From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============1399901089514739395==" MIME-Version: 1.0 From: Ming Lei To: lkp@lists.01.org Subject: Re: 07173c3ec2: stderr.fio:pid=#, err=#/file:filesetup.c:#, func=fsync, error=Input/output_error Date: Mon, 01 Jul 2019 15:14:11 +0800 Message-ID: <20190701071409.GD16809@ming.t460p> In-Reply-To: <14cf3658-21d5-a8fe-982e-60776476eb75@cn.fujitsu.com> List-Id: --===============1399901089514739395== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable On Mon, Jul 01, 2019 at 02:54:31PM +0800, Liu Yiding wrote: > = > =E5=9C=A8 2019/7/1 =E4=B8=8B=E5=8D=882:44, Ming Lei =E5=86=99=E9=81=93: > > On Mon, Jul 01, 2019 at 12:49:51PM +0800, Liu Yiding wrote: > > > =E5=9C=A8 2019/7/1 =E4=B8=8A=E5=8D=8810:47, Ming Lei =E5=86=99=E9=81= =93: > > > > [snip] > > > > > > > ------ > > > > > > > = > > > > > > > 2412 [=C2=A0 112.558155] XFS (sdb1): Ending clean mount > > > > > > > 2413 [ 1317.330820] WARNING: CPU: 63 PID: 1495 at > > > > > > > drivers/scsi/scsi_lib.c:1034 scsi_init_io+0x11a/0x2e0 > > > > > > > 2414 [ 1317.344000] Modules linked in: xfs dm_mod btrfs xor z= std_decompress > > > > > > > zstd_compress raid6_pq libcrc32c sd_mod sg intel_rapl sb_edac > > > > > > > x86_pkg_temp_thermal intel_powerclamp coretemp kvm_i=C2=A0=C2= =A0=C2=A0=C2=A0 ntel kvm irqbypass > > > > > > > crct10dif_pclmul crc32_pclmul crc32c_intel mgag200 snd_pcm > > > > > > > ghash_clmulni_intel ttm snd_timer drm_kms_helper ipmi_si ahci= snd > > > > > > > aesni_intel syscopyarea libahci ipmi_devi=C2=A0=C2=A0=C2=A0= =C2=A0 ntf sysfillrect sysimgblt > > > > > > > crypto_simd mei_me fb_sys_fops soundcore ioatdma cryptd ipmi_= msghandler > > > > > > > glue_helper pcspkr drm libata mei joydev lpc_ich i2c_i801 dca= wmi > > > > > > > acpi_power_meter acpi=C2=A0=C2=A0=C2=A0=C2=A0 _pad pcc_cpufre= q ip_tables > > > > > > > 2415 [ 1317.408135] CPU: 63 PID: 1495 Comm: kworker/63:1H Tai= nted: > > > > > > > G=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 I=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 5.2.0-rc5-00001-gc2d55e2 #1 > > > > > > > 2416 [ 1317.420350] Hardware name: Intel Corporation S2600WTT= /S2600WTT, BIOS > > > > > > > SE5C610.86B.01.01.0008.021120151325 02/11/2015 > > > > > > > 2417 [ 1317.433545] Workqueue: kblockd blk_mq_run_work_fn > > > > > > > 2418 [ 1317.440295] RIP: 0010:scsi_init_io+0x11a/0x2e0 > > > > > > > 2419 [ 1317.446746] Code: be 01 00 00 00 e9 44 ff ff ff 8b 45= 70 89 83 08 01 > > > > > > > 00 00 f6 45 1a 01 75 8a 31 c0 48 83 c4 08 5b 5d 41 5c 41 5d 4= 1 5e 41 5f c3 > > > > > > > <0f> 0b 8b 4d 18 8b 55 28 45 31 c9 48=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0 c7 c6 98 da 69 82 48 c7 c7 f0 > > > > > > > 2420 [ 1317.470863] RSP: 0018:ffffa5fa4f7e7c68 EFLAGS: 000102= 46 > > > > > > > 2421 [ 1317.478294] RAX: 000000003d17d1b8 RBX: ffff8ae6cb7426= 90 RCX: > > > > > > > 0000000000000000 > > > > > > > 2422 [ 1317.487878] RDX: 0000000000000000 RSI: 00000000000000= 00 RDI: > > > > > > > ffff8ae6cb742690 > > > > > > > 2423 [ 1317.497469] RBP: ffff8ae6cb742580 R08: 00000000000230= 03 R09: > > > > > > > ffff8ae6cb7426c8 > > > > > > > 2424 [ 1317.507057] R10: ffffa5fa4c7d7da0 R11: fefefefefefefe= ff R12: > > > > > > > 0000000000000801 > > > > > > > 2425 [ 1317.516650] R13: 0000000000000000 R14: 00000000000000= 00 R15: > > > > > > > ffff8ac800bd1800 > > > > > > > 2426 [ 1317.526253] FS:=C2=A0 0000000000000000(0000) GS:ffff8= b077fac0000(0000) > > > > > > > knlGS:0000000000000000 > > > > > > > 2427 [ 1317.536918] CS:=C2=A0 0010 DS: 0000 ES: 0000 CR0: 000= 0000080050033 > > > > > > > 2428 [ 1317.544952] CR2: 00007f39a5818004 CR3: 000000343440a0= 06 CR4: > > > > > > > 00000000001606e0 > > > > > > > 2429 [ 1317.554554] Call Trace: > > > > > > > 2430 [ 1317.558921]=C2=A0 sd_init_command+0x2ee/0xaf0 [sd_mod] > > > > > > > 2431 [ 1317.565800]=C2=A0 scsi_queue_rq+0x559/0xa80 > > > > > > > 2432 [ 1317.571607]=C2=A0 blk_mq_dispatch_rq_list+0x3da/0x5b0 > > > > > > > 2433 [ 1317.578375]=C2=A0 ? syscall_return_via_sysret+0x10/0x= 7f > > > > > > > 2434 [ 1317.585339]=C2=A0 ? __switch_to_asm+0x40/0x70 > > > > > > > 2435 [ 1317.591321]=C2=A0 ? elv_rb_del+0x1f/0x30 > > > > > > > 2436 [ 1317.596807]=C2=A0 ? deadline_remove_request+0x55/0xc0 > > > > > > > 2437 [ 1317.603560]=C2=A0 blk_mq_do_dispatch_sched+0x76/0x120 > > > > > > > 2438 [ 1317.610302]=C2=A0 blk_mq_sched_dispatch_requests+0x10= 0/0x170 > > > > > > > 2439 [ 1317.617733]=C2=A0 __blk_mq_run_hw_queue+0x60/0x130 > > > > > > > 2440 [ 1317.624200]=C2=A0 process_one_work+0x19c/0x3c0 > > > > > > > 2441 [ 1317.630258]=C2=A0 worker_thread+0x3c/0x3b0 > > > > > > > 2442 [ 1317.635937]=C2=A0 ? process_one_work+0x3c0/0x3c0 > > > > > > > 2443 [ 1317.642183]=C2=A0 kthread+0x11e/0x140 > > > > > > > 2444 [ 1317.647346]=C2=A0 ? kthread_park+0x90/0x90 > > > > > > > 2445 [ 1317.652986]=C2=A0 ret_from_fork+0x35/0x40 > > > > > > > 2446 [ 1317.658523] ---[ end trace 61026ff97261da6a ]--- > > > > > > > 2447 [ 1317.665327] scsi_dump_rq: rq 6(f:801/23003, seg: 0) > > > > > > > 2448 [ 1317.671601] bio 0: d602/801 0 1024971192 0 > > > > > > > 2449 [ 1317.676991] print_req_error: I/O error, dev sdb, sect= or 1024971192 > > > > > > > flags 801 > > > > > > > 2450 [ 1317.685722] scsi_dump_rq: rq 8(f:801/23003, seg: 0) > > > > > > > 2451 [ 1317.691994] bio 0: d602/801 0 1033359800 0 > > > > > > The above line shows the bio is zero-size, and this bio is subm= itted > > > > > > from bio, since the bio->bi_flags indicates that BIO_CLONED isn= 't set. > > > > > > = > > > > > > Please apply the following debug patch, and we may see where is= wrong. > > > > > > = > > > > > > diff --git a/block/blk-core.c b/block/blk-core.c > > > > > > index 8340f69670d8..86bb61b0df83 100644 > > > > > > --- a/block/blk-core.c > > > > > > +++ b/block/blk-core.c > > > > > > @@ -1125,6 +1125,7 @@ blk_qc_t submit_bio(struct bio *bio) > > > > > > bio_devname(bio, b), count); > > > > > > } > > > > > > } > > > > > > + WARN_ON(bio->bi_iter.bi_size =3D=3D 0 && bio_op(bio) =3D=3D R= EQ_OP_WRITE); > > > > > > return generic_make_request(bio); > > > > > Sorry for delay, > > > > > = > > > > > ------ > > > > > = > > > > > [=C2=A0 116.619136] 2019-06-29 16:12:41 mkdir -p /fs/sdb1 > > > > > [=C2=A0 116.619138] > > > > > [=C2=A0 116.628743] 2019-06-29 16:12:41 mount -t xfs -o inode64 /= dev/sdb1 > > > > > /fs/sdb1 > > > > > [=C2=A0 116.628745] > > > > > [=C2=A0 116.634699] SGI XFS with ACLs, security attributes, realt= ime, scrub, > > > > > debug enabled > > > > > [=C2=A0 116.653110] XFS (sdb1): Mounting V5 Filesystem > > > > > [=C2=A0 116.670081] XFS (sdb1): Ending clean mount > > > > > [=C2=A0 690.106779] WARNING: CPU: 53 PID: 1477 at block/blk-core.= c:1128 > > > > > submit_bio+0x93/0x150 > > > > > [=C2=A0 690.117972] Modules linked in: xfs dm_mod btrfs xor zstd_= decompress > > > > > zstd_compress raid6_pq libcrc32c sd_mod sg intel_rapl sb_edac > > > > > x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm irqb= ypass > > > > > mgag200 crct10dif_pclmul crc32_pclmul ttm crc32c_intel snd_pcm > > > > > ghash_clmulni_intel drm_kms_helper snd_timer ahci syscopyarea sys= fillrect > > > > > snd aesni_intel sysimgblt libahci ipmi_si soundcore fb_sys_fops m= ei_me > > > > > crypto_simd ipmi_devintf ioatdma cryptd pcspkr drm ipmi_msghandler > > > > > glue_helper libata mei joydev lpc_ich i2c_i801 dca wmi acpi_power= _meter > > > > > acpi_pad pcc_cpufreq ip_tables > > > > > [=C2=A0 690.176967] CPU: 53 PID: 1477 Comm: fio Tainted: G=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 W I > > > > > 5.2.0-rc5-00002-g9fa09bd #1 > > > > > [=C2=A0 690.187198] Hardware name: Intel Corporation S2600WTT/S26= 00WTT, BIOS > > > > > SE5C610.86B.01.01.0008.021120151325 02/11/2015 > > > > > [=C2=A0 690.199375] RIP: 0010:submit_bio+0x93/0x150 > > > > > [=C2=A0 690.204583] Code: ee 74 45 83 e1 01 74 26 65 48 01 35 bf = 39 3f 45 8b 0d > > > > > ad 96 c4 01 85 c9 75 56 8b 43 28 85 c0 75 b3 0f b6 53 10 83 fa 01= 75 aa <0f> > > > > > 0b eb a6 65 48 8b 14 25 c0 6b 01 00 65 48 01 35 88 39 3f 45 48 > > > > > [=C2=A0 690.226668] RSP: 0018:ffffb2d90efc3b40 EFLAGS: 00010246 > > > > > [=C2=A0 690.233072] RAX: 0000000000000000 RBX: ffff9654c6ad9f40 R= CX: > > > > > 0000000000000801 > > > > > [=C2=A0 690.241617] RDX: 0000000000000001 RSI: ffff9654c6ad9f00 R= DI: > > > > > ffff9654c6ad9f40 > > > > > [=C2=A0 690.250146] RBP: 0000000000000000 R08: 0000000000000002 R= 09: > > > > > 0000000000000000 > > > > > [=C2=A0 690.258667] R10: 00000000000303a0 R11: ffff96757ffd4000 R= 12: > > > > > ffffb2d90efc3e08 > > > > > [=C2=A0 690.267183] R13: ffff9654c6ad9f00 R14: ffffb2d90efc3e08 R= 15: > > > > > 0000000000000000 > > > > > [=C2=A0 690.275696] FS:=C2=A0 00007fa8e005da80(0000) GS:ffff9654f= fac0000(0000) > > > > > knlGS:0000000000000000 > > > > > [=C2=A0 690.285282] CS:=C2=A0 0010 DS: 0000 ES: 0000 CR0: 0000000= 080050033 > > > > > [=C2=A0 690.292252] CR2: 00007fa8cc05b3c0 CR3: 0000001fc8518001 C= R4: > > > > > 00000000001606e0 > > > > > [=C2=A0 690.300770] Call Trace: > > > > > [=C2=A0 690.304123]=C2=A0 xfs_submit_ioend+0x84/0x1e0 [xfs] > > > > > [=C2=A0 690.309668]=C2=A0 xfs_do_writepage+0x246/0x5a0 [xfs] > > > > > [=C2=A0 690.315279]=C2=A0 write_cache_pages+0x16b/0x440 > > > > > [=C2=A0 690.320420]=C2=A0 ? xfs_vm_releasepage+0x80/0x80 [xfs] > > > > > [=C2=A0 690.326238]=C2=A0 xfs_vm_writepages+0x6b/0xa0 [xfs] > > > > > [=C2=A0 690.331742]=C2=A0 do_writepages+0x4b/0xf0 > > > > > [=C2=A0 690.336275]=C2=A0 ? iomap_file_buffered_write+0x62/0x90 > > > > > [=C2=A0 690.342162]=C2=A0 ? iomap_write_end+0x240/0x240 > > > > > [=C2=A0 690.347295]=C2=A0 ? xfs_iunlock+0x165/0x220 [xfs] > > > > > [=C2=A0 690.352595]=C2=A0 ? __filemap_fdatawrite_range+0xc1/0x100 > > > > > [=C2=A0 690.358673]=C2=A0 __filemap_fdatawrite_range+0xc1/0x100 > > > > > [=C2=A0 690.364571]=C2=A0 ? __switch_to_asm+0x34/0x70 > > > > > [=C2=A0 690.369484]=C2=A0 file_write_and_wait_range+0x4c/0xa0 > > > > > [=C2=A0 690.375198]=C2=A0 xfs_file_fsync+0x5d/0x230 [xfs] > > > > > [=C2=A0 690.380499]=C2=A0 do_fsync+0x38/0x60 > > > > > [=C2=A0 690.384534]=C2=A0 __x64_sys_fsync+0x10/0x20 > > > > > [=C2=A0 690.389243]=C2=A0 do_syscall_64+0x5b/0x1e0 > > > > > [=C2=A0 690.393857]=C2=A0 entry_SYSCALL_64_after_hwframe+0x44/0xa9 > > > > > [=C2=A0 690.400020] RIP: 0033:0x7fa8cbe4f70d > > > > > [=C2=A0 690.404533] Code: bc 20 00 00 75 10 b8 4a 00 00 00 0f 05 = 48 3d 01 f0 ff > > > > > ff 73 31 c3 48 83 ec 08 e8 6e f7 ff ff 48 89 04 24 b8 4a 00 00 00= 0f 05 <48> > > > > > 8b 3c 24 48 89 c2 e8 b7 f7 ff ff 48 89 d0 48 83 c4 08 48 3d 01 > > > > > [=C2=A0 690.426581] RSP: 002b:00007ffed6954160 EFLAGS: 00000293 O= RIG_RAX: > > > > > 000000000000004a > > > > > [=C2=A0 690.435583] RAX: ffffffffffffffda RBX: 00005573363790f0 R= CX: > > > > > 00007fa8cbe4f70d > > > > > [=C2=A0 690.444102] RDX: 0000000000200000 RSI: 00005573363790f0 R= DI: > > > > > 0000000000000003 > > > > > [=C2=A0 690.452615] RBP: 00007fa8b337f418 R08: 0000000000000008 R= 09: > > > > > 0000000000000001 > > > > > [=C2=A0 690.461124] R10: 00000000897b0000 R11: 0000000000000293 R= 12: > > > > > 0000000000000000 > > > > > [=C2=A0 690.469631] R13: 0000001000000000 R14: 00007fa8c81dda50 R= 15: > > > > > 0000000000000000 > > > > > [=C2=A0 690.478130] ---[ end trace 90dd4ee6d5cdfcb8 ]--- > > > > I have just checked xfs code, and looks the only reason is that bio= size > > > > is overflowed, that means the whole bio may reach 4GB. That might be > > > > possible in case of writeback. But I have to double check if that is > > > > the case, could you test the following patch and provide the log? > > > > = > > > > diff --git a/block/blk-core.c b/block/blk-core.c > > > > index 8340f69670d8..9d61e408c608 100644 > > > > --- a/block/blk-core.c > > > > +++ b/block/blk-core.c > > > > @@ -1125,6 +1125,19 @@ blk_qc_t submit_bio(struct bio *bio) > > > > bio_devname(bio, b), count); > > > > } > > > > } > > > > + if (WARN_ON(bio->bi_iter.bi_size =3D=3D 0 && bio_op(bio) =3D=3D R= EQ_OP_WRITE)) { > > > > + int i; > > > > + printk("bio: %hx/%hx %u %llu %u\n", bio->bi_flags, bio->bi_opf, > > > > + bio->bi_vcnt, > > > > + (unsigned long long)bio->bi_iter.bi_sector, > > > > + bio->bi_iter.bi_size); > > > > + for (i =3D 0; i < bio->bi_vcnt; i++) { > > > > + struct bio_vec *bvec =3D &bio->bi_io_vec[i]; > > > > + printk("%d: pfn %lu offset %u len %u\n", i++, > > > > + (unsigned long)page_to_pfn(bvec->bv= _page), > > > > + bvec->bv_offset, bvec->bv_len); > > > > + } > > > > + } > > > > return generic_make_request(bio); > > > > } > > > the newest log and i also see you fix patch,=C2=A0 i will test your f= ix patch > > > more times to check it. > > > = > > > ------ > > > = > > > [=C2=A0=C2=A0 54.656016] 2019-07-01 11:48:52 mkfs -t xfs /dev/sdb1 > > > [=C2=A0=C2=A0 54.656017] > > > [=C2=A0 119.228467] WARNING: CPU: 5 PID: 1530 at block/blk-core.c:1128 > > > submit_bio+0xa5/0x1d0 > > > [=C2=A0 119.240810] Modules linked in: dm_mod btrfs xor zstd_decompre= ss > > > zstd_compress raid6_pq libcrc32c sd_mod intel_rapl sg sb_edac > > > x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm irqbypass > > > mgag200 crct10dif_pclmul snd_pcm crc32_pclmul ttm crc32c_intel snd_ti= mer > > > drm_kms_helper ghash_clmulni_intel snd syscopyarea ipmi_si sysfillrect > > > soundcore aesni_intel sysimgblt ipmi_devintf crypto_simd fb_sys_fops = ahci > > > mei_me libahci cryptd ipmi_msghandler pcspkr glue_helper drm mei liba= ta > > > joydev ioatdma lpc_ich i2c_i801 dca wmi acpi_power_meter acpi_pad > > > pcc_cpufreq ip_tables > > > [=C2=A0 119.306252] CPU: 5 PID: 1530 Comm: mkfs.xfs Tainted: G=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 W I > > > 5.2.0-rc5-00001-g82f8097 #1 > > > [=C2=A0 119.318274] Hardware name: Intel Corporation S2600WTT/S2600WT= T, BIOS > > > SE5C610.86B.01.01.0008.021120151325 02/11/2015 > > > [=C2=A0 119.331852] RIP: 0010:submit_bio+0xa5/0x1d0 > > > [=C2=A0 119.338422] Code: 00 00 65 48 01 35 b3 39 5f 76 8b 0d a1 96 c= 4 01 85 c9 > > > 0f 85 9b 00 00 00 8b 43 28 85 c0 75 a5 8b 53 10 0f b6 ca 83 f9 01 75 = 9a <0f> > > > 0b 0f b7 4b 68 0f b7 73 14 45 31 c9 4c 8b 43 20 48 c7 c7 f1 99 > > > [=C2=A0 119.363296] RSP: 0018:ffff940c4fb17e58 EFLAGS: 00010246 > > > [=C2=A0 119.371107] RAX: 0000000000000000 RBX: ffff88a53e3d6900 RCX: > > > 0000000000000001 > > > [=C2=A0 119.381068] RDX: 0000000000040801 RSI: ffff88868138c000 RDI: > > > ffff88a53e3d6900 > > > [=C2=A0 119.391041] RBP: ffff88a5448b0000 R08: ffff88a53e3d6900 R09: > > > 0000000000000000 > > > [=C2=A0 119.401003] R10: 0000000000000000 R11: 0000000000000000 R12: > > > 0000000000000000 > > > [=C2=A0 119.410967] R13: 0000000000000000 R14: 0000000000000000 R15: > > > 0000000000000000 > > > [=C2=A0 119.420881] FS:=C2=A0 00007f37eba81780(0000) GS:ffff88a57f340= 000(0000) > > > knlGS:0000000000000000 > > > [=C2=A0 119.431839] CS:=C2=A0 0010 DS: 0000 ES: 0000 CR0: 00000000800= 50033 > > > [=C2=A0 119.440161] CR2: 00005630eb09d000 CR3: 0000001fc711c002 CR4: > > > 00000000001606e0 > > > [=C2=A0 119.450017] Call Trace: > > > [=C2=A0 119.454595]=C2=A0 submit_bio_wait+0x53/0x80 > > > [=C2=A0 119.460636]=C2=A0 blkdev_issue_flush+0x81/0xb0 > > > [=C2=A0 119.466930]=C2=A0 blkdev_fsync+0x2f/0x40 > > > [=C2=A0 119.472590]=C2=A0 do_fsync+0x38/0x60 > > > [=C2=A0 119.477810]=C2=A0 __x64_sys_fsync+0x10/0x20 > > > [=C2=A0 119.483722]=C2=A0 do_syscall_64+0x5b/0x1e0 > > > [=C2=A0 119.489516]=C2=A0 entry_SYSCALL_64_after_hwframe+0x44/0xa9 > > > [=C2=A0 119.496857] RIP: 0033:0x7f37eb2146f0 > > > [=C2=A0 119.502502] Code: 48 3d 01 f0 ff ff 73 01 c3 48 8b 0d b2 78 2= 0 00 f7 d8 > > > 64 89 01 48 83 c8 ff c3 83 3d d9 bc 20 00 00 75 10 b8 4a 00 00 00 0f = 05 <48> > > > 3d 01 f0 ff ff 73 31 c3 48 83 ec 08 e8 6e f7 ff ff 48 89 04 24 > > > [=C2=A0 119.526842] RSP: 002b:00007fffb27f4878 EFLAGS: 00000246 ORIG_= RAX: > > > 000000000000004a > > > [=C2=A0 119.536975] RAX: ffffffffffffffda RBX: 0000000000000811 RCX: > > > 00007f37eb2146f0 > > > [=C2=A0 119.546592] RDX: 00005630e91ba8e0 RSI: 0000000000000000 RDI: > > > 0000000000000004 > > > [=C2=A0 119.556171] RBP: 0000000000000004 R08: 00005630eb02b8a0 R09: > > > 00007fffb27f4e00 > > > [=C2=A0 119.565753] R10: 000000000000009b R11: 0000000000000246 R12: > > > 00007fffb27f4c60 > > > [=C2=A0 119.575339] R13: 00007fffb27f4a30 R14: 00005630eb048800 R15: > > > 00005630eb05a800 > > > [=C2=A0 119.584940] ---[ end trace 46a2e1214c930529 ]--- > > > [=C2=A0 119.591765] bio: 0/801 0 0 0 > > > = > > Hi Yiding, > > = > > The above warning is false positive, and it is flush request, and > > it isn't same with your original report in which the following warning > > is triggered. > > = > > "drivers/scsi/scsi_lib.c:1034 scsi_init_io+0x11a/0x2e0" > > = > > So please run the test kernel with the debug patch I sent to you today > > and try to trigger the warning in scsi_init_io(). > = > = > ------ > = > [=C2=A0 728.863669] 230: pfn 66784668 offset 0 len 17031168 > [=C2=A0 728.869319] 232: pfn 66788826 offset 0 len 22708224 > [=C2=A0 728.874946] 234: pfn 66794370 offset 0 len 29933568 > [=C2=A0 728.880563] 236: pfn 66801678 offset 0 len 4136960 > [=C2=A0 728.886086] 238: pfn 66820096 offset 0 len 2506752 Hi Yiding, Could you include all logs dumped by the debug patch? + if (WARN_ON(bio->bi_iter.bi_size =3D=3D 0 && bio_op(bio) =3D=3D REQ= _OP_WRITE)) { + int i; + printk("bio: %hx/%hx %u %llu %u\n", bio->bi_flags, bio->bi_= opf, + bio->bi_vcnt, + (unsigned long long)bio->bi_iter.bi_sector, + bio->bi_iter.bi_size); + for (i =3D 0; i < bio->bi_vcnt; i++) { + struct bio_vec *bvec =3D &bio->bi_io_vec[i]; + printk("%d: pfn %lu offset %u len %u\n", i++, + (unsigned long)page_to_pfn(bvec->bv_page), + bvec->bv_offset, bvec->bv_len); + } + } Thanks, Ming --===============1399901089514739395==--