[Drbd-dev] size zero read from upper layers warning in 9.0.26-rc4

[Michael D Labriola]

Hmm.  I'm testing on Linux 5.10.1, along with the latest ZFS which
you'll see in the following WARNING I'm getting.  I seem to recall
seing a commit about zero-size reads w/ ZFS a few days ago while
perusing the drbd commit logs.  What I don't remember is if I should
be worried.

Was the WARNING we stuck in the code warning me, the user, whose data
might got kabloom, or a warning I should pass on to the ZFS folks
because they're doing something wrong that they should fix?  Or
perhaps both?

Thanks!

[  365.901499] ------------[ cut here ]------------
[  365.901502] size zero read from upper layers
[  365.901528] WARNING: CPU: 2 PID: 1794 at drbd_submit_bio+0xe6/0x100 [drbd]
[  365.901529] Modules linked in: wireguard curve25519_x86_64 libcurve25519_generic libchacha20
poly1305 chacha_x86_64 libchacha poly1305_x86_64 libblake2s blake2s_x86_64 libblake2s_generic d
rbd_transport_tcp drbd lru_cache rfkill xen_netfront crc32c_intel aesni_intel glue_helper crypt
o_simd cryptd parport_pc ppdev lp parport sunrpc autofs4 xen_blkfront loop
[  365.901553] CPU: 2 PID: 1794 Comm: txg_sync Not tainted 5.10.1-mdl+ #7
[  365.901562] RIP: 0010:drbd_submit_bio+0xe6/0x100 [drbd]
[  365.901565] Code: c0 99 9a c1 48 89 df e8 88 f3 ff ff 48 83 c4 08 b8 ff ff ff ff 5b c3 48 c7
 c7 a0 ce 2a c0 c6 05 09 7f 05 00 01 e8 2a 5f 22 c0 <0f> 0b 48 8b 34 24 e9 64 ff ff ff 0f 1f 44
 00 00 66 2e 0f 1f 84 00
[  365.901567] RSP: 0018:ffffc90001e27aa8 EFLAGS: 00010282
[  365.901570] RAX: 0000000000000000 RBX: ffff888005497000 RCX: 0000000000000001
[  365.901572] RDX: 0000000080000001 RSI: ffffffff818b1556 RDI: 00000000ffffffff
[  365.901573] RBP: ffffc90001e27b20 R08: 0000000000000000 R09: 0000000000000001
[  365.901574] R10: ffffc90001e27b40 R11: ffffc90001e278c8 R12: 00000000ffffffff
[  365.901575] R13: ffff8880051cdca0 R14: ffff888005494800 R15: ffff88801ff9f440
[  365.901630] FS:  0000000000000000(0000) GS:ffff888103d00000(0000) knlGS:0000000000000000
[  365.901632] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  365.901633] CR2: 0000556f4bb8a038 CR3: 0000000004f7a000 CR4: 00000000003506e0
[  365.901636] Call Trace:
[  365.901643]  submit_bio_noacct+0x14c/0x410
[  365.901647]  ? submit_bio+0x40/0x130
[  365.901650]  submit_bio+0x40/0x130
[  365.901654]  vdev_disk_io_start+0x946/0x9f0
[  365.901658]  ? kmem_cache_alloc+0x1c/0x140
[  365.901661]  ? preempt_count_add+0x63/0x90
[  365.901665]  ? _raw_spin_lock+0xe/0x30
[  365.901668]  ? zio_add_child+0x13f/0x160
[  365.901671]  ? taskq_init_ent+0x33/0x70
[  365.901674]  ? zio_create+0x405/0x4b0
[  365.901677]  ? zio_vdev_io_start+0x120/0x2d0
[  365.901679]  ? zio_ready+0x27d/0x450
[  365.901681]  zio_vdev_io_start+0x120/0x2d0
[  365.901685]  zio_nowait+0x9e/0x160
[  365.901688]  zio_ioctl+0x7c/0xe0
[  365.901691]  zio_flush+0x1f/0x30
[  365.901695]  vdev_config_sync+0xcc/0x230
[  365.901699]  spa_sync+0xbba/0xf70
[  365.901703]  txg_sync_thread+0x281/0x410
[  365.901707]  ? txg_quiesce_thread+0x380/0x380
[  365.901709]  ? __thread_exit+0x10/0x10
[  365.901712]  thread_generic_wrapper+0x6a/0x80
[  365.901714]  kthread+0x134/0x160
[  365.901717]  ? kthread_park+0x80/0x80
[  365.901720]  ret_from_fork+0x22/0x30
[  365.901724] ---[ end trace 628d60b3830d674a ]---
[  368.576735] EXT4-fs (zd0): mounted filesystem with ordered data mode. Opts: (null)


-- 
Michael D Labriola
21 Rip Van Winkle Cir
Warwick, RI 02886
401-316-9844 (cell)