Skip to contentSkip to conversation

Fix SIGILL psw.addr reporting

PATCH Superseded Ilya Leoshkevich 4 participants 10 messages May 21, 2021

v1 of 6 Superseded Latest is v6 →

lore
Share & tools
lore
#0 lore raw

cover Fix SIGILL psw.addr reporting

qemu-s390x puts a wrong value into SIGILL's siginfo_t's psw.addr: it should be a pointer to the instruction following the illegal instruction, but at the moment it is a pointer to the illegal instruction itself. This breaks OpenJDK, which relies on this value.

Patch 1 fixes the issue, patch 2 adds a test.

Ilya Leoshkevich (2):
target/s390x: Fix SIGILL psw.addr reporting
tests/tcg/s390x: Test SIGILL handling

 linux-user/s390x/cpu_loop.c     |  6 ++-
 target/s390x/excp_helper.c      | 69 ++++++++++++++++++---------------
 target/s390x/internal.h         |  1 +
 tests/tcg/s390x/Makefile.target |  1 +
 tests/tcg/s390x/sigill.c        | 41 ++++++++++++++++++++
 5 files changed, 85 insertions(+), 33 deletions(-)
 create mode 100644 tests/tcg/s390x/sigill.c
Signature · 1 line
2.31.1
#1 lore raw

PATCH 1/2 target/s390x: Fix SIGILL psw.addr reporting

When a s390x CPU attempts to execute an illegal instruction, an operation exception is recognized. This is a suppressing exception, which means that the PSW is advanced by the length of the illegal instruction.

On the real hardware or in qemu-system-s390x the kernel then raises SIGILL with si_addr pointing to the suppressed instruction and psw.addr containing the updated PSW.

Unfortunately qemu-s390x sets both to the address of the suppressed instruction at the moment. Fix by sharing the PSW advancement logic with qemu-system-s390x and setting si_addr to the address of the instruction that raised the exception.

---
 linux-user/s390x/cpu_loop.c |  6 +++-
 target/s390x/excp_helper.c  | 69 ++++++++++++++++++++-----------------
 target/s390x/internal.h     |  1 +
 3 files changed, 43 insertions(+), 33 deletions(-)
linux-user/s390x/cpu_loop.c +5 -1
index f2d1215fb1..6f5462d4f8 100644
@@ -21,6 +21,7 @@
 #include "qemu-common.h"
 #include "qemu.h"
 #include "cpu_loop-common.h"
+#include "internal.h"
 
 /* s390x masks the fault address it reports in si_addr for SIGSEGV and SIGBUS */
 #define S390X_FAIL_ADDR_MASK -4096LL
@@ -29,6 +30,7 @@ void cpu_loop(CPUS390XState *env)
 {
     CPUState *cs = env_cpu(env);
     int trapnr, n, sig;
+    target_ulong excp_psw_addr;
     target_siginfo_t info;
     target_ulong addr;
     abi_long ret;
@@ -38,6 +40,7 @@ void cpu_loop(CPUS390XState *env)
         trapnr = cpu_exec(cs);
         cpu_exec_end(cs);
         process_queued_cpu_work(cs);
+        excp_psw_addr = env->psw.addr;
 
         switch (trapnr) {
         case EXCP_INTERRUPT:
@@ -66,6 +69,7 @@ void cpu_loop(CPUS390XState *env)
             n = TARGET_TRAP_BRKPT;
             goto do_signal_pc;
         case EXCP_PGM:
+            s390_cpu_program_interrupt_advance_psw(env);
             n = env->int_pgm_code;
             switch (n) {
             case PGM_OPERATION:
@@ -131,7 +135,7 @@ void cpu_loop(CPUS390XState *env)
             break;
 
         do_signal_pc:
-            addr = env->psw.addr;
+            addr = excp_psw_addr;
         do_signal:
             info.si_signo = sig;
             info.si_errno = 0;
target/s390x/excp_helper.c +37 -32

