Skip to contentSkip to conversation

target/i386/kvm: Free xsave_buf when destroying vCPU

RFC Applied Philippe Mathieu-Daudé 4 participants 6 messages Mar 22, 2022 KVM

lore
Share & tools
lore
#0 lore raw

RFC v4 1/1 target/i386/kvm: Free xsave_buf when destroying vCPU

From: Philippe Mathieu-Daudé <[email protected]>

Fix vCPU hot-unplug related leak reported by Valgrind:

==132362== 4,096 bytes in 1 blocks are definitely lost in loss record 8,440 of 8,549

  ==132362==    at 0x4C3B15F: memalign (vg_replace_malloc.c:1265)
  ==132362==    by 0x4C3B288: posix_memalign (vg_replace_malloc.c:1429)
  ==132362==    by 0xB41195: qemu_try_memalign (memalign.c:53)
  ==132362==    by 0xB41204: qemu_memalign (memalign.c:73)
  ==132362==    by 0x7131CB: kvm_init_xsave (kvm.c:1601)
  ==132362==    by 0x7148ED: kvm_arch_init_vcpu (kvm.c:2031)
  ==132362==    by 0x91D224: kvm_init_vcpu (kvm-all.c:516)
  ==132362==    by 0x9242C9: kvm_vcpu_thread_fn (kvm-accel-ops.c:40)
  ==132362==    by 0xB2EB26: qemu_thread_start (qemu-thread-posix.c:556)
  ==132362==    by 0x7EB2159: start_thread (in /usr/lib64/libpthread-2.28.so)
  ==132362==    by 0x9D45DD2: clone (in /usr/lib64/libc-2.28.so)

---
Based on a series from Mark:
https://lore.kernel.org/qemu-devel/[email protected]/

RFC because currently no time to test

---
 target/i386/kvm/kvm.c | 2 ++
 1 file changed, 2 insertions(+)
target/i386/kvm/kvm.c +2 -0
index ef2c68a6f4..e93440e774 100644
@@ -2072,6 +2072,8 @@ int kvm_arch_destroy_vcpu(CPUState *cs)
     X86CPU *cpu = X86_CPU(cs);
     CPUX86State *env = &cpu->env;
 
+    g_free(env->xsave_buf);
+
     if (cpu->kvm_msr_buf) {
         g_free(cpu->kvm_msr_buf);
         cpu->kvm_msr_buf = NULL;
Signature · 1 line
2.35.1
#1 lore raw
Quoted 17 lines · Philippe Mathieu-Daudé

From: Philippe Mathieu-Daudé <[email protected]>

Fix vCPU hot-unplug related leak reported by Valgrind:

   ==132362== 4,096 bytes in 1 blocks are definitely lost in loss record 8,440 of 8,549
   ==132362==    at 0x4C3B15F: memalign (vg_replace_malloc.c:1265)
   ==132362==    by 0x4C3B288: posix_memalign (vg_replace_malloc.c:1429)
   ==132362==    by 0xB41195: qemu_try_memalign (memalign.c:53)
   ==132362==    by 0xB41204: qemu_memalign (memalign.c:73)
   ==132362==    by 0x7131CB: kvm_init_xsave (kvm.c:1601)
   ==132362==    by 0x7148ED: kvm_arch_init_vcpu (kvm.c:2031)
   ==132362==    by 0x91D224: kvm_init_vcpu (kvm-all.c:516)
   ==132362==    by 0x9242C9: kvm_vcpu_thread_fn (kvm-accel-ops.c:40)
   ==132362==    by 0xB2EB26: qemu_thread_start (qemu-thread-posix.c:556)
   ==132362==    by 0x7EB2159: start_thread (in /usr/lib64/libpthread-2.28.so)
   ==132362==    by 0x9D45DD2: clone (in /usr/lib64/libc-2.28.so)
Quoted 7 lines

---
Based on a series from Mark:
https://lore.kernel.org/qemu-devel/[email protected]/

RFC because currently no time to test

Mark, do you mind testing this patch?

Tested-by
#2 lore raw
Quoted 25 lines · Philippe Mathieu-Daudé
Quoted 24 lines · Philippe Mathieu-Daudé

From: Philippe Mathieu-Daudé <[email protected]>

Fix vCPU hot-unplug related leak reported by Valgrind:

   ==132362== 4,096 bytes in 1 blocks are definitely lost in loss record 8,440 of 8,549 ==132362==    at 0x4C3B15F: memalign (vg_replace_malloc.c:1265) ==132362==    by 0x4C3B288: posix_memalign (vg_replace_malloc.c:1429) ==132362==    by 0xB41195: qemu_try_memalign (memalign.c:53) ==132362==    by 0xB41204: qemu_memalign (memalign.c:73)
   ==132362==    by 0x7131CB: kvm_init_xsave (kvm.c:1601)
   ==132362==    by 0x7148ED: kvm_arch_init_vcpu (kvm.c:2031)
   ==132362==    by 0x91D224: kvm_init_vcpu (kvm-all.c:516)
   ==132362==    by 0x9242C9: kvm_vcpu_thread_fn (kvm-accel-ops.c:40) ==132362==    by 0xB2EB26: qemu_thread_start (qemu-thread-posix.c:556) ==132362==    by 0x7EB2159: start_thread (in /usr/lib64/libpthread-2.28.so) ==132362==    by 0x9D45DD2: clone (in /usr/lib64/libc-2.28.so)

---
Based on a series from Mark:
https://lore.kernel.org/qemu-devel/[email protected]/

Quoted 3 lines
Quoted 1 line

RFC because currently no time to test

Mark, do you mind testing this patch?

Sanity tested with x86_64 KVM. Valgrind confirms the leak is fixed upon the vCPU hotplug.

Thanks/regards,
-Mark

#3 lore raw
In reply to Philippe Mathieu-Daudé
Quoted 33 lines · On Tue, 22 Mar 2022 13:05:22 +0100 Philippe Mathieu-Daudé…

From: Philippe Mathieu-Daudé <[email protected]>

Fix vCPU hot-unplug related leak reported by Valgrind:

==132362== 4,096 bytes in 1 blocks are definitely lost in loss record 8,440 of 8,549

  ==132362==    at 0x4C3B15F: memalign (vg_replace_malloc.c:1265)
  ==132362==    by 0x4C3B288: posix_memalign (vg_replace_malloc.c:1429)
  ==132362==    by 0xB41195: qemu_try_memalign (memalign.c:53)
  ==132362==    by 0xB41204: qemu_memalign (memalign.c:73)
  ==132362==    by 0x7131CB: kvm_init_xsave (kvm.c:1601)
  ==132362==    by 0x7148ED: kvm_arch_init_vcpu (kvm.c:2031)
  ==132362==    by 0x91D224: kvm_init_vcpu (kvm-all.c:516)
  ==132362==    by 0x9242C9: kvm_vcpu_thread_fn (kvm-accel-ops.c:40)
  ==132362==    by 0xB2EB26: qemu_thread_start (qemu-thread-posix.c:556)
  ==132362==    by 0x7EB2159: start_thread (in /usr/lib64/libpthread-2.28.so)
  ==132362==    by 0x9D45DD2: clone (in /usr/lib64/libc-2.28.so)

---
Based on a series from Mark:
https://lore.kernel.org/qemu-devel/[email protected]/

RFC because currently no time to test

---
 target/i386/kvm/kvm.c | 2 ++
 1 file changed, 2 insertions(+)
target/i386/kvm/kvm.c +0 -0
index ef2c68a6f4..e93440e774 100644

target/i386/kvm/kvm.c int kvm_arch_destroy_vcpu(CPUState *cs) · line 2079

Quoted 8 lines
target/i386/kvm/kvm.c +2 -0
@@ -2072,6 +2072,8 @@ int kvm_arch_destroy_vcpu(CPUState *cs)
     X86CPU *cpu = X86_CPU(cs);
     CPUX86State *env = &cpu->env;
 
+    g_free(env->xsave_buf);
+
     if (cpu->kvm_msr_buf) {
         g_free(cpu->kvm_msr_buf);
         cpu->kvm_msr_buf = NULL;

shouldn't we do the same in hvf_arch_vcpu_destroy() ?

#4 lore raw
Quoted 39 lines · Igor Mammedov
Quoted 36 lines · On Tue, 22 Mar 2022 13:05:22 +0100 Philippe Mathieu-Daudé…

From: Philippe Mathieu-Daudé <[email protected]>

Fix vCPU hot-unplug related leak reported by Valgrind:

   ==132362== 4,096 bytes in 1 blocks are definitely lost in loss record 8,440 of 8,549
   ==132362==    at 0x4C3B15F: memalign (vg_replace_malloc.c:1265)
   ==132362==    by 0x4C3B288: posix_memalign (vg_replace_malloc.c:1429)
   ==132362==    by 0xB41195: qemu_try_memalign (memalign.c:53)
   ==132362==    by 0xB41204: qemu_memalign (memalign.c:73)
   ==132362==    by 0x7131CB: kvm_init_xsave (kvm.c:1601)
   ==132362==    by 0x7148ED: kvm_arch_init_vcpu (kvm.c:2031)
   ==132362==    by 0x91D224: kvm_init_vcpu (kvm-all.c:516)
   ==132362==    by 0x9242C9: kvm_vcpu_thread_fn (kvm-accel-ops.c:40)
   ==132362==    by 0xB2EB26: qemu_thread_start (qemu-thread-posix.c:556)
   ==132362==    by 0x7EB2159: start_thread (in /usr/lib64/libpthread-2.28.so)
   ==132362==    by 0x9D45DD2: clone (in /usr/lib64/libc-2.28.so)

---
Based on a series from Mark:
https://lore.kernel.org/qemu-devel/[email protected]/

RFC because currently no time to test

---
  target/i386/kvm/kvm.c | 2 ++
  1 file changed, 2 insertions(+)
target/i386/kvm/kvm.c +0 -0
index ef2c68a6f4..e93440e774 100644
@@ -2072,6 +2072,8 @@ int kvm_arch_destroy_vcpu(CPUState *cs)
      X86CPU *cpu = X86_CPU(cs);
      CPUX86State *env = &cpu->env;
 
Quoted 8 lines

target/i386/kvm/kvm.c int kvm_arch_destroy_vcpu(CPUState *cs) · line 2079

Quoted 5 lines
target/i386/kvm/kvm.c +2 -0
+    g_free(env->xsave_buf);
+
      if (cpu->kvm_msr_buf) {
          g_free(cpu->kvm_msr_buf);
          cpu->kvm_msr_buf = NULL;

shouldn't we do the same in hvf_arch_vcpu_destroy() ?

Yeah HVF needs a similar patch (at least hvf_caps needs to be released too, but I had no time to review it carefully yet).