[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <363f8293-23e3-44d3-8005-b31eb5b7f975@amd.com>
Date: Tue, 25 Mar 2025 14:40:00 +0530
From: "Aithal, Srikanth" <sraithal@....com>
To: Linux-Next Mailing List <linux-next@...r.kernel.org>,
open list <linux-kernel@...r.kernel.org>, kirill.shutemov@...ux.intel.com
Cc: "Lendacky, Thomas" <Thomas.Lendacky@....com>,
"Roth, Michael" <Michael.Roth@....com>
Subject: linux-next regression: SNP Guest boot hangs with certain cpu/mem
config combination
Hello,
Starting linux-next build next-20250312, including recent build
20250324, we are seeing an issue where the SNP guest boot hangs at the
"boot smp config" step:
[ 2.294722] smp: Bringing up secondary CPUs ...
[ 2.295211] smpboot: Parallel CPU startup disabled by the platform
[ 2.309687] smpboot: x86: Booting SMP configuration:
[ 2.310214] .... node #0, CPUs: #1 #2 #3 #4 #5 #6
#7 #8 #9 #10 #11 #12 #13 #14 #15 #16 #17 #18 #19 #20 #21
#22 #23 #24 #25 #26 #27 #28 #29 #30 #31 #32 #33 #34 #35
#36 #37 #38 #39 #40 #41 #42 #43 #44 #45 #46 #47 #48 #49
#50 #51 #52 #53 #54 #55 #56 #57 #58 #59 #60 #61 #62 #63
#64 #65 #66 #67 #68 #69 #70 #71 #72 #73 #74 #75 #76 #77
#78 #79 #80 #81 #82 #83 #84 #85 #86 #87 #88 #89 #90 #91
#92 #93 #94 #95 #96 #97 #98 #99 #100 #101 #102 #103 #104 #105 #106
#107 #108 #109 #110 #111 #112 #113 #114 #115 #116 #117 #118 #119 #120
#121 #122 #123 #124 #125 #126 #127 #128 #129 #130 #131 #132 #133 #134
#135 #136 #137 #138 #139 #140 #141 #142 #143 #144 #145 #146 #147 #148
#149 #150 #151 #152 #153 #154 #155 #156 #157 #158 #159 #160 #161 #162
#163 #164 #165 #166 #167 #168 #169 #170 #171 #172 #173 #174 #175 #176
#177 #178 #179 #180 #181 #182 #183 #184 #185 #186 #187 #188 #189 #190
#191 #192 #193 #194 #195 #196 #197 #198
--> The guest hangs forever at this point.
I have observed that certain vCPU and memory combinations work, while
others do not. The VM configuration I am using does not have any NUMA nodes.
vcpus Mem SNP guest boot
<=240 19456M Boots fine
>=241,<255 19456M Hangs
1-255 2048M Boots fine
1-255 4096M Boots fine
>71 8192M Hangs
>41 6144M Hangs
When I bisected this issue, it pointed to the following commit :
*commit 800f1059c99e2b39899bdc67a7593a7bea6375d8*
Author: Kirill A. Shutemov <kirill.shutemov@...ux.intel.com>
Date: Mon Mar 10 10:28:55 2025 +0200
mm/page_alloc: fix memory accept before watermarks gets initialized
Watermarks are initialized during the postcore initcall. Until
then, all
watermarks are set to zero. This causes cond_accept_memory() to
incorrectly skip memory acceptance because a watermark of 0 is
always met.
This can lead to a premature OOM on boot.
To ensure progress, accept one MAX_ORDER page if the watermark is zero.
Link:
https://lkml.kernel.org/r/20250310082855.2587122-1-kirill.shutemov@linux.intel.com
Fixes: dcdfdd40fa82 ("mm: Add support for unaccepted memory")
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@...ux.intel.com>
Tested-by: Farrah Chen <farrah.chen@...el.com>
Reported-by: Farrah Chen <farrah.chen@...el.com>
Acked-by: Vlastimil Babka <vbabka@...e.cz>
Reviewed-by: Pankaj Gupta <pankaj.gupta@....com>
Cc: Ashish Kalra <ashish.kalra@....com>
Cc: David Hildenbrand <david@...hat.com>
Cc: "Edgecombe, Rick P" <rick.p.edgecombe@...el.com>
Cc: Mel Gorman <mgorman@...hsingularity.net>
Cc: "Mike Rapoport (IBM)" <rppt@...nel.org>
Cc: Thomas Lendacky <thomas.lendacky@....com>
Cc: <stable@...r.kernel.org> [6.5+]
Signed-off-by: Andrew Morton <akpm@...ux-foundation.org>
Reproduction Steps:
Build the guest kernel with linux-next [starting next-20250324] using
the attached guest config.
Try booting the SNP guest using the following sample QEMU command line:
qemu-system-x86_64 \
-machine q35,confidential-guest-support=sev0,memory-backend=mem-upmsev0 \
-object
memory-backend-memfd,size=19456M,prealloc=false,share=true,reserve=false,id=mem-upmsev0
\
-object
sev-snp-guest,id=sev0,cbitpos=51,reduced-phys-bits=1,kernel-hashes=off,policy=0x30000
\
-name guest=vm,debug-threads=on \
-bios OVMF_X64/OVMF.fd \
-cpu EPYC-v4 \
-smp 255,maxcpus=255,cores=255,threads=1,dies=1,sockets=1 \
-kernel bzImage_0321 \
-append "root=/dev/vda rw console=ttyS0,115200n8
earlyprintk=ttyS0,115200 net.ifnames=0 biosdevname=0 movable_node
swiotlb=65536 " \
-device
virtio-blk-pci,id=blk0,drive=drive_image1,disable-legacy=on,iommu_platform=true
\
-blockdev
node-name=file_image1,driver=file,filename=/home/VT_BUILD/22.04-server.qcow2
\
-blockdev node-name=drive_image1,driver=qcow2,file=file_image1 \
--enable-kvm \
--nographic \
-qmp tcp:localhost:4444,server,wait=off
*Issue goes away if I build the guest kernel by reverting the
aforementioned commit or boot the guest with accept_memory=eager.*
Note that once you revert aforementioned commit or boot the guest with
accept_memory=eager the guest boots and ends up with another issue which
is already reported [2].
[1] https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git,
next-20250324
[2] Re: [syzbot] [pci?] linux-next test error: general protection fault
in msix_capability_init - Aithal, Srikanth
<https://lore.kernel.org/all/423b87d8-2ae3-48af-b368-657f1fbab88d@amd.com/>
Content of type "text/html" skipped
View attachment "guest_config" of type "text/plain" (154820 bytes)
Powered by blists - more mailing lists