Bug 208885

Summary:	Running sudo lcov --zerocounters on a gcov enabled kernel causes write to read-only error on kernel in a VM
Product:	Platform Specific/Hardware	Reporter:	Colin Ian King (colin.king)
Component:	ARM	Assignee:	linux-arm-kernel (linux-arm-kernel)
Status:	NEW ---
Severity:	blocking
Priority:	P1
Hardware:	ARM
OS:	Linux
Kernel Version:	5.8	Subsystem:
Regression:	No	Bisected commit-id:

Description Colin Ian King 2020-08-12 09:10:41 UTC

5.8 kernel running on a QEMU KVM Virtual Machine, arm64, gcov enabled:

CONFIG_GCOV_KERNEL=y
CONFIG_ARCH_HAS_GCOV_PROFILE_ALL=y
CONFIG_GCOV_PROFILE_ALL=y

running: sudo lcov --zerocounters

..causes the following splat:


[   78.634926] Unable to handle kernel write to read-only memory at virtual address ffffd764f255bb10
[   78.639076] Mem abort info:
[   78.640313]   ESR = 0x9600004e
[   78.641656]   EC = 0x25: DABT (current EL), IL = 32 bits
[   78.644351]   SET = 0, FnV = 0
[   78.645766]   EA = 0, S1PTW = 0
[   78.649526] Data abort info:
[   78.654341]   ISV = 0, ISS = 0x0000004e
[   78.656851]   CM = 0, WnR = 1
[   78.659043] swapper pgtable: 4k pages, 48-bit VAs, pgdp=00000000b775f000
[   78.665822] [ffffd764f255bb10] pgd=00000001b6fff003, p4d=00000001b6fff003, pud=00000001b6ffe003, pmd=00400000b5000f91
[   78.677367] Internal error: Oops: 9600004e [#1] SMP
[   78.681204] Modules linked in: nls_iso8859_1 dm_multipath scsi_dh_rdac scsi_dh_emc scsi_dh_alua efi_pstore qemu_fw_cfg sch_fq_codel drm virtio_rng ip_t
ables x_tables autofs4 btrfs blake2b_generic raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor xor_neon raid6_pq libcrc32c rai
d1 raid0 multipath linear crct10dif_ce ghash_ce sha2_ce sha256_arm64 sha1_ce virtio_net xhci_pci net_failover virtio_blk virtio_scsi failover xhci_pci_ren
esas aes_neon_bs aes_neon_blk aes_ce_blk crypto_simd cryptd aes_ce_cipher
[   78.722642] CPU: 14 PID: 1175 Comm: lcov Tainted: G        W         5.8.0-8-generic #9
[   78.729364] Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015
[   78.733467] pstate: 20400005 (nzCv daif +PAN -UAO BTYPE=--)
[   78.736219] pc : __memset+0x148/0x188
[   78.737993] lr : gcov_info_reset+0xa0/0x110
[   78.740037] sp : ffff800010e53bf0
[   78.741624] x29: ffff800010e53bf0 x28: ffff00015a648000 
[   78.744181] x27: 0000000000000000 x26: ffffd764e3df4060 
[   78.746776] x25: 0000000000000001 x24: ffffd764e3df4040 
[   78.749310] x23: 0000000000000000 x22: ffffd764e3df40a8 
[   78.751976] x21: ffffd764f6295db8 x20: ffffd764e3df4068 
[   78.754642] x19: ffffd764e3df4068 x18: 0000000000000000 
[   78.757265] x17: 0000000000000000 x16: 0000000000000000 
[   78.759922] x15: 0000000000000000 x14: 0000000000000000 
[   78.762585] x13: 0000000000000000 x12: 0000000000000000 
[   78.765209] x11: 0000000000000000 x10: 0000000000000000 
[   78.767861] x9 : 0000000000000000 x8 : ffffd764f255bb10 
[   78.770518] x7 : 0000000000000000 x6 : 000000000000003f 
[   78.773135] x5 : 0000000000000040 x4 : 0000000000000030 
[   78.775793] x3 : 000000071ef97fd0 x2 : 000000071ef97fd0 
[   78.778449] x1 : 0000000000000000 x0 : ffffd764f255bb10 
[   78.781075] Call trace:
[   78.782329]  __memset+0x148/0x188
[   78.783980]  reset_node+0x70/0xc0
[   78.785628]  reset_write+0xc0/0x130
[   78.787397]  full_proxy_write+0x98/0x110
[   78.789354]  vfs_write+0x18c/0x3e0
[   78.791086]  ksys_write+0xc4/0x1f0
[   78.792778]  __arm64_sys_write+0x44/0x60
[   78.794771]  invoke_syscall+0x74/0x150
[   78.796639]  el0_svc_common.constprop.0+0x94/0x264
[   78.799049]  do_el0_svc+0x5c/0x7c
[   78.800702]  el0_svc+0x70/0x114
[   78.802311]  el0_sync_handler+0x140/0x39c
[   78.804304]  el0_sync+0x17c/0x180
[   78.805967] Code: f101007f fa45a068 54fffc0b aa0303e2 (a9001d07) 
[   78.809019] ---[ end trace b9a826b707c133e4 ]---

Comment 1 Colin Ian King 2020-08-12 09:17:59 UTC

Just spotted that earlier in the boot we get:

[    8.808508] virtio_blk virtio3: [vda] 209715200 512-byte logical blocks (107 GB/100 GiB)
[    8.808649] ------------[ cut here ]------------
[    8.815777] vda: detected capacity change from 0 to 107374182400
[    8.816623] WARNING: CPU: 3 PID: 212 at mm/page_alloc.c:4813 __alloc_pages_nodemask+0x540/0x694
[    8.816630] Modules linked in: net_failover virtio_blk(+) xhci_pci_renesas virtio_scsi(+) failover fjes(-) aes_neon_bs aes_neon_blk aes_ce_blk crypto_s
imd cryptd aes_ce_cipher
[    8.837701] CPU: 3 PID: 212 Comm: systemd-udevd Not tainted 5.8.0-8-generic #9
[    8.842236] Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015
[    8.846585] pstate: 20400005 (nzCv daif +PAN -UAO BTYPE=--)
[    8.850112] pc : __alloc_pages_nodemask+0x540/0x694
[    8.853106] lr : alloc_pages_current+0xe0/0x17c
[    8.856024] sp : ffff8000106138c0
[    8.858175] x29: ffff8000106138c0 x28: ffff000170252dc0 
[    8.861565] x27: 0000000000000000 x26: 0000000000000015 
[    8.865006] x25: 00000001abcfef80 x24: 0000000000000015 
[    8.868466] x23: ffff000170252dc0 x22: 0000000000000000 
[    8.871794] x21: ffffdb2796703050 x20: 0000000000040dc0 
[    8.875143] x19: ffffdb2796703250 x18: 0000000000000000 
[    8.878529] x17: 0000000000000000 x16: 0000000000000000 
[    8.881917] x15: 0000000000000000 x14: 73656a662f74656e 
[    8.885218] x13: 2f73726576697264 x12: 2f636972656e6567 
[    8.888554] x11: 2d646c6975622f64 x10: 6c6975622f6e6169 
[    8.892032] x9 : ffffdb2792b6dfa4 x8 : 63672e7366677562 
[    8.895415] x7 : 65645f73656a662f x6 : ffff0001558ef353 
[    8.898814] x5 : 0000000000000018 x4 : 0000000000000dd5 
[    8.902124] x3 : ffffdb27966ea138 x2 : 0000000000000001 
[    8.905418] x1 : ffffdb27966ea030 x0 : 0000000000000001 
[    8.908698] Call trace:
[    8.910278]  __alloc_pages_nodemask+0x540/0x694
[    8.913278]  alloc_pages_current+0xe0/0x17c
[    8.915967]  kmalloc_order+0x48/0xec
[    8.918325]  kmalloc_order_trace+0x54/0x21c
[    8.921165]  __kmalloc+0x454/0x5c0
[    8.923368]  kmalloc_array.constprop.0+0x30/0x50
[    8.926301]  gcov_info_dup+0x98/0x2c0
[    8.928557]  remove_info+0x1e0/0x244
[    8.930876]  gcov_event+0xfc/0x170
[    8.933097]  gcov_module_notifier+0x164/0x174
[    8.935848]  notifier_call_chain+0xa0/0x110
[    8.938435]  blocking_notifier_call_chain+0x80/0xe0
[    8.941579]  do_init_module+0x360/0x43c
[    8.944198]  load_module+0x8fc/0xa50
[    8.946591]  __do_sys_finit_module+0x190/0x1c0
[    8.949485]  __arm64_sys_finit_module+0x48/0x64
[    8.952376]  invoke_syscall+0x74/0x150
[    8.954901]  el0_svc_common.constprop.0+0x94/0x264
[    8.956405] sr 0:0:0:0: Power-on or device reset occurred
[    8.957610]  do_el0_svc+0x5c/0x7c
[    8.960983] sr 0:0:0:0: [sr0] scsi3-mmc drive: 16x/50x cd/rw xa/form2 cdda tray
[    8.962169]  el0_svc+0x70/0x114
[    8.965694] cdrom: Uniform CD-ROM driver Revision: 3.20
[    8.967253]  el0_sync_handler+0x140/0x39c
[    8.971738]  el0_sync+0x17c/0x180
[    8.973450] ---[ end trace 267579295e47a82a ]---
[    8.976042] gcov: could not save data for '/home/cking/unstable/debian/build/build-generic/drivers/net/fjes/fjes_debugfs.gcda' (out of memory)
[    8.986263] gcov: could not save data for '/home/cking/unstable/debian/build/build-generic/drivers/net/fjes/fjes_trace.gcda' (out of memory)
[    8.996162] gcov: could not save data for '/home/cking/unstable/debian/build/build-generic/drivers/net/fjes/fjes_ethtool.gcda' (out of memory)
[    9.006231] gcov: could not save data for '/home/cking/unstable/debian/build/build-generic/drivers/net/fjes/fjes_hw.gcda' (out of memory)
[    9.010491] sr 0:0:0:0: Attached scsi CD-ROM sr0
[    9.013246] sr 0:0:0:0: Attached scsi generic sg0 type 5
[    9.014959] gcov: could not save data for '/home/cking/unstable/debian/build/build-generic/drivers/net/fjes/fjes_main.gcda' (out of memory)

Comment 2 Colin Ian King 2020-08-12 09:26:34 UTC

I upped the VM memory to 16GB and reduced the CPU count to 1 CPU and still hit the out of memory condition with gcov.