Skip to contentSkip to conversation

single step for KVM HV

RFC PATCH Superseded Fabiano Rosas 3 participants 6 messages Nov 19, 2018 – Jan 16, 2019

v1 of 6 Superseded Latest is v6 →

lore
Share & tools
lore
#0 lore raw

cover single step for KVM HV

Single stepping via GDB/gdbstub is currently not working with KVM HV. When asking for a single step (stepi), KVM simply ignores the request and execution continues.

This has the direct effect of breaking GDB's 'step', 'stepi', 'next', 'nexti' commands. The 'continue' command is also affected since continuing right after a breakpoint requires that GDB first perform a single step so that the breakpoint can be re-inserted before continuing - in this case the breakpoint is not re-inserted and it won't hit again.

The issue here is that single stepping in POWER makes use of an interrupt (Trace Interrupt [1]) that does not reach the hypervisor, so while the single step would happen if properly triggered, it would not cause an exit to KVM so there would be no way of handing control back to GDB. Aside from that, the guest kernel is not prepared to deal with such an interrupt in kernel mode (when not using KGDB, or some other debugging facility) and it causes an Oops.

This series implements a "software single step" approach that makes use of: i) the Trace Interrupt to perform the step inside the guest and ii) a breakpoint at the Trace Interrupt handler address to cause a vm exit (Emulation Assist) so that we can return control to QEMU.

With (i), we basically get the single step for free, without having to discover what are the possible targets of instructions that divert execution.

With (ii), we hide the single step from the guest and keep all of the step logic in QEMU.

This was so far tested with single and multiple vcpus and with GDB scheduler locking on and off [2].

I have not fully explored yet the potential issues when using debuggers simultaneously inside and outside the guest, however I was able to single step the ptrace code while single stepping a userspace program inside the guest with GDB.

I'm looking for feedback on the general approach before I develop this further.

1- PowerISA Section 6.5.15 - Trace Interrupt
2- https://sourceware.org/gdb/onlinedocs/gdb/All_002dStop-Mode.html

Fabiano Rosas (1):
target/ppc: support single stepping with KVM HV

 accel/kvm/kvm-all.c  | 10 +++++++
 exec.c               |  1 +
 include/sysemu/kvm.h |  4 +++
 target/arm/kvm.c     |  4 +++
 target/i386/kvm.c    |  4 +++
 target/ppc/kvm.c     | 65 +++++++++++++++++++++++++++++++++++++++++++-
 target/s390x/kvm.c   |  4 +++
 7 files changed, 91 insertions(+), 1 deletion(-)

--
2.17.1

#1 lore raw

RFC PATCH 1/1 target/ppc: support single stepping with KVM HV

The hardware singlestep mechanism in POWER works via a Trace Interrupt (0xd00) that happens after any instruction executes, whenever MSR_SE = 1 (PowerISA Section 6.5.15 - Trace Interrupt).

However, with kvm_hv, the Trace Interrupt happens inside the guest and KVM has no visibility of it. Therefore, when the gdbstub uses the KVM_SET_GUEST_DEBUG ioctl to enable singlestep, KVM simply ignores it.

This patch takes advantage of the Trace Interrupt to perform the step inside the guest, but uses a breakpoint at the Trace Interrupt handler to return control to KVM. The exit is treated by KVM as a regular breakpoint and it returns to the host (and QEMU eventually).

Before signalling GDB, QEMU sets the Next Instruction Pointer to the instruction following the one being stepped, effectively skipping the interrupt handler execution and hiding the trace interrupt breakpoint from GDB.

This approach works with both of GDB's 'scheduler-locking' options (off, step).

---
 accel/kvm/kvm-all.c  | 10 +++++++
 exec.c               |  1 +
 include/sysemu/kvm.h |  4 +++
 target/arm/kvm.c     |  4 +++
 target/i386/kvm.c    |  4 +++
 target/ppc/kvm.c     | 65 +++++++++++++++++++++++++++++++++++++++++++-
 target/s390x/kvm.c   |  4 +++
 7 files changed, 91 insertions(+), 1 deletion(-)
accel/kvm/kvm-all.c +10 -0
index 4880a05399..4fb7199a15 100644
@@ -2313,6 +2313,11 @@ int kvm_update_guest_debug(CPUState *cpu, unsigned long reinject_trap)
     return data.err;
 }
 
+void kvm_set_singlestep(CPUState *cs, int enabled)
+{
+    kvm_arch_set_singlestep(cs, enabled);
+}
+
 int kvm_insert_breakpoint(CPUState *cpu, target_ulong addr,
                           target_ulong len, int type)
 {
@@ -2439,6 +2444,11 @@ int kvm_remove_breakpoint(CPUState *cpu, target_ulong addr,
 void kvm_remove_all_breakpoints(CPUState *cpu)
 {
 }
+
+void kvm_set_singlestep(CPUState *cs, int enabled)
+{
+}
+
 #endif /* !KVM_CAP_SET_GUEST_DEBUG */
 
 static int kvm_set_signal_mask(CPUState *cpu, const sigset_t *sigset)
exec.c +1 -0
index bb6170dbff..55614822c3 100644
@@ -1233,6 +1233,7 @@ void cpu_single_step(CPUState *cpu, int enabled)
     if (cpu->singlestep_enabled != enabled) {
         cpu->singlestep_enabled = enabled;
         if (kvm_enabled()) {
+            kvm_set_singlestep(cpu, enabled);
             kvm_update_guest_debug(cpu, 0);
         } else {
             /* must flush all the translated code to avoid inconsistencies */
include/sysemu/kvm.h +4 -0
index 97d8d9d0d5..a01a8d58dd 100644
@@ -259,6 +259,8 @@ int kvm_remove_breakpoint(CPUState *cpu, target_ulong addr,
 void kvm_remove_all_breakpoints(CPUState *cpu);
 int kvm_update_guest_debug(CPUState *cpu, unsigned long reinject_trap);
 
