linux_dsm_epyc7002/arch/sparc/kernel
Jane Chu 9d53caec84 sparc64: Measure receiver forward progress to avoid send mondo timeout
A large sun4v SPARC system may have moments of intensive xcall activities,
usually caused by unmapping many pages on many CPUs concurrently. This can
flood receivers with CPU mondo interrupts for an extended period, causing
some unlucky senders to hit send-mondo timeout. This problem gets worse
as cpu count increases because sometimes mappings must be invalidated on
all CPUs, and sometimes all CPUs may gang up on a single CPU.

But a busy system is not a broken system. In the above scenario, as long
as the receiver is making forward progress processing mondo interrupts,
the sender should continue to retry.

This patch implements the receiver's forward progress meter by introducing
a per cpu counter 'cpu_mondo_counter[cpu]' where 'cpu' is in the range
of 0..NR_CPUS. The receiver increments its counter as soon as it receives
a mondo and the sender tracks the receiver's counter. If the receiver has
stopped making forward progress when the retry limit is reached, the sender
declares send-mondo-timeout and panic; otherwise, the receiver is allowed
to keep making forward progress.

In addition, it's been observed that PCIe hotplug events generate Correctable
Errors that are handled by hypervisor and then OS. Hypervisor 'borrows'
a guest cpu strand briefly to provide the service. If the cpu strand is
simultaneously the only cpu targeted by a mondo, it may not be available
for the mondo in 20msec, causing SUN4V mondo timeout. It appears that 1 second
is the agreed wait time between hypervisor and guest OS, this patch makes
the adjustment.

Orabug: 25476541
Orabug: 26417466

Signed-off-by: Jane Chu <jane.chu@oracle.com>
Reviewed-by: Steve Sistare <steven.sistare@oracle.com>
Reviewed-by: Anthony Yznaga <anthony.yznaga@oracle.com>
Reviewed-by: Rob Gardner <rob.gardner@oracle.com>
Reviewed-by: Thomas Tai <thomas.tai@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-07-14 11:18:02 -07:00
..
.gitignore
apc.c sparc: kernel: apc: make of_device_ids const 2017-06-25 13:39:02 -07:00
asm-offsets.c sched/headers: Prepare to remove the <linux/mm_types.h> dependency from <linux/sched.h> 2017-03-02 08:42:37 +01:00
audit.c sparc: Convert naked unsigned uses to unsigned int 2016-03-20 21:28:58 -07:00
auxio_32.c sparc32: fix sparse warning in auxio_32.c 2014-05-18 19:01:27 -07:00
auxio_64.c sparc: kernel: drop owner assignment from platform_drivers 2014-10-20 16:20:15 +02:00
btext.c sparc64: fix sparse warning in btext.c 2014-05-18 19:01:30 -07:00
central.c sparc: kernel: drop owner assignment from platform_drivers 2014-10-20 16:20:15 +02:00
cherrs.S sparc64: Fix bootup regressions on some Kconfig combinations. 2016-04-27 17:27:37 -04:00
chmc.c sparc: kernel: drop owner assignment from platform_drivers 2014-10-20 16:20:15 +02:00
compat_audit.c sparc: Convert naked unsigned uses to unsigned int 2016-03-20 21:28:58 -07:00
cpu.c sparc64: recognize and support Sonoma CPU type 2016-04-21 16:43:47 -04:00
cpumap.c sparc64: recognize and support Sonoma CPU type 2016-04-21 16:43:47 -04:00
cpumap.h sparc: drop use of extern for prototypes in arch/sparc/* 2014-05-18 19:01:29 -07:00
devices.c sparc32: drop tadpole specific code 2014-05-18 19:01:29 -07:00
dma.c sparc: remove several unnecessary module.h include instances 2011-10-31 19:30:54 -04:00
ds.c sparc64: Fix build warnings with gcc 7. 2017-06-01 09:42:46 -07:00
dtlb_miss.S
dtlb_prot.S sparc: Don't leak context bits into thread->fault_address 2016-07-27 17:50:26 -07:00
ebus.c sparc: delete non-required instances of include <linux/init.h> 2014-01-28 23:38:23 -08:00
entry.h sparc: Resolve conflict between sparc v9 and M7 on usage of bit 9 of TTE 2015-05-31 22:15:01 -07:00
entry.S sparc: move exports to definitions 2016-08-07 23:55:43 -04:00
etrap_32.S sparc32: introduce support for run-time patching for all shared assembler code 2012-05-27 23:52:49 -07:00
etrap_64.S sparc64: clear syscall_noerror on the entry to syscall, not on the exit 2012-10-14 19:26:52 -04:00
fpu_traps.S sparc64: Fix bootup regressions on some Kconfig combinations. 2016-04-27 17:27:37 -04:00
ftrace.c sparc/ftrace: Fix ftrace graph time measurement 2017-05-17 12:07:47 -07:00
getsetcc.S
head_32.S sparc32: kill __ret_efault() 2017-03-28 18:24:00 -04:00
head_64.S sparc64: fix fault handling in NGbzero.S and GENbzero.S 2017-05-09 12:16:25 -07:00
helpers.S sparc: move exports to definitions 2016-08-07 23:55:43 -04:00
hvapi.c sparc64: Add ATU (new IOMMU) support 2016-11-18 11:16:59 -08:00
hvcalls.S sparc: move exports to definitions 2016-08-07 23:55:43 -04:00
hvtramp.S sparc64: Fix register corruption in top-most kernel stack frame during boot. 2014-10-24 09:52:49 -07:00
idprom.c net: Add eth_platform_get_mac_address() helper. 2016-01-06 16:31:56 -05:00
iommu_common.h sparc: implement ->mapping_error 2017-06-28 06:54:32 -07:00
iommu.c sparc: remove arch specific dma_supported implementations 2017-06-28 06:54:39 -07:00
ioport.c sparc: remove arch specific dma_supported implementations 2017-06-28 06:54:39 -07:00
irq_32.c sparc32: off by ones in BUG_ON() 2016-07-27 22:53:17 -07:00
irq_64.c arch/sparc: support NR_CPUS = 4096 2017-06-06 16:41:47 -04:00
irq.h sparc: drop use of extern for prototypes in arch/sparc/* 2014-05-18 19:01:29 -07:00
itlb_miss.S
ivec.S
jump_label.c jump_label: Reorder hotplug lock and jump_label_lock 2017-05-26 10:10:45 +02:00
kernel.h sparc64: initialize time early 2017-06-12 15:44:03 -07:00
kgdb_32.c sparc32: use proper prototype for trapbase 2016-05-20 17:55:41 -07:00
kgdb_64.c sparc64: fix sparse warning in kgdb_64.c 2014-05-18 19:01:34 -07:00
kprobes.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
kstack.h sparc64: Run NMIs on the hardirq stack. 2010-04-14 02:04:29 -07:00
ktlb.S sparc: Don't leak context bits into thread->fault_address 2016-07-27 17:50:26 -07:00
ldc.c sparc64: skip handshake for LDC channels in RAW mode 2017-06-25 13:43:12 -07:00
led.c sparc: use memdup_user_nul in sun4m LED driver 2017-05-09 12:16:25 -07:00
leon_kernel.c sparc: leon: Fix a retry loop in leon_init_timers() 2016-12-11 18:15:49 -08:00
leon_pci_grpci1.c genirq: Remove irq argument from irq flow handlers 2015-09-16 15:47:51 +02:00
leon_pci_grpci2.c genirq: Remove irq argument from irq flow handlers 2015-09-16 15:47:51 +02:00
leon_pci.c PCI: Cleanup control flow 2015-03-19 10:17:22 -05:00
leon_pmc.c sparc32: fix sparse warnings in leon_pmc.c 2014-04-29 01:12:27 -04:00
leon_smp.c sched/headers: Move task->mm handling methods to <linux/sched/mm.h> 2017-03-03 01:43:28 +01:00
Makefile sparc64:Support User Probes for sparc 2016-12-11 18:01:51 -08:00
mdesc.c sparc64: Enhance search for VIO device in MDESC 2017-06-25 13:43:14 -07:00
misctrap.S arch/sparc: Avoid DCTI Couples 2017-03-27 21:51:40 -07:00
module.c mm: vmalloc: pass additional vm_flags to __vmalloc_node_range() 2015-02-13 21:21:42 -08:00
nmi.c sparc: implement watchdog_nmi_enable and watchdog_nmi_disable 2016-12-14 16:04:08 -08:00
of_device_32.c sparc32: rename sparc_irq_config to sparc_config 2012-04-15 10:28:49 -07:00
of_device_64.c devicetree: add helper inline for retrieving a node's full name 2012-07-06 07:16:34 -05:00
of_device_common.c sparc: fix sparse warnings in of_device_common.c 2014-04-29 01:12:27 -04:00
of_device_common.h
pci_common.c sparc/PCI: Add mem64 resource parsing for root bus 2015-10-29 17:35:46 -05:00
