Merge tag 'linux-user-for-7.2-pull-request' of https://gitlab.com/laurent_vivier/qemu into staging

linux-user pull request 20220928-v2

use 'max' instead of 'qemu32' / 'qemu64'
add  pidfd_open(), pidfd_send_signal() and pidfd_getfd()
Improve madvise(MADV_DONTNEED)
futex syscal rework
strace improvement
HP/PA fixes and improvement
Misc fixes

# -----BEGIN PGP SIGNATURE-----
#
# iQJGBAABCAAwFiEEzS913cjjpNwuT1Fz8ww4vT8vvjwFAmM0riISHGxhdXJlbnRA
# dml2aWVyLmV1AAoJEPMMOL0/L748gH4P/2wesXJKPMY2zQzP3Rld4iyefoPGG/Yp
# mdq59BbjO2jQMR8GBss/nl9l84cIzzkYRQIogaKsjljtZYm/OO5xRefqrzJY6apD
# eidxv20dAVjuaXHAIdGhbFlxot1ctExbZs9atB4uj5DWxfYGD6e/stoBy5/pSmr4
# M5EbGHhyrRI7tRbHGtVQVvG6AT6XGE0pT9tzT5JLaApF8UPMkgJwmez16PNWvcMm
# v8GEvKm/vEVS8CCpzLV4kfwVeo3f54VAOrEBDi29ph2Yo50IA21k8BvoRZaSp+Kn
# G6TMnnly/DkMspAs5EOVfat+kv3TziNNdDH7EnVU1vV1yTDdZgW/1204Uy/JY0Pw
# WotwAFuO9FYeHKmjY0CfnIIZZHYZpDYUOZ8M6dESD/O0EjoB8LMf5p9cbYlze4DE
# csJZCsVcz19HDv6QZXi5mvvDcJ83B2IDb8/PUAzSc0n62lXL9qjYD0wdb0QsLdAT
# I25qLDge1HCmQfCIKcaoHYvE0pDmvkF6ftuQUXLtIwtaV0Z/N5wDf2PEHikjOYHM
# gD2izz23/2wQx6KP/9ZNnCJ5QEBkEgm5wpHncsvjzSzi1uIdNlHyzJJwGTAcc5qZ
# hOeoJ7dT0D6g0BGnvOdg2W/bDx18KW65mNDxE4d+W0uzn0YmQtArk2YsnhKQNO46
# 12/0ltPFnSV/
# =DIzQ
# -----END PGP SIGNATURE-----
# gpg: Signature made Wed 28 Sep 2022 16:27:14 EDT
# gpg:                using RSA key CD2F75DDC8E3A4DC2E4F5173F30C38BD3F2FBE3C
# gpg:                issuer "laurent@vivier.eu"
# gpg: Good signature from "Laurent Vivier <lvivier@redhat.com>" [full]
# gpg:                 aka "Laurent Vivier <laurent@vivier.eu>" [full]
# gpg:                 aka "Laurent Vivier (Red Hat) <lvivier@redhat.com>" [full]
# Primary key fingerprint: CD2F 75DD C8E3 A4DC 2E4F  5173 F30C 38BD 3F2F BE3C

* tag 'linux-user-for-7.2-pull-request' of https://gitlab.com/laurent_vivier/qemu: (37 commits)
  linux-user: Add parameters of getrandom() syscall for strace
  linux-user: Lock log around strace
  linux-user: Update print_futex_op
  linux-user: Implement PI futexes
  linux-user: Convert signal number for FUTEX_FD
  linux-user: Implement FUTEX_WAKE_BITSET
  linux-user: Sink call to do_safe_futex
  linux-user: Combine do_futex and do_futex_time64
  linux-user: Set ELF_BASE_PLATFORM for MIPS
  linux-user: Introduce stubs for ELF AT_BASE_PLATFORM
  linux-user/s390x: Save/restore fpc when handling a signal
  linux-user: Don't assume 0 is not a valid host timer_t value
  linux-user: fix bug about missing signum convert of sigqueue
  linux-user/hppa: Fix setup_sigcontext()
  linux-user/hppa: Allow PROT_GROWSUP and PROT_GROWSDOWN in mprotect()
  linux-user/hppa: Increase guest stack size to 80MB for hppa target
  linux-user/hppa: Drop stack guard page on hppa target
  linux-user/hppa: Add signal trampoline for hppa target
  linux-user: Add proper strace format strings for getdents()/getdents64()
  linux-user: Fix TARGET_PROT_SEM for XTENSA
  ...