+void kvm_set_singlestep(CPUState *cpu, int enabled);
+
 int kvm_on_sigbus_vcpu(CPUState *cpu, int code, void *addr);
 int kvm_on_sigbus(int code, void *addr);
 
@@ -431,6 +433,8 @@ void kvm_arch_remove_all_hw_breakpoints(void);
 
 void kvm_arch_update_guest_debug(CPUState *cpu, struct kvm_guest_debug *dbg);
 
+void kvm_arch_set_singlestep(CPUState *cpu, int enabled);
+
 bool kvm_arch_stop_on_emulation_error(CPUState *cpu);
 
 int kvm_check_extension(KVMState *s, unsigned int extension);
target/arm/kvm.c +4 -0
index 44dd0ce6ce..dd8e43ab7e 100644
@@ -670,6 +670,10 @@ int kvm_arch_process_async_events(CPUState *cs)
     return 0;
 }
 
+void kvm_arch_set_singlestep(CPUState *cs, int enabled)
+{
+}
+
 /* The #ifdef protections are until 32bit headers are imported and can
  * be removed once both 32 and 64 bit reach feature parity.
  */
target/i386/kvm.c +4 -0
index f524e7d929..ba56f2ee1f 100644
@@ -3521,6 +3521,10 @@ static int kvm_handle_debug(X86CPU *cpu,
     return ret;
 }
 
+void kvm_arch_set_singlestep(CPUState *cs, int enabled)
+{
+}
+
 void kvm_arch_update_guest_debug(CPUState *cpu, struct kvm_guest_debug *dbg)
 {
     const uint8_t type_code[] = {
target/ppc/kvm.c +64 -1

95 lines not on this page · raw message

target/s390x/kvm.c +4 -0
index 2ebf26adfe..4bde183458 100644
@@ -975,6 +975,10 @@ void kvm_arch_remove_all_hw_breakpoints(void)
     hw_breakpoints = NULL;
 }
 
+void kvm_arch_set_singlestep(CPUState *cs, int enabled)
+{
+}
+
 void kvm_arch_update_guest_debug(CPUState *cpu, struct kvm_guest_debug *dbg)
 {
     int i;
Signature · 1 line
2.17.1
#2 lore raw

Re: [Qemu-devel] [RFC PATCH 1/1] target/ppc: support single stepping with KVM HV

Hi Fabiano,

You should Cc the relevant maintainers to get more attention. You can check this wiki page:
https://wiki.qemu.org/Contribute/SubmitAPatch#CC_the_relevant_maintainer

$ ./scripts/get_maintainer.pl -f accel/kvm/kvm-all.c include/sysemu/kvm.h target/*/kvm.c
Paolo Bonzini <[email protected]> (supporter:Overall)
Peter Maydell <[email protected]> (maintainer:ARM)
Marcelo Tosatti <[email protected]> (supporter:X86)
Richard Henderson <[email protected]> (maintainer:X86)
Eduardo Habkost <[email protected]> (maintainer:X86)
James Hogan <[email protected]> (maintainer:MIPS)
Stefan Markovic <[email protected]> (reviewer:MIPS)
Aurelien Jarno <[email protected]> (maintainer:MIPS)
Aleksandar Markovic <[email protected]> (maintainer:MIPS)
David Gibson <[email protected]> (maintainer:PPC)
Christian Borntraeger <[email protected]> (maintainer:S390)
Cornelia Huck <[email protected]> (maintainer:S390)
David Hildenbrand <[email protected]> (maintainer:S390)
[email protected] (open list:Overall)
[email protected] (open list:All patches CC here)
[email protected] (open list:ARM)
[email protected] (open list:PowerPC)
[email protected] (open list:S390)
Quoted 37 lines · Fabiano Rosas

The hardware singlestep mechanism in POWER works via a Trace Interrupt (0xd00) that happens after any instruction executes, whenever MSR_SE = 1 (PowerISA Section 6.5.15 - Trace Interrupt).

However, with kvm_hv, the Trace Interrupt happens inside the guest and KVM has no visibility of it. Therefore, when the gdbstub uses the KVM_SET_GUEST_DEBUG ioctl to enable singlestep, KVM simply ignores it.

This patch takes advantage of the Trace Interrupt to perform the step inside the guest, but uses a breakpoint at the Trace Interrupt handler to return control to KVM. The exit is treated by KVM as a regular breakpoint and it returns to the host (and QEMU eventually).

Before signalling GDB, QEMU sets the Next Instruction Pointer to the instruction following the one being stepped, effectively skipping the interrupt handler execution and hiding the trace interrupt breakpoint from GDB.

This approach works with both of GDB's 'scheduler-locking' options (off, step).

---
  accel/kvm/kvm-all.c  | 10 +++++++
  exec.c               |  1 +
  include/sysemu/kvm.h |  4 +++
  target/arm/kvm.c     |  4 +++
  target/i386/kvm.c    |  4 +++
  target/ppc/kvm.c     | 65 +++++++++++++++++++++++++++++++++++++++++++-
  target/s390x/kvm.c   |  4 +++
  7 files changed, 91 insertions(+), 1 deletion(-)
accel/kvm/kvm-all.c +0 -0
index 4880a05399..4fb7199a15 100644

accel/kvm/kvm-all.c int kvm_update_guest_debug(CPUState *cpu, unsigned long reinject_trap) · line 2318

Quoted 6 lines
accel/kvm/kvm-all.c +3 -0
@@ -2313,6 +2313,11 @@ int kvm_update_guest_debug(CPUState *cpu, unsigned long reinject_trap)
      return data.err;
  }
 
+void kvm_set_singlestep(CPUState *cs, int enabled)
+{
+    kvm_arch_set_singlestep(cs, enabled);

This won't link on MIPS, you miss the stub there.

