Found with scripts/clean-header-guards.pl.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Reviewed-by: NRichard Henderson <rth@twiddle.net>

Signed-off-by: NLaurent Vivier <laurent@vivier.eu>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>

Support the F_GETPIPE_SZ and F_SETPIPE_SZ fcntl operations.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NLaurent Vivier <laurent@vivier.eu>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

The l_start and l_len fields in the various target_flock structures are
supposed to be '__kernel_off_t' or '__kernel_loff_t', which means they
should be signed, not unsigned. Correcting the structure definitions means
that __get_user() and __put_user() will correctly sign extend them if
the guest is using 32 bit offsets and the host is using 64 bit offsets.

This fixes failures in the LTP 'fcntl14' tests where it checks that
negative seek offsets work correctly.

We reindent the structures to drop hard tabs since we're touching 40%
of the fields anyway.

RV: long long -> abi_llong as suggested by Laurent Vivier <laurent@vivier.eu>
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

The siginfo_t struct includes a union. The correct way to identify
which fields of the union are relevant is complicated, because we
have to use a combination of the si_code and si_signo to figure out
which of the union's members are valid.  (Within the host kernel it
is always possible to tell, but the kernel carefully avoids giving
userspace the high 16 bits of si_code, so we don't have the
information to do this the easy way...) We therefore make our best
guess, bearing in mind that a guest can spoof most of the si_codes
via rt_sigqueueinfo() if it likes.  Once we have made our guess, we
record it in the top 16 bits of the si_code, so that tswap_siginfo()
later can use it.  tswap_siginfo() then strips these top bits out
before writing si_code to the guest (sign-extending the lower bits).

This fixes a bug where fields were sometimes wrong; in particular
the LTP kill10 test went into an infinite loop because its signal
handler got a si_pid value of 0 rather than the pid of the sending
process.

As part of this change, we switch to using __put_user() in the
tswap_siginfo code which writes out the byteswapped values to
the target memory, in case the target memory pointer is not
sufficiently aligned for the host CPU's requirements.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

The 64-bit x86 syscall ABI uses 32-bit UIDs; only define
USE_UID16 for 32-bit x86.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

Adds the definitions for the socket calls SOCKOP_sendmmsg
and SOCKOP_recvmmsg and wires them up with the rest of the code.
The necessary function do_sendrecvmmsg() is already present in
linux-user/syscall.c. After adding these two definitions and wiring
them up, I no longer receive an error message about the
unimplemented socket calls when running "apt-get update" on Debian
unstable running on qemu with glibc_2.21 on m68k.
Signed-off-by: NJohn Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Reviewed-by: NLaurent Vivier <laurent@vivier.eu>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

According to comments in /usr/include/linux/eventpoll.h,
poll_event is packed only on x86_64.

And to be sure fields are correctly aligned in epoll_data,
use abi_XXX types for all of them.

Moreover, fd type is wrong: fd is int, not ulong.

This has been tested with a ppc guest on an x86_64 host:
without this patch, systemd crashes (core).

CC: Alexander Graf <agraf@suse.de>
CC: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: NLaurent Vivier <laurent@vivier.eu>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

They content several new macro members, also contents TARGET_N*.
Signed-off-by: NChen Gang <gang.chen.5i5j@gmail.com>
Message-Id: <1443240605-2924-1-git-send-email-gang.chen.5i5j@gmail.com>
Signed-off-by: NRichard Henderson <rth@twiddle.net>

Currently, __target_cmsg_nxthdr compares a pointer derived from
target_cmsg against the msg_control field of target_msgh (through
subtraction).  This failed for me when emulating i386 code under x86_64,
because pointers in the host address space and pointers in the guest
address space were not the same.  This patch passes the initial value of
target_cmsg into __target_cmsg_nxthdr.

I found and fixed two more related bugs:
- __target_cmsg_nxthdr now returns the new cmsg pointer instead of the
  old one.
- tgt_space (in host_to_target_cmsg) doesn't count "sizeof (struct
  target_cmsghdr)" twice anymore.
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NJonathan Neuschäfer <j.neuschaefer@gmx.net>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

Add main working flow feature, system call processing feature, and elf64
tilegx binary loading feature, based on Linux kernel tilegx 64-bit
implementation.