pci_fire.c sparc: kernel: drop owner assignment from platform_drivers 2014-10-20 16:20:15 +02:00
pci_impl.h sparc/PCI: Add mem64 resource parsing for root bus 2015-10-29 17:35:46 -05:00
pci_msi.c PCI/MSI: Rename mask/unmask_msi_irq treewide 2014-11-23 13:01:45 +01:00
pci_psycho.c sparc: kernel: drop owner assignment from platform_drivers 2014-10-20 16:20:15 +02:00
pci_sabre.c sparc: kernel: drop owner assignment from platform_drivers 2014-10-20 16:20:15 +02:00
pci_schizo.c Driver core patches for 3.19-rc1 2014-12-14 16:10:09 -08:00
pci_sun4v_asm.S sparc64: Enable sun4v dma ops to use IOMMU v2 APIs 2016-11-18 11:17:00 -08:00
pci_sun4v.c SPARC64: Fix sun4v DMA panic 2017-07-12 08:13:04 -07:00
pci_sun4v.h sparc64: Enable sun4v dma ops to use IOMMU v2 APIs 2016-11-18 11:17:00 -08:00
pci.c PCI: Add BAR index argument to pci_mmap_page_range() 2017-04-20 08:47:47 -05:00
pcic.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
pcr.c sparc: perf: Add support M7 processor 2015-03-19 18:54:49 -07:00
perf_event.c perf core: Add a 'nr' field to perf_event_callchain_context 2016-05-16 23:11:51 -03:00
pmc.c sparc: kernel: pmc: make of_device_ids const. 2017-07-03 02:53:07 -07:00
power.c sparc: kernel: use builtin_platform_driver 2016-12-11 18:15:48 -08:00
process_32.c arch: remove unused macro/function thread_saved_pc() 2017-06-28 16:13:57 -07:00
process_64.c arch: remove unused macro/function thread_saved_pc() 2017-06-28 16:13:57 -07:00
prom_32.c sparc: remove several unnecessary module.h include instances 2011-10-31 19:30:54 -04:00
prom_64.c sparc64: fix typo in property 2017-06-25 22:01:11 -04:00
prom_common.c of: Fix locking vs. interrupts 2013-06-13 22:12:14 +01:00
prom_irqtrans.c sparc: Remove unnecessary semicolons 2011-06-07 16:06:34 -07:00
prom.h sparc: drop use of extern for prototypes in arch/sparc/* 2014-05-18 19:01:29 -07:00
psycho_common.c sparc: Remove unnecessary semicolons 2011-06-07 16:06:34 -07:00
psycho_common.h sparc: drop use of extern for prototypes in arch/sparc/* 2014-05-18 19:01:29 -07:00
ptrace_32.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
ptrace_64.c sparc64: Fill in rest of HAVE_REGS_AND_STACK_ACCESS_API 2017-04-23 17:15:51 -07:00
reboot.c Disintegrate asm/system.h for Sparc 2012-03-28 18:30:03 +01:00
rtrap_32.S sparc32: introduce support for run-time patching for all shared assembler code 2012-05-27 23:52:49 -07:00
rtrap_64.S arch/sparc: Avoid DCTI Couples 2017-03-27 21:51:40 -07:00
sbus.c sparc: kernel/sbus.c: fix memory leakage 2013-01-21 14:33:00 -08:00
setup_32.c sparc: Remove redundant tests in boot_flags_init(). 2017-05-03 08:28:48 -07:00
setup_64.c sparc64: initialize time early 2017-06-12 15:44:03 -07:00
signal32.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
signal_32.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
signal_64.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
sigutil_32.c sparc: Harden signal return frame checks. 2016-05-29 11:24:05 -07:00
sigutil_64.c sparc: Harden signal return frame checks. 2016-05-29 11:24:05 -07:00
sigutil.h sparc: Allow handling signals when stack is corrupted. 2011-08-20 17:14:54 -07:00
smp_32.c cpu/hotplug: Remove CPU_STARTING and CPU_DYING notifier 2016-09-06 18:30:19 +02:00
smp_64.c sparc64: Measure receiver forward progress to avoid send mondo timeout 2017-07-14 11:18:02 -07:00
sparc_ksyms.c sparc: move exports to definitions 2016-08-07 23:55:43 -04:00
spiterrs.S arch/sparc: Avoid DCTI Couples 2017-03-27 21:51:40 -07:00
sstate.c sparc: Fixed typo in sstate.c. Replaced panicing with panicking 2017-01-16 22:31:51 -05:00
stacktrace.c sched/headers: Prepare for new header dependencies before moving code to <linux/sched/debug.h> 2017-03-02 08:42:34 +01:00
starfire.c arch: sparc: kernel: starfire.c: Remove unused function 2015-03-01 21:33:58 -08:00
sun4d_irq.c sparc/irq: Use helper irq_data_get_irq_handler_data() 2015-07-31 22:20:05 +02:00
sun4d_smp.c sched/headers: Move task->mm handling methods to <linux/sched/mm.h> 2017-03-03 01:43:28 +01:00
sun4m_irq.c sched/headers: Prepare for new header dependencies before moving code to <linux/sched/debug.h> 2017-03-02 08:42:34 +01:00
sun4m_smp.c sched/headers: Move task->mm handling methods to <linux/sched/mm.h> 2017-03-03 01:43:28 +01:00
sun4v_ivec.S sparc64: Measure receiver forward progress to avoid send mondo timeout 2017-07-14 11:18:02 -07:00
sun4v_tlb_miss.S arch/sparc: Avoid DCTI Couples 2017-03-27 21:51:40 -07:00
sys32.S sparc: Hook up renameat2 syscall. 2014-07-21 22:27:56 -07:00
sys_sparc32.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
sys_sparc_32.c sched/headers: Prepare for new header dependencies before moving code to <linux/sched/debug.h> 2017-03-02 08:42:34 +01:00
sys_sparc_64.c mm: larger stack guard gap, between vmas 2017-06-19 21:50:20 +08:00
syscalls.S sparc: Fix system call tracing register handling. 2016-01-21 16:06:00 -08:00
sysfs.c sparc/sysfs: Replace racy task affinity logic 2017-04-15 12:20:54 +02:00
systbls_32.S sparc: Update syscall tables. 2017-04-23 18:28:55 -07:00
systbls_64.S sparc: Update syscall tables. 2017-04-23 18:28:55 -07:00
systbls.h sparc64: fix sparse warnings in sys_sparc32.c 2014-05-18 19:01:31 -07:00
time_32.c sparc/time: make of_device_ids const 2017-06-15 08:25:19 -07:00
time_64.c sparc64: fix typo in property 2017-06-25 22:01:11 -04:00
trampoline_32.S sparc: delete non-required instances of include <linux/init.h> 2014-01-28 23:38:23 -08:00
trampoline_64.S sparc64: Fix register corruption in top-most kernel stack frame during boot. 2014-10-24 09:52:49 -07:00
traps_32.c sched/headers: Move task->mm handling methods to <linux/sched/mm.h> 2017-03-03 01:43:28 +01:00
traps_64.c sparc64: Measure receiver forward progress to avoid send mondo timeout 2017-07-14 11:18:02 -07:00
tsb.S sparc64: mm: fix copy_tsb to correctly copy huge page TSBs 2017-06-06 13:45:02 -07:00
ttable_32.S sparc32: move trap table to a separate file 2012-05-19 23:27:25 -07:00
ttable_64.S sparc64: delete old wrap code 2017-06-06 13:45:29 -07:00
una_asm_32.S sparc32: unaligned memory access (MNA) trap handler bug 2011-02-01 12:39:59 -08:00
una_asm_64.S sparc: Fix .size directive for do_int_load 2011-03-16 18:19:15 -07:00
unaligned_32.c sched/headers: Prepare for new header dependencies before moving code to <linux/sched/signal.h> 2017-03-02 08:42:29 +01:00
unaligned_64.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
uprobes.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
urtt_fill.S arch/sparc: Avoid DCTI Couples 2017-03-27 21:51:40 -07:00
utrap.S sparc64: Fix bootup regressions on some Kconfig combinations. 2016-04-27 17:27:37 -04:00
vio.c Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-next 2017-07-08 12:14:14 -07:00
viohs.c sparc64: skip handshake for LDC channels in RAW mode 2017-06-25 13:43:12 -07:00
visemul.c scripts/spelling.txt: add "partiton" pattern and fix typo instances 2017-02-27 18:43:46 -08:00
vmlinux.lds.S sparc64: add hot-patched and inlined get_tick() 2017-06-12 15:44:03 -07:00
windows.c Replace <asm/uaccess.h> with <linux/uaccess.h> globally 2016-12-24 11:46:01 -08:00
winfixup.S arch/sparc: Avoid DCTI Couples 2017-03-27 21:51:40 -07:00
wof.S sparc32: introduce support for run-time patching for all shared assembler code 2012-05-27 23:52:49 -07:00
wuf.S sparc32: introduce support for run-time patching for all shared assembler code 2012-05-27 23:52:49 -07:00