Quoted 94 lines
accel/kvm/kvm-all.c +7 -0
+}
+
  int kvm_insert_breakpoint(CPUState *cpu, target_ulong addr,
                            target_ulong len, int type)
  {
@@ -2439,6 +2444,11 @@ int kvm_remove_breakpoint(CPUState *cpu, target_ulong addr,
  void kvm_remove_all_breakpoints(CPUState *cpu)
  {
  }
+
+void kvm_set_singlestep(CPUState *cs, int enabled)
+{
+}
+
  #endif /* !KVM_CAP_SET_GUEST_DEBUG */
 
  static int kvm_set_signal_mask(CPUState *cpu, const sigset_t *sigset)
exec.c +1 -0
index bb6170dbff..55614822c3 100644
@@ -1233,6 +1233,7 @@ void cpu_single_step(CPUState *cpu, int enabled)
      if (cpu->singlestep_enabled != enabled) {
          cpu->singlestep_enabled = enabled;
          if (kvm_enabled()) {
+            kvm_set_singlestep(cpu, enabled);
              kvm_update_guest_debug(cpu, 0);
          } else {
              /* must flush all the translated code to avoid inconsistencies */
include/sysemu/kvm.h +4 -0
index 97d8d9d0d5..a01a8d58dd 100644
@@ -259,6 +259,8 @@ int kvm_remove_breakpoint(CPUState *cpu, target_ulong addr,
  void kvm_remove_all_breakpoints(CPUState *cpu);
  int kvm_update_guest_debug(CPUState *cpu, unsigned long reinject_trap);
 
+void kvm_set_singlestep(CPUState *cpu, int enabled);
+
  int kvm_on_sigbus_vcpu(CPUState *cpu, int code, void *addr);
  int kvm_on_sigbus(int code, void *addr);
 
@@ -431,6 +433,8 @@ void kvm_arch_remove_all_hw_breakpoints(void);
 
  void kvm_arch_update_guest_debug(CPUState *cpu, struct kvm_guest_debug *dbg);
 
+void kvm_arch_set_singlestep(CPUState *cpu, int enabled);
+
  bool kvm_arch_stop_on_emulation_error(CPUState *cpu);
 
  int kvm_check_extension(KVMState *s, unsigned int extension);
target/arm/kvm.c +4 -0
index 44dd0ce6ce..dd8e43ab7e 100644
@@ -670,6 +670,10 @@ int kvm_arch_process_async_events(CPUState *cs)
      return 0;
  }
 
+void kvm_arch_set_singlestep(CPUState *cs, int enabled)
+{
+}
+
  /* The #ifdef protections are until 32bit headers are imported and can
   * be removed once both 32 and 64 bit reach feature parity.
   */
target/i386/kvm.c +4 -0
index f524e7d929..ba56f2ee1f 100644
@@ -3521,6 +3521,10 @@ static int kvm_handle_debug(X86CPU *cpu,
      return ret;
  }
 
+void kvm_arch_set_singlestep(CPUState *cs, int enabled)
+{
+}
+
  void kvm_arch_update_guest_debug(CPUState *cpu, struct kvm_guest_debug *dbg)
  {
      const uint8_t type_code[] = {
target/ppc/kvm.c +1 -0
index f81327d6cd..3af5e91997 100644
@@ -94,6 +94,7 @@ static int cap_ppc_safe_indirect_branch;
  static int cap_ppc_nested_kvm_hv;
 
  static uint32_t debug_inst_opcode;
+static target_ulong trace_handler_addr;
 
  /* XXX We have a race condition where we actually have a level triggered
   *     interrupt, but the infrastructure can't expose that yet, so the guest

target/ppc/kvm.c int kvm_arch_init_vcpu(CPUState *cs) · line 514

Quoted 5 lines
target/ppc/kvm.c +2 -0
@@ -509,6 +510,9 @@ int kvm_arch_init_vcpu(CPUState *cs)
      kvm_get_one_reg(cs, KVM_REG_PPC_DEBUG_INST, &debug_inst_opcode);
      kvmppc_hw_debug_points_init(cenv);
 
+    trace_handler_addr = (cenv->excp_vectors[POWERPC_EXCP_TRACE] |
+                          0xc000000000004000ull);

Can you add a definition for this magic value?

Quoted 52 lines
target/ppc/kvm.c +38 -0
+
      return ret;
  }
 
@@ -1551,6 +1555,28 @@ void kvm_arch_remove_all_hw_breakpoints(void)
      nb_hw_breakpoint = nb_hw_watchpoint = 0;
  }
 
+void kvm_arch_set_singlestep(CPUState *cs, int enabled)
+{
+    PowerPCCPU *cpu = POWERPC_CPU(cs);
+    CPUPPCState *env = &cpu->env;
+
+    if (kvmppc_is_pr(kvm_state)) {
+            return;
+    }
+
+    if (enabled) {
+        /* MSR_SE = 1 will cause a Trace Interrupt in the guest after
+         * the next instruction executes. */
+        env->msr |= (1ULL << MSR_SE);
+
+        /* We set a breakpoint at the interrupt handler address so
+         * that the singlestep will be seen by KVM (this is treated by
+         * KVM like an ordinary breakpoint) and control is returned to
+         * QEMU. */
+        kvm_insert_breakpoint(cs, trace_handler_addr, 4, GDB_BREAKPOINT_SW);
+    }
+}
+
  void kvm_arch_update_guest_debug(CPUState *cs, struct kvm_guest_debug *dbg)
  {
      int n;
@@ -1590,6 +1616,43 @@ void kvm_arch_update_guest_debug(CPUState *cs, struct kvm_guest_debug *dbg)
      }
  }
 
