The pmem infrastructure uses memcpy_mcsafe in the pmem layer so as to
convert machine check exceptions into a return value on failure in case
a machine check exception is encountered during the memcpy. The return
value is the number of bytes remaining to be copied.
This patch largely borrows from the copyuser_power7 logic and does not add
the VMX optimizations, largely to keep the patch simple. If needed those
optimizations can be folded in.
Signed-off-by: Balbir Singh <bsingharora@gmail.com>
[arbab@linux.ibm.com: Added symbol export]
Co-developed-by: Santosh Sivaraj <santosh@fossix.org>
Signed-off-by: Santosh Sivaraj <santosh@fossix.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190820081352.8641-7-santosh@fossix.org
If we take a UE on one of the instructions with a fixup entry, set nip
to continue execution at the fixup entry. Stop processing the event
further or print it.
Co-developed-by: Reza Arbab <arbab@linux.ibm.com>
Signed-off-by: Reza Arbab <arbab@linux.ibm.com>
Signed-off-by: Balbir Singh <bsingharora@gmail.com>
Reviewed-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Reviewed-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Santosh Sivaraj <santosh@fossix.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190820081352.8641-6-santosh@fossix.org
The function doesn't get used outside this file, so make it static.
Signed-off-by: Reza Arbab <arbab@linux.ibm.com>
Signed-off-by: Santosh Sivaraj <santosh@fossix.org>
Reviewed-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190820081352.8641-4-santosh@fossix.org
The current code would fail on huge pages addresses, since the shift would
be incorrect. Use the correct page shift value returned by
__find_linux_pte() to get the correct physical address. The code is more
generic and can handle both regular and compound pages.
Fixes: ba41e1e1cc ("powerpc/mce: Hookup derror (load/store) UE errors")
Signed-off-by: Balbir Singh <bsingharora@gmail.com>
[arbab@linux.ibm.com: Fixup pseries_do_memory_failure()]
Signed-off-by: Reza Arbab <arbab@linux.ibm.com>
Tested-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Santosh Sivaraj <santosh@fossix.org>
Cc: stable@vger.kernel.org # v4.15+
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190820081352.8641-3-santosh@fossix.org
After a partition migration, pseries_devicetree_update() processes
changes to the device tree communicated from the platform to
Linux. This is a relatively heavyweight operation, with multiple
device tree searches, memory allocations, and conversations with
partition firmware.
There's a few levels of nested loops which are bounded only by
decisions made by the platform, outside of Linux's control, and indeed
we have seen RCU stalls on large systems while executing this call
graph. Use cond_resched() in these loops so that the cpu is yielded
when needed.
Signed-off-by: Nathan Lynch <nathanl@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190802192926.19277-4-nathanl@linux.ibm.com
rtas_cpu_state_change_mask() potentially operates on scores of cpus,
so explicitly allow rescheduling in the loop body.
Signed-off-by: Nathan Lynch <nathanl@linux.ibm.com>
Reviewed-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190802192926.19277-3-nathanl@linux.ibm.com
The LPAR migration implementation and userspace-initiated cpu hotplug
can interleave their executions like so:
1. Set cpu 7 offline via sysfs.
2. Begin a partition migration, whose implementation requires the OS
to ensure all present cpus are online; cpu 7 is onlined:
rtas_ibm_suspend_me -> rtas_online_cpus_mask -> cpu_up
This sets cpu 7 online in all respects except for the cpu's
corresponding struct device; dev->offline remains true.
3. Set cpu 7 online via sysfs. _cpu_up() determines that cpu 7 is
already online and returns success. The driver core (device_online)
sets dev->offline = false.
4. The migration completes and restores cpu 7 to offline state:
rtas_ibm_suspend_me -> rtas_offline_cpus_mask -> cpu_down
This leaves cpu7 in a state where the driver core considers the cpu
device online, but in all other respects it is offline and
unused. Attempts to online the cpu via sysfs appear to succeed but the
driver core actually does not pass the request to the lower-level
cpuhp support code. This makes the cpu unusable until the cpu device
is manually set offline and then online again via sysfs.
Instead of directly calling cpu_up/cpu_down, the migration code should
use the higher-level device core APIs to maintain consistent state and
serialize operations.
Fixes: 120496ac2d ("powerpc: Bring all threads online prior to migration/hibernation")
Signed-off-by: Nathan Lynch <nathanl@linux.ibm.com>
Reviewed-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190802192926.19277-2-nathanl@linux.ibm.com
If a page is already mapped RW without the DIRTY flag, the DIRTY
flag is never set and a TLB store miss exception is taken forever.
This is easily reproduced with the following app:
void main(void)
{
volatile char *ptr = mmap(0, 4096, PROT_READ | PROT_WRITE, MAP_SHARED | MAP_ANONYMOUS, -1, 0);
*ptr = *ptr;
}
When DIRTY flag is not set, bail out of TLB miss handler and take
a minor page fault which will set the DIRTY flag.
Fixes: f8b58c64ea ("powerpc/603: let's handle PAGE_DIRTY directly")
Cc: stable@vger.kernel.org # v5.1+
Reported-by: Doug Crawford <doug.crawford@intelight-its.com>
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/80432f71194d7ee75b2f5043ecf1501cf1cca1f3.1566196646.git.christophe.leroy@c-s.fr
pfn_pte is never given a pte above the addressable physical memory
limit, so the masking is redundant. In case of a software bug, it
is not obviously better to silently truncate the pfn than to corrupt
the pte (either one will result in memory corruption or crashes),
so there is no reason to add this to the fast path.
Add VM_BUG_ON to catch cases where the pfn is invalid. These would
catch the create_section_mapping bug fixed by a previous commit.
[16885.256466] ------------[ cut here ]------------
[16885.256492] kernel BUG at arch/powerpc/include/asm/book3s/64/pgtable.h:612!
cpu 0x0: Vector: 700 (Program Check) at [c0000000ee0a36d0]
pc: c000000000080738: __map_kernel_page+0x248/0x6f0
lr: c000000000080ac0: __map_kernel_page+0x5d0/0x6f0
sp: c0000000ee0a3960
msr: 9000000000029033
current = 0xc0000000ec63b400
paca = 0xc0000000017f0000 irqmask: 0x03 irq_happened: 0x01
pid = 85, comm = sh
kernel BUG at arch/powerpc/include/asm/book3s/64/pgtable.h:612!
Linux version 5.3.0-rc1-00001-g0fe93e5f3394
enter ? for help
[c0000000ee0a3a00] c000000000d37378 create_physical_mapping+0x260/0x360
[c0000000ee0a3b10] c000000000d370bc create_section_mapping+0x1c/0x3c
[c0000000ee0a3b30] c000000000071f54 arch_add_memory+0x74/0x130
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190724084638.24982-5-npiggin@gmail.com
Ensure __va is given a physical address below PAGE_OFFSET, and __pa is
given a virtual address above PAGE_OFFSET.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190724084638.24982-4-npiggin@gmail.com
The alloc_pages_node return value should be tested for failure
before being passed to page_address.
Tested-by: Anju T Sudhakar <anju@linux.vnet.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190724084638.24982-3-npiggin@gmail.com
create_physical_mapping expects physical addresses, but splitting
these mapping on hot unplug is supplying virtual (effective)
addresses.
Fixes: 4dd5f8a99e ("powerpc/mm/radix: Split linear mapping on hot-unplug")
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190724084638.24982-2-npiggin@gmail.com
create_physical_mapping expects physical addresses, but creating and
splitting these mappings after boot is supplying virtual (effective)
addresses. This can be irritated by booting with mem= to limit memory
then probing an unused physical memory range:
echo <addr> > /sys/devices/system/memory/probe
This mostly works by accident, firstly because __va(__va(x)) == __va(x)
so the virtual address does not get corrupted. Secondly because pfn_pte
masks out the upper bits of the pfn beyond the physical address limit,
so a pfn constructed with a 0xc000000000000000 virtual linear address
will be masked back to the correct physical address in the pte.
Fixes: 6cc27341b2 ("powerpc/mm: add radix__create_section_mapping()")
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190724084638.24982-1-npiggin@gmail.com
current may be cached by the compiler, so remove the volatile asm
restriction. This results in better generated code, as well as being
smaller and fewer dependent loads, it can avoid store-hit-load flushes
like this one that shows up in irq_exit():
preempt_count_sub(HARDIRQ_OFFSET);
if (!in_interrupt() && ...)
Which ends up as:
((struct thread_info *)current)->preempt_count -= HARDIRQ_OFFSET;
if (((struct thread_info *)current)->preempt_count ...
Evaluating current twice presently means it has to be loaded twice, and
here gcc happens to pick a different register each time, then
preempt_count is accessed via that base register:
1058: ld r10,2392(r13) <-- current
105c: lwz r9,0(r10) <-- preempt_count
1060: addis r9,r9,-1
1064: stw r9,0(r10) <-- preempt_count
1068: ld r9,2392(r13) <-- current
106c: lwz r9,0(r9) <-- preempt_count
1070: rlwinm. r9,r9,0,11,23
1074: bne 1090 <irq_exit+0x60>
This can frustrate store-hit-load detection heuristics and cause
flushes. Allowing the compiler to cache current in a reigster with this
patch results in the same base register being used for all accesses,
which is more likely to be detected as an alias:
1058: ld r31,2392(r13)
...
1070: lwz r9,0(r31)
1074: addis r9,r9,-1
1078: stw r9,0(r31)
107c: lwz r9,0(r31)
1080: rlwinm. r9,r9,0,11,23
1084: bne 10a0 <irq_exit+0x60>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190612140317.24490-1-npiggin@gmail.com
copy_page() and clear_page() expect page aligned destination, and
use dcbz instruction to clear entire cache lines based on the
assumption that the destination is cache aligned.
As shown during analysis of a bug in BTRFS filesystem, a misaligned
copy_page() can create bugs that are difficult to locate (see Link).
Add an explicit WARNING when copy_page() or clear_page() are called
with misaligned destination.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://bugzilla.kernel.org/show_bug.cgi?id=204371
Link: https://lore.kernel.org/r/c6cea38f90480268d439ca44a645647e260fff09.1565941808.git.christophe.leroy@c-s.fr
update_mmu_cache() is only for BOOK3S, and can be simplified for
BOOK3S32.
Move it out of mem.c into respective BOOK3S32 and BOOK3S64 files
containing hash utils.
BOOK3S64 version of hash_preload() is only used locally, declare it
static.
Remove the radix_enabled() stuff in BOOK3S32 version.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/107aaf43583a5f5d09e0d4e84c4c4390ecfcd512.1565933217.git.christophe.leroy@c-s.fr
Only BOOK3S and FSL_BOOK3E have a usefull update_mmu_cache().
For the others, just define it static inline.
In the meantime, simplify the FSL_BOOK3E related ifdef as
book3e_hugetlb_preload() only exists when CONFIG_PPC_FSL_BOOK3E
is selected.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/668aba4db6b9af6d8a151174e11a4289f1a6bbcd.1565933217.git.christophe.leroy@c-s.fr
When KASAN is selected, the definitive hash table has to be
set up later, but there is already an early temporary one.
When KASAN is not selected, there is no early hash table,
so the setup of the definitive hash table cannot be delayed.
Fixes: 72f208c6a8 ("powerpc/32s: move hash code patching out of MMU_init_hw()")
Cc: stable@vger.kernel.org # v5.2+
Reported-by: Jonathan Neuschafer <j.neuschaefer@gmx.net>
Tested-by: Jonathan Neuschafer <j.neuschaefer@gmx.net>
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/b7860c5e1e784d6b96ba67edf47dd6cbc2e78ab6.1565776892.git.christophe.leroy@c-s.fr
We see warnings such as:
kernel/futex.c: In function 'do_futex':
kernel/futex.c:1676:17: warning: 'oldval' may be used uninitialized in this function [-Wmaybe-uninitialized]
return oldval == cmparg;
^
kernel/futex.c:1651:6: note: 'oldval' was declared here
int oldval, ret;
^
This is because arch_futex_atomic_op_inuser() only sets *oval if ret
is 0 and GCC doesn't see that it will only use it when ret is 0.
Anyway, the non-zero ret path is an error path that won't suffer from
setting *oval, and as *oval is a local var in futex_atomic_op_inuser()
it will have no impact.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
[mpe: reword change log slightly]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/86b72f0c134367b214910b27b9a6dd3321af93bb.1565774657.git.christophe.leroy@c-s.fr
When loading modules, from time to time an Oops is encountered during
the init of shadow area for globals. This is due to the last page not
always being mapped depending on the exact distance between the start
and the end of the shadow area and the alignment with the page
addresses.
Fix this by aligning the starting address with the page address.
Fixes: 2edb16efc8 ("powerpc/32: Add KASAN support")
Cc: stable@vger.kernel.org # v5.2+
Reported-by: Erhard F. <erhard_f@mailbox.org>
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/4f887e9b77d0d725cbb52035c7ece485c1c5fc14.1565361881.git.christophe.leroy@c-s.fr
Parallel loading of modules may lead to bad setup of shadow page table
entries.
First, lets align modules so that two modules never share the same
shadow page.
Second, ensure that two modules cannot allocate two page tables for
the same PMD entry at the same time. This is done by using
init_mm.page_table_lock in the same way as __pte_alloc_kernel()
Fixes: 2edb16efc8 ("powerpc/32: Add KASAN support")
Cc: stable@vger.kernel.org # v5.2+
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/c97284f912128cbc3f2fe09d68e90e65fb3e6026.1565361876.git.christophe.leroy@c-s.fr
On 8xx, breakpoints stop after executing the instruction, so
stepping/emulation is not needed. Move it into a sub-function and
remove the #ifdefs.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Reviewed-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/f8cdc3f1c66ad3c43ebc568abcc6c39ed4676284.1561737231.git.christophe.leroy@c-s.fr
hashpagetable.c is only compiled when CONFIG_PPC_BOOK3S_64 is
defined, so drop the test and its 'else' branch.
Use IS_ENABLED(CONFIG_PPC_PSERIES) instead of #ifdef, this allows the
code to be checked at any build. It is still optimised out by GCC.
Use IS_ENABLED(CONFIG_PPC_64K_PAGES) instead of #ifdef.
Use IS_ENABLED(CONFIG_SPARSEMEN_VMEMMAP) instead of #ifdef.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/c8998ed32e4e3954b56a8dacecfe43319a2a0483.1565786091.git.christophe.leroy@c-s.fr
walk_pagetables() always walk the entire pgdir from address 0
but considers PAGE_OFFSET or KERN_VIRT_START as the starting
address of the walk, resulting in a possible mismatch in the
displayed addresses.
Ex: on PPC32, when KERN_VIRT_START was locally defined as
PAGE_OFFSET, ptdump displayed 0x80000000
instead of 0xc0000000 for the first kernel page,
because 0xc0000000 + 0xc0000000 = 0x80000000
Start the walk at st->start_address instead of starting at 0.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/5aa2ac513295f594cce8ddb1c649f61947bd063d.1565786091.git.christophe.leroy@c-s.fr
Modify the xmon 'dxi' command to query all interrupts if no IRQ number
is specified.
Signed-off-by: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190814154754.23682-4-clg@kaod.org
The xmon 'dxi' command calls OPAL to query the XIVE configuration of a
interrupt. This can only be done on baremetal (PowerNV) and it will
crash a pseries machine.
Introduce a new XIVE get_irq_config() operation which implements a
different query depending on the platform, PowerNV or pseries, and
modify xmon to use a top level wrapper.
Signed-off-by: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190814154754.23682-3-clg@kaod.org
Currently, the xmon 'dx' command calls OPAL to dump the XIVE state in
the OPAL logs and also outputs some of the fields of the internal XIVE
structures in Linux. The OPAL calls can only be done on baremetal
(PowerNV) and they crash a pseries machine. Fix by checking the
hypervisor feature of the CPU.
Signed-off-by: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190814154754.23682-2-clg@kaod.org
At the moment we create a small window only for 32bit devices, the window
maps 0..2GB of the PCI space only. For other devices we either use
a sketchy bypass or hardware bypass but the former can only work if
the amount of RAM is no bigger than the device's DMA mask and the latter
requires devices to support at least 59bit DMA.
This extends the default DMA window to the maximum size possible to allow
a wider DMA mask than just 32bit. The default window size is now limited
by the the iommu_table::it_map allocation bitmap which is a contiguous
array, 1 bit per an IOMMU page.
This increases the default IOMMU page size from hard coded 4K to
the system page size to allow wider DMA masks.
This increases the level number to not exceed the max order allocation
limit per TCE level. By the same time, this keeps minimal levels number
as 2 in order to save memory.
As the extended window now overlaps the 32bit MMIO region, this adds
an area reservation to iommu_init_table().
After this change the default window size is 0x80000000000==1<<43 so
devices limited to DMA mask smaller than the amount of system RAM can
still use more than just 2GB of memory for DMA.
This is an optimization and not a bug fix for DMA API usage.
With the on-demand allocation of indirect TCE table levels enabled and
2 levels, the first TCE level size is just
1<<ceil((log2(0x7ffffffffff+1)-16)/2)=16384 TCEs or 2 system pages.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190718051139.74787-5-aik@ozlabs.ru
We allocate only the first level of multilevel TCE tables for KVM
already (alloc_userspace_copy==true), and the rest is allocated on demand.
This is not enabled though for bare metal.
This removes the KVM limitation (implicit, via the alloc_userspace_copy
parameter) and always allocates just the first level. The on-demand
allocation of missing levels is already implemented.
As from now on DMA map might happen with disabled interrupts, this
allocates TCEs with GFP_ATOMIC; otherwise lockdep reports errors 1].
In practice just a single page is allocated there so chances for failure
are quite low.
To save time when creating a new clean table, this skips non-allocated
indirect TCE entries in pnv_tce_free just like we already do in
the VFIO IOMMU TCE driver.
This changes the default level number from 1 to 2 to reduce the amount
of memory required for the default 32bit DMA window at the boot time.
The default window size is up to 2GB which requires 4MB of TCEs which is
unlikely to be used entirely or at all as most devices these days are
64bit capable so by switching to 2 levels by default we save 4032KB of
RAM per a device.
While at this, add __GFP_NOWARN to alloc_pages_node() as the userspace
can trigger this path via VFIO, see the failure and try creating a table
again with different parameters which might succeed.
