There are two big uses of do_exit. The first is it's design use to be
the guts of the exit(2) system call. The second use is to terminate
a task after something catastrophic has happened like a NULL pointer
in kernel code.
Add a function make_task_dead that is initialy exactly the same as
do_exit to cover the cases where do_exit is called to handle
catastrophic failure. In time this can probably be reduced to just a
light wrapper around do_task_dead. For now keep it exactly the same so
that there will be no behavioral differences introducing this new
concept.
Replace all of the uses of do_exit that use it for catastraphic
task cleanup with make_task_dead to make it clear what the code
is doing.
As part of this rename rewind_stack_do_exit
rewind_stack_and_make_dead.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
In panic path, fadump is triggered via a panic notifier function.
Before calling panic notifier functions, smp_send_stop() gets called,
which stops all CPUs except the panic'ing CPU. Commit 8389b37dff
("powerpc: stop_this_cpu: remove the cpu from the online map.") and
again commit bab26238bb ("powerpc: Offline CPU in stop_this_cpu()")
started marking CPUs as offline while stopping them. So, if a kernel
has either of the above commits, vmcore captured with fadump via panic
path would not process register data for all CPUs except the panic'ing
CPU. Sample output of crash-utility with such vmcore:
# crash vmlinux vmcore
...
KERNEL: vmlinux
DUMPFILE: vmcore [PARTIAL DUMP]
CPUS: 1
DATE: Wed Nov 10 09:56:34 EST 2021
UPTIME: 00:00:42
LOAD AVERAGE: 2.27, 0.69, 0.24
TASKS: 183
NODENAME: XXXXXXXXX
RELEASE: 5.15.0+
VERSION: #974 SMP Wed Nov 10 04:18:19 CST 2021
MACHINE: ppc64le (2500 Mhz)
MEMORY: 8 GB
PANIC: "Kernel panic - not syncing: sysrq triggered crash"
PID: 3394
COMMAND: "bash"
TASK: c0000000150a5f80 [THREAD_INFO: c0000000150a5f80]
CPU: 1
STATE: TASK_RUNNING (PANIC)
crash> p -x __cpu_online_mask
__cpu_online_mask = $1 = {
bits = {0x2, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0}
}
crash>
crash>
crash> p -x __cpu_active_mask
__cpu_active_mask = $2 = {
bits = {0xff, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0}
}
crash>
While this has been the case since fadump was introduced, the issue
was not identified for two probable reasons:
- In general, the bulk of the vmcores analyzed were from crash
due to exception.
- The above did change since commit 8341f2f222 ("sysrq: Use
panic() to force a crash") started using panic() instead of
deferencing NULL pointer to force a kernel crash. But then
commit de6e5d3841 ("powerpc: smp_send_stop do not offline
stopped CPUs") stopped marking CPUs as offline till kernel
commit bab26238bb ("powerpc: Offline CPU in stop_this_cpu()")
reverted that change.
To ensure post processing register data of all other CPUs happens
as intended, let panic() function take the crash friendly path (read
crash_smp_send_stop()) with the help of crash_kexec_post_notifiers
option. Also, as register data for all CPUs is captured by f/w, skip
IPI callbacks here for fadump, to avoid any complications in finding
the right backtraces.
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211207103719.91117-2-hbathini@linux.ibm.com
Kdump can be triggered after panic_notifers since commit f06e5153f4
("kernel/panic.c: add "crash_kexec_post_notifiers" option for kdump
after panic_notifers") introduced crash_kexec_post_notifiers option.
But using this option would mean smp_send_stop(), that marks all other
CPUs as offline, gets called before kdump is triggered. As a result,
kdump routines fail to save other CPUs' registers. To fix this, kdump
friendly crash_smp_send_stop() function was introduced with kernel
commit 0ee59413c9 ("x86/panic: replace smp_send_stop() with kdump
friendly version in panic path"). Override this kdump friendly weak
function to handle crash_kexec_post_notifiers option appropriately
on powerpc.