+static int kvm_handle_singlestep(CPUState *cs,
+                                 struct kvm_debug_exit_arch *arch_info)
+{
+    PowerPCCPU *cpu = POWERPC_CPU(cs);
+    CPUPPCState *env = &cpu->env;
+    target_ulong msr = env->msr;
+    uint32_t insn;
+    int ret = 1;
+    int reg;
+
+    if (kvmppc_is_pr(kvm_state)) {
+        return ret;
+    }
+
+    if (arch_info->address == trace_handler_addr) {

target/ppc/kvm.c void kvm_arch_update_guest_debug(CPUState *cs, struct kvm_guest_debug *dbg) · line 1643

Quoted 10 lines
target/ppc/kvm.c +10 -0
+        cpu_synchronize_state(cs);
+        kvm_remove_breakpoint(cs, trace_handler_addr, 4, GDB_BREAKPOINT_SW);
+
+        cpu_memory_rw_debug(cs, env->spr[SPR_SRR0] - 4, (uint8_t *)&insn,
+                            sizeof(insn), 0);
+
+        /* If the last instruction was a mfmsr, make sure that the
+         * MSR_SE bit is not set to avoid the guest kernel knowing
+         * that it is being single-stepped */
+        if (((insn >> 26) & 0x3f) == 31 && ((insn >> 1) & 0x3ff) == 83) {

You can use the extract32() function to extract bits, it is easier to review:

    if (extract32(insn, 26, 6) == 31 && extract32(insn, 1, 10) == 83) {

target/ppc/kvm.c void kvm_arch_update_guest_debug(CPUState *cs, struct kvm_guest_debug *dbg) · line 1644

Quoted 1 line
target/ppc/kvm.c +1 -0
+            reg = (insn >> 21) & 0x1f;

Ditto.

Regards,

Phil.

Quoted 39 lines
target/ppc/kvm.c +12 -1
+            env->gpr[reg] &= ~(1ULL << MSR_SE);
+        }
+
+        env->nip = env->spr[SPR_SRR0];
+        env->msr = msr &= ~(1ULL << MSR_SE);
+        cpu_synchronize_state(cs);
+    }
+
+    return ret;
+}
+
  static int kvm_handle_debug(PowerPCCPU *cpu, struct kvm_run *run)
  {
      CPUState *cs = CPU(cpu);
@@ -1600,7 +1663,7 @@ static int kvm_handle_debug(PowerPCCPU *cpu, struct kvm_run *run)
      int flag = 0;
 
      if (cs->singlestep_enabled) {
-        handle = 1;
+        handle = kvm_handle_singlestep(cs, arch_info);
      } else if (arch_info->status) {
          if (nb_hw_breakpoint + nb_hw_watchpoint > 0) {
              if (arch_info->status & KVMPPC_DEBUG_BREAKPOINT) {
target/s390x/kvm.c +4 -0
index 2ebf26adfe..4bde183458 100644
@@ -975,6 +975,10 @@ void kvm_arch_remove_all_hw_breakpoints(void)
      hw_breakpoints = NULL;
  }
 
+void kvm_arch_set_singlestep(CPUState *cs, int enabled)
+{
+}
+
  void kvm_arch_update_guest_debug(CPUState *cpu, struct kvm_guest_debug *dbg)
  {
      int i;
#3 lore raw

Re: [Qemu-devel] [RFC PATCH 1/1] target/ppc: support single stepping with KVM HV

Quoted 273 lines · Philippe Mathieu-Daudé

Hi Fabiano,

You should Cc the relevant maintainers to get more attention. You can check this wiki page:
https://wiki.qemu.org/Contribute/SubmitAPatch#CC_the_relevant_maintainer

$ ./scripts/get_maintainer.pl -f accel/kvm/kvm-all.c include/sysemu/kvm.h target/*/kvm.c
Paolo Bonzini <[email protected]> (supporter:Overall)
Peter Maydell <[email protected]> (maintainer:ARM)
Marcelo Tosatti <[email protected]> (supporter:X86)
Richard Henderson <[email protected]> (maintainer:X86)
Eduardo Habkost <[email protected]> (maintainer:X86)
James Hogan <[email protected]> (maintainer:MIPS)
Stefan Markovic <[email protected]> (reviewer:MIPS)
Aurelien Jarno <[email protected]> (maintainer:MIPS)
Aleksandar Markovic <[email protected]> (maintainer:MIPS)
David Gibson <[email protected]> (maintainer:PPC)
Christian Borntraeger <[email protected]> (maintainer:S390)
Cornelia Huck <[email protected]> (maintainer:S390)
David Hildenbrand <[email protected]> (maintainer:S390)
[email protected] (open list:Overall)
[email protected] (open list:All patches CC here)
[email protected] (open list:ARM)
[email protected] (open list:PowerPC)
[email protected] (open list:S390)
Quoted 37 lines · Fabiano Rosas

The hardware singlestep mechanism in POWER works via a Trace Interrupt (0xd00) that happens after any instruction executes, whenever MSR_SE = 1 (PowerISA Section 6.5.15 - Trace Interrupt).

However, with kvm_hv, the Trace Interrupt happens inside the guest and KVM has no visibility of it. Therefore, when the gdbstub uses the KVM_SET_GUEST_DEBUG ioctl to enable singlestep, KVM simply ignores it.

This patch takes advantage of the Trace Interrupt to perform the step inside the guest, but uses a breakpoint at the Trace Interrupt handler to return control to KVM. The exit is treated by KVM as a regular breakpoint and it returns to the host (and QEMU eventually).

Before signalling GDB, QEMU sets the Next Instruction Pointer to the instruction following the one being stepped, effectively skipping the interrupt handler execution and hiding the trace interrupt breakpoint from GDB.

