The most interesting or most complicated part here is the syscall_nr.h
generators. In order to keep the generation logic all in meson.build,
I am adding to config_target the name of the .tbl file, and making the
generated file syscall<SUFFIX>_nr.h for input file syscall<SUFFIX>.tbl.
For architectures where the input file is not named syscall_nr.tbl,
syscall_nr.h has to be a source file; it's just a forwarder for x86
(i386/x86_64), while for MIPS64 it chooses between N32 and N64 ABIs.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Meson doesn't enjoy the same flexibility we have with Make in choosing
the include path. In particular the tracing headers are using
$(build_root)/$(<D).
In order to keep the include directives unchanged,
the simplest solution is to generate headers with patterns like
"trace/trace-audio.h" and place forwarding headers in the source tree
such that for example "audio/trace.h" includes "trace/trace-audio.h".
This patch is too ugly to be applied to the Makefiles now. It's only
a way to separate the changes to the tracing header files from the
Meson rewrite of the tracing logic.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The glibc getcwd function returns different errors than the getcwd
syscall, which triggers an assertion failure in the glibc getcwd function
when running under the emulation.
When the syscall returns ENAMETOOLONG, the glibc wrapper uses a fallback
implementation that potentially handles an unlimited path length, and
returns with ERANGE if the provided buffer is too small. The qemu
emulation cannot distinguish the two cases, and thus always returns ERANGE.
This is unexpected by the glibc wrapper.
Signed-off-by: Andreas Schwab <schwab@suse.de>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <mvmmu3qplvi.fsf@suse.de>
[lv: updated description]
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Implementation of 'rt_sigtimedwait()' in 'syscall.c' uses the
function 'target_to_host_timespec()' to transfer the value of
'struct timespec' from target to host. However, the implementation
doesn't check whether this conversion succeeds and thus can cause
an unaproppriate error instead of the 'EFAULT (Bad address)' which
is supposed to be set if the conversion from target to host fails.
This was confirmed with the LTP test for rt_sigtimedwait:
"/testcases/kernel/syscalls/rt_sigtimedwait/rt_sigtimedwait01.c"
which causes an unapropriate error in test case "test_bad_adress3"
which is run with a bad adress for the 'struct timespec' argument:
FAIL: test_bad_address3 (349): Unexpected failure: EAGAIN/EWOULDBLOCK (11)
The test fails with an unexptected errno 'EAGAIN/EWOULDBLOCK' instead
of the expected EFAULT.
After the changes from this patch, the test case is executed successfully
along with the other LTP test cases for 'rt_sigtimedwait()':
PASS: test_bad_address3 (349): Test passed
Signed-off-by: Filip Bozuta <Filip.Bozuta@syrmia.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200724181651.167819-1-Filip.Bozuta@syrmia.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
When the chroot does not have /proc mounted, we can read neither
/proc/sys/vm/mmap_min_addr nor /proc/sys/maps.
The enforcement of mmap_min_addr in the host kernel is done by
the security module, and so does not apply to processes owned
by root. Which leads pgd_find_hole_fallback to succeed in probing
a reservation at address 0. Which confuses pgb_reserved_va to
believe that guest_base has not actually been initialized.
We don't actually want NULL addresses to become accessible, so
make sure that mmap_min_addr is initialized with a non-zero value.
Buglink: https://bugs.launchpad.net/qemu/+bug/1888728
Reported-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Tested-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Acked-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200724212314.545877-1-richard.henderson@linaro.org>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Our safe_clock_nanosleep() returns -1 and updates errno.
We don't need to update the CRF bit in syscall.c because it will
be updated in ppc/cpu_loop.c as the return value is negative.
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200722174612.2917566-3-laurent@vivier.eu>
Message-Id: <20200724064509.331-14-alex.bennee@linaro.org>
If the call is interrupted by a signal handler, it fails with error EINTR
and if "remain" is not NULL and "flags" is not TIMER_ABSTIME, it returns
the remaining unslept time in "remain".
Update linux-user to not overwrite the "remain" structure if there is no
error.
Found with "make check-tcg", linux-test fails on nanosleep test:
TEST linux-test on x86_64
.../tests/tcg/multiarch/linux-test.c:242: nanosleep
Reported-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200722174612.2917566-2-laurent@vivier.eu>
Message-Id: <20200724064509.331-13-alex.bennee@linaro.org>
Plain MAP_FIXED has the undesirable behaviour of splatting exiting
maps so we don't actually achieve what we want when looking for gaps.
We should be using MAP_FIXED_NOREPLACE. As this isn't always available
we need to potentially check the returned address to see if the kernel
gave us what we asked for.
Fixes: ad592e37df ("linux-user: provide fallback pgd_find_hole for bare chroots")
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200724064509.331-9-alex.bennee@linaro.org>
print_syscall_err() relies on the sign of the returned value to know
if it is an errno value or not.
But in some cases the returned value can have the most signicant bit
set without being an errno.
This patch restores previous behaviour that was also checking if
we can decode the errno to validate it.
This patch fixes this kind of problem (qemu-m68k):
root@sid:/# QEMU_STRACE= ls
3 brk(NULL) = -1 errno=21473607683 uname(0x407fff8a) = 0
to become:
root@sid:/# QEMU_STRACE= ls
3 brk(NULL) = 0x8001e000
3 uname(0xffffdf8a) = 0
Fixes: c84be71f68 ("linux-user: Extend strace support to enable argument printing after syscall execution")
Cc: Filip.Bozuta@syrmia.com
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200708152435.706070-3-laurent@vivier.eu>
errno of the target is returned as a negative value by the syscall,
not in the host errno variable.
The emulation of the target syscall can return an error while the
host doesn't set an errno value. Target errnos and host errnos can
also differ in some cases.
Fixes: c84be71f68 ("linux-user: Extend strace support to enable argument printing after syscall execution")
Cc: Filip.Bozuta@syrmia.com
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Filip Bozuta <Filip.Bozuta@syrmia.com>
Message-Id: <20200708152435.706070-2-laurent@vivier.eu>
This command is needed to be able to boot systemd in a container.
$ sudo systemd-nspawn -D /chroot/armhf/sid/ -b
Spawning container sid on /chroot/armhf/sid.
Press ^] three times within 1s to kill container.
systemd 245.6-2 running in system mode.
Detected virtualization systemd-nspawn.
Detected architecture arm.
Welcome to Debian GNU/Linux bullseye/sid!
Set hostname to <virt-arm>.
Failed to enqueue loopback interface start request: Operation not supported
Caught <SEGV>, dumped core as pid 3.
Exiting PID 1...
Container sid failed with error code 255.
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200709072332.890440-2-laurent@vivier.eu>
Only implement IFLA_PERM_ADDRESS to fix the following error:
Unknown host QEMU_IFLA type: 54
The couple of other ones, IFLA_PROP_LIST and IFLA_ALT_IFNAME, have
been introduced to be used with RTM_NEWLINKPROP, RTM_DELLINKPROP and
RTM_GETLINKPROP that are not implemented by QEMU.
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200709072332.890440-1-laurent@vivier.eu>
In new functions print_ioctl() and print_syscall_ret_ioctl(), we don't
check if lock_user() returns NULL and this would cause a segfault in
thunk_print().
If lock_user() returns NULL don't call thunk_print() but prints only the
value of the (invalid) pointer.
Tested with:
# cat ioctl.c
#include <unistd.h>
#include <sys/ioctl.h>
int main(void)
{
int ret;
ret = ioctl(STDOUT_FILENO, TCGETS, 0xdeadbeef);
ret = ioctl(STDOUT_FILENO, TCSETSF, 0xdeadbeef);
return 0;
}
# QEMU_STRACE= ./ioctl
...
578 ioctl(1,TCGETS,0xdeadbeef) = -1 errno=2 (Bad address)
578 ioctl(1,TCSETSF,0xdeadbeef) = -1 errno=2 (Bad address)
...
# QEMU_STRACE= passwd
...
623 ioctl(0,TCGETS,0x3fffed04) = 0 ({})
623 ioctl(0,TCSETSF,{}) = 0
...
Reported-by: Peter Maydell <peter.maydell@linaro.org>
Fixes: 79482e5987 ("linux-user: Add strace support for printing arguments of ioctl()")
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Refactoring ipc syscall for s390x and SPARC, so it matches glibc implementation
We should add support of semtimedop syscall as new version of glibc
2.31 uses semop based on semtimedop
(commit: 765cdd0bff ).
Signed-off-by: Matus Kysel <mkysel@tachyum.com>
Message-Id: <20200626124612.58593-2-mkysel@tachyum.com>
Message-Id: <20200626124612.58593-3-mkysel@tachyum.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
[lv: merged PATCH 1 & 2 to avoid build break on PATCH 1]
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Linux uses the EPROTONOSUPPORT error code[1] if the users requests a
netlink socket with an unsupported netlink protocol. This change
switches linux-user to use the same code as Linux, instead of
EPFNOSUPPORT (which AFAIK is just an anachronistic version of
EAFNOSUPPORT).
Tested by compiling all linux-user targets on x86.
[1]:
bfe91da29b/net/netlink/af_netlink.c (L683)
Signed-off-by: Josh Kunz <jkz@google.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200707001036.1671982-1-jkz@google.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Given we assert the requested address matches what we asked we should
also make that clear in the mmap flags. Otherwise we see failures in
the GitLab environment for some currently unknown but allowable
reason. We use MAP_FIXED_NOREPLACE if we can so we don't just clobber
an existing mapping. Also include the strerror string for a bit more
info on failure.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20200701135652.1366-34-alex.bennee@linaro.org>
The object_property_set_FOO() setters take property name and value in
an unusual order:
void object_property_set_FOO(Object *obj, FOO_TYPE value,
const char *name, Error **errp)
Having to pass value before name feels grating. Swap them.
Same for object_property_set(), object_property_get(), and
object_property_parse().
Convert callers with this Coccinelle script:
@@
identifier fun = {
object_property_get, object_property_parse, object_property_set_str,
object_property_set_link, object_property_set_bool,
object_property_set_int, object_property_set_uint, object_property_set,
object_property_set_qobject
};
expression obj, v, name, errp;
@@
- fun(obj, v, name, errp)
+ fun(obj, name, v, errp)
Chokes on hw/arm/musicpal.c's lcd_refresh() with the unhelpful error
message "no position information". Convert that one manually.
Fails to convert hw/arm/armsse.c, because Coccinelle gets confused by
ARMSSE being used both as typedef and function-like macro there.
Convert manually.
Fails to convert hw/rx/rx-gdbsim.c, because Coccinelle gets confused
by RXCPU being used both as typedef and function-like macro there.
Convert manually. The other files using RXCPU that way don't need
conversion.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20200707160613.848843-27-armbru@redhat.com>
[Straightforwad conflict with commit 2336172d9b "audio: set default
value for pcspk.iobase property" resolved]
This patch implements functionality for strace argument printing for ioctls.
When running ioctls through qemu with "-strace", they get printed in format:
"ioctl(fd_num,0x*,0x*) = ret_value"
where the request code an the ioctl's third argument get printed in a hexadicemal
format. This patch changes that by enabling strace to print both the request code
name and the contents of the third argument. For example, when running ioctl
RTC_SET_TIME with "-strace", with changes from this patch, it gets printed in
this way:
"ioctl(3,RTC_SET_TIME,{12,13,15,20,10,119,0,0,0}) = 0"
In case of IOC_R type ioctls, the contents of the third argument get printed
after the return value, and the argument inside the ioctl call gets printed
as pointer in hexadecimal format. For example, when running RTC_RD_TIME with
"-strace", with changes from this patch, it gets printed in this way:
"ioctl(3,RTC_RD_TIME,0x40800374) = 0 ({22,9,13,11,5,120,0,0,0})"
In case of IOC_RW type ioctls, the contents of the third argument get printed
both inside the ioctl call and after the return value.
Implementation notes:
Functions "print_ioctl()" and "print_syscall_ret_ioctl()", that are defined
in "strace.c", are listed in file "strace.list" as "call" and "result"
value for ioctl. Structure definition "IOCTLEntry" as well as predefined
values for IOC_R, IOC_W and IOC_RW were cut and pasted from file "syscall.c"
to file "qemu.h" so that they can be used by these functions to print the
contents of the third ioctl argument. Also, the "static" identifier for array
"ioctl_entries[]" was removed and this array was declared as "extern" in "qemu.h"
so that it can also be used by these functions. To decode the structure type
of the ioctl third argument, function "thunk_print()" was defined in file
"thunk.c" and its definition is somewhat simillar to that of function
"thunk_convert()".
Signed-off-by: Filip Bozuta <Filip.Bozuta@syrmia.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200619124727.18080-3-filip.bozuta@syrmia.com>
[lv: fix close-bracket]
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Socket ioctls SIOCGSTAMP and SIOCGSTAMPNS, used for timestamping the socket
connection, are defined in file "ioctls.h" differently from other ioctls.
The reason for this difference is explained in the comments above their definition.
These ioctls didn't have defined thunk argument types before changes from this
patch. They have special handling functions ("do_ioctl_SIOCGSTAMP" and
"do_ioctl_SIOCGSTAMPNS") that take care of setting values for approppriate argument
types (struct timeval and struct timespec) and thus no thunk argument types were
needed for their implementation. But this patch adds those argument type definitions
in file "syscall_types.h" and "ioctls.h" as it is needed for printing arguments
of these ioctls with strace.
Implementation notes:
There are two variants of these ioctls: SIOCGSTAMP_OLD/SIOCGSTAM_NEW and
SIOCGSTAMPNS_OLD/SIOCGSTAMPNS_NEW. One is the old existing definition and the
other is the 2038 safe variant used for 32-bit architectures. Corresponding
structure definitions STRUCT_timespec/STRUCT__kernel_timespec and
STRUCT_timeval/STRUCT__kernel_sock_timeval were added for these variants.
STRUCT_timeval definition was already inside the file as it is used by
another implemented ioctl. Two cases were added for definitions
STRUCT_timeval/STRUCT__kernel_sock_timeval to manage the case when the
"u_sec" field of the timeval structure is of type int.
Signed-off-by: Filip Bozuta <Filip.Bozuta@syrmia.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200619124727.18080-2-filip.bozuta@syrmia.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
This patch implements strace argument printing functionality for following syscall:
*fallocate - manipulate file space
int fallocate(int fd, int mode, off_t offset, off_t len)
man page: https://www.man7.org/linux/man-pages/man2/fallocate.2.html
Implementation notes:
This syscall's second argument "mode" is composed of predefined values
which represent flags that determine the type of operation that is
to be performed on the file space. For that reason, a printing
function "print_fallocate" was stated in file "strace.list". This printing
function uses an already existing function "print_flags()" to print flags of
the "mode" argument. These flags are stated inside an array "falloc_flags"
that contains values of type "struct flags". These values are instantiated
using an existing macro "FLAG_GENERIC()". Most of these flags are defined
after kernel version 3.0 which is why they are enwrapped in an #ifdef
directive.
The syscall's third ant fourth argument are of type "off_t" which can
cause variations between 32/64-bit architectures. To handle this variation,
function "target_offset64()" was copied from file "strace.c" and used in
"print_fallocate" to print "off_t" arguments for 32-bit architectures.
Signed-off-by: Filip Bozuta <Filip.Bozuta@syrmia.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200619123331.17387-7-filip.bozuta@syrmia.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
This patch implements strace argument printing functionality for syscalls:
*chown, lchown - change ownership of a file
int chown(const char *pathname, uid_t owner, gid_t group)
int lchown(const char *pathname, uid_t owner, gid_t group)
man page: https://www.man7.org/linux/man-pages/man2/lchown.2.html
Implementation notes:
Both syscalls use strings as arguments and thus a separate
printing function was stated in "strace.list" for them.
Both syscalls share the same number and types of arguments
and thus share a same definition in file "syscall.c".
This defintion uses existing functions "print_string()" to
print the string argument and "print_raw_param()" to print
other two arguments that are of basic types.
Signed-off-by: Filip Bozuta <Filip.Bozuta@syrmia.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200619123331.17387-6-filip.bozuta@syrmia.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
This patch implements strace argument printing functionality for syscall:
*lseek - reposition read/write file offset
off_t lseek(int fd, off_t offset, int whence)
man page: https://www.man7.org/linux/man-pages/man2/lseek.2.html
Implementation notes:
The syscall's third argument "whence" has predefined values:
"SEEK_SET","SEEK_CUR","SEEK_END","SEEK_DATA","SEEK_HOLE"
and thus a separate printing function "print_lseek" was stated
in file "strace.list". This function is defined in "strace.c"
by using an existing function "print_raw_param()" to print
the first and second argument and a switch(case) statement
for the predefined values of the third argument.
Values "SEEK_DATA" and "SEEK_HOLE" are defined in kernel version 3.1.
That is the reason why case statements for these values are
enwrapped in #ifdef directive.