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
[Fixed signature of crash_stop_this_cpu() - reported by lkp@intel.com]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211207103719.91117-1-hbathini@linux.ibm.com
Unlike PPC64 ABI, PPC32 uses the stack to pass a parameter defined
as a struct, even when the struct has a single simple element.
To avoid that, define ppc_inst_t as u32 on PPC32.
Keep it as 'struct ppc_inst' when __CHECKER__ is defined so that
sparse can perform type checking.
Also revert commit 511eea5e2c ("powerpc/kprobes: Fix Oops by passing
ppc_inst as a pointer to emulate_step() on ppc32") as now the
instruction to be emulated is passed as a register to emulate_step().
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/c6d0c46f598f76ad0b0a88bc0d84773bd921b17c.1638208156.git.christophe.leroy@csgroup.eu
On booke/40x we don't have segments like book3s/32.
On booke/40x we don't have access protection groups like 8xx.
Use the PID register to provide user access protection.
Kernel address space can be accessed with any PID.
User address space has to be accessed with the PID of the user.
User PID is always not null.
Everytime the kernel is entered, set PID register to 0 and
restore PID register when returning to user.
Everytime kernel needs to access user data, PID is restored
for the access.
In TLB miss handlers, check the PID and bail out to data storage
exception when PID is 0 and accessed address is in user space.
Note that also forbids execution of user text by kernel except
when user access is unlocked. But this shouldn't be a problem
as the kernel is not supposed to ever run user text.
This patch prepares the infrastructure but the real activation of KUAP
is done by following patches for each processor type one by one.
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/5d65576a8e31e9480415785a180c92dd4e72306d.1634627931.git.christophe.leroy@csgroup.eu
Calling 'mfsr' to get the content of segment registers is heavy,
in addition it requires clearing of the 'reserved' bits.
In order to avoid this operation, save it in mm context and in
thread struct.
The saved sr0 is the one used by kernel, this means that on
locking entry it can be used as is.
For unlocking, the only thing to do is to clear SR_NX.
This improves null_syscall selftest by 12 cycles, ie 4%.
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/b02baf2ed8f09bad910dfaeeb7353b2ae6830525.1634627931.git.christophe.leroy@csgroup.eu
When interrupt and syscall entries where converted to C, KUEP locking
and unlocking was also converted. It improved performance by unrolling
the loop, and allowed easily implementing boot time deactivation of
KUEP.
However, null_syscall selftest shows that KUEP is still heavy
(361 cycles with KUEP, 212 cycles without).
A way to improve more is to group 'mtsr's together, instead of
repeating 'addi' + 'mtsr' several times.
In order to do that, more registers need to be available. In C, GCC
will always be able to provide the requested number of registers, but
at the cost of saving some data on the stack, which is counter
performant here.
So let's do it in assembly, when we have full control of which
register can be used. It also has the advantage of locking earlier
and unlocking later and it helps GCC generating less tricky code.
The only drawback is to make boot time deactivation less straight
forward and require 'hand' instruction patching.
Group 'mtsr's by 4.
With this change, null_syscall selftest reports 336 cycles. Without
the change it was 361 cycles, that's a 7% reduction.
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/115cb279e9b9948dfd93a065e047081c59e3a2a6.1634627931.git.christophe.leroy@csgroup.eu
Compiling out hash support code when CONFIG_PPC_64S_HASH_MMU=n saves
128kB kernel image size (90kB text) on powernv_defconfig minus KVM,
350kB on pseries_defconfig minus KVM, 40kB on a tiny config.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
[mpe: Fixup defined(ARCH_HAS_MEMREMAP_COMPAT_ALIGN), which needs CONFIG.
Fix radix_enabled() use in setup_initial_memory_limit(). Add some
stubs to reduce number of ifdefs.]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211201144153.2456614-18-npiggin@gmail.com
The radix test can exclude slb_flush_all_realmode() from being called
because flush_and_reload_slb() is only expected to flush ERAT when
called by flush_erat(), which is only on pre-ISA v3.0 CPUs that do not
support radix.
