On Tue, Dec 3, 2024 at 4:51 PM bibo mao maobibo@loongson.cn wrote:

On 2024/12/3 下午2:50, Huacai Chen wrote:

...

When we enable lockdep we get such a warning:

============================= WARNING: suspicious RCU usage 6.12.0-rc7+ #1891 Tainted: G W

---

arch/loongarch/kvm/../../../virt/kvm/kvm_main.c:5945 suspicious rcu_dereference_check() usage! other info that might help us debug this: rcu_scheduler_active = 2, debug_locks = 1 1 lock held by qemu-system-loo/948: #0: 90000001184a00a8 (&vcpu->mutex){+.+.}-{4:4}, at: kvm_vcpu_ioctl+0xf4/0xe20 [kvm] stack backtrace: CPU: 2 UID: 0 PID: 948 Comm: qemu-system-loo Tainted: G W 6.12.0-rc7+ #1891 Tainted: [W]=WARN Hardware name: Loongson Loongson-3A5000-7A1000-1w-CRB/Loongson-LS3A5000-7A1000-1w-CRB, BIOS vUDK2018-LoongArch-V2.0.0-prebeta9 10/21/2022 Stack : 0000000000000089 9000000005a0db9c 90000000071519c8 900000012c578000 900000012c57b940 0000000000000000 900000012c57b948 9000000007e53788 900000000815bcc8 900000000815bcc0 900000012c57b7b0 0000000000000001 0000000000000001 4b031894b9d6b725 0000000005dec000 9000000100427b00 00000000000003d2 0000000000000001 000000000000002d 0000000000000003 0000000000000030 00000000000003b4 0000000005dec000 0000000000000000 900000000806d000 9000000007e53788 00000000000000b4 0000000000000004 0000000000000004 0000000000000000 0000000000000000 9000000107baf600 9000000008916000 9000000007e53788 9000000005924778 000000001fe001e5 00000000000000b0 0000000000000007 0000000000000000 0000000000071c1d ... Call Trace: [<9000000005924778>] show_stack+0x38/0x180 [<90000000071519c4>] dump_stack_lvl+0x94/0xe4 [<90000000059eb754>] lockdep_rcu_suspicious+0x194/0x240 [<ffff80000221f47c>] kvm_io_bus_read+0x19c/0x1e0 [kvm] [<ffff800002225118>] kvm_emu_mmio_read+0xd8/0x440 [kvm] [<ffff8000022254bc>] kvm_handle_read_fault+0x3c/0xe0 [kvm] [<ffff80000222b3c8>] kvm_handle_exit+0x228/0x480 [kvm]

Fix it by protecting kvm_io_bus_{read,write}() with SRCU.

Cc: stable@vger.kernel.org Signed-off-by: Huacai Chen chenhuacai@loongson.cn

---

arch/loongarch/kvm/exit.c | 31 +++++++++++++++++++++---------- arch/loongarch/kvm/intc/ipi.c | 6 +++++- 2 files changed, 26 insertions(+), 11 deletions(-)

diff --git a/arch/loongarch/kvm/exit.c b/arch/loongarch/kvm/exit.c index 69f3e3782cc9..a7893bd01e73 100644 --- a/arch/loongarch/kvm/exit.c +++ b/arch/loongarch/kvm/exit.c @@ -156,7 +156,7 @@ static int kvm_handle_csr(struct kvm_vcpu *vcpu, larch_inst inst)

int kvm_emu_iocsr(larch_inst inst, struct kvm_run *run, struct kvm_vcpu *vcpu) {

• int ret;
  

• int idx, ret;
  unsigned long *val;
  u32 addr, rd, rj, opcode;
  
  

@@ -167,7 +167,6 @@ int kvm_emu_iocsr(larch_inst inst, struct kvm_run *run, struct kvm_vcpu *vcpu) rj = inst.reg2_format.rj; opcode = inst.reg2_format.opcode; addr = vcpu->arch.gprs[rj];

• ret = EMULATE_DO_IOCSR;
  run->iocsr_io.phys_addr = addr;
  run->iocsr_io.is_write = 0;
  val = &vcpu->arch.gprs[rd];
  

@@ -207,20 +206,28 @@ int kvm_emu_iocsr(larch_inst inst, struct kvm_run *run, struct kvm_vcpu *vcpu) }

  if (run->iocsr_io.is_write) {

•         if (!kvm_io_bus_write(vcpu, KVM_IOCSR_BUS, addr, run->iocsr_io.len, val))
  

•         idx = srcu_read_lock(&vcpu->kvm->srcu);
  

•         ret = kvm_io_bus_write(vcpu, KVM_IOCSR_BUS, addr, run->iocsr_io.len, val);
  

•         srcu_read_unlock(&vcpu->kvm->srcu, idx);
  

•         if (ret == 0)
                  ret = EMULATE_DONE;
  

•         else
  

•         else {
  

•                 ret = EMULATE_DO_IOCSR;
                  /* Save data and let user space to write it */
                  memcpy(run->iocsr_io.data, val, run->iocsr_io.len);
  