Signed-off-by: Filip Bozuta <Filip.Bozuta@syrmia.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200619123331.17387-5-filip.bozuta@syrmia.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
This patch implements strace argument printing functionality for following syscalls:
*getxattr, lgetxattr, fgetxattr - retrieve an extended attribute value
ssize_t getxattr(const char *path, const char *name, void *value, size_t size)
ssize_t lgetxattr(const char *path, const char *name, void *value, size_t size)
ssize_t fgetxattr(int fd, const char *name, void *value, size_t size)
man page: https://www.man7.org/linux/man-pages/man2/getxattr.2.html
*listxattr, llistxattr, flistxattr - list extended attribute names
ssize_t listxattr(const char *path, char *list, size_t size)
ssize_t llistxattr(const char *path, char *list, size_t size)
ssize_t flistxattr(int fd, char *list, size_t size)
man page: https://www.man7.org/linux/man-pages/man2/listxattr.2.html
*removexattr, lremovexattr, fremovexattr - remove an extended attribute
int removexattr(const char *path, const char *name)
int lremovexattr(const char *path, const char *name)
int fremovexattr(int fd, const char *name)
man page: https://www.man7.org/linux/man-pages/man2/removexattr.2.html
Implementation notes:
All of the syscalls have strings as argument types and thus a separate
printing function was stated in file "strace.list" for every one of them.
All of these printing functions were defined in "strace.c" using existing
printing functions for appropriate argument types:
"print_string()" - for (const char*) type
"print_pointer()" - for (char*) and (void *) type
"print_raw_param()" for (int) and (size_t) type
Syscalls "getxattr()" and "lgetxattr()" have the same number and type of
arguments and thus their print functions ("print_getxattr", "print_lgetxattr")
share a same definition. The same statement applies to syscalls "listxattr()"
and "llistxattr()".
Function "print_syscall_ret_listxattr()" was added to print the returned list
of extended attributes for syscalls "print_listxattr(), print_llistxattr() and
print_flistxattr()".
Signed-off-by: Filip Bozuta <Filip.Bozuta@syrmia.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200619123331.17387-4-filip.bozuta@syrmia.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
This patch implements strace argument printing functionality for following syscalls:
*acct - switch process accounting on or off
int acct(const char *filename)
man page: https://www.man7.org/linux/man-pages/man2/acct.2.html
*fsync, fdatasync - synchronize a file's in-core state with storage device
int fsync(int fd)
int fdatasync(int fd)
man page: https://www.man7.org/linux/man-pages/man2/fsync.2.html
*listen - listen for connections on a socket
int listen(int sockfd, int backlog)
man page: https://www.man7.org/linux/man-pages/man2/listen.2.html
Implementation notes:
Syscall acct() takes string as its only argument and thus a separate
print function "print_acct" is stated in file "strace.list". This
function is defined and implemented in "strace.c" by using an
existing function used to print string arguments: "print_string()".
All the other syscalls have only primitive argument types, so the
rest of the implementation was handled by stating an appropriate
printing format in file "strace.list".
Signed-off-by: Filip Bozuta <Filip.Bozuta@syrmia.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200619123331.17387-3-filip.bozuta@syrmia.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Structure "struct syscallname" in file "strace.c" is used for "-strace"
to print arguments and return values of syscalls. The last field of
this structure "result" represents the calling function that prints the
return values. This field was extended in this patch so that this function
takes all syscalls arguments beside the return value. In this way, it enables
"-strace" to print arguments of syscalls that have changed after the syscall
execution. This extension will be useful as there are many syscalls that
return values inside their arguments (i.e. listxattr() that returns the list
of extended attributes inside the "list" argument).
Implementation notes:
Since there are already three existing "print_syscall_ret*" functions inside
"strace.c" ("print_syscall_ret_addr()", "print_syscall_ret_adjtimex()",
"print_syscall_ret_newselect()"), they were changed to have all syscall arguments
beside the return value. This was done so that these functions don't cause build
errors (even though syscall arguments are not used in these functions).
There is code repetition in these functions for checking the return value
and printing the approppriate error message (this code is also located in
print_syscall_ret() at the end of "strace.c"). That is the reason why a
function "syscall_print_err()" was added for this code and put inside these
functions. Functions "print_newselect()" and "print_syscall_ret_newselect()"
were changed to use this new implemented functionality and not store the syscall
argument values in separate static variables.