[rth: Moved all of the implementation of atomic instructions to a later patch.]
Signed-off-by: NChen Gang <gang.chen.5i5j@gmail.com>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Message-Id: <BLU436-SMTP938552D42808AA60634582B9660@phx.gbl>
Signed-off-by: NRichard Henderson <rth@twiddle.net>

Signed-off-by: NVeres Lajos <vlajos@gmail.com>
Signed-off-by: NMichael Tokarev <mjt@tls.msk.ru>

Simple "hello world" MIPS N32 userland program crashes with segfault due to
incorrectly defined stat structure in QEMU.

Correct "target_stat" definition to match kernel's "stat64" as in MIPS N32
there are only plain "stat" syscalls using 64-bit structure.
Reported-by: NDaniel Sanders <daniel.sanders@imgtec.com>
Signed-off-by: NLeon Alrae <leon.alrae@imgtec.com>
Tested-by: NDaniel Sanders <daniel.sanders@imgtec.com>
Reviewed-by: NJames Hogan <james.hogan@imgtec.com>

Fix TARGET_SI_PAD_SIZE calculation to match the way the kernel does it.
Use different TARGET_SI_PREAMBLE_SIZE for 32-bit and 64-bit targets.
Signed-off-by: NMaxim Ostapenko <m.ostapenko@partner.samsung.com>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NMichael Tokarev <mjt@tls.msk.ru>

When creating a timer handle, we give the timer id a special magic offset
of 0xcafe0000. However, we never mask that offset out of the timer id before
we start using it to dereference our timer array. So we always end up aborting
timer operations because the timer id is out of bounds.

This was not an issue before my patch e52a99f7 ("linux-user: Simplify
timerid checks on g_posix_timers range") because before we would blindly mask
anything above the first 16 bits.

This patch simplifies the code around timer id creation by introducing a proper
target_timer_id typedef that is s32, just like Linux has it. It also changes the
magic offset to a value that makes all timer ids be positive.
Reported-by: NTom Musta <tommusta@gmail.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NTom Musta <tommusta@gmail.com>
Tested-by: NTom Musta <tommusta@gmail.com>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

Implement conversion of the AF_PACKET sockaddr subtype
in target_to_host_sockaddr.
Signed-off-by: NJoakim Tjernlund <Joakim.Tjernlund@transmode.se>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

Add a definition of the SIOCGIFINDEX ioctl, allowing its use by target
programs.
Signed-off-by: NPaul Burton <paul@archlinuxmips.org>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

Add a definition of the KDSIGACCEPT ioctl & allow its use by target
programs.
Signed-off-by: NPaul Burton <paul@archlinuxmips.org>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

The settimeofday syscall accepts a tz argument indicating the desired
timezone to the kernel. QEMU previously ignored any argument provided
by the target program & always passed NULL to the kernel. Instead,
translate the argument & pass along the data userland provided.

Although this argument is described by the settimeofday man page as
obsolete, it is used by systemd as of version 213.
Signed-off-by: NPaul Burton <paul@archlinuxmips.org>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

MIPS requires the pad field to 64b-align the data field just as ARM
does.
Signed-off-by: NPaul Burton <paul@archlinuxmips.org>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

Use the public sigset_t instead of the glibc specific internal
__sigset_t in _syscall.

Calculate the sigevent pad size is calculated in similar way as kernel
does it instead of using glibc internal field _pad.

This is needed for building with musl libc.
Signed-off-by: NNatanael Copa <ncopa@alpinelinux.org>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>

Implement the capget and capset syscalls. This is useful because
simple programs like 'ls' try to use it in AArch64, and otherwise
we emit a lot of noise about it being unimplemented.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

F_GETOWN is replaced by F_GETOWN_EX inside the glibc fcntl wrapper
Signed-off-by: NAndreas Schwab <schwab@suse.de>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

Glibc when built for newer kernels assumes that the sendmmsg syscall is
available. Without it, dns resolution simply fails to work.

Wrap the syscall with existing infrastructure so that we don't have a host
dependency on sendmmsg.

