linux-kernel - Re: [PATCH v3 3/4] lib/find_bit: optimize find_next

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
Hash Suite for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:   Fri, 09 Sep 2022 14:24:31 +0200
From:   Sven Schnelle <svens@...ux.ibm.com>
To:     Yury Norov <yury.norov@...il.com>
Cc:     Valentin Schneider <vschneid@...hat.com>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        linux-kernel@...r.kernel.org,
        Alexey Klimov <klimov.linux@...il.com>,
        Andy Shevchenko <andriy.shevchenko@...ux.intel.com>,
        Andy Whitcroft <apw@...onical.com>,
        Catalin Marinas <catalin.marinas@....com>,
        David Laight <David.Laight@...lab.com>,
        Dennis Zhou <dennis@...nel.org>,
        Guenter Roeck <linux@...ck-us.net>,
        Kees Cook <keescook@...omium.org>,
        Rasmus Villemoes <linux@...musvillemoes.dk>,
        Russell King <linux@...linux.org.uk>
Subject: Re: [PATCH v3 3/4] lib/find_bit: optimize find_next_bit() functions

Hi Yury,

Yury Norov <yury.norov@...il.com> writes:

> On Wed, Sep 07, 2022 at 05:27:08PM +0100, Valentin Schneider wrote:
>> On 27/08/22 10:58, Yury Norov wrote:
>> > +#define FIND_NEXT_BIT(FETCH, MUNGE, size, start)				\
>> > +({										\
>> > +	unsigned long mask, idx, tmp, sz = (size), __start = (start);		\
>> > +										\
>> > +	if (unlikely(__start >= sz))						\
>> > +		goto out;							\
>> > +										\
>> > +	mask = MUNGE(BITMAP_FIRST_WORD_MASK(__start));				\
>> > +	idx = __start / BITS_PER_LONG;						\
>> > +										\
>> > +	for (tmp = (FETCH) & mask; !tmp; tmp = (FETCH)) {			\
>> > +		if (idx > sz / BITS_PER_LONG)					\
>> 
>> Does that want to be
>
> Yes, I already fixed this.
>  
>>                 if (idx + 1 >= sz / BITS_PER_LONG)
>> 
>> ?

Did you push that already? We're still seeing crashes in CI, and the
'idx + 1' doesnt seem to be in next-20220908. Adding it makes the
out-of-bound access go away, but the kernel will crash later in the
block mq code:

