| Age | Commit message (Collapse) | Author |
|---|
|
All callers of force_sig_info pass info.si_signo in for the signal
by definition as well as in practice.
Further all callers of force_sig_info except force_sig_fault_to_task
pass current as the target task to force_sig_info.
Factor out a static force_sig_info_to_task that
force_sig_fault_to_task can call.
This prepares the way for force_sig_info to have it's task and signal
parameters removed.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
In preparation for removing the special case in force_sig_info for
only having a signal number generate an appropriate siginfo in
force_sig the last caller of force_sig_info that does not
pass a filled out siginfo.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
Forcing a signal or not allowing a pid namespace init to ignore
SIGKILL or SIGSTOP is more cleanly computed in send_signal.
There are two cases where we don't allow a pid namespace init
to ignore SIGKILL or SIGSTOP. If the sending process is
from an ancestor pid namespace and as such is effectively
the god to the target process, and if the it is the kernel
that is sending the signal, not another application.
It is known that a process is from an ancestor pid namespace if
it can see it's target but it's target does not have a pid for
the sender in it's pid namespace.
It is know that a signal is sent from the kernel if si_code is set to
SI_KERNEL or info is SEND_SIG_PRIV (which ultimately generates
a signal with si_code == SI_KERNEL).
The only signals that matter are SIGKILL and SIGSTOP neither of
which can really be caught, and both of which always have a siginfo
layout that includes si_uid and si_pid. Therefore we never need
to worry about forcing a signal when si_pid and si_uid are absent.
So handle the two special cases of info and the case when si_pid and
si_uid are present.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
Any time siginfo is not stored in the signal queue information is
lost. Therefore set TRACE_SIGNAL_LOSE_INFO every time the code does
not allocate a signal queue entry, and a queue overflow abort is not
triggered.
Fixes: ba005e1f4172 ("tracepoint: Add signal loss events")
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
As synchronous exceptions really only make sense against the current
task (otherwise how are you synchronous) remove the task parameter
from from force_sig_fault to make it explicit that is what is going
on.
The two known exceptions that deliver a synchronous exception to a
stopped ptraced task have already been changed to
force_sig_fault_to_task.
The callers have been changed with the following emacs regular expression
(with obvious variations on the architectures that take more arguments)
to avoid typos:
force_sig_fault[(]\([^,]+\)[,]\([^,]+\)[,]\([^,]+\)[,]\W+current[)]
->
force_sig_fault(\1,\2,\3)
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
current
In preparation for removing the task parameter from force_sig_fault
introduce force_sig_fault_to_task and use it for the two cases where
it matters.
On mips force_fcr31_sig calls force_sig_fault and is called on either
the current task, or a task that is suspended and is being switched to
by the scheduler. This is safe because the task being switched to by
the scheduler is guaranteed to be suspended. This ensures that
task->sighand is stable while the signal is delivered to it.
On parisc user_enable_single_step calls force_sig_fault and is in turn
called by ptrace_request. The function ptrace_request always calls
user_enable_single_step on a child that is stopped for tracing. The
child being traced and not reaped ensures that child->sighand is not
NULL, and that the child will not change child->sighand.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
Update the calls of force_sig_fault that pass in a variable that is
set to current earlier to explicitly use current.
This is to make the next change that removes the task parameter
from force_sig_fault easier to verify.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
The __do_user_fault function is always called with tsk == current.
Make that obvious by removing the tsk parameter.
This makes it clear that __do_user_fault calls force_sig_fault
on the current task.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
The __do_user_fault function is always called with tsk == current.
Make that obvious by removing the tsk parameter.
This makes it clear that __do_user_fault calls force_sig_fault
on the current task.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
The ptrace_break function is always called with tsk == current.
Make that obvious by removing the tsk parameter.
This also makes it clear that ptrace_break calls force_sig_fault
on the current task.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
The send_sigtrap function is always called with tsk == current.
Make that obvious by removing the tsk parameter.
This also makes it clear that send_sigtrap always calls
force_sig_fault on the current task.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
The do_trap function is always called with tsk == current.
Make that obvious by removing the tsk parameter.
This also makes it clear that do_trap calls force_sig_fault
on the current task.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
The force_sig_info_fault function is always called with tsk == current.
Make that explicit by removing the tsk parameter.
This also makes it clear that the sh force_sig_info_fault passes
current into force_sig_fault.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
The send_sigtrap function is always called with task == current. Make
that explicit by removing the task parameter.
This also makes it clear that the uml send_sigtrap passes current
into force_sig_fault.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
The send_sigtrap function is always called with task == current. Make
that explicit by removing the task parameter.
This also makes it clear that the x86 send_sigtrap passes current
into force_sig_fault.
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
All of the callers pass current into force_sig_mceer so remove the
task parameter to make this obvious.
This also makes it clear that force_sig_mceerr passes current
into force_sig_info.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
All of the remaining callers pass current into force_sig so
remove the task parameter to make this obvious and to make
misuse more difficult in the future.
This also makes it clear force_sig passes current into force_sig_info.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
The function force_sigsegv is always called on the current task
so passing in current is redundant and not passing in current
makes this fact obvious.
This also makes it clear force_sigsegv always calls force_sig
on the current task.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
The locking in force_sig_info is not prepared to deal with a task that
exits or execs (as sighand may change). The is not a locking problem
in force_sig as force_sig is only built to handle synchronous
exceptions.
Further the function force_sig_info changes the signal state if the
signal is ignored, or blocked or if SIGNAL_UNKILLABLE will prevent the
delivery of the signal. The signal SIGKILL can not be ignored and can
not be blocked and SIGNAL_UNKILLABLE won't prevent it from being
delivered.
So using force_sig rather than send_sig for SIGKILL is confusing
and pointless.
Because it won't impact the sending of the signal and and because
using force_sig is wrong, replace force_sig with send_sig.
Cc: Namjae Jeon <namjae.jeon@samsung.com>
Cc: Jeff Layton <jlayton@primarydata.com>
Cc: Steve French <smfrench@gmail.com>
Fixes: a5c3e1c725af ("Revert "cifs: No need to send SIGKILL to demux_thread during umount"")
Fixes: e7ddee9037e7 ("cifs: disable sharing session and tcon and add new TCP sharing code")
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
The locking in force_sig_info is not prepared to deal with a task that
exits or execs (as sighand may change). The is not a locking problem
in force_sig as force_sig is only built to handle synchronous
exceptions.