This helps the later change to make hash support configurable to not
introduce runtime changes to radix mode behaviour.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211201144153.2456614-9-npiggin@gmail.com
In preparation for FORTIFY_SOURCE performing compile-time and run-time
field bounds checking for memset(), avoid intentionally writing across
neighboring fields.
Add a struct_group() for the spe registers so that memset() can correctly reason
about the size:
In function 'fortify_memset_chk',
inlined from 'restore_user_regs.part.0' at arch/powerpc/kernel/signal_32.c:539:3:
>> include/linux/fortify-string.h:195:4: error: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Werror=attribute-warning]
195 | __write_overflow_field();
| ^~~~~~~~~~~~~~~~~~~~~~~~
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Kees Cook <keescook@chromium.org>
Reviewed-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211118203604.1288379-1-keescook@chromium.org
Some thread flags can be set remotely, and so even when IRQs are disabled,
the flags can change under our feet. Generally this is unlikely to cause a
problem in practice, but it is somewhat unsound, and KCSAN will
legitimately warn that there is a data race.
To avoid such issues, a snapshot of the flags has to be taken prior to
using them. Some places already use READ_ONCE() for that, others do not.
Convert them all to the new flag accessor helpers.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Paul Mackerras <paulus@samba.org>
Link: https://lore.kernel.org/r/20211129130653.2037928-11-mark.rutland@arm.com
Some thread flags can be set remotely, and so even when IRQs are disabled,
the flags can change under our feet. Thus, when setting flags we must use
an atomic operation rather than a plain read-modify-write sequence, as a
plain read-modify-write may discard flags which are concurrently set by a
remote thread, e.g.
// task A // task B
tmp = A->thread_info.flags;
set_tsk_thread_flag(A, NEWFLAG_B);
tmp |= NEWFLAG_A;
A->thread_info.flags = tmp;
arch/powerpc/kernel/interrupt.c's system_call_exception() sets
_TIF_RESTOREALL in the thread info flags with a read-modify-write, which
may result in other flags being discarded.
Elsewhere in the file it uses clear_bits() to atomically remove flag bits,
so use set_bits() here for consistency with those.
There may be reasons (e.g. instrumentation) that prevent the use of
set_thread_flag() and clear_thread_flag() here, which would otherwise be
preferable.
Fixes: ae7aaecc3f ("powerpc/64s: system call rfscv workaround for TM bugs")
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Eirik Fuller <efuller@redhat.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nicholas Piggin <npiggin@gmail.com>
Link: https://lore.kernel.org/r/20211129130653.2037928-10-mark.rutland@arm.com
module_alloc() first tries to allocate module text within 24 bits direct
jump from kernel text, and tries a wider allocation if first one fails.
When first allocation fails the following is observed in kernel logs:
vmap allocation for size 2400256 failed: use vmalloc=<size> to increase size
systemd-udevd: vmalloc error: size 2395133, vm_struct allocation failed, mode:0xcc0(GFP_KERNEL), nodemask=(null)
CPU: 0 PID: 127 Comm: systemd-udevd Tainted: G W 5.15.5-gentoo-PowerMacG4 #9
Call Trace:
[e2a53a50] [c0ba0048] dump_stack_lvl+0x80/0xb0 (unreliable)
[e2a53a70] [c0540128] warn_alloc+0x11c/0x2b4
[e2a53b50] [c0531be8] __vmalloc_node_range+0xd8/0x64c
[e2a53c10] [c00338c0] module_alloc+0xa0/0xac
[e2a53c40] [c027a368] load_module+0x2ae0/0x8148
[e2a53e30] [c027fc78] sys_finit_module+0xfc/0x130
[e2a53f30] [c0035098] ret_from_syscall+0x0/0x28
...
Add __GFP_NOWARN flag to first allocation so that no warning appears
when it fails.
Reported-by: Erhard Furtner <erhard_f@mailbox.org>
Fixes: 2ec13df167 ("powerpc/modules: Load modules closer to kernel text")
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/93c9b84d6ec76aaf7b4f03468e22433a6d308674.1638267035.git.christophe.leroy@csgroup.eu
The printk layer at the moment does not seem to have a good way to force
flush printk messages that are created in NMI context, except in the
panic path.
NMI-context printk messages normally get to the console with irq_work,
but that won't help if the CPU is stuck with irqs disabled, as can be
the case for hard lockup watchdog messages.
The watchdog currently flushes the printk buffers after detecting a
lockup on remote CPUs, but they may not have processed their NMI IPI
yet by that stage, or they may have self-detected a lockup in which
case they won't go via this NMI IPI path.
Improve the situation by having NMI-context mark a flag if it called
printk, and have watchdog timer interrupts check if that flag was set
and try to flush if it was. Latency is not a big problem because we
were already stuck for a while, just need to try to make sure the
messages eventually make it out.
Depends-on: 5d5e4522a7 ("printk: restore flushing of NMI buffers on remote CPUs after NMI backtraces")
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Laurent Dufour <ldufour@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211119113146.752759-6-npiggin@gmail.com
All functions calling _mcount do it exactly the same way, with the
following sequence of instructions:
c07de788: 7c 08 02 a6 mflr r0
c07de78c: 90 01 00 04 stw r0,4(r1)
c07de790: 4b 84 13 65 bl c001faf4 <_mcount>
Allthough LR is pushed on stack, it is still in r0 while entering
_mcount().
Function arguments are in r3-r10, so r11 and r12 are still available
at that point.
Do like PPC64 and use r12 to move LR into CTR, so that r0 is preserved
and doesn't need to be restored from the stack.
While at it, bring back the EXPORT_SYMBOL at the end of _mcount.
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/24a3ba7db388537c44a038026f926d885372e6d3.1635423081.git.christophe.leroy@csgroup.eu
Prior to commit b1923caa6e ("powerpc: Merge 32-bit and 64-bit
setup_arch()") probe_machine() was called from setup_32/64.c and lived
in setup-common.c. But now it's only called from setup-common.c so it
can be static and __init, and we don't need the declaration in
machdep.h either.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211124093254.1054750-6-mpe@ellerman.id.au
setup_profiling_timer() is only needed when CONFIG_PROFILING is enabled.
Fixes the following W=1 warning when CONFIG_PROFILING=n:
linux/arch/powerpc/kernel/smp.c:1638:5: error: no previous prototype for ‘setup_profiling_timer’
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211124093254.1054750-5-mpe@ellerman.id.au
Pull powerpc fixes from Michael Ellerman:
"Fix KVM using a Power9 instruction on earlier CPUs, which could lead
to the host SLB being incorrectly invalidated and a subsequent host
crash.
Fix kernel hardlockup on vmap stack overflow on 32-bit.
Thanks to Christophe Leroy, Nicholas Piggin, and Fabiano Rosas"
* tag 'powerpc-5.16-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
powerpc/32: Fix hardlockup on vmap stack overflow
KVM: PPC: Book3S HV: Prevent POWER7/8 TLB flush flushing SLB
Pull asm-generic syscall table update from Arnd Bergmann:
"André Almeida sends an update for the newly added futex_waitv syscall
that was initially only added to a few architectures.
Some additional ones have since made it through architecture
maintainer trees, this finishes the remaining ones"
* tag 'asm-generic-5.16-2' of git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic:
futex: Wireup futex_waitv syscall
wd_smp_last_reset_tb now gets reset by watchdog_smp_panic() as part of
marking CPUs stuck and removing them from the pending mask before it
begins any printing. This causes last reset times reported to be off.
Fix this by reading it into a local variable before it gets reset.
Fixes: 76521c4b02 ("powerpc/watchdog: Avoid holding wd_smp_lock over printk and smp_send_nmi_ipi")
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211125103346.1188958-1-npiggin@gmail.com
When taking watchdog actions, printing messages, comparing and
re-setting wd_smp_last_reset_tb, etc., read TB close to the point of use
and under wd_smp_lock or printing lock (if applicable).
This should keep timebase mostly monotonic with kernel log messages, and
could prevent (in theory) a laggy CPU updating wd_smp_last_reset_tb to
something a long way in the past, and causing other CPUs to appear to be
stuck.
These additional TB reads are all slowpath (lockup has been detected),
so performance does not matter.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Laurent Dufour <ldufour@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211110025056.2084347-5-npiggin@gmail.com
There is a deadlock with the console_owner lock and the wd_smp_lock:
CPU x takes the console_owner lock
CPU y takes a watchdog timer interrupt and takes __wd_smp_lock
CPU x takes a soft-NMI interrupt, detects deadlock, spins on __wd_smp_lock
CPU y detects deadlock, tries to print something and spins on console_owner
-> deadlock
Change the watchdog locking scheme so wd_smp_lock protects the watchdog
internal data, but "reporting" (printing, issuing NMI IPIs, taking any
action outside of watchdog) uses a non-waiting exclusion. If a CPU detects
a problem but can not take the reporting lock, it just returns because
something else is already reporting. It will try again at some point.
Typically hard lockup watchdog report usefulness is not impacted due to
failure to spewing a large enough amount of data in as short a time as
possible, but by messages getting garbled.
Laurent debugged this and found the deadlock, and this patch is based on
his general approach to avoid expensive operations while holding the lock.
With the addition of the reporting exclusion.
Signed-off-by: Laurent Dufour <ldufour@linux.ibm.com>
[np: rework to add reporting exclusion update changelog]
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211110025056.2084347-4-npiggin@gmail.com
Most updates to wd_smp_cpus_pending are under lock except the watchdog
interrupt bit clear.
This can race with non-atomic RMW updates to the mask under lock, which
can happen in two instances:
Firstly, if another CPU detects this one is stuck, removes it from the
mask, mask becomes empty and is re-filled with non-atomic stores. This
is okay because it would re-fill the mask with this CPU's bit clear
anyway (because this CPU is now stuck), so it doesn't matter that the
bit clear update got "lost". Add a comment for this.
Secondly, if another CPU detects a different CPU is stuck and removes it
from the pending mask with a non-atomic store to bytes which also
include the bit of this CPU. This case can result in the bit clear being
lost and the end result being the bit is set. This should be so rare it
hardly matters, but to make things simpler to reason about just avoid
the non-atomic access for that case.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Laurent Dufour <ldufour@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211110025056.2084347-3-npiggin@gmail.com
It is possible for all CPUs to miss the pending cpumask becoming clear,
and then nobody resetting it, which will cause the lockup detector to
stop working. It will eventually expire, but watchdog_smp_panic will
avoid doing anything if the pending mask is clear and it will never be
reset.
Order the cpumask clear vs the subsequent test to close this race.
Add an extra check for an empty pending mask when the watchdog fires and
finds its bit still clear, to try to catch any other possible races or
bugs here and keep the watchdog working. The extra test in
arch_touch_nmi_watchdog is required to prevent the new warning from
firing off.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Laurent Dufour <ldufour@linux.ibm.com>
Debugged-by: Laurent Dufour <ldufour@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211110025056.2084347-2-npiggin@gmail.com
Generally RTAS cannot block, and in PAPR it is required to return control
to the OS within a few tens of microseconds. In order to support operations
which may take longer to complete, many RTAS primitives can return
intermediate -2 ("busy") or 990x ("extended delay") values, which indicate
that the OS should reattempt the same call with the same arguments at some
point in the future.
Current versions of PAPR are less than clear about this, but the intended
meanings of these values in more detail are:
RTAS_BUSY (-2): RTAS has suspended a potentially long-running operation in
order to meet its latency obligation and give the OS the opportunity to
perform other work. RTAS can resume making progress as soon as the OS
reattempts the call.
RTAS_EXTENDED_DELAY_{MIN...MAX} (9900-9905): RTAS must wait for an external
event to occur or for internal contention to resolve before it can complete
the requested operation. The value encodes a non-binding hint as to roughly
how long the OS should wait before calling again, but the OS is allowed to
reattempt the call sooner or even immediately.
Linux of course must take its own CPU scheduling obligations into account
when handling these statuses; e.g. a task which receives an RTAS_BUSY
status should check whether to reschedule before it attempts the RTAS call
again to avoid starving other tasks.
rtas_busy_delay() is a helper function that "consumes" a busy or extended
delay status. Common usage:
int rc;
do {
rc = rtas_call(rtas_token("some-function"), ...);
} while (rtas_busy_delay(rc));
/* convert rc to Linux error value, etc */
If rc is a busy or extended delay status, the caller can rely on
rtas_busy_delay() to perform an appropriate sleep or reschedule and return
nonzero. Other statuses are handled normally by the caller.
The current implementation of rtas_busy_delay() both oversleeps and
overuses the CPU:
* It performs msleep() for all 990x and even when no delay is
suggested (-2), but this is understood to actually sleep for two jiffies
minimum in practice (20ms with HZ=100). 9900 (1ms) and 9901 (10ms)
appear to be the most common extended delay statuses, and the
oversleeping measurably lengthens DLPAR operations, which perform
many RTAS calls.
* It does not sleep on 990x unless need_resched() is true, causing code
like the loop above to needlessly retry, wasting CPU time.
Alter the logic to align better with the intended meanings:
* When passed RTAS_BUSY, perform cond_resched() and return without
sleeping. The caller should reattempt immediately
* Always sleep when passed an extended delay status, using usleep_range()
for precise shorter sleeps. Limit the sleep time to one second even
though there are higher architected values.
Change rtas_busy_delay()'s return type to bool to better reflect its usage,
and add kernel-doc.
rtas_busy_delay_time() is unchanged, even though it "incorrectly" returns 1
for RTAS_BUSY. There are users of that API with open-coded delay loops in
sensitive contexts that will have to be taken on an individual basis.
Brief results for addition and removal of 5GB memory on a small P9 PowerVM
partition follow. Load was generated with stress-ng --cpu N. For add,
elapsed time is greatly reduced without significant change in the number of
RTAS calls or time spent on CPU. For remove, elapsed time is modestly
reduced, with significant reductions in RTAS calls and time spent on CPU.
With no competing workload (- before, + after):
Performance counter stats for 'bash -c echo "memory add count 20" > /sys/kernel/dlpar' (10 runs):
- 1,935 probe:rtas_call # 0.003 M/sec ( +- 0.22% )
- 609.99 msec task-clock # 0.183 CPUs utilized ( +- 0.19% )
+ 1,956 probe:rtas_call # 0.003 M/sec ( +- 0.17% )
+ 618.56 msec task-clock # 0.278 CPUs utilized ( +- 0.11% )
- 3.3322 +- 0.0670 seconds time elapsed ( +- 2.01% )
+ 2.2222 +- 0.0416 seconds time elapsed ( +- 1.87% )
Performance counter stats for 'bash -c echo "memory remove count 20" > /sys/kernel/dlpar' (10 runs):
- 6,224 probe:rtas_call # 0.008 M/sec ( +- 2.57% )
- 750.36 msec task-clock # 0.190 CPUs utilized ( +- 2.01% )
+ 843 probe:rtas_call # 0.003 M/sec ( +- 0.12% )
+ 250.66 msec task-clock # 0.068 CPUs utilized ( +- 0.17% )
- 3.9394 +- 0.0890 seconds time elapsed ( +- 2.26% )
+ 3.678 +- 0.113 seconds time elapsed ( +- 3.07% )
With all CPUs 100% busy (- before, + after):
Performance counter stats for 'bash -c echo "memory add count 20" > /sys/kernel/dlpar' (10 runs):
- 2,979 probe:rtas_call # 0.003 M/sec ( +- 0.12% )
- 1,096.62 msec task-clock # 0.105 CPUs utilized ( +- 0.10% )
+ 2,981 probe:rtas_call # 0.003 M/sec ( +- 0.22% )
+ 1,095.26 msec task-clock # 0.154 CPUs utilized ( +- 0.21% )
- 10.476 +- 0.104 seconds time elapsed ( +- 1.00% )
+ 7.1124 +- 0.0865 seconds time elapsed ( +- 1.22% )
Performance counter stats for 'bash -c echo "memory remove count 20" > /sys/kernel/dlpar' (10 runs):
- 2,702 probe:rtas_call # 0.004 M/sec ( +- 4.00% )
- 722.71 msec task-clock # 0.067 CPUs utilized ( +- 2.41% )
+ 1,246 probe:rtas_call # 0.003 M/sec ( +- 0.25% )
+ 487.73 msec task-clock # 0.049 CPUs utilized ( +- 0.20% )
- 10.829 +- 0.163 seconds time elapsed ( +- 1.51% )
+ 9.9887 +- 0.0866 seconds time elapsed ( +- 0.87% )
Signed-off-by: Nathan Lynch <nathanl@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211117060259.957178-2-nathanl@linux.ibm.com
Fix the following issues reported by kernel-doc:
$ scripts/kernel-doc -v -none arch/powerpc/kernel/rtas.c
arch/powerpc/kernel/rtas.c:810: info: Scanning doc for function rtas_activate_firmware
arch/powerpc/kernel/rtas.c:818: warning: contents before sections
arch/powerpc/kernel/rtas.c:841: info: Scanning doc for function rtas_call_reentrant
arch/powerpc/kernel/rtas.c:893: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
* Find a specific pseries error log in an RTAS extended event log.
Signed-off-by: Nathan Lynch <nathanl@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211116215806.928235-1-nathanl@linux.ibm.com
The EEH recovery logic in eeh_handle_normal_event() has some pretty strange
flow control. If we remove all the actual recovery logic we're left with
the following skeleton:
if (result != PCI_ERS_RESULT_DISCONNECT) {
...
}
if (result != PCI_ERS_RESULT_DISCONNECT) {
...
}
if (result == PCI_ERS_RESULT_NONE) {
...
}
if (result == PCI_ERS_RESULT_CAN_RECOVER) {
...
}
if (result == PCI_ERS_RESULT_CAN_RECOVER) {
...
}
if (result == PCI_ERS_RESULT_NEED_RESET) {
...
}
if ((result == PCI_ERS_RESULT_RECOVERED) ||
(result == PCI_ERS_RESULT_NONE)) {
...
goto out;
}
/*
* unsuccessful recovery / PCI_ERS_RESULT_DISCONECTED
* handling is here.
*/
...
out:
...
Most of the "if () { ... }" blocks above change "result" to
PCI_ERS_RESULT_DISCONNECTED if an error occurs in that recovery step. This
makes the control flow a bit confusing since it breaks the early-exit
pattern that is generally used in Linux. In any case we end up handling the
error in the final else block so why not just jump there directly? Doing so
also allows us to de-indent a bunch of code.
No functional changes.
[dja: rebase on top of linux-next + my preceeding refactor,
move clearing the EEH_DEV_NO_HANDLER bit above the first goto so that
it is always clear in the error handler code as it was before.]
Signed-off-by: Oliver O'Halloran <oohall@gmail.com>
Signed-off-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211015070628.1331635-2-dja@axtens.net
The control flow of eeh_handle_normal_event() is a bit tricky.
Break out one of the error handling paths - rather than be in an else
block, we'll make it part of the regular body of the function and put a
'goto out;' in the true limb of the if.
Signed-off-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211015070628.1331635-1-dja@axtens.net
for_each_node_by_type performs an of_node_get on each iteration, so
a break out of the loop requires an of_node_put.
A simplified version of the semantic patch that fixes this problem is as
follows (http://coccinelle.lip6.fr):
// <smpl>
@@
local idexpression n;
expression e;
@@
for_each_node_by_type(n,...) {
...
(
of_node_put(n);
|
e = n
|
+ of_node_put(n);
? break;
)
...
}
... when != n
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/1448051604-25256-6-git-send-email-Julia.Lawall@lip6.fr