[1]:
===
BUG: sleeping function called from invalid context at mm/page_alloc.c:4596
in_atomic(): 1, irqs_disabled(): 1, pid: 1038, name: scsi_eh_1
2 locks held by scsi_eh_1/1038:
#0: 000000005efd659a (&host->eh_mutex){+.+.}, at: ata_eh_acquire+0x34/0x80
#1: 0000000006cf56a6 (&(&host->lock)->rlock){....}, at: ata_exec_internal_sg+0xb0/0x5c0
irq event stamp: 500
hardirqs last enabled at (499): [<c000000000cb8a74>] _raw_spin_unlock_irqrestore+0x94/0xd0
hardirqs last disabled at (500): [<c000000000cb85c4>] _raw_spin_lock_irqsave+0x44/0x120
softirqs last enabled at (0): [<c000000000101120>] copy_process.isra.4.part.5+0x640/0x1a80
softirqs last disabled at (0): [<0000000000000000>] 0x0
CPU: 73 PID: 1038 Comm: scsi_eh_1 Not tainted 5.2.0-rc6-le_nv2_aikATfstn1-p1 #634
Call Trace:
[c000003d064cef50] [c000000000c8e6c4] dump_stack+0xe8/0x164 (unreliable)
[c000003d064cefa0] [c00000000014ed78] ___might_sleep+0x2f8/0x310
[c000003d064cf020] [c0000000003ca084] __alloc_pages_nodemask+0x2a4/0x1560
[c000003d064cf220] [c0000000000c2530] pnv_alloc_tce_level.isra.0+0x90/0x130
[c000003d064cf290] [c0000000000c2888] pnv_tce+0x128/0x3b0
[c000003d064cf360] [c0000000000c2c00] pnv_tce_build+0xb0/0xf0
[c000003d064cf3c0] [c0000000000bbd9c] pnv_ioda2_tce_build+0x3c/0xb0
[c000003d064cf400] [c00000000004cfe0] ppc_iommu_map_sg+0x210/0x550
[c000003d064cf510] [c00000000004b7a4] dma_iommu_map_sg+0x74/0xb0
[c000003d064cf530] [c000000000863944] ata_qc_issue+0x134/0x470
[c000003d064cf5b0] [c000000000863ec4] ata_exec_internal_sg+0x244/0x5c0
[c000003d064cf700] [c0000000008642d0] ata_exec_internal+0x90/0xe0
[c000003d064cf780] [c0000000008650ac] ata_dev_read_id+0x2ec/0x640
[c000003d064cf8d0] [c000000000878e28] ata_eh_recover+0x948/0x16d0
[c000003d064cfa10] [c00000000087d760] sata_pmp_error_handler+0x480/0xbf0
[c000003d064cfbc0] [c000000000884624] ahci_error_handler+0x74/0xe0
[c000003d064cfbf0] [c000000000879fa8] ata_scsi_port_error_handler+0x2d8/0x7c0
[c000003d064cfca0] [c00000000087a544] ata_scsi_error+0xb4/0x100
[c000003d064cfd00] [c000000000802450] scsi_error_handler+0x120/0x510
[c000003d064cfdb0] [c000000000140c48] kthread+0x1b8/0x1c0
[c000003d064cfe20] [c00000000000bd8c] ret_from_kernel_thread+0x5c/0x70
ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
irq event stamp: 2305
========================================================
hardirqs last enabled at (2305): [<c00000000000e4c8>] fast_exc_return_irq+0x28/0x34
hardirqs last disabled at (2303): [<c000000000cb9fd0>] __do_softirq+0x4a0/0x654
WARNING: possible irq lock inversion dependency detected
5.2.0-rc6-le_nv2_aikATfstn1-p1 #634 Tainted: G W
softirqs last enabled at (2304): [<c000000000cba054>] __do_softirq+0x524/0x654
softirqs last disabled at (2297): [<c00000000010f278>] irq_exit+0x128/0x180
--------------------------------------------------------
swapper/0/0 just changed the state of lock:
0000000006cf56a6 (&(&host->lock)->rlock){-...}, at: ahci_single_level_irq_intr+0xac/0x120
but this lock took another, HARDIRQ-unsafe lock in the past:
(fs_reclaim){+.+.}
and interrupts could create inverse lock ordering between them.
other info that might help us debug this:
Possible interrupt unsafe locking scenario:
CPU0 CPU1
---- ----
lock(fs_reclaim);
local_irq_disable();
lock(&(&host->lock)->rlock);
lock(fs_reclaim);
<Interrupt>
lock(&(&host->lock)->rlock);
*** DEADLOCK ***
no locks held by swapper/0/0.
the shortest dependencies between 2nd lock and 1st lock:
-> (fs_reclaim){+.+.} ops: 167579 {
HARDIRQ-ON-W at:
lock_acquire+0xf8/0x2a0
fs_reclaim_acquire.part.23+0x44/0x60
kmem_cache_alloc_node_trace+0x80/0x590
alloc_desc+0x64/0x270
__irq_alloc_descs+0x2e4/0x3a0
irq_domain_alloc_descs+0xb0/0x150
irq_create_mapping+0x168/0x2c0
xics_smp_probe+0x2c/0x98
pnv_smp_probe+0x40/0x9c
smp_prepare_cpus+0x524/0x6c4
kernel_init_freeable+0x1b4/0x650
kernel_init+0x2c/0x148
ret_from_kernel_thread+0x5c/0x70
SOFTIRQ-ON-W at:
lock_acquire+0xf8/0x2a0
fs_reclaim_acquire.part.23+0x44/0x60
kmem_cache_alloc_node_trace+0x80/0x590
alloc_desc+0x64/0x270
__irq_alloc_descs+0x2e4/0x3a0
irq_domain_alloc_descs+0xb0/0x150
irq_create_mapping+0x168/0x2c0
xics_smp_probe+0x2c/0x98
pnv_smp_probe+0x40/0x9c
smp_prepare_cpus+0x524/0x6c4
kernel_init_freeable+0x1b4/0x650
kernel_init+0x2c/0x148
ret_from_kernel_thread+0x5c/0x70
INITIAL USE at:
lock_acquire+0xf8/0x2a0
fs_reclaim_acquire.part.23+0x44/0x60
kmem_cache_alloc_node_trace+0x80/0x590
alloc_desc+0x64/0x270
__irq_alloc_descs+0x2e4/0x3a0
irq_domain_alloc_descs+0xb0/0x150
irq_create_mapping+0x168/0x2c0
xics_smp_probe+0x2c/0x98
pnv_smp_probe+0x40/0x9c
smp_prepare_cpus+0x524/0x6c4
kernel_init_freeable+0x1b4/0x650
kernel_init+0x2c/0x148
ret_from_kernel_thread+0x5c/0x70
}
===
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: Alistair Popple <alistair@popple.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190718051139.74787-4-aik@ozlabs.ru
POWER8 and newer support a bypass mode which maps all host memory to
PCI buses so an IOMMU table is not always required. However if we fail to
create such a table, the DMA setup fails and the kernel does not boot.
This skips the 32bit DMA setup check if the bypass is selected.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190718051139.74787-3-aik@ozlabs.ru
pnv_tce() returns a pointer to a TCE entry and originally a TCE table
would be pre-allocated. For the default case of 2GB window the table
needs only a single level and that is fine. However if more levels are
requested, it is possible to get a race when 2 threads want a pointer
to a TCE entry from the same page of TCEs.
This adds cmpxchg to handle the race. Note that once TCE is non-zero,
it cannot become zero again.
Fixes: a68bd1267b ("powerpc/powernv/ioda: Allocate indirect TCE levels on demand")
CC: stable@vger.kernel.org # v4.19+
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190718051139.74787-2-aik@ozlabs.ru
The calls to arch_add_memory()/arch_remove_memory() are always made
with the read-side cpu_hotplug_lock acquired via memory_hotplug_begin().
On pSeries, arch_add_memory()/arch_remove_memory() eventually call
resize_hpt() which in turn calls stop_machine() which acquires the
read-side cpu_hotplug_lock again, thereby resulting in the recursive
acquisition of this lock.
In the absence of CONFIG_PROVE_LOCKING, we hadn't observed a system
lockup during a memory hotplug operation because cpus_read_lock() is a
per-cpu rwsem read, which, in the fast-path (in the absence of the
writer, which in our case is a CPU-hotplug operation) simply
increments the read_count on the semaphore. Thus a recursive read in
the fast-path doesn't cause any problems.
However, we can hit this problem in practice if there is a concurrent
CPU-Hotplug operation in progress which is waiting to acquire the
write-side of the lock. This will cause the second recursive read to
block until the writer finishes. While the writer is blocked since the
first read holds the lock. Thus both the reader as well as the writers
fail to make any progress thereby blocking both CPU-Hotplug as well as
Memory Hotplug operations.
Memory-Hotplug CPU-Hotplug
CPU 0 CPU 1
------ ------
1. down_read(cpu_hotplug_lock.rw_sem)
[memory_hotplug_begin]
2. down_write(cpu_hotplug_lock.rw_sem)
[cpu_up/cpu_down]
3. down_read(cpu_hotplug_lock.rw_sem)
[stop_machine()]
Lockdep complains as follows in these code-paths.
swapper/0/1 is trying to acquire lock:
(____ptrval____) (cpu_hotplug_lock.rw_sem){++++}, at: stop_machine+0x2c/0x60
but task is already holding lock:
(____ptrval____) (cpu_hotplug_lock.rw_sem){++++}, at: mem_hotplug_begin+0x20/0x50
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(cpu_hotplug_lock.rw_sem);
lock(cpu_hotplug_lock.rw_sem);
*** DEADLOCK ***
May be due to missing lock nesting notation
3 locks held by swapper/0/1:
#0: (____ptrval____) (&dev->mutex){....}, at: __driver_attach+0x12c/0x1b0
#1: (____ptrval____) (cpu_hotplug_lock.rw_sem){++++}, at: mem_hotplug_begin+0x20/0x50
#2: (____ptrval____) (mem_hotplug_lock.rw_sem){++++}, at: percpu_down_write+0x54/0x1a0
stack backtrace:
CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.0.0-rc5-58373-gbc99402235f3-dirty #166
Call Trace:
dump_stack+0xe8/0x164 (unreliable)
__lock_acquire+0x1110/0x1c70
lock_acquire+0x240/0x290
cpus_read_lock+0x64/0xf0
stop_machine+0x2c/0x60
pseries_lpar_resize_hpt+0x19c/0x2c0
resize_hpt_for_hotplug+0x70/0xd0
arch_add_memory+0x58/0xfc
devm_memremap_pages+0x5e8/0x8f0
pmem_attach_disk+0x764/0x830
nvdimm_bus_probe+0x118/0x240
really_probe+0x230/0x4b0
driver_probe_device+0x16c/0x1e0
__driver_attach+0x148/0x1b0
bus_for_each_dev+0x90/0x130
driver_attach+0x34/0x50
bus_add_driver+0x1a8/0x360
driver_register+0x108/0x170
__nd_driver_register+0xd0/0xf0
nd_pmem_driver_init+0x34/0x48
do_one_initcall+0x1e0/0x45c
kernel_init_freeable+0x540/0x64c
kernel_init+0x2c/0x160
ret_from_kernel_thread+0x5c/0x68
Fix this issue by
1) Requiring all the calls to pseries_lpar_resize_hpt() be made
with cpu_hotplug_lock held.
