LinuxPPC-Dev Archive on lore.kernel.org
 help / color / mirror / Atom feed

* Re: switch the remaining architectures to use generic GUP v3
From: Jason Gunthorpe @ 2019-06-21 14:43 UTC (permalink / raw)
  To: Christoph Hellwig
  Cc: x86, Rich Felker, Yoshinori Sato, linux-sh, James Hogan,
	linuxppc-dev, Khalid Aziz, Nicholas Piggin, linux-mips, linux-mm,
	Paul Burton, Paul Mackerras, Andrey Konovalov, sparclinux,
	Linus Torvalds, David S. Miller, linux-kernel
In-Reply-To: <20190611144102.8848-1-hch@lst.de>

On Tue, Jun 11, 2019 at 04:40:46PM +0200, Christoph Hellwig wrote:
> Hi Linus and maintainers,
> 
> below is a series to switch mips, sh and sparc64 to use the generic
> GUP code so that we only have one codebase to touch for further
> improvements to this code.  I don't have hardware for any of these
> architectures, and generally no clue about their page table
> management, so handle with care.

I like this series, ther are alot of people talking about work for GUP
and this will make any of that so much easier to do.

Jason

^ permalink raw reply

* Re: [PATCH 7/7] powerpc/kprobes: Allow probing on any ftrace address
From: Masami Hiramatsu @ 2019-06-21 14:50 UTC (permalink / raw)
  To: Naveen N. Rao
  Cc: linux-kernel, Nicholas Piggin, Steven Rostedt, linuxppc-dev,
	Ingo Molnar
In-Reply-To: <d26f5467577ff0aeecea55e7035ea64e303bdf17.1560868106.git.naveen.n.rao@linux.vnet.ibm.com>

On Tue, 18 Jun 2019 20:17:06 +0530
"Naveen N. Rao" <naveen.n.rao@linux.vnet.ibm.com> wrote:

> With KPROBES_ON_FTRACE, kprobe is allowed to be inserted on instructions
> that branch to _mcount (referred to as ftrace location). With
> -mprofile-kernel, we now include the preceding 'mflr r0' as being part
> of the ftrace location.
> 
> However, by default, probing on an instruction that is not actually the
> branch to _mcount() is prohibited, as that is considered to not be at an
> instruction boundary. This is not the case on powerpc, so allow the same
> by overriding arch_check_ftrace_location()
> 
> In addition, we update kprobe_ftrace_handler() to detect this scenarios
> and to pass the proper nip to the pre and post probe handlers.
> 
> Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
> ---
>  arch/powerpc/kernel/kprobes-ftrace.c | 30 ++++++++++++++++++++++++++++
>  1 file changed, 30 insertions(+)
> 
> diff --git a/arch/powerpc/kernel/kprobes-ftrace.c b/arch/powerpc/kernel/kprobes-ftrace.c
> index 972cb28174b2..6a0bd3c16cb6 100644
> --- a/arch/powerpc/kernel/kprobes-ftrace.c
> +++ b/arch/powerpc/kernel/kprobes-ftrace.c
> @@ -12,14 +12,34 @@
>  #include <linux/preempt.h>
>  #include <linux/ftrace.h>
>  
> +/*
> + * With -mprofile-kernel, we patch two instructions -- the branch to _mcount
> + * as well as the preceding 'mflr r0'. Both these instructions are claimed
> + * by ftrace and we should allow probing on either instruction.
> + */
> +int arch_check_ftrace_location(struct kprobe *p)
> +{
> +	if (ftrace_location((unsigned long)p->addr))
> +		p->flags |= KPROBE_FLAG_FTRACE;
> +	return 0;
> +}
> +
>  /* Ftrace callback handler for kprobes */
>  void kprobe_ftrace_handler(unsigned long nip, unsigned long parent_nip,
>  			   struct ftrace_ops *ops, struct pt_regs *regs)
>  {
>  	struct kprobe *p;
> +	int mflr_kprobe = 0;
>  	struct kprobe_ctlblk *kcb;
>  
>  	p = get_kprobe((kprobe_opcode_t *)nip);
> +	if (unlikely(!p)) {

Hmm, is this really unlikely? If we put a kprobe on the second instruction address,
we will see p == NULL always.

> +		p = get_kprobe((kprobe_opcode_t *)(nip - MCOUNT_INSN_SIZE));
> +		if (!p)

Here will be unlikely, because we can not find kprobe at both of nip and
nip - MCOUNT_INSN_SIZE.

> +			return;
> +		mflr_kprobe = 1;
> +	}
> +
>  	if (unlikely(!p) || kprobe_disabled(p))

"unlikely(!p)" is not needed here.

Thank you,

>  		return;
>  
> @@ -33,6 +53,9 @@ void kprobe_ftrace_handler(unsigned long nip, unsigned long parent_nip,
>  		 */
>  		regs->nip -= MCOUNT_INSN_SIZE;
>  
> +		if (mflr_kprobe)
> +			regs->nip -= MCOUNT_INSN_SIZE;
> +
>  		__this_cpu_write(current_kprobe, p);
>  		kcb->kprobe_status = KPROBE_HIT_ACTIVE;
>  		if (!p->pre_handler || !p->pre_handler(p, regs)) {
> @@ -45,6 +68,8 @@ void kprobe_ftrace_handler(unsigned long nip, unsigned long parent_nip,
>  				kcb->kprobe_status = KPROBE_HIT_SSDONE;
>  				p->post_handler(p, regs, 0);
>  			}
> +			if (mflr_kprobe)
> +				regs->nip += MCOUNT_INSN_SIZE;
>  		}
>  		/*
>  		 * If pre_handler returns !0, it changes regs->nip. We have to
> @@ -57,6 +82,11 @@ NOKPROBE_SYMBOL(kprobe_ftrace_handler);
>  
>  int arch_prepare_kprobe_ftrace(struct kprobe *p)
>  {
> +	if ((unsigned long)p->addr & 0x03) {
> +		printk("Attempt to register kprobe at an unaligned address\n");
> +		return -EILSEQ;
> +	}
> +
>  	p->ainsn.insn = NULL;
>  	p->ainsn.boostable = -1;
>  	return 0;
> -- 
> 2.22.0
> 


-- 
Masami Hiramatsu <mhiramat@kernel.org>

^ permalink raw reply

* Re: [PATCH v3 0/6] mm: Further memory block device cleanups
From: Qian Cai @ 2019-06-21 15:15 UTC (permalink / raw)
  To: David Hildenbrand, linux-kernel
  Cc: Oscar Salvador, Michal Hocko, Rafael J. Wysocki, Wei Yang,
	Keith Busch, linux-mm, Arun KS, Rashmica Gupta, Thomas Gleixner,
	Stephen Rothwell, Michael Neuling, Baoquan He, Pavel Tatashin,
	linux-acpi, Len Brown, Pavel Tatashin, Pavel Tatashin,
	Anshuman Khandual, mike.travis@hpe.com, linuxppc-dev,
	Mike Rapoport, Dan Williams, Vlastimil Babka, Oscar Salvador,
	Juergen Gross, Andrew Banman, Greg Kroah-Hartman,
	Rafael J. Wysocki, Johannes Weiner, Paul Mackerras, Andrew Morton,
	Mel Gorman
In-Reply-To: <20190620183139.4352-1-david@redhat.com>

On Thu, 2019-06-20 at 20:31 +0200, David Hildenbrand wrote:
> @Andrew: Only patch 1, 4 and 6 changed compared to v1.
> 
> Some further cleanups around memory block devices. Especially, clean up
> and simplify walk_memory_range(). Including some other minor cleanups.
> 
> Compiled + tested on x86 with DIMMs under QEMU. Compile-tested on ppc64.
> 
> v2 -> v3:
> - "mm/memory_hotplug: Rename walk_memory_range() and pass start+size .."
> -- Avoid warning on ppc.
> - "drivers/base/memory.c: Get rid of find_memory_block_hinted()"
> -- Fixup a comment regarding hinted devices.
> 
> v1 -> v2:
> - "mm: Section numbers use the type "unsigned long""
> -- "unsigned long i" -> "unsigned long nr", in one case -> "int i"
> - "drivers/base/memory.c: Get rid of find_memory_block_hinted("
> -- Fix compilation error
> -- Get rid of the "hint" parameter completely
> 
> David Hildenbrand (6):
>   mm: Section numbers use the type "unsigned long"
>   drivers/base/memory: Use "unsigned long" for block ids
>   mm: Make register_mem_sect_under_node() static
>   mm/memory_hotplug: Rename walk_memory_range() and pass start+size
>     instead of pfns
>   mm/memory_hotplug: Move and simplify walk_memory_blocks()
>   drivers/base/memory.c: Get rid of find_memory_block_hinted()
> 
>  arch/powerpc/platforms/powernv/memtrace.c |  23 ++---
>  drivers/acpi/acpi_memhotplug.c            |  19 +---
>  drivers/base/memory.c                     | 120 +++++++++++++---------
>  drivers/base/node.c                       |   8 +-
>  include/linux/memory.h                    |   5 +-
>  include/linux/memory_hotplug.h            |   2 -
>  include/linux/mmzone.h                    |   4 +-
>  include/linux/node.h                      |   7 --
>  mm/memory_hotplug.c                       |  57 +---------
>  mm/sparse.c                               |  12 +--
>  10 files changed, 106 insertions(+), 151 deletions(-)
> 

This series causes a few machines are unable to boot triggering endless soft
lockups. Reverted those commits fixed the issue.

97f4217d1da0 Revert "mm/memory_hotplug: rename walk_memory_range() and pass
start+size instead of pfns"
c608eebf33c6 Revert "mm-memory_hotplug-rename-walk_memory_range-and-pass-
startsize-instead-of-pfns-fix"
34b5e4ab7558 Revert "mm/memory_hotplug: move and simplify walk_memory_blocks()"
59a9f3eec5d1 Revert "drivers/base/memory.c: Get rid of
find_memory_block_hinted()"
5cfcd52288b6 Revert "drivers-base-memoryc-get-rid-of-find_memory_block_hinted-
v3"

[    4.582081][    T1] ACPI FADT declares the system doesn't support PCIe ASPM,
so disable it
[    4.590405][    T1] ACPI: bus type PCI registered
[    4.592908][    T1] PCI: MMCONFIG for domain 0000 [bus 00-ff] at [mem
0x80000000-0x8fffffff] (base 0x80000000)
[    4.601860][    T1] PCI: MMCONFIG at [mem 0x80000000-0x8fffffff] reserved in
E820
[    4.601860][    T1] PCI: Using configuration type 1 for base access
[   28.661336][   C16] watchdog: BUG: soft lockup - CPU#16 stuck for 22s!
[swapper/0:1]
[   28.671351][   C16] Modules linked in:
[   28.671354][   C16] CPU: 16 PID: 1 Comm: swapper/0 Not tainted 5.2.0-rc5-
next-20190621+ #1
[   28.681366][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385
Gen10, BIOS A40 03/09/2018
[   28.691334][   C16] RIP: 0010:_raw_spin_unlock_irqrestore+0x2f/0x40
[   28.701334][   C16] Code: 55 48 89 e5 41 54 49 89 f4 be 01 00 00 00 53 48 8b
55 08 48 89 fb 48 8d 7f 18 e8 4c 89 7d ff 48 89 df e8 94 f9 7d ff 41 54 9d <65>
ff 0d c2 44 8d 48 5b 41 5c 5d c3 0f 1f 44 00 00 0f 1f 44 00 00
[   28.711354][   C16] RSP: 0018:ffff888205b27bf8 EFLAGS: 00000246 ORIG_RAX:
ffffffffffffff13
[   28.721372][   C16] RAX: 0000000000000000 RBX: ffff8882053d6138 RCX:
ffffffffb6f2a3b8
[   28.731371][   C16] RDX: 1ffff11040a7ac27 RSI: dffffc0000000000 RDI:
ffff8882053d6138
[   28.741371][   C16] RBP: ffff888205b27c08 R08: ffffed1040a7ac28 R09:
ffffed1040a7ac27
[   28.751334][   C16] R10: ffffed1040a7ac27 R11: ffff8882053d613b R12:
0000000000000246
[   28.751370][   C16] R13: ffff888205b27c98 R14: ffff8884504d0a20 R15:
0000000000000000
[   28.761368][   C16] FS:  0000000000000000(0000) GS:ffff888454500000(0000)
knlGS:0000000000000000
[   28.771373][   C16] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   28.781334][   C16] CR2: 0000000000000000 CR3: 00000007c9012000 CR4:
00000000001406a0
[   28.791333][   C16] Call Trace:
[   28.791374][   C16]  klist_next+0xd8/0x1c0
[   28.791374][   C16]  subsys_find_device_by_id+0x13b/0x1f0
[   28.801334][   C16]  ? bus_find_device_by_name+0x20/0x20
[   28.801370][   C16]  ? kobject_put+0x23/0x250
[   28.811333][   C16]  walk_memory_blocks+0x6c/0xb8
[   28.811353][   C16]  ? write_policy_show+0x40/0x40
[   28.821334][   C16]  link_mem_sections+0x7e/0xa0
[   28.821369][   C16]  ? unregister_memory_block_under_nodes+0x210/0x210
[   28.831353][   C16]  ? __register_one_node+0x3bd/0x600
[   28.831353][   C16]  topology_init+0xbf/0x126
[   28.841364][   C16]  ? enable_cpu0_hotplug+0x1a/0x1a
[   28.841368][   C16]  do_one_initcall+0xfe/0x45a
[   28.851334][   C16]  ? initcall_blacklisted+0x150/0x150
[   28.851353][   C16]  ? kasan_check_write+0x14/0x20
[   28.861333][   C16]  ? up_write+0x75/0x140
[   28.861369][   C16]  kernel_init_freeable+0x619/0x6ac
[   28.871333][   C16]  ? rest_init+0x188/0x188
[   28.871353][   C16]  kernel_init+0x11/0x138
[   28.881363][   C16]  ? rest_init+0x188/0x188
[   28.881363][   C16]  ret_from_fork+0x22/0x40
[   56.661336][   C16] watchdog: BUG: soft lockup - CPU#16 stuck for 22s!
[swapper/0:1]
[   56.671352][   C16] Modules linked in:
[   56.671354][   C16] CPU: 16 PID: 1 Comm: swapper/0 Tainted:
G             L    5.2.0-rc5-next-20190621+ #1
[   56.681357][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385
Gen10, BIOS A40 03/09/2018
[   56.691356][   C16] RIP: 0010:subsys_find_device_by_id+0x168/0x1f0
[   56.701334][   C16] Code: 48 85 c0 74 3e 48 8d 78 58 e8 14 77 ca ff 4d 8b 7e
58 4d 85 ff 74 2c 49 8d bf a0 03 00 00 e8 bf 75 ca ff 45 39 a7 a0 03 00 00 <75>
c9 4c 89 ff e8 0e 89 ff ff 48 85 c0 74 bc 48 89 df e8 21 3b 24
[   56.721333][   C16] RSP: 0018:ffff888205b27c68 EFLAGS: 00000287 ORIG_RAX:
ffffffffffffff13
[   56.721370][   C16] RAX: 0000000000000000 RBX: ffff888205b27c90 RCX:
ffffffffb74c9dc1
[   56.731370][   C16] RDX: 0000000000000003 RSI: dffffc0000000000 RDI:
ffff8888774ec3e0
[   56.741371][   C16] RBP: ffff888205b27cf8 R08: ffffed1040a7ac28 R09:
ffffed1040a7ac27
[   56.751335][   C16] R10: ffffed1040a7ac27 R11: ffff8882053d613b R12:
0000000000085c1b
[   56.761334][   C16] R13: 1ffff11040b64f8e R14: ffff888450de4a20 R15:
ffff8888774ec040
[   56.761372][   C16] FS:  0000000000000000(0000) GS:ffff888454500000(0000)
knlGS:0000000000000000
[   56.771374][   C16] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   56.781370][   C16] CR2: 0000000000000000 CR3: 00000007c9012000 CR4:
00000000001406a0
[   56.791373][   C16] Call Trace:
[   56.791373][   C16]  ? bus_find_device_by_name+0x20/0x20
[   56.801334][   C16]  ? kobject_put+0x23/0x250
[   56.801334][   C16]  walk_memory_blocks+0x6c/0xb8
[   56.811333][   C16]  ? write_policy_show+0x40/0x40
[   56.811353][   C16]  link_mem_sections+0x7e/0xa0
[   56.811353][   C16]  ? unregister_memory_block_under_nodes+0x210/0x210
[   56.821333][   C16]  ? __register_one_node+0x3bd/0x600
[   56.831333][   C16]  topology_init+0xbf/0x126
[   56.831355][   C16]  ? enable_cpu0_hotplug+0x1a/0x1a
[   56.841334][   C16]  do_one_initcall+0xfe/0x45a
[   56.841334][   C16]  ? initcall_blacklisted+0x150/0x150
[   56.851333][   C16]  ? kasan_check_write+0x14/0x20
[   56.851354][   C16]  ? up_write+0x75/0x140
[   56.861333][   C16]  kernel_init_freeable+0x619/0x6ac
[   56.861333][   C16]  ? rest_init+0x188/0x188
[   56.861369][   C16]  kernel_init+0x11/0x138
[   56.871333][   C16]  ? rest_init+0x188/0x188
[   56.871354][   C16]  ret_from_fork+0x22/0x40
[   64.601362][   C16] rcu: INFO: rcu_sched self-detected stall on CPU
[   64.611335][   C16] rcu: 	16-....: (5958 ticks this GP)
idle=37e/1/0x4000000000000002 softirq=27/27 fqs=3000 
[   64.621334][   C16] 	(t=6002 jiffies g=-1079 q=25)
[   64.621334][   C16] NMI backtrace for cpu 16
[   64.621374][   C16] CPU: 16 PID: 1 Comm: swapper/0 Tainted:
G             L    5.2.0-rc5-next-20190621+ #1
[   64.631372][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385
Gen10, BIOS A40 03/09/2018
[   64.641371][   C16] Call Trace:
[   64.651337][   C16]  <IRQ>
[   64.651376][   C16]  dump_stack+0x62/0x9a
[   64.651376][   C16]  nmi_cpu_backtrace.cold.0+0x2e/0x33
[   64.661337][   C16]  ? nmi_cpu_backtrace_handler+0x20/0x20
[   64.661337][   C16]  nmi_trigger_cpumask_backtrace+0x1a6/0x1b9
[   64.671353][   C16]  arch_trigger_cpumask_backtrace+0x19/0x20
[   64.681366][   C16]  rcu_dump_cpu_stacks+0x18b/0x1d6
[   64.681366][   C16]  rcu_sched_clock_irq.cold.64+0x368/0x791
[   64.691336][   C16]  ? kasan_check_read+0x11/0x20
[   64.691354][   C16]  ? __raise_softirq_irqoff+0x66/0x150
[   64.701336][   C16]  update_process_times+0x2f/0x60
[   64.701362][   C16]  tick_periodic+0x38/0xe0
[   64.711334][   C16]  tick_handle_periodic+0x2e/0x80
[   64.711353][   C16]  smp_apic_timer_interrupt+0xfb/0x370
[   64.721367][   C16]  apic_timer_interrupt+0xf/0x20
[   64.721367][   C16]  </IRQ>
[   64.721367][   C16] RIP: 0010:_raw_spin_unlock_irqrestore+0x2f/0x40
[   64.731370][   C16] Code: 55 48 89 e5 41 54 49 89 f4 be 01 00 00 00 53 

^ permalink raw reply

* Re: [PATCH v3 0/6] mm: Further memory block device cleanups
From: David Hildenbrand @ 2019-06-21 15:22 UTC (permalink / raw)
  To: Qian Cai, linux-kernel
  Cc: Oscar Salvador, Michal Hocko, Rafael J. Wysocki, Wei Yang,
	Keith Busch, linux-mm, Arun KS, Rashmica Gupta, Thomas Gleixner,
	Stephen Rothwell, Michael Neuling, Baoquan He, Pavel Tatashin,
	linux-acpi, Len Brown, Pavel Tatashin, Pavel Tatashin,
	Anshuman Khandual, mike.travis@hpe.com, linuxppc-dev,
	Mike Rapoport, Dan Williams, Vlastimil Babka, Oscar Salvador,
	Juergen Gross, Andrew Banman, Greg Kroah-Hartman,
	Rafael J. Wysocki, Johannes Weiner, Paul Mackerras, Andrew Morton,
	Mel Gorman
In-Reply-To: <1561130120.5154.47.camel@lca.pw>

On 21.06.19 17:15, Qian Cai wrote:
> On Thu, 2019-06-20 at 20:31 +0200, David Hildenbrand wrote:
>> @Andrew: Only patch 1, 4 and 6 changed compared to v1.
>>
>> Some further cleanups around memory block devices. Especially, clean up
>> and simplify walk_memory_range(). Including some other minor cleanups.
>>
>> Compiled + tested on x86 with DIMMs under QEMU. Compile-tested on ppc64.
>>
>> v2 -> v3:
>> - "mm/memory_hotplug: Rename walk_memory_range() and pass start+size .."
>> -- Avoid warning on ppc.
>> - "drivers/base/memory.c: Get rid of find_memory_block_hinted()"
>> -- Fixup a comment regarding hinted devices.
>>
>> v1 -> v2:
>> - "mm: Section numbers use the type "unsigned long""
>> -- "unsigned long i" -> "unsigned long nr", in one case -> "int i"
>> - "drivers/base/memory.c: Get rid of find_memory_block_hinted("
>> -- Fix compilation error
>> -- Get rid of the "hint" parameter completely
>>
>> David Hildenbrand (6):
>>   mm: Section numbers use the type "unsigned long"
>>   drivers/base/memory: Use "unsigned long" for block ids
>>   mm: Make register_mem_sect_under_node() static
>>   mm/memory_hotplug: Rename walk_memory_range() and pass start+size
>>     instead of pfns
>>   mm/memory_hotplug: Move and simplify walk_memory_blocks()
>>   drivers/base/memory.c: Get rid of find_memory_block_hinted()
>>
>>  arch/powerpc/platforms/powernv/memtrace.c |  23 ++---
>>  drivers/acpi/acpi_memhotplug.c            |  19 +---
>>  drivers/base/memory.c                     | 120 +++++++++++++---------
>>  drivers/base/node.c                       |   8 +-
>>  include/linux/memory.h                    |   5 +-
>>  include/linux/memory_hotplug.h            |   2 -
>>  include/linux/mmzone.h                    |   4 +-
>>  include/linux/node.h                      |   7 --
>>  mm/memory_hotplug.c                       |  57 +---------
>>  mm/sparse.c                               |  12 +--
>>  10 files changed, 106 insertions(+), 151 deletions(-)
>>
> 
> This series causes a few machines are unable to boot triggering endless soft
> lockups. Reverted those commits fixed the issue.
> 
> 97f4217d1da0 Revert "mm/memory_hotplug: rename walk_memory_range() and pass
> start+size instead of pfns"
> c608eebf33c6 Revert "mm-memory_hotplug-rename-walk_memory_range-and-pass-
> startsize-instead-of-pfns-fix"
> 34b5e4ab7558 Revert "mm/memory_hotplug: move and simplify walk_memory_blocks()"
> 59a9f3eec5d1 Revert "drivers/base/memory.c: Get rid of
> find_memory_block_hinted()"
> 5cfcd52288b6 Revert "drivers-base-memoryc-get-rid-of-find_memory_block_hinted-
> v3"
> 
> [    4.582081][    T1] ACPI FADT declares the system doesn't support PCIe ASPM,
> so disable it
> [    4.590405][    T1] ACPI: bus type PCI registered
> [    4.592908][    T1] PCI: MMCONFIG for domain 0000 [bus 00-ff] at [mem
> 0x80000000-0x8fffffff] (base 0x80000000)
> [    4.601860][    T1] PCI: MMCONFIG at [mem 0x80000000-0x8fffffff] reserved in
> E820
> [    4.601860][    T1] PCI: Using configuration type 1 for base access
> [   28.661336][   C16] watchdog: BUG: soft lockup - CPU#16 stuck for 22s!
> [swapper/0:1]
> [   28.671351][   C16] Modules linked in:
> [   28.671354][   C16] CPU: 16 PID: 1 Comm: swapper/0 Not tainted 5.2.0-rc5-
> next-20190621+ #1
> [   28.681366][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385
> Gen10, BIOS A40 03/09/2018
> [   28.691334][   C16] RIP: 0010:_raw_spin_unlock_irqrestore+0x2f/0x40
> [   28.701334][   C16] Code: 55 48 89 e5 41 54 49 89 f4 be 01 00 00 00 53 48 8b
> 55 08 48 89 fb 48 8d 7f 18 e8 4c 89 7d ff 48 89 df e8 94 f9 7d ff 41 54 9d <65>
> ff 0d c2 44 8d 48 5b 41 5c 5d c3 0f 1f 44 00 00 0f 1f 44 00 00
> [   28.711354][   C16] RSP: 0018:ffff888205b27bf8 EFLAGS: 00000246 ORIG_RAX:
> ffffffffffffff13
> [   28.721372][   C16] RAX: 0000000000000000 RBX: ffff8882053d6138 RCX:
> ffffffffb6f2a3b8
> [   28.731371][   C16] RDX: 1ffff11040a7ac27 RSI: dffffc0000000000 RDI:
> ffff8882053d6138
> [   28.741371][   C16] RBP: ffff888205b27c08 R08: ffffed1040a7ac28 R09:
> ffffed1040a7ac27
> [   28.751334][   C16] R10: ffffed1040a7ac27 R11: ffff8882053d613b R12:
> 0000000000000246
> [   28.751370][   C16] R13: ffff888205b27c98 R14: ffff8884504d0a20 R15:
> 0000000000000000
> [   28.761368][   C16] FS:  0000000000000000(0000) GS:ffff888454500000(0000)
> knlGS:0000000000000000
> [   28.771373][   C16] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [   28.781334][   C16] CR2: 0000000000000000 CR3: 00000007c9012000 CR4:
> 00000000001406a0
> [   28.791333][   C16] Call Trace:
> [   28.791374][   C16]  klist_next+0xd8/0x1c0
> [   28.791374][   C16]  subsys_find_device_by_id+0x13b/0x1f0
> [   28.801334][   C16]  ? bus_find_device_by_name+0x20/0x20
> [   28.801370][   C16]  ? kobject_put+0x23/0x250
> [   28.811333][   C16]  walk_memory_blocks+0x6c/0xb8
> [   28.811353][   C16]  ? write_policy_show+0x40/0x40
> [   28.821334][   C16]  link_mem_sections+0x7e/0xa0
> [   28.821369][   C16]  ? unregister_memory_block_under_nodes+0x210/0x210
> [   28.831353][   C16]  ? __register_one_node+0x3bd/0x600
> [   28.831353][   C16]  topology_init+0xbf/0x126
> [   28.841364][   C16]  ? enable_cpu0_hotplug+0x1a/0x1a
> [   28.841368][   C16]  do_one_initcall+0xfe/0x45a
> [   28.851334][   C16]  ? initcall_blacklisted+0x150/0x150
> [   28.851353][   C16]  ? kasan_check_write+0x14/0x20
> [   28.861333][   C16]  ? up_write+0x75/0x140
> [   28.861369][   C16]  kernel_init_freeable+0x619/0x6ac
> [   28.871333][   C16]  ? rest_init+0x188/0x188
> [   28.871353][   C16]  kernel_init+0x11/0x138
> [   28.881363][   C16]  ? rest_init+0x188/0x188
> [   28.881363][   C16]  ret_from_fork+0x22/0x40
> [   56.661336][   C16] watchdog: BUG: soft lockup - CPU#16 stuck for 22s!
> [swapper/0:1]
> [   56.671352][   C16] Modules linked in:
> [   56.671354][   C16] CPU: 16 PID: 1 Comm: swapper/0 Tainted:
> G             L    5.2.0-rc5-next-20190621+ #1
> [   56.681357][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385
> Gen10, BIOS A40 03/09/2018
> [   56.691356][   C16] RIP: 0010:subsys_find_device_by_id+0x168/0x1f0
> [   56.701334][   C16] Code: 48 85 c0 74 3e 48 8d 78 58 e8 14 77 ca ff 4d 8b 7e
> 58 4d 85 ff 74 2c 49 8d bf a0 03 00 00 e8 bf 75 ca ff 45 39 a7 a0 03 00 00 <75>
> c9 4c 89 ff e8 0e 89 ff ff 48 85 c0 74 bc 48 89 df e8 21 3b 24
> [   56.721333][   C16] RSP: 0018:ffff888205b27c68 EFLAGS: 00000287 ORIG_RAX:
> ffffffffffffff13
> [   56.721370][   C16] RAX: 0000000000000000 RBX: ffff888205b27c90 RCX:
> ffffffffb74c9dc1
> [   56.731370][   C16] RDX: 0000000000000003 RSI: dffffc0000000000 RDI:
> ffff8888774ec3e0
> [   56.741371][   C16] RBP: ffff888205b27cf8 R08: ffffed1040a7ac28 R09:
> ffffed1040a7ac27
> [   56.751335][   C16] R10: ffffed1040a7ac27 R11: ffff8882053d613b R12:
> 0000000000085c1b
> [   56.761334][   C16] R13: 1ffff11040b64f8e R14: ffff888450de4a20 R15:
> ffff8888774ec040
> [   56.761372][   C16] FS:  0000000000000000(0000) GS:ffff888454500000(0000)
> knlGS:0000000000000000
> [   56.771374][   C16] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [   56.781370][   C16] CR2: 0000000000000000 CR3: 00000007c9012000 CR4:
> 00000000001406a0
> [   56.791373][   C16] Call Trace:
> [   56.791373][   C16]  ? bus_find_device_by_name+0x20/0x20
> [   56.801334][   C16]  ? kobject_put+0x23/0x250
> [   56.801334][   C16]  walk_memory_blocks+0x6c/0xb8
> [   56.811333][   C16]  ? write_policy_show+0x40/0x40
> [   56.811353][   C16]  link_mem_sections+0x7e/0xa0
> [   56.811353][   C16]  ? unregister_memory_block_under_nodes+0x210/0x210
> [   56.821333][   C16]  ? __register_one_node+0x3bd/0x600
> [   56.831333][   C16]  topology_init+0xbf/0x126
> [   56.831355][   C16]  ? enable_cpu0_hotplug+0x1a/0x1a
> [   56.841334][   C16]  do_one_initcall+0xfe/0x45a
> [   56.841334][   C16]  ? initcall_blacklisted+0x150/0x150
> [   56.851333][   C16]  ? kasan_check_write+0x14/0x20
> [   56.851354][   C16]  ? up_write+0x75/0x140
> [   56.861333][   C16]  kernel_init_freeable+0x619/0x6ac
> [   56.861333][   C16]  ? rest_init+0x188/0x188
> [   56.861369][   C16]  kernel_init+0x11/0x138
> [   56.871333][   C16]  ? rest_init+0x188/0x188
> [   56.871354][   C16]  ret_from_fork+0x22/0x40
> [   64.601362][   C16] rcu: INFO: rcu_sched self-detected stall on CPU
> [   64.611335][   C16] rcu: 	16-....: (5958 ticks this GP)
> idle=37e/1/0x4000000000000002 softirq=27/27 fqs=3000 
> [   64.621334][   C16] 	(t=6002 jiffies g=-1079 q=25)
> [   64.621334][   C16] NMI backtrace for cpu 16
> [   64.621374][   C16] CPU: 16 PID: 1 Comm: swapper/0 Tainted:
> G             L    5.2.0-rc5-next-20190621+ #1
> [   64.631372][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385
> Gen10, BIOS A40 03/09/2018
> [   64.641371][   C16] Call Trace:
> [   64.651337][   C16]  <IRQ>
> [   64.651376][   C16]  dump_stack+0x62/0x9a
> [   64.651376][   C16]  nmi_cpu_backtrace.cold.0+0x2e/0x33
> [   64.661337][   C16]  ? nmi_cpu_backtrace_handler+0x20/0x20
> [   64.661337][   C16]  nmi_trigger_cpumask_backtrace+0x1a6/0x1b9
> [   64.671353][   C16]  arch_trigger_cpumask_backtrace+0x19/0x20
> [   64.681366][   C16]  rcu_dump_cpu_stacks+0x18b/0x1d6
> [   64.681366][   C16]  rcu_sched_clock_irq.cold.64+0x368/0x791
> [   64.691336][   C16]  ? kasan_check_read+0x11/0x20
> [   64.691354][   C16]  ? __raise_softirq_irqoff+0x66/0x150
> [   64.701336][   C16]  update_process_times+0x2f/0x60
> [   64.701362][   C16]  tick_periodic+0x38/0xe0
> [   64.711334][   C16]  tick_handle_periodic+0x2e/0x80
> [   64.711353][   C16]  smp_apic_timer_interrupt+0xfb/0x370
> [   64.721367][   C16]  apic_timer_interrupt+0xf/0x20
> [   64.721367][   C16]  </IRQ>
> [   64.721367][   C16] RIP: 0010:_raw_spin_unlock_irqrestore+0x2f/0x40
> [   64.731370][   C16] Code: 55 48 89 e5 41 54 49 89 f4 be 01 00 00 00 53 
> 

Thanks for the report. Man, this series is nastier than I thought. This
is making more noise than I was hoping for.

@Andrew can you revert patch 4-6 for now? I'll be on vacation soon and
don't want cleanups to constantly break things. Just nasty.

-- 

Thanks,

David / dhildenb

^ permalink raw reply

* Re: [PATCH v3 5/6] mm/memory_hotplug: Move and simplify walk_memory_blocks()
From: David Hildenbrand @ 2019-06-21 15:26 UTC (permalink / raw)
  To: linux-kernel
  Cc: Oscar Salvador, Stephen Rothwell, Michal Hocko, Pavel Tatashin,
	linux-mm, mike.travis@hpe.com, Greg Kroah-Hartman,
	Rafael J. Wysocki, Wei Yang, linux-acpi, Andrew Banman, Arun KS,
	Qian Cai, Dan Williams, linuxppc-dev, Andrew Morton
In-Reply-To: <20190620183139.4352-6-david@redhat.com>

On 20.06.19 20:31, David Hildenbrand wrote:
> Let's move walk_memory_blocks() to the place where memory block logic
> resides and simplify it. While at it, add a type for the callback function.
> 
> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> Cc: "Rafael J. Wysocki" <rafael@kernel.org>
> Cc: David Hildenbrand <david@redhat.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Stephen Rothwell <sfr@canb.auug.org.au>
> Cc: Pavel Tatashin <pasha.tatashin@soleen.com>
> Cc: Andrew Banman <andrew.banman@hpe.com>
> Cc: "mike.travis@hpe.com" <mike.travis@hpe.com>
> Cc: Oscar Salvador <osalvador@suse.com>
> Cc: Michal Hocko <mhocko@suse.com>
> Cc: Wei Yang <richard.weiyang@gmail.com>
> Cc: Arun KS <arunks@codeaurora.org>
> Cc: Qian Cai <cai@lca.pw>
> Signed-off-by: David Hildenbrand <david@redhat.com>
> ---
>  drivers/base/memory.c          | 42 ++++++++++++++++++++++++++
>  include/linux/memory.h         |  3 ++
>  include/linux/memory_hotplug.h |  2 --
>  mm/memory_hotplug.c            | 55 ----------------------------------
>  4 files changed, 45 insertions(+), 57 deletions(-)
> 
> diff --git a/drivers/base/memory.c b/drivers/base/memory.c
> index c54e80fd25a8..0204384b4d1d 100644
> --- a/drivers/base/memory.c
> +++ b/drivers/base/memory.c
> @@ -44,6 +44,11 @@ static inline unsigned long pfn_to_block_id(unsigned long pfn)
>  	return base_memory_block_id(pfn_to_section_nr(pfn));
>  }
>  
> +static inline unsigned long phys_to_block_id(unsigned long phys)
> +{
> +	return pfn_to_block_id(PFN_DOWN(phys));
> +}
> +
>  static int memory_subsys_online(struct device *dev);
>  static int memory_subsys_offline(struct device *dev);
>  
> @@ -851,3 +856,40 @@ int __init memory_dev_init(void)
>  		printk(KERN_ERR "%s() failed: %d\n", __func__, ret);
>  	return ret;
>  }
> +
> +/**
> + * walk_memory_blocks - walk through all present memory blocks overlapped
> + *			by the range [start, start + size)
> + *
> + * @start: start address of the memory range
> + * @size: size of the memory range
> + * @arg: argument passed to func
> + * @func: callback for each memory section walked
> + *
> + * This function walks through all present memory blocks overlapped by the
> + * range [start, start + size), calling func on each memory block.
> + *
> + * In case func() returns an error, walking is aborted and the error is
> + * returned.
> + */
> +int walk_memory_blocks(unsigned long start, unsigned long size,
> +		       void *arg, walk_memory_blocks_func_t func)
> +{
> +	const unsigned long start_block_id = phys_to_block_id(start);
> +	const unsigned long end_block_id = phys_to_block_id(start + size - 1);
> +	struct memory_block *mem;
> +	unsigned long block_id;
> +	int ret = 0;

I *guess* the stall we are seeing is when size = 0.

(via ACPI, if info->length is 0)

if (!size)
	return 0;

... but that is just a wild guess. Will have a look after my vacation.

-- 

Thanks,

David / dhildenb

^ permalink raw reply

* Re: [PATCH 01/16] mm: use untagged_addr() for get_user_pages_fast addresses
From: Khalid Aziz @ 2019-06-21 15:35 UTC (permalink / raw)
  To: Jason Gunthorpe, Christoph Hellwig
  Cc: x86, Rich Felker, Yoshinori Sato, linux-sh, James Hogan,
	linuxppc-dev, linux-mips, Nicholas Piggin, linux-kernel, linux-mm,
	Paul Burton, Paul Mackerras, Andrey Konovalov, sparclinux,
	Linus Torvalds, David S. Miller
In-Reply-To: <20190621133911.GL19891@ziepe.ca>

On 6/21/19 7:39 AM, Jason Gunthorpe wrote:
> On Tue, Jun 11, 2019 at 04:40:47PM +0200, Christoph Hellwig wrote:
>> This will allow sparc64 to override its ADI tags for
>> get_user_pages and get_user_pages_fast.
>>
>> Signed-off-by: Christoph Hellwig <hch@lst.de>
>>  mm/gup.c | 4 ++--
>>  1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/mm/gup.c b/mm/gup.c
>> index ddde097cf9e4..6bb521db67ec 100644
>> +++ b/mm/gup.c
>> @@ -2146,7 +2146,7 @@ int __get_user_pages_fast(unsigned long start, int nr_pages, int write,
>>  	unsigned long flags;
>>  	int nr = 0;
>>  
>> -	start &= PAGE_MASK;
>> +	start = untagged_addr(start) & PAGE_MASK;
>>  	len = (unsigned long) nr_pages << PAGE_SHIFT;
>>  	end = start + len;
> 
> Hmm, this function, and the other, goes on to do:
> 
>         if (unlikely(!access_ok((void __user *)start, len)))
>                 return 0;
> 
> and I thought that access_ok takes in the tagged pointer?
> 
> How about re-order it a bit?

access_ok() can handle tagged or untagged pointers. It just strips the
tag bits from the top bits. Current order doesn't really matter from
functionality point of view. There might be minor gain in delaying
untagging in __get_user_pages_fast() but I could go either way.

--
Khalid

> 
> diff --git a/mm/gup.c b/mm/gup.c
> index ddde097cf9e410..f48747ced4723b 100644
> --- a/mm/gup.c
> +++ b/mm/gup.c
> @@ -2148,11 +2148,12 @@ int __get_user_pages_fast(unsigned long start, int nr_pages, int write,
>  
>  	start &= PAGE_MASK;
>  	len = (unsigned long) nr_pages << PAGE_SHIFT;
> -	end = start + len;
> -
>  	if (unlikely(!access_ok((void __user *)start, len)))
>  		return 0;
>  
> +	start = untagged_ptr(start);
> +	end = start + len;
> +
>  	/*
>  	 * Disable interrupts.  We use the nested form as we can already have
>  	 * interrupts disabled by get_futex_key.
> 



^ permalink raw reply

* Re: [PATCH 01/16] mm: use untagged_addr() for get_user_pages_fast addresses
From: Jason Gunthorpe @ 2019-06-21 15:54 UTC (permalink / raw)
  To: Khalid Aziz
  Cc: x86, Rich Felker, Yoshinori Sato, linux-sh, James Hogan,
	linuxppc-dev, linux-mips, Nicholas Piggin, David S. Miller,
	linux-mm, Paul Burton, Paul Mackerras, Andrey Konovalov,
	sparclinux, Linus Torvalds, Christoph Hellwig, linux-kernel
In-Reply-To: <9a4e1485-4683-92b0-3d26-73f26896d646@oracle.com>

On Fri, Jun 21, 2019 at 09:35:11AM -0600, Khalid Aziz wrote:
> On 6/21/19 7:39 AM, Jason Gunthorpe wrote:
> > On Tue, Jun 11, 2019 at 04:40:47PM +0200, Christoph Hellwig wrote:
> >> This will allow sparc64 to override its ADI tags for
> >> get_user_pages and get_user_pages_fast.
> >>
> >> Signed-off-by: Christoph Hellwig <hch@lst.de>
> >>  mm/gup.c | 4 ++--
> >>  1 file changed, 2 insertions(+), 2 deletions(-)
> >>
> >> diff --git a/mm/gup.c b/mm/gup.c
> >> index ddde097cf9e4..6bb521db67ec 100644
> >> +++ b/mm/gup.c
> >> @@ -2146,7 +2146,7 @@ int __get_user_pages_fast(unsigned long start, int nr_pages, int write,
> >>  	unsigned long flags;
> >>  	int nr = 0;
> >>  
> >> -	start &= PAGE_MASK;
> >> +	start = untagged_addr(start) & PAGE_MASK;
> >>  	len = (unsigned long) nr_pages << PAGE_SHIFT;
> >>  	end = start + len;
> > 
> > Hmm, this function, and the other, goes on to do:
> > 
> >         if (unlikely(!access_ok((void __user *)start, len)))
> >                 return 0;
> > 
> > and I thought that access_ok takes in the tagged pointer?
> > 
> > How about re-order it a bit?
> 
> access_ok() can handle tagged or untagged pointers. It just strips the
> tag bits from the top bits. Current order doesn't really matter from
> functionality point of view. There might be minor gain in delaying
> untagging in __get_user_pages_fast() but I could go either way.

I understand the current ARM and SPARC implementations don't do much
with the tags, but it feels like a really big assumption for the core
code that all future uses of tags will be fine to have them stripped
out of 'void __user *' pointers. IMHO that is something we should not
be doing in the core kernel..

Jason

^ permalink raw reply

* Re: [PATCH v3 0/6] mm: Further memory block device cleanups
From: David Hildenbrand @ 2019-06-21 18:24 UTC (permalink / raw)
  To: Qian Cai, linux-kernel
  Cc: Oscar Salvador, Michal Hocko, Rafael J. Wysocki, Wei Yang,
	Keith Busch, linux-mm, Arun KS, Rashmica Gupta, Thomas Gleixner,
	Stephen Rothwell, Michael Neuling, Baoquan He, Pavel Tatashin,
	linux-acpi, Len Brown, Pavel Tatashin, Pavel Tatashin,
	Anshuman Khandual, mike.travis@hpe.com, linuxppc-dev,
	Mike Rapoport, Dan Williams, Vlastimil Babka, Oscar Salvador,
	Juergen Gross, Andrew Banman, Greg Kroah-Hartman,
	Rafael J. Wysocki, Johannes Weiner, Paul Mackerras, Andrew Morton,
	Mel Gorman
In-Reply-To: <1561130120.5154.47.camel@lca.pw>

On 21.06.19 17:15, Qian Cai wrote:
> On Thu, 2019-06-20 at 20:31 +0200, David Hildenbrand wrote:
>> @Andrew: Only patch 1, 4 and 6 changed compared to v1.
>>
>> Some further cleanups around memory block devices. Especially, clean up
>> and simplify walk_memory_range(). Including some other minor cleanups.
>>
>> Compiled + tested on x86 with DIMMs under QEMU. Compile-tested on ppc64.
>>
>> v2 -> v3:
>> - "mm/memory_hotplug: Rename walk_memory_range() and pass start+size .."
>> -- Avoid warning on ppc.
>> - "drivers/base/memory.c: Get rid of find_memory_block_hinted()"
>> -- Fixup a comment regarding hinted devices.
>>
>> v1 -> v2:
>> - "mm: Section numbers use the type "unsigned long""
>> -- "unsigned long i" -> "unsigned long nr", in one case -> "int i"
>> - "drivers/base/memory.c: Get rid of find_memory_block_hinted("
>> -- Fix compilation error
>> -- Get rid of the "hint" parameter completely
>>
>> David Hildenbrand (6):
>>   mm: Section numbers use the type "unsigned long"
>>   drivers/base/memory: Use "unsigned long" for block ids
>>   mm: Make register_mem_sect_under_node() static
>>   mm/memory_hotplug: Rename walk_memory_range() and pass start+size
>>     instead of pfns
>>   mm/memory_hotplug: Move and simplify walk_memory_blocks()
>>   drivers/base/memory.c: Get rid of find_memory_block_hinted()
>>
>>  arch/powerpc/platforms/powernv/memtrace.c |  23 ++---
>>  drivers/acpi/acpi_memhotplug.c            |  19 +---
>>  drivers/base/memory.c                     | 120 +++++++++++++---------
>>  drivers/base/node.c                       |   8 +-
>>  include/linux/memory.h                    |   5 +-
>>  include/linux/memory_hotplug.h            |   2 -
>>  include/linux/mmzone.h                    |   4 +-
>>  include/linux/node.h                      |   7 --
>>  mm/memory_hotplug.c                       |  57 +---------
>>  mm/sparse.c                               |  12 +--
>>  10 files changed, 106 insertions(+), 151 deletions(-)
>>
> 
> This series causes a few machines are unable to boot triggering endless soft
> lockups. Reverted those commits fixed the issue.
> 
> 97f4217d1da0 Revert "mm/memory_hotplug: rename walk_memory_range() and pass
> start+size instead of pfns"
> c608eebf33c6 Revert "mm-memory_hotplug-rename-walk_memory_range-and-pass-
> startsize-instead-of-pfns-fix"
> 34b5e4ab7558 Revert "mm/memory_hotplug: move and simplify walk_memory_blocks()"
> 59a9f3eec5d1 Revert "drivers/base/memory.c: Get rid of
> find_memory_block_hinted()"
> 5cfcd52288b6 Revert "drivers-base-memoryc-get-rid-of-find_memory_block_hinted-
> v3"
> 
> [    4.582081][    T1] ACPI FADT declares the system doesn't support PCIe ASPM,
> so disable it
> [    4.590405][    T1] ACPI: bus type PCI registered
> [    4.592908][    T1] PCI: MMCONFIG for domain 0000 [bus 00-ff] at [mem
> 0x80000000-0x8fffffff] (base 0x80000000)
> [    4.601860][    T1] PCI: MMCONFIG at [mem 0x80000000-0x8fffffff] reserved in
> E820
> [    4.601860][    T1] PCI: Using configuration type 1 for base access
> [   28.661336][   C16] watchdog: BUG: soft lockup - CPU#16 stuck for 22s!
> [swapper/0:1]
> [   28.671351][   C16] Modules linked in:
> [   28.671354][   C16] CPU: 16 PID: 1 Comm: swapper/0 Not tainted 5.2.0-rc5-
> next-20190621+ #1
> [   28.681366][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385
> Gen10, BIOS A40 03/09/2018
> [   28.691334][   C16] RIP: 0010:_raw_spin_unlock_irqrestore+0x2f/0x40
> [   28.701334][   C16] Code: 55 48 89 e5 41 54 49 89 f4 be 01 00 00 00 53 48 8b
> 55 08 48 89 fb 48 8d 7f 18 e8 4c 89 7d ff 48 89 df e8 94 f9 7d ff 41 54 9d <65>
> ff 0d c2 44 8d 48 5b 41 5c 5d c3 0f 1f 44 00 00 0f 1f 44 00 00
> [   28.711354][   C16] RSP: 0018:ffff888205b27bf8 EFLAGS: 00000246 ORIG_RAX:
> ffffffffffffff13
> [   28.721372][   C16] RAX: 0000000000000000 RBX: ffff8882053d6138 RCX:
> ffffffffb6f2a3b8
> [   28.731371][   C16] RDX: 1ffff11040a7ac27 RSI: dffffc0000000000 RDI:
> ffff8882053d6138
> [   28.741371][   C16] RBP: ffff888205b27c08 R08: ffffed1040a7ac28 R09:
> ffffed1040a7ac27
> [   28.751334][   C16] R10: ffffed1040a7ac27 R11: ffff8882053d613b R12:
> 0000000000000246
> [   28.751370][   C16] R13: ffff888205b27c98 R14: ffff8884504d0a20 R15:
> 0000000000000000
> [   28.761368][   C16] FS:  0000000000000000(0000) GS:ffff888454500000(0000)
> knlGS:0000000000000000
> [   28.771373][   C16] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [   28.781334][   C16] CR2: 0000000000000000 CR3: 00000007c9012000 CR4:
> 00000000001406a0
> [   28.791333][   C16] Call Trace:
> [   28.791374][   C16]  klist_next+0xd8/0x1c0
> [   28.791374][   C16]  subsys_find_device_by_id+0x13b/0x1f0
> [   28.801334][   C16]  ? bus_find_device_by_name+0x20/0x20
> [   28.801370][   C16]  ? kobject_put+0x23/0x250
> [   28.811333][   C16]  walk_memory_blocks+0x6c/0xb8
> [   28.811353][   C16]  ? write_policy_show+0x40/0x40
> [   28.821334][   C16]  link_mem_sections+0x7e/0xa0
> [   28.821369][   C16]  ? unregister_memory_block_under_nodes+0x210/0x210
> [   28.831353][   C16]  ? __register_one_node+0x3bd/0x600
> [   28.831353][   C16]  topology_init+0xbf/0x126
> [   28.841364][   C16]  ? enable_cpu0_hotplug+0x1a/0x1a
> [   28.841368][   C16]  do_one_initcall+0xfe/0x45a
> [   28.851334][   C16]  ? initcall_blacklisted+0x150/0x150
> [   28.851353][   C16]  ? kasan_check_write+0x14/0x20
> [   28.861333][   C16]  ? up_write+0x75/0x140
> [   28.861369][   C16]  kernel_init_freeable+0x619/0x6ac
> [   28.871333][   C16]  ? rest_init+0x188/0x188
> [   28.871353][   C16]  kernel_init+0x11/0x138
> [   28.881363][   C16]  ? rest_init+0x188/0x188
> [   28.881363][   C16]  ret_from_fork+0x22/0x40
> [   56.661336][   C16] watchdog: BUG: soft lockup - CPU#16 stuck for 22s!
> [swapper/0:1]
> [   56.671352][   C16] Modules linked in:
> [   56.671354][   C16] CPU: 16 PID: 1 Comm: swapper/0 Tainted:
> G             L    5.2.0-rc5-next-20190621+ #1
> [   56.681357][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385
> Gen10, BIOS A40 03/09/2018
> [   56.691356][   C16] RIP: 0010:subsys_find_device_by_id+0x168/0x1f0
> [   56.701334][   C16] Code: 48 85 c0 74 3e 48 8d 78 58 e8 14 77 ca ff 4d 8b 7e
> 58 4d 85 ff 74 2c 49 8d bf a0 03 00 00 e8 bf 75 ca ff 45 39 a7 a0 03 00 00 <75>
> c9 4c 89 ff e8 0e 89 ff ff 48 85 c0 74 bc 48 89 df e8 21 3b 24
> [   56.721333][   C16] RSP: 0018:ffff888205b27c68 EFLAGS: 00000287 ORIG_RAX:
> ffffffffffffff13
> [   56.721370][   C16] RAX: 0000000000000000 RBX: ffff888205b27c90 RCX:
> ffffffffb74c9dc1
> [   56.731370][   C16] RDX: 0000000000000003 RSI: dffffc0000000000 RDI:
> ffff8888774ec3e0
> [   56.741371][   C16] RBP: ffff888205b27cf8 R08: ffffed1040a7ac28 R09:
> ffffed1040a7ac27
> [   56.751335][   C16] R10: ffffed1040a7ac27 R11: ffff8882053d613b R12:
> 0000000000085c1b
> [   56.761334][   C16] R13: 1ffff11040b64f8e R14: ffff888450de4a20 R15:
> ffff8888774ec040
> [   56.761372][   C16] FS:  0000000000000000(0000) GS:ffff888454500000(0000)
> knlGS:0000000000000000
> [   56.771374][   C16] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [   56.781370][   C16] CR2: 0000000000000000 CR3: 00000007c9012000 CR4:
> 00000000001406a0
> [   56.791373][   C16] Call Trace:
> [   56.791373][   C16]  ? bus_find_device_by_name+0x20/0x20
> [   56.801334][   C16]  ? kobject_put+0x23/0x250
> [   56.801334][   C16]  walk_memory_blocks+0x6c/0xb8
> [   56.811333][   C16]  ? write_policy_show+0x40/0x40
> [   56.811353][   C16]  link_mem_sections+0x7e/0xa0
> [   56.811353][   C16]  ? unregister_memory_block_under_nodes+0x210/0x210
> [   56.821333][   C16]  ? __register_one_node+0x3bd/0x600
> [   56.831333][   C16]  topology_init+0xbf/0x126
> [   56.831355][   C16]  ? enable_cpu0_hotplug+0x1a/0x1a
> [   56.841334][   C16]  do_one_initcall+0xfe/0x45a
> [   56.841334][   C16]  ? initcall_blacklisted+0x150/0x150
> [   56.851333][   C16]  ? kasan_check_write+0x14/0x20
> [   56.851354][   C16]  ? up_write+0x75/0x140
> [   56.861333][   C16]  kernel_init_freeable+0x619/0x6ac
> [   56.861333][   C16]  ? rest_init+0x188/0x188
> [   56.861369][   C16]  kernel_init+0x11/0x138
> [   56.871333][   C16]  ? rest_init+0x188/0x188
> [   56.871354][   C16]  ret_from_fork+0x22/0x40
> [   64.601362][   C16] rcu: INFO: rcu_sched self-detected stall on CPU
> [   64.611335][   C16] rcu: 	16-....: (5958 ticks this GP)
> idle=37e/1/0x4000000000000002 softirq=27/27 fqs=3000 
> [   64.621334][   C16] 	(t=6002 jiffies g=-1079 q=25)
> [   64.621334][   C16] NMI backtrace for cpu 16
> [   64.621374][   C16] CPU: 16 PID: 1 Comm: swapper/0 Tainted:
> G             L    5.2.0-rc5-next-20190621+ #1
> [   64.631372][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385
> Gen10, BIOS A40 03/09/2018
> [   64.641371][   C16] Call Trace:
> [   64.651337][   C16]  <IRQ>
> [   64.651376][   C16]  dump_stack+0x62/0x9a
> [   64.651376][   C16]  nmi_cpu_backtrace.cold.0+0x2e/0x33
> [   64.661337][   C16]  ? nmi_cpu_backtrace_handler+0x20/0x20
> [   64.661337][   C16]  nmi_trigger_cpumask_backtrace+0x1a6/0x1b9
> [   64.671353][   C16]  arch_trigger_cpumask_backtrace+0x19/0x20
> [   64.681366][   C16]  rcu_dump_cpu_stacks+0x18b/0x1d6
> [   64.681366][   C16]  rcu_sched_clock_irq.cold.64+0x368/0x791
> [   64.691336][   C16]  ? kasan_check_read+0x11/0x20
> [   64.691354][   C16]  ? __raise_softirq_irqoff+0x66/0x150
> [   64.701336][   C16]  update_process_times+0x2f/0x60
> [   64.701362][   C16]  tick_periodic+0x38/0xe0
> [   64.711334][   C16]  tick_handle_periodic+0x2e/0x80
> [   64.711353][   C16]  smp_apic_timer_interrupt+0xfb/0x370
> [   64.721367][   C16]  apic_timer_interrupt+0xf/0x20
> [   64.721367][   C16]  </IRQ>
> [   64.721367][   C16] RIP: 0010:_raw_spin_unlock_irqrestore+0x2f/0x40
> [   64.731370][   C16] Code: 55 48 89 e5 41 54 49 89 f4 be 01 00 00 00 53 
> 

@Qian Cai, unfortunately I can't reproduce.

If you get the chance, it would be great if you could retry with

diff --git a/drivers/base/memory.c b/drivers/base/memory.c
index 972c5336bebf..742f99ddd148 100644
--- a/drivers/base/memory.c
+++ b/drivers/base/memory.c
@@ -868,6 +868,9 @@ int walk_memory_blocks(unsigned long start, unsigned
long size,
        unsigned long block_id;
        int ret = 0;

+       if (!size)
+               return;
+
        for (block_id = start_block_id; block_id <= end_block_id;
block_id++) {
                mem = find_memory_block_by_id(block_id);
                if (!mem)



If both, start and size are 0, we would get a veeeery long loop. This
would mean that we have an online node that does not span any pages at
all (pgdat->node_start_pfn = 0, start_pfn + pgdat->node_spanned_pages = 0).

-- 

Thanks,

David / dhildenb

^ permalink raw reply related

* Re: [PATCH v3 0/6] mm: Further memory block device cleanups
From: David Hildenbrand @ 2019-06-21 18:56 UTC (permalink / raw)
  To: Qian Cai, linux-kernel
  Cc: Oscar Salvador, Michal Hocko, Rafael J. Wysocki, Wei Yang,
	Keith Busch, linux-mm, Arun KS, Rashmica Gupta, Thomas Gleixner,
	Stephen Rothwell, Michael Neuling, Baoquan He, Pavel Tatashin,
	linux-acpi, Len Brown, Pavel Tatashin, Pavel Tatashin,
	Anshuman Khandual, mike.travis@hpe.com, linuxppc-dev,
	Mike Rapoport, Dan Williams, Vlastimil Babka, Oscar Salvador,
	Juergen Gross, Andrew Banman, Greg Kroah-Hartman,
	Rafael J. Wysocki, Johannes Weiner, Paul Mackerras, Andrew Morton,
	Mel Gorman
In-Reply-To: <1c2edc22-afd7-2211-c4c7-40e54e5007e8@redhat.com>

On 21.06.19 20:24, David Hildenbrand wrote:
> On 21.06.19 17:15, Qian Cai wrote:
>> On Thu, 2019-06-20 at 20:31 +0200, David Hildenbrand wrote:
>>> @Andrew: Only patch 1, 4 and 6 changed compared to v1.
>>>
>>> Some further cleanups around memory block devices. Especially, clean up
>>> and simplify walk_memory_range(). Including some other minor cleanups.
>>>
>>> Compiled + tested on x86 with DIMMs under QEMU. Compile-tested on ppc64.
>>>
>>> v2 -> v3:
>>> - "mm/memory_hotplug: Rename walk_memory_range() and pass start+size .."
>>> -- Avoid warning on ppc.
>>> - "drivers/base/memory.c: Get rid of find_memory_block_hinted()"
>>> -- Fixup a comment regarding hinted devices.
>>>
>>> v1 -> v2:
>>> - "mm: Section numbers use the type "unsigned long""
>>> -- "unsigned long i" -> "unsigned long nr", in one case -> "int i"
>>> - "drivers/base/memory.c: Get rid of find_memory_block_hinted("
>>> -- Fix compilation error
>>> -- Get rid of the "hint" parameter completely
>>>
>>> David Hildenbrand (6):
>>>   mm: Section numbers use the type "unsigned long"
>>>   drivers/base/memory: Use "unsigned long" for block ids
>>>   mm: Make register_mem_sect_under_node() static
>>>   mm/memory_hotplug: Rename walk_memory_range() and pass start+size
>>>     instead of pfns
>>>   mm/memory_hotplug: Move and simplify walk_memory_blocks()
>>>   drivers/base/memory.c: Get rid of find_memory_block_hinted()
>>>
>>>  arch/powerpc/platforms/powernv/memtrace.c |  23 ++---
>>>  drivers/acpi/acpi_memhotplug.c            |  19 +---
>>>  drivers/base/memory.c                     | 120 +++++++++++++---------
>>>  drivers/base/node.c                       |   8 +-
>>>  include/linux/memory.h                    |   5 +-
>>>  include/linux/memory_hotplug.h            |   2 -
>>>  include/linux/mmzone.h                    |   4 +-
>>>  include/linux/node.h                      |   7 --
>>>  mm/memory_hotplug.c                       |  57 +---------
>>>  mm/sparse.c                               |  12 +--
>>>  10 files changed, 106 insertions(+), 151 deletions(-)
>>>
>>
>> This series causes a few machines are unable to boot triggering endless soft
>> lockups. Reverted those commits fixed the issue.
>>
>> 97f4217d1da0 Revert "mm/memory_hotplug: rename walk_memory_range() and pass
>> start+size instead of pfns"
>> c608eebf33c6 Revert "mm-memory_hotplug-rename-walk_memory_range-and-pass-
>> startsize-instead-of-pfns-fix"
>> 34b5e4ab7558 Revert "mm/memory_hotplug: move and simplify walk_memory_blocks()"
>> 59a9f3eec5d1 Revert "drivers/base/memory.c: Get rid of
>> find_memory_block_hinted()"
>> 5cfcd52288b6 Revert "drivers-base-memoryc-get-rid-of-find_memory_block_hinted-
>> v3"
>>
>> [    4.582081][    T1] ACPI FADT declares the system doesn't support PCIe ASPM,
>> so disable it
>> [    4.590405][    T1] ACPI: bus type PCI registered
>> [    4.592908][    T1] PCI: MMCONFIG for domain 0000 [bus 00-ff] at [mem
>> 0x80000000-0x8fffffff] (base 0x80000000)
>> [    4.601860][    T1] PCI: MMCONFIG at [mem 0x80000000-0x8fffffff] reserved in
>> E820
>> [    4.601860][    T1] PCI: Using configuration type 1 for base access
>> [   28.661336][   C16] watchdog: BUG: soft lockup - CPU#16 stuck for 22s!
>> [swapper/0:1]
>> [   28.671351][   C16] Modules linked in:
>> [   28.671354][   C16] CPU: 16 PID: 1 Comm: swapper/0 Not tainted 5.2.0-rc5-
>> next-20190621+ #1
>> [   28.681366][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385
>> Gen10, BIOS A40 03/09/2018
>> [   28.691334][   C16] RIP: 0010:_raw_spin_unlock_irqrestore+0x2f/0x40
>> [   28.701334][   C16] Code: 55 48 89 e5 41 54 49 89 f4 be 01 00 00 00 53 48 8b
>> 55 08 48 89 fb 48 8d 7f 18 e8 4c 89 7d ff 48 89 df e8 94 f9 7d ff 41 54 9d <65>
>> ff 0d c2 44 8d 48 5b 41 5c 5d c3 0f 1f 44 00 00 0f 1f 44 00 00
>> [   28.711354][   C16] RSP: 0018:ffff888205b27bf8 EFLAGS: 00000246 ORIG_RAX:
>> ffffffffffffff13
>> [   28.721372][   C16] RAX: 0000000000000000 RBX: ffff8882053d6138 RCX:
>> ffffffffb6f2a3b8
>> [   28.731371][   C16] RDX: 1ffff11040a7ac27 RSI: dffffc0000000000 RDI:
>> ffff8882053d6138
>> [   28.741371][   C16] RBP: ffff888205b27c08 R08: ffffed1040a7ac28 R09:
>> ffffed1040a7ac27
>> [   28.751334][   C16] R10: ffffed1040a7ac27 R11: ffff8882053d613b R12:
>> 0000000000000246
>> [   28.751370][   C16] R13: ffff888205b27c98 R14: ffff8884504d0a20 R15:
>> 0000000000000000
>> [   28.761368][   C16] FS:  0000000000000000(0000) GS:ffff888454500000(0000)
>> knlGS:0000000000000000
>> [   28.771373][   C16] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>> [   28.781334][   C16] CR2: 0000000000000000 CR3: 00000007c9012000 CR4:
>> 00000000001406a0
>> [   28.791333][   C16] Call Trace:
>> [   28.791374][   C16]  klist_next+0xd8/0x1c0
>> [   28.791374][   C16]  subsys_find_device_by_id+0x13b/0x1f0
>> [   28.801334][   C16]  ? bus_find_device_by_name+0x20/0x20
>> [   28.801370][   C16]  ? kobject_put+0x23/0x250
>> [   28.811333][   C16]  walk_memory_blocks+0x6c/0xb8
>> [   28.811353][   C16]  ? write_policy_show+0x40/0x40
>> [   28.821334][   C16]  link_mem_sections+0x7e/0xa0
>> [   28.821369][   C16]  ? unregister_memory_block_under_nodes+0x210/0x210
>> [   28.831353][   C16]  ? __register_one_node+0x3bd/0x600
>> [   28.831353][   C16]  topology_init+0xbf/0x126
>> [   28.841364][   C16]  ? enable_cpu0_hotplug+0x1a/0x1a
>> [   28.841368][   C16]  do_one_initcall+0xfe/0x45a
>> [   28.851334][   C16]  ? initcall_blacklisted+0x150/0x150
>> [   28.851353][   C16]  ? kasan_check_write+0x14/0x20
>> [   28.861333][   C16]  ? up_write+0x75/0x140
>> [   28.861369][   C16]  kernel_init_freeable+0x619/0x6ac
>> [   28.871333][   C16]  ? rest_init+0x188/0x188
>> [   28.871353][   C16]  kernel_init+0x11/0x138
>> [   28.881363][   C16]  ? rest_init+0x188/0x188
>> [   28.881363][   C16]  ret_from_fork+0x22/0x40
>> [   56.661336][   C16] watchdog: BUG: soft lockup - CPU#16 stuck for 22s!
>> [swapper/0:1]
>> [   56.671352][   C16] Modules linked in:
>> [   56.671354][   C16] CPU: 16 PID: 1 Comm: swapper/0 Tainted:
>> G             L    5.2.0-rc5-next-20190621+ #1
>> [   56.681357][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385
>> Gen10, BIOS A40 03/09/2018
>> [   56.691356][   C16] RIP: 0010:subsys_find_device_by_id+0x168/0x1f0
>> [   56.701334][   C16] Code: 48 85 c0 74 3e 48 8d 78 58 e8 14 77 ca ff 4d 8b 7e
>> 58 4d 85 ff 74 2c 49 8d bf a0 03 00 00 e8 bf 75 ca ff 45 39 a7 a0 03 00 00 <75>
>> c9 4c 89 ff e8 0e 89 ff ff 48 85 c0 74 bc 48 89 df e8 21 3b 24
>> [   56.721333][   C16] RSP: 0018:ffff888205b27c68 EFLAGS: 00000287 ORIG_RAX:
>> ffffffffffffff13
>> [   56.721370][   C16] RAX: 0000000000000000 RBX: ffff888205b27c90 RCX:
>> ffffffffb74c9dc1
>> [   56.731370][   C16] RDX: 0000000000000003 RSI: dffffc0000000000 RDI:
>> ffff8888774ec3e0
>> [   56.741371][   C16] RBP: ffff888205b27cf8 R08: ffffed1040a7ac28 R09:
>> ffffed1040a7ac27
>> [   56.751335][   C16] R10: ffffed1040a7ac27 R11: ffff8882053d613b R12:
>> 0000000000085c1b
>> [   56.761334][   C16] R13: 1ffff11040b64f8e R14: ffff888450de4a20 R15:
>> ffff8888774ec040
>> [   56.761372][   C16] FS:  0000000000000000(0000) GS:ffff888454500000(0000)
>> knlGS:0000000000000000
>> [   56.771374][   C16] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>> [   56.781370][   C16] CR2: 0000000000000000 CR3: 00000007c9012000 CR4:
>> 00000000001406a0
>> [   56.791373][   C16] Call Trace:
>> [   56.791373][   C16]  ? bus_find_device_by_name+0x20/0x20
>> [   56.801334][   C16]  ? kobject_put+0x23/0x250
>> [   56.801334][   C16]  walk_memory_blocks+0x6c/0xb8
>> [   56.811333][   C16]  ? write_policy_show+0x40/0x40
>> [   56.811353][   C16]  link_mem_sections+0x7e/0xa0
>> [   56.811353][   C16]  ? unregister_memory_block_under_nodes+0x210/0x210
>> [   56.821333][   C16]  ? __register_one_node+0x3bd/0x600
>> [   56.831333][   C16]  topology_init+0xbf/0x126
>> [   56.831355][   C16]  ? enable_cpu0_hotplug+0x1a/0x1a
>> [   56.841334][   C16]  do_one_initcall+0xfe/0x45a
>> [   56.841334][   C16]  ? initcall_blacklisted+0x150/0x150
>> [   56.851333][   C16]  ? kasan_check_write+0x14/0x20
>> [   56.851354][   C16]  ? up_write+0x75/0x140
>> [   56.861333][   C16]  kernel_init_freeable+0x619/0x6ac
>> [   56.861333][   C16]  ? rest_init+0x188/0x188
>> [   56.861369][   C16]  kernel_init+0x11/0x138
>> [   56.871333][   C16]  ? rest_init+0x188/0x188
>> [   56.871354][   C16]  ret_from_fork+0x22/0x40
>> [   64.601362][   C16] rcu: INFO: rcu_sched self-detected stall on CPU
>> [   64.611335][   C16] rcu: 	16-....: (5958 ticks this GP)
>> idle=37e/1/0x4000000000000002 softirq=27/27 fqs=3000 
>> [   64.621334][   C16] 	(t=6002 jiffies g=-1079 q=25)
>> [   64.621334][   C16] NMI backtrace for cpu 16
>> [   64.621374][   C16] CPU: 16 PID: 1 Comm: swapper/0 Tainted:
>> G             L    5.2.0-rc5-next-20190621+ #1
>> [   64.631372][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385
>> Gen10, BIOS A40 03/09/2018
>> [   64.641371][   C16] Call Trace:
>> [   64.651337][   C16]  <IRQ>
>> [   64.651376][   C16]  dump_stack+0x62/0x9a
>> [   64.651376][   C16]  nmi_cpu_backtrace.cold.0+0x2e/0x33
>> [   64.661337][   C16]  ? nmi_cpu_backtrace_handler+0x20/0x20
>> [   64.661337][   C16]  nmi_trigger_cpumask_backtrace+0x1a6/0x1b9
>> [   64.671353][   C16]  arch_trigger_cpumask_backtrace+0x19/0x20
>> [   64.681366][   C16]  rcu_dump_cpu_stacks+0x18b/0x1d6
>> [   64.681366][   C16]  rcu_sched_clock_irq.cold.64+0x368/0x791
>> [   64.691336][   C16]  ? kasan_check_read+0x11/0x20
>> [   64.691354][   C16]  ? __raise_softirq_irqoff+0x66/0x150
>> [   64.701336][   C16]  update_process_times+0x2f/0x60
>> [   64.701362][   C16]  tick_periodic+0x38/0xe0
>> [   64.711334][   C16]  tick_handle_periodic+0x2e/0x80
>> [   64.711353][   C16]  smp_apic_timer_interrupt+0xfb/0x370
>> [   64.721367][   C16]  apic_timer_interrupt+0xf/0x20
>> [   64.721367][   C16]  </IRQ>
>> [   64.721367][   C16] RIP: 0010:_raw_spin_unlock_irqrestore+0x2f/0x40
>> [   64.731370][   C16] Code: 55 48 89 e5 41 54 49 89 f4 be 01 00 00 00 53 
>>
> 
> @Qian Cai, unfortunately I can't reproduce.
> 
> If you get the chance, it would be great if you could retry with
> 
> diff --git a/drivers/base/memory.c b/drivers/base/memory.c
> index 972c5336bebf..742f99ddd148 100644
> --- a/drivers/base/memory.c
> +++ b/drivers/base/memory.c
> @@ -868,6 +868,9 @@ int walk_memory_blocks(unsigned long start, unsigned
> long size,
>         unsigned long block_id;
>         int ret = 0;
> 
> +       if (!size)
> +               return;
> +
>         for (block_id = start_block_id; block_id <= end_block_id;
> block_id++) {
>                 mem = find_memory_block_by_id(block_id);
>                 if (!mem)
> 
> 
> 
> If both, start and size are 0, we would get a veeeery long loop. This
> would mean that we have an online node that does not span any pages at
> all (pgdat->node_start_pfn = 0, start_pfn + pgdat->node_spanned_pages = 0).
> 


...trying to reproduce with QEMU (setting 0MB for the second node):

qemu-system-x86_64 --enable-kvm -m 4G,maxmem=20G,slots=2 \
	-smp sockets=2,cores=1 \
	-numa node,nodeid=0,cpus=0,mem=4G \
	-numa node,nodeid=1,cpus=1,mem=0 ...

I can indeed see that the node is online and
"pgdat->node_start_pfn == 0 && start_pfn + pgdat->node_spanned_pages == 0".

However, the kernel segfaults in an unrelated code path, so I can't
verify if this solves this problem:

[    0.313284] BUG: kernel NULL pointer dereference, address: 00000000000000a0
[    0.313479] #PF: supervisor read access in kernel mode
[    0.313479] #PF: error_code(0x0000) - not-present page
[    0.313479] PGD 0 P4D 0 
[    0.313479] Oops: 0000 [#1] SMP PTI
[    0.313479] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.2.0-rc5-next-20190620+ #56
[    0.313479] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.4
[    0.313479] RIP: 0010:bus_add_device+0x59/0x110
[    0.313479] Code: 20 48 89 df e8 f8 b4 ff ff 41 89 c4 85 c0 0f 85 81 00 00 00 48 8b 53 50 48 85 d2 75 03 48 8b 135
[    0.313479] RSP: 0000:ffffb4a6c0013e20 EFLAGS: 00010246
[    0.313479] RAX: 0000000000000000 RBX: ffff8b61bac23800 RCX: 0000000000000000
[    0.313479] RDX: ffff8b61bac29038 RSI: ffff8b61bac23800 RDI: ffff8b61bac23800
[    0.313479] RBP: ffffffff9d2f4500 R08: 0000000000000000 R09: 0000000000000001
[    0.313479] R10: 0000000000000000 R11: ffff8b61bad20878 R12: 0000000000000000
[    0.313479] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
[    0.313479] FS:  0000000000000000(0000) GS:ffff8b61bba00000(0000) knlGS:0000000000000000
[    0.313479] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[    0.313479] CR2: 00000000000000a0 CR3: 0000000013c24000 CR4: 00000000000006f0
[    0.313479] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[    0.313479] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[    0.313479] Call Trace:
[    0.313479]  device_add+0x304/0x660
[    0.313479]  ? __init_waitqueue_head+0x31/0x50
[    0.313479]  __register_one_node+0x67/0x170
[    0.313479]  __try_online_node.cold+0x3e/0x78
[    0.313479]  try_online_node+0x25/0x40
[    0.313479]  do_cpu_up+0x36/0xc0
[    0.313479]  smp_init+0x59/0xb3
[    0.313479]  kernel_init_freeable+0x11a/0x247
[    0.313479]  ? rest_init+0x23f/0x23f
[    0.313479]  kernel_init+0x5/0xf1
[    0.313479]  ret_from_fork+0x3a/0x50
[    0.313479] Modules linked in:

Figuring out what goes wrong here (maybe QEMU creating a weird
system configuration) is a different journey :)

-- 

Thanks,

David / dhildenb

^ permalink raw reply

* Re: [PATCH v3 0/6] mm: Further memory block device cleanups
From: Qian Cai @ 2019-06-21 19:07 UTC (permalink / raw)
  To: David Hildenbrand, linux-kernel
  Cc: Oscar Salvador, Michal Hocko, Rafael J. Wysocki, Wei Yang,
	Keith Busch, linux-mm, Arun KS, Rashmica Gupta, Thomas Gleixner,
	Stephen Rothwell, Michael Neuling, Baoquan He, Pavel Tatashin,
	linux-acpi, Len Brown, Pavel Tatashin, Pavel Tatashin,
	Anshuman Khandual, mike.travis@hpe.com, linuxppc-dev,
	Mike Rapoport, Dan Williams, Vlastimil Babka, Oscar Salvador,
	Juergen Gross, Andrew Banman, Greg Kroah-Hartman,
	Rafael J. Wysocki, Johannes Weiner, Paul Mackerras, Andrew Morton,
	Mel Gorman
In-Reply-To: <5b4e4f85-8f3e-0a15-6c85-704512205a42@redhat.com>

On Fri, 2019-06-21 at 20:56 +0200, David Hildenbrand wrote:
> On 21.06.19 20:24, David Hildenbrand wrote:
> > On 21.06.19 17:15, Qian Cai wrote:
> > > On Thu, 2019-06-20 at 20:31 +0200, David Hildenbrand wrote:
> > > > @Andrew: Only patch 1, 4 and 6 changed compared to v1.
> > > > 
> > > > Some further cleanups around memory block devices. Especially, clean up
> > > > and simplify walk_memory_range(). Including some other minor cleanups.
> > > > 
> > > > Compiled + tested on x86 with DIMMs under QEMU. Compile-tested on ppc64.
> > > > 
> > > > v2 -> v3:
> > > > - "mm/memory_hotplug: Rename walk_memory_range() and pass start+size .."
> > > > -- Avoid warning on ppc.
> > > > - "drivers/base/memory.c: Get rid of find_memory_block_hinted()"
> > > > -- Fixup a comment regarding hinted devices.
> > > > 
> > > > v1 -> v2:
> > > > - "mm: Section numbers use the type "unsigned long""
> > > > -- "unsigned long i" -> "unsigned long nr", in one case -> "int i"
> > > > - "drivers/base/memory.c: Get rid of find_memory_block_hinted("
> > > > -- Fix compilation error
> > > > -- Get rid of the "hint" parameter completely
> > > > 
> > > > David Hildenbrand (6):
> > > >   mm: Section numbers use the type "unsigned long"
> > > >   drivers/base/memory: Use "unsigned long" for block ids
> > > >   mm: Make register_mem_sect_under_node() static
> > > >   mm/memory_hotplug: Rename walk_memory_range() and pass start+size
> > > >     instead of pfns
> > > >   mm/memory_hotplug: Move and simplify walk_memory_blocks()
> > > >   drivers/base/memory.c: Get rid of find_memory_block_hinted()
> > > > 
> > > >  arch/powerpc/platforms/powernv/memtrace.c |  23 ++---
> > > >  drivers/acpi/acpi_memhotplug.c            |  19 +---
> > > >  drivers/base/memory.c                     | 120 +++++++++++++---------
> > > >  drivers/base/node.c                       |   8 +-
> > > >  include/linux/memory.h                    |   5 +-
> > > >  include/linux/memory_hotplug.h            |   2 -
> > > >  include/linux/mmzone.h                    |   4 +-
> > > >  include/linux/node.h                      |   7 --
> > > >  mm/memory_hotplug.c                       |  57 +---------
> > > >  mm/sparse.c                               |  12 +--
> > > >  10 files changed, 106 insertions(+), 151 deletions(-)
> > > > 
> > > 
> > > This series causes a few machines are unable to boot triggering endless
> > > soft
> > > lockups. Reverted those commits fixed the issue.
> > > 
> > > 97f4217d1da0 Revert "mm/memory_hotplug: rename walk_memory_range() and
> > > pass
> > > start+size instead of pfns"
> > > c608eebf33c6 Revert "mm-memory_hotplug-rename-walk_memory_range-and-pass-
> > > startsize-instead-of-pfns-fix"
> > > 34b5e4ab7558 Revert "mm/memory_hotplug: move and simplify
> > > walk_memory_blocks()"
> > > 59a9f3eec5d1 Revert "drivers/base/memory.c: Get rid of
> > > find_memory_block_hinted()"
> > > 5cfcd52288b6 Revert "drivers-base-memoryc-get-rid-of-
> > > find_memory_block_hinted-
> > > v3"
> > > 
> > > [    4.582081][    T1] ACPI FADT declares the system doesn't support PCIe
> > > ASPM,
> > > so disable it
> > > [    4.590405][    T1] ACPI: bus type PCI registered
> > > [    4.592908][    T1] PCI: MMCONFIG for domain 0000 [bus 00-ff] at [mem
> > > 0x80000000-0x8fffffff] (base 0x80000000)
> > > [    4.601860][    T1] PCI: MMCONFIG at [mem 0x80000000-0x8fffffff]
> > > reserved in
> > > E820
> > > [    4.601860][    T1] PCI: Using configuration type 1 for base access
> > > [   28.661336][   C16] watchdog: BUG: soft lockup - CPU#16 stuck for 22s!
> > > [swapper/0:1]
> > > [   28.671351][   C16] Modules linked in:
> > > [   28.671354][   C16] CPU: 16 PID: 1 Comm: swapper/0 Not tainted 5.2.0-
> > > rc5-
> > > next-20190621+ #1
> > > [   28.681366][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant
> > > DL385
> > > Gen10, BIOS A40 03/09/2018
> > > [   28.691334][   C16] RIP: 0010:_raw_spin_unlock_irqrestore+0x2f/0x40
> > > [   28.701334][   C16] Code: 55 48 89 e5 41 54 49 89 f4 be 01 00 00 00 53
> > > 48 8b
> > > 55 08 48 89 fb 48 8d 7f 18 e8 4c 89 7d ff 48 89 df e8 94 f9 7d ff 41 54 9d
> > > <65>
> > > ff 0d c2 44 8d 48 5b 41 5c 5d c3 0f 1f 44 00 00 0f 1f 44 00 00
> > > [   28.711354][   C16] RSP: 0018:ffff888205b27bf8 EFLAGS: 00000246
> > > ORIG_RAX:
> > > ffffffffffffff13
> > > [   28.721372][   C16] RAX: 0000000000000000 RBX: ffff8882053d6138 RCX:
> > > ffffffffb6f2a3b8
> > > [   28.731371][   C16] RDX: 1ffff11040a7ac27 RSI: dffffc0000000000 RDI:
> > > ffff8882053d6138
> > > [   28.741371][   C16] RBP: ffff888205b27c08 R08: ffffed1040a7ac28 R09:
> > > ffffed1040a7ac27
> > > [   28.751334][   C16] R10: ffffed1040a7ac27 R11: ffff8882053d613b R12:
> > > 0000000000000246
> > > [   28.751370][   C16] R13: ffff888205b27c98 R14: ffff8884504d0a20 R15:
> > > 0000000000000000
> > > [   28.761368][   C16] FS:  0000000000000000(0000)
> > > GS:ffff888454500000(0000)
> > > knlGS:0000000000000000
> > > [   28.771373][   C16] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > > [   28.781334][   C16] CR2: 0000000000000000 CR3: 00000007c9012000 CR4:
> > > 00000000001406a0
> > > [   28.791333][   C16] Call Trace:
> > > [   28.791374][   C16]  klist_next+0xd8/0x1c0
> > > [   28.791374][   C16]  subsys_find_device_by_id+0x13b/0x1f0
> > > [   28.801334][   C16]  ? bus_find_device_by_name+0x20/0x20
> > > [   28.801370][   C16]  ? kobject_put+0x23/0x250
> > > [   28.811333][   C16]  walk_memory_blocks+0x6c/0xb8
> > > [   28.811353][   C16]  ? write_policy_show+0x40/0x40
> > > [   28.821334][   C16]  link_mem_sections+0x7e/0xa0
> > > [   28.821369][   C16]  ? unregister_memory_block_under_nodes+0x210/0x210
> > > [   28.831353][   C16]  ? __register_one_node+0x3bd/0x600
> > > [   28.831353][   C16]  topology_init+0xbf/0x126
> > > [   28.841364][   C16]  ? enable_cpu0_hotplug+0x1a/0x1a
> > > [   28.841368][   C16]  do_one_initcall+0xfe/0x45a
> > > [   28.851334][   C16]  ? initcall_blacklisted+0x150/0x150
> > > [   28.851353][   C16]  ? kasan_check_write+0x14/0x20
> > > [   28.861333][   C16]  ? up_write+0x75/0x140
> > > [   28.861369][   C16]  kernel_init_freeable+0x619/0x6ac
> > > [   28.871333][   C16]  ? rest_init+0x188/0x188
> > > [   28.871353][   C16]  kernel_init+0x11/0x138
> > > [   28.881363][   C16]  ? rest_init+0x188/0x188
> > > [   28.881363][   C16]  ret_from_fork+0x22/0x40
> > > [   56.661336][   C16] watchdog: BUG: soft lockup - CPU#16 stuck for 22s!
> > > [swapper/0:1]
> > > [   56.671352][   C16] Modules linked in:
> > > [   56.671354][   C16] CPU: 16 PID: 1 Comm: swapper/0 Tainted:
> > > G             L    5.2.0-rc5-next-20190621+ #1
> > > [   56.681357][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant
> > > DL385
> > > Gen10, BIOS A40 03/09/2018
> > > [   56.691356][   C16] RIP: 0010:subsys_find_device_by_id+0x168/0x1f0
> > > [   56.701334][   C16] Code: 48 85 c0 74 3e 48 8d 78 58 e8 14 77 ca ff 4d
> > > 8b 7e
> > > 58 4d 85 ff 74 2c 49 8d bf a0 03 00 00 e8 bf 75 ca ff 45 39 a7 a0 03 00 00
> > > <75>
> > > c9 4c 89 ff e8 0e 89 ff ff 48 85 c0 74 bc 48 89 df e8 21 3b 24
> > > [   56.721333][   C16] RSP: 0018:ffff888205b27c68 EFLAGS: 00000287
> > > ORIG_RAX:
> > > ffffffffffffff13
> > > [   56.721370][   C16] RAX: 0000000000000000 RBX: ffff888205b27c90 RCX:
> > > ffffffffb74c9dc1
> > > [   56.731370][   C16] RDX: 0000000000000003 RSI: dffffc0000000000 RDI:
> > > ffff8888774ec3e0
> > > [   56.741371][   C16] RBP: ffff888205b27cf8 R08: ffffed1040a7ac28 R09:
> > > ffffed1040a7ac27
> > > [   56.751335][   C16] R10: ffffed1040a7ac27 R11: ffff8882053d613b R12:
> > > 0000000000085c1b
> > > [   56.761334][   C16] R13: 1ffff11040b64f8e R14: ffff888450de4a20 R15:
> > > ffff8888774ec040
> > > [   56.761372][   C16] FS:  0000000000000000(0000)
> > > GS:ffff888454500000(0000)
> > > knlGS:0000000000000000
> > > [   56.771374][   C16] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > > [   56.781370][   C16] CR2: 0000000000000000 CR3: 00000007c9012000 CR4:
> > > 00000000001406a0
> > > [   56.791373][   C16] Call Trace:
> > > [   56.791373][   C16]  ? bus_find_device_by_name+0x20/0x20
> > > [   56.801334][   C16]  ? kobject_put+0x23/0x250
> > > [   56.801334][   C16]  walk_memory_blocks+0x6c/0xb8
> > > [   56.811333][   C16]  ? write_policy_show+0x40/0x40
> > > [   56.811353][   C16]  link_mem_sections+0x7e/0xa0
> > > [   56.811353][   C16]  ? unregister_memory_block_under_nodes+0x210/0x210
> > > [   56.821333][   C16]  ? __register_one_node+0x3bd/0x600
> > > [   56.831333][   C16]  topology_init+0xbf/0x126
> > > [   56.831355][   C16]  ? enable_cpu0_hotplug+0x1a/0x1a
> > > [   56.841334][   C16]  do_one_initcall+0xfe/0x45a
> > > [   56.841334][   C16]  ? initcall_blacklisted+0x150/0x150
> > > [   56.851333][   C16]  ? kasan_check_write+0x14/0x20
> > > [   56.851354][   C16]  ? up_write+0x75/0x140
> > > [   56.861333][   C16]  kernel_init_freeable+0x619/0x6ac
> > > [   56.861333][   C16]  ? rest_init+0x188/0x188
> > > [   56.861369][   C16]  kernel_init+0x11/0x138
> > > [   56.871333][   C16]  ? rest_init+0x188/0x188
> > > [   56.871354][   C16]  ret_from_fork+0x22/0x40
> > > [   64.601362][   C16] rcu: INFO: rcu_sched self-detected stall on CPU
> > > [   64.611335][   C16] rcu: 	16-....: (5958 ticks this GP)
> > > idle=37e/1/0x4000000000000002 softirq=27/27 fqs=3000 
> > > [   64.621334][   C16] 	(t=6002 jiffies g=-1079 q=25)
> > > [   64.621334][   C16] NMI backtrace for cpu 16
> > > [   64.621374][   C16] CPU: 16 PID: 1 Comm: swapper/0 Tainted:
> > > G             L    5.2.0-rc5-next-20190621+ #1
> > > [   64.631372][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant
> > > DL385
> > > Gen10, BIOS A40 03/09/2018
> > > [   64.641371][   C16] Call Trace:
> > > [   64.651337][   C16]  <IRQ>
> > > [   64.651376][   C16]  dump_stack+0x62/0x9a
> > > [   64.651376][   C16]  nmi_cpu_backtrace.cold.0+0x2e/0x33
> > > [   64.661337][   C16]  ? nmi_cpu_backtrace_handler+0x20/0x20
> > > [   64.661337][   C16]  nmi_trigger_cpumask_backtrace+0x1a6/0x1b9
> > > [   64.671353][   C16]  arch_trigger_cpumask_backtrace+0x19/0x20
> > > [   64.681366][   C16]  rcu_dump_cpu_stacks+0x18b/0x1d6
> > > [   64.681366][   C16]  rcu_sched_clock_irq.cold.64+0x368/0x791
> > > [   64.691336][   C16]  ? kasan_check_read+0x11/0x20
> > > [   64.691354][   C16]  ? __raise_softirq_irqoff+0x66/0x150
> > > [   64.701336][   C16]  update_process_times+0x2f/0x60
> > > [   64.701362][   C16]  tick_periodic+0x38/0xe0
> > > [   64.711334][   C16]  tick_handle_periodic+0x2e/0x80
> > > [   64.711353][   C16]  smp_apic_timer_interrupt+0xfb/0x370
> > > [   64.721367][   C16]  apic_timer_interrupt+0xf/0x20
> > > [   64.721367][   C16]  </IRQ>
> > > [   64.721367][   C16] RIP: 0010:_raw_spin_unlock_irqrestore+0x2f/0x40
> > > [   64.731370][   C16] Code: 55 48 89 e5 41 54 49 89 f4 be 01 00 00 00 53 
> > > 
> > 
> > @Qian Cai, unfortunately I can't reproduce.
> > 
> > If you get the chance, it would be great if you could retry with
> > 
> > diff --git a/drivers/base/memory.c b/drivers/base/memory.c
> > index 972c5336bebf..742f99ddd148 100644
> > --- a/drivers/base/memory.c
> > +++ b/drivers/base/memory.c
> > @@ -868,6 +868,9 @@ int walk_memory_blocks(unsigned long start, unsigned
> > long size,
> >         unsigned long block_id;
> >         int ret = 0;
> > 
> > +       if (!size)
> > +               return;
> > +
> >         for (block_id = start_block_id; block_id <= end_block_id;
> > block_id++) {
> >                 mem = find_memory_block_by_id(block_id);
> >                 if (!mem)
> > 
> > 
> > 
> > If both, start and size are 0, we would get a veeeery long loop. This
> > would mean that we have an online node that does not span any pages at
> > all (pgdat->node_start_pfn = 0, start_pfn + pgdat->node_spanned_pages = 0).
> > 
> 
> 
> ...trying to reproduce with QEMU (setting 0MB for the second node):
> 
> qemu-system-x86_64 --enable-kvm -m 4G,maxmem=20G,slots=2 \
> 	-smp sockets=2,cores=1 \
> 	-numa node,nodeid=0,cpus=0,mem=4G \
> 	-numa node,nodeid=1,cpus=1,mem=0 ...
> 
> I can indeed see that the node is online and
> "pgdat->node_start_pfn == 0 && start_pfn + pgdat->node_spanned_pages == 0".
> 
> However, the kernel segfaults in an unrelated code path, so I can't
> verify if this solves this problem:
> 
> [    0.313284] BUG: kernel NULL pointer dereference, address: 00000000000000a0
> [    0.313479] #PF: supervisor read access in kernel mode
> [    0.313479] #PF: error_code(0x0000) - not-present page
> [    0.313479] PGD 0 P4D 0 
> [    0.313479] Oops: 0000 [#1] SMP PTI
> [    0.313479] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.2.0-rc5-next-
> 20190620+ #56
> [    0.313479] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
> rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.4
> [    0.313479] RIP: 0010:bus_add_device+0x59/0x110
> [    0.313479] Code: 20 48 89 df e8 f8 b4 ff ff 41 89 c4 85 c0 0f 85 81 00 00
> 00 48 8b 53 50 48 85 d2 75 03 48 8b 135
> [    0.313479] RSP: 0000:ffffb4a6c0013e20 EFLAGS: 00010246
> [    0.313479] RAX: 0000000000000000 RBX: ffff8b61bac23800 RCX:
> 0000000000000000
> [    0.313479] RDX: ffff8b61bac29038 RSI: ffff8b61bac23800 RDI:
> ffff8b61bac23800
> [    0.313479] RBP: ffffffff9d2f4500 R08: 0000000000000000 R09:
> 0000000000000001
> [    0.313479] R10: 0000000000000000 R11: ffff8b61bad20878 R12:
> 0000000000000000
> [    0.313479] R13: 0000000000000000 R14: 0000000000000000 R15:
> 0000000000000000
> [    0.313479] FS:  0000000000000000(0000) GS:ffff8b61bba00000(0000)
> knlGS:0000000000000000
> [    0.313479] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [    0.313479] CR2: 00000000000000a0 CR3: 0000000013c24000 CR4:
> 00000000000006f0
> [    0.313479] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
> 0000000000000000
> [    0.313479] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7:
> 0000000000000400
> [    0.313479] Call Trace:
> [    0.313479]  device_add+0x304/0x660
> [    0.313479]  ? __init_waitqueue_head+0x31/0x50
> [    0.313479]  __register_one_node+0x67/0x170
> [    0.313479]  __try_online_node.cold+0x3e/0x78
> [    0.313479]  try_online_node+0x25/0x40
> [    0.313479]  do_cpu_up+0x36/0xc0
> [    0.313479]  smp_init+0x59/0xb3
> [    0.313479]  kernel_init_freeable+0x11a/0x247
> [    0.313479]  ? rest_init+0x23f/0x23f
> [    0.313479]  kernel_init+0x5/0xf1
> [    0.313479]  ret_from_fork+0x3a/0x50
> [    0.313479] Modules linked in:
> 
> Figuring out what goes wrong here (maybe QEMU creating a weird
> system configuration) is a different journey :)
> 

That is a separate issue need to revert,

"x86, numa: always initialize all possible nodes"

and then, you should be able to reproduce.


