Distinguish host SIGABRT from guest SIGABRT by mapping
the guest signal onto one of the host RT signals.
This prevents a cycle by which a host assertion failure
is caught and handled by host_signal_handler, queued for
the guest, and then we attempt to continue past the
host abort. What happens next depends on the host libc,
but is neither good nor helpful.
Acked-by: Helge Deller <deller@gmx.de>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
These signals, when not spoofed via kill(), are always bugs.
Use die_from_signal to report this sensibly.
Acked-by: Helge Deller <deller@gmx.de>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Install the host signal handler at the same time we are
probing the target signals for SIG_IGN/SIG_DFL. Ignore
unmapped target signals.
Acked-by: Helge Deller <deller@gmx.de>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Do not return a valid signal number in one domain
when given an invalid signal number in the other domain.
Acked-by: Helge Deller <deller@gmx.de>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
The set of fatal signals is really immaterial. If one arrives,
and is unhandled, then the qemu process dies and the parent gets
the correct signal.
It is only for those signals which we would like to perform a
guest core dump instead of a host core dump that we need to catch.
Acked-by: Helge Deller <deller@gmx.de>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
If there is an internal program error in the qemu source code which
raises SIGSEGV or SIGBUS, we currently assume the signal belongs to
the guest. With an artificial error introduced, we will now print
QEMU internal SIGSEGV {code=MAPERR, addr=(nil)}
Signed-off-by: Helge Deller <deller@gmx.de>
Message-Id: <20230812164314.352131-1-deller@gmx.de>
[rth: Use in_code_gen_buffer and die_with_signal; drop backtrace]
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This line is supposed to be unreachable, but if we're going to
have it at all, SIGABRT via abort() is subject to the same signal
peril that created this function in the first place.
We can _exit immediately without peril.
Acked-by: Helge Deller <deller@gmx.de>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Because we trap so many signals for use by the guest,
we have to take extra steps to exit properly.
Acked-by: Helge Deller <deller@gmx.de>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This build option has been deprecated since 8.0.
Remove all CONFIG_GPROF code that depends on that,
including one errant check using TARGET_GPROF.
Acked-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
The process was pretty similar to the softmmu move except we take the
time to split stuff between user.c and user-target.c to avoid as much
target specific compilation as possible. We also start to make use of
our shiny new header scheme so the user-only helpers can be included
without the rest of the exec/gsbstub.h cruft.
As before we split some functions into user and softmmu versions
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230302190846.2593720-12-alex.bennee@linaro.org>
Message-Id: <20230303025805.625589-12-richard.henderson@linaro.org>
When exiting due to an exit() syscall, qemu-user calls
preexit_cleanup(), but this is currently not the case when exiting due
to a signal. This leads to various buffers not being flushed (e.g.,
for gprof, for gcov, and for the upcoming perf support).
Add the missing call.
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230112152013.125680-2-iii@linux.ibm.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Some of the guest signal numbers are currently not converted to
their representative names in the strace output, e.g. SIGVTALRM.
This patch introduces a smart way to generate and keep in sync the
host-to-guest and guest-to-host signal conversion tables for usage in
the qemu signal and strace code. This ensures that any signals
will now show up in both tables.
There is no functional change in this patch - with the exception that yet
missing signal names now show up in the strace code too.
Signed-off-by: Helge Deller <deller@gmx.de>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20220918194555.83535-2-deller@gmx.de>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
In linux-user/signal.c we have two FIXME comments claiming that
parts of the signal-handling code are not threadsafe. These are
very old, as they were first introduced in commit 624f797905
in 2008. Since then we've radically overhauled the signal-handling
logic, while carefully preserving these FIXME comments.
It's unclear exactly what thread-safety issue the original
author was trying to point out -- the relevant data structures
are in the TaskStruct, which makes them per-thread and only
operated on by that thread. The old code at the time of that
commit did have various races involving signal handlers being
invoked at awkward times; possibly this was what was meant.
Delete these FIXME comments:
* they were written at a time when the way we handled
signals was completely different
* the code today appears to us to not have thread-safety issues
* nobody knows what the problem the comments were trying to
point out was
so they are serving no useful purpose for us today.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Warner Losh <imp@bsdimp.com>
Message-Id: <20220114155032.3767771-1-peter.maydell@linaro.org>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Do not directly access ucontext_t as the third signal parameter.
This is preparation for a sparc64 fix.
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
The linux-user queue_signal() function always returns 1, and none of
its callers check the return value. Give it a void return type
instead.
The return value is a leftover from the old pre-2016 linux-user
signal handling code, which really did have a queue of signals and so
might return a failure indication if too many signals were queued at
once. The current design avoids having to ever have more than one
signal queued via queue_signal() at once, so it can never fail.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20220114153732.3767229-4-peter.maydell@linaro.org>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
When converting a siginfo_t from waitid(), the interpretation of si_status
depends on the value of si_code: For CLD_EXITED, it is an exit code and
should be copied verbatim. For other codes, it is a signal number
(possibly with additional high bits from ptrace) that should be mapped.
This code was previously changed in commit 1c3dfb506e
("linux-user/signal: Decode waitid si_code"), but the fix was
incomplete.
Tested with the following test program:
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <sys/wait.h>
int main() {
pid_t pid = fork();
if (pid == 0) {
exit(12);
} else {
siginfo_t siginfo = {};
waitid(P_PID, pid, &siginfo, WEXITED);
printf("Code: %d, status: %d\n", (int)siginfo.si_code, (int)siginfo.si_status);
}
pid = fork();
if (pid == 0) {
raise(SIGUSR2);
} else {
siginfo_t siginfo = {};
waitid(P_PID, pid, &siginfo, WEXITED);
printf("Code: %d, status: %d\n", (int)siginfo.si_code, (int)siginfo.si_status);
}
}
Output with an x86_64 host and mips64el target before 1c3dfb506e
(incorrect: exit code 12 is translated like a signal):
Code: 1, status: 17
Code: 2, status: 17
After 1c3dfb506e (incorrect: signal number is not translated):
Code: 1, status: 12
Code: 2, status: 12
With this patch:
Code: 1, status: 12
Code: 2, status: 17
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <81534fde7cdfc6acea4889d886fbefdd606630fb.1635019124.git.mschiffer@universe-factory.net>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Move linux-user safe-syscall.S and safe-syscall-error.c to common-user
so that bsd-user can also use it. Also move safe-syscall.h to
include/user/. Since there is nothing here that is related to the guest,
as opposed to the host, build it once.
Reviewed-by: Warner Losh <imp@bsdimp.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This value is fully internal to qemu, and so is not a TARGET define.
We use this as an extra marker for both host and target errno.
Reviewed-by: Warner Losh <imp@bsdimp.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
All supported hosts now define HAVE_SAFE_SYSCALL, so remove
the ifdefs. This leaves hostdep.h empty, so remove it.
Reviewed-by: Warner Losh <imp@bsdimp.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
All instances of rewind_if_in_safe_syscall are the same, differing only
in how the instruction point is fetched from the ucontext and the size
of the registers. Use host_signal_pc and new host_signal_set_pc
interfaces to fetch the pointer to the PC and adjust if needed. Delete
all the old copies of rewind_if_in_safe_syscall.
Acked-by: Laurent Vivier <laurent@vivier.eu>
Signed-off-by: Warner Losh <imp@bsdimp.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20211113045603.60391-3-imp@bsdimp.com>
[rth: include safe-syscall.h, simplify ifdefs]
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Handle BUS_ADRALN via cpu_loop_exit_sigbus, but allow other SIGBUS
si_codes to continue into the host-to-guest signal conversion code.
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This is a new interface to be provided by the os emulator for
raising SIGBUS on fault. Use the new record_sigbus target hook.
Reviewed-by: Warner Losh <imp@bsdimp.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
We have replaced tlb_fill with record_sigsegv for user mode.
Move the declaration to restrict it to system emulation.
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This is a new interface to be provided by the os emulator for
raising SIGSEGV on fault. Use the new record_sigsegv target hook.
Reviewed by: Warner Losh <imp@bsdimp.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Now that all of the linux-user hosts have been converted
to host-signal.h, drop the compatibility code.
Reviewed by: Warner Losh <imp@bsdimp.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Add stub host-signal.h for all linux-user hosts.
Add new code replacing cpu_signal_handler.
Full migration will happen one host at a time.
Reviewed-by: Warner Losh <imp@bsdimp.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
In many places in the linux-user code we need to queue a signal for
the guest using the QEMU_SI_FAULT si_type. This requires that the
caller sets up and passes us a target_siginfo, including setting the
appropriate part of the _sifields union for the si_type. In a number
of places the code forgets to set the _sifields union field.
Provide a new force_sig_fault() function, which does the same thing
as the Linux kernel function of that name -- it takes the signal
number, the si_code value and the address to use in
_sifields._sigfault, and assembles the target_siginfo itself. This
makes the callsites simpler and means it's harder to forget to pass
in an address value.
We follow force_sig() and the kernel's force_sig_fault() in not
requiring the caller to pass in the CPU pointer but always acting
on the CPU of the current thread.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20210813131809.28655-6-peter.maydell@linaro.org>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Currently the linux-user qemu.h pulls in gdbstub.h. There's no real reason
why it should do this; include it directly from the C files which require
it, and drop the include line in qemu.h.
(Note that several of the C files previously relying on this indirect
include were going out of their way to only include gdbstub.h conditionally
on not CONFIG_USER_ONLY!)
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20210908154405.15417-9-peter.maydell@linaro.org>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
qemu.h is included in various non-linux-user files (which
mostly want the TaskState struct and the functions for
doing usermode access to guest addresses like lock_user(),
unlock_user(), get_user*(), etc).
Split out the parts that are only used in linux-user itself
into a new user-internals.h. This leaves qemu.h with basically
three things:
* the definition of the TaskState struct
* the user-access functions and macros
* do_brk()
all of which are needed by code outside linux-user that
includes qemu.h.
The addition of all the extra #include lines was done with
sed -i '/include.*qemu\.h/a #include "user-internals.h"' $(git grep -l 'include.*qemu\.h' linux-user)
(and then undoing the change to fpa11.h).
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20210908154405.15417-8-peter.maydell@linaro.org>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
This check is to ensure that the loop in signal_table_init() from
SIGRTMIN to SIGRTMAX falls within the bounds of host_to_target_signal_table
(_NSIG). However, it is not critical, since _NSIG is already defined
to be the one larger than the largest signal supported by the system
(as specified in the upcoming POSIX revision[0]).
musl libc does not define __SIGRTMAX, so disabling this check when
it is missing fixes one of the last remaining errors when building
qemu.
[0] https://www.austingroupbugs.net/view.php?id=741
Signed-off-by: Michael Forney <mforney@mforney.org>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20210526190203.4255-1-mforney@mforney.org>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
In most cases we were already passing get_sp_from_cpustate
directly to the function. In other cases, we were passing
a local variable which already contained the same value.
In the rest of the cases, we were passing the stack pointer
out of env directly.
Reviewed by: Warner Losh <imp@bsdimp.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20210426025334.1168495-5-richard.henderson@linaro.org>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>