Further the function force_sig_info changes the signal state if the
signal is ignored, or blocked or if SIGNAL_UNKILLABLE will prevent the
delivery of the signal. The signal SIGKILL can not be ignored and can
not be blocked and SIGNAL_UNKILLABLE won't prevent it from being
delivered.
So using force_sig rather than send_sig for SIGKILL is confusing
and pointless.
Because it won't impact the sending of the signal and and because
using force_sig is wrong, replace force_sig with send_sig.
Cc: Daniel Lezcano <daniel.lezcano@free.fr>
Cc: Serge Hallyn <serge@hallyn.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Fixes: cf3f89214ef6 ("pidns: add reboot_pid_ns() to handle the reboot syscall")
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
The locking in force_sig_info is not prepared to deal with
a task that exits or execs (as sighand may change). As force_sig
is only built to handle synchronous exceptions.
Further the function force_sig_info changes the signal state if the
signal is ignored, or blocked or if SIGNAL_UNKILLABLE will prevent the
delivery of the signal. The signal SIGKILL can not be ignored and can
not be blocked and SIGNAL_UNKILLABLE won't prevent it from being
delivered.
So using force_sig rather than send_sig for SIGKILL is pointless.
Because it won't impact the sending of the signal and and because
using force_sig is wrong, replace force_sig with send_sig.
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: David S. Miller <davem@davemloft.net>
Fixes: d2ba09c17a06 ("net: add skeleton of bpfilter kernel module")
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
The drbd module exclusively sends signals to kernel threads it creates with
kthread_create. These kernel threads do not block or ignore signals (only
flush signals after they have been delivered), nor can drbd threads
possibly be pid namespace init processes so the extra work that force_sig
performs that send_sig does not is unnecessary.
Further force_sig is for delivering synchronous signals (aka exceptions).
The locking in force_sig is not prepared to deal with running processes, as
tsk->sighand may change during exec for a running process.
In short it is not only unnecessary for drbd to use force_sig it is
semantically wrong.
With drbd using send_sig it becomes easier to maintain force_sig as only
synchronous signals need to be considered.
Cc: Philipp Reisner <philipp.reisner@linbit.com>
Cc: Lars Ellenberg <lars.ellenberg@linbit.com>
Cc: drbd-dev@lists.linbit.com
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
I don't think this is userspace visible but SIGKILL does not have
any si_codes that use the fault member of the siginfo union. Correct
this the simple way and call force_sig instead of force_sig_fault when
the signal is SIGKILL.
The two know places where synchronous SIGKILL are generated are
do_bad_area and fpsimd_save. The call paths to force_sig_fault are:
do_bad_area
arm64_force_sig_fault
force_sig_fault
force_signal_inject
arm64_notify_die
arm64_force_sig_fault
force_sig_fault
Which means correcting this in arm64_force_sig_fault is enough
to ensure the arm64 code is not misusing the generic code, which
could lead to maintenance problems later.
Cc: stable@vger.kernel.org
Cc: Dave Martin <Dave.Martin@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
Acked-by: Will Deacon <will.deacon@arm.com>
Fixes: af40ff687bc9 ("arm64: signal: Ensure si_code is valid for all fault signals")
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
The function send_signal was split from __send_signal so that it would
be possible to bypass the namespace logic based upon current[1]. As it
turns out the si_pid and the si_uid fixup are both inappropriate in
the case of kill_pid_usb_asyncio so move that logic into send_signal.
It is difficult to arrange but possible for a signal with an si_code
of SI_TIMER or SI_SIGIO to be sent across namespace boundaries. In
which case tests for when it is ok to change si_pid and si_uid based
on SI_FROMUSER are incorrect. Replace the use of SI_FROMUSER with a
new test has_si_pid_and_used based on siginfo_layout.
Now that the uid fixup is no longer present after expanding
SEND_SIG_NOINFO properly calculate the si_uid that the target
task needs to read.
[1] 7978b567d315 ("signals: add from_ancestor_ns parameter to send_signal()")
Cc: stable@vger.kernel.org
Fixes: 6588c1e3ff01 ("signals: SI_USER: Masquerade si_pid when crossing pid ns boundary")
Fixes: 6b550f949594 ("user namespace: make signal.c respect user namespaces")
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
The usb support for asyncio encoded one of it's values in the wrong
field. It should have used si_value but instead used si_addr which is
not present in the _rt union member of struct siginfo.
The practical result of this is that on a 64bit big endian kernel
when delivering a signal to a 32bit process the si_addr field
is set to NULL, instead of the expected pointer value.
This issue can not be fixed in copy_siginfo_to_user32 as the usb
usage of the the _sigfault (aka si_addr) member of the siginfo
union when SI_ASYNCIO is set is incompatible with the POSIX and
glibc usage of the _rt member of the siginfo union.
Therefore replace kill_pid_info_as_cred with kill_pid_usb_asyncio a
dedicated function for this one specific case. There are no other
users of kill_pid_info_as_cred so this specialization should have no
impact on the amount of code in the kernel. Have kill_pid_usb_asyncio
take instead of a siginfo_t which is difficult and error prone, 3
arguments, a signal number, an errno value, and an address enconded as
a sigval_t. The encoding of the address as a sigval_t allows the
code that reads the userspace request for a signal to handle this
compat issue along with all of the other compat issues.
Add BUILD_BUG_ONs in kernel/signal.c to ensure that we can now place
the pointer value at the in si_pid (instead of si_addr). That is the
code now verifies that si_pid and si_addr always occur at the same
location. Further the code veries that for native structures a value
placed in si_pid and spilling into si_uid will appear in userspace in
si_addr (on a byte by byte copy of siginfo or a field by field copy of
siginfo). The code also verifies that for a 64bit kernel and a 32bit
userspace the 32bit pointer will fit in si_pid.
I have used the usbsig.c program below written by Alan Stern and
slightly tweaked by me to run on a big endian machine to verify the
issue exists (on sparc64) and to confirm the patch below fixes the issue.
/* usbsig.c -- test USB async signal delivery */
#define _GNU_SOURCE
#include <stdio.h>
#include <fcntl.h>
#include <signal.h>
#include <string.h>
#include <sys/ioctl.h>
#include <unistd.h>
#include <endian.h>
#include <linux/usb/ch9.h>
#include <linux/usbdevice_fs.h>
static struct usbdevfs_urb urb;
static struct usbdevfs_disconnectsignal ds;
static volatile sig_atomic_t done = 0;
void urb_handler(int sig, siginfo_t *info , void *ucontext)
{
printf("Got signal %d, signo %d errno %d code %d addr: %p urb: %p\n",
sig, info->si_signo, info->si_errno, info->si_code,
info->si_addr, &urb);
printf("%s\n", (info->si_addr == &urb) ? "Good" : "Bad");
}
void ds_handler(int sig, siginfo_t *info , void *ucontext)
{
printf("Got signal %d, signo %d errno %d code %d addr: %p ds: %p\n",
sig, info->si_signo, info->si_errno, info->si_code,
info->si_addr, &ds);
printf("%s\n", (info->si_addr == &ds) ? "Good" : "Bad");
done = 1;
}
int main(int argc, char **argv)
{
char *devfilename;
int fd;
int rc;
struct sigaction act;
struct usb_ctrlrequest *req;
void *ptr;
char buf[80];
if (argc != 2) {
fprintf(stderr, "Usage: usbsig device-file-name\n");
return 1;
}
devfilename = argv[1];
fd = open(devfilename, O_RDWR);
if (fd == -1) {
perror("Error opening device file");
return 1;
}
act.sa_sigaction = urb_handler;
sigemptyset(&act.sa_mask);
act.sa_flags = SA_SIGINFO;
rc = sigaction(SIGUSR1, &act, NULL);
if (rc == -1) {
perror("Error in sigaction");
return 1;
}
act.sa_sigaction = ds_handler;
sigemptyset(&act.sa_mask);
act.sa_flags = SA_SIGINFO;
rc = sigaction(SIGUSR2, &act, NULL);
if (rc == -1) {
perror("Error in sigaction");
return 1;
}
memset(&urb, 0, sizeof(urb));
urb.type = USBDEVFS_URB_TYPE_CONTROL;
urb.endpoint = USB_DIR_IN | 0;
urb.buffer = buf;
urb.buffer_length = sizeof(buf);
urb.signr = SIGUSR1;
req = (struct usb_ctrlrequest *) buf;
req->bRequestType = USB_DIR_IN | USB_TYPE_STANDARD | USB_RECIP_DEVICE;
req->bRequest = USB_REQ_GET_DESCRIPTOR;
req->wValue = htole16(USB_DT_DEVICE << 8);
req->wIndex = htole16(0);
req->wLength = htole16(sizeof(buf) - sizeof(*req));
rc = ioctl(fd, USBDEVFS_SUBMITURB, &urb);
if (rc == -1) {
perror("Error in SUBMITURB ioctl");
return 1;
}
rc = ioctl(fd, USBDEVFS_REAPURB, &ptr);
if (rc == -1) {
perror("Error in REAPURB ioctl");
return 1;
}
memset(&ds, 0, sizeof(ds));
ds.signr = SIGUSR2;
ds.context = &ds;
rc = ioctl(fd, USBDEVFS_DISCSIGNAL, &ds);
if (rc == -1) {
perror("Error in DISCSIGNAL ioctl");
return 1;
}
printf("Waiting for usb disconnect\n");
while (!done) {
sleep(1);
}
close(fd);
return 0;
}
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: linux-usb@vger.kernel.org
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Oliver Neukum <oneukum@suse.com>
Fixes: v2.3.39
Cc: stable@vger.kernel.org
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
|
|
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/rw/ubifs
Pull UBIFS fixes from Richard Weinberger:
- build errors wrt xattrs
- mismerge which lead to a wrong Kconfig ifdef
- missing endianness conversion
* tag 'upstream-5.2-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rw/ubifs:
ubifs: Convert xattr inum to host order
ubifs: Use correct config name for encryption
ubifs: Fix build error without CONFIG_UBIFS_FS_XATTR
|
|
Merge yet more updates from Andrew Morton:
"A few final bits:
- large changes to vmalloc, yielding large performance benefits
- tweak the console-flush-on-panic code
- a few fixes"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
panic: add an option to replay all the printk message in buffer
initramfs: don't free a non-existent initrd
fs/writeback.c: use rcu_barrier() to wait for inflight wb switches going into workqueue when umount
mm/compaction.c: correct zone boundary handling when isolating pages from a pageblock
mm/vmap: add DEBUG_AUGMENT_LOWEST_MATCH_CHECK macro
mm/vmap: add DEBUG_AUGMENT_PROPAGATE_CHECK macro
mm/vmalloc.c: keep track of free blocks for vmap allocation
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild
Pull more Kbuild updates from Masahiro Yamada:
- remove unneeded use of cc-option, cc-disable-warning, cc-ldoption
- exclude tracked files from .gitignore
- re-enable -Wint-in-bool-context warning
- refactor samples/Makefile
- stop building immediately if syncconfig fails
- do not sprinkle error messages when $(CC) does not exist
- move arch/alpha/defconfig to the configs subdirectory
- remove crappy header search path manipulation
- add comment lines to .config to clarify the end of menu blocks
- check uniqueness of module names (adding new warnings intentionally)
* tag 'kbuild-v5.2-2' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild: (24 commits)
kconfig: use 'else ifneq' for Makefile to improve readability
kbuild: check uniqueness of module names
kconfig: Terminate menu blocks with a comment in the generated config
kbuild: add LICENSES to KBUILD_ALLDIRS
kbuild: remove 'addtree' and 'flags' magic for header search paths
treewide: prefix header search paths with $(srctree)/
media: prefix header search paths with $(srctree)/
media: remove unneeded header search paths
alpha: move arch/alpha/defconfig to arch/alpha/configs/defconfig
kbuild: terminate Kconfig when $(CC) or $(LD) is missing
kbuild: turn auto.conf.cmd into a mandatory include file
.gitignore: exclude .get_maintainer.ignore and .gitattributes
kbuild: add all Clang-specific flags unconditionally
kbuild: Don't try to add '-fcatch-undefined-behavior' flag
kbuild: add some extra warning flags unconditionally
kbuild: add -Wvla flag unconditionally
arch: remove dangling asm-generic wrappers
samples: guard sub-directories with CONFIG options
kbuild: re-enable int-in-bool-context warning
MAINTAINERS: kbuild: Add pattern for scripts/*vmlinux*
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux
Pull i2c updates from Wolfram Sang:
"Some I2C core API additions which are kind of simple but enhance error
checking for users a lot, especially by returning errno now.