•         }
          trace_kvm_iocsr(KVM_TRACE_IOCSR_WRITE, run->iocsr_io.len, addr, val);
  } else {
  

•         if (!kvm_io_bus_read(vcpu, KVM_IOCSR_BUS, addr, run->iocsr_io.len, val))
  

•         idx = srcu_read_lock(&vcpu->kvm->srcu);
  

•         ret = kvm_io_bus_read(vcpu, KVM_IOCSR_BUS, addr, run->iocsr_io.len, val);
  

•         srcu_read_unlock(&vcpu->kvm->srcu, idx);
  

•         if (ret == 0)
                  ret = EMULATE_DONE;
  

•         else
  

•         else {
  

•                 ret = EMULATE_DO_IOCSR;
                  /* Save register id for iocsr read completion */
                  vcpu->arch.io_gpr = rd;
  

•         }
          trace_kvm_iocsr(KVM_TRACE_IOCSR_READ, run->iocsr_io.len, addr, NULL);
  }
  
  

@@ -359,7 +366,7 @@ static int kvm_handle_gspr(struct kvm_vcpu *vcpu)

int kvm_emu_mmio_read(struct kvm_vcpu *vcpu, larch_inst inst) {

• int ret;
  

• int idx, ret;
  unsigned int op8, opcode, rd;
  struct kvm_run *run = vcpu->run;
  
  

@@ -464,8 +471,10 @@ int kvm_emu_mmio_read(struct kvm_vcpu *vcpu, larch_inst inst) * it need not return to user space to handle the mmio * exception. */

•         idx = srcu_read_lock(&vcpu->kvm->srcu);
          ret = kvm_io_bus_read(vcpu, KVM_MMIO_BUS, vcpu->arch.badv,
                                run->mmio.len, &vcpu->arch.gprs[rd]);
  

•         srcu_read_unlock(&vcpu->kvm->srcu, idx);
          if (!ret) {
                  update_pc(&vcpu->arch);
                  vcpu->mmio_needed = 0;
  

@@ -531,7 +540,7 @@ int kvm_complete_mmio_read(struct kvm_vcpu *vcpu, struct kvm_run *run)

int kvm_emu_mmio_write(struct kvm_vcpu *vcpu, larch_inst inst) {

• int ret;
  

• int idx, ret;
  unsigned int rd, op8, opcode;
  unsigned long curr_pc, rd_val = 0;
  struct kvm_run *run = vcpu->run;
  

@@ -631,7 +640,9 @@ int kvm_emu_mmio_write(struct kvm_vcpu *vcpu, larch_inst inst) * it need not return to user space to handle the mmio * exception. */

•         idx = srcu_read_lock(&vcpu->kvm->srcu);
          ret = kvm_io_bus_write(vcpu, KVM_MMIO_BUS, vcpu->arch.badv, run->mmio.len, data);
  

•         srcu_read_unlock(&vcpu->kvm->srcu, idx);
          if (!ret)
                  return EMULATE_DONE;
  
  

diff --git a/arch/loongarch/kvm/intc/ipi.c b/arch/loongarch/kvm/intc/ipi.c index a233a323e295..4b7ff20ed438 100644 --- a/arch/loongarch/kvm/intc/ipi.c +++ b/arch/loongarch/kvm/intc/ipi.c @@ -98,7 +98,7 @@ static void write_mailbox(struct kvm_vcpu *vcpu, int offset, uint64_t data, int

static int send_ipi_data(struct kvm_vcpu *vcpu, gpa_t addr, uint64_t data) {

• int i, ret;
  

• int i, idx, ret;
  uint32_t val = 0, mask = 0;
  
  /*
  

@@ -107,7 +107,9 @@ static int send_ipi_data(struct kvm_vcpu *vcpu, gpa_t addr, uint64_t data) */ if ((data >> 27) & 0xf) { /* Read the old val */

•         srcu_read_unlock(&vcpu->kvm->srcu, idx);
  

here should be idx = srcu_read_lock(&vcpu->kvm->srcu) ?

...

          ret = kvm_io_bus_read(vcpu, KVM_IOCSR_BUS, addr, sizeof(val), &val);

•         srcu_read_unlock(&vcpu->kvm->srcu, idx);
          if (unlikely(ret)) {
                  kvm_err("%s: : read date from addr %llx failed\n", __func__, addr);
                  return ret;
  

@@ -121,7 +123,9 @@ static int send_ipi_data(struct kvm_vcpu *vcpu, gpa_t addr, uint64_t data) val &= mask; } val |= ((uint32_t)(data >> 32) & ~mask);

• srcu_read_unlock(&vcpu->kvm->srcu, idx);
  

here should be idx = srcu_read_lock(&vcpu->kvm->srcu)

Yes, this is my mistake, thank you.

Huacai

...

  ret = kvm_io_bus_write(vcpu, KVM_IOCSR_BUS, addr, sizeof(val), &val);

• srcu_read_unlock(&vcpu->kvm->srcu, idx);
  if (unlikely(ret))
          kvm_err("%s: : write date to addr %llx failed\n", __func__, addr);
  
  
  

otherwise looks good to me.