Signed-off-by: Filip Bozuta <Filip.Bozuta@syrmia.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200619123331.17387-2-filip.bozuta@syrmia.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Fix the handling of window spill traps by keeping cansave into account
when calculating the new CWP.
Signed-off-by: Giuseppe Musacchio <thatlemon@gmail.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200625091204.3186186-3-laurent@vivier.eu>
Relaxing the restrictions on 64 bit guests leads to the user being
able to attempt to map right at the edge of addressable memory. This
in turn lead to address overflow tripping the assert in page_set_flags
when the end address wrapped around.
Detect the wrap earlier and correctly -ENOMEM the guest (in the
reported case LTP mmap15).
Fixes: 7d8cbbabcb
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reported-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200605154929.26910-15-alex.bennee@linaro.org>
We rely on the pointer to wrap when accessing the high address of the
COMMPAGE so it lands somewhere reasonable. However on 32 bit hosts we
cannot afford just to map the entire 4gb address range. The old mmap
trial and error code handled this by just checking we could map both
the guest_base and the computed COMMPAGE address.
We can't just manipulate loadaddr to get what we want so we introduce
an offset which pgb_find_hole can apply when looking for a gap for
guest_base that ensures there is space left to map the COMMPAGE
afterwards.
This is arguably a little inefficient for the one 32 bit
value (kuser_helper_version) we need to keep there given all the
actual code entries are picked up during the translation phase.
Fixes: ee94743034
Bug: https://bugs.launchpad.net/qemu/+bug/1880225
Cc: Bug 1880225 <1880225@bugs.launchpad.net>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Tested-by: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
Cc: Richard Henderson <richard.henderson@linaro.org>
Cc: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20200605154929.26910-13-alex.bennee@linaro.org>
When running QEMU out of a chroot environment we may not have access
to /proc/self/maps. As there is no other "official" way to introspect
our memory map we need to fall back to the original technique of
repeatedly trying to mmap an address range until we find one that
works.
Fortunately it's not quite as ugly as the original code given we
already re-factored the complications of dealing with the
ARM_COMMPAGE. We do make an attempt to skip over brk() which is about
the only concrete piece of information we have about the address map
at this moment.
Fixes: ee9474303
Reported-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20200605154929.26910-12-alex.bennee@linaro.org>
Fixes: https://bugs.launchpad.net/bugs/1876373
This code path in mmap occurs when a page size is decreased with mremap. When a section of pages is shrunk, qemu calls mmap_reserve on the pages that were released. However, it has the diff operation reversed, subtracting the larger old_size from the smaller new_size. Instead, it should be subtracting the smaller new_size from the larger old_size. You can also see in the previous line of the change that this mmap_reserve call only occurs when old_size > new_size.
Bug: https://bugs.launchpad.net/qemu/+bug/1876373
Signed-off-by: Jonathan Marler <johnnymarler@gmail.com>
Reviewded-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200502161225.14346-1-johnnymarler@gmail.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
- Build with other images instead of the broken Debian containers
- Fix building with the latest version of Clang (at least wrt. to
the gitlab-CI pipeline)
- Add Philippe, Alex and Wainer to the Gitlab-CI section in MAINTAINERS
-----BEGIN PGP SIGNATURE-----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=TThW
-----END PGP SIGNATURE-----
Merge remote-tracking branch 'remotes/huth-gitlab/tags/pull-request-2020-05-28' into staging
Fixes and improvements for the gitlab-CI:
- Build with other images instead of the broken Debian containers
- Fix building with the latest version of Clang (at least wrt. to
the gitlab-CI pipeline)
- Add Philippe, Alex and Wainer to the Gitlab-CI section in MAINTAINERS
# gpg: Signature made Thu 28 May 2020 10:16:15 BST
# gpg: using RSA key 27B88847EEE0250118F3EAB92ED9D774FE702DB5
# gpg: issuer "thuth@redhat.com"
# gpg: Good signature from "Thomas Huth <th.huth@gmx.de>" [full]
# gpg: aka "Thomas Huth <thuth@redhat.com>" [full]
# gpg: aka "Thomas Huth <huth@tuxfamily.org>" [full]
# gpg: aka "Thomas Huth <th.huth@posteo.de>" [unknown]
# Primary key fingerprint: 27B8 8847 EEE0 2501 18F3 EAB9 2ED9 D774 FE70 2DB5
* remotes/huth-gitlab/tags/pull-request-2020-05-28:
gitlab-ci: Determine the number of jobs dynamically
gitlab-ci: Do not use the standard container images from gitlab
gitlab-ci: Move edk2 and opensbi YAML files to .gitlab-ci.d folder
GitLab CI: avoid calling before_scripts on unintended jobs
gitlab-ci: Remove flex/bison packages
MAINTAINERS: Add Philippe, Alex and Wainer to the Gitlab-CI section
linux-user: limit check to HOST_LONG_BITS < TARGET_ABI_BITS
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Newer clangs rightly spot that you can never exceed the full address
space of 64 bit hosts with:
linux-user/elfload.c:2076:41: error: result of comparison 'unsigned
long' > 18446744073709551615 is always false
[-Werror,-Wtautological-type-limit-compare]
4685 if ((guest_hiaddr - guest_base) > ~(uintptr_t)0) {
4686 ~~~~~~~~~~~~~~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~
4687 1 error generated.
