| 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: <1724912338.229802-1-xuanzhuo@linux.alibaba.com>
Date: Thu, 29 Aug 2024 14:18:58 +0800
From: Xuan Zhuo <xuanzhuo@...ux.alibaba.com>
To: "Si-Wei Liu" <si-wei.liu@...cle.com>
Cc: netdev@...r.kernel.org,
Jason Wang <jasowang@...hat.com>,
Eugenio Pérez <eperezma@...hat.com>,
"David S. Miller" <davem@...emloft.net>,
Eric Dumazet <edumazet@...gle.com>,
Jakub Kicinski <kuba@...nel.org>,
Paolo Abeni <pabeni@...hat.com>,
virtualization@...ts.linux.dev,
"Linux regression tracking (Thorsten Leemhuis)" <regressions@...mhuis.info>,
Darren Kenny <darren.kenny@...cle.com>,
"Michael S. Tsirkin" <mst@...hat.com>
Subject: Re: [PATCH net] virtio-net: fix overflow inside virtnet_rq_alloc
On Wed, 28 Aug 2024 12:57:56 -0700, "Si-Wei Liu" <si-wei.liu@...cle.com> wrote:
> Just in case Xuan missed the last email while his email server kept
> rejecting incoming emails in the last week.: the patch doesn't seem fix
> the regression.
>
> Xuan, given this is not very hard to reproduce and we have clearly
> stated how to, could you try to get the patch verified in house before
> posting to upstream? Or you were unable to reproduce locally?
Of course I have tested it locally first. And the crash reported this time is
different from the previous one. I really don't know where the problem is. I may
make a new patch based on Jason's suggestion, which should solve the problem of
occupied frag space that you have been worried about.
Of course, before sending the patch, I will reproduce and test it.
Thanks.
>
> Thanks,
> -Siwei
>
> On 8/21/2024 9:47 AM, Darren Kenny wrote:
> > Hi Michael,
> >
> > On Tuesday, 2024-08-20 at 12:50:39 -04, Michael S. Tsirkin wrote:
> >> On Tue, Aug 20, 2024 at 03:19:13PM +0800, Xuan Zhuo wrote:
> >>> leads to regression on VM with the sysctl value of:
> >>>
> >>> - net.core.high_order_alloc_disable=1
> >>
> >>
> >>
> >>> which could see reliable crashes or scp failure (scp a file 100M in size
> >>> to VM):
> >>>
> >>> The issue is that the virtnet_rq_dma takes up 16 bytes at the beginning
> >>> of a new frag. When the frag size is larger than PAGE_SIZE,
> >>> everything is fine. However, if the frag is only one page and the
> >>> total size of the buffer and virtnet_rq_dma is larger than one page, an
> >>> overflow may occur. In this case, if an overflow is possible, I adjust
> >>> the buffer size. If net.core.high_order_alloc_disable=1, the maximum
> >>> buffer size is 4096 - 16. If net.core.high_order_alloc_disable=0, only
> >>> the first buffer of the frag is affected.
> >>>
> >>> Fixes: f9dac92ba908 ("virtio_ring: enable premapped mode whatever use_dma_api")
> >>> Reported-by: "Si-Wei Liu" <si-wei.liu@...cle.com>
> >>> Closes: http://lore.kernel.org/all/8b20cc28-45a9-4643-8e87-ba164a540c0a@oracle.com
> >>> Signed-off-by: Xuan Zhuo <xuanzhuo@...ux.alibaba.com>
> >>
> >> Darren, could you pls test and confirm?
> > Unfortunately with this change I seem to still get a panic as soon as I start a
> > download using wget:
> >
> > [ 144.055630] Kernel panic - not syncing: corrupted stack end detected inside scheduler
> > [ 144.056249] CPU: 8 PID: 37894 Comm: sleep Kdump: loaded Not tainted 6.10.0-1.el8uek.x86_64 #2
> > [ 144.056850] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.0-4.module+el8.9.0+90173+a3f3e83a 04/01/2014
> > [ 144.057585] Call Trace:
> > [ 144.057791] <TASK>
> > [ 144.057973] panic+0x347/0x370
> > [ 144.058223] schedule_debug.isra.0+0xfb/0x100
> > [ 144.058565] __schedule+0x58/0x6a0
> > [ 144.058838] ? refill_stock+0x26/0x50
> > [ 144.059120] ? srso_alias_return_thunk+0x5/0xfbef5
> > [ 144.059473] do_task_dead+0x42/0x50
> > [ 144.059752] do_exit+0x31e/0x4b0
> > [ 144.060011] ? __audit_syscall_entry+0xee/0x150
> > [ 144.060352] do_group_exit+0x30/0x80
> > [ 144.060633] __x64_sys_exit_group+0x18/0x20
> > [ 144.060946] do_syscall_64+0x8c/0x1c0
> > [ 144.061228] ? srso_alias_return_thunk+0x5/0xfbef5
> > [ 144.061570] ? __audit_filter_op+0xbe/0x140
> > [ 144.061873] ? srso_alias_return_thunk+0x5/0xfbef5
> > [ 144.062204] ? audit_reset_context+0x232/0x310
> > [ 144.062514] ? srso_alias_return_thunk+0x5/0xfbef5
> > [ 144.062851] ? syscall_exit_work+0x103/0x130
> > [ 144.063148] ? srso_alias_return_thunk+0x5/0xfbef5
> > [ 144.063473] ? syscall_exit_to_user_mode+0x77/0x220
> > [ 144.063813] ? srso_alias_return_thunk+0x5/0xfbef5
> > [ 144.064142] ? do_syscall_64+0xb9/0x1c0
> > [ 144.064411] ? srso_alias_return_thunk+0x5/0xfbef5
> > [ 144.064747] ? do_syscall_64+0xb9/0x1c0
> > [ 144.065018] ? srso_alias_return_thunk+0x5/0xfbef5
> > [ 144.065345] ? do_read_fault+0x109/0x1b0
> > [ 144.065628] ? srso_alias_return_thunk+0x5/0xfbef5
> > [ 144.065961] ? do_fault+0x1aa/0x2f0
> > [ 144.066212] ? handle_pte_fault+0x102/0x1a0
> > [ 144.066503] ? srso_alias_return_thunk+0x5/0xfbef5
> > [ 144.066836] ? __handle_mm_fault+0x5ed/0x710
> > [ 144.067137] ? srso_alias_return_thunk+0x5/0xfbef5
> > [ 144.067464] ? __count_memcg_events+0x72/0x110
> > [ 144.067779] ? srso_alias_return_thunk+0x5/0xfbef5
> > [ 144.068106] ? count_memcg_events.constprop.0+0x26/0x50
> > [ 144.068457] ? srso_alias_return_thunk+0x5/0xfbef5
> > [ 144.068788] ? handle_mm_fault+0xae/0x320
> > [ 144.069068] ? srso_alias_return_thunk+0x5/0xfbef5
> > [ 144.069395] ? do_user_addr_fault+0x34a/0x6b0
> > [ 144.069708] entry_SYSCALL_64_after_hwframe+0x76/0x7e
> > [ 144.070049] RIP: 0033:0x7fc5524f9c66
> > [ 144.070307] Code: Unable to access opcode bytes at 0x7fc5524f9c3c.
> > [ 144.070720] RSP: 002b:00007ffee052beb8 EFLAGS: 00000246 ORIG_RAX: 00000000000000e7
> > [ 144.071214] RAX: ffffffffffffffda RBX: 00007fc5527bb860 RCX: 00007fc5524f9c66
> > [ 144.071684] RDX: 0000000000000000 RSI: 000000000000003c RDI: 0000000000000000
> > [ 144.072146] RBP: 0000000000000000 R08: 00000000000000e7 R09: ffffffffffffff78
> > [ 144.072608] R10: 00007ffee052bdef R11: 0000000000000246 R12: 00007fc5527bb860
> > [ 144.073076] R13: 0000000000000002 R14: 00007fc5527c4528 R15: 0000000000000000
> > [ 144.073543] </TASK>
> > [ 144.074780] Kernel Offset: 0x37c00000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff)
> >
> > Thanks,
> >
> > Darren.
> >
> >>> ---
> >>> drivers/net/virtio_net.c | 12 +++++++++---
> >>> 1 file changed, 9 insertions(+), 3 deletions(-)
> >>>
> >>> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> >>> index c6af18948092..e5286a6da863 100644
> >>> --- a/drivers/net/virtio_net.c
> >>> +++ b/drivers/net/virtio_net.c
> >>> @@ -918,9 +918,6 @@ static void *virtnet_rq_alloc(struct receive_queue *rq, u32 size, gfp_t gfp)
> >>> void *buf, *head;
> >>> dma_addr_t addr;
> >>>
> >>> - if (unlikely(!skb_page_frag_refill(size, alloc_frag, gfp)))
> >>> - return NULL;
> >>> -
> >>> head = page_address(alloc_frag->page);
> >>>
> >>> dma = head;
> >>> @@ -2421,6 +2418,9 @@ static int add_recvbuf_small(struct virtnet_info *vi, struct receive_queue *rq,
> >>> len = SKB_DATA_ALIGN(len) +
> >>> SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
> >>>
> >>> + if (unlikely(!skb_page_frag_refill(len, &rq->alloc_frag, gfp)))
> >>> + return -ENOMEM;
> >>> +
> >>> buf = virtnet_rq_alloc(rq, len, gfp);
> >>> if (unlikely(!buf))
> >>> return -ENOMEM;
> >>> @@ -2521,6 +2521,12 @@ static int add_recvbuf_mergeable(struct virtnet_info *vi,
> >>> */
> >>> len = get_mergeable_buf_len(rq, &rq->mrg_avg_pkt_len, room);
> >>>
> >>> + if (unlikely(!skb_page_frag_refill(len + room, alloc_frag, gfp)))
> >>> + return -ENOMEM;
> >>> +
> >>> + if (!alloc_frag->offset && len + room + sizeof(struct virtnet_rq_dma) > alloc_frag->size)
> >>> + len -= sizeof(struct virtnet_rq_dma);
> >>> +
> >>> buf = virtnet_rq_alloc(rq, len + room, gfp);
> >>> if (unlikely(!buf))
> >>> return -ENOMEM;
> >>> --
> >>> 2.32.0.3.g01195cf9f
>
Powered by blists - more mailing lists