^ permalink raw reply

* Re: [PATCH v3 0/6] mm: Further memory block device cleanups
From: David Hildenbrand @ 2019-06-21 19:25 UTC (permalink / raw)
  To: Qian Cai, linux-kernel
  Cc: Oscar Salvador, Michal Hocko, Rafael J. Wysocki, Wei Yang,
	Keith Busch, linux-mm, Arun KS, Rashmica Gupta, Thomas Gleixner,
	Stephen Rothwell, Michael Neuling, Baoquan He, Pavel Tatashin,
	linux-acpi, Len Brown, Pavel Tatashin, Pavel Tatashin,
	Anshuman Khandual, mike.travis@hpe.com, linuxppc-dev,
	Mike Rapoport, Dan Williams, Vlastimil Babka, Oscar Salvador,
	Juergen Gross, Andrew Banman, Greg Kroah-Hartman,
	Rafael J. Wysocki, Johannes Weiner, Paul Mackerras, Andrew Morton,
	Mel Gorman
In-Reply-To: <1561144059.5154.52.camel@lca.pw>

On 21.06.19 21:07, Qian Cai wrote:
> On Fri, 2019-06-21 at 20:56 +0200, David Hildenbrand wrote:
>> On 21.06.19 20:24, David Hildenbrand wrote:
>>> On 21.06.19 17:15, Qian Cai wrote:
>>>> On Thu, 2019-06-20 at 20:31 +0200, David Hildenbrand wrote:
>>>>> @Andrew: Only patch 1, 4 and 6 changed compared to v1.
>>>>>
>>>>> Some further cleanups around memory block devices. Especially, clean up
>>>>> and simplify walk_memory_range(). Including some other minor cleanups.
>>>>>
>>>>> Compiled + tested on x86 with DIMMs under QEMU. Compile-tested on ppc64.
>>>>>
>>>>> v2 -> v3:
>>>>> - "mm/memory_hotplug: Rename walk_memory_range() and pass start+size .."
>>>>> -- Avoid warning on ppc.
>>>>> - "drivers/base/memory.c: Get rid of find_memory_block_hinted()"
>>>>> -- Fixup a comment regarding hinted devices.
>>>>>
>>>>> v1 -> v2:
>>>>> - "mm: Section numbers use the type "unsigned long""
>>>>> -- "unsigned long i" -> "unsigned long nr", in one case -> "int i"
>>>>> - "drivers/base/memory.c: Get rid of find_memory_block_hinted("
>>>>> -- Fix compilation error
>>>>> -- Get rid of the "hint" parameter completely
>>>>>
>>>>> David Hildenbrand (6):
>>>>>   mm: Section numbers use the type "unsigned long"
>>>>>   drivers/base/memory: Use "unsigned long" for block ids
>>>>>   mm: Make register_mem_sect_under_node() static
>>>>>   mm/memory_hotplug: Rename walk_memory_range() and pass start+size
>>>>>     instead of pfns
>>>>>   mm/memory_hotplug: Move and simplify walk_memory_blocks()
>>>>>   drivers/base/memory.c: Get rid of find_memory_block_hinted()
>>>>>
>>>>>  arch/powerpc/platforms/powernv/memtrace.c |  23 ++---
>>>>>  drivers/acpi/acpi_memhotplug.c            |  19 +---
>>>>>  drivers/base/memory.c                     | 120 +++++++++++++---------
>>>>>  drivers/base/node.c                       |   8 +-
>>>>>  include/linux/memory.h                    |   5 +-
>>>>>  include/linux/memory_hotplug.h            |   2 -
>>>>>  include/linux/mmzone.h                    |   4 +-
>>>>>  include/linux/node.h                      |   7 --
>>>>>  mm/memory_hotplug.c                       |  57 +---------
>>>>>  mm/sparse.c                               |  12 +--
>>>>>  10 files changed, 106 insertions(+), 151 deletions(-)
>>>>>
>>>>
>>>> This series causes a few machines are unable to boot triggering endless
>>>> soft
>>>> lockups. Reverted those commits fixed the issue.
>>>>
>>>> 97f4217d1da0 Revert "mm/memory_hotplug: rename walk_memory_range() and
>>>> pass
>>>> start+size instead of pfns"
>>>> c608eebf33c6 Revert "mm-memory_hotplug-rename-walk_memory_range-and-pass-
>>>> startsize-instead-of-pfns-fix"
>>>> 34b5e4ab7558 Revert "mm/memory_hotplug: move and simplify
>>>> walk_memory_blocks()"
>>>> 59a9f3eec5d1 Revert "drivers/base/memory.c: Get rid of
>>>> find_memory_block_hinted()"
>>>> 5cfcd52288b6 Revert "drivers-base-memoryc-get-rid-of-
>>>> find_memory_block_hinted-
>>>> v3"
>>>>
>>>> [    4.582081][    T1] ACPI FADT declares the system doesn't support PCIe
>>>> ASPM,
>>>> so disable it
>>>> [    4.590405][    T1] ACPI: bus type PCI registered
>>>> [    4.592908][    T1] PCI: MMCONFIG for domain 0000 [bus 00-ff] at [mem
>>>> 0x80000000-0x8fffffff] (base 0x80000000)
>>>> [    4.601860][    T1] PCI: MMCONFIG at [mem 0x80000000-0x8fffffff]
>>>> reserved in
>>>> E820
>>>> [    4.601860][    T1] PCI: Using configuration type 1 for base access
>>>> [   28.661336][   C16] watchdog: BUG: soft lockup - CPU#16 stuck for 22s!
>>>> [swapper/0:1]
>>>> [   28.671351][   C16] Modules linked in:
>>>> [   28.671354][   C16] CPU: 16 PID: 1 Comm: swapper/0 Not tainted 5.2.0-
>>>> rc5-
>>>> next-20190621+ #1
>>>> [   28.681366][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant
>>>> DL385
>>>> Gen10, BIOS A40 03/09/2018
>>>> [   28.691334][   C16] RIP: 0010:_raw_spin_unlock_irqrestore+0x2f/0x40
>>>> [   28.701334][   C16] Code: 55 48 89 e5 41 54 49 89 f4 be 01 00 00 00 53
>>>> 48 8b
>>>> 55 08 48 89 fb 48 8d 7f 18 e8 4c 89 7d ff 48 89 df e8 94 f9 7d ff 41 54 9d
>>>> <65>
>>>> ff 0d c2 44 8d 48 5b 41 5c 5d c3 0f 1f 44 00 00 0f 1f 44 00 00
>>>> [   28.711354][   C16] RSP: 0018:ffff888205b27bf8 EFLAGS: 00000246
>>>> ORIG_RAX:
>>>> ffffffffffffff13
>>>> [   28.721372][   C16] RAX: 0000000000000000 RBX: ffff8882053d6138 RCX:
>>>> ffffffffb6f2a3b8
>>>> [   28.731371][   C16] RDX: 1ffff11040a7ac27 RSI: dffffc0000000000 RDI:
>>>> ffff8882053d6138
>>>> [   28.741371][   C16] RBP: ffff888205b27c08 R08: ffffed1040a7ac28 R09:
>>>> ffffed1040a7ac27
>>>> [   28.751334][   C16] R10: ffffed1040a7ac27 R11: ffff8882053d613b R12:
>>>> 0000000000000246
>>>> [   28.751370][   C16] R13: ffff888205b27c98 R14: ffff8884504d0a20 R15:
>>>> 0000000000000000
>>>> [   28.761368][   C16] FS:  0000000000000000(0000)
>>>> GS:ffff888454500000(0000)
>>>> knlGS:0000000000000000
>>>> [   28.771373][   C16] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>>>> [   28.781334][   C16] CR2: 0000000000000000 CR3: 00000007c9012000 CR4:
>>>> 00000000001406a0
>>>> [   28.791333][   C16] Call Trace:
>>>> [   28.791374][   C16]  klist_next+0xd8/0x1c0
>>>> [   28.791374][   C16]  subsys_find_device_by_id+0x13b/0x1f0
>>>> [   28.801334][   C16]  ? bus_find_device_by_name+0x20/0x20
>>>> [   28.801370][   C16]  ? kobject_put+0x23/0x250
>>>> [   28.811333][   C16]  walk_memory_blocks+0x6c/0xb8
>>>> [   28.811353][   C16]  ? write_policy_show+0x40/0x40
>>>> [   28.821334][   C16]  link_mem_sections+0x7e/0xa0
>>>> [   28.821369][   C16]  ? unregister_memory_block_under_nodes+0x210/0x210
>>>> [   28.831353][   C16]  ? __register_one_node+0x3bd/0x600
>>>> [   28.831353][   C16]  topology_init+0xbf/0x126
>>>> [   28.841364][   C16]  ? enable_cpu0_hotplug+0x1a/0x1a
>>>> [   28.841368][   C16]  do_one_initcall+0xfe/0x45a
>>>> [   28.851334][   C16]  ? initcall_blacklisted+0x150/0x150
>>>> [   28.851353][   C16]  ? kasan_check_write+0x14/0x20
>>>> [   28.861333][   C16]  ? up_write+0x75/0x140
>>>> [   28.861369][   C16]  kernel_init_freeable+0x619/0x6ac
>>>> [   28.871333][   C16]  ? rest_init+0x188/0x188
>>>> [   28.871353][   C16]  kernel_init+0x11/0x138
>>>> [   28.881363][   C16]  ? rest_init+0x188/0x188
>>>> [   28.881363][   C16]  ret_from_fork+0x22/0x40
>>>> [   56.661336][   C16] watchdog: BUG: soft lockup - CPU#16 stuck for 22s!
>>>> [swapper/0:1]
>>>> [   56.671352][   C16] Modules linked in:
>>>> [   56.671354][   C16] CPU: 16 PID: 1 Comm: swapper/0 Tainted:
>>>> G             L    5.2.0-rc5-next-20190621+ #1
>>>> [   56.681357][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant
>>>> DL385
>>>> Gen10, BIOS A40 03/09/2018
>>>> [   56.691356][   C16] RIP: 0010:subsys_find_device_by_id+0x168/0x1f0
>>>> [   56.701334][   C16] Code: 48 85 c0 74 3e 48 8d 78 58 e8 14 77 ca ff 4d
>>>> 8b 7e
>>>> 58 4d 85 ff 74 2c 49 8d bf a0 03 00 00 e8 bf 75 ca ff 45 39 a7 a0 03 00 00
>>>> <75>
>>>> c9 4c 89 ff e8 0e 89 ff ff 48 85 c0 74 bc 48 89 df e8 21 3b 24
>>>> [   56.721333][   C16] RSP: 0018:ffff888205b27c68 EFLAGS: 00000287
>>>> ORIG_RAX:
>>>> ffffffffffffff13
>>>> [   56.721370][   C16] RAX: 0000000000000000 RBX: ffff888205b27c90 RCX:
>>>> ffffffffb74c9dc1
>>>> [   56.731370][   C16] RDX: 0000000000000003 RSI: dffffc0000000000 RDI:
>>>> ffff8888774ec3e0
>>>> [   56.741371][   C16] RBP: ffff888205b27cf8 R08: ffffed1040a7ac28 R09:
>>>> ffffed1040a7ac27
>>>> [   56.751335][   C16] R10: ffffed1040a7ac27 R11: ffff8882053d613b R12:
>>>> 0000000000085c1b
>>>> [   56.761334][   C16] R13: 1ffff11040b64f8e R14: ffff888450de4a20 R15:
>>>> ffff8888774ec040
>>>> [   56.761372][   C16] FS:  0000000000000000(0000)
>>>> GS:ffff888454500000(0000)
>>>> knlGS:0000000000000000
>>>> [   56.771374][   C16] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>>>> [   56.781370][   C16] CR2: 0000000000000000 CR3: 00000007c9012000 CR4:
>>>> 00000000001406a0
>>>> [   56.791373][   C16] Call Trace:
>>>> [   56.791373][   C16]  ? bus_find_device_by_name+0x20/0x20
>>>> [   56.801334][   C16]  ? kobject_put+0x23/0x250
>>>> [   56.801334][   C16]  walk_memory_blocks+0x6c/0xb8
>>>> [   56.811333][   C16]  ? write_policy_show+0x40/0x40
>>>> [   56.811353][   C16]  link_mem_sections+0x7e/0xa0
>>>> [   56.811353][   C16]  ? unregister_memory_block_under_nodes+0x210/0x210
>>>> [   56.821333][   C16]  ? __register_one_node+0x3bd/0x600
>>>> [   56.831333][   C16]  topology_init+0xbf/0x126
>>>> [   56.831355][   C16]  ? enable_cpu0_hotplug+0x1a/0x1a
>>>> [   56.841334][   C16]  do_one_initcall+0xfe/0x45a
>>>> [   56.841334][   C16]  ? initcall_blacklisted+0x150/0x150
>>>> [   56.851333][   C16]  ? kasan_check_write+0x14/0x20
>>>> [   56.851354][   C16]  ? up_write+0x75/0x140
>>>> [   56.861333][   C16]  kernel_init_freeable+0x619/0x6ac
>>>> [   56.861333][   C16]  ? rest_init+0x188/0x188
>>>> [   56.861369][   C16]  kernel_init+0x11/0x138
>>>> [   56.871333][   C16]  ? rest_init+0x188/0x188
>>>> [   56.871354][   C16]  ret_from_fork+0x22/0x40
>>>> [   64.601362][   C16] rcu: INFO: rcu_sched self-detected stall on CPU
>>>> [   64.611335][   C16] rcu: 	16-....: (5958 ticks this GP)
>>>> idle=37e/1/0x4000000000000002 softirq=27/27 fqs=3000 
>>>> [   64.621334][   C16] 	(t=6002 jiffies g=-1079 q=25)
>>>> [   64.621334][   C16] NMI backtrace for cpu 16
>>>> [   64.621374][   C16] CPU: 16 PID: 1 Comm: swapper/0 Tainted:
>>>> G             L    5.2.0-rc5-next-20190621+ #1
>>>> [   64.631372][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant
>>>> DL385
>>>> Gen10, BIOS A40 03/09/2018
>>>> [   64.641371][   C16] Call Trace:
>>>> [   64.651337][   C16]  <IRQ>
>>>> [   64.651376][   C16]  dump_stack+0x62/0x9a
>>>> [   64.651376][   C16]  nmi_cpu_backtrace.cold.0+0x2e/0x33
>>>> [   64.661337][   C16]  ? nmi_cpu_backtrace_handler+0x20/0x20
>>>> [   64.661337][   C16]  nmi_trigger_cpumask_backtrace+0x1a6/0x1b9
>>>> [   64.671353][   C16]  arch_trigger_cpumask_backtrace+0x19/0x20
>>>> [   64.681366][   C16]  rcu_dump_cpu_stacks+0x18b/0x1d6
>>>> [   64.681366][   C16]  rcu_sched_clock_irq.cold.64+0x368/0x791
>>>> [   64.691336][   C16]  ? kasan_check_read+0x11/0x20
>>>> [   64.691354][   C16]  ? __raise_softirq_irqoff+0x66/0x150
>>>> [   64.701336][   C16]  update_process_times+0x2f/0x60
>>>> [   64.701362][   C16]  tick_periodic+0x38/0xe0
>>>> [   64.711334][   C16]  tick_handle_periodic+0x2e/0x80
>>>> [   64.711353][   C16]  smp_apic_timer_interrupt+0xfb/0x370
>>>> [   64.721367][   C16]  apic_timer_interrupt+0xf/0x20
>>>> [   64.721367][   C16]  </IRQ>
>>>> [   64.721367][   C16] RIP: 0010:_raw_spin_unlock_irqrestore+0x2f/0x40
>>>> [   64.731370][   C16] Code: 55 48 89 e5 41 54 49 89 f4 be 01 00 00 00 53 
>>>>
>>>
>>> @Qian Cai, unfortunately I can't reproduce.
>>>
>>> If you get the chance, it would be great if you could retry with
>>>
>>> diff --git a/drivers/base/memory.c b/drivers/base/memory.c
>>> index 972c5336bebf..742f99ddd148 100644
>>> --- a/drivers/base/memory.c
>>> +++ b/drivers/base/memory.c
>>> @@ -868,6 +868,9 @@ int walk_memory_blocks(unsigned long start, unsigned
>>> long size,
>>>         unsigned long block_id;
>>>         int ret = 0;
>>>
>>> +       if (!size)
>>> +               return;
>>> +
>>>         for (block_id = start_block_id; block_id <= end_block_id;
>>> block_id++) {
>>>                 mem = find_memory_block_by_id(block_id);
>>>                 if (!mem)
>>>
>>>
>>>
>>> If both, start and size are 0, we would get a veeeery long loop. This
>>> would mean that we have an online node that does not span any pages at
>>> all (pgdat->node_start_pfn = 0, start_pfn + pgdat->node_spanned_pages = 0).
>>>
>>
>>
>> ...trying to reproduce with QEMU (setting 0MB for the second node):
>>
>> qemu-system-x86_64 --enable-kvm -m 4G,maxmem=20G,slots=2 \
>> 	-smp sockets=2,cores=1 \
>> 	-numa node,nodeid=0,cpus=0,mem=4G \
>> 	-numa node,nodeid=1,cpus=1,mem=0 ...
>>
>> I can indeed see that the node is online and
>> "pgdat->node_start_pfn == 0 && start_pfn + pgdat->node_spanned_pages == 0".
>>
>> However, the kernel segfaults in an unrelated code path, so I can't
>> verify if this solves this problem:
>>
>> [    0.313284] BUG: kernel NULL pointer dereference, address: 00000000000000a0
>> [    0.313479] #PF: supervisor read access in kernel mode
>> [    0.313479] #PF: error_code(0x0000) - not-present page
>> [    0.313479] PGD 0 P4D 0 
>> [    0.313479] Oops: 0000 [#1] SMP PTI
>> [    0.313479] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.2.0-rc5-next-
>> 20190620+ #56
>> [    0.313479] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
>> rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.4
>> [    0.313479] RIP: 0010:bus_add_device+0x59/0x110
>> [    0.313479] Code: 20 48 89 df e8 f8 b4 ff ff 41 89 c4 85 c0 0f 85 81 00 00
>> 00 48 8b 53 50 48 85 d2 75 03 48 8b 135
>> [    0.313479] RSP: 0000:ffffb4a6c0013e20 EFLAGS: 00010246
>> [    0.313479] RAX: 0000000000000000 RBX: ffff8b61bac23800 RCX:
>> 0000000000000000
>> [    0.313479] RDX: ffff8b61bac29038 RSI: ffff8b61bac23800 RDI:
>> ffff8b61bac23800
>> [    0.313479] RBP: ffffffff9d2f4500 R08: 0000000000000000 R09:
>> 0000000000000001
>> [    0.313479] R10: 0000000000000000 R11: ffff8b61bad20878 R12:
>> 0000000000000000
>> [    0.313479] R13: 0000000000000000 R14: 0000000000000000 R15:
>> 0000000000000000
>> [    0.313479] FS:  0000000000000000(0000) GS:ffff8b61bba00000(0000)
>> knlGS:0000000000000000
>> [    0.313479] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>> [    0.313479] CR2: 00000000000000a0 CR3: 0000000013c24000 CR4:
>> 00000000000006f0
>> [    0.313479] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
>> 0000000000000000
>> [    0.313479] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7:
>> 0000000000000400
>> [    0.313479] Call Trace:
>> [    0.313479]  device_add+0x304/0x660
>> [    0.313479]  ? __init_waitqueue_head+0x31/0x50
>> [    0.313479]  __register_one_node+0x67/0x170
>> [    0.313479]  __try_online_node.cold+0x3e/0x78
>> [    0.313479]  try_online_node+0x25/0x40
>> [    0.313479]  do_cpu_up+0x36/0xc0
>> [    0.313479]  smp_init+0x59/0xb3
>> [    0.313479]  kernel_init_freeable+0x11a/0x247
>> [    0.313479]  ? rest_init+0x23f/0x23f
>> [    0.313479]  kernel_init+0x5/0xf1
>> [    0.313479]  ret_from_fork+0x3a/0x50
>> [    0.313479] Modules linked in:
>>
>> Figuring out what goes wrong here (maybe QEMU creating a weird
>> system configuration) is a different journey :)
>>
> 
> That is a separate issue need to revert,
> 
> "x86, numa: always initialize all possible nodes"
> 
> and then, you should be able to reproduce.
> 

Thanks, reproduced and verified that this is the fix.

-- 

Thanks,

David / dhildenb

^ permalink raw reply

* Re: [PATCH v3 0/6] mm: Further memory block device cleanups
From: Qian Cai @ 2019-06-21 19:29 UTC (permalink / raw)
  To: David Hildenbrand, linux-kernel
  Cc: Oscar Salvador, Michal Hocko, Rafael J. Wysocki, Wei Yang,
	Keith Busch, linux-mm, Arun KS, Rashmica Gupta, Thomas Gleixner,
	Stephen Rothwell, Michael Neuling, Baoquan He, Pavel Tatashin,
	linux-acpi, Len Brown, Pavel Tatashin, Pavel Tatashin,
	Anshuman Khandual, mike.travis@hpe.com, linuxppc-dev,
	Mike Rapoport, Dan Williams, Vlastimil Babka, Oscar Salvador,
	Juergen Gross, Andrew Banman, Greg Kroah-Hartman,
	Rafael J. Wysocki, Johannes Weiner, Paul Mackerras, Andrew Morton,
	Mel Gorman
In-Reply-To: <1c2edc22-afd7-2211-c4c7-40e54e5007e8@redhat.com>

On Fri, 2019-06-21 at 20:24 +0200, David Hildenbrand wrote:
> On 21.06.19 17:15, Qian Cai wrote:
> > On Thu, 2019-06-20 at 20:31 +0200, David Hildenbrand wrote:
> > > @Andrew: Only patch 1, 4 and 6 changed compared to v1.
> > > 
> > > Some further cleanups around memory block devices. Especially, clean up
> > > and simplify walk_memory_range(). Including some other minor cleanups.
> > > 
> > > Compiled + tested on x86 with DIMMs under QEMU. Compile-tested on ppc64.
> > > 
> > > v2 -> v3:
> > > - "mm/memory_hotplug: Rename walk_memory_range() and pass start+size .."
> > > -- Avoid warning on ppc.
> > > - "drivers/base/memory.c: Get rid of find_memory_block_hinted()"
> > > -- Fixup a comment regarding hinted devices.
> > > 
> > > v1 -> v2:
> > > - "mm: Section numbers use the type "unsigned long""
> > > -- "unsigned long i" -> "unsigned long nr", in one case -> "int i"
> > > - "drivers/base/memory.c: Get rid of find_memory_block_hinted("
> > > -- Fix compilation error
> > > -- Get rid of the "hint" parameter completely
> > > 
> > > David Hildenbrand (6):
> > >   mm: Section numbers use the type "unsigned long"
> > >   drivers/base/memory: Use "unsigned long" for block ids
> > >   mm: Make register_mem_sect_under_node() static
> > >   mm/memory_hotplug: Rename walk_memory_range() and pass start+size
> > >     instead of pfns
> > >   mm/memory_hotplug: Move and simplify walk_memory_blocks()
> > >   drivers/base/memory.c: Get rid of find_memory_block_hinted()
> > > 
> > >  arch/powerpc/platforms/powernv/memtrace.c |  23 ++---
> > >  drivers/acpi/acpi_memhotplug.c            |  19 +---
> > >  drivers/base/memory.c                     | 120 +++++++++++++---------
> > >  drivers/base/node.c                       |   8 +-
> > >  include/linux/memory.h                    |   5 +-
> > >  include/linux/memory_hotplug.h            |   2 -
> > >  include/linux/mmzone.h                    |   4 +-
> > >  include/linux/node.h                      |   7 --
> > >  mm/memory_hotplug.c                       |  57 +---------
> > >  mm/sparse.c                               |  12 +--
> > >  10 files changed, 106 insertions(+), 151 deletions(-)
> > > 
> > 
> > This series causes a few machines are unable to boot triggering endless soft
> > lockups. Reverted those commits fixed the issue.
> > 
> > 97f4217d1da0 Revert "mm/memory_hotplug: rename walk_memory_range() and pass
> > start+size instead of pfns"
> > c608eebf33c6 Revert "mm-memory_hotplug-rename-walk_memory_range-and-pass-
> > startsize-instead-of-pfns-fix"
> > 34b5e4ab7558 Revert "mm/memory_hotplug: move and simplify
> > walk_memory_blocks()"
> > 59a9f3eec5d1 Revert "drivers/base/memory.c: Get rid of
> > find_memory_block_hinted()"
> > 5cfcd52288b6 Revert "drivers-base-memoryc-get-rid-of-
> > find_memory_block_hinted-
> > v3"
> > 
> > [    4.582081][    T1] ACPI FADT declares the system doesn't support PCIe
> > ASPM,
> > so disable it
> > [    4.590405][    T1] ACPI: bus type PCI registered
> > [    4.592908][    T1] PCI: MMCONFIG for domain 0000 [bus 00-ff] at [mem
> > 0x80000000-0x8fffffff] (base 0x80000000)
> > [    4.601860][    T1] PCI: MMCONFIG at [mem 0x80000000-0x8fffffff] reserved
> > in
> > E820
> > [    4.601860][    T1] PCI: Using configuration type 1 for base access
> > [   28.661336][   C16] watchdog: BUG: soft lockup - CPU#16 stuck for 22s!
> > [swapper/0:1]
> > [   28.671351][   C16] Modules linked in:
> > [   28.671354][   C16] CPU: 16 PID: 1 Comm: swapper/0 Not tainted 5.2.0-rc5-
> > next-20190621+ #1
> > [   28.681366][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant
> > DL385
> > Gen10, BIOS A40 03/09/2018
> > [   28.691334][   C16] RIP: 0010:_raw_spin_unlock_irqrestore+0x2f/0x40
> > [   28.701334][   C16] Code: 55 48 89 e5 41 54 49 89 f4 be 01 00 00 00 53 48
> > 8b
> > 55 08 48 89 fb 48 8d 7f 18 e8 4c 89 7d ff 48 89 df e8 94 f9 7d ff 41 54 9d
> > <65>
> > ff 0d c2 44 8d 48 5b 41 5c 5d c3 0f 1f 44 00 00 0f 1f 44 00 00
> > [   28.711354][   C16] RSP: 0018:ffff888205b27bf8 EFLAGS: 00000246 ORIG_RAX:
> > ffffffffffffff13
> > [   28.721372][   C16] RAX: 0000000000000000 RBX: ffff8882053d6138 RCX:
> > ffffffffb6f2a3b8
> > [   28.731371][   C16] RDX: 1ffff11040a7ac27 RSI: dffffc0000000000 RDI:
> > ffff8882053d6138
> > [   28.741371][   C16] RBP: ffff888205b27c08 R08: ffffed1040a7ac28 R09:
> > ffffed1040a7ac27
> > [   28.751334][   C16] R10: ffffed1040a7ac27 R11: ffff8882053d613b R12:
> > 0000000000000246
> > [   28.751370][   C16] R13: ffff888205b27c98 R14: ffff8884504d0a20 R15:
> > 0000000000000000
> > [   28.761368][   C16] FS:  0000000000000000(0000) GS:ffff888454500000(0000)
> > knlGS:0000000000000000
> > [   28.771373][   C16] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > [   28.781334][   C16] CR2: 0000000000000000 CR3: 00000007c9012000 CR4:
> > 00000000001406a0
> > [   28.791333][   C16] Call Trace:
> > [   28.791374][   C16]  klist_next+0xd8/0x1c0
> > [   28.791374][   C16]  subsys_find_device_by_id+0x13b/0x1f0
> > [   28.801334][   C16]  ? bus_find_device_by_name+0x20/0x20
> > [   28.801370][   C16]  ? kobject_put+0x23/0x250
> > [   28.811333][   C16]  walk_memory_blocks+0x6c/0xb8
> > [   28.811353][   C16]  ? write_policy_show+0x40/0x40
> > [   28.821334][   C16]  link_mem_sections+0x7e/0xa0
> > [   28.821369][   C16]  ? unregister_memory_block_under_nodes+0x210/0x210
> > [   28.831353][   C16]  ? __register_one_node+0x3bd/0x600
> > [   28.831353][   C16]  topology_init+0xbf/0x126
> > [   28.841364][   C16]  ? enable_cpu0_hotplug+0x1a/0x1a
> > [   28.841368][   C16]  do_one_initcall+0xfe/0x45a
> > [   28.851334][   C16]  ? initcall_blacklisted+0x150/0x150
> > [   28.851353][   C16]  ? kasan_check_write+0x14/0x20
> > [   28.861333][   C16]  ? up_write+0x75/0x140
> > [   28.861369][   C16]  kernel_init_freeable+0x619/0x6ac
> > [   28.871333][   C16]  ? rest_init+0x188/0x188
> > [   28.871353][   C16]  kernel_init+0x11/0x138
> > [   28.881363][   C16]  ? rest_init+0x188/0x188
> > [   28.881363][   C16]  ret_from_fork+0x22/0x40
> > [   56.661336][   C16] watchdog: BUG: soft lockup - CPU#16 stuck for 22s!
> > [swapper/0:1]
> > [   56.671352][   C16] Modules linked in:
> > [   56.671354][   C16] CPU: 16 PID: 1 Comm: swapper/0 Tainted:
> > G             L    5.2.0-rc5-next-20190621+ #1
> > [   56.681357][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant
> > DL385
> > Gen10, BIOS A40 03/09/2018
> > [   56.691356][   C16] RIP: 0010:subsys_find_device_by_id+0x168/0x1f0
> > [   56.701334][   C16] Code: 48 85 c0 74 3e 48 8d 78 58 e8 14 77 ca ff 4d 8b
> > 7e
> > 58 4d 85 ff 74 2c 49 8d bf a0 03 00 00 e8 bf 75 ca ff 45 39 a7 a0 03 00 00
> > <75>
> > c9 4c 89 ff e8 0e 89 ff ff 48 85 c0 74 bc 48 89 df e8 21 3b 24
> > [   56.721333][   C16] RSP: 0018:ffff888205b27c68 EFLAGS: 00000287 ORIG_RAX:
> > ffffffffffffff13
> > [   56.721370][   C16] RAX: 0000000000000000 RBX: ffff888205b27c90 RCX:
> > ffffffffb74c9dc1
> > [   56.731370][   C16] RDX: 0000000000000003 RSI: dffffc0000000000 RDI:
> > ffff8888774ec3e0
> > [   56.741371][   C16] RBP: ffff888205b27cf8 R08: ffffed1040a7ac28 R09:
> > ffffed1040a7ac27
> > [   56.751335][   C16] R10: ffffed1040a7ac27 R11: ffff8882053d613b R12:
> > 0000000000085c1b
> > [   56.761334][   C16] R13: 1ffff11040b64f8e R14: ffff888450de4a20 R15:
> > ffff8888774ec040
> > [   56.761372][   C16] FS:  0000000000000000(0000) GS:ffff888454500000(0000)
> > knlGS:0000000000000000
> > [   56.771374][   C16] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > [   56.781370][   C16] CR2: 0000000000000000 CR3: 00000007c9012000 CR4:
> > 00000000001406a0
> > [   56.791373][   C16] Call Trace:
> > [   56.791373][   C16]  ? bus_find_device_by_name+0x20/0x20
> > [   56.801334][   C16]  ? kobject_put+0x23/0x250
> > [   56.801334][   C16]  walk_memory_blocks+0x6c/0xb8
> > [   56.811333][   C16]  ? write_policy_show+0x40/0x40
> > [   56.811353][   C16]  link_mem_sections+0x7e/0xa0
> > [   56.811353][   C16]  ? unregister_memory_block_under_nodes+0x210/0x210
> > [   56.821333][   C16]  ? __register_one_node+0x3bd/0x600
> > [   56.831333][   C16]  topology_init+0xbf/0x126
> > [   56.831355][   C16]  ? enable_cpu0_hotplug+0x1a/0x1a
> > [   56.841334][   C16]  do_one_initcall+0xfe/0x45a
> > [   56.841334][   C16]  ? initcall_blacklisted+0x150/0x150
> > [   56.851333][   C16]  ? kasan_check_write+0x14/0x20
> > [   56.851354][   C16]  ? up_write+0x75/0x140
> > [   56.861333][   C16]  kernel_init_freeable+0x619/0x6ac
> > [   56.861333][   C16]  ? rest_init+0x188/0x188
> > [   56.861369][   C16]  kernel_init+0x11/0x138
> > [   56.871333][   C16]  ? rest_init+0x188/0x188
> > [   56.871354][   C16]  ret_from_fork+0x22/0x40
> > [   64.601362][   C16] rcu: INFO: rcu_sched self-detected stall on CPU
> > [   64.611335][   C16] rcu: 	16-....: (5958 ticks this GP)
> > idle=37e/1/0x4000000000000002 softirq=27/27 fqs=3000 
> > [   64.621334][   C16] 	(t=6002 jiffies g=-1079 q=25)
> > [   64.621334][   C16] NMI backtrace for cpu 16
> > [   64.621374][   C16] CPU: 16 PID: 1 Comm: swapper/0 Tainted:
> > G             L    5.2.0-rc5-next-20190621+ #1
> > [   64.631372][   C16] Hardware name: HPE ProLiant DL385 Gen10/ProLiant
> > DL385
> > Gen10, BIOS A40 03/09/2018
> > [   64.641371][   C16] Call Trace:
> > [   64.651337][   C16]  <IRQ>
> > [   64.651376][   C16]  dump_stack+0x62/0x9a
> > [   64.651376][   C16]  nmi_cpu_backtrace.cold.0+0x2e/0x33
> > [   64.661337][   C16]  ? nmi_cpu_backtrace_handler+0x20/0x20
> > [   64.661337][   C16]  nmi_trigger_cpumask_backtrace+0x1a6/0x1b9
> > [   64.671353][   C16]  arch_trigger_cpumask_backtrace+0x19/0x20
> > [   64.681366][   C16]  rcu_dump_cpu_stacks+0x18b/0x1d6
> > [   64.681366][   C16]  rcu_sched_clock_irq.cold.64+0x368/0x791
> > [   64.691336][   C16]  ? kasan_check_read+0x11/0x20
> > [   64.691354][   C16]  ? __raise_softirq_irqoff+0x66/0x150
> > [   64.701336][   C16]  update_process_times+0x2f/0x60
> > [   64.701362][   C16]  tick_periodic+0x38/0xe0
> > [   64.711334][   C16]  tick_handle_periodic+0x2e/0x80
> > [   64.711353][   C16]  smp_apic_timer_interrupt+0xfb/0x370
> > [   64.721367][   C16]  apic_timer_interrupt+0xf/0x20
> > [   64.721367][   C16]  </IRQ>
> > [   64.721367][   C16] RIP: 0010:_raw_spin_unlock_irqrestore+0x2f/0x40
> > [   64.731370][   C16] Code: 55 48 89 e5 41 54 49 89 f4 be 01 00 00 00 53 
> > 
> 
> @Qian Cai, unfortunately I can't reproduce.
> 
> If you get the chance, it would be great if you could retry with
> 
> diff --git a/drivers/base/memory.c b/drivers/base/memory.c
> index 972c5336bebf..742f99ddd148 100644
> --- a/drivers/base/memory.c
> +++ b/drivers/base/memory.c
> @@ -868,6 +868,9 @@ int walk_memory_blocks(unsigned long start, unsigned
> long size,
>         unsigned long block_id;
>         int ret = 0;
> 
> +       if (!size)
> +               return;
> +
>         for (block_id = start_block_id; block_id <= end_block_id;
> block_id++) {
>                 mem = find_memory_block_by_id(block_id);
>                 if (!mem)
> 
> 
> 
> If both, start and size are 0, we would get a veeeery long loop. This
> would mean that we have an online node that does not span any pages at
> all (pgdat->node_start_pfn = 0, start_pfn + pgdat->node_spanned_pages = 0).
> 

It works fine here.

^ permalink raw reply

* [PATCH] powerpc/64s/exception: Fix machine check early corrupting AMR
From: Nicholas Piggin @ 2019-06-21 22:55 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Nicholas Piggin

The early machine check runs in real mode, so locking is unnecessary.
Worse, the windup does not restore AMR, so this can result in a false
KUAP fault after a recoverable machine check hits inside a user copy
operation.

Fix this similarly to HMI by just avoiding the kuap lock in the
early machine check handler (it will be set by the late handler that
runs in virtual mode if that runs). If the virtual mode handler is
reached, it will lock and restore the AMR.

Fixes: 890274c2dc4c0 ("powerpc/64s: Implement KUAP for Radix MMU")
Cc: Russell Currey <ruscur@russell.cc>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
 arch/powerpc/kernel/exceptions-64s.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 6b86055e5251..73ba246ca11d 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -315,7 +315,7 @@ TRAMP_REAL_BEGIN(machine_check_common_early)
 	mfspr	r11,SPRN_DSISR		/* Save DSISR */
 	std	r11,_DSISR(r1)
 	std	r9,_CCR(r1)		/* Save CR in stackframe */
-	kuap_save_amr_and_lock r9, r10, cr1
+	/* We don't touch AMR here, we never go to virtual mode */
 	/* Save r9 through r13 from EXMC save area to stack frame. */
 	EXCEPTION_PROLOG_COMMON_2(PACA_EXMC)
 	mfmsr	r11			/* get MSR value */
-- 
2.20.1


^ permalink raw reply related

* Re: [PATCH 11/13] powerpc/64s: Save r13 in machine_check_common_early
From: Nicholas Piggin @ 2019-06-21 23:21 UTC (permalink / raw)
  To: mahesh, Santosh Sivaraj
  Cc: Aneesh Kumar K.V, linuxppc-dev, Mahesh Salgaonkar,
	Chandan Rajendra, Reza Arbab
In-Reply-To: <20190621114725.xg6cogv4ecejz6pj@in.ibm.com>

Mahesh J Salgaonkar's on June 21, 2019 9:47 pm:
> On 2019-06-21 06:27:15 Fri, Santosh Sivaraj wrote:
>> From: Reza Arbab <arbab@linux.ibm.com>
>> 
>> Testing my memcpy_mcsafe() work in progress with an injected UE, I get
>> an error like this immediately after the function returns:
>> 
>> BUG: Unable to handle kernel data access at 0x7fff84dec8f8
>> Faulting instruction address: 0xc0080000009c00b0
>> Oops: Kernel access of bad area, sig: 11 [#1]
>> LE PAGE_SIZE=64K MMU=Radix MMU=Hash SMP NR_CPUS=2048 NUMA PowerNV
>> Modules linked in: mce(O+) vmx_crypto crc32c_vpmsum
>> CPU: 0 PID: 1375 Comm: modprobe Tainted: G           O      5.1.0-rc6 #267
>> NIP:  c0080000009c00b0 LR: c0080000009c00a8 CTR: c000000000095f90
>> REGS: c0000000ee197790 TRAP: 0300   Tainted: G           O       (5.1.0-rc6)
>> MSR:  900000000280b033 <SF,HV,VEC,VSX,EE,FP,ME,IR,DR,RI,LE>  CR: 88002826  XER: 00040000
>> CFAR: c000000000095f8c DAR: 00007fff84dec8f8 DSISR: 40000000 IRQMASK: 0
>> GPR00: 000000006c6c6568 c0000000ee197a20 c0080000009c8400 fffffffffffffff2
>> GPR04: c0080000009c02e0 0000000000000006 0000000000000000 c000000003c834c8
>> GPR08: 0080000000000000 776a6681b7fb5100 0000000000000000 c0080000009c01c8
>> GPR12: c000000000095f90 00007fff84debc00 000000004d071440 0000000000000000
>> GPR16: 0000000100000601 c0080000009e0000 c000000000c98dd8 c000000000c98d98
>> GPR20: c000000003bba970 c0080000009c04d0 c0080000009c0618 c0000000001e5820
>> GPR24: 0000000000000000 0000000000000100 0000000000000001 c000000003bba958
>> GPR28: c0080000009c02e8 c0080000009c0318 c0080000009c02e0 0000000000000000
>> NIP [c0080000009c00b0] cause_ue+0xa8/0xe8 [mce]
>> LR [c0080000009c00a8] cause_ue+0xa0/0xe8 [mce]
>> 
>> To fix, ensure that r13 is properly restored after an MCE.
>> 
>> Signed-off-by: Reza Arbab <arbab@linux.ibm.com>
>> ---
>>  arch/powerpc/kernel/exceptions-64s.S | 1 +
>>  1 file changed, 1 insertion(+)
>> 
>> diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
>> index 311f1392a2ec..932d8d05892c 100644
>> --- a/arch/powerpc/kernel/exceptions-64s.S
>> +++ b/arch/powerpc/kernel/exceptions-64s.S
>> @@ -265,6 +265,7 @@ ALT_FTR_SECTION_END_IFSET(CPU_FTR_HVMODE)
>>  EXC_REAL_END(machine_check, 0x200, 0x100)
>>  EXC_VIRT_NONE(0x4200, 0x100)
>>  TRAMP_REAL_BEGIN(machine_check_common_early)
>> +	SET_SCRATCH0(r13)		/* save r13 */
>>  	EXCEPTION_PROLOG_1(PACA_EXMC, NOTEST, 0x200)
>>  	/*
>>  	 * Register contents:
> 
> We do save r13 before we call machine_check_common_early(). I don't
> think I understood clearly how this change fixes the issue you are
> seeing. What am I missing here ?
> 
> Above change will push the paca pointer to scratch0 overwriting the
> original saved r13.
> 
> EXC_REAL_BEGIN(machine_check, 0x200, 0x100)
> 	/* This is moved out of line as it can be patched by FW, but
> 	 * some code path might still want to branch into the original
> 	 * vector
> 	 */
> 	SET_SCRATCH0(r13)		/* save r13 */
> 	EXCEPTION_PROLOG_0(PACA_EXMC)
> BEGIN_FTR_SECTION
> 	b	machine_check_common_early

Yep, from the stack trace, r13 is corrupted. So r13 must have got
corrupted before the machine check and this just happens to have
corrected it.

How does cause_ue work? It or memcpy_mcsafe must be corrupting
r13.

Thanks,
Nick


^ permalink raw reply

* Re: [PATCH v3 0/6] mm: Further memory block device cleanups
From: Andrew Morton @ 2019-06-21 23:42 UTC (permalink / raw)
  To: David Hildenbrand
  Cc: Oscar Salvador, Michal Hocko, Rafael J. Wysocki, Wei Yang,
	Keith Busch, linux-mm, Arun KS, Rashmica Gupta, Thomas Gleixner,
	Stephen Rothwell, Michael Neuling, Baoquan He, Pavel Tatashin,
	linux-acpi, Len Brown, Pavel Tatashin, Pavel Tatashin,
	Anshuman Khandual, mike.travis@hpe.com, linuxppc-dev,
	Mike Rapoport, Qian Cai, Dan Williams, Vlastimil Babka,
	Oscar Salvador, Juergen Gross, Andrew Banman, Greg Kroah-Hartman,
	Rafael J. Wysocki, linux-kernel, Johannes Weiner, Paul Mackerras,
	Mel Gorman
In-Reply-To: <1c2edc22-afd7-2211-c4c7-40e54e5007e8@redhat.com>

On Fri, 21 Jun 2019 20:24:59 +0200 David Hildenbrand <david@redhat.com> wrote:

> @Qian Cai, unfortunately I can't reproduce.
> 
> If you get the chance, it would be great if you could retry with
> 
> diff --git a/drivers/base/memory.c b/drivers/base/memory.c
> index 972c5336bebf..742f99ddd148 100644
> --- a/drivers/base/memory.c
> +++ b/drivers/base/memory.c
> @@ -868,6 +868,9 @@ int walk_memory_blocks(unsigned long start, unsigned
> long size,
>         unsigned long block_id;
>         int ret = 0;
> 
> +       if (!size)
> +               return;
> +
>         for (block_id = start_block_id; block_id <= end_block_id;
> block_id++) {
>                 mem = find_memory_block_by_id(block_id);
>                 if (!mem)
> 
> 
> 
> If both, start and size are 0, we would get a veeeery long loop. This
> would mean that we have an online node that does not span any pages at
> all (pgdat->node_start_pfn = 0, start_pfn + pgdat->node_spanned_pages = 0).

I think I'll make that a `return 0' and I won't drop patches 4-6 for
now, as we appear to have this fixed.



From: David Hildenbrand <david@redhat.com>
Subject: drivers-base-memoryc-get-rid-of-find_memory_block_hinted-v3-fix

handle zero-length walks

Link: http://lkml.kernel.org/r/1c2edc22-afd7-2211-c4c7-40e54e5007e8@redhat.com
Reported-by: Qian Cai <cai@lca.pw>
Tested-by: Qian Cai <cai@lca.pw>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 drivers/base/memory.c |    3 +++
 1 file changed, 3 insertions(+)

--- a/drivers/base/memory.c~drivers-base-memoryc-get-rid-of-find_memory_block_hinted-v3-fix
+++ a/drivers/base/memory.c
@@ -866,6 +866,9 @@ int walk_memory_blocks(unsigned long sta
 	unsigned long block_id;
 	int ret = 0;
 
+	if (!size)
+		return 0;
+
 	for (block_id = start_block_id; block_id <= end_block_id; block_id++) {
 		mem = find_memory_block_by_id(block_id);
 		if (!mem)



^ permalink raw reply

* Re: [PATCH 16/16] mm: pass get_user_pages_fast iterator arguments in a structure
From: Nicholas Piggin @ 2019-06-21 23:55 UTC (permalink / raw)
  To: Christoph Hellwig, Linus Torvalds
  Cc: Rich Felker, Yoshinori Sato, Linux-sh list, James Hogan,
	the arch/x86 maintainers, Linux List Kernel Mailing, linux-mips,
	Linux-MM, Khalid Aziz, Paul Burton, Andrey Konovalov, sparclinux,
	Paul Mackerras, linuxppc-dev, David S. Miller
In-Reply-To: <20190621081501.GA17718@lst.de>

Christoph Hellwig's on June 21, 2019 6:15 pm:
> On Thu, Jun 20, 2019 at 10:21:46AM -0700, Linus Torvalds wrote:
>> Hmm. Honestly, I've never seen anything like that in any kernel profiles.
>> 
>> Compared to the problems I _do_ see (which is usually the obvious
>> cache misses, and locking), it must either be in the noise or it's
>> some problem specific to whatever CPU you are doing performance work
>> on?
>> 
>> I've occasionally seen pipeline hiccups in profiles, but it's usually
>> been either some serious glass jaw of the core, or it's been something
>> really stupid we did (or occasionally that the compiler did: one in
>> particular I remember was how there was a time when gcc would narrow
>> stores when it could, so if you set a bit in a word, it would do it
>> with a byte store, and then when you read the whole word afterwards
>> you'd get a major pipeline stall and it happened to show up in some
>> really hot paths).
> 
> I've not seen any difference in the GUP bench output here ar all.
> 
> But I'm fine with skipping this patch for now, I have a potential
> series I'm looking into that would benefit a lot from it, but we
> can discusss it in that context and make sure all the other works gets in
> in time.
> 

If you can, that would be good. I don't like to object based on
handwaving so I'll see if I can find any benchmarks that will give
better confidence. Those old TPC-C tests were good, and there was
some DB2 workload that was the reason I added gup fast in the first
place. I'll do some digging.

Thanks,
Nick

^ permalink raw reply

* Re: [PATCH 7/7] powerpc/kprobes: Allow probing on any ftrace address
From: Joe Perches @ 2019-06-22  3:49 UTC (permalink / raw)
  To: Masami Hiramatsu, Naveen N. Rao
  Cc: linux-kernel, Nicholas Piggin, Steven Rostedt, linuxppc-dev,
	Ingo Molnar
In-Reply-To: <20190621235034.acc00fc3e2b2c7e89caa1fd5@kernel.org>

On Fri, 2019-06-21 at 23:50 +0900, Masami Hiramatsu wrote:
> On Tue, 18 Jun 2019 20:17:06 +0530
> "Naveen N. Rao" <naveen.n.rao@linux.vnet.ibm.com> wrote:

trivia:

> > diff --git a/arch/powerpc/kernel/kprobes-ftrace.c b/arch/powerpc/kernel/kprobes-ftrace.c
[]
> > @@ -57,6 +82,11 @@ NOKPROBE_SYMBOL(kprobe_ftrace_handler);
> >  
> >  int arch_prepare_kprobe_ftrace(struct kprobe *p)
> >  {
> > +	if ((unsigned long)p->addr & 0x03) {
> > +		printk("Attempt to register kprobe at an unaligned address\n");

Please use the appropriate KERN_<LEVEL> or pr_<level>



^ permalink raw reply

* Re: [PATCH 1/4] mm: Move ioremap page table mapping function to mm/
From: Nicholas Piggin @ 2019-06-22  9:42 UTC (permalink / raw)
  To: Christophe Leroy, linux-mm; +Cc: linuxppc-dev, linux-arm-kernel
In-Reply-To: <7218a243-0d9c-ad90-d409-87663893799e@c-s.fr>

Christophe Leroy's on June 19, 2019 11:18 pm:
> 
> 
> Le 19/06/2019 à 05:43, Nicholas Piggin a écrit :
>> Christophe Leroy's on June 11, 2019 3:24 pm:
>>>
>>>
>>> Le 10/06/2019 à 06:38, Nicholas Piggin a écrit :
> 
> [snip]
> 
>>>> diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h
>>>> index 51e131245379..812bea5866d6 100644
>>>> --- a/include/linux/vmalloc.h
>>>> +++ b/include/linux/vmalloc.h
>>>> @@ -147,6 +147,9 @@ extern struct vm_struct *find_vm_area(const void *addr);
>>>>    extern int map_vm_area(struct vm_struct *area, pgprot_t prot,
>>>>    			struct page **pages);
>>>>    #ifdef CONFIG_MMU
>>>> +extern int vmap_range(unsigned long addr,
>>>> +		       unsigned long end, phys_addr_t phys_addr, pgprot_t prot,
>>>> +		       unsigned int max_page_shift);
>>>
>>> Drop extern keyword here.
>> 
>> I don't know if I was going crazy but at one point I was getting
>> duplicate symbol errors that were fixed by adding extern somewhere.
> 
> probably not on a function name ...

I know it sounds crazy :P

>>> As checkpatch tells you, 'CHECK:AVOID_EXTERNS: extern prototypes should
>>> be avoided in .h files'
>> 
>> I prefer to follow existing style in surrounding code at the expense
>> of some checkpatch warnings. If somebody later wants to "fix" it
>> that's fine.
> 
> I don't think that's fine to 'fix' later things that could be done right 
> from the begining. 'Cosmetic only' fixes never happen because they are a 
> nightmare for backports, and a shame for 'git blame'.
> 
> In some patches, you add cleanups to make the code look nicer, and here 
> you have the opportunity to make the code nice from the begining and you 
> prefer repeating the errors done in the past ? You're surprising me.

Well I never claimed to be consistent. I actually don't mind the
extern keyword so it's probably just my personal preference that
makes me notice something nearby. I have dropped those "cleanup"
changes though, so there.

Thanks,
Nick

^ permalink raw reply

* [GIT PULL] Please pull powerpc/linux.git powerpc-5.2-5 tag
From: Michael Ellerman @ 2019-06-22 11:52 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: mikey, linux-kernel, sjitindarsingh, linuxppc-dev, hch,
	Larry.Finger

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Linus,

Please pull some more powerpc fixes for 5.2.

This is a frustratingly large batch at rc5. Some of these were sent earlier but
were missed by me due to being distracted by other things, and some took a while
to track down due to needing manual bisection on old hardware. But still we
clearly need to improve our testing of KVM, and of 32-bit, so that we catch
these earlier.

cheers


The following changes since commit c21f5a9ed85ca3e914ca11f421677ae9ae0d04b0:

  powerpc/32s: fix booting with CONFIG_PPC_EARLY_DEBUG_BOOTX (2019-06-07 19:00:14 +1000)

are available in the git repository at:

  https://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux.git tags/powerpc-5.2-5

for you to fetch changes up to 50087112592016a3fc10b394a55f1f1a1bde6908:

  KVM: PPC: Book3S HV: Invalidate ERAT when flushing guest TLB entries (2019-06-20 22:11:25 +1000)

- ------------------------------------------------------------------
powerpc fixes for 5.2 #5

Seven fixes, all for bugs introduced this cycle.

The commit to add KASAN support broke booting on 32-bit SMP machines, due to a
refactoring that moved some setup out of the secondary CPU path.

A fix for another 32-bit SMP bug introduced by the fast syscall entry
implementation for 32-bit BOOKE. And a build fix for the same commit.

Our change to allow the DAWR to be force enabled on Power9 introduced a bug in
KVM, where we clobber r3 leading to a host crash.

The same commit also exposed a previously unreachable bug in the nested KVM
handling of DAWR, which could lead to an oops in a nested host.

One of the DMA reworks broke the b43legacy WiFi driver on some people's
powermacs, fix it by enabling a 30-bit ZONE_DMA on 32-bit.

A fix for TLB flushing in KVM introduced a new bug, as it neglected to also
flush the ERAT, this could lead to memory corruption in the guest.

Thanks to:
  Aaro Koskinen, Christoph Hellwig, Christophe Leroy, Larry Finger, Michael
  Neuling, Suraj Jitindar Singh.

- ------------------------------------------------------------------
Christoph Hellwig (1):
      powerpc: enable a 30-bit ZONE_DMA for 32-bit pmac

Christophe Leroy (3):
      powerpc/32s: fix initial setup of segment registers on secondary CPU
      powerpc/booke: fix fast syscall entry on SMP
      powerpc/32: fix build failure on book3e with KVM

Michael Neuling (1):
      KVM: PPC: Book3S HV: Fix r3 corruption in h_set_dabr()

Suraj Jitindar Singh (2):
      KVM: PPC: Book3S HV: Only write DAWR[X] when handling h_set_dawr in real mode
      KVM: PPC: Book3S HV: Invalidate ERAT when flushing guest TLB entries


 arch/powerpc/include/asm/page.h         |  7 +++++++
 arch/powerpc/kernel/head_32.S           |  1 +
 arch/powerpc/kernel/head_booke.h        | 10 +++++-----
 arch/powerpc/kernel/head_fsl_booke.S    |  2 +-
 arch/powerpc/kvm/book3s_hv_builtin.c    |  1 +
 arch/powerpc/kvm/book3s_hv_rmhandlers.S | 15 +++++++++++++--
 arch/powerpc/mm/mem.c                   |  3 ++-
 arch/powerpc/platforms/powermac/Kconfig |  1 +
 8 files changed, 31 insertions(+), 9 deletions(-)
-----BEGIN PGP SIGNATURE-----
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=5RHe
-----END PGP SIGNATURE-----

^ permalink raw reply

* [PATCH v3 00/25] powerpc/64s interrupt handler cleanups, gasification
From: Nicholas Piggin @ 2019-06-22 13:15 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Nicholas Piggin

One more try at this, sorry for the spam. To reduce the chance of more
problems I've gone back the other way and actually pulled out almost
all the patches that altered generated code, leaving just a couple of
simple cases.

Thanks,
Nick

Nicholas Piggin (25):
  powerpc/64s/exception: remove H concatenation for EXC_HV variants
  powerpc/64s/exception: consolidate EXCEPTION_PROLOG_2 with _NORI
    variant
  powerpc/64s/exception: move and tidy EXCEPTION_PROLOG_2 variants
  powerpc/64s/exception: fix sreset KVM test code
  powerpc/64s/exception: remove the "extra" macro parameter
  powerpc/64s/exception: consolidate maskable and non-maskable prologs
  powerpc/64s/exception: merge KVM handler and skip variants
  powerpc/64s/exception: KVM handler can set the HSRR trap bit
  powerpc/64s/exception: Make EXCEPTION_PROLOG_0 a gas macro for
    consistency with others
  powerpc/64s/exception: Move EXCEPTION_COMMON handler and return
    branches into callers
  powerpc/64s/exception: Move EXCEPTION_COMMON additions into callers
  powerpc/64s/exception: unwind exception-64s.h macros
  powerpc/64s/exception: improve 0x500 handler code
  powerpc/64s/exception: move EXCEPTION_PROLOG_2* to a more logical
    place
  powerpc/64s/exception: remove STD_EXCEPTION_COMMON variants
  powerpc/64s/exception: move KVM related code together
  powerpc/64s/exception: move exception-64s.h code to exception-64s.S
    where it is used
  powerpc/64s/exception: move head-64.h code to exception-64s.S where it
    is used
  powerpc/64s/exception: remove __BRANCH_TO_KVM
  powerpc/64s/exception: remove unused BRANCH_TO_COMMON
  powerpc/64s/exception: use a gas macro for system call handler code
  powerpc/64s/exception: fix indenting irregularities
  powerpc/64s/exception: generate regs clear instructions using .rept
  powerpc/64s/exception: remove pointless EXCEPTION_PROLOG macro
    indirection
  powerpc/64s/exception: move paca save area offsets into
    exception-64s.S

 arch/powerpc/include/asm/exception-64s.h |  604 +----------
 arch/powerpc/include/asm/head-64.h       |  202 ----
 arch/powerpc/kernel/exceptions-64s.S     | 1170 +++++++++++++++++-----
 3 files changed, 941 insertions(+), 1035 deletions(-)

-- 
2.20.1


^ permalink raw reply

* [PATCH v3 02/25] powerpc/64s/exception: consolidate EXCEPTION_PROLOG_2 with _NORI variant
From: Nicholas Piggin @ 2019-06-22 13:15 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Nicholas Piggin
In-Reply-To: <20190622131535.20996-1-npiggin@gmail.com>

Switch to a gas macro that conditionally expands the RI clearing
instruction.

No generated code change.

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
 arch/powerpc/include/asm/exception-64s.h | 43 ++++++------------------
 arch/powerpc/kernel/exceptions-64s.S     | 12 +++----
 2 files changed, 17 insertions(+), 38 deletions(-)

diff --git a/arch/powerpc/include/asm/exception-64s.h b/arch/powerpc/include/asm/exception-64s.h
index 1496e4089cee..94c4992188a7 100644
--- a/arch/powerpc/include/asm/exception-64s.h
+++ b/arch/powerpc/include/asm/exception-64s.h
@@ -319,32 +319,11 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 #define EXCEPTION_PROLOG_1(area, extra, vec)				\
 	_EXCEPTION_PROLOG_1(area, extra, vec)
 
-.macro EXCEPTION_PROLOG_2 label, hsrr
-	ld	r10,PACAKMSR(r13)	/* get MSR value for kernel */
-	.if \hsrr
-	mfspr	r11,SPRN_HSRR0		/* save HSRR0 */
-	.else
-	mfspr	r11,SPRN_SRR0		/* save SRR0 */
-	.endif
-	LOAD_HANDLER(r12,\label\())
-	.if \hsrr
-	mtspr	SPRN_HSRR0,r12
-	mfspr	r12,SPRN_HSRR1		/* and HSRR1 */
-	mtspr	SPRN_HSRR1,r10
-	HRFI_TO_KERNEL
-	.else
-	mtspr	SPRN_SRR0,r12
-	mfspr	r12,SPRN_SRR1		/* and SRR1 */
-	mtspr	SPRN_SRR1,r10
-	RFI_TO_KERNEL
-	.endif
-	b	.	/* prevent speculative execution */
-.endm
-
-/* _NORI variant keeps MSR_RI clear */
-.macro EXCEPTION_PROLOG_2_NORI label, hsrr
+.macro EXCEPTION_PROLOG_2 label, hsrr, set_ri
 	ld	r10,PACAKMSR(r13)	/* get MSR value for kernel */
+	.if ! \set_ri
 	xori	r10,r10,MSR_RI		/* Clear MSR_RI */
+	.endif
 	.if \hsrr
 	mfspr	r11,SPRN_HSRR0		/* save HSRR0 */
 	.else
@@ -369,7 +348,7 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 	SET_SCRATCH0(r13);		/* save r13 */			\
 	EXCEPTION_PROLOG_0(area);					\
 	EXCEPTION_PROLOG_1(area, extra, vec);				\
-	EXCEPTION_PROLOG_2 label, h
+	EXCEPTION_PROLOG_2 label, h, 1
 
 #ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 /*
@@ -438,7 +417,7 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 #define EXCEPTION_PROLOG_NORI(area, label, h, extra, vec)		\
 	EXCEPTION_PROLOG_0(area);					\
 	EXCEPTION_PROLOG_1(area, extra, vec);				\
-	EXCEPTION_PROLOG_2_NORI label, h
+	EXCEPTION_PROLOG_2 label, h, 0
 
 #ifdef CONFIG_KVM_BOOK3S_64_HANDLER
 .macro KVMTEST hsrr, n
@@ -595,14 +574,14 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 
 #define STD_EXCEPTION_OOL(vec, label)				\
 	EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_PR, vec);	\
-	EXCEPTION_PROLOG_2 label, EXC_STD
+	EXCEPTION_PROLOG_2 label, EXC_STD, 1
 
 #define STD_EXCEPTION_HV(loc, vec, label)			\
 	EXCEPTION_PROLOG(PACA_EXGEN, label, EXC_HV, KVMTEST_HV, vec)
 
 #define STD_EXCEPTION_HV_OOL(vec, label)			\
 	EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_HV, vec);	\
-	EXCEPTION_PROLOG_2 label, EXC_HV
+	EXCEPTION_PROLOG_2 label, EXC_HV, 1
 
 #define STD_RELON_EXCEPTION(loc, vec, label)		\
 	/* No guest interrupts come through here */	\
@@ -666,21 +645,21 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 	SET_SCRATCH0(r13);    /* save r13 */				\
 	EXCEPTION_PROLOG_0(PACA_EXGEN);					\
 	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, extra, vec, bitmask);	\
-	EXCEPTION_PROLOG_2 label, h
+	EXCEPTION_PROLOG_2 label, h, 1
 
 #define MASKABLE_EXCEPTION(vec, label, bitmask)				\
 	__MASKABLE_EXCEPTION(vec, label, EXC_STD, SOFTEN_TEST_PR, bitmask)
 
 #define MASKABLE_EXCEPTION_OOL(vec, label, bitmask)			\
 	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_TEST_PR, vec, bitmask);\
-	EXCEPTION_PROLOG_2 label, EXC_STD
+	EXCEPTION_PROLOG_2 label, EXC_STD, 1
 
 #define MASKABLE_EXCEPTION_HV(vec, label, bitmask)			\
 	__MASKABLE_EXCEPTION(vec, label, EXC_HV, SOFTEN_TEST_HV, bitmask)
 
 #define MASKABLE_EXCEPTION_HV_OOL(vec, label, bitmask)			\
 	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_TEST_HV, vec, bitmask);\
-	EXCEPTION_PROLOG_2 label, EXC_HV
+	EXCEPTION_PROLOG_2 label, EXC_HV, 1
 
 #define __MASKABLE_RELON_EXCEPTION(vec, label, h, extra, bitmask)	\
 	SET_SCRATCH0(r13);    /* save r13 */				\
@@ -693,7 +672,7 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 
 #define MASKABLE_RELON_EXCEPTION_OOL(vec, label, bitmask)		\
 	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_NOTEST_PR, vec, bitmask);\
-	EXCEPTION_PROLOG_2 label, EXC_STD
+	EXCEPTION_PROLOG_2 label, EXC_STD, 1
 
 #define MASKABLE_RELON_EXCEPTION_HV(vec, label, bitmask)		\
 	__MASKABLE_RELON_EXCEPTION(vec, label, EXC_HV, SOFTEN_TEST_HV, bitmask)
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 8996165fd357..32d1d1b89a2e 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -356,7 +356,7 @@ machine_check_pSeries_0:
 	 * nested machine check corrupts it. machine_check_common enables
 	 * MSR_RI.
 	 */
-	EXCEPTION_PROLOG_2_NORI machine_check_common, EXC_STD
+	EXCEPTION_PROLOG_2 machine_check_common, EXC_STD, 0
 
 TRAMP_KVM_SKIP(PACA_EXMC, 0x200)
 
@@ -598,7 +598,7 @@ EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_PR, 0x300)
 	mfspr	r11,SPRN_DSISR
 	std	r10,PACA_EXGEN+EX_DAR(r13)
 	stw	r11,PACA_EXGEN+EX_DSISR(r13)
-EXCEPTION_PROLOG_2 data_access_common, EXC_STD
+EXCEPTION_PROLOG_2 data_access_common, EXC_STD, 1
 
 EXC_VIRT_BEGIN(data_access, 0x4300, 0x80)
 SET_SCRATCH0(r13)		/* save r13 */
@@ -645,7 +645,7 @@ TRAMP_REAL_BEGIN(tramp_real_data_access_slb)
 EXCEPTION_PROLOG_1(PACA_EXSLB, KVMTEST_PR, 0x380)
 	mfspr	r10,SPRN_DAR
 	std	r10,PACA_EXSLB+EX_DAR(r13)
-EXCEPTION_PROLOG_2 data_access_slb_common, EXC_STD
+EXCEPTION_PROLOG_2 data_access_slb_common, EXC_STD, 1
 
 EXC_VIRT_BEGIN(data_access_slb, 0x4380, 0x80)
 SET_SCRATCH0(r13)		/* save r13 */
@@ -774,7 +774,7 @@ EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_PR, 0x600)
 	mfspr	r11,SPRN_DSISR
 	std	r10,PACA_EXGEN+EX_DAR(r13)
 	stw	r11,PACA_EXGEN+EX_DSISR(r13)
-EXCEPTION_PROLOG_2 alignment_common, EXC_STD
+EXCEPTION_PROLOG_2 alignment_common, EXC_STD, 1
 EXC_REAL_END(alignment, 0x600, 0x100)
 
 EXC_VIRT_BEGIN(alignment, 0x4600, 0x100)
@@ -1320,7 +1320,7 @@ EXC_REAL_BEGIN(denorm_exception_hv, 0x1500, 0x100)
 #endif
 
 	KVMTEST_HV(0x1500)
-	EXCEPTION_PROLOG_2 denorm_common, EXC_HV
+	EXCEPTION_PROLOG_2 denorm_common, EXC_HV, 1
 EXC_REAL_END(denorm_exception_hv, 0x1500, 0x100)
 
 #ifdef CONFIG_PPC_DENORMALISATION
@@ -1442,7 +1442,7 @@ EXC_VIRT_NONE(0x5800, 0x100)
 	std	r12,PACA_EXGEN+EX_R12(r13);		\
 	GET_SCRATCH0(r10);				\
 	std	r10,PACA_EXGEN+EX_R13(r13);		\
-	EXCEPTION_PROLOG_2 soft_nmi_common, _H
+	EXCEPTION_PROLOG_2 soft_nmi_common, _H, 1
 
 /*
  * Branch to soft_nmi_interrupt using the emergency stack. The emergency
-- 
2.20.1


^ permalink raw reply related

* [PATCH v3 01/25] powerpc/64s/exception: remove H concatenation for EXC_HV variants
From: Nicholas Piggin @ 2019-06-22 13:15 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Nicholas Piggin
In-Reply-To: <20190622131535.20996-1-npiggin@gmail.com>

Replace all instances of this with gas macros that test the hsrr
parameter and use the appropriate register names / labels.

No generated code change.

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
 arch/powerpc/include/asm/exception-64s.h | 333 +++++++++++++----------
 arch/powerpc/include/asm/head-64.h       |   8 +-
 arch/powerpc/kernel/exceptions-64s.S     |  97 ++++---
 3 files changed, 253 insertions(+), 185 deletions(-)

diff --git a/arch/powerpc/include/asm/exception-64s.h b/arch/powerpc/include/asm/exception-64s.h
index d3987ce65857..1496e4089cee 100644
--- a/arch/powerpc/include/asm/exception-64s.h
+++ b/arch/powerpc/include/asm/exception-64s.h
@@ -63,6 +63,8 @@
  */
 #define EX_R3		EX_DAR
 
+#ifdef __ASSEMBLY__
+
 #define STF_ENTRY_BARRIER_SLOT						\
 	STF_ENTRY_BARRIER_FIXUP_SECTION;				\
 	nop;								\
@@ -144,38 +146,6 @@
 	hrfid;								\
 	b	hrfi_flush_fallback
 
-#ifdef CONFIG_RELOCATABLE
-#define __EXCEPTION_PROLOG_2_RELON(label, h)				\
-	mfspr	r11,SPRN_##h##SRR0;	/* save SRR0 */			\
-	LOAD_HANDLER(r12,label);					\
-	mtctr	r12;							\
-	mfspr	r12,SPRN_##h##SRR1;	/* and SRR1 */			\
-	li	r10,MSR_RI;						\
-	mtmsrd 	r10,1;			/* Set RI (EE=0) */		\
-	bctr;
-#else
-/* If not relocatable, we can jump directly -- and save messing with LR */
-#define __EXCEPTION_PROLOG_2_RELON(label, h)				\
-	mfspr	r11,SPRN_##h##SRR0;	/* save SRR0 */			\
-	mfspr	r12,SPRN_##h##SRR1;	/* and SRR1 */			\
-	li	r10,MSR_RI;						\
-	mtmsrd 	r10,1;			/* Set RI (EE=0) */		\
-	b	label;
-#endif
-#define EXCEPTION_PROLOG_2_RELON(label, h)				\
-	__EXCEPTION_PROLOG_2_RELON(label, h)
-
-/*
- * As EXCEPTION_PROLOG(), except we've already got relocation on so no need to
- * rfid. Save LR in case we're CONFIG_RELOCATABLE, in which case
- * EXCEPTION_PROLOG_2_RELON will be using LR.
- */
-#define EXCEPTION_RELON_PROLOG(area, label, h, extra, vec)		\
-	SET_SCRATCH0(r13);		/* save r13 */			\
-	EXCEPTION_PROLOG_0(area);					\
-	EXCEPTION_PROLOG_1(area, extra, vec);				\
-	EXCEPTION_PROLOG_2_RELON(label, h)
-
 /*
  * We're short on space and time in the exception prolog, so we can't
  * use the normal LOAD_REG_IMMEDIATE macro to load the address of label.
@@ -200,9 +170,54 @@
 	ori	reg,reg,(ABS_ADDR(label))@l;				\
 	addis	reg,reg,(ABS_ADDR(label))@h
 
+#ifdef CONFIG_RELOCATABLE
+.macro EXCEPTION_PROLOG_2_RELON label, hsrr
+	.if \hsrr
+	mfspr	r11,SPRN_HSRR0	/* save HSRR0 */
+	.else
+	mfspr	r11,SPRN_SRR0	/* save SRR0 */
+	.endif
+	LOAD_HANDLER(r12, \label\())
+	mtctr	r12
+	.if \hsrr
+	mfspr	r12,SPRN_HSRR1	/* and HSRR1 */
+	.else
+	mfspr	r12,SPRN_SRR1	/* and HSRR1 */
+	.endif
+	li	r10,MSR_RI
+	mtmsrd 	r10,1		/* Set RI (EE=0) */
+	bctr
+.endm
+#else
+/* If not relocatable, we can jump directly -- and save messing with LR */
+.macro EXCEPTION_PROLOG_2_RELON label, hsrr
+	.if \hsrr
+	mfspr	r11,SPRN_HSRR0		/* save HSRR0 */
+	mfspr	r12,SPRN_HSRR1		/* and HSRR1 */
+	.else
+	mfspr	r11,SPRN_SRR0		/* save SRR0 */
+	mfspr	r12,SPRN_SRR1		/* and SRR1 */
+	.endif
+	li	r10,MSR_RI
+	mtmsrd 	r10,1			/* Set RI (EE=0) */
+	b	\label
+.endm
+#endif
+
+/*
+ * As EXCEPTION_PROLOG(), except we've already got relocation on so no need to
+ * rfid. Save LR in case we're CONFIG_RELOCATABLE, in which case
+ * EXCEPTION_PROLOG_2_RELON will be using LR.
+ */
+#define EXCEPTION_RELON_PROLOG(area, label, hsrr, extra, vec)		\
+	SET_SCRATCH0(r13);		/* save r13 */			\
+	EXCEPTION_PROLOG_0(area);					\
+	EXCEPTION_PROLOG_1(area, extra, vec);				\
+	EXCEPTION_PROLOG_2_RELON label, hsrr
+
 /* Exception register prefixes */
-#define EXC_HV	H
-#define EXC_STD
+#define EXC_HV		1
+#define EXC_STD		0
 
 #if defined(CONFIG_RELOCATABLE)
 /*
@@ -304,43 +319,57 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 #define EXCEPTION_PROLOG_1(area, extra, vec)				\
 	_EXCEPTION_PROLOG_1(area, extra, vec)
 
-#define __EXCEPTION_PROLOG_2(label, h)					\
-	ld	r10,PACAKMSR(r13);	/* get MSR value for kernel */	\
-	mfspr	r11,SPRN_##h##SRR0;	/* save SRR0 */			\
-	LOAD_HANDLER(r12,label);					\
-	mtspr	SPRN_##h##SRR0,r12;					\
-	mfspr	r12,SPRN_##h##SRR1;	/* and SRR1 */			\
-	mtspr	SPRN_##h##SRR1,r10;					\
-	h##RFI_TO_KERNEL;						\
+.macro EXCEPTION_PROLOG_2 label, hsrr
+	ld	r10,PACAKMSR(r13)	/* get MSR value for kernel */
+	.if \hsrr
+	mfspr	r11,SPRN_HSRR0		/* save HSRR0 */
+	.else
+	mfspr	r11,SPRN_SRR0		/* save SRR0 */
+	.endif
+	LOAD_HANDLER(r12,\label\())
+	.if \hsrr
+	mtspr	SPRN_HSRR0,r12
+	mfspr	r12,SPRN_HSRR1		/* and HSRR1 */
+	mtspr	SPRN_HSRR1,r10
+	HRFI_TO_KERNEL
+	.else
+	mtspr	SPRN_SRR0,r12
+	mfspr	r12,SPRN_SRR1		/* and SRR1 */
+	mtspr	SPRN_SRR1,r10
+	RFI_TO_KERNEL
+	.endif
 	b	.	/* prevent speculative execution */
-#define EXCEPTION_PROLOG_2(label, h)					\
-	__EXCEPTION_PROLOG_2(label, h)
+.endm
 
 /* _NORI variant keeps MSR_RI clear */
-#define __EXCEPTION_PROLOG_2_NORI(label, h)				\
-	ld	r10,PACAKMSR(r13);	/* get MSR value for kernel */	\
-	xori	r10,r10,MSR_RI;		/* Clear MSR_RI */		\
-	mfspr	r11,SPRN_##h##SRR0;	/* save SRR0 */			\
-	LOAD_HANDLER(r12,label);					\
-	mtspr	SPRN_##h##SRR0,r12;					\
-	mfspr	r12,SPRN_##h##SRR1;	/* and SRR1 */			\
-	mtspr	SPRN_##h##SRR1,r10;					\
-	h##RFI_TO_KERNEL;						\
+.macro EXCEPTION_PROLOG_2_NORI label, hsrr
+	ld	r10,PACAKMSR(r13)	/* get MSR value for kernel */
+	xori	r10,r10,MSR_RI		/* Clear MSR_RI */
+	.if \hsrr
+	mfspr	r11,SPRN_HSRR0		/* save HSRR0 */
+	.else
+	mfspr	r11,SPRN_SRR0		/* save SRR0 */
+	.endif
+	LOAD_HANDLER(r12,\label\())
+	.if \hsrr
+	mtspr	SPRN_HSRR0,r12
+	mfspr	r12,SPRN_HSRR1		/* and HSRR1 */
+	mtspr	SPRN_HSRR1,r10
+	HRFI_TO_KERNEL
+	.else
+	mtspr	SPRN_SRR0,r12
+	mfspr	r12,SPRN_SRR1		/* and SRR1 */
+	mtspr	SPRN_SRR1,r10
+	RFI_TO_KERNEL
+	.endif
 	b	.	/* prevent speculative execution */
-
-#define EXCEPTION_PROLOG_2_NORI(label, h)				\
-	__EXCEPTION_PROLOG_2_NORI(label, h)
+.endm
 
 #define EXCEPTION_PROLOG(area, label, h, extra, vec)			\
 	SET_SCRATCH0(r13);		/* save r13 */			\
 	EXCEPTION_PROLOG_0(area);					\
 	EXCEPTION_PROLOG_1(area, extra, vec);				\
-	EXCEPTION_PROLOG_2(label, h)
-
-#define __KVMTEST(h, n)							\
-	lbz	r10,HSTATE_IN_GUEST(r13);				\
-	cmpwi	r10,0;							\
-	bne	do_kvm_##h##n
+	EXCEPTION_PROLOG_2 label, h
 
 #ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 /*
@@ -409,52 +438,66 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 #define EXCEPTION_PROLOG_NORI(area, label, h, extra, vec)		\
 	EXCEPTION_PROLOG_0(area);					\
 	EXCEPTION_PROLOG_1(area, extra, vec);				\
-	EXCEPTION_PROLOG_2_NORI(label, h)
-
-
-#define __KVM_HANDLER(area, h, n)					\
-	BEGIN_FTR_SECTION_NESTED(947)					\
-	ld	r10,area+EX_CFAR(r13);					\
-	std	r10,HSTATE_CFAR(r13);					\
-	END_FTR_SECTION_NESTED(CPU_FTR_CFAR,CPU_FTR_CFAR,947);		\
-	BEGIN_FTR_SECTION_NESTED(948)					\
-	ld	r10,area+EX_PPR(r13);					\
-	std	r10,HSTATE_PPR(r13);					\
-	END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948);	\
-	ld	r10,area+EX_R10(r13);					\
-	std	r12,HSTATE_SCRATCH0(r13);				\
-	sldi	r12,r9,32;						\
-	ori	r12,r12,(n);						\
-	/* This reloads r9 before branching to kvmppc_interrupt */	\
-	__BRANCH_TO_KVM_EXIT(area, kvmppc_interrupt)
-
-#define __KVM_HANDLER_SKIP(area, h, n)					\
-	cmpwi	r10,KVM_GUEST_MODE_SKIP;				\
-	beq	89f;							\
-	BEGIN_FTR_SECTION_NESTED(948)					\
-	ld	r10,area+EX_PPR(r13);					\
-	std	r10,HSTATE_PPR(r13);					\
-	END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948);	\
-	ld	r10,area+EX_R10(r13);					\
-	std	r12,HSTATE_SCRATCH0(r13);				\
-	sldi	r12,r9,32;						\
-	ori	r12,r12,(n);						\
-	/* This reloads r9 before branching to kvmppc_interrupt */	\
-	__BRANCH_TO_KVM_EXIT(area, kvmppc_interrupt);			\
-89:	mtocrf	0x80,r9;						\
-	ld	r9,area+EX_R9(r13);					\
-	ld	r10,area+EX_R10(r13);					\
-	b	kvmppc_skip_##h##interrupt
+	EXCEPTION_PROLOG_2_NORI label, h
 
 #ifdef CONFIG_KVM_BOOK3S_64_HANDLER
-#define KVMTEST(h, n)			__KVMTEST(h, n)
-#define KVM_HANDLER(area, h, n)		__KVM_HANDLER(area, h, n)
-#define KVM_HANDLER_SKIP(area, h, n)	__KVM_HANDLER_SKIP(area, h, n)
+.macro KVMTEST hsrr, n
+	lbz	r10,HSTATE_IN_GUEST(r13)
+	cmpwi	r10,0
+	.if \hsrr
+	bne	do_kvm_H\n
+	.else
+	bne	do_kvm_\n
+	.endif
+.endm
+
+.macro KVM_HANDLER area, hsrr, n
+	BEGIN_FTR_SECTION_NESTED(947)
+	ld	r10,\area+EX_CFAR(r13)
+	std	r10,HSTATE_CFAR(r13)
+	END_FTR_SECTION_NESTED(CPU_FTR_CFAR,CPU_FTR_CFAR,947)
+	BEGIN_FTR_SECTION_NESTED(948)
+	ld	r10,\area+EX_PPR(r13)
+	std	r10,HSTATE_PPR(r13)
+	END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948)
+	ld	r10,\area+EX_R10(r13)
+	std	r12,HSTATE_SCRATCH0(r13)
+	sldi	r12,r9,32
+	ori	r12,r12,(\n)
+	/* This reloads r9 before branching to kvmppc_interrupt */
+	__BRANCH_TO_KVM_EXIT(\area, kvmppc_interrupt)
+.endm
+
+.macro KVM_HANDLER_SKIP area, hsrr, n
+	cmpwi	r10,KVM_GUEST_MODE_SKIP
+	beq	89f
+	BEGIN_FTR_SECTION_NESTED(948)
+	ld	r10,\area+EX_PPR(r13)
+	std	r10,HSTATE_PPR(r13)
+	END_FTR_SECTION_NESTED(CPU_FTR_HAS_PPR,CPU_FTR_HAS_PPR,948)
+	ld	r10,\area+EX_R10(r13)
+	std	r12,HSTATE_SCRATCH0(r13)
+	sldi	r12,r9,32
+	ori	r12,r12,(\n)
+	/* This reloads r9 before branching to kvmppc_interrupt */
+	__BRANCH_TO_KVM_EXIT(\area, kvmppc_interrupt)
+89:	mtocrf	0x80,r9
+	ld	r9,\area+EX_R9(r13)
+	ld	r10,\area+EX_R10(r13)
+	.if \hsrr
+	b	kvmppc_skip_Hinterrupt
+	.else
+	b	kvmppc_skip_interrupt
+	.endif
+.endm
 
 #else
-#define KVMTEST(h, n)
-#define KVM_HANDLER(area, h, n)
-#define KVM_HANDLER_SKIP(area, h, n)
+.macro KVMTEST hsrr, n
+.endm
+.macro KVM_HANDLER area, hsrr, n
+.endm
+.macro KVM_HANDLER_SKIP area, hsrr, n
+.endm
 #endif
 
 #define NOTEST(n)
@@ -552,14 +595,14 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 
 #define STD_EXCEPTION_OOL(vec, label)				\
 	EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_PR, vec);	\
-	EXCEPTION_PROLOG_2(label, EXC_STD)
+	EXCEPTION_PROLOG_2 label, EXC_STD
 
 #define STD_EXCEPTION_HV(loc, vec, label)			\
 	EXCEPTION_PROLOG(PACA_EXGEN, label, EXC_HV, KVMTEST_HV, vec)
 
 #define STD_EXCEPTION_HV_OOL(vec, label)			\
 	EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_HV, vec);	\
-	EXCEPTION_PROLOG_2(label, EXC_HV)
+	EXCEPTION_PROLOG_2 label, EXC_HV
 
 #define STD_RELON_EXCEPTION(loc, vec, label)		\
 	/* No guest interrupts come through here */	\
@@ -567,89 +610,97 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 
 #define STD_RELON_EXCEPTION_OOL(vec, label)			\
 	EXCEPTION_PROLOG_1(PACA_EXGEN, NOTEST, vec);		\
-	EXCEPTION_PROLOG_2_RELON(label, EXC_STD)
+	EXCEPTION_PROLOG_2_RELON label, EXC_STD
 
 #define STD_RELON_EXCEPTION_HV(loc, vec, label)		\
 	EXCEPTION_RELON_PROLOG(PACA_EXGEN, label, EXC_HV, KVMTEST_HV, vec)
 
 #define STD_RELON_EXCEPTION_HV_OOL(vec, label)			\
 	EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_HV, vec);	\
-	EXCEPTION_PROLOG_2_RELON(label, EXC_HV)
-
-/* This associate vector numbers with bits in paca->irq_happened */
-#define SOFTEN_VALUE_0x500	PACA_IRQ_EE
-#define SOFTEN_VALUE_0x900	PACA_IRQ_DEC
-#define SOFTEN_VALUE_0x980	PACA_IRQ_DEC
-#define SOFTEN_VALUE_0xa00	PACA_IRQ_DBELL
-#define SOFTEN_VALUE_0xe80	PACA_IRQ_DBELL
-#define SOFTEN_VALUE_0xe60	PACA_IRQ_HMI
-#define SOFTEN_VALUE_0xea0	PACA_IRQ_EE
-#define SOFTEN_VALUE_0xf00	PACA_IRQ_PMI
-
-#define __SOFTEN_TEST(h, vec, bitmask)					\
-	lbz	r10,PACAIRQSOFTMASK(r13);				\
-	andi.	r10,r10,bitmask;					\
-	li	r10,SOFTEN_VALUE_##vec;					\
-	bne	masked_##h##interrupt
-
-#define _SOFTEN_TEST(h, vec, bitmask)	__SOFTEN_TEST(h, vec, bitmask)
+	EXCEPTION_PROLOG_2_RELON label, EXC_HV
+
+.macro SOFTEN_TEST hsrr, vec, bitmask
+	lbz	r10, PACAIRQSOFTMASK(r13)
+	andi.	r10, r10, \bitmask
+	/* This associates vector numbers with bits in paca->irq_happened */
+	.if \vec == 0x500 || \vec == 0xea0
+	li	r10, PACA_IRQ_EE
+	.elseif \vec == 0x900 || \vec == 0xea0
+	li	r10, PACA_IRQ_DEC
+	.elseif \vec == 0xa00 || \vec == 0xe80
+	li	r10, PACA_IRQ_DBELL
+	.elseif \vec == 0xe60
+	li	r10, PACA_IRQ_HMI
+	.elseif \vec == 0xf00
+	li	r10, PACA_IRQ_PMI
+	.else
+	.abort "Bad maskable vector"
+	.endif
+
+
+	.if \hsrr
+	bne	masked_Hinterrupt
+	.else
+	bne	masked_interrupt
+	.endif
+.endm
 
 #define SOFTEN_TEST_PR(vec, bitmask)					\
-	KVMTEST(EXC_STD, vec);						\
-	_SOFTEN_TEST(EXC_STD, vec, bitmask)
+	KVMTEST EXC_STD, vec ;						\
+	SOFTEN_TEST EXC_STD, vec, bitmask
 
 #define SOFTEN_TEST_HV(vec, bitmask)					\
-	KVMTEST(EXC_HV, vec);						\
-	_SOFTEN_TEST(EXC_HV, vec, bitmask)
+	KVMTEST EXC_HV, vec ;						\
+	SOFTEN_TEST EXC_HV, vec, bitmask
 
 #define KVMTEST_PR(vec)							\
-	KVMTEST(EXC_STD, vec)
+	KVMTEST EXC_STD, vec
 
 #define KVMTEST_HV(vec)							\
-	KVMTEST(EXC_HV, vec)
+	KVMTEST EXC_HV, vec
 
-#define SOFTEN_NOTEST_PR(vec, bitmask)	_SOFTEN_TEST(EXC_STD, vec, bitmask)
-#define SOFTEN_NOTEST_HV(vec, bitmask)	_SOFTEN_TEST(EXC_HV, vec, bitmask)
+#define SOFTEN_NOTEST_PR(vec, bitmask)	SOFTEN_TEST EXC_STD, vec, bitmask
+#define SOFTEN_NOTEST_HV(vec, bitmask)	SOFTEN_TEST EXC_HV, vec, bitmask
 
 #define __MASKABLE_EXCEPTION(vec, label, h, extra, bitmask)		\
 	SET_SCRATCH0(r13);    /* save r13 */				\
 	EXCEPTION_PROLOG_0(PACA_EXGEN);					\
 	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, extra, vec, bitmask);	\
-	EXCEPTION_PROLOG_2(label, h)
+	EXCEPTION_PROLOG_2 label, h
 
 #define MASKABLE_EXCEPTION(vec, label, bitmask)				\
 	__MASKABLE_EXCEPTION(vec, label, EXC_STD, SOFTEN_TEST_PR, bitmask)
 
 #define MASKABLE_EXCEPTION_OOL(vec, label, bitmask)			\
 	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_TEST_PR, vec, bitmask);\
-	EXCEPTION_PROLOG_2(label, EXC_STD)
+	EXCEPTION_PROLOG_2 label, EXC_STD
 
 #define MASKABLE_EXCEPTION_HV(vec, label, bitmask)			\
 	__MASKABLE_EXCEPTION(vec, label, EXC_HV, SOFTEN_TEST_HV, bitmask)
 
 #define MASKABLE_EXCEPTION_HV_OOL(vec, label, bitmask)			\
 	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_TEST_HV, vec, bitmask);\
-	EXCEPTION_PROLOG_2(label, EXC_HV)
+	EXCEPTION_PROLOG_2 label, EXC_HV
 
 #define __MASKABLE_RELON_EXCEPTION(vec, label, h, extra, bitmask)	\
 	SET_SCRATCH0(r13);    /* save r13 */				\
 	EXCEPTION_PROLOG_0(PACA_EXGEN);					\
 	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, extra, vec, bitmask);	\
-	EXCEPTION_PROLOG_2_RELON(label, h)
+	EXCEPTION_PROLOG_2_RELON label, h
 
 #define MASKABLE_RELON_EXCEPTION(vec, label, bitmask)			\
 	__MASKABLE_RELON_EXCEPTION(vec, label, EXC_STD, SOFTEN_NOTEST_PR, bitmask)
 
 #define MASKABLE_RELON_EXCEPTION_OOL(vec, label, bitmask)		\
 	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_NOTEST_PR, vec, bitmask);\
-	EXCEPTION_PROLOG_2(label, EXC_STD)
+	EXCEPTION_PROLOG_2 label, EXC_STD
 
 #define MASKABLE_RELON_EXCEPTION_HV(vec, label, bitmask)		\
 	__MASKABLE_RELON_EXCEPTION(vec, label, EXC_HV, SOFTEN_TEST_HV, bitmask)
 
 #define MASKABLE_RELON_EXCEPTION_HV_OOL(vec, label, bitmask)		\
 	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_TEST_HV, vec, bitmask);\
-	EXCEPTION_PROLOG_2_RELON(label, EXC_HV)
+	EXCEPTION_PROLOG_2_RELON label, EXC_HV
 
 /*
  * Our exception common code can be passed various "additions"
@@ -728,4 +779,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_CAN_NAP)
 #define FINISH_NAP
 #endif
 
+#endif /* __ASSEMBLY__ */
+
 #endif	/* _ASM_POWERPC_EXCEPTION_H */
diff --git a/arch/powerpc/include/asm/head-64.h b/arch/powerpc/include/asm/head-64.h
index e34b3d06bf61..4767d6c7b8fa 100644
--- a/arch/powerpc/include/asm/head-64.h
+++ b/arch/powerpc/include/asm/head-64.h
@@ -387,22 +387,22 @@ end_##sname:
 
 #define TRAMP_KVM(area, n)						\
 	TRAMP_KVM_BEGIN(do_kvm_##n);					\
-	KVM_HANDLER(area, EXC_STD, n);					\
+	KVM_HANDLER area, EXC_STD, n
 
 #define TRAMP_KVM_SKIP(area, n)						\
 	TRAMP_KVM_BEGIN(do_kvm_##n);					\
-	KVM_HANDLER_SKIP(area, EXC_STD, n);				\
+	KVM_HANDLER_SKIP area, EXC_STD, n
 
 /*
  * HV variant exceptions get the 0x2 bit added to their trap number.
  */
 #define TRAMP_KVM_HV(area, n)						\
 	TRAMP_KVM_BEGIN(do_kvm_H##n);					\
-	KVM_HANDLER(area, EXC_HV, n + 0x2);				\
+	KVM_HANDLER area, EXC_HV, n + 0x2
 
 #define TRAMP_KVM_HV_SKIP(area, n)					\
 	TRAMP_KVM_BEGIN(do_kvm_H##n);					\
-	KVM_HANDLER_SKIP(area, EXC_HV, n + 0x2);			\
+	KVM_HANDLER_SKIP area, EXC_HV, n + 0x2
 
 #define EXC_COMMON(name, realvec, hdlr)					\
 	EXC_COMMON_BEGIN(name);						\
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 73ba246ca11d..8996165fd357 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -356,7 +356,7 @@ machine_check_pSeries_0:
 	 * nested machine check corrupts it. machine_check_common enables
 	 * MSR_RI.
 	 */
-	EXCEPTION_PROLOG_2_NORI(machine_check_common, EXC_STD)
+	EXCEPTION_PROLOG_2_NORI machine_check_common, EXC_STD
 
 TRAMP_KVM_SKIP(PACA_EXMC, 0x200)
 
@@ -598,7 +598,7 @@ EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_PR, 0x300)
 	mfspr	r11,SPRN_DSISR
 	std	r10,PACA_EXGEN+EX_DAR(r13)
 	stw	r11,PACA_EXGEN+EX_DSISR(r13)
-EXCEPTION_PROLOG_2(data_access_common, EXC_STD)
+EXCEPTION_PROLOG_2 data_access_common, EXC_STD
 
 EXC_VIRT_BEGIN(data_access, 0x4300, 0x80)
 SET_SCRATCH0(r13)		/* save r13 */
@@ -608,7 +608,7 @@ EXCEPTION_PROLOG_1(PACA_EXGEN, NOTEST, 0x300)
 	mfspr	r11,SPRN_DSISR
 	std	r10,PACA_EXGEN+EX_DAR(r13)
 	stw	r11,PACA_EXGEN+EX_DSISR(r13)
-EXCEPTION_PROLOG_2_RELON(data_access_common, EXC_STD)
+EXCEPTION_PROLOG_2_RELON data_access_common, EXC_STD
 EXC_VIRT_END(data_access, 0x4300, 0x80)
 
 TRAMP_KVM_SKIP(PACA_EXGEN, 0x300)
@@ -645,7 +645,7 @@ TRAMP_REAL_BEGIN(tramp_real_data_access_slb)
 EXCEPTION_PROLOG_1(PACA_EXSLB, KVMTEST_PR, 0x380)
 	mfspr	r10,SPRN_DAR
 	std	r10,PACA_EXSLB+EX_DAR(r13)
-EXCEPTION_PROLOG_2(data_access_slb_common, EXC_STD)
+EXCEPTION_PROLOG_2 data_access_slb_common, EXC_STD
 
 EXC_VIRT_BEGIN(data_access_slb, 0x4380, 0x80)
 SET_SCRATCH0(r13)		/* save r13 */
@@ -653,7 +653,7 @@ EXCEPTION_PROLOG_0(PACA_EXSLB)
 EXCEPTION_PROLOG_1(PACA_EXSLB, NOTEST, 0x380)
 	mfspr	r10,SPRN_DAR
 	std	r10,PACA_EXSLB+EX_DAR(r13)
-EXCEPTION_PROLOG_2_RELON(data_access_slb_common, EXC_STD)
+EXCEPTION_PROLOG_2_RELON data_access_slb_common, EXC_STD
 EXC_VIRT_END(data_access_slb, 0x4380, 0x80)
 
 TRAMP_KVM_SKIP(PACA_EXSLB, 0x380)
@@ -774,7 +774,7 @@ EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_PR, 0x600)
 	mfspr	r11,SPRN_DSISR
 	std	r10,PACA_EXGEN+EX_DAR(r13)
 	stw	r11,PACA_EXGEN+EX_DSISR(r13)
-EXCEPTION_PROLOG_2(alignment_common, EXC_STD)
+EXCEPTION_PROLOG_2 alignment_common, EXC_STD
 EXC_REAL_END(alignment, 0x600, 0x100)
 
 EXC_VIRT_BEGIN(alignment, 0x4600, 0x100)
@@ -785,7 +785,7 @@ EXCEPTION_PROLOG_1(PACA_EXGEN, NOTEST, 0x600)
 	mfspr	r11,SPRN_DSISR
 	std	r10,PACA_EXGEN+EX_DAR(r13)
 	stw	r11,PACA_EXGEN+EX_DSISR(r13)
-EXCEPTION_PROLOG_2_RELON(alignment_common, EXC_STD)
+EXCEPTION_PROLOG_2_RELON alignment_common, EXC_STD
 EXC_VIRT_END(alignment, 0x4600, 0x100)
 
 TRAMP_KVM(PACA_EXGEN, 0x600)
@@ -1053,7 +1053,7 @@ TRAMP_KVM_BEGIN(do_kvm_0xc00)
 	SET_SCRATCH0(r10)
 	std	r9,PACA_EXGEN+EX_R9(r13)
 	mfcr	r9
-	KVM_HANDLER(PACA_EXGEN, EXC_STD, 0xc00)
+	KVM_HANDLER PACA_EXGEN, EXC_STD, 0xc00
 #endif
 
 
@@ -1320,7 +1320,7 @@ EXC_REAL_BEGIN(denorm_exception_hv, 0x1500, 0x100)
 #endif
 
 	KVMTEST_HV(0x1500)
-	EXCEPTION_PROLOG_2(denorm_common, EXC_HV)
+	EXCEPTION_PROLOG_2 denorm_common, EXC_HV
 EXC_REAL_END(denorm_exception_hv, 0x1500, 0x100)
 
 #ifdef CONFIG_PPC_DENORMALISATION
@@ -1442,7 +1442,7 @@ EXC_VIRT_NONE(0x5800, 0x100)
 	std	r12,PACA_EXGEN+EX_R12(r13);		\
 	GET_SCRATCH0(r10);				\
 	std	r10,PACA_EXGEN+EX_R13(r13);		\
-	EXCEPTION_PROLOG_2(soft_nmi_common, _H)
+	EXCEPTION_PROLOG_2 soft_nmi_common, _H
 
 /*
  * Branch to soft_nmi_interrupt using the emergency stack. The emergency
@@ -1477,35 +1477,50 @@ EXC_COMMON_BEGIN(soft_nmi_common)
  * - Else it is one of PACA_IRQ_MUST_HARD_MASK, so hard disable and return.
  * This is called with r10 containing the value to OR to the paca field.
  */
-#define MASKED_INTERRUPT(_H)				\
-masked_##_H##interrupt:					\
-	std	r11,PACA_EXGEN+EX_R11(r13);		\
-	lbz	r11,PACAIRQHAPPENED(r13);		\
-	or	r11,r11,r10;				\
-	stb	r11,PACAIRQHAPPENED(r13);		\
-	cmpwi	r10,PACA_IRQ_DEC;			\
-	bne	1f;					\
-	lis	r10,0x7fff;				\
-	ori	r10,r10,0xffff;				\
-	mtspr	SPRN_DEC,r10;				\
-	b	MASKED_DEC_HANDLER_LABEL;		\
-1:	andi.	r10,r10,PACA_IRQ_MUST_HARD_MASK;	\
-	beq	2f;					\
-	mfspr	r10,SPRN_##_H##SRR1;			\
-	xori	r10,r10,MSR_EE; /* clear MSR_EE */	\
-	mtspr	SPRN_##_H##SRR1,r10;			\
-	ori	r11,r11,PACA_IRQ_HARD_DIS;		\
-	stb	r11,PACAIRQHAPPENED(r13);		\
-2:	/* done */					\
-	mtcrf	0x80,r9;				\
-	std	r1,PACAR1(r13);				\
-	ld	r9,PACA_EXGEN+EX_R9(r13);		\
-	ld	r10,PACA_EXGEN+EX_R10(r13);		\
-	ld	r11,PACA_EXGEN+EX_R11(r13);		\
-	/* returns to kernel where r13 must be set up, so don't restore it */ \
-	##_H##RFI_TO_KERNEL;				\
-	b	.;					\
-	MASKED_DEC_HANDLER(_H)
+.macro MASKED_INTERRUPT hsrr
+	.if \hsrr
+masked_Hinterrupt:
+	.else
+masked_interrupt:
+	.endif
+	std	r11,PACA_EXGEN+EX_R11(r13)
+	lbz	r11,PACAIRQHAPPENED(r13)
+	or	r11,r11,r10
+	stb	r11,PACAIRQHAPPENED(r13)
+	cmpwi	r10,PACA_IRQ_DEC
+	bne	1f
+	lis	r10,0x7fff
+	ori	r10,r10,0xffff
+	mtspr	SPRN_DEC,r10
+	b	MASKED_DEC_HANDLER_LABEL
+1:	andi.	r10,r10,PACA_IRQ_MUST_HARD_MASK
+	beq	2f
+	.if \hsrr
+	mfspr	r10,SPRN_HSRR1
+	xori	r10,r10,MSR_EE	/* clear MSR_EE */
+	mtspr	SPRN_HSRR1,r10
+	.else
+	mfspr	r10,SPRN_SRR1
+	xori	r10,r10,MSR_EE	/* clear MSR_EE */
+	mtspr	SPRN_SRR1,r10
+	.endif
+	ori	r11,r11,PACA_IRQ_HARD_DIS
+	stb	r11,PACAIRQHAPPENED(r13)
+2:	/* done */
+	mtcrf	0x80,r9
+	std	r1,PACAR1(r13)
+	ld	r9,PACA_EXGEN+EX_R9(r13)
+	ld	r10,PACA_EXGEN+EX_R10(r13)
+	ld	r11,PACA_EXGEN+EX_R11(r13)
+	/* returns to kernel where r13 must be set up, so don't restore it */
+	.if \hsrr
+	HRFI_TO_KERNEL
+	.else
+	RFI_TO_KERNEL
+	.endif
+	b	.
+	MASKED_DEC_HANDLER(\hsrr\())
+.endm
 
 TRAMP_REAL_BEGIN(stf_barrier_fallback)
 	std	r9,PACA_EXRFI+EX_R9(r13)
@@ -1612,8 +1627,8 @@ TRAMP_REAL_BEGIN(hrfi_flush_fallback)
  * cannot reach these if they are put there.
  */
 USE_FIXED_SECTION(virt_trampolines)
-	MASKED_INTERRUPT()
-	MASKED_INTERRUPT(H)
+	MASKED_INTERRUPT EXC_STD
+	MASKED_INTERRUPT EXC_HV
 
 #ifdef CONFIG_KVM_BOOK3S_64_HANDLER
 TRAMP_REAL_BEGIN(kvmppc_skip_interrupt)
-- 
2.20.1


^ permalink raw reply related

* [PATCH v3 03/25] powerpc/64s/exception: move and tidy EXCEPTION_PROLOG_2 variants
From: Nicholas Piggin @ 2019-06-22 13:15 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Nicholas Piggin
In-Reply-To: <20190622131535.20996-1-npiggin@gmail.com>

- Re-name the macros to _REAL and _VIRT suffixes rather than no and
  _RELON suffix.

- Move the macro definitions together in the file.

- Move RELOCATABLE ifdef inside the _VIRT macro.

Further consolidation between variants does not buy much here.

No generated code change.

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
 arch/powerpc/include/asm/exception-64s.h | 87 ++++++++++++------------
 arch/powerpc/kernel/exceptions-64s.S     | 18 ++---
 2 files changed, 51 insertions(+), 54 deletions(-)

diff --git a/arch/powerpc/include/asm/exception-64s.h b/arch/powerpc/include/asm/exception-64s.h
index 94c4992188a7..4aef70defcdd 100644
--- a/arch/powerpc/include/asm/exception-64s.h
+++ b/arch/powerpc/include/asm/exception-64s.h
@@ -170,8 +170,33 @@
 	ori	reg,reg,(ABS_ADDR(label))@l;				\
 	addis	reg,reg,(ABS_ADDR(label))@h
 
+.macro EXCEPTION_PROLOG_2_REAL label, hsrr, set_ri
+	ld	r10,PACAKMSR(r13)	/* get MSR value for kernel */
+	.if ! \set_ri
+	xori	r10,r10,MSR_RI		/* Clear MSR_RI */
+	.endif
+	.if \hsrr
+	mfspr	r11,SPRN_HSRR0		/* save HSRR0 */
+	.else
+	mfspr	r11,SPRN_SRR0		/* save SRR0 */
+	.endif
+	LOAD_HANDLER(r12, \label\())
+	.if \hsrr
+	mtspr	SPRN_HSRR0,r12
+	mfspr	r12,SPRN_HSRR1		/* and HSRR1 */
+	mtspr	SPRN_HSRR1,r10
+	HRFI_TO_KERNEL
+	.else
+	mtspr	SPRN_SRR0,r12
+	mfspr	r12,SPRN_SRR1		/* and SRR1 */
+	mtspr	SPRN_SRR1,r10
+	RFI_TO_KERNEL
+	.endif
+	b	.	/* prevent speculative execution */
+.endm
+
+.macro EXCEPTION_PROLOG_2_VIRT label, hsrr
 #ifdef CONFIG_RELOCATABLE
-.macro EXCEPTION_PROLOG_2_RELON label, hsrr
 	.if \hsrr
 	mfspr	r11,SPRN_HSRR0	/* save HSRR0 */
 	.else
@@ -187,10 +212,7 @@
 	li	r10,MSR_RI
 	mtmsrd 	r10,1		/* Set RI (EE=0) */
 	bctr
-.endm
 #else
-/* If not relocatable, we can jump directly -- and save messing with LR */
-.macro EXCEPTION_PROLOG_2_RELON label, hsrr
 	.if \hsrr
 	mfspr	r11,SPRN_HSRR0		/* save HSRR0 */
 	mfspr	r12,SPRN_HSRR1		/* and HSRR1 */
@@ -201,19 +223,19 @@
 	li	r10,MSR_RI
 	mtmsrd 	r10,1			/* Set RI (EE=0) */
 	b	\label
-.endm
 #endif
+.endm
 
 /*
  * As EXCEPTION_PROLOG(), except we've already got relocation on so no need to
- * rfid. Save LR in case we're CONFIG_RELOCATABLE, in which case
- * EXCEPTION_PROLOG_2_RELON will be using LR.
+ * rfid. Save CTR in case we're CONFIG_RELOCATABLE, in which case
+ * EXCEPTION_PROLOG_2_VIRT will be using CTR.
  */
 #define EXCEPTION_RELON_PROLOG(area, label, hsrr, extra, vec)		\
 	SET_SCRATCH0(r13);		/* save r13 */			\
 	EXCEPTION_PROLOG_0(area);					\
 	EXCEPTION_PROLOG_1(area, extra, vec);				\
-	EXCEPTION_PROLOG_2_RELON label, hsrr
+	EXCEPTION_PROLOG_2_VIRT label, hsrr
 
 /* Exception register prefixes */
 #define EXC_HV		1
@@ -319,36 +341,11 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 #define EXCEPTION_PROLOG_1(area, extra, vec)				\
 	_EXCEPTION_PROLOG_1(area, extra, vec)
 
-.macro EXCEPTION_PROLOG_2 label, hsrr, set_ri
-	ld	r10,PACAKMSR(r13)	/* get MSR value for kernel */
-	.if ! \set_ri
-	xori	r10,r10,MSR_RI		/* Clear MSR_RI */
-	.endif
-	.if \hsrr
-	mfspr	r11,SPRN_HSRR0		/* save HSRR0 */
-	.else
-	mfspr	r11,SPRN_SRR0		/* save SRR0 */
-	.endif
-	LOAD_HANDLER(r12,\label\())
-	.if \hsrr
-	mtspr	SPRN_HSRR0,r12
-	mfspr	r12,SPRN_HSRR1		/* and HSRR1 */
-	mtspr	SPRN_HSRR1,r10
-	HRFI_TO_KERNEL
-	.else
-	mtspr	SPRN_SRR0,r12
-	mfspr	r12,SPRN_SRR1		/* and SRR1 */
-	mtspr	SPRN_SRR1,r10
-	RFI_TO_KERNEL
-	.endif
-	b	.	/* prevent speculative execution */
-.endm
-
 #define EXCEPTION_PROLOG(area, label, h, extra, vec)			\
 	SET_SCRATCH0(r13);		/* save r13 */			\
 	EXCEPTION_PROLOG_0(area);					\
 	EXCEPTION_PROLOG_1(area, extra, vec);				\
-	EXCEPTION_PROLOG_2 label, h, 1
+	EXCEPTION_PROLOG_2_REAL label, h, 1
 
 #ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 /*
@@ -417,7 +414,7 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 #define EXCEPTION_PROLOG_NORI(area, label, h, extra, vec)		\
 	EXCEPTION_PROLOG_0(area);					\
 	EXCEPTION_PROLOG_1(area, extra, vec);				\
-	EXCEPTION_PROLOG_2 label, h, 0
+	EXCEPTION_PROLOG_2_REAL label, h, 0
 
 #ifdef CONFIG_KVM_BOOK3S_64_HANDLER
 .macro KVMTEST hsrr, n
@@ -574,14 +571,14 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 
 #define STD_EXCEPTION_OOL(vec, label)				\
 	EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_PR, vec);	\
-	EXCEPTION_PROLOG_2 label, EXC_STD, 1
+	EXCEPTION_PROLOG_2_REAL label, EXC_STD, 1
 
 #define STD_EXCEPTION_HV(loc, vec, label)			\
 	EXCEPTION_PROLOG(PACA_EXGEN, label, EXC_HV, KVMTEST_HV, vec)
 
 #define STD_EXCEPTION_HV_OOL(vec, label)			\
 	EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_HV, vec);	\
-	EXCEPTION_PROLOG_2 label, EXC_HV, 1
+	EXCEPTION_PROLOG_2_REAL label, EXC_HV, 1
 
 #define STD_RELON_EXCEPTION(loc, vec, label)		\
 	/* No guest interrupts come through here */	\
@@ -589,14 +586,14 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 
 #define STD_RELON_EXCEPTION_OOL(vec, label)			\
 	EXCEPTION_PROLOG_1(PACA_EXGEN, NOTEST, vec);		\
-	EXCEPTION_PROLOG_2_RELON label, EXC_STD
+	EXCEPTION_PROLOG_2_VIRT label, EXC_STD
 
 #define STD_RELON_EXCEPTION_HV(loc, vec, label)		\
 	EXCEPTION_RELON_PROLOG(PACA_EXGEN, label, EXC_HV, KVMTEST_HV, vec)
 
 #define STD_RELON_EXCEPTION_HV_OOL(vec, label)			\
 	EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_HV, vec);	\
-	EXCEPTION_PROLOG_2_RELON label, EXC_HV
+	EXCEPTION_PROLOG_2_VIRT label, EXC_HV
 
 .macro SOFTEN_TEST hsrr, vec, bitmask
 	lbz	r10, PACAIRQSOFTMASK(r13)
@@ -645,41 +642,41 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 	SET_SCRATCH0(r13);    /* save r13 */				\
 	EXCEPTION_PROLOG_0(PACA_EXGEN);					\
 	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, extra, vec, bitmask);	\
-	EXCEPTION_PROLOG_2 label, h, 1
+	EXCEPTION_PROLOG_2_REAL label, h, 1
 
 #define MASKABLE_EXCEPTION(vec, label, bitmask)				\
 	__MASKABLE_EXCEPTION(vec, label, EXC_STD, SOFTEN_TEST_PR, bitmask)
 
 #define MASKABLE_EXCEPTION_OOL(vec, label, bitmask)			\
 	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_TEST_PR, vec, bitmask);\
-	EXCEPTION_PROLOG_2 label, EXC_STD, 1
+	EXCEPTION_PROLOG_2_REAL label, EXC_STD, 1
 
 #define MASKABLE_EXCEPTION_HV(vec, label, bitmask)			\
 	__MASKABLE_EXCEPTION(vec, label, EXC_HV, SOFTEN_TEST_HV, bitmask)
 
 #define MASKABLE_EXCEPTION_HV_OOL(vec, label, bitmask)			\
 	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_TEST_HV, vec, bitmask);\
-	EXCEPTION_PROLOG_2 label, EXC_HV, 1
+	EXCEPTION_PROLOG_2_REAL label, EXC_HV, 1
 
 #define __MASKABLE_RELON_EXCEPTION(vec, label, h, extra, bitmask)	\
 	SET_SCRATCH0(r13);    /* save r13 */				\
 	EXCEPTION_PROLOG_0(PACA_EXGEN);					\
 	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, extra, vec, bitmask);	\
-	EXCEPTION_PROLOG_2_RELON label, h
+	EXCEPTION_PROLOG_2_VIRT label, h
 
 #define MASKABLE_RELON_EXCEPTION(vec, label, bitmask)			\
 	__MASKABLE_RELON_EXCEPTION(vec, label, EXC_STD, SOFTEN_NOTEST_PR, bitmask)
 
 #define MASKABLE_RELON_EXCEPTION_OOL(vec, label, bitmask)		\
 	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_NOTEST_PR, vec, bitmask);\
-	EXCEPTION_PROLOG_2 label, EXC_STD, 1
+	EXCEPTION_PROLOG_2_REAL label, EXC_STD, 1
 
 #define MASKABLE_RELON_EXCEPTION_HV(vec, label, bitmask)		\
 	__MASKABLE_RELON_EXCEPTION(vec, label, EXC_HV, SOFTEN_TEST_HV, bitmask)
 
 #define MASKABLE_RELON_EXCEPTION_HV_OOL(vec, label, bitmask)		\
 	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_TEST_HV, vec, bitmask);\
-	EXCEPTION_PROLOG_2_RELON label, EXC_HV
+	EXCEPTION_PROLOG_2_VIRT label, EXC_HV
 
 /*
  * Our exception common code can be passed various "additions"
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 32d1d1b89a2e..4ebe39c74aa0 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -356,7 +356,7 @@ machine_check_pSeries_0:
 	 * nested machine check corrupts it. machine_check_common enables
 	 * MSR_RI.
 	 */
-	EXCEPTION_PROLOG_2 machine_check_common, EXC_STD, 0
+	EXCEPTION_PROLOG_2_REAL machine_check_common, EXC_STD, 0
 
 TRAMP_KVM_SKIP(PACA_EXMC, 0x200)
 
@@ -598,7 +598,7 @@ EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_PR, 0x300)
 	mfspr	r11,SPRN_DSISR
 	std	r10,PACA_EXGEN+EX_DAR(r13)
 	stw	r11,PACA_EXGEN+EX_DSISR(r13)
-EXCEPTION_PROLOG_2 data_access_common, EXC_STD, 1
+EXCEPTION_PROLOG_2_REAL data_access_common, EXC_STD, 1
 
 EXC_VIRT_BEGIN(data_access, 0x4300, 0x80)
 SET_SCRATCH0(r13)		/* save r13 */
@@ -608,7 +608,7 @@ EXCEPTION_PROLOG_1(PACA_EXGEN, NOTEST, 0x300)
 	mfspr	r11,SPRN_DSISR
 	std	r10,PACA_EXGEN+EX_DAR(r13)
 	stw	r11,PACA_EXGEN+EX_DSISR(r13)
-EXCEPTION_PROLOG_2_RELON data_access_common, EXC_STD
+EXCEPTION_PROLOG_2_VIRT data_access_common, EXC_STD
 EXC_VIRT_END(data_access, 0x4300, 0x80)
 
 TRAMP_KVM_SKIP(PACA_EXGEN, 0x300)
@@ -645,7 +645,7 @@ TRAMP_REAL_BEGIN(tramp_real_data_access_slb)
 EXCEPTION_PROLOG_1(PACA_EXSLB, KVMTEST_PR, 0x380)
 	mfspr	r10,SPRN_DAR
 	std	r10,PACA_EXSLB+EX_DAR(r13)
-EXCEPTION_PROLOG_2 data_access_slb_common, EXC_STD, 1
+EXCEPTION_PROLOG_2_REAL data_access_slb_common, EXC_STD, 1
 
 EXC_VIRT_BEGIN(data_access_slb, 0x4380, 0x80)
 SET_SCRATCH0(r13)		/* save r13 */
@@ -653,7 +653,7 @@ EXCEPTION_PROLOG_0(PACA_EXSLB)
 EXCEPTION_PROLOG_1(PACA_EXSLB, NOTEST, 0x380)
 	mfspr	r10,SPRN_DAR
 	std	r10,PACA_EXSLB+EX_DAR(r13)
-EXCEPTION_PROLOG_2_RELON data_access_slb_common, EXC_STD
+EXCEPTION_PROLOG_2_VIRT data_access_slb_common, EXC_STD
 EXC_VIRT_END(data_access_slb, 0x4380, 0x80)
 
 TRAMP_KVM_SKIP(PACA_EXSLB, 0x380)
@@ -774,7 +774,7 @@ EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_PR, 0x600)
 	mfspr	r11,SPRN_DSISR
 	std	r10,PACA_EXGEN+EX_DAR(r13)
 	stw	r11,PACA_EXGEN+EX_DSISR(r13)
-EXCEPTION_PROLOG_2 alignment_common, EXC_STD, 1
+EXCEPTION_PROLOG_2_REAL alignment_common, EXC_STD, 1
 EXC_REAL_END(alignment, 0x600, 0x100)
 
 EXC_VIRT_BEGIN(alignment, 0x4600, 0x100)
@@ -785,7 +785,7 @@ EXCEPTION_PROLOG_1(PACA_EXGEN, NOTEST, 0x600)
 	mfspr	r11,SPRN_DSISR
 	std	r10,PACA_EXGEN+EX_DAR(r13)
 	stw	r11,PACA_EXGEN+EX_DSISR(r13)
-EXCEPTION_PROLOG_2_RELON alignment_common, EXC_STD
+EXCEPTION_PROLOG_2_VIRT alignment_common, EXC_STD
 EXC_VIRT_END(alignment, 0x4600, 0x100)
 
 TRAMP_KVM(PACA_EXGEN, 0x600)
@@ -1320,7 +1320,7 @@ EXC_REAL_BEGIN(denorm_exception_hv, 0x1500, 0x100)
 #endif
 
 	KVMTEST_HV(0x1500)
-	EXCEPTION_PROLOG_2 denorm_common, EXC_HV, 1
+	EXCEPTION_PROLOG_2_REAL denorm_common, EXC_HV, 1
 EXC_REAL_END(denorm_exception_hv, 0x1500, 0x100)
 
 #ifdef CONFIG_PPC_DENORMALISATION
@@ -1442,7 +1442,7 @@ EXC_VIRT_NONE(0x5800, 0x100)
 	std	r12,PACA_EXGEN+EX_R12(r13);		\
 	GET_SCRATCH0(r10);				\
 	std	r10,PACA_EXGEN+EX_R13(r13);		\
-	EXCEPTION_PROLOG_2 soft_nmi_common, _H, 1
+	EXCEPTION_PROLOG_2_REAL soft_nmi_common, _H, 1
 
 /*
  * Branch to soft_nmi_interrupt using the emergency stack. The emergency
-- 
2.20.1


^ permalink raw reply related

* [PATCH v3 04/25] powerpc/64s/exception: fix sreset KVM test code
From: Nicholas Piggin @ 2019-06-22 13:15 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Nicholas Piggin
In-Reply-To: <20190622131535.20996-1-npiggin@gmail.com>

The sreset handler KVM test theoretically should not depend on P7.
In practice KVM now only supports P7 and up so no real bug fix, but
this change is made now so the quirk is not propagated through
cleanup patches.

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
 arch/powerpc/kernel/exceptions-64s.S | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 4ebe39c74aa0..1a2ad5a022e8 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -126,10 +126,10 @@ EXC_VIRT_NONE(0x4000, 0x100)
 	bltlr	cr1 ;	/* no state loss, return to idle caller */	\
 	BRANCH_TO_C000(r10, system_reset_idle_common) ;			\
 1:									\
-	KVMTEST_PR(n) ;							\
-	END_FTR_SECTION_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206)
+	END_FTR_SECTION_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206) ;	\
+	KVMTEST_PR(n)
 #else
-#define IDLETEST NOTEST
+#define IDLETEST KVMTEST_PR
 #endif
 
 EXC_REAL_BEGIN(system_reset, 0x100, 0x100)
-- 
2.20.1


^ permalink raw reply related

* [PATCH v3 05/25] powerpc/64s/exception: remove the "extra" macro parameter
From: Nicholas Piggin @ 2019-06-22 13:15 UTC (permalink / raw)
  To: linuxppc-dev; +Cc: Nicholas Piggin
In-Reply-To: <20190622131535.20996-1-npiggin@gmail.com>

Rather than pass in the soft-masking and KVM tests via macro that is
passed to another macro to expand it, switch to usig gas macros and
conditionally expand the soft-masking and KVM tests.

The system reset with its idle test is open coded as it is a one-off.

No generated code change.

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
 arch/powerpc/include/asm/exception-64s.h | 158 ++++++++++-------------
 arch/powerpc/kernel/exceptions-64s.S     |  78 ++++++-----
 2 files changed, 114 insertions(+), 122 deletions(-)

diff --git a/arch/powerpc/include/asm/exception-64s.h b/arch/powerpc/include/asm/exception-64s.h
index 4aef70defcdd..e1b449e2c9ea 100644
--- a/arch/powerpc/include/asm/exception-64s.h
+++ b/arch/powerpc/include/asm/exception-64s.h
@@ -231,10 +231,10 @@
  * rfid. Save CTR in case we're CONFIG_RELOCATABLE, in which case
  * EXCEPTION_PROLOG_2_VIRT will be using CTR.
  */
-#define EXCEPTION_RELON_PROLOG(area, label, hsrr, extra, vec)		\
+#define EXCEPTION_RELON_PROLOG(area, label, hsrr, kvm, vec)		\
 	SET_SCRATCH0(r13);		/* save r13 */			\
 	EXCEPTION_PROLOG_0(area);					\
-	EXCEPTION_PROLOG_1(area, extra, vec);				\
+	EXCEPTION_PROLOG_1 hsrr, area, kvm, vec ;			\
 	EXCEPTION_PROLOG_2_VIRT label, hsrr
 
 /* Exception register prefixes */
@@ -321,31 +321,58 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 /*
  * This version of the EXCEPTION_PROLOG_1 will carry
  * addition parameter called "bitmask" to support
- * checking of the interrupt maskable level in the SOFTEN_TEST.
+ * checking of the interrupt maskable level.
  * Intended to be used in MASKABLE_EXCPETION_* macros.
  */
-#define MASKABLE_EXCEPTION_PROLOG_1(area, extra, vec, bitmask)			\
-	__EXCEPTION_PROLOG_1_PRE(area);					\
-	extra(vec, bitmask);						\
-	__EXCEPTION_PROLOG_1_POST(area)
+.macro MASKABLE_EXCEPTION_PROLOG_1 hsrr, area, kvm, vec, bitmask
+	__EXCEPTION_PROLOG_1_PRE(\area\())
+	.if \kvm
+		KVMTEST \hsrr \vec
+	.endif
+
+	lbz	r10,PACAIRQSOFTMASK(r13)
+	andi.	r10,r10,\bitmask
+	/* This associates vector numbers with bits in paca->irq_happened */
+	.if \vec == 0x500 || \vec == 0xea0
+	li	r10,PACA_IRQ_EE
+	.elseif \vec == 0x900 || \vec == 0xea0
+	li	r10,PACA_IRQ_DEC
+	.elseif \vec == 0xa00 || \vec == 0xe80
+	li	r10,PACA_IRQ_DBELL
+	.elseif \vec == 0xe60
+	li	r10,PACA_IRQ_HMI
+	.elseif \vec == 0xf00
+	li	r10,PACA_IRQ_PMI
+	.else
+	.abort "Bad maskable vector"
+	.endif
+
+	.if \hsrr
+	bne	masked_Hinterrupt
+	.else
+	bne	masked_interrupt
+	.endif
+
+	__EXCEPTION_PROLOG_1_POST(\area\())
+.endm
 
 /*
  * This version of the EXCEPTION_PROLOG_1 is intended
  * to be used in STD_EXCEPTION* macros
  */
-#define _EXCEPTION_PROLOG_1(area, extra, vec)				\
-	__EXCEPTION_PROLOG_1_PRE(area);					\
-	extra(vec);							\
-	__EXCEPTION_PROLOG_1_POST(area)
-
-#define EXCEPTION_PROLOG_1(area, extra, vec)				\
-	_EXCEPTION_PROLOG_1(area, extra, vec)
+.macro EXCEPTION_PROLOG_1 hsrr, area, kvm, vec
+	__EXCEPTION_PROLOG_1_PRE(\area\())
+	.if \kvm
+		KVMTEST \hsrr \vec
+	.endif
+	__EXCEPTION_PROLOG_1_POST(\area\())
+.endm
 
-#define EXCEPTION_PROLOG(area, label, h, extra, vec)			\
+#define EXCEPTION_PROLOG(area, label, hsrr, kvm, vec)			\
 	SET_SCRATCH0(r13);		/* save r13 */			\
 	EXCEPTION_PROLOG_0(area);					\
-	EXCEPTION_PROLOG_1(area, extra, vec);				\
-	EXCEPTION_PROLOG_2_REAL label, h, 1
+	EXCEPTION_PROLOG_1 hsrr, area, kvm, vec ;			\
+	EXCEPTION_PROLOG_2_REAL label, hsrr, 1
 
 #ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 /*
@@ -411,10 +438,10 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 #endif
 
 /* Do not enable RI */
-#define EXCEPTION_PROLOG_NORI(area, label, h, extra, vec)		\
+#define EXCEPTION_PROLOG_NORI(area, label, hsrr, kvm, vec)		\
 	EXCEPTION_PROLOG_0(area);					\
-	EXCEPTION_PROLOG_1(area, extra, vec);				\
-	EXCEPTION_PROLOG_2_REAL label, h, 0
+	EXCEPTION_PROLOG_1 hsrr, area, kvm, vec ;			\
+	EXCEPTION_PROLOG_2_REAL label, hsrr, 0
 
 #ifdef CONFIG_KVM_BOOK3S_64_HANDLER
 .macro KVMTEST hsrr, n
@@ -476,8 +503,6 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 .endm
 #endif
 
-#define NOTEST(n)
-
 #define EXCEPTION_PROLOG_COMMON_1()					   \
 	std	r9,_CCR(r1);		/* save CR in stackframe	*/ \
 	std	r11,_NIP(r1);		/* save SRR0 in stackframe	*/ \
@@ -561,7 +586,7 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
  * Exception vectors.
  */
 #define STD_EXCEPTION(vec, label)				\
-	EXCEPTION_PROLOG(PACA_EXGEN, label, EXC_STD, KVMTEST_PR, vec);
+	EXCEPTION_PROLOG(PACA_EXGEN, label, EXC_STD, 1, vec);
 
 /* Version of above for when we have to branch out-of-line */
 #define __OOL_EXCEPTION(vec, label, hdlr)			\
@@ -570,112 +595,69 @@ END_FTR_SECTION_NESTED(ftr,ftr,943)
 	b hdlr
 
 #define STD_EXCEPTION_OOL(vec, label)				\
-	EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_PR, vec);	\
+	EXCEPTION_PROLOG_1 EXC_STD, PACA_EXGEN, 1, vec ;	\
 	EXCEPTION_PROLOG_2_REAL label, EXC_STD, 1
 
 #define STD_EXCEPTION_HV(loc, vec, label)			\
-	EXCEPTION_PROLOG(PACA_EXGEN, label, EXC_HV, KVMTEST_HV, vec)
+	EXCEPTION_PROLOG(PACA_EXGEN, label, EXC_HV, 1, vec)
 
 #define STD_EXCEPTION_HV_OOL(vec, label)			\
-	EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_HV, vec);	\
+	EXCEPTION_PROLOG_1 EXC_HV, PACA_EXGEN, 1, vec ;		\
 	EXCEPTION_PROLOG_2_REAL label, EXC_HV, 1
 
 #define STD_RELON_EXCEPTION(loc, vec, label)		\
 	/* No guest interrupts come through here */	\
-	EXCEPTION_RELON_PROLOG(PACA_EXGEN, label, EXC_STD, NOTEST, vec)
+	EXCEPTION_RELON_PROLOG(PACA_EXGEN, label, EXC_STD, 0, vec)
 
 #define STD_RELON_EXCEPTION_OOL(vec, label)			\
-	EXCEPTION_PROLOG_1(PACA_EXGEN, NOTEST, vec);		\
+	EXCEPTION_PROLOG_1 EXC_STD, PACA_EXGEN, 0, vec ;	\
 	EXCEPTION_PROLOG_2_VIRT label, EXC_STD
 
-#define STD_RELON_EXCEPTION_HV(loc, vec, label)		\
-	EXCEPTION_RELON_PROLOG(PACA_EXGEN, label, EXC_HV, KVMTEST_HV, vec)
+#define STD_RELON_EXCEPTION_HV(loc, vec, label)			\
+	EXCEPTION_RELON_PROLOG(PACA_EXGEN, label, EXC_HV, 1, vec)
 
 #define STD_RELON_EXCEPTION_HV_OOL(vec, label)			\
-	EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_HV, vec);	\
+	EXCEPTION_PROLOG_1 EXC_HV, PACA_EXGEN, 1, vec;		\
 	EXCEPTION_PROLOG_2_VIRT label, EXC_HV
 
-.macro SOFTEN_TEST hsrr, vec, bitmask
-	lbz	r10, PACAIRQSOFTMASK(r13)
-	andi.	r10, r10, \bitmask
-	/* This associates vector numbers with bits in paca->irq_happened */
-	.if \vec == 0x500 || \vec == 0xea0
-	li	r10, PACA_IRQ_EE
-	.elseif \vec == 0x900 || \vec == 0xea0
-	li	r10, PACA_IRQ_DEC
-	.elseif \vec == 0xa00 || \vec == 0xe80
-	li	r10, PACA_IRQ_DBELL
-	.elseif \vec == 0xe60
-	li	r10, PACA_IRQ_HMI
-	.elseif \vec == 0xf00
-	li	r10, PACA_IRQ_PMI
-	.else
-	.abort "Bad maskable vector"
-	.endif
-
-
-	.if \hsrr
-	bne	masked_Hinterrupt
-	.else
-	bne	masked_interrupt
-	.endif
-.endm
-
-#define SOFTEN_TEST_PR(vec, bitmask)					\
-	KVMTEST EXC_STD, vec ;						\
-	SOFTEN_TEST EXC_STD, vec, bitmask
-
-#define SOFTEN_TEST_HV(vec, bitmask)					\
-	KVMTEST EXC_HV, vec ;						\
-	SOFTEN_TEST EXC_HV, vec, bitmask
-
-#define KVMTEST_PR(vec)							\
-	KVMTEST EXC_STD, vec
-
-#define KVMTEST_HV(vec)							\
-	KVMTEST EXC_HV, vec
-
-#define SOFTEN_NOTEST_PR(vec, bitmask)	SOFTEN_TEST EXC_STD, vec, bitmask
-#define SOFTEN_NOTEST_HV(vec, bitmask)	SOFTEN_TEST EXC_HV, vec, bitmask
-
-#define __MASKABLE_EXCEPTION(vec, label, h, extra, bitmask)		\
+#define __MASKABLE_EXCEPTION(vec, label, hsrr, kvm, bitmask)		\
 	SET_SCRATCH0(r13);    /* save r13 */				\
 	EXCEPTION_PROLOG_0(PACA_EXGEN);					\
-	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, extra, vec, bitmask);	\
-	EXCEPTION_PROLOG_2_REAL label, h, 1
+	MASKABLE_EXCEPTION_PROLOG_1 hsrr, PACA_EXGEN, kvm, vec, bitmask ; \
+	EXCEPTION_PROLOG_2_REAL label, hsrr, 1
 
 #define MASKABLE_EXCEPTION(vec, label, bitmask)				\
-	__MASKABLE_EXCEPTION(vec, label, EXC_STD, SOFTEN_TEST_PR, bitmask)
+	__MASKABLE_EXCEPTION(vec, label, EXC_STD, 1, bitmask)
 
 #define MASKABLE_EXCEPTION_OOL(vec, label, bitmask)			\
-	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_TEST_PR, vec, bitmask);\
+	MASKABLE_EXCEPTION_PROLOG_1 EXC_STD, PACA_EXGEN, 1, vec, bitmask ; \
 	EXCEPTION_PROLOG_2_REAL label, EXC_STD, 1
 
 #define MASKABLE_EXCEPTION_HV(vec, label, bitmask)			\
-	__MASKABLE_EXCEPTION(vec, label, EXC_HV, SOFTEN_TEST_HV, bitmask)
+	__MASKABLE_EXCEPTION(vec, label, EXC_HV, 1, bitmask)
 
 #define MASKABLE_EXCEPTION_HV_OOL(vec, label, bitmask)			\
-	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_TEST_HV, vec, bitmask);\
+	MASKABLE_EXCEPTION_PROLOG_1 EXC_HV, PACA_EXGEN, 1, vec, bitmask ; \
 	EXCEPTION_PROLOG_2_REAL label, EXC_HV, 1
 
-#define __MASKABLE_RELON_EXCEPTION(vec, label, h, extra, bitmask)	\
+#define __MASKABLE_RELON_EXCEPTION(vec, label, hsrr, kvm, bitmask)	\
 	SET_SCRATCH0(r13);    /* save r13 */				\
 	EXCEPTION_PROLOG_0(PACA_EXGEN);					\
-	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, extra, vec, bitmask);	\
-	EXCEPTION_PROLOG_2_VIRT label, h
+	MASKABLE_EXCEPTION_PROLOG_1 hsrr, PACA_EXGEN, kvm, vec, bitmask ; \
+	EXCEPTION_PROLOG_2_VIRT label, hsrr
 
 #define MASKABLE_RELON_EXCEPTION(vec, label, bitmask)			\
-	__MASKABLE_RELON_EXCEPTION(vec, label, EXC_STD, SOFTEN_NOTEST_PR, bitmask)
+	__MASKABLE_RELON_EXCEPTION(vec, label, EXC_STD, 0, bitmask)
 
 #define MASKABLE_RELON_EXCEPTION_OOL(vec, label, bitmask)		\
-	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_NOTEST_PR, vec, bitmask);\
+	MASKABLE_EXCEPTION_PROLOG_1 EXC_STD, PACA_EXGEN, 0, vec, bitmask ; \
 	EXCEPTION_PROLOG_2_REAL label, EXC_STD, 1
 
 #define MASKABLE_RELON_EXCEPTION_HV(vec, label, bitmask)		\
-	__MASKABLE_RELON_EXCEPTION(vec, label, EXC_HV, SOFTEN_TEST_HV, bitmask)
+	__MASKABLE_RELON_EXCEPTION(vec, label, EXC_HV, 1, bitmask)
 
 #define MASKABLE_RELON_EXCEPTION_HV_OOL(vec, label, bitmask)		\
-	MASKABLE_EXCEPTION_PROLOG_1(PACA_EXGEN, SOFTEN_TEST_HV, vec, bitmask);\
+	MASKABLE_EXCEPTION_PROLOG_1 EXC_HV, PACA_EXGEN, 1, vec, bitmask ; \
 	EXCEPTION_PROLOG_2_VIRT label, EXC_HV
 
 /*
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 1a2ad5a022e8..f44169239e51 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -107,6 +107,17 @@ __start_interrupts:
 EXC_VIRT_NONE(0x4000, 0x100)
 
 
+EXC_REAL_BEGIN(system_reset, 0x100, 0x100)
+	SET_SCRATCH0(r13)
+	EXCEPTION_PROLOG_0(PACA_EXNMI)
+
+	/* This is EXCEPTION_PROLOG_1 with the idle feature section added */
+	OPT_SAVE_REG_TO_PACA(PACA_EXNMI+EX_PPR, r9, CPU_FTR_HAS_PPR)
+	OPT_SAVE_REG_TO_PACA(PACA_EXNMI+EX_CFAR, r10, CPU_FTR_CFAR)
+	INTERRUPT_TO_KERNEL
+	SAVE_CTR(r10, PACA_EXNMI)
+	mfcr	r9
+
 #ifdef CONFIG_PPC_P7_NAP
 	/*
 	 * If running native on arch 2.06 or later, check if we are waking up
@@ -116,30 +127,29 @@ EXC_VIRT_NONE(0x4000, 0x100)
 	 * but we branch to the 0xc000... address so we can turn on relocation
 	 * with mtmsr.
 	 */
-#define IDLETEST(n)							\
-	BEGIN_FTR_SECTION ;						\
-	mfspr	r10,SPRN_SRR1 ;						\
-	rlwinm.	r10,r10,47-31,30,31 ;					\
-	beq-	1f ;							\
-	cmpwi	cr1,r10,2 ;						\
-	mfspr	r3,SPRN_SRR1 ;						\
-	bltlr	cr1 ;	/* no state loss, return to idle caller */	\
-	BRANCH_TO_C000(r10, system_reset_idle_common) ;			\
-1:									\
-	END_FTR_SECTION_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206) ;	\
-	KVMTEST_PR(n)
-#else
-#define IDLETEST KVMTEST_PR
+	BEGIN_FTR_SECTION
+	mfspr	r10,SPRN_SRR1
+	rlwinm.	r10,r10,47-31,30,31
+	beq-	1f
+	cmpwi	cr1,r10,2
+	mfspr	r3,SPRN_SRR1
+	bltlr	cr1	/* no state loss, return to idle caller */
+	BRANCH_TO_C000(r10, system_reset_idle_common)
+1:
+	END_FTR_SECTION_IFSET(CPU_FTR_HVMODE | CPU_FTR_ARCH_206)
 #endif
 
-EXC_REAL_BEGIN(system_reset, 0x100, 0x100)
-	SET_SCRATCH0(r13)
+	KVMTEST EXC_STD 0x100
+	std	r11,PACA_EXNMI+EX_R11(r13)
+	std	r12,PACA_EXNMI+EX_R12(r13)
+	GET_SCRATCH0(r10)
+	std	r10,PACA_EXNMI+EX_R13(r13)
+
+	EXCEPTION_PROLOG_2_REAL system_reset_common, EXC_STD, 0
 	/*
 	 * MSR_RI is not enabled, because PACA_EXNMI and nmi stack is
 	 * being used, so a nested NMI exception would corrupt it.
 	 */
-	EXCEPTION_PROLOG_NORI(PACA_EXNMI, system_reset_common, EXC_STD,
-			      IDLETEST, 0x100)
 
 EXC_REAL_END(system_reset, 0x100, 0x100)
 EXC_VIRT_NONE(0x4100, 0x100)
@@ -246,7 +256,7 @@ TRAMP_REAL_BEGIN(system_reset_fwnmi)
 	SET_SCRATCH0(r13)		/* save r13 */
 	/* See comment at system_reset exception */
 	EXCEPTION_PROLOG_NORI(PACA_EXNMI, system_reset_common, EXC_STD,
-			      NOTEST, 0x100)
+			      0, 0x100)
 #endif /* CONFIG_PPC_PSERIES */
 
 
@@ -265,7 +275,7 @@ ALT_FTR_SECTION_END_IFSET(CPU_FTR_HVMODE)
 EXC_REAL_END(machine_check, 0x200, 0x100)
 EXC_VIRT_NONE(0x4200, 0x100)
 TRAMP_REAL_BEGIN(machine_check_common_early)
-	EXCEPTION_PROLOG_1(PACA_EXMC, NOTEST, 0x200)
+	EXCEPTION_PROLOG_1 EXC_STD, PACA_EXMC, 0, 0x200
 	/*
 	 * Register contents:
 	 * R13		= PACA
@@ -350,7 +360,7 @@ BEGIN_FTR_SECTION
 	b	machine_check_common_early
 END_FTR_SECTION_IFCLR(CPU_FTR_HVMODE)
 machine_check_pSeries_0:
-	EXCEPTION_PROLOG_1(PACA_EXMC, KVMTEST_PR, 0x200)
+	EXCEPTION_PROLOG_1 EXC_STD, PACA_EXMC, 1, 0x200
 	/*
 	 * MSR_RI is not enabled, because PACA_EXMC is being used, so a
 	 * nested machine check corrupts it. machine_check_common enables
@@ -588,7 +598,7 @@ EXCEPTION_PROLOG_0(PACA_EXGEN)
 EXC_REAL_END(data_access, 0x300, 0x80)
 
 TRAMP_REAL_BEGIN(tramp_real_data_access)
-EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_PR, 0x300)
+EXCEPTION_PROLOG_1 EXC_STD, PACA_EXGEN, 1, 0x300
 	/*
 	 * DAR/DSISR must be read before setting MSR[RI], because
 	 * a d-side MCE will clobber those registers so is not
@@ -603,7 +613,7 @@ EXCEPTION_PROLOG_2_REAL data_access_common, EXC_STD, 1
 EXC_VIRT_BEGIN(data_access, 0x4300, 0x80)
 SET_SCRATCH0(r13)		/* save r13 */
 EXCEPTION_PROLOG_0(PACA_EXGEN)
-EXCEPTION_PROLOG_1(PACA_EXGEN, NOTEST, 0x300)
+EXCEPTION_PROLOG_1 EXC_STD, PACA_EXGEN, 0, 0x300
 	mfspr	r10,SPRN_DAR
 	mfspr	r11,SPRN_DSISR
 	std	r10,PACA_EXGEN+EX_DAR(r13)
@@ -642,7 +652,7 @@ EXCEPTION_PROLOG_0(PACA_EXSLB)
 EXC_REAL_END(data_access_slb, 0x380, 0x80)
 
 TRAMP_REAL_BEGIN(tramp_real_data_access_slb)
-EXCEPTION_PROLOG_1(PACA_EXSLB, KVMTEST_PR, 0x380)
+EXCEPTION_PROLOG_1 EXC_STD, PACA_EXSLB, 1, 0x380
 	mfspr	r10,SPRN_DAR
 	std	r10,PACA_EXSLB+EX_DAR(r13)
 EXCEPTION_PROLOG_2_REAL data_access_slb_common, EXC_STD, 1
@@ -650,7 +660,7 @@ EXCEPTION_PROLOG_2_REAL data_access_slb_common, EXC_STD, 1
 EXC_VIRT_BEGIN(data_access_slb, 0x4380, 0x80)
 SET_SCRATCH0(r13)		/* save r13 */
 EXCEPTION_PROLOG_0(PACA_EXSLB)
-EXCEPTION_PROLOG_1(PACA_EXSLB, NOTEST, 0x380)
+EXCEPTION_PROLOG_1 EXC_STD, PACA_EXSLB, 0, 0x380
 	mfspr	r10,SPRN_DAR
 	std	r10,PACA_EXSLB+EX_DAR(r13)
 EXCEPTION_PROLOG_2_VIRT data_access_slb_common, EXC_STD
@@ -705,11 +715,11 @@ ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_TYPE_RADIX)
 
 
 EXC_REAL_BEGIN(instruction_access_slb, 0x480, 0x80)
-EXCEPTION_PROLOG(PACA_EXSLB, instruction_access_slb_common, EXC_STD, KVMTEST_PR, 0x480);
+EXCEPTION_PROLOG(PACA_EXSLB, instruction_access_slb_common, EXC_STD, 1, 0x480);
 EXC_REAL_END(instruction_access_slb, 0x480, 0x80)
 
 EXC_VIRT_BEGIN(instruction_access_slb, 0x4480, 0x80)
-EXCEPTION_RELON_PROLOG(PACA_EXSLB, instruction_access_slb_common, EXC_STD, NOTEST, 0x480);
+EXCEPTION_RELON_PROLOG(PACA_EXSLB, instruction_access_slb_common, EXC_STD, 0, 0x480);
 EXC_VIRT_END(instruction_access_slb, 0x4480, 0x80)
 
 TRAMP_KVM(PACA_EXSLB, 0x480)
@@ -757,7 +767,7 @@ hardware_interrupt_relon_hv:
 					    IRQS_DISABLED)
 	FTR_SECTION_ELSE
 		__MASKABLE_RELON_EXCEPTION(0x500, hardware_interrupt_common,
-					   EXC_STD, SOFTEN_TEST_PR, IRQS_DISABLED)
+					   EXC_STD, 1, IRQS_DISABLED)
 	ALT_FTR_SECTION_END_IFSET(CPU_FTR_HVMODE)
 EXC_VIRT_END(hardware_interrupt, 0x4500, 0x100)
 
@@ -769,7 +779,7 @@ EXC_COMMON_ASYNC(hardware_interrupt_common, 0x500, do_IRQ)
 EXC_REAL_BEGIN(alignment, 0x600, 0x100)
 SET_SCRATCH0(r13)		/* save r13 */
 EXCEPTION_PROLOG_0(PACA_EXGEN)
-EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_PR, 0x600)
+EXCEPTION_PROLOG_1 EXC_STD, PACA_EXGEN, 1, 0x600
 	mfspr	r10,SPRN_DAR
 	mfspr	r11,SPRN_DSISR
 	std	r10,PACA_EXGEN+EX_DAR(r13)
@@ -780,7 +790,7 @@ EXC_REAL_END(alignment, 0x600, 0x100)
 EXC_VIRT_BEGIN(alignment, 0x4600, 0x100)
 SET_SCRATCH0(r13)		/* save r13 */
 EXCEPTION_PROLOG_0(PACA_EXGEN)
-EXCEPTION_PROLOG_1(PACA_EXGEN, NOTEST, 0x600)
+EXCEPTION_PROLOG_1 EXC_STD, PACA_EXGEN, 0, 0x600
 	mfspr	r10,SPRN_DAR
 	mfspr	r11,SPRN_DSISR
 	std	r10,PACA_EXGEN+EX_DAR(r13)
@@ -946,7 +956,7 @@ EXC_COMMON(trap_0b_common, 0xb00, unknown_exception)
 	GET_PACA(r13);							\
 	std	r10,PACA_EXGEN+EX_R10(r13);				\
 	INTERRUPT_TO_KERNEL;						\
-	KVMTEST_PR(0xc00); /* uses r10, branch to do_kvm_0xc00_system_call */ \
+	KVMTEST EXC_STD 0xc00 ; /* uses r10, branch to do_kvm_0xc00_system_call */ \
 	HMT_MEDIUM;							\
 	mfctr	r9;
 
@@ -1109,7 +1119,7 @@ __TRAMP_REAL_OOL_MASKABLE_HV(hmi_exception, 0xe60, IRQS_DISABLED)
 EXC_VIRT_NONE(0x4e60, 0x20)
 TRAMP_KVM_HV(PACA_EXGEN, 0xe60)
 TRAMP_REAL_BEGIN(hmi_exception_early)
-	EXCEPTION_PROLOG_1(PACA_EXGEN, KVMTEST_HV, 0xe60)
+	EXCEPTION_PROLOG_1 EXC_HV, PACA_EXGEN, 1, 0xe60
 	mr	r10,r1			/* Save r1 */
 	ld	r1,PACAEMERGSP(r13)	/* Use emergency stack for realmode */
 	subi	r1,r1,INT_FRAME_SIZE	/* alloc stack frame		*/
@@ -1311,7 +1321,7 @@ EXC_VIRT_NONE(0x5400, 0x100)
 EXC_REAL_BEGIN(denorm_exception_hv, 0x1500, 0x100)
 	mtspr	SPRN_SPRG_HSCRATCH0,r13
 	EXCEPTION_PROLOG_0(PACA_EXGEN)
-	EXCEPTION_PROLOG_1(PACA_EXGEN, NOTEST, 0x1500)
+	EXCEPTION_PROLOG_1 EXC_HV, PACA_EXGEN, 0, 0x1500
 
 #ifdef CONFIG_PPC_DENORMALISATION
 	mfspr	r10,SPRN_HSRR1
@@ -1319,7 +1329,7 @@ EXC_REAL_BEGIN(denorm_exception_hv, 0x1500, 0x100)
 	bne+	denorm_assist
 #endif
 
-	KVMTEST_HV(0x1500)
+	KVMTEST EXC_HV 0x1500
 	EXCEPTION_PROLOG_2_REAL denorm_common, EXC_HV, 1
 EXC_REAL_END(denorm_exception_hv, 0x1500, 0x100)
 
-- 
2.20.1


^ permalink raw reply related

page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox