On 2020-04-29 2:37 pm, Salil Mehta wrote:
Hi Bin,
From: Bin [mailto:anole1...@gmail.com]
Sent: Wednesday, April 29, 2020 5:14 AM
To: Salil Mehta <salil.me...@huawei.com>
Hi Shlil:
Thank you for your attention, and these are my answers:
1. I don't really understand what you're saying. What's the difference between
DMA buffer and DMA mapping?
It's like a memory block pool and a memory block or something like that?
DMA Mapping: Mapping are translations/associations [IOVA<->HPA OR
IOVA<->GPA(further translated
to HPA by Stage-2)] which are created by the NIC driver. IOMMU hardware
responsible for NIC
IOVA translations is populated with the mappings by the driver before
submitting the DMA buffer
to the hardware for TX/RX.
DMA buffers: Actual Memory allocated by the driver where data could be DMA'ed
(RX'ed or TX'ed)
I think you have missed the important point I mentioned earlier:
If there is a leak of IOVA mapping due to dma_unmap_* not being called
somewhere then at
certain point the throughput will drastically fall and will almost become equal
to zero.
This is due to the exhaustion of available IOVA mapping space in the IOMMU
hardware.
With 64-bit address spaces, you're still likely to run out of memory for
the IOVA structures and pagetables before you run out of the actual
address space that they represent. The slowdown comes from having to
walk the whole the rbtree to search for free space or free a PFN, but
depending on how the allocation pattern interacts with the caching
mechanism that may never happen to a significant degree.
Above condition is very much different than a *memory leak* of the DMA buffer
itself which
will eventually lead to OOM.
Salil.
FYI:
I found an interesting phenomenon that it's just a small part of the running
hosts has this issue, even though they all
have the same kernel, configuration and hardwares, I don't know if this really
mean something.
Another thought for a debugging sanity check is to look at the
intel-iommu tracepoints on a misbehaving system and see whether maps vs.
unmaps look significantly out of balance. You could probably do
something clever with ftrace to look for that kind of pattern in teh DMA
API calls, too.
Robin.
Salil Mehta <salil.me...@huawei.com> 于2020年4月28日周二 下午5:17写道：
Hi Bin,
Few questions:
1. If there is a leak of IOVA due to dma_unmap_* not being called somewhere then
at certain point the throughput will drastically fall and will almost become
equal
to zero. This should be due to unavailability of the mapping anymore. But in
your
case VM is getting killed so this could be actual DMA buffer leak not DMA
mapping
leak. I doubt VM will get killed due to exhaustion of the DMA mappings in the
IOMMU
Layer for a transient reason or even due to mapping/unmapping leak.
2. Could you check if you have TSO offload enabled on Intel 82599? It will help
in reducing the number of mappings and will take off IOVA mapping pressure from
the IOMMU/VT-d? Though I am not sure it will help in reducing the amount of
memory
required for the buffers.
3. Also, have you checked the cpu-usage while your experiment is going on?
Thanks
Salil.
-----Original Message-----
From: iommu [mailto:iommu-boun...@lists.linux-foundation.org] On Behalf Of
Robin Murphy
Sent: Friday, April 24, 2020 5:31 PM
To: Bin <anole1...@gmail.com>
Cc: iommu@lists.linux-foundation.org
Subject: Re: iommu_iova slab eats too much memory
On 2020-04-24 2:20 pm, Bin wrote:
Dear Robin:
Thank you for your explanation. Now, I understand that this could be
NIC driver's fault, but how could I confirm it? Do I have to debug the
driver myself?
I'd start with CONFIG_DMA_API_DEBUG - of course it will chew through
memory about an order of magnitude faster than the IOVAs alone, but it
should shed some light on whether DMA API usage looks suspicious, and
dumping the mappings should help track down the responsible driver(s).
Although the debugfs code doesn't show the stacktrace of where each
mapping was made, I guess it would be fairly simple to tweak that for a
quick way to narrow down where to start looking in an offending driver.
Robin.
Robin Murphy <robin.mur...@arm.com> 于2020年4月24日周五 下午8:15写道：
On 2020-04-24 1:06 pm, Bin wrote:
I'm not familiar with the mmu stuff, so what you mean by "some driver
leaking DMA mappings", is it possible that some other kernel module like
KVM or NIC driver leads to the leaking problem instead of the iommu
module
itself?
Yes - I doubt that intel-iommu itself is failing to free IOVAs when it
should, since I'd expect a lot of people to have noticed that. It's far
more likely that some driver is failing to call dma_unmap_* when it's
finished with a buffer - with the IOMMU disabled that would be a no-op
on x86 with a modern 64-bit-capable device, so such a latent bug could
have been easily overlooked.
Robin.
Bin <anole1...@gmail.com> 于 2020年4月24日周五 20:00写道：
Well, that's the problem! I'm assuming the iommu kernel module is
leaking
memory. But I don't know why and how.
Do you have any idea about it? Or any further information is needed?
Robin Murphy <robin.mur...@arm.com> 于 2020年4月24日周五 19:20写道：
On 2020-04-24 1:40 am, Bin wrote:
Hello? anyone there?
Bin <anole1...@gmail.com> 于2020年4月23日周四 下午5:14写道：
Forget to mention, I've already disabled the slab merge, so this is
what
it is.
Bin <anole1...@gmail.com> 于2020年4月23日周四 下午5:11写道：
Hey, guys:
I'm running a batch of CoreOS boxes, the lsb_release is:
```
# cat /etc/lsb-release
DISTRIB_ID="Container Linux by CoreOS"
DISTRIB_RELEASE=2303.3.0
DISTRIB_CODENAME="Rhyolite"
DISTRIB_DESCRIPTION="Container Linux by CoreOS 2303.3.0 (Rhyolite)"
```
```
# uname -a
Linux cloud-worker-25 4.19.86-coreos #1 SMP Mon Dec 2 20:13:38 -00
2019
x86_64 Intel(R) Xeon(R) CPU E5-2640 v2 @ 2.00GHz GenuineIntel
GNU/Linux
```
Recently, I found my vms constently being killed due to OOM, and
after
digging into the problem, I finally realized that the kernel is
leaking
memory.
Here's my slabinfo:
Active / Total Objects (% used) : 83818306 / 84191607 (99.6%)
Active / Total Slabs (% used) : 1336293 / 1336293 (100.0%)
Active / Total Caches (% used) : 152 / 217 (70.0%)
Active / Total Size (% used) : 5828768.08K / 5996848.72K
(97.2%)
Minimum / Average / Maximum Object : 0.01K / 0.07K / 23.25K
OBJS ACTIVE USE OBJ SIZE SLABS OBJ/SLAB CACHE SIZE NAME
80253888 80253888 100% 0.06K 1253967 64 5015868K
iommu_iova
Do you really have a peak demand of ~80 million simultaneous DMA
buffers, or is some driver leaking DMA mappings?
Robin.
489472 489123 99% 0.03K 3824 128 15296K kmalloc-32
297444 271112 91% 0.19K 7082 42 56656K dentry
254400 252784 99% 0.06K 3975 64 15900K
anon_vma_chain
222528 39255 17% 0.50K 6954 32 111264K kmalloc-512
202482 201814 99% 0.19K 4821 42 38568K
vm_area_struct
200192 200192 100% 0.01K 391 512 1564K kmalloc-8
170528 169359 99% 0.25K 5329 32 42632K filp
158144 153508 97% 0.06K 2471 64 9884K kmalloc-64
149914 149365 99% 0.09K 3259 46 13036K anon_vma
146640 143123 97% 0.10K 3760 39 15040K buffer_head
130368 32791 25% 0.09K 3104 42 12416K kmalloc-96
129752 129752 100% 0.07K 2317 56 9268K Acpi-Operand
105468 105106 99% 0.04K 1034 102 4136K
selinux_inode_security
73080 73080 100% 0.13K 2436 30 9744K
kernfs_node_cache
72360 70261 97% 0.59K 1340 54 42880K inode_cache
71040 71040 100% 0.12K 2220 32 8880K
eventpoll_epi
68096 59262 87% 0.02K 266 256 1064K kmalloc-16
53652 53652 100% 0.04K 526 102 2104K pde_opener
50496 31654 62% 2.00K 3156 16 100992K
kmalloc-2048
46242 46242 100% 0.19K 1101 42 8808K cred_jar
44496 43013 96% 0.66K 927 48 29664K
proc_inode_cache
44352 44352 100% 0.06K 693 64 2772K
task_delay_info
43516 43471 99% 0.69K 946 46 30272K
sock_inode_cache
37856 27626 72% 1.00K 1183 32 37856K
kmalloc-1024
36736 36736 100% 0.07K 656 56 2624K
eventpoll_pwq
34076 31282 91% 0.57K 1217 28 19472K
radix_tree_node
33660 30528 90% 1.05K 1122 30 35904K
ext4_inode_cache
32760 30959 94% 0.19K 780 42 6240K kmalloc-192
32028 32028 100% 0.04K 314 102 1256K
ext4_extent_status
30048 30048 100% 0.25K 939 32 7512K
skbuff_head_cache
28736 28736 100% 0.06K 449 64 1796K fs_cache
24702 24702 100% 0.69K 537 46 17184K files_cache
23808 23808 100% 0.66K 496 48 15872K ovl_inode
23104 22945 99% 0.12K 722 32 2888K kmalloc-128
22724 21307 93% 0.69K 494 46 15808K
shmem_inode_cache
21472 21472 100% 0.12K 671 32 2684K seq_file
19904 19904 100% 1.00K 622 32 19904K UNIX
17340 17340 100% 1.06K 578 30 18496K mm_struct
15980 15980 100% 0.02K 94 170 376K avtab_node
14070 14070 100% 1.06K 469 30 15008K
signal_cache
13248 13248 100% 0.12K 414 32 1656K pid
12128 11777 97% 0.25K 379 32 3032K kmalloc-256
11008 11008 100% 0.02K 43 256 172K
selinux_file_security
10812 10812 100% 0.04K 106 102 424K
Acpi-Namespace
These information shows that the 'iommu_iova' is the top memory
consumer.
In order to optimize the network performence of Openstack virtual
machines,
I enabled the vt-d feature in bios and sriov feature of Intel 82599
10G
NIC. I'm assuming this is the root cause of this issue.
Is there anything I can do to fix it?
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu