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

1 files changed, 116 insertions, 90 deletions

diff --git a/linux-user/syscall.c b/linux-user/syscall.c
index f409121202..2e954d8dbd 100644
--- a/linux-user/syscall.c
+++ b/linux-user/syscall.c
@@ -346,6 +346,16 @@ _syscall6(int,sys_futex,int *,uaddr,int,op,int,val,
 _syscall6(int,sys_futex_time64,int *,uaddr,int,op,int,val,
           const struct timespec *,timeout,int *,uaddr2,int,val3)
 #endif
+#if defined(__NR_pidfd_open) && defined(TARGET_NR_pidfd_open)
+_syscall2(int, pidfd_open, pid_t, pid, unsigned int, flags);
+#endif
+#if defined(__NR_pidfd_send_signal) && defined(TARGET_NR_pidfd_send_signal)
+_syscall4(int, pidfd_send_signal, int, pidfd, int, sig, siginfo_t *, info,
+                             unsigned int, flags);
+#endif
+#if defined(__NR_pidfd_getfd) && defined(TARGET_NR_pidfd_getfd)
+_syscall3(int, pidfd_getfd, int, pidfd, int, targetfd, unsigned int, flags);
+#endif
 #define __NR_sys_sched_getaffinity __NR_sched_getaffinity
 _syscall3(int, sys_sched_getaffinity, pid_t, pid, unsigned int, len,
           unsigned long *, user_mask_ptr);
@@ -515,20 +525,25 @@ _syscall4(int, sys_prlimit64, pid_t, pid, int, resource,
 
 #if defined(TARGET_NR_timer_create)
 /* Maximum of 32 active POSIX timers allowed at any one time. */
-static timer_t g_posix_timers[32] = { 0, } ;
+#define GUEST_TIMER_MAX 32
+static timer_t g_posix_timers[GUEST_TIMER_MAX];
+static int g_posix_timer_allocated[GUEST_TIMER_MAX];
 
 static inline int next_free_host_timer(void)
 {
-    int k ;
-    /* FIXME: Does finding the next free slot require a lock? */
-    for (k = 0; k < ARRAY_SIZE(g_posix_timers); k++) {
-        if (g_posix_timers[k] == 0) {
-            g_posix_timers[k] = (timer_t) 1;
+    int k;
+    for (k = 0; k < ARRAY_SIZE(g_posix_timer_allocated); k++) {
+        if (qatomic_xchg(g_posix_timer_allocated + k, 1) == 0) {
             return k;
         }
     }
     return -1;
 }
+
+static inline void free_host_timer_slot(int id)
+{
+    qatomic_store_release(g_posix_timer_allocated + id, 0);
+}
 #endif
 
 static inline int host_to_target_errno(int host_errno)
@@ -7753,15 +7768,16 @@ static int do_safe_futex(int *uaddr, int op, int val,
    futexes locally would make futexes shared between multiple processes
    tricky.  However they're probably useless because guest atomic
    operations won't work either.  */
-#if defined(TARGET_NR_futex)
-static int do_futex(CPUState *cpu, target_ulong uaddr, int op, int val,
-                    target_ulong timeout, target_ulong uaddr2, int val3)
+#if defined(TARGET_NR_futex) || defined(TARGET_NR_futex_time64)
+static int do_futex(CPUState *cpu, bool time64, target_ulong uaddr,
+                    int op, int val, target_ulong timeout,
+                    target_ulong uaddr2, int val3)
 {
-    struct timespec ts, *pts;
+    struct timespec ts, *pts = NULL;
+    void *haddr2 = NULL;
     int base_op;
 