[    0.760803] NET: Registered PF_PACKET protocol family
[    0.760808] bridge: filtering via arp/ip/ip6tables is no longeravailable by default. Update your scripts to load br_netfilter if youneed this.
[    0.760938] Key type dns_resolver registered
[    0.760967] cio: Channel measurement facility initialized using format extended (mode autodetected)
[    0.761789] dasd-eckd 0.0.3850: A channel path to the device has become operational
[    0.762577] dasd-eckd 0.0.3850: A channel path to the device has become operational
[    0.772623] dasd-eckd 0.0.3850: New DASD 3390/0C (CU 3990/01) with 30051 cylinders, 15 heads, 224 sectors
[    0.806105] dasd-eckd 0.0.3850: DASD with 4 KB/block, 21636720 KB total size, 48 KB/track, compatible disk layout
[    0.806131] Unable to handle kernel pointer dereference in virtual kernel address space
[    0.806132] Failing address: fffffffffffff000 TEID: fffffffffffff803
[    0.806134] Fault in home space mode while using kernel ASCE.
[    0.806137] AS:0000000001f14007 R3:0000000081360007 S:0000000000000020
[    0.806154] Oops: 0038 ilc:2 [#1] SMP
[    0.806333] Modules linked in:
[    0.806337] CPU: 0 PID: 42 Comm: kworker/0:3 Not tainted 6.0.0-rc4-next-20220908-dirty #463
[    0.806339] Hardware name: IBM 3906 M04 704 (z/VM 7.1.0)
[    0.806340] Workqueue: events do_kick_device
[    0.806344] Krnl PSW : 0704c00180000000 0000000000811ee4 (blk_rq_merge_ok+0x1c/0x120)
[    0.806362]            R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:3 CC:0 PM:0 RI:0 EA:3
[    0.806368] Krnl GPRS: 8000000000000001 00000000018461e0 ffffffffffffffb8 0000000080ac6200
[    0.806379]            00000000ffffbfff 00000000fee80000 0000000001fbe4a8 00000000822cbda0
[    0.806387]            0000000000000008 0000000080ac6200 000003ff7f842108 ffffffffffffffb8
[    0.806390]            00000000822aa100 00000380003b78d0 00000380003b7590 00000380003b7550
[    0.806413] Krnl Code: 0000000000811ed4: b90400ef            lgr     %r14,%r15
[    0.806413]            0000000000811ed8: e3f0ffc0ff71        lay     %r15,-64(%r15)
[    0.806413]           #0000000000811ede: e3e0f0980024        stg     %r14,152(%r15)
[    0.806413]           >0000000000811ee4: 58102018            l       %r1,24(%r2)
[    0.806413]            0000000000811ee8: 4210f0a7            stc     %r1,167(%r15)
[    0.806413]            0000000000811eec: ec4138bf0055        risbg   %r4,%r1,56,191,0
[    0.806413]            0000000000811ef2: b90400b2            lgr     %r11,%r2
[    0.806413]            0000000000811ef6: b90400a3            lgr     %r10,%r3
[    0.806427] Call Trace:
[    0.806431]  [<0000000000811ee4>] blk_rq_merge_ok+0x1c/0x120
[    0.806470]  [<0000000000d07a5a>] blk_bio_list_merge+0x7a/0xd0
[    0.806474]  [<0000000000820456>] blk_mq_sched_bio_merge+0xde/0x150
[    0.806478]  [<0000000000815110>] blk_mq_get_new_requests+0x88/0x190
[    0.806481]  [<000000000081a346>] blk_mq_submit_bio+0x286/0x438
[    0.806484]  [<00000000008090d2>] __submit_bio+0x12a/0x1d8
[    0.806498]  [<0000000000809738>] submit_bio_noacct_nocheck+0xa0/0xf8
[    0.806502]  [<000000000047e02e>] submit_bh_wbc+0x16e/0x1b0
[    0.806505]  [<0000000000481384>] block_read_full_folio+0x2ec/0x400
[    0.806508]  [<0000000000348518>] filemap_read_folio+0x38/0xc8
[    0.806512]  [<000000000034998e>] do_read_cache_folio+0x13e/0x1e8
[    0.806516]  [<0000000000349a60>] read_cache_folio+0x28/0x38
[    0.806520]  [<0000000000826d56>] read_part_sector+0x5e/0xe8
[    0.806524]  [<0000000000828ed4>] read_lba+0xb4/0x170
[    0.806528]  [<00000000008294d6>] find_valid_gpt.constprop.0+0xde/0x568
[    0.806532]  [<0000000000829c92>] efi_partition+0x332/0x398
[    0.806559]  [<0000000000826586>] check_partition+0x12e/0x248
[    0.806567]  [<0000000000826a1c>] bdev_disk_changed.part.0+0xcc/0x378
[    0.806578]  [<0000000000ca8016>] dasd_scan_partitions+0x76/0x120
[    0.806582]  [<0000000000ca1018>] dasd_change_state+0x310/0x3a0
[    0.806584]  [<0000000000ca10e8>] do_kick_device+0x40/0xc8
[    0.806587]  [<0000000000174ae0>] process_one_work+0x200/0x458
[    0.806591]  [<000000000017526e>] worker_thread+0x66/0x480
[    0.806594]  [<000000000017cf98>] kthread+0x108/0x110
[    0.806596]  [<0000000000103354>] __ret_from_fork+0x3c/0x58
[    0.806600]  [<0000000000d1eb0a>] ret_from_fork+0xa/0x40
[    0.806604] Last Breaking-Event-Address:
[    0.806606]  [<0000000000d07a54>] blk_bio_list_merge+0x74/0xd0
[    0.806610] Kernel panic - not syncing: Fatal exception: panic_on_oops

Reverting the 'lib/find_bit: optimize find_next_bit() functions' patch
'fixes' these issues.

Regards
Sven