Reviewed-by: Bibo Mao maobibo@loongson.cn

On Tue, Dec 3, 2024 at 4:27 PM bibo mao maobibo@loongson.cn wrote:

On 2024/12/3 下午2:50, Huacai Chen wrote:

...

When we enable lockdep we get such a warning:

============================= WARNING: suspicious RCU usage 6.12.0-rc7+ #1891 Tainted: G W

---

include/linux/kvm_host.h:1043 suspicious rcu_dereference_check() usage! other info that might help us debug this: rcu_scheduler_active = 2, debug_locks = 1 1 lock held by qemu-system-loo/948: #0: 90000001184a00a8 (&vcpu->mutex){+.+.}-{4:4}, at: kvm_vcpu_ioctl+0xf4/0xe20 [kvm] stack backtrace: CPU: 0 UID: 0 PID: 948 Comm: qemu-system-loo Tainted: G W 6.12.0-rc7+ #1891 Tainted: [W]=WARN Hardware name: Loongson Loongson-3A5000-7A1000-1w-CRB/Loongson-LS3A5000-7A1000-1w-CRB, BIOS vUDK2018-LoongArch-V2.0.0-prebeta9 10/21/2022 Stack : 0000000000000089 9000000005a0db9c 90000000071519c8 900000012c578000 900000012c57b920 0000000000000000 900000012c57b928 9000000007e53788 900000000815bcc8 900000000815bcc0 900000012c57b790 0000000000000001 0000000000000001 4b031894b9d6b725 0000000004dec000 90000001003299c0 0000000000000414 0000000000000001 000000000000002d 0000000000000003 0000000000000030 00000000000003b4 0000000004dec000 90000001184a0000 900000000806d000 9000000007e53788 00000000000000b4 0000000000000004 0000000000000004 0000000000000000 0000000000000000 9000000107baf600 9000000008916000 9000000007e53788 9000000005924778 0000000010000044 00000000000000b0 0000000000000004 0000000000000000 0000000000071c1d ... Call Trace: [<9000000005924778>] show_stack+0x38/0x180 [<90000000071519c4>] dump_stack_lvl+0x94/0xe4 [<90000000059eb754>] lockdep_rcu_suspicious+0x194/0x240 [<ffff8000022143bc>] kvm_gfn_to_hva_cache_init+0xfc/0x120 [kvm] [<ffff80000222ade4>] kvm_pre_enter_guest+0x3a4/0x520 [kvm] [<ffff80000222b3dc>] kvm_handle_exit+0x23c/0x480 [kvm]

Fix it by protecting kvm_check_requests() with SRCU.

Cc: stable@vger.kernel.org Signed-off-by: Huacai Chen chenhuacai@loongson.cn

---

arch/loongarch/kvm/vcpu.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/loongarch/kvm/vcpu.c b/arch/loongarch/kvm/vcpu.c index cab1818be68d..d18a4a270415 100644 --- a/arch/loongarch/kvm/vcpu.c +++ b/arch/loongarch/kvm/vcpu.c @@ -240,7 +240,7 @@ static void kvm_late_check_requests(struct kvm_vcpu *vcpu) */ static int kvm_enter_guest_check(struct kvm_vcpu *vcpu) {

• int ret;
  

• int idx, ret;
  
  /*
   * Check conditions before entering the guest
  

@@ -249,7 +249,9 @@ static int kvm_enter_guest_check(struct kvm_vcpu *vcpu) if (ret < 0) return ret;

• idx = srcu_read_lock(&vcpu->kvm->srcu);
  ret = kvm_check_requests(vcpu);
  

• srcu_read_unlock(&vcpu->kvm->srcu, idx);
  
  return ret;
  

  }

How about adding rcu readlock with closest function kvm_update_stolen_time()?

I have considered this method before. But then I read vcpu_run() of x86, it protect the whole vcpu_run() except the subroutine xfer_to_guest_mode_handle_work(), so I think protect the whole kvm_check_requests() is more like x86.

Huacai

static int kvm_check_requests(struct kvm_vcpu *vcpu) {

•   int idx;
  

•    if (!kvm_request_pending(vcpu))
             return RESUME_GUEST;
  
  

@@ -213,8 +215,11 @@ static int kvm_check_requests(struct kvm_vcpu *vcpu) if (kvm_dirty_ring_check_request(vcpu)) return RESUME_HOST;

•   if (kvm_check_request(KVM_REQ_STEAL_UPDATE, vcpu))
  

•   if (kvm_check_request(KVM_REQ_STEAL_UPDATE, vcpu)) {
  

•           idx = srcu_read_lock(&vcpu->kvm->srcu);
             kvm_update_stolen_time(vcpu);
  

•           srcu_read_unlock(&vcpu->kvm->srcu, idx);
  

•   }
  
     return RESUME_GUEST;
  

  }

Both method look good to me.

Regards Bibo Mao