There are wrappers to still support the old API but it will be removed
once all users are converted"
* 'i2c/for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
i2c: core: add device-managed version of i2c_new_dummy
i2c: core: improve return value handling of i2c_new_device and i2c_new_dummy
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4
Pull ext4 fixes from Ted Ts'o:
"Some bug fixes, and an update to the URL's for the final version of
Unicode 12.1.0"
* tag 'ext4_for_linus_stable' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
ext4: avoid panic during forced reboot due to aborted journal
ext4: fix block validity checks for journal inodes using indirect blocks
unicode: update to Unicode 12.1.0 final
unicode: add missing check for an error return from utf8lookup()
ext4: fix miscellaneous sparse warnings
ext4: unsigned int compared against zero
ext4: fix use-after-free in dx_release()
ext4: fix data corruption caused by overlapping unaligned and aligned IO
jbd2: fix potential double free
ext4: zero out the unused memory region in the extent tree block
|
|
Pull cifs fixes from Steve French:
"Minor cleanup and fixes, one for stable, four rdma (smbdirect)
related. Also adds SEEK_HOLE support"
* tag '5.2-rc-smb3-fixes' of git://git.samba.org/sfrench/cifs-2.6:
cifs: add support for SEEK_DATA and SEEK_HOLE
Fixed https://bugzilla.kernel.org/show_bug.cgi?id=202935 allow write on the same file
cifs: Allocate memory for all iovs in smb2_ioctl
cifs: Don't match port on SMBDirect transport
cifs:smbd Use the correct DMA direction when sending data
cifs:smbd When reconnecting to server, call smbd_destroy() after all MIDs have been called
cifs: use the right include for signal_pending()
smb3: trivial cleanup to smb2ops.c
cifs: cleanup smb2ops.c and normalize strings
smb3: display session id in debug data
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf tooling updates from Ingo Molnar:
"perf.data:
- Streaming compression of perf ring buffer into
PERF_RECORD_COMPRESSED user space records, resulting in ~3-5x
perf.data file size reduction on variety of tested workloads what
saves storage space on larger server systems where perf.data size
can easily reach several tens or even hundreds of GiBs, especially
when profiling with DWARF-based stacks and tracing of context
switches.
perf record:
- Improve -user-regs/intr-regs suggestions to overcome errors
perf annotate:
- Remove hist__account_cycles() from callback, speeding up branch
processing (perf record -b)
perf stat:
- Add a 'percore' event qualifier, e.g.: -e
cpu/event=0,umask=0x3,percore=1/, that sums up the event counts for
both hardware threads in a core.
We can already do this with --per-core, but it's often useful to do
this together with other metrics that are collected per hardware
thread.
I.e. now its possible to do this per-event, and have it mixed with
other events not aggregated by core.
arm64:
- Map Brahma-B53 CPUID to cortex-a53 events.
- Add Cortex-A57 and Cortex-A72 events.
csky:
- Add DWARF register mappings for libdw, allowing --call-graph=dwarf
to work on the C-SKY arch.
x86:
- Add support for recording and printing XMM registers, available,
for instance, on Icelake.
- Add uncore_upi (Intel's "Ultra Path Interconnect" events) JSON
support. UPI replaced the Intel QuickPath Interconnect (QPI) in
Xeon Skylake-SP.
Intel PT:
- Fix instructions sampling rate.
- Timestamp fixes.
- Improve exported-sql-viewer GUI, allowing, for instance, to
copy'n'paste the trees, useful for e-mailing"
* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (73 commits)
perf stat: Support 'percore' event qualifier
perf stat: Factor out aggregate counts printing
perf tools: Add a 'percore' event qualifier
perf docs: Add description for stderr
perf intel-pt: Fix sample timestamp wrt non-taken branches
perf intel-pt: Fix improved sample timestamp
perf intel-pt: Fix instructions sampling rate
perf regs x86: Add X86 specific arch__intr_reg_mask()
perf parse-regs: Add generic support for arch__intr/user_reg_mask()
perf parse-regs: Split parse_regs
perf vendor events arm64: Add Cortex-A57 and Cortex-A72 events
perf vendor events arm64: Map Brahma-B53 CPUID to cortex-a53 events
perf vendor events arm64: Remove [[:xdigit:]] wildcard
perf jevents: Remove unused variable
perf test zstd: Fixup verbose mode output
perf tests: Implement Zstd comp/decomp integration test
perf inject: Enable COMPRESSED record decompression
perf report: Implement perf.data record decompression
perf record: Implement -z,--compression_level[=<n>] option
perf report: Add stub processing of compressed events for -D
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull clocksource updates from Ingo Molnar:
"Misc clocksource/clockevent driver updates that came in a bit late but
are ready for v5.2"
* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
misc: atmel_tclib: Do not probe already used TCBs
clocksource/drivers/timer-atmel-tcb: Convert tc_clksrc_suspend|resume() to static
clocksource/drivers/tcb_clksrc: Rename the file for consistency
clocksource/drivers/timer-atmel-pit: Rework Kconfig option
clocksource/drivers/tcb_clksrc: Move Kconfig option
ARM: at91: Implement clocksource selection
clocksource/drivers/tcb_clksrc: Use tcb as sched_clock
clocksource/drivers/tcb_clksrc: Stop depending on atmel_tclib
ARM: at91: move SoC specific definitions to SoC folder
clocksource/drivers/timer-milbeaut: Cleanup common register accesses
clocksource/drivers/timer-milbeaut: Add shutdown function
clocksource/drivers/timer-milbeaut: Fix to enable one-shot timer
clocksource/drivers/tegra: Rework for compensation of suspend time
clocksource/drivers/sp804: Add COMPILE_TEST to CONFIG_ARM_TIMER_SP804
clocksource/drivers/sun4i: Add a compatible for suniv
dt-bindings: timer: Add Allwinner suniv timer
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull IRQ chip updates from Ingo Molnar:
"A late irqchips update:
- New TI INTR/INTA set of drivers
- Rewrite of the stm32mp1-exti driver as a platform driver
- Update the IOMMU MSI mapping API to be RT friendly
- A number of cleanups and other low impact fixes"
* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (34 commits)
iommu/dma-iommu: Remove iommu_dma_map_msi_msg()
irqchip/gic-v3-mbi: Don't map the MSI page in mbi_compose_m{b, s}i_msg()
irqchip/ls-scfg-msi: Don't map the MSI page in ls_scfg_msi_compose_msg()
irqchip/gic-v3-its: Don't map the MSI page in its_irq_compose_msi_msg()
irqchip/gicv2m: Don't map the MSI page in gicv2m_compose_msi_msg()
iommu/dma-iommu: Split iommu_dma_map_msi_msg() in two parts
genirq/msi: Add a new field in msi_desc to store an IOMMU cookie
arm64: arch_k3: Enable interrupt controller drivers
irqchip/ti-sci-inta: Add msi domain support
soc: ti: Add MSI domain bus support for Interrupt Aggregator
irqchip/ti-sci-inta: Add support for Interrupt Aggregator driver
dt-bindings: irqchip: Introduce TISCI Interrupt Aggregator bindings
irqchip/ti-sci-intr: Add support for Interrupt Router driver
dt-bindings: irqchip: Introduce TISCI Interrupt router bindings
gpio: thunderx: Use the default parent apis for {request,release}_resources
genirq: Introduce irq_chip_{request,release}_resource_parent() apis
firmware: ti_sci: Add helper apis to manage resources
firmware: ti_sci: Add RM mapping table for am654
firmware: ti_sci: Add support for IRQ management
firmware: ti_sci: Add support for RM core ops
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull EFI fix from Ingo Molnar:
"Fix an EFI-fb regression that affects certain x86 systems"
* 'efi-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
fbdev/efifb: Ignore framebuffer memmap entries that lack any memory types
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull core fixes from Ingo Molnar:
"This fixes a particularly thorny munmap() bug with MPX, plus fixes a
host build environment assumption in objtool"
* 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
objtool: Allow AR to be overridden with HOSTAR
x86/mpx, mm/core: Fix recursive munmap() corruption
|
|
Pull ARM SoC late updates from Olof Johansson:
"This is some material that we picked up into our tree late. Most of it
are smaller fixes and additions, some defconfig updates due to recent
development, etc.
Code-wise the largest portion is a series of PM updates for the at91
platform, and those have been in linux-next a while through the at91
tree before we picked them up"
* tag 'armsoc-late' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc: (29 commits)
arm64: dts: sprd: Add clock properties for serial devices
Opt out of scripts/get_maintainer.pl
ARM: ixp4xx: Remove duplicated include from common.c
soc: ixp4xx: qmgr: Fix an NULL vs IS_ERR() check in probe
arm64: tegra: Disable XUSB support on Jetson TX2
arm64: tegra: Enable SMMU translation for PCI on Tegra186
arm64: tegra: Fix insecure SMMU users for Tegra186
arm64: tegra: Select ARM_GIC_PM
amba: tegra-ahb: Mark PM functions as __maybe_unused
ARM: dts: logicpd-som-lv: Fix MMC1 card detect
ARM: mvebu: drop return from void function
ARM: mvebu: prefix coprocessor operand with p
ARM: mvebu: drop unnecessary label
ARM: mvebu: fix a leaked reference by adding missing of_node_put
ARM: socfpga_defconfig: enable LTC2497
ARM: mvebu: kirkwood: remove error message when retrieving mac address
ARM: at91: sama5: make ov2640 as a module
ARM: OMAP1: ams-delta: fix early boot crash when LED support is disabled
ARM: at91: remove HAVE_FB_ATMEL for sama5 SoC as they use DRM
soc/fsl/qe: Fix an error code in qe_pin_request()
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
Pull powerpc fixes from Michael Ellerman:
"One fix going back to stable, for a bug on 32-bit introduced when we
added support for THREAD_INFO_IN_TASK.
A fix for a typo in a recent rework of our hugetlb code that leads to
crashes on 64-bit when using hugetlbfs with a 4K PAGE_SIZE.
Two fixes for our recent rework of the address layout on 64-bit hash
CPUs, both only triggered when userspace tries to access addresses
outside the user or kernel address ranges.
Finally a fix for a recently introduced double free in an error path
in our cacheinfo code.
Thanks to: Aneesh Kumar K.V, Christophe Leroy, Sachin Sant, Tobin C.
Harding"
* tag 'powerpc-5.2-2' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
powerpc/cacheinfo: Remove double free
powerpc/mm/hash: Fix get_region_id() for invalid addresses
powerpc/mm: Drop VM_BUG_ON in get_region_id()
powerpc/mm: Fix crashes with hugepages & 4K pages
powerpc/32s: fix flush_hash_pages() on SMP
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux
Pull a few more MIPS updates from Paul Burton:
"Some SGI IP27 specific PCI rework and a batch of fixes:
- A build fix for BMIPS5000 configurations with
CONFIG_HW_PERF_EVENTS=y, which also neatly removes some #ifdefery.
- A fix to report supported ISAs correctly on older Ingenic SoCs
which incorrectly indicate MIPSr2 support in their cop0 Config
register.
- Some PCI modernization for SGI IP27 systems as part of ongoing work
to support some other SGI systems.
- A fix allowing use of appended DTB files with generic kernels.
- DMA mask fixes for SGI IP22 & Alchemy systems"
* tag 'mips_5.2_2' of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux:
MIPS: Alchemy: add DMA masks for on-chip ethernet
MIPS: SGI-IP22: provide missing dma_mask/coherent_dma_mask
generic: fix appended dtb support
MIPS: SGI-IP27: abstract chipset irq from bridge
MIPS: SGI-IP27: use generic PCI driver
MIPS: Fix Ingenic SoCs sometimes reporting wrong ISA
MIPS: perf: Fix build with CONFIG_CPU_BMIPS5000 enabled
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/palmer/riscv-linux
Pull RISC-V updates from Palmer Dabbelt:
"This contains an assortment of RISC-V related patches that I'd like to
target for the 5.2 merge window. Most of the patches are cleanups, but
there are a handful of user-visible changes:
- The nosmp and nr_cpus command-line arguments are now supported,
which work like normal.
- The SBI console no longer installs itself as a preferred console,
we rely on standard mechanisms (/chosen, command-line, hueristics)
instead.
- sfence_remove_sfence_vma{,_asid} now pass their arguments along to
the SBI call.
- Modules now support BUG().
- A missing sfence.vma during boot has been added. This bug only
manifests during boot.
- The arch/riscv support for SiFive's L2 cache controller has been
merged, which should un-block the EDAC framework work.
I've only tested this on QEMU again, as I didn't have time to get
things running on the Unleashed. The latest master from this morning
merges in cleanly and passes the tests as well"
* tag 'riscv-for-linus-5.2-mw2' of git://git.kernel.org/pub/scm/linux/kernel/git/palmer/riscv-linux: (31 commits)
riscv: fix locking violation in page fault handler
RISC-V: sifive_l2_cache: Add L2 cache controller driver for SiFive SoCs
RISC-V: Add DT documentation for SiFive L2 Cache Controller
RISC-V: Avoid using invalid intermediate translations
riscv: Support BUG() in kernel module
riscv: Add the support for c.ebreak check in is_valid_bugaddr()
riscv: support trap-based WARN()
riscv: fix sbi_remote_sfence_vma{,_asid}.
riscv: move switch_mm to its own file
riscv: move flush_icache_{all,mm} to cacheflush.c
tty: Don't force RISCV SBI console as preferred console
RISC-V: Access CSRs using CSR numbers
RISC-V: Add interrupt related SCAUSE defines in asm/csr.h
RISC-V: Use tabs to align macro values in asm/csr.h
RISC-V: Fix minor checkpatch issues.
RISC-V: Support nr_cpus command line option.
RISC-V: Implement nosmp commandline option.
RISC-V: Add RISC-V specific arch_match_cpu_phys_id
riscv: vdso: drop unnecessary cc-ldoption
riscv: call pm_power_off from machine_halt / machine_power_off
...
|
|
'ifeq ... else ifneq ... endif' notation is supported by GNU Make 3.81
or later, which is the requirement for building the kernel since
commit 37d69ee30808 ("docs: bump minimal GNU Make version to 3.81").
Use it to improve the readability.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
|
|
Currently on panic, kernel will lower the loglevel and print out pending
printk msg only with console_flush_on_panic().
Add an option for users to configure the "panic_print" to replay all
dmesg in buffer, some of which they may have never seen due to the
loglevel setting, which will help panic debugging .
[feng.tang@intel.com: keep the original console_flush_on_panic() inside panic()]
Link: http://lkml.kernel.org/r/1556199137-14163-1-git-send-email-feng.tang@intel.com
[feng.tang@intel.com: use logbuf lock to protect the console log index]
Link: http://lkml.kernel.org/r/1556269868-22654-1-git-send-email-feng.tang@intel.com
Link: http://lkml.kernel.org/r/1556095872-36838-1-git-send-email-feng.tang@intel.com
Signed-off-by: Feng Tang <feng.tang@intel.com>
Reviewed-by: Petr Mladek <pmladek@suse.com>
Cc: Aaro Koskinen <aaro.koskinen@nokia.com>
Cc: Petr Mladek <pmladek@suse.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Borislav Petkov <bp@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Since commit 54c7a8916a88 ("initramfs: free initrd memory if opening
/initrd.image fails"), the kernel has unconditionally attempted to free
the initrd even if it doesn't exist.
In the non-existent case this causes a boot-time splat if
CONFIG_DEBUG_VIRTUAL is enabled due to a call to virt_to_phys() with a
NULL address.
Instead we should check that the initrd actually exists and only attempt
to free it if it does.
Link: http://lkml.kernel.org/r/20190516143125.48948-1-steven.price@arm.com
Fixes: 54c7a8916a88 ("initramfs: free initrd memory if opening /initrd.image fails")
Signed-off-by: Steven Price <steven.price@arm.com>
Reported-by: Mark Rutland <mark.rutland@arm.com>
Tested-by: Mark Rutland <mark.rutland@arm.com>
Reviewed-by: Mike Rapoport <rppt@linux.ibm.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
into workqueue when umount
synchronize_rcu() didn't wait for call_rcu() callbacks, so inode wb
switch may not go to the workqueue after synchronize_rcu(). Thus
previous scheduled switches was not finished even flushing the
workqueue, which will cause a NULL pointer dereferenced followed below.
VFS: Busy inodes after unmount of vdd. Self-destruct in 5 seconds. Have a nice day...
BUG: unable to handle kernel NULL pointer dereference at 0000000000000278
evict+0xb3/0x180
iput+0x1b0/0x230
inode_switch_wbs_work_fn+0x3c0/0x6a0
worker_thread+0x4e/0x490
? process_one_work+0x410/0x410
kthread+0xe6/0x100
ret_from_fork+0x39/0x50
Replace the synchronize_rcu() call with a rcu_barrier() to wait for all
pending callbacks to finish. And inc isw_nr_in_flight after call_rcu()
in inode_switch_wbs() to make more sense.
Link: http://lkml.kernel.org/r/20190429024108.54150-1-jiufei.xue@linux.alibaba.com
Signed-off-by: Jiufei Xue <jiufei.xue@linux.alibaba.com>
Acked-by: Tejun Heo <tj@kernel.org>
Suggested-by: Tejun Heo <tj@kernel.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
pageblock
syzbot reported the following error from a tree with a head commit of
baf76f0c58ae ("slip: make slhc_free() silently accept an error pointer")
BUG: unable to handle kernel paging request at ffffea0003348000
#PF error: [normal kernel read fault]
PGD 12c3f9067 P4D 12c3f9067 PUD 12c3f8067 PMD 0
Oops: 0000 [#1] PREEMPT SMP KASAN
CPU: 1 PID: 28916 Comm: syz-executor.2 Not tainted 5.1.0-rc6+ #89
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:constant_test_bit arch/x86/include/asm/bitops.h:314 [inline]
RIP: 0010:PageCompound include/linux/page-flags.h:186 [inline]
RIP: 0010:isolate_freepages_block+0x1c0/0xd40 mm/compaction.c:579
Code: 01 d8 ff 4d 85 ed 0f 84 ef 07 00 00 e8 29 00 d8 ff 4c 89 e0 83 85 38 ff
ff ff 01 48 c1 e8 03 42 80 3c 38 00 0f 85 31 0a 00 00 <4d> 8b 2c 24 31 ff 49
c1 ed 10 41 83 e5 01 44 89 ee e8 3a 01 d8 ff
RSP: 0018:ffff88802b31eab8 EFLAGS: 00010246
RAX: 1ffffd4000669000 RBX: 00000000000cd200 RCX: ffffc9000a235000
RDX: 000000000001ca5e RSI: ffffffff81988cc7 RDI: 0000000000000001
RBP: ffff88802b31ebd8 R08: ffff88805af700c0 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000000 R12: ffffea0003348000
R13: 0000000000000000 R14: ffff88802b31f030 R15: dffffc0000000000
FS: 00007f61648dc700(0000) GS:ffff8880ae900000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: ffffea0003348000 CR3: 0000000037c64000 CR4: 00000000001426e0
Call Trace:
fast_isolate_around mm/compaction.c:1243 [inline]
fast_isolate_freepages mm/compaction.c:1418 [inline]
isolate_freepages mm/compaction.c:1438 [inline]
compaction_alloc+0x1aee/0x22e0 mm/compaction.c:1550
There is no reproducer and it is difficult to hit -- 1 crash every few
days. The issue is very similar to the fix in commit 6b0868c820ff
("mm/compaction.c: correct zone boundary handling when resetting pageblock
skip hints"). When isolating free pages around a target pageblock, the
boundary handling is off by one and can stray into the next pageblock.
Triggering the syzbot error requires that the end of pageblock is section
or zone aligned, and that the next section is unpopulated.
A more subtle consequence of the bug is that pageblocks were being
improperly used as migration targets which potentially hurts fragmentation
avoidance in the long-term one page at a time.
A debugging patch revealed that it's definitely possible to stray outside
of a pageblock which is not intended. While syzbot cannot be used to
verify this patch, it was confirmed that the debugging warning no longer
triggers with this patch applied. It has also been confirmed that the THP
allocation stress tests are not degraded by this patch.
Link: http://lkml.kernel.org/r/20190510182124.GI18914@techsingularity.net
Fixes: e332f741a8dd ("mm, compaction: be selective about what pageblocks to clear skip hints")
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Reported-by: syzbot+d84c80f9fe26a0f7a734@syzkaller.appspotmail.com
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Qian Cai <cai@lca.pw>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: <stable@vger.kernel.org> # v5.1+
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
This macro adds some debug code to check that vmap allocations are
happened in ascending order.
By default this option is set to 0 and not active. It requires
recompilation of the kernel to activate it. Set to 1, compile the
kernel.
[urezki@gmail.com: v4]
Link: http://lkml.kernel.org/r/20190406183508.25273-4-urezki@gmail.com
Link: http://lkml.kernel.org/r/20190402162531.10888-4-urezki@gmail.com
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Reviewed-by: Roman Gushchin <guro@fb.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Oleksiy Avramchenko <oleksiy.avramchenko@sonymobile.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Thomas Garnier <thgarnie@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
This macro adds some debug code to check that the augment tree is
maintained correctly, meaning that every node contains valid
subtree_max_size value.
By default this option is set to 0 and not active. It requires
recompilation of the kernel to activate it. Set to 1, compile the
kernel.
[urezki@gmail.com: v4]
Link: http://lkml.kernel.org/r/20190406183508.25273-3-urezki@gmail.com
Link: http://lkml.kernel.org/r/20190402162531.10888-3-urezki@gmail.com
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Reviewed-by: Roman Gushchin <guro@fb.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Oleksiy Avramchenko <oleksiy.avramchenko@sonymobile.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Thomas Garnier <thgarnie@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Patch series "improve vmap allocation", v3.
Objective
---------
Please have a look for the description at:
https://lkml.org/lkml/2018/10/19/786
but let me also summarize it a bit here as well.
The current implementation has O(N) complexity. Requests with different
permissive parameters can lead to long allocation time. When i say
"long" i mean milliseconds.
Description
-----------
This approach organizes the KVA memory layout into free areas of the
1-ULONG_MAX range, i.e. an allocation is done over free areas lookups,
instead of finding a hole between two busy blocks. It allows to have
lower number of objects which represent the free space, therefore to have
less fragmented memory allocator. Because free blocks are always as large
as possible.
It uses the augment tree where all free areas are sorted in ascending
order of va->va_start address in pair with linked list that provides
O(1) access to prev/next elements.
Since the tree is augment, we also maintain the "subtree_max_size" of VA
that reflects a maximum available free block in its left or right
sub-tree. Knowing that, we can easily traversal toward the lowest (left
most path) free area.
Allocation: ~O(log(N)) complexity. It is sequential allocation method
therefore tends to maximize locality. The search is done until a first
suitable block is large enough to encompass the requested parameters.
Bigger areas are split.
I copy paste here the description of how the area is split, since i
described it in https://lkml.org/lkml/2018/10/19/786
<snip>
A free block can be split by three different ways. Their names are
FL_FIT_TYPE, LE_FIT_TYPE/RE_FIT_TYPE and NE_FIT_TYPE, i.e. they
correspond to how requested size and alignment fit to a free block.
FL_FIT_TYPE - in this case a free block is just removed from the free
list/tree because it fully fits. Comparing with current design there is
an extra work with rb-tree updating.
LE_FIT_TYPE/RE_FIT_TYPE - left/right edges fit. In this case what we do
is just cutting a free block. It is as fast as a current design. Most of
the vmalloc allocations just end up with this case, because the edge is
always aligned to 1.
NE_FIT_TYPE - Is much less common case. Basically it happens when
requested size and alignment does not fit left nor right edges, i.e. it
is between them. In this case during splitting we have to build a
remaining left free area and place it back to the free list/tree.
Comparing with current design there are two extra steps. First one is we
have to allocate a new vmap_area structure. Second one we have to insert
that remaining free block to the address sorted list/tree.
In order to optimize a first case there is a cache with free_vmap objects.
Instead of allocating from slab we just take an object from the cache and
reuse it.
Second one is pretty optimized. Since we know a start point in the tree
we do not do a search from the top. Instead a traversal begins from a
rb-tree node we split.
<snip>
De-allocation. ~O(log(N)) complexity. An area is not inserted straight
away to the tree/list, instead we identify the spot first, checking if it
can be merged around neighbors. The list provides O(1) access to
prev/next, so it is pretty fast to check it. Summarizing. If merged then
large coalesced areas are created, if not the area is just linked making
more fragments.
There is one more thing that i should mention here. After modification of
VA node, its subtree_max_size is updated if it was/is the biggest area in
its left or right sub-tree. Apart of that it can also be populated back
to upper levels to fix the tree. For more details please have a look at
the __augment_tree_propagate_from() function and the description.
Tests and stressing
-------------------
I use the "test_vmalloc.sh" test driver available under
"tools/testing/selftests/vm/" since 5.1-rc1 kernel. Just trigger "sudo
./test_vmalloc.sh" to find out how to deal with it.
Tested on different platforms including x86_64/i686/ARM64/x86_64_NUMA.
Regarding last one, i do not have any physical access to NUMA system,
therefore i emulated it. The time of stressing is days.
If you run the test driver in "stress mode", you also need the patch that
is in Andrew's tree but not in Linux 5.1-rc1. So, please apply it:
http://git.cmpxchg.org/cgit.cgi/linux-mmotm.git/commit/?id=e0cf7749bade6da318e98e934a24d8b62fab512c
After massive testing, i have not identified any problems like memory
leaks, crashes or kernel panics. I find it stable, but more testing would
be good.
Performance analysis
--------------------
I have used two systems to test. One is i5-3320M CPU @ 2.60GHz and
another is HiKey960(arm64) board. i5-3320M runs on 4.20 kernel, whereas
Hikey960 uses 4.15 kernel. I have both system which could run on 5.1-rc1
as well, but the results have not been ready by time i an writing this.