This approach works with both of GDB's 'scheduler-locking' options (off, step).

---
  accel/kvm/kvm-all.c  | 10 +++++++
  exec.c               |  1 +
  include/sysemu/kvm.h |  4 +++
  target/arm/kvm.c     |  4 +++
  target/i386/kvm.c    |  4 +++
  target/ppc/kvm.c     | 65 +++++++++++++++++++++++++++++++++++++++++++- target/s390x/kvm.c   |  4 +++

  7 files changed, 91 insertions(+), 1 deletion(-)
accel/kvm/kvm-all.c +0 -0
index 4880a05399..4fb7199a15 100644

accel/kvm/kvm-all.c int kvm_update_guest_debug(CPUState *cpu, · line 2318

Quoted 6 lines
accel/kvm/kvm-all.c +3 -0
@@ -2313,6 +2313,11 @@ int kvm_update_guest_debug(CPUState *cpu, 
 unsigned long reinject_trap)
       return data.err;
   }
+void kvm_set_singlestep(CPUState *cs, int enabled)
+{
+    kvm_arch_set_singlestep(cs, enabled);

This won't link on MIPS, you miss the stub there.

Quoted 86 lines
accel/kvm/kvm-all.c +7 -0
+}
+
   int kvm_insert_breakpoint(CPUState *cpu, target_ulong addr,
                             target_ulong len, int type)
   {
@@ -2439,6 +2444,11 @@ int kvm_remove_breakpoint(CPUState *cpu, 
 target_ulong addr,
   void kvm_remove_all_breakpoints(CPUState *cpu)
   {
   }
+
+void kvm_set_singlestep(CPUState *cs, int enabled)
+{
+}
+
   #endif /* !KVM_CAP_SET_GUEST_DEBUG */
   static int kvm_set_signal_mask(CPUState *cpu, const sigset_t *sigset)
exec.c +1 -0
index bb6170dbff..55614822c3 100644
@@ -1233,6 +1233,7 @@ void cpu_single_step(CPUState *cpu, int enabled)
       if (cpu->singlestep_enabled != enabled) {
           cpu->singlestep_enabled = enabled;
           if (kvm_enabled()) {
+            kvm_set_singlestep(cpu, enabled);
               kvm_update_guest_debug(cpu, 0);
           } else {
               /* must flush all the translated code to avoid inconsistencies */
include/sysemu/kvm.h +4 -0
index 97d8d9d0d5..a01a8d58dd 100644
@@ -259,6 +259,8 @@ int kvm_remove_breakpoint(CPUState *cpu, 
 target_ulong addr,
   void kvm_remove_all_breakpoints(CPUState *cpu);
   int kvm_update_guest_debug(CPUState *cpu, unsigned long reinject_trap);
+void kvm_set_singlestep(CPUState *cpu, int enabled);
+
   int kvm_on_sigbus_vcpu(CPUState *cpu, int code, void *addr);
   int kvm_on_sigbus(int code, void *addr);
@@ -431,6 +433,8 @@ void kvm_arch_remove_all_hw_breakpoints(void);
   void kvm_arch_update_guest_debug(CPUState *cpu, struct kvm_guest_debug *dbg);
+void kvm_arch_set_singlestep(CPUState *cpu, int enabled);
+
   bool kvm_arch_stop_on_emulation_error(CPUState *cpu);
   int kvm_check_extension(KVMState *s, unsigned int extension);
target/arm/kvm.c +4 -0
index 44dd0ce6ce..dd8e43ab7e 100644
@@ -670,6 +670,10 @@ int kvm_arch_process_async_events(CPUState *cs)
       return 0;
   }
+void kvm_arch_set_singlestep(CPUState *cs, int enabled)
+{
+}
+
   /* The #ifdef protections are until 32bit headers are imported and can
    * be removed once both 32 and 64 bit reach feature parity.
    */
target/i386/kvm.c +4 -0
index f524e7d929..ba56f2ee1f 100644
@@ -3521,6 +3521,10 @@ static int kvm_handle_debug(X86CPU *cpu,
       return ret;
   }
+void kvm_arch_set_singlestep(CPUState *cs, int enabled)
+{
+}
+
   void kvm_arch_update_guest_debug(CPUState *cpu, struct kvm_guest_debug *dbg)
   {
       const uint8_t type_code[] = {
target/ppc/kvm.c +1 -0
index f81327d6cd..3af5e91997 100644
@@ -94,6 +94,7 @@ static int cap_ppc_safe_indirect_branch;
   static int cap_ppc_nested_kvm_hv;
   static uint32_t debug_inst_opcode;
+static target_ulong trace_handler_addr;
   /* XXX We have a race condition where we actually have a level triggered
    *     interrupt, but the infrastructure can't expose that yet, so the guest

target/ppc/kvm.c int kvm_arch_init_vcpu(CPUState *cs) · line 513

Quoted 4 lines
target/ppc/kvm.c +2 -0
@@ -509,6 +510,9 @@ int kvm_arch_init_vcpu(CPUState *cs)
       kvm_get_one_reg(cs, KVM_REG_PPC_DEBUG_INST, &debug_inst_opcode);
       kvmppc_hw_debug_points_init(cenv);
+    trace_handler_addr = (cenv->excp_vectors[POWERPC_EXCP_TRACE] |
+                          0xc000000000004000ull);

Can you add a definition for this magic value?