To avoid locking the same area of guest memory twice (which will break if
DEBUG_REMAP is defined) we pull the lock/unlock part of do_sendrecvmsg()
out into its own function so the actual implementation can be shared.
Signed-off-by: NAlexander Graf <agraf@suse.de>
[PMM: add recvmmsg support;
 handle errors (which also implies support for non-blocking operations);
 cap the vector length as the kernel implementation does;
 don't lock guest memory twice;
 support MSG_WAITFORONE flag]
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>
Reviewed-by: NRichard Henderson <rth@twiddle.net>

The AArch64 kernel defines its __kernel_uid_t type as 32 bits, unlike
32 bit ARM, so don't enable our 16-bit UID wrapper handling.
Signed-off-by: NAndreas Schwab <schwab@suse.de>
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>
Reviewed-by: NRichard Henderson <rth@twiddle.net>

Signed-off-by: NAndreas Färber <afaerber@suse.de>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

Cc: Riku Voipio <riku.voipio@iki.fi>
Signed-off-by: NAndré Hentschel <nerv@dawncrow.de>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NErik de Castro Lopo <erikd@mega-nerd.com>
Reviewed-by: NLaurent Vivier <laurent@vivier.eu>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

Signed-off-by: NErik de Castro Lopo <erikd@mega-nerd.com>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

Some targets use a stat64 structure for the stat64 syscall while others
use a stat structure. SPARC64 used the wrong kind.

Instead of extending the conditional compilation in syscall.c, now a
macro TARGET_HAS_STRUCT_STAT64 is defined whenever a target has a
target_stat64.
Signed-off-by: NStefan Weil <sw@weilnetz.de>
Reviewed-by: NErik de Castro Lopo <erikd@mega-nerd.com>

This is needed to be able to run dhclient.
Signed-off-by: NLaurent Vivier <laurent@vivier.eu>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

Some syscall handlers have special code for ARM enabled that we don't
need on AArch64. Exclude AArch64 in those cases. In other places we
can share struct definitions with other targets or have to provide our
own.

With this patch applied, most syscall definitions in linux-user should
be sound for AArch64.
Signed-off-by: NAlexander Graf <agraf@suse.de>
Signed-off-by: NJohn Rigby <john.rigby@linaro.org>
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Message-id: 1378235544-22290-16-git-send-email-peter.maydell@linaro.org
Message-id: 1368505980-17151-9-git-send-email-john.rigby@linaro.org
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

The epoll emulation uses data structures without packing them, so the
compiler might choose to add padding inside.

This patch makes the most offending one (target_epoll_event) a packed
structure to make sure we don't pad it by accident. ARM would pad it,
so declare the padding mandatory for ARM targets.

This fixes i386-on-ARM epoll emulation for me.
Signed-off-by: NAlexander Graf <agraf@suse.de>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

OpenRISC uses the asm-generic versions of target_stat and
target_stat64, but it was incorrectly using the x86/ARM/etc version
due to a misplaced defined(TARGET_OPENRISC).  The previously unused
OpenRISC section of the ifdef ladder also defined an incorrect
target_stat and omitted the target_stat64 definition.  Fix
target_stat, provide target_stat64, and add a comment noting that
these are the asm-generic versions for the benefit of future ports.
Reviewed-by: NJia Liu <proljc@gmail.com>
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

The values of F_RDLCK, F_WRLCK, F_UNLCK, F_EXLCK, F_SHLCK
differ between alpha and other linux architectures.

This patch allows to run "dpkg" (database lock).
Signed-off-by: NLaurent Vivier <laurent@vivier.eu>
Signed-off-by: NRichard Henderson <rth@twiddle.net>

Unconditional bswap replaced by __get_user/__put_user.
Signed-off-by: NRichard Henderson <rth@twiddle.net>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

In case when TARGET_ABI_BITS == 32 && HOST_LONG_BITS == 64, the last
byte of the target dirent structure (aka d_type byte) was never copied
from the host dirent structure, thus breaking everything that relies
on valid d_type value, e.g. glob(3).
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NDmitry V. Levin <ldv@altlinux.org>
Signed-off-by: NRiku Voipio <riku.voipio@linaro.org>

Fix the SNDCTL_DSP_MAP{IN,OUT}BUF ioctl definitions so that they
refer to a suitably defined target struct layout rather than hardcoding
the ioctl number. This fixes complaints from the syscall_init()
consistency check when running an x86_64-to-x86_64 linux-user qemu.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

The definitions for the ioctl numbers TARGET_BLKBSZGET and
TARGET_BLKBSZSET had the wrong size parameters (they are defined
with size_t, not int, even though the ioctl implementations themselves
read and write integers). Since commit 354a0008 we now have an
ioctl wrapper definition for BLKBSZGET and so on an x86-64-to-x86-64
linux-user binary we were triggering the mismatch warning in
syscall_init().
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

For each target, only define the bits that appear in
arch/target/include/asm/fcntl.h.  Mirror the kernel's
asm-generic layout by handling anything undefined afterward.
Signed-off-by: NRichard Henderson <rth@twiddle.net>