Currently it consist of 8 tests. There are three of them which correspond
to different types of splitting(to compare with default). We have 3
ones(see above). Another 5 do allocations in different conditions.
a) sudo ./test_vmalloc.sh performance
When the test driver is run in "performance" mode, it runs all available
tests pinned to first online CPU with sequential execution test order. We
do it in order to get stable and repeatable results. Take a look at time
difference in "long_busy_list_alloc_test". It is not surprising because
the worst case is O(N).
# i5-3320M
How many cycles all tests took:
CPU0=646919905370(default) cycles vs CPU0=193290498550(patched) cycles
# See detailed table with results here:
ftp://vps418301.ovh.net/incoming/vmap_test_results_v2/i5-3320M_performance_default.txt
ftp://vps418301.ovh.net/incoming/vmap_test_results_v2/i5-3320M_performance_patched.txt
# Hikey960 8x CPUs
How many cycles all tests took:
CPU0=3478683207 cycles vs CPU0=463767978 cycles
# See detailed table with results here:
ftp://vps418301.ovh.net/incoming/vmap_test_results_v2/HiKey960_performance_default.txt
ftp://vps418301.ovh.net/incoming/vmap_test_results_v2/HiKey960_performance_patched.txt
b) time sudo ./test_vmalloc.sh test_repeat_count=1
With this configuration, all tests are run on all available online CPUs.
Before running each CPU shuffles its tests execution order. It gives
random allocation behaviour. So it is rough comparison, but it puts in
the picture for sure.
# i5-3320M
<default> vs <patched>
real 101m22.813s real 0m56.805s
user 0m0.011s user 0m0.015s
sys 0m5.076s sys 0m0.023s
# See detailed table with results here:
ftp://vps418301.ovh.net/incoming/vmap_test_results_v2/i5-3320M_test_repeat_count_1_default.txt
ftp://vps418301.ovh.net/incoming/vmap_test_results_v2/i5-3320M_test_repeat_count_1_patched.txt
# Hikey960 8x CPUs
<default> vs <patched>
real unknown real 4m25.214s
user unknown user 0m0.011s
sys unknown sys 0m0.670s
I did not manage to complete this test on "default Hikey960" kernel
version. After 24 hours it was still running, therefore i had to cancel
it. That is why real/user/sys are "unknown".
This patch (of 3):
Currently an allocation of the new vmap area is done over busy list
iteration(complexity O(n)) until a suitable hole is found between two busy
areas. Therefore each new allocation causes the list being grown. Due to
over fragmented list and different permissive parameters an allocation can
take a long time. For example on embedded devices it is milliseconds.
This patch organizes the KVA memory layout into free areas of the
1-ULONG_MAX range. It uses an augment red-black tree that keeps blocks
sorted by their offsets in pair with linked list keeping the free space in
order of increasing addresses.
Nodes are augmented with the size of the maximum available free block in
its left or right sub-tree. Thus, that allows to take a decision and
traversal toward the block that will fit and will have the lowest start
address, i.e. it is sequential allocation.
Allocation: to allocate a new block a search is done over the tree until a
suitable lowest(left most) block is large enough to encompass: the
requested size, alignment and vstart point. If the block is bigger than
requested size - it is split.
De-allocation: when a busy vmap area is freed it can either be merged or
inserted to the tree. Red-black tree allows efficiently find a spot
whereas a linked list provides a constant-time access to previous and next
blocks to check if merging can be done. In case of merging of
de-allocated memory chunk a large coalesced area is created.
Complexity: ~O(log(N))
[urezki@gmail.com: v3]
Link: http://lkml.kernel.org/r/20190402162531.10888-2-urezki@gmail.com
[urezki@gmail.com: v4]
Link: http://lkml.kernel.org/r/20190406183508.25273-2-urezki@gmail.com
Link: http://lkml.kernel.org/r/20190321190327.11813-2-urezki@gmail.com
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Reviewed-by: Roman Gushchin <guro@fb.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Thomas Garnier <thgarnie@google.com>
Cc: Oleksiy Avramchenko <oleksiy.avramchenko@sonymobile.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core
Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo:
perf.data:
Alexey Budankov:
- Streaming compression of perf ring buffer into PERF_RECORD_COMPRESSED
user space records, resulting in ~3-5x perf.data file size reduction
on variety of tested workloads what saves storage space on larger
server systems where perf.data size can easily reach several tens or
even hundreds of GiBs, especially when profiling with DWARF-based
stacks and tracing of context switches.
perf record:
Arnaldo Carvalho de Melo
- Improve -user-regs/intr-regs suggestions to overcome errors.
perf annotate:
Jin Yao:
- Remove hist__account_cycles() from callback, speeding up branch processing
(perf record -b).
perf stat:
- Add a 'percore' event qualifier, e.g.: -e cpu/event=0,umask=0x3,percore=1/,
that sums up the event counts for both hardware threads in a core.
We can already do this with --per-core, but it's often useful to do
this together with other metrics that are collected per hardware thread.
I.e. now its possible to do this per-event, and have it mixed with other
events not aggregated by core.
core libraries:
Donald Yandt:
- Check for errors when doing fgets(/proc/version).
Jiri Olsa:
- Speed up report for perf compiled with linbunwind.
tools headers:
Arnaldo Carvalho de Melo
- Update memcpy_64.S, x86's kvm.h and pt_regs.h.
arm64:
Florian Fainelli:
- Map Brahma-B53 CPUID to cortex-a53 events.
- Add Cortex-A57 and Cortex-A72 events.
csky:
Mao Han:
- Add DWARF register mappings for libdw, allowing --call-graph=dwarf to work
on the C-SKY arch.
x86:
Andi Kleen/Kan Liang:
- Add support for recording and printing XMM registers, available, for
instance, on Icelake.
Kan Liang:
- Add uncore_upi (Intel's "Ultra Path Interconnect" events) JSON support.
UPI replaced the Intel QuickPath Interconnect (QPI) in Xeon Skylake-SP.
Intel PT:
Adrian Hunter
. Fix instructions sampling rate.
. Timestamp fixes.
. Improve exported-sql-viewer GUI, allowing, for instance, to copy'n'paste
the trees, useful for e-mailing.
Documentation:
Thomas Richter:
- Add description for 'perf --debug stderr=1', which redirects stderr to stdout.
libtraceevent:
Tzvetomir Stoyanov:
- Add man pages for the various APIs.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