2) In pseries_lpar_resize_hpt() invoke stop_machine_cpuslocked()
as a consequence of 1)
3) To satisfy 1), in hpt_order_set(), call mmu_hash_ops.resize_hpt()
with cpu_hotplug_lock held.
Fixes: dbcf929c00 ("powerpc/pseries: Add support for hash table resizing")
Cc: stable@vger.kernel.org # v4.11+
Reported-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/1557906352-29048-1-git-send-email-ego@linux.vnet.ibm.com
Testing has revealed the existence of a race condition where a XIVE
interrupt being shut down can be in one of the XIVE interrupt queues
(of which there are up to 8 per CPU, one for each priority) at the
point where free_irq() is called. If this happens, can return an
interrupt number which has been shut down. This can lead to various
symptoms:
- irq_to_desc(irq) can be NULL. In this case, no end-of-interrupt
function gets called, resulting in the CPU's elevated interrupt
priority (numerically lowered CPPR) never gets reset. That then
means that the CPU stops processing interrupts, causing device
timeouts and other errors in various device drivers.
- The irq descriptor or related data structures can be in the process
of being freed as the interrupt code is using them. This typically
leads to crashes due to bad pointer dereferences.
This race is basically what commit 62e0468650 ("genirq: Add optional
hardware synchronization for shutdown", 2019-06-28) is intended to
fix, given a get_irqchip_state() method for the interrupt controller
being used. It works by polling the interrupt controller when an
interrupt is being freed until the controller says it is not pending.
With XIVE, the PQ bits of the interrupt source indicate the state of
the interrupt source, and in particular the P bit goes from 0 to 1 at
the point where the hardware writes an entry into the interrupt queue
that this interrupt is directed towards. Normally, the code will then
process the interrupt and do an end-of-interrupt (EOI) operation which
will reset PQ to 00 (assuming another interrupt hasn't been generated
in the meantime). However, there are situations where the code resets
P even though a queue entry exists (for example, by setting PQ to 01,
which disables the interrupt source), and also situations where the
code leaves P at 1 after removing the queue entry (for example, this
is done for escalation interrupts so they cannot fire again until
they are explicitly re-enabled).
The code already has a 'saved_p' flag for the interrupt source which
indicates that a queue entry exists, although it isn't maintained
consistently. This patch adds a 'stale_p' flag to indicate that
P has been left at 1 after processing a queue entry, and adds code
to set and clear saved_p and stale_p as necessary to maintain a
consistent indication of whether a queue entry may or may not exist.
With this, we can implement xive_get_irqchip_state() by looking at
stale_p, saved_p and the ESB PQ bits for the interrupt.
There is some additional code to handle escalation interrupts
properly; because they are enabled and disabled in KVM assembly code,
which does not have access to the xive_irq_data struct for the
escalation interrupt. Hence, stale_p may be incorrect when the
escalation interrupt is freed in kvmppc_xive_{,native_}cleanup_vcpu().
Fortunately, we can fix it up by looking at vcpu->arch.xive_esc_on,
with some careful attention to barriers in order to ensure the correct
result if xive_esc_irq() races with kvmppc_xive_cleanup_vcpu().
Finally, this adds code to make noise on the console (pr_crit and
WARN_ON(1)) if we find an interrupt queue entry for an interrupt
which does not have a descriptor. While this won't catch the race
reliably, if it does get triggered it will be an indication that
the race is occurring and needs to be debugged.
Fixes: 243e25112d ("powerpc/xive: Native exploitation of the XIVE interrupt controller")
Cc: stable@vger.kernel.org # v4.12+
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190813100648.GE9567@blackberry
At present, when running a guest on POWER9 using HV KVM but not using
an in-kernel interrupt controller (XICS or XIVE), for example if QEMU
is run with the kernel_irqchip=off option, the guest entry code goes
ahead and tries to load the guest context into the XIVE hardware, even
though no context has been set up.
To fix this, we check that the "CAM word" is non-zero before pushing
it to the hardware. The CAM word is initialized to a non-zero value
in kvmppc_xive_connect_vcpu() and kvmppc_xive_native_connect_vcpu(),
and is now cleared in kvmppc_xive_{,native_}cleanup_vcpu.
Fixes: 5af5099385 ("KVM: PPC: Book3S HV: Native usage of the XIVE interrupt controller")
Cc: stable@vger.kernel.org # v4.12+
Reported-by: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190813100100.GC9567@blackberry
Escalation interrupts are interrupts sent to the host by the XIVE
hardware when it has an interrupt to deliver to a guest VCPU but that
VCPU is not running anywhere in the system. Hence we disable the
escalation interrupt for the VCPU being run when we enter the guest
and re-enable it when the guest does an H_CEDE hypercall indicating
it is idle.
It is possible that an escalation interrupt gets generated just as we
are entering the guest. In that case the escalation interrupt may be
using a queue entry in one of the interrupt queues, and that queue
entry may not have been processed when the guest exits with an H_CEDE.
The existing entry code detects this situation and does not clear the
vcpu->arch.xive_esc_on flag as an indication that there is a pending
queue entry (if the queue entry gets processed, xive_esc_irq() will
clear the flag). There is a comment in the code saying that if the
flag is still set on H_CEDE, we have to abort the cede rather than
re-enabling the escalation interrupt, lest we end up with two
occurrences of the escalation interrupt in the interrupt queue.
However, the exit code doesn't do that; it aborts the cede in the sense
that vcpu->arch.ceded gets cleared, but it still enables the escalation
interrupt by setting the source's PQ bits to 00. Instead we need to
set the PQ bits to 10, indicating that an interrupt has been triggered.
We also need to avoid setting vcpu->arch.xive_esc_on in this case
(i.e. vcpu->arch.xive_esc_on seen to be set on H_CEDE) because
xive_esc_irq() will run at some point and clear it, and if we race with
that we may end up with an incorrect result (i.e. xive_esc_on set when
the escalation interrupt has just been handled).
It is extremely unlikely that having two queue entries would cause
observable problems; theoretically it could cause queue overflow, but
the CPU would have to have thousands of interrupts targetted to it for
that to be possible. However, this fix will also make it possible to
determine accurately whether there is an unhandled escalation
interrupt in the queue, which will be needed by the following patch.
Fixes: 9b9b13a6d1 ("KVM: PPC: Book3S HV: Keep XIVE escalation interrupt masked unless ceded")
Cc: stable@vger.kernel.org # v4.16+
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190813100349.GD9567@blackberry
When a vCPU is brought done, the XIVE VP (Virtual Processor) is first
disabled and then the event notification queues are freed. When freeing
the queues, we check for possible escalation interrupts and free them
also.
But when a XIVE VP is disabled, the underlying XIVE ENDs also are
disabled in OPAL. When an END (Event Notification Descriptor) is
disabled, its ESB pages (ESn and ESe) are disabled and loads return all
1s. Which means that any access on the ESB page of the escalation
interrupt will return invalid values.
When an interrupt is freed, the shutdown handler computes a 'saved_p'
field from the value returned by a load in xive_do_source_set_mask().
This value is incorrect for escalation interrupts for the reason
described above.
This has no impact on Linux/KVM today because we don't make use of it
but we will introduce in future changes a xive_get_irqchip_state()
handler. This handler will use the 'saved_p' field to return the state
of an interrupt and 'saved_p' being incorrect, softlockup will occur.
Fix the vCPU cleanup sequence by first freeing the escalation interrupts
if any, then disable the XIVE VP and last free the queues.
Fixes: 90c73795af ("KVM: PPC: Book3S HV: Add a new KVM device for the XIVE native exploitation mode")
Fixes: 5af5099385 ("KVM: PPC: Book3S HV: Native usage of the XIVE interrupt controller")
Cc: stable@vger.kernel.org # v4.12+
Signed-off-by: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190806172538.5087-1-clg@kaod.org
There is no need to use GFP_ATOMIC here. GFP_KERNEL should be enough.
GFP_KERNEL is also already used for another allocation just a few lines
below.
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Reviewed-by: Greg Kurz <groug@kaod.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/85d5d247ce753befd6aa63c473f7823de6520ccd.1564647619.git.christophe.jaillet@wanadoo.fr
Commit ebb9d30a6a ("powerpc/mm: any thread in one core can be the
first to setup TLB1") removed the need to know the cpu_id in
early_init_this_mmu(), but the call to smp_processor_id() which was
marked __maybe_used remained.
Since commit ed1cd6deb0 ("powerpc: Activate CONFIG_THREAD_INFO_IN_TASK")
thread_info cannot be reached before MMU is properly set up.
Drop this stale call to smp_processor_id() which makes SMP hang when
CONFIG_PREEMPT is set.
Fixes: ebb9d30a6a ("powerpc/mm: any thread in one core can be the first to setup TLB1")
Fixes: ed1cd6deb0 ("powerpc: Activate CONFIG_THREAD_INFO_IN_TASK")
Cc: stable@vger.kernel.org # v5.1+
Reported-by: Chris Packham <Chris.Packham@alliedtelesis.co.nz>
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Tested-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/bef479514f4c08329fa649f67735df8918bc0976.1565268248.git.christophe.leroy@c-s.fr
Balbir Singh
Reza Arbab
Santosh Sivaraj
Nathan Lynch
Christophe Leroy
Nicholas Piggin
Cédric Le Goater
Alexey Kardashevskiy
Gautham R. Shenoy
Michael Ellerman
Paul Mackerras
Christophe JAILLET