So lets limit the check to 32 bit hosts only.
Fixes: ee94743034
Reported-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20200525131823.715-8-thuth@redhat.com>
[thuth: Use HOST_LONG_BITS < TARGET_ABI_BITS instead of HOST_LONG_BITS == 32]
Signed-off-by: Thomas Huth <thuth@redhat.com>
When we try to bootstrap debian/lenny for alpha, it fails because
it cannot umount /.root directory:
...
Setting up initscripts (2.86.ds1-61) ...
umount: /.root: Function not implemented
dpkg: error processing initscripts (--configure):
subprocess post-installation script returned error exit status 1
dpkg: sysvinit: dependency problems, but configuring anyway as you request:
sysvinit depends on initscripts; however:
Package initscripts is not configured yet.
This is because, when we switched from syscall_nr.h to syscall.tbl,
the syscall #321 has been renamed from umount to oldumount and
syscall.c has not been updated to manage the new name.
oldumount has been introduced in linux 2.1.116pre1 by:
7d32756b2 ("Import 2.1.116pre1")
...
* We now support a flag for forced unmount like the other 'big iron'
* unixes. Our API is identical to OSF/1 to avoid making a mess of AMD
...
Fixes: 6116aea994 ("linux-user, alpha: add syscall table generation support")
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20200502194642.32823-1-laurent@vivier.eu>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
We shouldn't be messing around with the CPU list in linux-user save
for the very special case of do_fork(). When threads end we need to
properly follow QOM object lifetime handling and allow the eventual
cpu_common_unrealizefn to both remove the CPU and ensure any clean-up
actions are taken place, for example calling plugin exit hooks.
There is still a race condition to avoid so use the linux-user
specific clone_lock instead of the cpu_list_lock to avoid it.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Cc: Nikolay Igotti <igotti@gmail.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Daniel P. Berrange <berrange@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Markus Armbruster <armbru@redhat.com>
Message-Id: <20200520140541.30256-14-alex.bennee@linaro.org>
POWER9 adds scv and rfscv instructions and the system call vectored
interrupt. Linux does not support this instruction yet but it has
been tested with a modified kernel that runs on real hardware.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Message-Id: <20200507115328.789175-1-npiggin@gmail.com>
[dwg: Corrected an overlong line]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The Arm signal-handling code has some parts ifdeffed with a
TARGET_CONFIG_CPU_32, which is always defined. This is a leftover
from when this code's structure was based on the Linux kernel
signal handling code, where it was intended to support 26-bit
Arm CPUs. The kernel dropped its CONFIG_CPU_32 in kernel commit
4da8b8208eded0ba21e3 in 2009.
QEMU has never had 26-bit CPU support and is unlikely to ever
add it; we certainly aren't going to support 26-bit Linux
binaries via linux-user mode. The ifdef is just unhelpful
noise, so remove it entirely.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200518143014.20689-1-peter.maydell@linaro.org
This fixes signal handlers running with the wrong endianness if the
interrupted code used SETEND to dynamically switch endianness.
Signed-off-by: Amanieu d'Antras <amanieu@gmail.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20200511131117.2486486-1-amanieu@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Using the MSR instruction to write to CPSR.E is deprecated, but it is
required to work from any mode including unprivileged code. We were
incorrectly forbidding usermode code from writing it because
CPSR_USER did not include the CPSR_E bit.