Quoted 54 lines
target/ppc/kvm.c +40 -0
+
       return ret;
   }
@@ -1551,6 +1555,28 @@ void kvm_arch_remove_all_hw_breakpoints(void)
       nb_hw_breakpoint = nb_hw_watchpoint = 0;
   }
+void kvm_arch_set_singlestep(CPUState *cs, int enabled)
+{
+    PowerPCCPU *cpu = POWERPC_CPU(cs);
+    CPUPPCState *env = &cpu->env;
+
+    if (kvmppc_is_pr(kvm_state)) {
+            return;
+    }
+
+    if (enabled) {
+        /* MSR_SE = 1 will cause a Trace Interrupt in the guest after
+         * the next instruction executes. */
+        env->msr |= (1ULL << MSR_SE);
+
+        /* We set a breakpoint at the interrupt handler address so
+         * that the singlestep will be seen by KVM (this is treated by
+         * KVM like an ordinary breakpoint) and control is returned to
+         * QEMU. */
+        kvm_insert_breakpoint(cs, trace_handler_addr, 4, 
 GDB_BREAKPOINT_SW);
+    }
+}
+
   void kvm_arch_update_guest_debug(CPUState *cs, struct kvm_guest_debug *dbg)
   {
       int n;
@@ -1590,6 +1616,43 @@ void kvm_arch_update_guest_debug(CPUState *cs, 
 struct kvm_guest_debug *dbg)
       }
   }
+static int kvm_handle_singlestep(CPUState *cs,
+                                 struct kvm_debug_exit_arch *arch_info)
+{
+    PowerPCCPU *cpu = POWERPC_CPU(cs);
+    CPUPPCState *env = &cpu->env;
+    target_ulong msr = env->msr;
+    uint32_t insn;
+    int ret = 1;
+    int reg;
+
+    if (kvmppc_is_pr(kvm_state)) {
+        return ret;
+    }
+
+    if (arch_info->address == trace_handler_addr) {
+        cpu_synchronize_state(cs);
+        kvm_remove_breakpoint(cs, trace_handler_addr, 4, 
 GDB_BREAKPOINT_SW);

target/ppc/kvm.c void kvm_arch_update_guest_debug(CPUState *cs, · line 1646

Quoted 10 lines
target/ppc/kvm.c +8 -0
+
+        cpu_memory_rw_debug(cs, env->spr[SPR_SRR0] - 4, (uint8_t 
 *)&insn,
+                            sizeof(insn), 0);
+
+        /* If the last instruction was a mfmsr, make sure that the
+         * MSR_SE bit is not set to avoid the guest kernel knowing
+         * that it is being single-stepped */
+        if (((insn >> 26) & 0x3f) == 31 && ((insn >> 1) & 0x3ff) == 
 83) {

You can use the extract32() function to extract bits, it is easier to review:

    if (extract32(insn, 26, 6) == 31 && extract32(insn, 1, 10) == 83) {

target/ppc/kvm.c void kvm_arch_update_guest_debug(CPUState *cs, struct kvm_guest_debug *dbg) · line 1644

Quoted 1 line
target/ppc/kvm.c +1 -0
+            reg = (insn >> 21) & 0x1f;

Ditto.

Regards,

Phil.

Quoted 30 lines
target/ppc/kvm.c +11 -0
+            env->gpr[reg] &= ~(1ULL << MSR_SE);
+        }
+
+        env->nip = env->spr[SPR_SRR0];
+        env->msr = msr &= ~(1ULL << MSR_SE);
+        cpu_synchronize_state(cs);
+    }
+
+    return ret;
+}
+

  static int kvm_handle_debug(PowerPCCPU *cpu, struct kvm_run *run) {
      CPUState *cs = CPU(cpu);

target/ppc/kvm.c +1 -1
@@ -1600,7 +1663,7 @@ static int kvm_handle_debug(PowerPCCPU *cpu, 
 struct kvm_run *run)
       int flag = 0;
       if (cs->singlestep_enabled) {
-        handle = 1;
+        handle = kvm_handle_singlestep(cs, arch_info);
       } else if (arch_info->status) {
           if (nb_hw_breakpoint + nb_hw_watchpoint > 0) {
               if (arch_info->status & KVMPPC_DEBUG_BREAKPOINT) {
target/s390x/kvm.c +0 -0
index 2ebf26adfe..4bde183458 100644
@@ -975,6 +975,10 @@ void kvm_arch_remove_all_hw_breakpoints(void)
       hw_breakpoints = NULL;
   }
Quoted 8 lines

target/s390x/kvm.c void kvm_arch_remove_all_hw_breakpoints(void)

Quoted 8 lines
target/s390x/kvm.c +4 -0
+void kvm_arch_set_singlestep(CPUState *cs, int enabled)
+{
+}
+
   void kvm_arch_update_guest_debug(CPUState *cpu, struct kvm_guest_debug *dbg)
   {
       int i;
 

I'd split this patch in 2:

- Add kvm_set_singlestep() and kvm_arch_set_singlestep() stubs
- Implement PPC kvm_arch_set_singlestep()

#4 lore raw

Re: [Qemu-devel] [RFC PATCH 1/1] target/ppc: support single stepping with KVM HV

In reply to Philippe Mathieu-Daudé

Fabiano,

Are you planning on reposting this any time soon? I am interested in the feature. Thanks.

Quoted 291 lines · Philippe Mathieu-Daudé

Hi Fabiano,

You should Cc the relevant maintainers to get more attention. You can check this wiki page:
https://wiki.qemu.org/Contribute/SubmitAPatch#CC_the_relevant_maintainer

$ ./scripts/get_maintainer.pl -f accel/kvm/kvm-all.c
include/sysemu/kvm.h target/*/kvm.c
Paolo Bonzini <[email protected]> (supporter:Overall)
Peter Maydell <[email protected]> (maintainer:ARM)
Marcelo Tosatti <[email protected]> (supporter:X86)
Richard Henderson <[email protected]> (maintainer:X86)
Eduardo Habkost <[email protected]> (maintainer:X86)
James Hogan <[email protected]> (maintainer:MIPS)
Stefan Markovic <[email protected]> (reviewer:MIPS)
Aurelien Jarno <[email protected]> (maintainer:MIPS)
Aleksandar Markovic <[email protected]> (maintainer:MIPS)
David Gibson <[email protected]> (maintainer:PPC)
Christian Borntraeger <[email protected]> (maintainer:S390)
Cornelia Huck <[email protected]> (maintainer:S390)
David Hildenbrand <[email protected]> (maintainer:S390)
[email protected] (open list:Overall)
[email protected] (open list:All patches CC here)
[email protected] (open list:ARM)
[email protected] (open list:PowerPC)
[email protected] (open list:S390)
Quoted 37 lines · Fabiano Rosas

The hardware singlestep mechanism in POWER works via a Trace Interrupt (0xd00) that happens after any instruction executes, whenever MSR_SE = 1 (PowerISA Section 6.5.15 - Trace Interrupt).

However, with kvm_hv, the Trace Interrupt happens inside the guest and KVM has no visibility of it. Therefore, when the gdbstub uses the KVM_SET_GUEST_DEBUG ioctl to enable singlestep, KVM simply ignores it.

This patch takes advantage of the Trace Interrupt to perform the step inside the guest, but uses a breakpoint at the Trace Interrupt handler to return control to KVM. The exit is treated by KVM as a regular breakpoint and it returns to the host (and QEMU eventually).

Before signalling GDB, QEMU sets the Next Instruction Pointer to the instruction following the one being stepped, effectively skipping the interrupt handler execution and hiding the trace interrupt breakpoint from GDB.

This approach works with both of GDB's 'scheduler-locking' options (off, step).

---
  accel/kvm/kvm-all.c  | 10 +++++++
  exec.c               |  1 +
  include/sysemu/kvm.h |  4 +++
  target/arm/kvm.c     |  4 +++
  target/i386/kvm.c    |  4 +++
  target/ppc/kvm.c     | 65 +++++++++++++++++++++++++++++++++++++++++++- target/s390x/kvm.c   |  4 +++

  7 files changed, 91 insertions(+), 1 deletion(-)
accel/kvm/kvm-all.c +0 -0
index 4880a05399..4fb7199a15 100644

accel/kvm/kvm-all.c int kvm_update_guest_debug(CPUState *cpu, · line 2318

Quoted 6 lines
accel/kvm/kvm-all.c +2 -0
@@ -2313,6 +2313,11 @@ int kvm_update_guest_debug(CPUState *cpu,
 unsigned long reinject_trap)
       return data.err;
   }
   +void kvm_set_singlestep(CPUState *cs, int enabled)
+{
+    kvm_arch_set_singlestep(cs, enabled);

This won't link on MIPS, you miss the stub there.

Quoted 93 lines
accel/kvm/kvm-all.c +2 -0
+}
+
   int kvm_insert_breakpoint(CPUState *cpu, target_ulong addr,
                             target_ulong len, int type)

  {

accel/kvm/kvm-all.c +5 -0
@@ -2439,6 +2444,11 @@ int kvm_remove_breakpoint(CPUState *cpu,
 target_ulong addr,
   void kvm_remove_all_breakpoints(CPUState *cpu)
   {
   }
+
+void kvm_set_singlestep(CPUState *cs, int enabled)
+{
+}
+
   #endif /* !KVM_CAP_SET_GUEST_DEBUG */
     static int kvm_set_signal_mask(CPUState *cpu, const sigset_t *sigset)
exec.c +1 -0
index bb6170dbff..55614822c3 100644
@@ -1233,6 +1233,7 @@ void cpu_single_step(CPUState *cpu, int enabled)
       if (cpu->singlestep_enabled != enabled) {
           cpu->singlestep_enabled = enabled;
           if (kvm_enabled()) {
+            kvm_set_singlestep(cpu, enabled);
               kvm_update_guest_debug(cpu, 0);
           } else {
               /* must flush all the translated code to avoid

inconsistencies */

include/sysemu/kvm.h +1 -0
index 97d8d9d0d5..a01a8d58dd 100644
@@ -259,6 +259,8 @@ int kvm_remove_breakpoint(CPUState *cpu,
 target_ulong addr,
   void kvm_remove_all_breakpoints(CPUState *cpu);
   int kvm_update_guest_debug(CPUState *cpu, unsigned long reinject_trap);
   +void kvm_set_singlestep(CPUState *cpu, int enabled);
+
   int kvm_on_sigbus_vcpu(CPUState *cpu, int code, void *addr);
   int kvm_on_sigbus(int code, void *addr);

  @@ -431,6 +433,8 @@ void kvm_arch_remove_all_hw_breakpoints(void); void kvm_arch_update_guest_debug(CPUState *cpu, struct kvm_guest_debug *dbg);
  +void kvm_arch_set_singlestep(CPUState *cpu, int enabled); +
  bool kvm_arch_stop_on_emulation_error(CPUState *cpu); int kvm_check_extension(KVMState *s, unsigned int extension);

target/arm/kvm.c +3 -0
index 44dd0ce6ce..dd8e43ab7e 100644
@@ -670,6 +670,10 @@ int kvm_arch_process_async_events(CPUState *cs)
       return 0;
   }
   +void kvm_arch_set_singlestep(CPUState *cs, int enabled)
+{
+}
+
   /* The #ifdef protections are until 32bit headers are imported and can
    * be removed once both 32 and 64 bit reach feature parity.
    */
target/i386/kvm.c +3 -0
index f524e7d929..ba56f2ee1f 100644
@@ -3521,6 +3521,10 @@ static int kvm_handle_debug(X86CPU *cpu,
       return ret;
   }
   +void kvm_arch_set_singlestep(CPUState *cs, int enabled)
+{
+}
+

  void kvm_arch_update_guest_debug(CPUState *cpu, struct kvm_guest_debug *dbg)
  {
      const uint8_t type_code[] = {

target/ppc/kvm.c +1 -0
index f81327d6cd..3af5e91997 100644
@@ -94,6 +94,7 @@ static int cap_ppc_safe_indirect_branch;
   static int cap_ppc_nested_kvm_hv;
     static uint32_t debug_inst_opcode;
+static target_ulong trace_handler_addr;
     /* XXX We have a race condition where we actually have a level

triggered
   *     interrupt, but the infrastructure can't expose that yet, so the guest

target/ppc/kvm.c +0 -0
@@ -509,6 +510,9 @@ int kvm_arch_init_vcpu(CPUState *cs)
       kvm_get_one_reg(cs, KVM_REG_PPC_DEBUG_INST, &debug_inst_opcode);
       kvmppc_hw_debug_points_init(cenv);
Quoted 2 lines

  +    trace_handler_addr = (cenv->excp_vectors[POWERPC_EXCP_TRACE] | +                          0xc000000000004000ull);

Can you add a definition for this magic value?

Quoted 64 lines

+
      return ret;
  }
  @@ -1551,6 +1555,28 @@ void kvm_arch_remove_all_hw_breakpoints(void) nb_hw_breakpoint = nb_hw_watchpoint = 0;
  }
  +void kvm_arch_set_singlestep(CPUState *cs, int enabled)

target/ppc/kvm.c +18 -0
+{
+    PowerPCCPU *cpu = POWERPC_CPU(cs);
+    CPUPPCState *env = &cpu->env;
+
+    if (kvmppc_is_pr(kvm_state)) {
+            return;
+    }
+
+    if (enabled) {
+        /* MSR_SE = 1 will cause a Trace Interrupt in the guest after
+         * the next instruction executes. */
+        env->msr |= (1ULL << MSR_SE);
+
+        /* We set a breakpoint at the interrupt handler address so
+         * that the singlestep will be seen by KVM (this is treated by
+         * KVM like an ordinary breakpoint) and control is returned to
+         * QEMU. */
+        kvm_insert_breakpoint(cs, trace_handler_addr, 4,

GDB_BREAKPOINT_SW);
+    }

target/ppc/kvm.c +2 -0
+}
+

  void kvm_arch_update_guest_debug(CPUState *cs, struct kvm_guest_debug *dbg)
  {
      int n;

target/ppc/kvm.c +24 -0
@@ -1590,6 +1616,43 @@ void kvm_arch_update_guest_debug(CPUState *cs,
 struct kvm_guest_debug *dbg)
       }
   }
   +static int kvm_handle_singlestep(CPUState *cs,
+                                 struct kvm_debug_exit_arch *arch_info)
+{
+    PowerPCCPU *cpu = POWERPC_CPU(cs);
+    CPUPPCState *env = &cpu->env;
+    target_ulong msr = env->msr;
+    uint32_t insn;
+    int ret = 1;
+    int reg;
+
+    if (kvmppc_is_pr(kvm_state)) {
+        return ret;
+    }
+
+    if (arch_info->address == trace_handler_addr) {
+        cpu_synchronize_state(cs);
+        kvm_remove_breakpoint(cs, trace_handler_addr, 4,
 GDB_BREAKPOINT_SW);
+
+        cpu_memory_rw_debug(cs, env->spr[SPR_SRR0] - 4, (uint8_t
 *)&insn,
+                            sizeof(insn), 0);
+
+        /* If the last instruction was a mfmsr, make sure that the
+         * MSR_SE bit is not set to avoid the guest kernel knowing
+         * that it is being single-stepped */
+        if (((insn >> 26) & 0x3f) == 31 && ((insn >> 1) & 0x3ff) ==
Quoted 1 line

83) {

You can use the extract32() function to extract bits, it is easier to review:

    if (extract32(insn, 26, 6) == 31 && extract32(insn, 1, 10) == 83) {

Quoted 1 line

+            reg = (insn >> 21) & 0x1f;

Ditto.

Regards,

Phil.

Quoted 39 lines

+            env->gpr[reg] &= ~(1ULL << MSR_SE);
+        }

target/ppc/kvm.c +9 -0
+
+        env->nip = env->spr[SPR_SRR0];
+        env->msr = msr &= ~(1ULL << MSR_SE);
+        cpu_synchronize_state(cs);
+    }
+
+    return ret;
+}
+

  static int kvm_handle_debug(PowerPCCPU *cpu, struct kvm_run *run) {
      CPUState *cs = CPU(cpu);

target/ppc/kvm.c +1 -1
@@ -1600,7 +1663,7 @@ static int kvm_handle_debug(PowerPCCPU *cpu,
 struct kvm_run *run)
       int flag = 0;
         if (cs->singlestep_enabled) {
-        handle = 1;
+        handle = kvm_handle_singlestep(cs, arch_info);
       } else if (arch_info->status) {
           if (nb_hw_breakpoint + nb_hw_watchpoint > 0) {
               if (arch_info->status & KVMPPC_DEBUG_BREAKPOINT) {
target/s390x/kvm.c +3 -0
index 2ebf26adfe..4bde183458 100644
@@ -975,6 +975,10 @@ void kvm_arch_remove_all_hw_breakpoints(void)
       hw_breakpoints = NULL;
   }
   +void kvm_arch_set_singlestep(CPUState *cs, int enabled)
+{
+}
+

  void kvm_arch_update_guest_debug(CPUState *cpu, struct kvm_guest_debug *dbg)
  {
      int i;

Signature · 1 line
Alexey
#5 lore raw

Re: [Qemu-devel] [RFC PATCH 1/1] target/ppc: support single stepping with KVM HV

Quoted 4 lines · Alexey Kardashevskiy

Fabiano,

Are you planning on reposting this any time soon? I am interested in the feature. Thanks.

Yes, I'm almost done with v3 of the series. Will probably send it later this week.