On Tue, 9 Sep 2025 10:17:17 +0800 Changhui Zhong <czhong@xxxxxxxxxx> wrote: > I still see this issue on v6.17.0-rc5, please take a look if you have a chance Well it might be related to zram, although we haven't done much at all with zram or zsmalloc in this cycle. > [ 6915.287305] BUG zs_handle-zram0 (Tainted: G S ): > Objects remaining on __kmem_cache_shutdown() > [ 6915.298379] ----------------------------------------------------------------------------- > [ 6915.298379] > [ 6915.309166] Object 0x0000000023ce3ee6 @offset=960 > [ 6915.314421] Slab 0x00000000913eaa62 objects=128 used=1 > fp=0x0000000064bf6df4 > flags=0x17ffffc0000200(workingset|node=0|zone=2|lastcpupid=0x1fffff) > [ 6915.329002] ------------[ cut here ]------------ > [ 6915.334159] WARNING: CPU: 2 PID: 7198 at mm/slub.c:1176 > __slab_err.part.0+0x19/0x20 > [ 6915.342713] Modules linked in: zram 842_decompress lz4hc_compress > 842_compress lz4_compress zstd_compress tls rpcsec_gss_krb5 > auth_rpcgss nfsv4 dns_resolver nfs lockd grace nfs_localio netfs > intel_rapl_msr intel_rapl_common intel_uncore_frequency > intel_uncore_frequency_common i10nm_edac skx_edac_common nfit > libnvdimm x86_pkg_temp_thermal intel_powerclamp rfkill coretemp > kvm_intel kvm dax_hmem cxl_acpi cxl_port irqbypass cxl_core iTCO_wdt > cdc_ether rapl iTCO_vendor_support usbnet intel_cstate mii > intel_uncore mgag200 intel_th_gth isst_if_mbox_pci einj i2c_i801 > i2c_algo_bit isst_if_mmio pcspkr mei_me isst_if_common intel_th_pci > ioatdma i2c_smbus mei intel_vsec acpi_power_meter ipmi_ssif > intel_pch_thermal intel_th dca ipmi_si acpi_ipmi ipmi_devintf > ipmi_msghandler acpi_pad sg fuse loop xfs sd_mod ghash_clmulni_intel > ahci libahci tg3 libata wmi sunrpc dm_mirror dm_region_hash dm_log > dm_multipath dm_mod nfnetlink [last unloaded: brd] > [ 6915.434755] CPU: 2 UID: 0 PID: 7198 Comm: bash Kdump: loaded > Tainted: G S B 6.17.0-rc5 #1 PREEMPT(voluntary) > [ 6915.447281] Tainted: [S]=CPU_OUT_OF_SPEC, [B]=BAD_PAGE > [ 6915.453016] Hardware name: Lenovo ThinkSystem SR650 V2/7Z73CTO1WW, > BIOS AFE118M-1.32 06/29/2022 > [ 6915.462728] RIP: 0010:__slab_err.part.0+0x19/0x20 > [ 6915.467981] Code: 00 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 > 90 0f 1f 44 00 00 e8 f6 fc ff ff be 01 00 00 00 bf 05 00 00 00 e8 77 > bf 29 00 <0f> 0b c3 cc cc cc cc 48 89 ee 48 c7 c7 eb a2 86 ab c6 05 10 > 13 c4 > [ 6915.488939] RSP: 0018:ffa0000002fcf918 EFLAGS: 00010046 > [ 6915.494773] RAX: 0000000000000000 RBX: ffd400000b6c3b80 RCX: ffffffffa80d3a9a > [ 6915.502740] RDX: 0000000000000001 RSI: 0000000000000008 RDI: ffffffffadd3f180 > [ 6915.510706] RBP: ff11000149c51140 R08: 0000000000000001 R09: fffffbfff5ba7e30 > [ 6915.518672] R10: ffffffffadd3f187 R11: 0000000062616c53 R12: ff11000285ff0000 > [ 6915.526640] R13: ff11000285ff1000 R14: ff11000149c4f040 R15: ffd400000a17fc00 > [ 6915.534605] FS: 00007f9aa117a740(0000) GS:ff1100043db3d000(0000) > knlGS:0000000000000000 > [ 6915.543637] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 > [ 6915.550051] CR2: 000056332051e698 CR3: 00000001485c1002 CR4: 0000000000773ef0 > [ 6915.558017] PKRU: 55555554 > [ 6915.561037] Call Trace: > [ 6915.563766] <TASK> > [ 6915.566108] __kmem_cache_shutdown.cold+0x1c/0x43 > [ 6915.571364] kmem_cache_destroy+0x68/0x170 > [ 6915.575938] zs_destroy_pool+0x18b/0x250 > [ 6915.580319] zram_reset_device+0x233/0x600 [zram] > [ 6915.585576] reset_store+0x20b/0x300 [zram] > [ 6915.590252] ? __pfx_reset_store+0x10/0x10 [zram] > [ 6915.595506] ? sysfs_file_kobj+0xb3/0x1c0 > [ 6915.599982] ? sysfs_file_kobj+0xbd/0x1c0 > [ 6915.604449] ? __pfx_sysfs_kf_write+0x10/0x10 > [ 6915.609315] kernfs_fop_write_iter+0x3a3/0x5a0 > [ 6915.614278] vfs_write+0x522/0xfd0 > [ 6915.618078] ? __pfx_vfs_write+0x10/0x10 > [ 6915.622459] ? local_clock_noinstr+0xd/0xe0 > [ 6915.627130] ? __lock_release.isra.0+0x1a4/0x2c0 > [ 6915.632287] ksys_write+0xf9/0x1d0 > [ 6915.636086] ? __pfx_ksys_write+0x10/0x10 > [ 6915.640561] ? lockdep_hardirqs_on+0x78/0x100 > [ 6915.645426] do_syscall_64+0x94/0x8d0 > [ 6915.649515] ? ktime_get_coarse_real_ts64+0x121/0x180 > [ 6915.655155] ? lockdep_hardirqs_on+0x78/0x100 > [ 6915.660022] ? entry_SYSCALL_64_after_hwframe+0x76/0x7e > [ 6915.665855] ? lockdep_hardirqs_on+0x78/0x100 > [ 6915.670719] ? do_syscall_64+0x139/0x8d0 > [ 6915.675089] ? __x64_sys_openat+0x108/0x1d0 > [ 6915.679762] ? __pfx___x64_sys_openat+0x10/0x10 > [ 6915.684819] ? lockdep_hardirqs_on+0x78/0x100 > [ 6915.689684] ? syscall_trace_enter+0x13e/0x240 > [ 6915.694648] ? entry_SYSCALL_64_after_hwframe+0x76/0x7e > [ 6915.700481] ? lockdep_hardirqs_on+0x78/0x100 > [ 6915.705347] ? do_syscall_64+0x139/0x8d0 > [ 6915.709726] ? rcu_is_watching+0x15/0xb0 > [ 6915.714106] ? clear_bhb_loop+0x50/0xa0 > > ... > > > > reproducer: > > > # modprobe zram > > > # zramctl --find --size 4G --algorithm zstd > > > # fio --name=test \ > > > --filename=/dev/zram0 \ > > > --rw=randrw \ > > > --bs=4k \ > > > --ioengine=libaio \ > > > --iodepth=16 \ > > > --numjobs=4 \ > > > --runtime=60 \ > > > --time_based \ > > > --group_reporting \ > > > --direct=1 > > > # echo 1 > /sys/block/zram0/reset > > > > > > dmesg log: > > > [ 4861.143371] zsmalloc: Class-80 fullness group 1 is not empty > > > [ 4861.149696] zsmalloc: Class-112 fullness group 1 is not empty > > > [ 4861.156121] zsmalloc: Class-144 fullness group 1 is not empty > > > [ 4861.162541] zsmalloc: Class-160 fullness group 1 is not empty > > > [ 4861.168963] zsmalloc: Class-176 fullness group 1 is not empty > > > [ 4861.175379] zsmalloc: Class-192 fullness group 1 is not empty > > > [ 4861.181797] zsmalloc: Class-224 fullness group 1 is not empty Minchan & Sergey, could you please give this a try?