-    /* ??? We assume FUTEX_* constants are the same on both host
-       and target.  */
+    /* We assume FUTEX_* constants are the same on both host and target. */
 #ifdef FUTEX_CMD_MASK
     base_op = op & FUTEX_CMD_MASK;
 #else
@@ -7770,85 +7786,53 @@ static int do_futex(CPUState *cpu, target_ulong uaddr, int op, int val,
     switch (base_op) {
     case FUTEX_WAIT:
     case FUTEX_WAIT_BITSET:
-        if (timeout) {
-            pts = &ts;
-            target_to_host_timespec(pts, timeout);
-        } else {
-            pts = NULL;
-        }
-        return do_safe_futex(g2h(cpu, uaddr),
-                             op, tswap32(val), pts, NULL, val3);
+        val = tswap32(val);
+        break;
+    case FUTEX_WAIT_REQUEUE_PI:
+        val = tswap32(val);
+        haddr2 = g2h(cpu, uaddr2);
+        break;
+    case FUTEX_LOCK_PI:
+    case FUTEX_LOCK_PI2:
+        break;
     case FUTEX_WAKE:
-        return do_safe_futex(g2h(cpu, uaddr),
-                             op, val, NULL, NULL, 0);
+    case FUTEX_WAKE_BITSET:
+    case FUTEX_TRYLOCK_PI:
+    case FUTEX_UNLOCK_PI:
+        timeout = 0;
+        break;
     case FUTEX_FD:
-        return do_safe_futex(g2h(cpu, uaddr),
-                             op, val, NULL, NULL, 0);
-    case FUTEX_REQUEUE:
+        val = target_to_host_signal(val);
+        timeout = 0;
+        break;
     case FUTEX_CMP_REQUEUE:
+    case FUTEX_CMP_REQUEUE_PI:
+        val3 = tswap32(val3);
+        /* fall through */
+    case FUTEX_REQUEUE:
     case FUTEX_WAKE_OP:
-        /* For FUTEX_REQUEUE, FUTEX_CMP_REQUEUE, and FUTEX_WAKE_OP, the
-           TIMEOUT parameter is interpreted as a uint32_t by the kernel.
-           But the prototype takes a `struct timespec *'; insert casts
-           to satisfy the compiler.  We do not need to tswap TIMEOUT
-           since it's not compared to guest memory.  */
-        pts = (struct timespec *)(uintptr_t) timeout;
-        return do_safe_futex(g2h(cpu, uaddr), op, val, pts, g2h(cpu, uaddr2),
-                             (base_op == FUTEX_CMP_REQUEUE
-                              ? tswap32(val3) : val3));
+        /*
+         * For these, the 4th argument is not TIMEOUT, but VAL2.
+         * But the prototype of do_safe_futex takes a pointer, so
+         * insert casts to satisfy the compiler.  We do not need
+         * to tswap VAL2 since it's not compared to guest memory.
+          */
+        pts = (struct timespec *)(uintptr_t)timeout;
+        timeout = 0;
+        haddr2 = g2h(cpu, uaddr2);
+        break;
     default:
         return -TARGET_ENOSYS;
     }
-}
-#endif
-
-#if defined(TARGET_NR_futex_time64)
-static int do_futex_time64(CPUState *cpu, target_ulong uaddr, int op,
-                           int val, target_ulong timeout,
-                           target_ulong uaddr2, int val3)
-{
-    struct timespec ts, *pts;
-    int base_op;
-
-    /* ??? We assume FUTEX_* constants are the same on both host
-       and target.  */
-#ifdef FUTEX_CMD_MASK
-    base_op = op & FUTEX_CMD_MASK;
-#else
-    base_op = op;
-#endif
-    switch (base_op) {
-    case FUTEX_WAIT:
-    case FUTEX_WAIT_BITSET:
-        if (timeout) {
-            pts = &ts;
-            if (target_to_host_timespec64(pts, timeout)) {
-                return -TARGET_EFAULT;
-            }
-        } else {
-            pts = NULL;
+    if (timeout) {
+        pts = &ts;
+        if (time64
+            ? target_to_host_timespec64(pts, timeout)
+            : target_to_host_timespec(pts, timeout)) {
+            return -TARGET_EFAULT;
         }
-        return do_safe_futex(g2h(cpu, uaddr), op,
-                             tswap32(val), pts, NULL, val3);
-    case FUTEX_WAKE:
-        return do_safe_futex(g2h(cpu, uaddr), op, val, NULL, NULL, 0);
-    case FUTEX_FD:
-        return do_safe_futex(g2h(cpu, uaddr), op, val, NULL, NULL, 0);
-    case FUTEX_REQUEUE:
-    case FUTEX_CMP_REQUEUE:
-    case FUTEX_WAKE_OP:
-        /* For FUTEX_REQUEUE, FUTEX_CMP_REQUEUE, and FUTEX_WAKE_OP, the
-           TIMEOUT parameter is interpreted as a uint32_t by the kernel.
-           But the prototype takes a `struct timespec *'; insert casts
-           to satisfy the compiler.  We do not need to tswap TIMEOUT
-           since it's not compared to guest memory.  */
-        pts = (struct timespec *)(uintptr_t) timeout;
-        return do_safe_futex(g2h(cpu, uaddr), op, val, pts, g2h(cpu, uaddr2),
-                             (base_op == FUTEX_CMP_REQUEUE
-                              ? tswap32(val3) : val3));
-    default:
-        return -TARGET_ENOSYS;
     }
+    return do_safe_futex(g2h(cpu, uaddr), op, val, pts, haddr2, val3);
 }
 #endif
 
@@ -8026,7 +8010,11 @@ static int open_self_maps(CPUArchState *cpu_env, int fd)
                 continue;
             }
 
+#ifdef TARGET_HPPA
+            if (h2g(max) == ts->info->stack_limit) {
+#else
             if (h2g(min) == ts->info->stack_limit) {
+#endif
                 path = "[stack]";
             } else {
                 path = e->path;
@@ -8684,6 +8672,30 @@ static abi_long do_syscall1(CPUArchState *cpu_env, int num, abi_long arg1,
         fd_trans_unregister(ret);
         return ret;
 #endif
+#if defined(__NR_pidfd_open) && defined(TARGET_NR_pidfd_open)
+    case TARGET_NR_pidfd_open:
+        return get_errno(pidfd_open(arg1, arg2));
+#endif
+#if defined(__NR_pidfd_send_signal) && defined(TARGET_NR_pidfd_send_signal)
+    case TARGET_NR_pidfd_send_signal:
+        {
+            siginfo_t uinfo;
+
+            p = lock_user(VERIFY_READ, arg3, sizeof(target_siginfo_t), 1);
+            if (!p) {
+                return -TARGET_EFAULT;
+            }
+            target_to_host_siginfo(&uinfo, p);
+            unlock_user(p, arg3, 0);
+            ret = get_errno(pidfd_send_signal(arg1, target_to_host_signal(arg2),
+                &uinfo, arg4));
+        }
+        return ret;
+#endif
+#if defined(__NR_pidfd_getfd) && defined(TARGET_NR_pidfd_getfd)
+    case TARGET_NR_pidfd_getfd:
+        return get_errno(pidfd_getfd(arg1, arg2, arg3));
+#endif
     case TARGET_NR_close:
         fd_trans_unregister(arg1);
         return get_errno(close(arg1));
@@ -9690,7 +9702,7 @@ static abi_long do_syscall1(CPUArchState *cpu_env, int num, abi_long arg1,
             }
             target_to_host_siginfo(&uinfo, p);
             unlock_user(p, arg3, 0);
-            ret = get_errno(sys_rt_sigqueueinfo(arg1, arg2, &uinfo));
+            ret = get_errno(sys_rt_sigqueueinfo(arg1, target_to_host_signal(arg2), &uinfo));
         }
         return ret;
     case TARGET_NR_rt_tgsigqueueinfo:
@@ -9703,7 +9715,7 @@ static abi_long do_syscall1(CPUArchState *cpu_env, int num, abi_long arg1,
             }
             target_to_host_siginfo(&uinfo, p);
             unlock_user(p, arg4, 0);
-            ret = get_errno(sys_rt_tgsigqueueinfo(arg1, arg2, arg3, &uinfo));
+            ret = get_errno(sys_rt_tgsigqueueinfo(arg1, arg2, target_to_host_signal(arg3), &uinfo));
         }
         return ret;
 #ifdef TARGET_NR_sigreturn
@@ -9912,11 +9924,22 @@ static abi_long do_syscall1(CPUArchState *cpu_env, int num, abi_long arg1,
             p2 = lock_user(VERIFY_WRITE, arg3, arg4, 0);
             if (!p || !p2) {
                 ret = -TARGET_EFAULT;
+            } else if (!arg4) {
+                /* Short circuit this for the magic exe check. */
+                ret = -TARGET_EINVAL;
             } else if (is_proc_myself((const char *)p, "exe")) {
                 char real[PATH_MAX], *temp;
                 temp = realpath(exec_path, real);
-                ret = temp == NULL ? get_errno(-1) : strlen(real) ;
-                snprintf((char *)p2, arg4, "%s", real);
+                /* Return value is # of bytes that we wrote to the buffer. */
+                if (temp == NULL) {
+                    ret = get_errno(-1);
+                } else {
+                    /* Don't worry about sign mismatch as earlier mapping
+                     * logic would have thrown a bad address error. */
+                    ret = MIN(strlen(real), arg4);
+                    /* We cannot NUL terminate the string. */
+                    memcpy(p2, real, ret);
+                }
             } else {
                 ret = get_errno(readlinkat(arg1, path(p), p2, arg4));
             }
@@ -12318,11 +12341,11 @@ static abi_long do_syscall1(CPUArchState *cpu_env, int num, abi_long arg1,
 #endif
 #ifdef TARGET_NR_futex
     case TARGET_NR_futex:
-        return do_futex(cpu, arg1, arg2, arg3, arg4, arg5, arg6);
+        return do_futex(cpu, false, arg1, arg2, arg3, arg4, arg5, arg6);
 #endif
 #ifdef TARGET_NR_futex_time64
     case TARGET_NR_futex_time64:
-        return do_futex_time64(cpu, arg1, arg2, arg3, arg4, arg5, arg6);
+        return do_futex(cpu, true, arg1, arg2, arg3, arg4, arg5, arg6);
 #endif
 #ifdef CONFIG_INOTIFY
 #if defined(TARGET_NR_inotify_init)
@@ -12847,15 +12870,18 @@ static abi_long do_syscall1(CPUArchState *cpu_env, int num, abi_long arg1,
                 phost_sevp = &host_sevp;
                 ret = target_to_host_sigevent(phost_sevp, arg2);
                 if (ret != 0) {
+                    free_host_timer_slot(timer_index);
                     return ret;
                 }
             }
 
             ret = get_errno(timer_create(clkid, phost_sevp, phtimer));
             if (ret) {
-                phtimer = NULL;
+                free_host_timer_slot(timer_index);
             } else {
                 if (put_user(TIMER_MAGIC | timer_index, arg3, target_timer_t)) {
+                    timer_delete(*phtimer);
+                    free_host_timer_slot(timer_index);
                     return -TARGET_EFAULT;
                 }
             }
@@ -12991,7 +13017,7 @@ static abi_long do_syscall1(CPUArchState *cpu_env, int num, abi_long arg1,
         } else {
             timer_t htimer = g_posix_timers[timerid];
             ret = get_errno(timer_delete(htimer));
-            g_posix_timers[timerid] = 0;
+            free_host_timer_slot(timerid);
         }
         return ret;
     }