81 lines not on this page · raw message

target/s390x/internal.h +1 -0
index 11515bb617..9f1665ccbf 100644
@@ -272,6 +272,7 @@ bool s390_cpu_tlb_fill(CPUState *cs, vaddr address, int size,
 void s390x_cpu_do_unaligned_access(CPUState *cs, vaddr addr,
                                    MMUAccessType access_type,
                                    int mmu_idx, uintptr_t retaddr);
+void s390_cpu_program_interrupt_advance_psw(CPUS390XState *cpu);
 
 
 /* fpu_helper.c */
Signature · 1 line
2.31.1
Reviewed-by on v1 1/2
#2 lore raw

Re: [PATCH 1/2] target/s390x: Fix SIGILL psw.addr reporting

Quoted 160 lines · Ilya Leoshkevich

When a s390x CPU attempts to execute an illegal instruction, an operation exception is recognized. This is a suppressing exception, which means that the PSW is advanced by the length of the illegal instruction.

On the real hardware or in qemu-system-s390x the kernel then raises SIGILL with si_addr pointing to the suppressed instruction and psw.addr containing the updated PSW.

Unfortunately qemu-s390x sets both to the address of the suppressed instruction at the moment. Fix by sharing the PSW advancement logic with qemu-system-s390x and setting si_addr to the address of the instruction that raised the exception.

---
  linux-user/s390x/cpu_loop.c |  6 +++-
  target/s390x/excp_helper.c  | 69 ++++++++++++++++++++-----------------
  target/s390x/internal.h     |  1 +
  3 files changed, 43 insertions(+), 33 deletions(-)
linux-user/s390x/cpu_loop.c +5 -1
index f2d1215fb1..6f5462d4f8 100644
@@ -21,6 +21,7 @@
  #include "qemu-common.h"
  #include "qemu.h"
  #include "cpu_loop-common.h"
+#include "internal.h"
 
  /* s390x masks the fault address it reports in si_addr for SIGSEGV and SIGBUS */
  #define S390X_FAIL_ADDR_MASK -4096LL
@@ -29,6 +30,7 @@ void cpu_loop(CPUS390XState *env)
  {
      CPUState *cs = env_cpu(env);
      int trapnr, n, sig;
+    target_ulong excp_psw_addr;
      target_siginfo_t info;
      target_ulong addr;
      abi_long ret;
@@ -38,6 +40,7 @@ void cpu_loop(CPUS390XState *env)
          trapnr = cpu_exec(cs);
          cpu_exec_end(cs);
          process_queued_cpu_work(cs);
+        excp_psw_addr = env->psw.addr;
 
          switch (trapnr) {
          case EXCP_INTERRUPT:
@@ -66,6 +69,7 @@ void cpu_loop(CPUS390XState *env)
              n = TARGET_TRAP_BRKPT;
              goto do_signal_pc;
          case EXCP_PGM:
+            s390_cpu_program_interrupt_advance_psw(env);
              n = env->int_pgm_code;
              switch (n) {
              case PGM_OPERATION:
@@ -131,7 +135,7 @@ void cpu_loop(CPUS390XState *env)
              break;
 
          do_signal_pc:
-            addr = env->psw.addr;
+            addr = excp_psw_addr;
          do_signal:
              info.si_signo = sig;
              info.si_errno = 0;
target/s390x/excp_helper.c +37 -32
index 20625c2c8f..0a323967ae 100644
@@ -82,6 +82,42 @@ void HELPER(data_exception)(CPUS390XState *env, uint32_t dxc)
      tcg_s390_data_exception(env, dxc, GETPC());
  }
 
+void s390_cpu_program_interrupt_advance_psw(CPUS390XState *env)
+{
+    switch (env->int_pgm_code) {
+    case PGM_PER:
+        if (env->per_perc_atmid & PER_CODE_EVENT_NULLIFICATION) {
+            break;
+        }
+        /* FALL THROUGH */
+    case PGM_OPERATION:
+    case PGM_PRIVILEGED:
+    case PGM_EXECUTE:
+    case PGM_PROTECTION:
+    case PGM_ADDRESSING:
+    case PGM_SPECIFICATION:
+    case PGM_DATA:
+    case PGM_FIXPT_OVERFLOW:
+    case PGM_FIXPT_DIVIDE:
+    case PGM_DEC_OVERFLOW:
+    case PGM_DEC_DIVIDE:
+    case PGM_HFP_EXP_OVERFLOW:
+    case PGM_HFP_EXP_UNDERFLOW:
+    case PGM_HFP_SIGNIFICANCE:
+    case PGM_HFP_DIVIDE:
+    case PGM_TRANS_SPEC:
+    case PGM_SPECIAL_OP:
+    case PGM_OPERAND:
+    case PGM_HFP_SQRT:
+    case PGM_PC_TRANS_SPEC:
+    case PGM_ALET_SPEC:
+    case PGM_MONITOR:
+        /* advance the PSW if our exception is not nullifying */
+        env->psw.addr += env->int_pgm_ilen;
+        break;
+    }
+}
+
  #if defined(CONFIG_USER_ONLY)
 
  void s390_cpu_do_interrupt(CPUState *cs)
@@ -202,38 +238,7 @@ static void do_program_interrupt(CPUS390XState *env)
 
      assert(ilen == 2 || ilen == 4 || ilen == 6);
 
-    switch (env->int_pgm_code) {
-    case PGM_PER:
-        if (env->per_perc_atmid & PER_CODE_EVENT_NULLIFICATION) {
-            break;
-        }
-        /* FALL THROUGH */
-    case PGM_OPERATION:
-    case PGM_PRIVILEGED:
-    case PGM_EXECUTE:
-    case PGM_PROTECTION:
-    case PGM_ADDRESSING:
-    case PGM_SPECIFICATION:
-    case PGM_DATA:
-    case PGM_FIXPT_OVERFLOW:
-    case PGM_FIXPT_DIVIDE:
-    case PGM_DEC_OVERFLOW:
-    case PGM_DEC_DIVIDE:
-    case PGM_HFP_EXP_OVERFLOW:
-    case PGM_HFP_EXP_UNDERFLOW:
-    case PGM_HFP_SIGNIFICANCE:
-    case PGM_HFP_DIVIDE:
-    case PGM_TRANS_SPEC:
-    case PGM_SPECIAL_OP:
-    case PGM_OPERAND:
-    case PGM_HFP_SQRT:
-    case PGM_PC_TRANS_SPEC:
-    case PGM_ALET_SPEC:
-    case PGM_MONITOR:
-        /* advance the PSW if our exception is not nullifying */
-        env->psw.addr += ilen;
-        break;
-    }
+    s390_cpu_program_interrupt_advance_psw(env);
 
      qemu_log_mask(CPU_LOG_INT,
                    "%s: code=0x%x ilen=%d psw: %" PRIx64 " %" PRIx64 "\n",
target/s390x/internal.h +0 -0
index 11515bb617..9f1665ccbf 100644

target/s390x/internal.h bool s390_cpu_tlb_fill(CPUState *cs, vaddr address, int size, · line 278

Quoted 7 lines
target/s390x/internal.h +1 -0
@@ -272,6 +272,7 @@ bool s390_cpu_tlb_fill(CPUState *cs, vaddr address, int size,
  void s390x_cpu_do_unaligned_access(CPUState *cs, vaddr addr,
                                     MMUAccessType access_type,
                                     int mmu_idx, uintptr_t retaddr);
+void s390_cpu_program_interrupt_advance_psw(CPUS390XState *cpu);
 
 
  /* fpu_helper.c */

LGTM, thanks

Signature · 3 lines
Thanks,

David / dhildenb
#3 lore raw

PATCH 2/2 tests/tcg/s390x: Test SIGILL handling

In reply to Ilya Leoshkevich

Verify that s390x-specific uc_mcontext.psw.addr is reported correctly.

---
 tests/tcg/s390x/Makefile.target |  1 +
 tests/tcg/s390x/sigill.c        | 41 +++++++++++++++++++++++++++++++++
 2 files changed, 42 insertions(+)
 create mode 100644 tests/tcg/s390x/sigill.c
tests/tcg/s390x/Makefile.target +1 -0
index 241ef28f61..8699d829a5 100644
@@ -8,3 +8,4 @@ TESTS+=exrl-trtr
 TESTS+=pack
 TESTS+=mvo
 TESTS+=mvc
+TESTS+=sigill
tests/tcg/s390x/sigill.c new file +41 -0

41 lines not on this page · raw message

Signature · 1 line
2.31.1
#4 lore raw

Re: [PATCH 2/2] tests/tcg/s390x: Test SIGILL handling

Quoted 24 lines · Ilya Leoshkevich

Verify that s390x-specific uc_mcontext.psw.addr is reported correctly.

---
  tests/tcg/s390x/Makefile.target |  1 +
  tests/tcg/s390x/sigill.c        | 41 +++++++++++++++++++++++++++++++++
  2 files changed, 42 insertions(+)

create mode 100644 tests/tcg/s390x/sigill.c

tests/tcg/s390x/Makefile.target +1 -0
index 241ef28f61..8699d829a5 100644
@@ -8,3 +8,4 @@ TESTS+=exrl-trtr
  TESTS+=pack
  TESTS+=mvo
  TESTS+=mvc
+TESTS+=sigill
tests/tcg/s390x/sigill.c new file +0 -0
new file mode 100644
index 0000000000..f8021dc6af

tests/tcg/s390x/sigill.c line 8

Quoted 8 lines
tests/tcg/s390x/sigill.c new file +8 -0
@@ -0,0 +1,41 @@
+#include <assert.h>
+#include <signal.h>
+#include <string.h>
+#include <ucontext.h>
+#include <unistd.h>
+
+extern char expected_si_addr[];
+extern char expected_psw_addr[];

Why "extern" ? For the magic inline asm below to work?

Quoted 16 lines
tests/tcg/s390x/sigill.c new file +16 -0
+
+static void handle_signal(int sig, siginfo_t *info, void *ucontext)
+{
+    if (sig != SIGILL) {
+        _exit(1);
+    }
+
+    if (info->si_addr != expected_si_addr) {
+        _exit(2);
+    }
+
+    if (((ucontext_t *)ucontext)->uc_mcontext.psw.addr !=
+            (unsigned long)expected_psw_addr) {
+        _exit(3);
+    }
+}

tests/tcg/s390x/sigill.c line 34

Quoted 10 lines
tests/tcg/s390x/sigill.c new file +10 -0
+
+int main(void)
+{
+    struct sigaction act;
+
+    memset(&act, 0, sizeof(act));
+    act.sa_sigaction = handle_signal;
+    act.sa_flags = SA_SIGINFO;
+
+    int ret = sigaction(SIGILL, &act, NULL);

Mixing code and declaration.

tests/tcg/s390x/sigill.c line 38

Quoted 4 lines
tests/tcg/s390x/sigill.c new file +4 -0
+    assert(ret == 0);
+
+    asm volatile("expected_si_addr:\t.byte\t0x00,0x00\n"
+                 "expected_psw_addr:");

At least I am confused how the right values actually end up in expected_si_addr and expected_psw_addr.

Can we maybe add a comment? This looks quite hacky ;)

Signature · 3 lines
Thanks,

David / dhildenb
#5 lore raw

Re: [PATCH 2/2] tests/tcg/s390x: Test SIGILL handling

Quoted 30 lines · David Hildenbrand
Quoted 29 lines · Ilya Leoshkevich

Verify that s390x-specific uc_mcontext.psw.addr is reported correctly.

---
  tests/tcg/s390x/Makefile.target |  1 +
  tests/tcg/s390x/sigill.c        | 41
+++++++++++++++++++++++++++++++++

  2 files changed, 42 insertions(+)

  create mode 100644 tests/tcg/s390x/sigill.c

tests/tcg/s390x/Makefile.target +0 -0
index 241ef28f61..8699d829a5 100644
@@ -8,3 +8,4 @@ TESTS+=exrl-trtr

  TESTS+=pack
  TESTS+=mvo
  TESTS+=mvc

tests/tcg/s390x/Makefile.target +1 -0
+TESTS+=sigill
tests/tcg/s390x/sigill.c new file +2 -0
new file mode 100644
index 0000000000..f8021dc6af
@@ -0,0 +1,41 @@
+#include <assert.h>
+#include <signal.h>
Quoted 8 lines

tests/tcg/s390x/sigill.c line 8

Quoted 6 lines
tests/tcg/s390x/sigill.c new file +6 -0
+#include <string.h>
+#include <ucontext.h>
+#include <unistd.h>
+
+extern char expected_si_addr[];
+extern char expected_psw_addr[];

Why "extern" ? For the magic inline asm below to work?

Yes - it cannot be static, because AFAIK there is no such thing as static variable declaration (one can only define static variables).

Quoted 21 lines
Quoted 21 lines
tests/tcg/s390x/sigill.c new file +2 -0
+
+static void handle_signal(int sig, siginfo_t *info, void

*ucontext)

tests/tcg/s390x/sigill.c new file +18 -0
+{
+    if (sig != SIGILL) {
+        _exit(1);
+    }
+
+    if (info->si_addr != expected_si_addr) {
+        _exit(2);
+    }
+
+    if (((ucontext_t *)ucontext)->uc_mcontext.psw.addr !=
+            (unsigned long)expected_psw_addr) {
+        _exit(3);
+    }
+}
+
+int main(void)
+{
+    struct sigaction act;
Quoted 8 lines

tests/tcg/s390x/sigill.c line 34

Quoted 6 lines
tests/tcg/s390x/sigill.c new file +6 -0
+
+    memset(&act, 0, sizeof(act));
+    act.sa_sigaction = handle_signal;
+    act.sa_flags = SA_SIGINFO;
+
+    int ret = sigaction(SIGILL, &act, NULL);

Mixing code and declaration.

Ouch, will fix.

Quoted 9 lines
Quoted 4 lines

+    assert(ret == 0);
+
+    asm volatile("expected_si_addr:\t.byte\t0x00,0x00\n" +                 "expected_psw_addr:");

At least I am confused how the right values actually end up in expected_si_addr and expected_psw_addr.

Can we maybe add a comment? This looks quite hacky ;)

This whole construction is roughly the same as having sigill.s file with:

.globl expected_si_addr
expected_si_addr: .byte 0,0
.globl expected_psw_addr
expected_psw_addr: br 14

and sigill.h file with:

void expected_si_addr(void);
extern char expected_psw_addr[];

Doing it this way would complicate the build, so I thought it would be better to just put everything into a single file.

#6 lore raw
In reply to Ilya Leoshkevich
Patchew URL: https://patchew.org/QEMU/[email protected]/

Patchew URL: https://patchew.org/QEMU/[email protected]/

Hi,

This series seems to have some coding style problems. See output below for more information:

Type: series
Message-id: [email protected]
Subject: [PATCH 0/2] Fix SIGILL psw.addr reporting

=== TEST SCRIPT BEGIN ===
#!/bin/bash
git rev-parse base > /dev/null || exit 0
git config --local diff.renamelimit 0
git config --local diff.renames True
git config --local diff.algorithm histogram
./scripts/checkpatch.pl --mailback base..
=== TEST SCRIPT END ===

Updating 3c8cf5a9c21ff8782164d1def7f44bd888713384 From https://github.com/patchew-project/qemu
* [new tag] patchew/[email protected] -> patchew/[email protected]
Switched to a new branch 'test'
069d02b tests/tcg/s390x: Test SIGILL handling a869643 target/s390x: Fix SIGILL psw.addr reporting

=== OUTPUT BEGIN ===
1/2 Checking commit a86964364d4d (target/s390x: Fix SIGILL psw.addr reporting) 2/2 Checking commit 069d02b0edf2 (tests/tcg/s390x: Test SIGILL handling)
WARNING: added, moved or deleted file(s), does MAINTAINERS need updating? #26:
new file mode 100644

ERROR: externs should be avoided in .c files
#37: FILE: tests/tcg/s390x/sigill.c:7:
+extern char expected_si_addr[];

ERROR: externs should be avoided in .c files
#38: FILE: tests/tcg/s390x/sigill.c:8:
+extern char expected_psw_addr[];

total: 2 errors, 1 warnings, 45 lines checked

Patch 2/2 has style problems, please review. If any of these errors are false positives report them to the maintainer, see CHECKPATCH in MAINTAINERS.

=== OUTPUT END ===

Test command exited with code: 1

The full log is available at
http://patchew.org/logs/[email protected]/testing.checkpatch/?type=message.
---
Email generated automatically by Patchew [https://patchew.org/].
Please send your feedback to [email protected]

#7 lore raw
In reply to Ilya Leoshkevich
Quoted 6 lines · Ilya Leoshkevich

qemu-s390x puts a wrong value into SIGILL's siginfo_t's psw.addr: it should be a pointer to the instruction following the illegal instruction, but at the moment it is a pointer to the illegal instruction itself. This breaks OpenJDK, which relies on this value.

Patch 1 fixes the issue, patch 2 adds a test.

I assume that should fix

https://bugs.launchpad.net/qemu/+bug/1920913

right?

Signature · 3 lines
Thanks,

David / dhildenb
#8 lore raw
Quoted 8 lines · David Hildenbrand
Quoted 7 lines · Ilya Leoshkevich

qemu-s390x puts a wrong value into SIGILL's siginfo_t's psw.addr:
it
should be a pointer to the instruction following the illegal instruction, but at the moment it is a pointer to the illegal instruction itself. This breaks OpenJDK, which relies on this value.

Quoted 7 lines
Quoted 1 line

Patch 1 fixes the issue, patch 2 adds a test.

I assume that should fix

https://bugs.launchpad.net/qemu/+bug/1920913

right?

Yes, I have this Buglink in the fix commit message.
I have to admit I did not test Java more extensively - there might be more things going on - but at least with this the SIGILL on startup is gone, and very simple programs work.

#9 lore raw
Quoted 11 lines · On Fri, 21 May 2021 12:45:30 +0200 Ilya Leoshkevich <[email protected]> wrote:
Quoted 10 lines · David Hildenbrand
Quoted 8 lines · Ilya Leoshkevich

qemu-s390x puts a wrong value into SIGILL's siginfo_t's psw.addr:
it
should be a pointer to the instruction following the illegal instruction, but at the moment it is a pointer to the illegal instruction itself. This breaks OpenJDK, which relies on this value.

Patch 1 fixes the issue, patch 2 adds a test.

Quoted 7 lines
Quoted 5 lines

I assume that should fix

https://bugs.launchpad.net/qemu/+bug/1920913

right?

Yes, I have this Buglink in the fix commit message.

Better add https://gitlab.com/qemu-project/qemu/-/issues/319 as the Buglink :)

Quoted 5 lines

I have to admit I did not test Java more extensively - there might be more things going on - but at least with this the SIGILL on startup is gone, and very simple programs work.