We use CPSR_USER in only three places:
* as the mask of what to allow userspace MSR to write to CPSR
* when deciding what bits a linux-user signal-return should be
able to write from the sigcontext structure
* in target_user_copy_regs() when we set up the initial
registers for the linux-user process
In the first two cases not being able to update CPSR.E is a bug, and
in the third case it doesn't matter because CPSR.E is always 0 there.
So we can fix both bugs by adding CPSR_E to CPSR_USER.
Because the cpsr_write() in restore_sigcontext() is now changing
a CPSR bit which is cached in hflags, we need to add an
arm_rebuild_hflags() call there; the callsite in
target_user_copy_regs() was already rebuilding hflags for other
reasons.
(The recommended way to change CPSR.E is to use the 'SETEND'
instruction, which we do correctly allow from usermode code.)
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20200518142801.20503-1-peter.maydell@linaro.org
Our code to identify syscall numbers has some issues:
* for Thumb mode, we never need the immediate value from the insn,
but we always read it anyway
* bad immediate values in the svc insn should cause a SIGILL, but we
were abort()ing instead (via "goto error")
We can fix both these things by refactoring the code that identifies
the syscall number to more closely follow the kernel COMPAT_OABI code:
* for Thumb it is always r7
* for Arm, if the immediate value is 0, then this is an EABI call
with the syscall number in r7
* otherwise, we XOR the immediate value with 0x900000
(ARM_SYSCALL_BASE for QEMU; __NR_OABI_SYSCALL_BASE in the kernel),
which converts valid syscall immediates into the desired value,
and puts all invalid immediates in the range 0x100000 or above
* then we can just let the existing "value too large, deliver
SIGILL" case handle invalid numbers, and drop the 'goto error'
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Message-id: 20200420212206.12776-5-peter.maydell@linaro.org
The kernel has different handling for syscalls with invalid
numbers that are in the "arm-specific" range 0x9f0000 and up:
* 0x9f0000..0x9f07ff return -ENOSYS if not implemented
* other out of range syscalls cause a SIGILL
(see the kernel's arch/arm/kernel/traps.c:arm_syscall())
Implement this distinction. (Note that our code doesn't look
quite like the kernel's, because we have removed the
0x900000 prefix by this point, whereas the kernel retains
it in arm_syscall().)
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-id: 20200420212206.12776-4-peter.maydell@linaro.org
We incorrectly treat SVC 0xf0002 as a cacheflush request (which is a
NOP for QEMU). This is the wrong syscall number, because in the
svc-immediate OABI syscall numbers are all offset by the
ARM_SYSCALL_BASE value and so the correct insn is SVC 0x9f0002.
(This is handled further down in the code with the other Arm-specific
syscalls like NR_breakpoint.)
When this code was initially added in commit 6f1f31c069 in
2004, ARM_NR_cacheflush was defined as (ARM_SYSCALL_BASE + 0xf0000 + 2)
so the value in the comparison took account of the extra 0x900000
offset. In commit fbb4a2e371 in 2008, the ARM_SYSCALL_BASE
was removed from the definition of ARM_NR_cacheflush and handling
for this group of syscalls was added below the point where we subtract
ARM_SYSCALL_BASE from the SVC immediate value. However that commit
forgot to remove the now-obsolete earlier handling code.
Remove the spurious ARM_NR_cacheflush condition.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Message-id: 20200420212206.12776-3-peter.maydell@linaro.org
In linux-user/arm/cpu-loop.c we incorrectly treat EXCP_BKPT similarly
to EXCP_SWI, which means that if the guest executes a BKPT insn then
QEMU will perform a syscall for it (which syscall depends on what
value happens to be in r7...). The correct behaviour is that the
guest process should take a SIGTRAP.
This code has been like this (more or less) since commit
06c949e62a in 2006 which added BKPT in the first place. This is
probably because at the time the same code path was used to handle
both Linux syscalls and semihosting calls, and (on M profile) BKPT
with a suitable magic number is used for semihosting calls. But
these days we've moved handling of semihosting out to an entirely
different codepath, so we can fix this bug by simply removing this
handling of EXCP_BKPT and instead making it deliver a SIGTRAP like
EXCP_DEBUG (as we do already on aarch64).
Reported-by: <omerg681@gmail.com>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20200420212206.12776-2-peter.maydell@linaro.org
Fixes: https://bugs.launchpad.net/qemu/+bug/1873898
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
