,
Nick

>
>
> >   	mfsrr1	r0
> >   	std	r0, _MSR(r1)
> > @@ -337,6 +339,7 @@ handler_htrampoline:
> >   	/* nip and msr */
> >   	mfspr	r0, SPR_HSRR0
> >   	std	r0, _NIP(r1)
> > +	std	r0, INT_FRAME_SIZE+16(r1)
>
> dito.
>
> >   	mfspr	r0, SPR_HSRR1
> >   	std	r0, _MSR(r1)
>
>   Thomas


  reply	other threads:[~2024-03-05  2:12 UTC|newest]

Thread overview: 71+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-02-26 10:11 [kvm-unit-tests PATCH 00/32] powerpc improvements Nicholas Piggin
2024-02-26 10:11 ` [kvm-unit-tests PATCH 01/32] powerpc: Fix KVM caps on POWER9 hosts Nicholas Piggin
2024-02-26 11:33   ` Thomas Huth
2024-02-26 10:11 ` [kvm-unit-tests PATCH 02/32] powerpc: Fix pseries getchar return value Nicholas Piggin
2024-02-26 11:35   ` Thomas Huth
2024-02-26 10:11 ` [kvm-unit-tests PATCH 03/32] powerpc: Fix stack backtrace termination Nicholas Piggin
2024-02-27  8:50   ` Thomas Huth
2024-03-01  9:45     ` Thomas Huth
2024-03-05  2:08       ` Nicholas Piggin
2024-03-05  6:29     ` Nicholas Piggin
2024-03-05  6:59       ` Thomas Huth
2024-02-26 10:11 ` [kvm-unit-tests PATCH 04/32] powerpc: interrupt stack backtracing Nicholas Piggin
2024-02-28 11:46   ` Andrew Jones
2024-02-29  3:50     ` Nicholas Piggin
2024-03-01  9:53   ` Thomas Huth
2024-03-05  2:12     ` Nicholas Piggin [this message]
2024-02-26 10:11 ` [kvm-unit-tests PATCH 05/32] powerpc: Cleanup SPR and MSR definitions Nicholas Piggin
2024-03-01 11:11   ` Thomas Huth
2024-02-26 10:11 ` [kvm-unit-tests PATCH 06/32] powerpc/sprs: Specify SPRs with data rather than code Nicholas Piggin
2024-02-26 10:11 ` [kvm-unit-tests PATCH 07/32] powerpc/sprs: Don't fail changed SPRs that are used by the test harness Nicholas Piggin
2024-03-01 11:15   ` Thomas Huth
2024-03-05  2:14     ` Nicholas Piggin
2024-02-26 10:11 ` [kvm-unit-tests PATCH 08/32] powerpc/sprs: Avoid taking PMU interrupts caused by register fuzzing Nicholas Piggin
2024-03-01 11:25   ` Thomas Huth
2024-02-26 10:11 ` [kvm-unit-tests PATCH 09/32] scripts: allow machine option to be specified in unittests.cfg Nicholas Piggin
2024-02-28 11:47   ` Andrew Jones
2024-02-29  3:52     ` Nicholas Piggin
2024-02-26 10:11 ` [kvm-unit-tests PATCH 10/32] scripts: Accommodate powerpc powernv machine differences Nicholas Piggin
2024-02-28 11:52   ` Andrew Jones
2024-02-26 10:11 ` [kvm-unit-tests PATCH 11/32] powerpc: Support powernv machine with QEMU TCG Nicholas Piggin
2024-02-26 10:11 ` [kvm-unit-tests PATCH 12/32] powerpc: Fix emulator illegal instruction test for powernv Nicholas Piggin
2024-03-01 11:50   ` Thomas Huth
2024-03-05  2:14     ` Nicholas Piggin
2024-02-26 10:11 ` [kvm-unit-tests PATCH 13/32] powerpc/sprs: Test hypervisor registers on powernv machine Nicholas Piggin
2024-02-26 10:12 ` [kvm-unit-tests PATCH 14/32] powerpc: general interrupt tests Nicholas Piggin
2024-03-01 12:41   ` Thomas Huth
2024-03-01 13:45     ` Andrew Jones
2024-03-01 13:57       ` Thomas Huth
2024-03-01 14:14         ` Andrew Jones
2024-03-05  2:35           ` Nicholas Piggin
2024-03-05  2:30       ` Nicholas Piggin
2024-03-05  6:18         ` Thomas Huth
2024-03-05  2:19     ` Nicholas Piggin
2024-03-05  6:26       ` Thomas Huth
2024-03-05 12:12         ` Andrew Jones
2024-02-26 10:12 ` [kvm-unit-tests PATCH 15/32] powerpc: Add rtas stop-self support Nicholas Piggin
2024-02-26 10:12 ` [kvm-unit-tests PATCH 16/32] powerpc: Remove broken SMP exception stack setup Nicholas Piggin
2024-02-26 10:12 ` [kvm-unit-tests PATCH 17/32] arch-run: Fix handling multiple exit status messages Nicholas Piggin
2024-02-28 11:51   ` Andrew Jones
2024-02-26 10:12 ` [kvm-unit-tests PATCH 18/32] powerpc: add SMP and IPI support Nicholas Piggin
2024-02-26 10:12 ` [kvm-unit-tests PATCH 19/32] powerpc: Permit ACCEL=tcg,thread=single Nicholas Piggin
2024-02-26 10:12 ` [kvm-unit-tests PATCH 20/32] powerpc: Avoid using larx/stcx. in spinlocks when only one CPU is running Nicholas Piggin
2024-02-26 10:12 ` [kvm-unit-tests PATCH 21/32] powerpc: Add atomics tests Nicholas Piggin
2024-02-26 10:12 ` [kvm-unit-tests PATCH 22/32] powerpc: Add timebase tests Nicholas Piggin
2024-02-26 10:12 ` [kvm-unit-tests PATCH 23/32] powerpc: Add MMU support Nicholas Piggin
2024-02-28 12:01   ` Andrew Jones
2024-02-26 10:12 ` [kvm-unit-tests PATCH 24/32] common/sieve: Use vmalloc.h for setup_mmu definition Nicholas Piggin
2024-02-28 12:03   ` Andrew Jones
2024-02-26 10:12 ` [kvm-unit-tests PATCH 25/32] common/sieve: Support machines without MMU Nicholas Piggin
2024-02-28 12:04   ` Andrew Jones
2024-02-26 10:12 ` [kvm-unit-tests PATCH 26/32] powerpc: Add sieve.c common test Nicholas Piggin
2024-02-26 10:12 ` [kvm-unit-tests PATCH 27/32] powerpc: add usermode support Nicholas Piggin
2024-02-26 10:12 ` [kvm-unit-tests PATCH 28/32] powerpc: add pmu tests Nicholas Piggin
2024-02-26 10:12 ` [kvm-unit-tests PATCH 29/32] configure: Fail on unknown arch Nicholas Piggin
2024-02-28 12:08   ` Andrew Jones
2024-02-26 10:12 ` [kvm-unit-tests PATCH 30/32] configure: Make arch_libdir a first-class entity Nicholas Piggin
2024-02-28 12:13   ` Andrew Jones
2024-02-26 10:12 ` [kvm-unit-tests PATCH 31/32] powerpc: Remove remnants of ppc64 directory and build structure Nicholas Piggin
2024-02-26 10:12 ` [kvm-unit-tests PATCH 32/32] powerpc: gitlab CI update Nicholas Piggin
2024-02-28 12:16   ` Andrew Jones
2024-02-29  3:55     ` Nicholas Piggin

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CZLGB4BCREUU.SZ1AQ0LTGNKB@wheely \
    --to=npiggin@gmail.com \
    --cc=andrew.jones@linux.dev \
    --cc=joel@jms.id.au \
    --cc=kvm@vger.kernel.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=lvivier@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=thuth@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox