| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <9deb3725-8991-43d1-8c3d-56523fabff28@redhat.com>
Date: Thu, 10 Apr 2025 15:18:00 +0200
From: David Hildenbrand <david@...hat.com>
To: Gavin Shan <gshan@...hat.com>, linux-mm@...ck.org
Cc: linux-kernel@...r.kernel.org, adityag@...ux.ibm.com,
donettom@...ux.ibm.com, osalvador@...e.de, gregkh@...uxfoundation.org,
rafael@...nel.org, dakr@...nel.org, akpm@...ux-foundation.org,
shan.gavin@...il.com
Subject: Re: [PATCH] drivers/base/memory: Avoid overhead from
for_each_present_section_nr()
On 10.04.25 14:51, Gavin Shan wrote:
> for_each_present_section_nr() was introduced to add_boot_memory_block()
> by commit 61659efdb35c ("drivers/base/memory: improve add_boot_memory_block()").
> It causes unnecessary overhead when the present sections are really
> sparse. next_present_section_nr() called by the macro to find the next
> present section, which is far away from the spanning sections in the
> specified block. Too much time consumed by next_present_section_nr()
> in this case, which can lead to softlockup as observed by Aditya Gupta
> on IBM Power10 machine.
>
> watchdog: BUG: soft lockup - CPU#248 stuck for 22s! [swapper/248:1]
> Modules linked in:
> CPU: 248 UID: 0 PID: 1 Comm: swapper/248 Not tainted 6.15.0-rc1-next-20250408 #1 VOLUNTARY
> Hardware name: 9105-22A POWER10 (raw) 0x800200 opal:v7.1-107-gfda75d121942 PowerNV
> NIP: c00000000209218c LR: c000000002092204 CTR: 0000000000000000
> REGS: c00040000418fa30 TRAP: 0900 Not tainted (6.15.0-rc1-next-20250408)
> MSR: 9000000002009033 <SF,HV,VEC,EE,ME,IR,DR,RI,LE> CR: 28000428 XER: 00000000
> CFAR: 0000000000000000 IRQMASK: 0
> GPR00: c000000002092204 c00040000418fcd0 c000000001b08100 0000000000000040
> GPR04: 0000000000013e00 c000c03ffebabb00 0000000000c03fff c000400fff587f80
> GPR08: 0000000000000000 00000000001196f7 0000000000000000 0000000028000428
> GPR12: 0000000000000000 c000000002e80000 c00000000001007c 0000000000000000
> GPR16: 0000000000000000 0000000000000000 0000000000000000 0000000000000000
> GPR20: 0000000000000000 0000000000000000 0000000000000000 0000000000000000
> GPR24: 0000000000000000 0000000000000000 0000000000000000 0000000000000000
> GPR28: c000000002df7f70 0000000000013dc0 c0000000011dd898 0000000008000000
> NIP [c00000000209218c] memory_dev_init+0x114/0x1e0
> LR [c000000002092204] memory_dev_init+0x18c/0x1e0
> Call Trace:
> [c00040000418fcd0] [c000000002092204] memory_dev_init+0x18c/0x1e0 (unreliable)
> [c00040000418fd50] [c000000002091348] driver_init+0x78/0xa4
> [c00040000418fd70] [c0000000020063ac] kernel_init_freeable+0x22c/0x370
> [c00040000418fde0] [c0000000000100a8] kernel_init+0x34/0x25c
> [c00040000418fe50] [c00000000000cd94] ret_from_kernel_user_thread+0x14/0x1c
>
> Avoid the overhead by folding for_each_present_section_nr() to the outer
> loop. add_boot_memory_block() is dropped after that.
>
> Fixes: 61659efdb35c ("drivers/base/memory: improve add_boot_memory_block()")
> Closes: https://lore.kernel.org/linux-mm/20250409180344.477916-1-adityag@linux.ibm.com
> Reported-by: Aditya Gupta <adityag@...ux.ibm.com>
> Signed-off-by: Gavin Shan <gshan@...hat.com>
> ---
> drivers/base/memory.c | 41 +++++++++++++++++------------------------
> 1 file changed, 17 insertions(+), 24 deletions(-)
>
> diff --git a/drivers/base/memory.c b/drivers/base/memory.c
> index 8f3a41d9bfaa..19469e7f88c2 100644
> --- a/drivers/base/memory.c
> +++ b/drivers/base/memory.c
> @@ -816,21 +816,6 @@ static int add_memory_block(unsigned long block_id, unsigned long state,
> return 0;
> }
>
> -static int __init add_boot_memory_block(unsigned long base_section_nr)
> -{
> - unsigned long nr;
> -
> - for_each_present_section_nr(base_section_nr, nr) {
> - if (nr >= (base_section_nr + sections_per_block))
> - break;
> -
> - return add_memory_block(memory_block_id(base_section_nr),
> - MEM_ONLINE, NULL, NULL);
> - }
> -
> - return 0;
> -}
> -
> static int add_hotplug_memory_block(unsigned long block_id,
> struct vmem_altmap *altmap,
> struct memory_group *group)
> @@ -957,7 +942,7 @@ static const struct attribute_group *memory_root_attr_groups[] = {
> void __init memory_dev_init(void)
> {
> int ret;
> - unsigned long block_sz, nr;
> + unsigned long block_sz, block_id, nr;
>
> /* Validate the configured memory block size */
> block_sz = memory_block_size_bytes();
> @@ -970,15 +955,23 @@ void __init memory_dev_init(void)
> panic("%s() failed to register subsystem: %d\n", __func__, ret);
>
> /*
> - * Create entries for memory sections that were found
> - * during boot and have been initialized
> + * Create entries for memory sections that were found during boot
> + * and have been initialized. Use @block_id to track the last
> + * handled block and initialize it to an invalid value (ULONG_MAX)
> + * to bypass the block ID matching check for the first present
> + * block so that it can be covered.
> */
> - for (nr = 0; nr <= __highest_present_section_nr;
> - nr += sections_per_block) {
> - ret = add_boot_memory_block(nr);
> - if (ret)
> - panic("%s() failed to add memory block: %d\n", __func__,
> - ret);
> + block_id = ULONG_MAX;
> + for_each_present_section_nr(0, nr) {
> + if (block_id != ULONG_MAX && memory_block_id(nr) == block_id)
> + continue;
> +
> + block_id = memory_block_id(nr);
> + ret = add_memory_block(block_id, MEM_ONLINE, NULL, NULL);
> + if (ret) {
> + panic("%s() failed to add memory block: %d\n",
> + __func__, ret);
> + }
> }
> }
>
Staring at the end result and the particularly long comment, are we now
really any better than before 61659efdb35c?
Revert instead?
--
Cheers,
David / dhildenb
Powered by blists - more mailing lists