Netdev Archive on lore.kernel.org
 help / color / Atom feed
* [igb] netconsole triggers warning in netpoll_poll_dev
@ 2021-04-06 12:36 Oleksandr Natalenko
  2021-04-06 18:48 ` Jakub Kicinski
  0 siblings, 1 reply; 7+ messages in thread
From: Oleksandr Natalenko @ 2021-04-06 12:36 UTC (permalink / raw)
  To: linux-kernel
  Cc: Jesse Brandeburg, Tony Nguyen, David S. Miller, Jakub Kicinski,
	intel-wired-lan, netdev

Hello.

I've raised this here [1] first, but was suggested to engage igb devs,
so here we are.

I'm experiencing the following woes while using netconsole regularly:

```
[22038.710800] ------------[ cut here ]------------
[22038.710801] igb_poll+0x0/0x1440 [igb] exceeded budget in poll
[22038.710802] WARNING: CPU: 12 PID: 40362 at net/core/netpoll.c:155 netpoll_poll_dev+0x18a/0x1a0
[22038.710802] Modules linked in: blocklayoutdriver rpcsec_gss_krb5 auth_rpcgss nfsv4 dns_resolver nfs lockd grace sunrpc nfs_ssc fscache uinput netconsole rfcomm nft_ct nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 cmac algif_hash algif_skcipher af_alg nf_tables tun bnep nfnetlink iwlmvm snd_hda_codec_realtek snd_hda_codec_generic ledtrig_audio mac80211 snd_hda_codec_hdmi snd_hda_intel nls_iso8859_1 snd_intel_dspcfg soundwire_intel vfat libarc4 soundwire_generic_allocation soundwire_cadence fat intel_rapl_msr snd_hda_codec intel_rapl_common iwlwifi snd_hda_core soundwire_bus snd_usb_audio snd_soc_core eeepc_wmi btusb asus_wmi snd_usbmidi_lib btrtl edac_mce_amd sparse_keymap btbcm video wmi_bmof mxm_wmi kvm_amd snd_hwdep snd_compress btintel snd_rawmidi ac97_bus cfg80211 uvcvideo bluetooth snd_pcm_dmaengine snd_seq_device snd_pcm videobuf2_vmalloc kvm videobuf2_memops snd_timer videobuf2_v4l2 joydev igb mousedev ecdh_generic snd irqbypass pl2303 ipmi_devintf ecc videobuf2_common soundcore
[22038.710820]  r8169 ipmi_msghandler rapl crc16 k10temp rfkill sp5100_tco dca i2c_piix4 tpm_crb realtek mdio_devres tpm_tis libphy tpm_tis_core pinctrl_amd wmi acpi_cpufreq mac_hid tcp_bbr2 vhost_vsock vmw_vsock_virtio_transport_common vhost vhost_iotlb vsock v4l2loopback videodev mc nct6775 hwmon_vid crypto_user fuse ip_tables x_tables xfs dm_thin_pool dm_persistent_data dm_bio_prison dm_bufio libcrc32c crc32c_generic dm_crypt cbc encrypted_keys trusted tpm hid_logitech_hidpp hid_logitech_dj usbhid dm_mod raid10 crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel aesni_intel md_mod crypto_simd cryptd glue_helper ccp amdgpu rng_core xhci_pci xhci_pci_renesas drm_ttm_helper ttm gpu_sched i2c_algo_bit drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops cec drm agpgart
[22038.710835] CPU: 12 PID: 40362 Comm: systemd-sleep Not tainted 5.11.0-pf7 #1
[22038.710836] Hardware name: ASUS System Product Name/Pro WS X570-ACE, BIOS 3302 03/05/2021
[22038.710836] RIP: 0010:netpoll_poll_dev+0x18a/0x1a0
[22038.710837] Code: 6e ff 80 3d d2 9d f8 00 00 0f 85 5c ff ff ff 48 8b 73 28 48 c7 c7 0c b8 21 84 89 44 24 04 c6 05 b6 9d f8 00 01 e8 84 21 1c 00 <0f> 0b 8b 54 24 04 e9 36 ff ff ff 66 66 2e 0f 1f 84 00 00 00 00 00
[22038.710838] RSP: 0018:ffffb24106e37ba0 EFLAGS: 00010086
[22038.710838] RAX: 0000000000000000 RBX: ffff9599d2929c50 RCX: ffff959f8ed1ac30
[22038.710839] RDX: 0000000000000000 RSI: 0000000000000023 RDI: ffff959f8ed1ac28
[22038.710839] RBP: ffff9598981d4058 R08: 0000000000000019 R09: ffffb24206e3796d
[22038.710839] R10: ffffffffffffffff R11: ffffb24106e37968 R12: ffff959887e51ec8
[22038.710840] R13: 000000000000000c R14: 00000000ffffffff R15: ffff9599d2929c60
[22038.710840] FS:  00007f3ade370a40(0000) GS:ffff959f8ed00000(0000) knlGS:0000000000000000
[22038.710841] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[22038.710841] CR2: 0000000000000000 CR3: 00000003017b0000 CR4: 0000000000350ee0
[22038.710841] Call Trace:
[22038.710842]  netpoll_send_skb+0x185/0x240
[22038.710842]  write_msg+0xe5/0x100 [netconsole]
[22038.710842]  console_unlock+0x37d/0x640
[22038.710842]  ? __schedule+0x2e5/0xc90
[22038.710843]  suspend_devices_and_enter+0x2ac/0x7f0
[22038.710843]  pm_suspend.cold+0x321/0x36c
[22038.710843]  state_store+0xa6/0x140
[22038.710844]  kernfs_fop_write_iter+0x124/0x1b0
[22038.710844]  new_sync_write+0x16a/0x200
[22038.710844]  vfs_write+0x21c/0x2e0
[22038.710844]  __x64_sys_write+0x6d/0xf0
[22038.710845]  do_syscall_64+0x33/0x40
[22038.710845]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
[22038.710845] RIP: 0033:0x7f3adece10f7
[22038.710846] Code: 0d 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b7 0f 1f 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 48 89 54 24 18 48 89 74 24
[22038.710847] RSP: 002b:00007ffc51c555b8 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
[22038.710847] RAX: ffffffffffffffda RBX: 0000000000000004 RCX: 00007f3adece10f7
[22038.710848] RDX: 0000000000000004 RSI: 00007ffc51c556a0 RDI: 0000000000000004
[22038.710848] RBP: 00007ffc51c556a0 R08: 000055ea374302a0 R09: 00007f3aded770c0
[22038.710849] R10: 00007f3aded76fc0 R11: 0000000000000246 R12: 0000000000000004
[22038.710849] R13: 000055ea3742c430 R14: 0000000000000004 R15: 00007f3adedb3700
[22038.710849] ---[ end trace 6eae54fbf23807f8 ]---
```

This one happened during suspend/resume cycle (on resume), followed by:

```
[22038.868669] igb 0000:05:00.0 enp5s0: Reset adapter
[22040.998673] igb 0000:05:00.0 enp5s0: Reset adapter
[22043.819198] igb 0000:05:00.0 enp5s0: igb: enp5s0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX
```

I've bumped into a similar issue in BZ 211911 [2] (see c#16),
and in c#17 it was suggested it was a separate unrelated issue,
hence I'm raising a new concern.

Please help in finding out why this woe happens and in fixing it.

Thanks.

[1] https://bugzilla.kernel.org/show_bug.cgi?id=212573
[2] https://bugzilla.kernel.org/show_bug.cgi?id=211911

-- 
  Oleksandr Natalenko (post-factum)

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [igb] netconsole triggers warning in netpoll_poll_dev
  2021-04-06 12:36 [igb] netconsole triggers warning in netpoll_poll_dev Oleksandr Natalenko
@ 2021-04-06 18:48 ` Jakub Kicinski
  2021-04-07  6:00   ` Oleksandr Natalenko
  0 siblings, 1 reply; 7+ messages in thread
From: Jakub Kicinski @ 2021-04-06 18:48 UTC (permalink / raw)
  To: Oleksandr Natalenko
  Cc: linux-kernel, Jesse Brandeburg, Tony Nguyen, David S. Miller,
	intel-wired-lan, netdev, Alexander Duyck

On Tue, 6 Apr 2021 14:36:19 +0200 Oleksandr Natalenko wrote:
> Hello.
> 
> I've raised this here [1] first, but was suggested to engage igb devs,
> so here we are.
> 
> I'm experiencing the following woes while using netconsole regularly:
> 
> ```
> [22038.710800] ------------[ cut here ]------------
> [22038.710801] igb_poll+0x0/0x1440 [igb] exceeded budget in poll
> [22038.710802] WARNING: CPU: 12 PID: 40362 at net/core/netpoll.c:155 netpoll_poll_dev+0x18a/0x1a0
> [22038.710802] Modules linked in: ...
> [22038.710835] CPU: 12 PID: 40362 Comm: systemd-sleep Not tainted 5.11.0-pf7 #1
> [22038.710836] Hardware name: ASUS System Product Name/Pro WS X570-ACE, BIOS 3302 03/05/2021
> [22038.710836] RIP: 0010:netpoll_poll_dev+0x18a/0x1a0
> [22038.710837] Code: 6e ff 80 3d d2 9d f8 00 00 0f 85 5c ff ff ff 48 8b 73 28 48 c7 c7 0c b8 21 84 89 44 24 04 c6 05 b6 9d f8 00 01 e8 84 21 1c 00 <0f> 0b 8b 54 24 04 e9 36 ff ff ff 66 66 2e 0f 1f 84 00 00 00 00 00
> [22038.710838] RSP: 0018:ffffb24106e37ba0 EFLAGS: 00010086
> [22038.710838] RAX: 0000000000000000 RBX: ffff9599d2929c50 RCX: ffff959f8ed1ac30
> [22038.710839] RDX: 0000000000000000 RSI: 0000000000000023 RDI: ffff959f8ed1ac28
> [22038.710839] RBP: ffff9598981d4058 R08: 0000000000000019 R09: ffffb24206e3796d
> [22038.710839] R10: ffffffffffffffff R11: ffffb24106e37968 R12: ffff959887e51ec8
> [22038.710840] R13: 000000000000000c R14: 00000000ffffffff R15: ffff9599d2929c60
> [22038.710840] FS:  00007f3ade370a40(0000) GS:ffff959f8ed00000(0000) knlGS:0000000000000000
> [22038.710841] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [22038.710841] CR2: 0000000000000000 CR3: 00000003017b0000 CR4: 0000000000350ee0
> [22038.710841] Call Trace:
> [22038.710842]  netpoll_send_skb+0x185/0x240
> [22038.710842]  write_msg+0xe5/0x100 [netconsole]
> [22038.710842]  console_unlock+0x37d/0x640
> [22038.710842]  ? __schedule+0x2e5/0xc90
> [22038.710843]  suspend_devices_and_enter+0x2ac/0x7f0
> [22038.710843]  pm_suspend.cold+0x321/0x36c
> [22038.710843]  state_store+0xa6/0x140
> [22038.710844]  kernfs_fop_write_iter+0x124/0x1b0
> [22038.710844]  new_sync_write+0x16a/0x200
> [22038.710844]  vfs_write+0x21c/0x2e0
> [22038.710844]  __x64_sys_write+0x6d/0xf0
> [22038.710845]  do_syscall_64+0x33/0x40
> [22038.710845]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
> [22038.710845] RIP: 0033:0x7f3adece10f7
> [22038.710846] Code: 0d 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b7 0f 1f 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 48 89 54 24 18 48 89 74 24
> [22038.710847] RSP: 002b:00007ffc51c555b8 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
> [22038.710847] RAX: ffffffffffffffda RBX: 0000000000000004 RCX: 00007f3adece10f7
> [22038.710848] RDX: 0000000000000004 RSI: 00007ffc51c556a0 RDI: 0000000000000004
> [22038.710848] RBP: 00007ffc51c556a0 R08: 000055ea374302a0 R09: 00007f3aded770c0
> [22038.710849] R10: 00007f3aded76fc0 R11: 0000000000000246 R12: 0000000000000004
> [22038.710849] R13: 000055ea3742c430 R14: 0000000000000004 R15: 00007f3adedb3700
> [22038.710849] ---[ end trace 6eae54fbf23807f8 ]---
> ```
> 
> This one happened during suspend/resume cycle (on resume), followed by:
> 
> ```
> [22038.868669] igb 0000:05:00.0 enp5s0: Reset adapter
> [22040.998673] igb 0000:05:00.0 enp5s0: Reset adapter
> [22043.819198] igb 0000:05:00.0 enp5s0: igb: enp5s0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX
> ```
> 
> I've bumped into a similar issue in BZ 211911 [2] (see c#16),
> and in c#17 it was suggested it was a separate unrelated issue,
> hence I'm raising a new concern.
> 
> Please help in finding out why this woe happens and in fixing it.
> 
> Thanks.
> 
> [1] https://bugzilla.kernel.org/show_bug.cgi?id=212573
> [2] https://bugzilla.kernel.org/show_bug.cgi?id=211911

Looks like igb_clean_tx_irq() should not return true if budget is 0,
ever, otherwise we risk hitting the min(work, budget - 1) which may
go negative.

So something like this? 

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index c9e8c65a3cfe..7a237b5311ca 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -8028,7 +8028,7 @@ static bool igb_clean_tx_irq(struct igb_q_vector *q_vector, int napi_budget)
        unsigned int i = tx_ring->next_to_clean;
 
        if (test_bit(__IGB_DOWN, &adapter->state))
-               return true;
+               goto out;
 
        tx_buffer = &tx_ring->tx_buffer_info[i];
        tx_desc = IGB_TX_DESC(tx_ring, i);
@@ -8157,7 +8157,7 @@ static bool igb_clean_tx_irq(struct igb_q_vector *q_vector, int napi_budget)
                                            tx_ring->queue_index);
 
                        /* we are about to reset, no point in enabling stuff */
-                       return true;
+                       goto out;
                }
        }
 
@@ -8180,7 +8180,7 @@ static bool igb_clean_tx_irq(struct igb_q_vector *q_vector, int napi_budget)
                        u64_stats_update_end(&tx_ring->tx_syncp);
                }
        }
-
+out:
        return !!budget;
 }

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [igb] netconsole triggers warning in netpoll_poll_dev
  2021-04-06 18:48 ` Jakub Kicinski
@ 2021-04-07  6:00   ` Oleksandr Natalenko
  2021-04-07 15:37     ` Jakub Kicinski
  0 siblings, 1 reply; 7+ messages in thread
From: Oleksandr Natalenko @ 2021-04-07  6:00 UTC (permalink / raw)
  To: Jakub Kicinski
  Cc: linux-kernel, Jesse Brandeburg, Tony Nguyen, David S. Miller,
	intel-wired-lan, netdev, Alexander Duyck

Hello.

On Tue, Apr 06, 2021 at 11:48:02AM -0700, Jakub Kicinski wrote:
> On Tue, 6 Apr 2021 14:36:19 +0200 Oleksandr Natalenko wrote:
> > Hello.
> > 
> > I've raised this here [1] first, but was suggested to engage igb devs,
> > so here we are.
> > 
> > I'm experiencing the following woes while using netconsole regularly:
> > 
> > ```
> > [22038.710800] ------------[ cut here ]------------
> > [22038.710801] igb_poll+0x0/0x1440 [igb] exceeded budget in poll
> > [22038.710802] WARNING: CPU: 12 PID: 40362 at net/core/netpoll.c:155 netpoll_poll_dev+0x18a/0x1a0
> > [22038.710802] Modules linked in: ...
> > [22038.710835] CPU: 12 PID: 40362 Comm: systemd-sleep Not tainted 5.11.0-pf7 #1
> > [22038.710836] Hardware name: ASUS System Product Name/Pro WS X570-ACE, BIOS 3302 03/05/2021
> > [22038.710836] RIP: 0010:netpoll_poll_dev+0x18a/0x1a0
> > [22038.710837] Code: 6e ff 80 3d d2 9d f8 00 00 0f 85 5c ff ff ff 48 8b 73 28 48 c7 c7 0c b8 21 84 89 44 24 04 c6 05 b6 9d f8 00 01 e8 84 21 1c 00 <0f> 0b 8b 54 24 04 e9 36 ff ff ff 66 66 2e 0f 1f 84 00 00 00 00 00
> > [22038.710838] RSP: 0018:ffffb24106e37ba0 EFLAGS: 00010086
> > [22038.710838] RAX: 0000000000000000 RBX: ffff9599d2929c50 RCX: ffff959f8ed1ac30
> > [22038.710839] RDX: 0000000000000000 RSI: 0000000000000023 RDI: ffff959f8ed1ac28
> > [22038.710839] RBP: ffff9598981d4058 R08: 0000000000000019 R09: ffffb24206e3796d
> > [22038.710839] R10: ffffffffffffffff R11: ffffb24106e37968 R12: ffff959887e51ec8
> > [22038.710840] R13: 000000000000000c R14: 00000000ffffffff R15: ffff9599d2929c60
> > [22038.710840] FS:  00007f3ade370a40(0000) GS:ffff959f8ed00000(0000) knlGS:0000000000000000
> > [22038.710841] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > [22038.710841] CR2: 0000000000000000 CR3: 00000003017b0000 CR4: 0000000000350ee0
> > [22038.710841] Call Trace:
> > [22038.710842]  netpoll_send_skb+0x185/0x240
> > [22038.710842]  write_msg+0xe5/0x100 [netconsole]
> > [22038.710842]  console_unlock+0x37d/0x640
> > [22038.710842]  ? __schedule+0x2e5/0xc90
> > [22038.710843]  suspend_devices_and_enter+0x2ac/0x7f0
> > [22038.710843]  pm_suspend.cold+0x321/0x36c
> > [22038.710843]  state_store+0xa6/0x140
> > [22038.710844]  kernfs_fop_write_iter+0x124/0x1b0
> > [22038.710844]  new_sync_write+0x16a/0x200
> > [22038.710844]  vfs_write+0x21c/0x2e0
> > [22038.710844]  __x64_sys_write+0x6d/0xf0
> > [22038.710845]  do_syscall_64+0x33/0x40
> > [22038.710845]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
> > [22038.710845] RIP: 0033:0x7f3adece10f7
> > [22038.710846] Code: 0d 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b7 0f 1f 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 48 89 54 24 18 48 89 74 24
> > [22038.710847] RSP: 002b:00007ffc51c555b8 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
> > [22038.710847] RAX: ffffffffffffffda RBX: 0000000000000004 RCX: 00007f3adece10f7
> > [22038.710848] RDX: 0000000000000004 RSI: 00007ffc51c556a0 RDI: 0000000000000004
> > [22038.710848] RBP: 00007ffc51c556a0 R08: 000055ea374302a0 R09: 00007f3aded770c0
> > [22038.710849] R10: 00007f3aded76fc0 R11: 0000000000000246 R12: 0000000000000004
> > [22038.710849] R13: 000055ea3742c430 R14: 0000000000000004 R15: 00007f3adedb3700
> > [22038.710849] ---[ end trace 6eae54fbf23807f8 ]---
> > ```
> > 
> > This one happened during suspend/resume cycle (on resume), followed by:
> > 
> > ```
> > [22038.868669] igb 0000:05:00.0 enp5s0: Reset adapter
> > [22040.998673] igb 0000:05:00.0 enp5s0: Reset adapter
> > [22043.819198] igb 0000:05:00.0 enp5s0: igb: enp5s0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX
> > ```
> > 
> > I've bumped into a similar issue in BZ 211911 [2] (see c#16),
> > and in c#17 it was suggested it was a separate unrelated issue,
> > hence I'm raising a new concern.
> > 
> > Please help in finding out why this woe happens and in fixing it.
> > 
> > Thanks.
> > 
> > [1] https://bugzilla.kernel.org/show_bug.cgi?id=212573
> > [2] https://bugzilla.kernel.org/show_bug.cgi?id=211911
> 
> Looks like igb_clean_tx_irq() should not return true if budget is 0,
> ever, otherwise we risk hitting the min(work, budget - 1) which may
> go negative.
> 
> So something like this? 
> 
> diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
> index c9e8c65a3cfe..7a237b5311ca 100644
> --- a/drivers/net/ethernet/intel/igb/igb_main.c
> +++ b/drivers/net/ethernet/intel/igb/igb_main.c
> @@ -8028,7 +8028,7 @@ static bool igb_clean_tx_irq(struct igb_q_vector *q_vector, int napi_budget)
>         unsigned int i = tx_ring->next_to_clean;
>  
>         if (test_bit(__IGB_DOWN, &adapter->state))
> -               return true;
> +               goto out;
>  
>         tx_buffer = &tx_ring->tx_buffer_info[i];
>         tx_desc = IGB_TX_DESC(tx_ring, i);
> @@ -8157,7 +8157,7 @@ static bool igb_clean_tx_irq(struct igb_q_vector *q_vector, int napi_budget)
>                                             tx_ring->queue_index);
>  
>                         /* we are about to reset, no point in enabling stuff */
> -                       return true;
> +                       goto out;
>                 }
>         }
>  
> @@ -8180,7 +8180,7 @@ static bool igb_clean_tx_irq(struct igb_q_vector *q_vector, int napi_budget)
>                         u64_stats_update_end(&tx_ring->tx_syncp);
>                 }
>         }
> -
> +out:
>         return !!budget;
>  }

Thanks for the effort, but reportedly [1] it made no difference,
unfortunately.

[1] https://bugzilla.kernel.org/show_bug.cgi?id=212573#c8

-- 
  Oleksandr Natalenko (post-factum)

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [igb] netconsole triggers warning in netpoll_poll_dev
  2021-04-07  6:00   ` Oleksandr Natalenko
@ 2021-04-07 15:37     ` Jakub Kicinski
  2021-04-07 16:25       ` Alexander Duyck
  0 siblings, 1 reply; 7+ messages in thread
From: Jakub Kicinski @ 2021-04-07 15:37 UTC (permalink / raw)
  To: Oleksandr Natalenko
  Cc: linux-kernel, Jesse Brandeburg, Tony Nguyen, David S. Miller,
	intel-wired-lan, netdev, Alexander Duyck

On Wed, 7 Apr 2021 08:00:53 +0200 Oleksandr Natalenko wrote:
> Thanks for the effort, but reportedly [1] it made no difference,
> unfortunately.
> 
> [1] https://bugzilla.kernel.org/show_bug.cgi?id=212573#c8

The only other option I see is that somehow the NAPI has no rings.

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index a45cd2b416c8..24568adc2fb1 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -7980,7 +7980,7 @@ static int igb_poll(struct napi_struct *napi, int budget)
        struct igb_q_vector *q_vector = container_of(napi,
                                                     struct igb_q_vector,
                                                     napi);
-       bool clean_complete = true;
+       bool clean_complete = q_vector->tx.ring || q_vector->rx.ring;
        int work_done = 0;
 
 #ifdef CONFIG_IGB_DCA

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [igb] netconsole triggers warning in netpoll_poll_dev
  2021-04-07 15:37     ` Jakub Kicinski
@ 2021-04-07 16:25       ` Alexander Duyck
  2021-04-07 18:07         ` Jakub Kicinski
  0 siblings, 1 reply; 7+ messages in thread
From: Alexander Duyck @ 2021-04-07 16:25 UTC (permalink / raw)
  To: Jakub Kicinski
  Cc: Oleksandr Natalenko, linux-kernel, Jesse Brandeburg, Tony Nguyen,
	David S. Miller, intel-wired-lan, netdev

On Wed, Apr 7, 2021 at 8:37 AM Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Wed, 7 Apr 2021 08:00:53 +0200 Oleksandr Natalenko wrote:
> > Thanks for the effort, but reportedly [1] it made no difference,
> > unfortunately.
> >
> > [1] https://bugzilla.kernel.org/show_bug.cgi?id=212573#c8
>
> The only other option I see is that somehow the NAPI has no rings.
>
> diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
> index a45cd2b416c8..24568adc2fb1 100644
> --- a/drivers/net/ethernet/intel/igb/igb_main.c
> +++ b/drivers/net/ethernet/intel/igb/igb_main.c
> @@ -7980,7 +7980,7 @@ static int igb_poll(struct napi_struct *napi, int budget)
>         struct igb_q_vector *q_vector = container_of(napi,
>                                                      struct igb_q_vector,
>                                                      napi);
> -       bool clean_complete = true;
> +       bool clean_complete = q_vector->tx.ring || q_vector->rx.ring;
>         int work_done = 0;
>
>  #ifdef CONFIG_IGB_DCA

It might make sense to just cast the work_done as a unsigned int, and
then on the end of igb_poll use:
  return min_t(unsigned int, work_done, budget - 1);

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [igb] netconsole triggers warning in netpoll_poll_dev
  2021-04-07 16:25       ` Alexander Duyck
@ 2021-04-07 18:07         ` Jakub Kicinski
  2021-04-07 23:06           ` Alexander Duyck
  0 siblings, 1 reply; 7+ messages in thread
From: Jakub Kicinski @ 2021-04-07 18:07 UTC (permalink / raw)
  To: Alexander Duyck
  Cc: Oleksandr Natalenko, linux-kernel, Jesse Brandeburg, Tony Nguyen,
	David S. Miller, intel-wired-lan, netdev

On Wed, 7 Apr 2021 09:25:28 -0700 Alexander Duyck wrote:
> On Wed, Apr 7, 2021 at 8:37 AM Jakub Kicinski <kuba@kernel.org> wrote:
> >
> > On Wed, 7 Apr 2021 08:00:53 +0200 Oleksandr Natalenko wrote:  
> > > Thanks for the effort, but reportedly [1] it made no difference,
> > > unfortunately.
> > >
> > > [1] https://bugzilla.kernel.org/show_bug.cgi?id=212573#c8  
> >
> > The only other option I see is that somehow the NAPI has no rings.
> >
> > diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
> > index a45cd2b416c8..24568adc2fb1 100644
> > --- a/drivers/net/ethernet/intel/igb/igb_main.c
> > +++ b/drivers/net/ethernet/intel/igb/igb_main.c
> > @@ -7980,7 +7980,7 @@ static int igb_poll(struct napi_struct *napi, int budget)
> >         struct igb_q_vector *q_vector = container_of(napi,
> >                                                      struct igb_q_vector,
> >                                                      napi);
> > -       bool clean_complete = true;
> > +       bool clean_complete = q_vector->tx.ring || q_vector->rx.ring;
> >         int work_done = 0;
> >
> >  #ifdef CONFIG_IGB_DCA  
> 
> It might make sense to just cast the work_done as a unsigned int, and
> then on the end of igb_poll use:
>   return min_t(unsigned int, work_done, budget - 1);

Sure, that's simplest. I wasn't sure something is supposed to prevent
this condition or if it's okay to cover it up.

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [igb] netconsole triggers warning in netpoll_poll_dev
  2021-04-07 18:07         ` Jakub Kicinski
@ 2021-04-07 23:06           ` Alexander Duyck
  0 siblings, 0 replies; 7+ messages in thread
From: Alexander Duyck @ 2021-04-07 23:06 UTC (permalink / raw)
  To: Jakub Kicinski
  Cc: Oleksandr Natalenko, linux-kernel, Jesse Brandeburg, Tony Nguyen,
	David S. Miller, intel-wired-lan, netdev

On Wed, Apr 7, 2021 at 11:07 AM Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Wed, 7 Apr 2021 09:25:28 -0700 Alexander Duyck wrote:
> > On Wed, Apr 7, 2021 at 8:37 AM Jakub Kicinski <kuba@kernel.org> wrote:
> > >
> > > On Wed, 7 Apr 2021 08:00:53 +0200 Oleksandr Natalenko wrote:
> > > > Thanks for the effort, but reportedly [1] it made no difference,
> > > > unfortunately.
> > > >
> > > > [1] https://bugzilla.kernel.org/show_bug.cgi?id=212573#c8
> > >
> > > The only other option I see is that somehow the NAPI has no rings.
> > >
> > > diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
> > > index a45cd2b416c8..24568adc2fb1 100644
> > > --- a/drivers/net/ethernet/intel/igb/igb_main.c
> > > +++ b/drivers/net/ethernet/intel/igb/igb_main.c
> > > @@ -7980,7 +7980,7 @@ static int igb_poll(struct napi_struct *napi, int budget)
> > >         struct igb_q_vector *q_vector = container_of(napi,
> > >                                                      struct igb_q_vector,
> > >                                                      napi);
> > > -       bool clean_complete = true;
> > > +       bool clean_complete = q_vector->tx.ring || q_vector->rx.ring;
> > >         int work_done = 0;
> > >
> > >  #ifdef CONFIG_IGB_DCA
> >
> > It might make sense to just cast the work_done as a unsigned int, and
> > then on the end of igb_poll use:
> >   return min_t(unsigned int, work_done, budget - 1);
>
> Sure, that's simplest. I wasn't sure something is supposed to prevent
> this condition or if it's okay to cover it up.

I'm pretty sure it is okay to cover it up. In this case the "budget -
1" is supposed to be the upper limit on what can be reported. I think
it was assuming an unsigned value anyway.

Another alternative would be to default clean_complete to !!budget.
Then if budget is 0 clean_complete would always return false.

^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, back to index

Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-04-06 12:36 [igb] netconsole triggers warning in netpoll_poll_dev Oleksandr Natalenko
2021-04-06 18:48 ` Jakub Kicinski
2021-04-07  6:00   ` Oleksandr Natalenko
2021-04-07 15:37     ` Jakub Kicinski
2021-04-07 16:25       ` Alexander Duyck
2021-04-07 18:07         ` Jakub Kicinski
2021-04-07 23:06           ` Alexander Duyck

Netdev Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/netdev/0 netdev/git/0.git
	git clone --mirror https://lore.kernel.org/netdev/1 netdev/git/1.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 netdev netdev/ https://lore.kernel.org/netdev \
		netdev@vger.kernel.org
	public-inbox-index netdev

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.netdev


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git