Il 24/03/2015 11:24, Christian Kanzian ha scritto:
> An old horse won't win the race? 
> It's a card from 2008? Please correct me if I am wrong.
> Disable OpenCL or get a newer GPU.

I am sorry for my previous encrypted message.

I do massive GPU calculus on BOINC and the same Einstein@home working
unit, compared with a calculus using only the CPU, is at least ten times
faster. So this discussion cannot end up with "your GPU is obsolete"
even if not all algorithms are tailored to work very fast on GPUs,
especially looking at the huge difference of timings of the benchmark I
posted.
The following dmesg errors make me think that it could be a bug of the nVidia 
drivers, so I would like
to get a feedback from Ulrich that is our OpenCL expert



==============================

[ 8343.366016] NVRM: Xid (PCI:0000:01:00): 8, Channel 00000003
[ 8343.370765] NVRM: Xid (PCI:0000:01:00): 13, Graphics Exception: ChID 0003, 
Class 000050c0, Offset 00000368, Data 00000000
[ 8343.373317] NVRM: Xid (PCI:0000:01:00): 13, Graphics Exception: ChID 0003, 
Class 000050c0, Offset 00000368, Data 00000000
[ 8343.375874] NVRM: Xid (PCI:0000:01:00): 13, Graphics Exception: ChID 0003, 
Class 000050c0, Offset 00000368, Data 00000000
[ 8345.370701] NVRM: os_schedule: Attempted to yield the CPU while in atomic or 
interrupt context
[ 8347.372127] NVRM: os_schedule: Attempted to yield the CPU while in atomic or 
interrupt context
[ 8349.373547] NVRM: os_schedule: Attempted to yield the CPU while in atomic or 
interrupt context
[ 8351.374972] NVRM: os_schedule: Attempted to yield the CPU while in atomic or 
interrupt context
[ 8353.376397] NVRM: os_schedule: Attempted to yield the CPU while in atomic or 
interrupt context
[ 8355.377816] NVRM: os_schedule: Attempted to yield the CPU while in atomic or 
interrupt context
[ 8357.379239] NVRM: os_schedule: Attempted to yield the CPU while in atomic or 
interrupt context
[ 8359.380661] NVRM: os_schedule: Attempted to yield the CPU while in atomic or 
interrupt context
[ 8361.382082] NVRM: os_schedule: Attempted to yield the CPU while in atomic or 
interrupt context
[ 8363.383504] NVRM: os_schedule: Attempted to yield the CPU while in atomic or 
interrupt context
[ 8365.384924] NVRM: os_schedule: Attempted to yield the CPU while in atomic or 
interrupt context
[ 8367.386346] NVRM: os_schedule: Attempted to yield the CPU while in atomic or 
interrupt context
[ 8369.387767] NVRM: os_schedule: Attempted to yield the CPU while in atomic or 
interrupt context
[ 8369.838481] NMI watchdog: BUG: soft lockup - CPU#1 stuck for 23s! 
[swapper/1:0]
[ 8369.838485] Modules linked in: nvidia_uvm(POE) bnep bluetooth cfg80211 
rfkill ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 xt_conntrack ebtable_nat 
ebtable_broute bridge stp llc ebtable_filter ebtables ip6table_nat 
nf_conntrack_ipv6 nf_defrag_ipv6 nf_nat_ipv6 ip6table_mangle ip6table_security 
ip6table_raw ip6table_filter ip6_tables iptable_nat nf_conntrack_ipv4 
nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security 
iptable_raw nvidia(POE) uvcvideo videobuf2_vmalloc videobuf2_core 
videobuf2_memops v4l2_common videodev snd_usb_audio snd_usbmidi_lib snd_hwdep 
media kvm_amd kvm ppdev snd_virtuoso snd_oxygen_lib snd_mpu401_uart snd_rawmidi 
snd_seq snd_seq_device snd_pcm snd_timer snd serio_raw soundcore edac_core drm 
k10temp microcode edac_mce_amd sp5100_tco shpchp parport_pc i2c_piix4
[ 8369.838574]  parport acpi_cpufreq binfmt_misc vboxnetadp(OE) vboxnetflt(OE) 
vboxdrv(OE) firewire_ohci firewire_core ata_generic pata_acpi crc_itu_t 
pata_jmicron uas usb_storage r8169 mii wmi
[ 8369.838601] CPU: 1 PID: 0 Comm: swapper/1 Tainted: P           OEL 
3.18.9-100.fc20.x86_64 #1
[ 8369.838607] Hardware name: Gigabyte Technology Co., Ltd. 
GA-890FXA-UD5/GA-890FXA-UD5, BIOS F6 11/24/2010
[ 8369.838613] task: ffff8804195575c0 ti: ffff880419644000 task.ti: 
ffff880419644000
[ 8369.838617] RIP: 0010:[<ffffffffa0774a5e>]  [<ffffffffa0774a5e>] 
_nv008626rm+0x54e/0xae0 [nvidia]
[ 8369.838968] RSP: 0018:ffff88042fc43c28  EFLAGS: 00000286
[ 8369.838972] RAX: 0000000000000000 RBX: ffff880418a7c008 RCX: ffff8800caf2d808
[ 8369.838977] RDX: ffff880408a2dc18 RSI: ffff880408a2dc60 RDI: ffff880418a7c008
[ 8369.838981] RBP: ffff880408a2dc28 R08: 0000000000000020 R09: ffff880408a2db74
[ 8369.838985] R10: 0000000000000504 R11: ffffffffa07700a0 R12: ffff88042fc43b98
[ 8369.838989] R13: ffffffff8174e1fd R14: ffff880408a2dc28 R15: ffff880408a2dc60
[ 8369.838995] FS:  00007f0ae2c98700(0000) GS:ffff88042fc40000(0000) 
knlGS:00000000ea3feb40
[ 8369.838999] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[ 8369.839003] CR2: 00007f09f8393d58 CR3: 0000000001c14000 CR4: 00000000000007e0
[ 8369.839007] Stack:
[ 8369.839010]  000000000040c258 ffff880408a2dc60 0000000008003800 
ffffffffa079ff9e
[ 8369.839017]  ffff880417353008 000000000000000f ffff880418a7c008 
ffff880417353230
[ 8369.839023]  ffff880408a2dd08 ffffffffa08d0a50 ffff880417354008 
ffff880418a7c008
[ 8369.839030] Call Trace:
[ 8369.839033]  <IRQ> 

[ 8369.839358]  [<ffffffffa079ff9e>] ? _nv009958rm+0x1ae/0x1f0 [nvidia]
[ 8369.839598]  [<ffffffffa08d0a50>] ? _nv014935rm+0x1f0/0x530 [nvidia]
[ 8369.839836]  [<ffffffffa08d522b>] ? _nv014985rm+0x9b/0x120 [nvidia]
[ 8369.840074]  [<ffffffffa08d0349>] ? _nv014919rm+0x349/0x5b0 [nvidia]
[ 8369.840312]  [<ffffffffa08cfcd2>] ? _nv014920rm+0xaa2/0xb60 [nvidia]
[ 8369.840548]  [<ffffffffa08d3950>] ? _nv015004rm+0xb0/0xd0 [nvidia]
[ 8369.840784]  [<ffffffffa08d3914>] ? _nv015004rm+0x74/0xd0 [nvidia]
[ 8369.841021]  [<ffffffffa08d4f19>] ? _nv015009rm+0x229/0x250 [nvidia]
[ 8369.841257]  [<ffffffffa08d3707>] ? _nv015005rm+0x77/0x210 [nvidia]
[ 8369.841441]  [<ffffffffa099b0b4>] ? _nv013578rm+0x1b4/0x6e0 [nvidia]
[ 8369.841604]  [<ffffffffa09f4590>] ? nvidia_ioctl+0x490/0x490 [nvidia]
[ 8369.841770]  [<ffffffffa09e655f>] ? rm_run_rc_callback+0x8f/0xf0 [nvidia]
[ 8369.841933]  [<ffffffffa09f45da>] ? nvidia_rc_timer+0x4a/0x80 [nvidia]
[ 8369.841943]  [<ffffffff810f795a>] ? call_timer_fn+0x3a/0x110
[ 8369.842106]  [<ffffffffa09f4590>] ? nvidia_ioctl+0x490/0x490 [nvidia]
[ 8369.842114]  [<ffffffff810f9382>] ? run_timer_softirq+0x212/0x2f0
[ 8369.842124]  [<ffffffff810992a5>] ? __do_softirq+0xf5/0x2d0
[ 8369.842133]  [<ffffffff8109978d>] ? irq_exit+0x12d/0x140
[ 8369.842140]  [<ffffffff8175012a>] ? smp_apic_timer_interrupt+0x4a/0x60
[ 8369.842148]  [<ffffffff8174e1fd>] ? apic_timer_interrupt+0x6d/0x80
[ 8369.842152]  <EOI> 

[ 8369.842161]  [<ffffffff81048fe9>] ? lapic_timer_setup+0x19/0x20
[ 8369.842168]  [<ffffffff810599f6>] ? native_safe_halt+0x6/0x10
[ 8369.842176]  [<ffffffff8101e62f>] ? default_idle+0x1f/0xc0
[ 8369.842184]  [<ffffffff8101e750>] ? amd_e400_idle+0x80/0x120
[ 8369.842191]  [<ffffffff8101efef>] ? arch_cpu_idle+0xf/0x20
[ 8369.842200]  [<ffffffff810d5642>] ? cpu_startup_entry+0x382/0x3c0
[ 8369.842208]  [<ffffffff81106a7c>] ? clockevents_register_device+0xbc/0x120
[ 8369.842215]  [<ffffffff810471f2>] ? start_secondary+0x192/0x200
[ 8369.842220] Code: 66 66 66 66 2e 0f 1f 84 00 00 00 00 00 41 54 45 31 e4 53 
48 89 f3 48 83 ec 08 48 83 ed 08 48 85 f6 48 89 7d 00 0f 84 2b 01 00 00 <48> 83 
7b 10 00 0f 84 05 01 00 00 48 8b 7d 00 48 85 ff 74 0a 66 
[ 8371.389189] NVRM: os_schedule: Attempted to yield the CPU while in atomic or 
interrupt context
[ 8373.390645] NVRM: os_schedule: Attempted to yield the CPU while in atomic or 
interrupt context
[ 8373.393680] NVRM: Xid (PCI:0000:01:00): 69, Class Error: ChId 0001, Class 
0000502d, Offset 000008dc, Data 00000094, ErrorCode 00000034
[ 8650.748017] EXT4-fs (dm-0): mounted filesystem with ordered data mode. Opts: 
(null)
[ 8650.748027] SELinux: initialized (dev dm-0, type ext4), uses xattr


------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
darktable-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/darktable-devel

Reply via email to