Hi!
In mid-merge-window kernel I'm getting this warning in mt7601. My tree,
skipping certainly irrelevant extra patches, has merge-base with master at
289cf155d95df07ac1e692dfaa7036f10235c77c.
static void mt7601u_agc_tune(struct mt7601u_dev *dev)
{
u8 val = mt7601u_agc_default(dev);
long avg_rssi;
if (test_bit(MT7601U_STATE_SCANNING, &dev->state))
return;
/* Note: only in STA mode and not dozing; perhaps do this only if
* there is enough rssi updates since last run?
* Rssi updates are only on beacons and U2M so should work...
*/
spin_lock_bh(&dev->con_mon_lock);
avg_rssi = ewma_rssi_read(&dev->avg_rssi);
====> WARN_ON_ONCE(avg_rssi == 0);
avg_rssi = -avg_rssi;
WARNING: CPU: 4 PID: 109 at drivers/net/wireless/mediatek/mt7601u/phy.c:989 mt7601u_phy_calibrate+0x58b/0xaa0 [mt7601u]
Modules linked in: arc4 mt7601u mac80211 sha256_generic snd_usb_audio snd_usbmidi_lib cfg80211 rfkill nouveau video ttm
CPU: 4 PID: 109 Comm: kworker/u12:7 Not tainted 4.17.0-debug-07836-ga3643eb8ca3a #1
Hardware name: System manufacturer System Product Name/M4A77T, BIOS 2401 05/18/2011
Workqueue: phy0 mt7601u_phy_calibrate [mt7601u]
RIP: 0010:mt7601u_phy_calibrate+0x58b/0xaa0 [mt7601u]
Code: be b2 00 00 00 48 89 ef 0f 84 da 00 00 00 e8 1c f2 ff
ff 83 f8 01 19 c0 25 00 e0 ff ff 45 8d bc 07 66 46 00 00 e9 93 fc ff ff <0f>
0b e9 ba fe ff ff 8b 83 f0 07 00 00 e9 5d fc ff ff c0 e8 04 4d
RSP: 0018:ffffc90001febe10 EFLAGS: 00010256
RAX: 0000000000000000 RBX: ffff8802200f95d0 RCX: ffff88022700c020
RDX: 0000000000000001 RSI: ffff88021ffac0b0 RDI: ffff8802200f9d20
RBP: ffff8802200f9500 R08: 8080808080808080 R09: fefefefefefefeff
R10: ffffffff8287baf0 R11: 000000000000001f R12: 0000000000000034
R13: ffff8802200f9d20 R14: ffff88021f80de00 R15: 0000000000000000
FS: 0000000000000000(0000) GS:ffff88022fd00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000563bf575c168 CR3: 00000002259ed000 CR4: 00000000000006e0
Call Trace:
? process_one_work+0x239/0x680
? pwq_activate_delayed_work+0x5c/0x100
? worker_thread+0x4a/0x4b0
? process_one_work+0x680/0x680
? kthread+0x16b/0x220
? kthread_create_worker_on_cpu+0x40/0x40
? ret_from_fork+0x22/0x40
---[ end trace e40c73f71a371719 ]---
As bisecting on real metal is more involved than a VM, I'm reporting first
-- in case you already have a fix pending. If not, how may I help? Bisect?
Some extra debug data? Or, would you prefer to wait until the merge window
is over?
Meow!
--
⢀⣴⠾⠻⢶⣦⠀ I've read an article about how lively happy music boosts
⣾⠁⢰⠒⠀⣿⡁ productivity. You can read it, too, you just need the
⢿⡄⠘⠷⠚⠋⠀ right music while doing so. I recommend Skepticism
⠈⠳⣄⠀⠀⠀⠀ (funeral doom metal).
On Fri, Jun 08, 2018 at 12:43:23PM +0200, Adam Borowski wrote:
> Hi!
> In mid-merge-window kernel I'm getting this warning in mt7601. My tree,
> skipping certainly irrelevant extra patches, has merge-base with master at
> 289cf155d95df07ac1e692dfaa7036f10235c77c.
>
> static void mt7601u_agc_tune(struct mt7601u_dev *dev)
> {
> u8 val = mt7601u_agc_default(dev);
> long avg_rssi;
>
> if (test_bit(MT7601U_STATE_SCANNING, &dev->state))
> return;
>
> /* Note: only in STA mode and not dozing; perhaps do this only if
> * there is enough rssi updates since last run?
> * Rssi updates are only on beacons and U2M so should work...
> */
> spin_lock_bh(&dev->con_mon_lock);
> avg_rssi = ewma_rssi_read(&dev->avg_rssi);
> ====> WARN_ON_ONCE(avg_rssi == 0);
This WARNING by added by my commit
b305a6ab02475f52ef604b36e4cecd3bf4aa5eb7
I assumed that we will never run calibration code without receiving
some frames that will provide us RSSI. This turned out not to be true.
I will post the fix - just remove warning and return from the function
if avg_rssi == 0.
Thanks
Stanislaw