* [PATCH] mm/memblock: use memblock_insert_region() for the empty array
@ 2015-12-05 7:23 Alexander Kuleshov
2015-12-09 18:46 ` Dan Williams
2015-12-09 18:59 ` Tyler Baker
0 siblings, 2 replies; 3+ messages in thread
From: Alexander Kuleshov @ 2015-12-05 7:23 UTC (permalink / raw
To: Andrew Morton
Cc: Tony Luck, Tang Chen, Pekka Enberg, Wei Yang, linux-mm,
linux-kernel, Alexander Kuleshov
We have the special case for an empty array in the memblock_add_range()
function. In the same time we have almost the same functional in the
memblock_insert_region() function. Let's use the memblock_insert_region()
instead of direct initialization.
Signed-off-by: Alexander Kuleshov <kuleshovmail@gmail.com>
---
mm/memblock.c | 14 +++++++-------
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/mm/memblock.c b/mm/memblock.c
index d300f13..e8a897d 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -496,12 +496,16 @@ static void __init_memblock memblock_insert_region(struct memblock_type *type,
struct memblock_region *rgn = &type->regions[idx];
BUG_ON(type->cnt >= type->max);
- memmove(rgn + 1, rgn, (type->cnt - idx) * sizeof(*rgn));
+ /* special case for empty array */
+ if (idx)
+ {
+ memmove(rgn + 1, rgn, (type->cnt - idx) * sizeof(*rgn));
+ type->cnt++;
+ }
rgn->base = base;
rgn->size = size;
rgn->flags = flags;
memblock_set_region_node(rgn, nid);
- type->cnt++;
type->total_size += size;
}
@@ -536,11 +540,7 @@ int __init_memblock memblock_add_range(struct memblock_type *type,
/* special case for empty array */
if (type->regions[0].size == 0) {
WARN_ON(type->cnt != 1 || type->total_size);
- type->regions[0].base = base;
- type->regions[0].size = size;
- type->regions[0].flags = flags;
- memblock_set_region_node(&type->regions[0], nid);
- type->total_size = size;
+ memblock_insert_region(type, 0, base, size, nid, flags);
return 0;
}
repeat:
--
2.5.0
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH] mm/memblock: use memblock_insert_region() for the empty array
2015-12-05 7:23 [PATCH] mm/memblock: use memblock_insert_region() for the empty array Alexander Kuleshov
@ 2015-12-09 18:46 ` Dan Williams
2015-12-09 18:59 ` Tyler Baker
1 sibling, 0 replies; 3+ messages in thread
From: Dan Williams @ 2015-12-09 18:46 UTC (permalink / raw
To: Alexander Kuleshov
Cc: Andrew Morton, Tony Luck, Tang Chen, Pekka Enberg, Wei Yang,
linux-mm, Linux Kernel Mailing List
On Fri, Dec 4, 2015 at 11:23 PM, Alexander Kuleshov
<kuleshovmail@gmail.com> wrote:
> We have the special case for an empty array in the memblock_add_range()
> function. In the same time we have almost the same functional in the
> memblock_insert_region() function. Let's use the memblock_insert_region()
> instead of direct initialization.
>
> Signed-off-by: Alexander Kuleshov <kuleshovmail@gmail.com>
> ---
> mm/memblock.c | 14 +++++++-------
> 1 file changed, 7 insertions(+), 7 deletions(-)
>
> diff --git a/mm/memblock.c b/mm/memblock.c
> index d300f13..e8a897d 100644
> --- a/mm/memblock.c
> +++ b/mm/memblock.c
> @@ -496,12 +496,16 @@ static void __init_memblock memblock_insert_region(struct memblock_type *type,
> struct memblock_region *rgn = &type->regions[idx];
>
> BUG_ON(type->cnt >= type->max);
> - memmove(rgn + 1, rgn, (type->cnt - idx) * sizeof(*rgn));
> + /* special case for empty array */
> + if (idx)
> + {
> + memmove(rgn + 1, rgn, (type->cnt - idx) * sizeof(*rgn));
> + type->cnt++;
> + }
> rgn->base = base;
> rgn->size = size;
> rgn->flags = flags;
> memblock_set_region_node(rgn, nid);
> - type->cnt++;
> type->total_size += size;
> }
>
> @@ -536,11 +540,7 @@ int __init_memblock memblock_add_range(struct memblock_type *type,
> /* special case for empty array */
> if (type->regions[0].size == 0) {
> WARN_ON(type->cnt != 1 || type->total_size);
> - type->regions[0].base = base;
> - type->regions[0].size = size;
> - type->regions[0].flags = flags;
> - memblock_set_region_node(&type->regions[0], nid);
> - type->total_size = size;
> + memblock_insert_region(type, 0, base, size, nid, flags);
> return 0;
> }
> repeat:
> --
Latest -next (20151209) fails to boot due to this patch. Here's the
backlog. Also reported here with a different failing signature:
https://lkml.org/lkml/2015/12/9/340
[ 0.860371] BUG: unable to handle kernel paging request at ffff880000099000
[ 0.862642] IP: [<ffffffff814686f4>] __memset+0x24/0x30
[ 0.864010] PGD 2f6a067 PUD 2f6b067 PMD 2f6c067 PTE 8000000000099161
[ 0.865787] Oops: 0003 [#1] SMP
[ 0.866856] Dumping ftrace buffer:
[ 0.867823] (ftrace buffer empty)
[ 0.868850] Modules linked in:
[ 0.869870] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.4.0-rc4+ #2187
[ 0.871322] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[ 0.872761] task: ffff88030e490000 ti: ffff88030e498000 task.ti:
ffff88030e498000
[ 0.874818] RIP: 0010:[<ffffffff814686f4>] [<ffffffff814686f4>]
__memset+0x24/0x30
[ 0.876979] RSP: 0000:ffff88030e49b878 EFLAGS: 00010206
[ 0.878331] RAX: 5a5a5a5a5a5a5a5a RBX: ffff88031314e300 RCX: 0000000000000600
[ 0.879945] RDX: 0000000000000000 RSI: 000000000000005a RDI: ffff880000099000
[ 0.881559] RBP: ffff88030e49b8c8 R08: ffffffff81cbf9b4 R09: ffff880000098000
[ 0.883185] R10: 0000000000000000 R11: ffffffff81cbf992 R12: 0000000000000002
[ 0.884797] R13: 0000000000020022 R14: 0000000000098000 R15: ffffea0000002600
[ 0.886344] FS: 0000000000000000(0000) GS:ffff88031fc00000(0000)
knlGS:0000000000000000
[ 0.888462] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[ 0.889870] CR2: ffff880000099000 CR3: 0000000001e09000 CR4: 00000000000006f0
[ 0.892214] Stack:
[ 0.892964] ffffffff8122a2a7 ffff88030e49b8b0 000000028121e4f4
00400000ffffffff
[ 0.895271] ffff880000098000 00000000024080c0 ffff8803131988c0
0000000000000001
[ 0.897563] ffff88031314e300 ffff88031ffeb120 ffff88030e49b9a8
ffffffff8122bf83
[ 0.899872] Call Trace:
[ 0.900686] [<ffffffff8122a2a7>] ? new_slab+0x4a7/0x530
[ 0.901970] [<ffffffff8122bf83>] ___slab_alloc+0x353/0x550
[ 0.903291] [<ffffffff812d89b1>] ? __kernfs_new_node+0x41/0xc0
[ 0.904765] [<ffffffff810f5de1>] ? mark_held_locks+0x71/0x90
[ 0.906076] [<ffffffff812d89b1>] ? __kernfs_new_node+0x41/0xc0
[ 0.907548] [<ffffffff8122c1d1>] __slab_alloc+0x51/0x90
[ 0.908835] [<ffffffff812d89b1>] ? __kernfs_new_node+0x41/0xc0
[ 0.910193] [<ffffffff8122c3b6>] kmem_cache_alloc+0x1a6/0x1f0
[ 0.911658] [<ffffffff812d89b1>] __kernfs_new_node+0x41/0xc0
[ 0.912987] [<ffffffff812d9be6>] kernfs_new_node+0x26/0x50
[ 0.914336] [<ffffffff812db605>] __kernfs_create_file+0x35/0xd0
[ 0.915800] [<ffffffff812dbee0>] sysfs_add_file_mode_ns+0x90/0x1b0
[ 0.917243] [<ffffffff812dc188>] sysfs_add_file+0x18/0x20
[ 0.918584] [<ffffffff812dc946>] sysfs_merge_group+0x56/0xc0
[ 0.919937] [<ffffffff8159b406>] dpm_sysfs_add+0x76/0xd0
[ 0.921983] [<ffffffff8159027c>] device_add+0x45c/0x6a0
[ 0.923295] [<ffffffff810f5fbd>] ? trace_hardirqs_on+0xd/0x10
[ 0.924639] [<ffffffff814e25a5>] acpi_device_add+0x1fd/0x297
[ 0.926004] [<ffffffff814e27c7>] ? acpi_free_pnp_ids+0x50/0x50
[ 0.927331] [<ffffffff814e3233>] acpi_add_single_object+0x4ef/0x55c
[ 0.928812] [<ffffffff814dc9ea>] ? acpi_os_signal_semaphore+0x29/0x35
[ 0.930282] [<ffffffff814e3372>] acpi_bus_check_add+0xd2/0x195
[ 0.931712] [<ffffffff8150190b>] acpi_ns_walk_namespace+0xdf/0x18f
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] mm/memblock: use memblock_insert_region() for the empty array
2015-12-05 7:23 [PATCH] mm/memblock: use memblock_insert_region() for the empty array Alexander Kuleshov
2015-12-09 18:46 ` Dan Williams
@ 2015-12-09 18:59 ` Tyler Baker
1 sibling, 0 replies; 3+ messages in thread
From: Tyler Baker @ 2015-12-09 18:59 UTC (permalink / raw
To: Alexander Kuleshov
Cc: Andrew Morton, Tony Luck, Tang Chen, Pekka Enberg, Wei Yang,
Linux MM, linux-kernel@vger.kernel.org, Kevin's boot bot,
Dan Williams
On 4 December 2015 at 23:23, Alexander Kuleshov <kuleshovmail@gmail.com> wrote:
> We have the special case for an empty array in the memblock_add_range()
> function. In the same time we have almost the same functional in the
> memblock_insert_region() function. Let's use the memblock_insert_region()
> instead of direct initialization.
>
> Signed-off-by: Alexander Kuleshov <kuleshovmail@gmail.com>
Just to add on to the report from Dan, the kernelci.org boot bot has
also detected ~65 new boot failures in next-20151209[1], which have
been bisected to this patch[2]. It doesn't revert cleanly, so I'm
going to try to clean it up by hand and see if it resolves the issue.
Cheers,
Tyler
[1] http://kernelci.org/boot/all/job/next/kernel/next-20151209/
[2] http://hastebin.com/ufiwonuraw.vhdl
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2015-12-09 18:59 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-12-05 7:23 [PATCH] mm/memblock: use memblock_insert_region() for the empty array Alexander Kuleshov
2015-12-09 18:46 ` Dan Williams
2015-12-09 18:59 ` Tyler Baker
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).