net: mscc: ocelot: replace readx_poll_timeout with readx_poll_timeout_atomic
Subsystems (3)
PATCH 1/1 net: mscc: ocelot: replace readx_poll_timeout with readx_poll_timeout_atomic
This fixes call trace like below to use atomic safe API:
BUG: sleeping function called from invalid context at drivers/net/ethernet/mscc/ocelot.c:59 in_atomic(): 1, irqs_disabled(): 0, pid: 3778, name: ifconfig INFO: lockdep is turned off. Preemption disabled at: [<ffff2b163c83b78c>] dev_set_rx_mode+0x24/0x40 Hardware name: LS1028A RDB Board (DT) Call trace: dump_backtrace+0x0/0x140 show_stack+0x24/0x30 dump_stack+0xc4/0x10c ___might_sleep+0x194/0x230 __might_sleep+0x58/0x90 ocelot_mact_forget+0x74/0xf8 ocelot_mc_unsync+0x2c/0x38 __hw_addr_sync_dev+0x6c/0x130 ocelot_set_rx_mode+0x8c/0xa0 __dev_set_rx_mode+0x58/0xa0 dev_set_rx_mode+0x2c/0x40 __dev_open+0x120/0x190 __dev_change_flags+0x168/0x1c0 dev_change_flags+0x3c/0x78 devinet_ioctl+0x6c4/0x7c8 inet_ioctl+0x2b8/0x2f8 sock_do_ioctl+0x54/0x260 sock_ioctl+0x21c/0x4d0 do_vfs_ioctl+0x6d4/0x968 ksys_ioctl+0x84/0xb8 __arm64_sys_ioctl+0x28/0x38 el0_svc_common.constprop.0+0x78/0x190 el0_svc_handler+0x70/0x90 el0_svc+0x8/0xc
--- drivers/net/ethernet/mscc/ocelot.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/mscc/ocelot.c +1 -1
@@ -56,7 +56,7 @@ static inline int ocelot_mact_wait_for_completion(struct ocelot *ocelot) {u32 val; - return readx_poll_timeout(ocelot_mact_read_macaccess, + return readx_poll_timeout_atomic(ocelot_mact_read_macaccess, ocelot, val, (val & ANA_TABLES_MACACCESS_MAC_TABLE_CMD_M) == MACACCESS_CMD_IDLE,
Signature · 1 line
2.17.1
Quoted 34 lines · Xulin Sun
This fixes call trace like below to use atomic safe API:
BUG: sleeping function called from invalid context at drivers/net/ethernet/mscc/ocelot.c:59 in_atomic(): 1, irqs_disabled(): 0, pid: 3778, name: ifconfig INFO: lockdep is turned off. Preemption disabled at: [<ffff2b163c83b78c>] dev_set_rx_mode+0x24/0x40 Hardware name: LS1028A RDB Board (DT) Call trace: dump_backtrace+0x0/0x140 show_stack+0x24/0x30 dump_stack+0xc4/0x10c ___might_sleep+0x194/0x230 __might_sleep+0x58/0x90 ocelot_mact_forget+0x74/0xf8 ocelot_mc_unsync+0x2c/0x38 __hw_addr_sync_dev+0x6c/0x130 ocelot_set_rx_mode+0x8c/0xa0 __dev_set_rx_mode+0x58/0xa0 dev_set_rx_mode+0x2c/0x40 __dev_open+0x120/0x190 __dev_change_flags+0x168/0x1c0 dev_change_flags+0x3c/0x78 devinet_ioctl+0x6c4/0x7c8 inet_ioctl+0x2b8/0x2f8 sock_do_ioctl+0x54/0x260 sock_ioctl+0x21c/0x4d0 do_vfs_ioctl+0x6d4/0x968 ksys_ioctl+0x84/0xb8 __arm64_sys_ioctl+0x28/0x38 el0_svc_common.constprop.0+0x78/0x190 el0_svc_handler+0x70/0x90 el0_svc+0x8/0xc
Quoted 1 line
Signed-off-by Xulin Sun <[email protected]>
Andrew
Hi Xulin,
Quoted 35 lines · Xulin Sun
This fixes call trace like below to use atomic safe API:
BUG: sleeping function called from invalid context at drivers/net/ethernet/mscc/ocelot.c:59 in_atomic(): 1, irqs_disabled(): 0, pid: 3778, name: ifconfig INFO: lockdep is turned off. Preemption disabled at: [<ffff2b163c83b78c>] dev_set_rx_mode+0x24/0x40 Hardware name: LS1028A RDB Board (DT) Call trace: dump_backtrace+0x0/0x140 show_stack+0x24/0x30 dump_stack+0xc4/0x10c ___might_sleep+0x194/0x230 __might_sleep+0x58/0x90 ocelot_mact_forget+0x74/0xf8 ocelot_mc_unsync+0x2c/0x38 __hw_addr_sync_dev+0x6c/0x130 ocelot_set_rx_mode+0x8c/0xa0 __dev_set_rx_mode+0x58/0xa0 dev_set_rx_mode+0x2c/0x40 __dev_open+0x120/0x190 __dev_change_flags+0x168/0x1c0 dev_change_flags+0x3c/0x78 devinet_ioctl+0x6c4/0x7c8 inet_ioctl+0x2b8/0x2f8 sock_do_ioctl+0x54/0x260 sock_ioctl+0x21c/0x4d0 do_vfs_ioctl+0x6d4/0x968 ksys_ioctl+0x84/0xb8 __arm64_sys_ioctl+0x28/0x38 el0_svc_common.constprop.0+0x78/0x190 el0_svc_handler+0x70/0x90 el0_svc+0x8/0xc
Quoted 2 lines
Signed-off-by Xulin Sun <[email protected]>---
The code path which you presented in your stack trace is impossible using code available currently in mainline.
Quoted 17 lines
drivers/net/ethernet/mscc/ocelot.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)drivers/net/ethernet/mscc/ocelot.c +1
-2@@ -56,7 +56,7 @@ static inline int ocelot_mact_wait_for_completion(struct ocelot *ocelot) {u32 val; - return readx_poll_timeout(ocelot_mact_read_macaccess, + return readx_poll_timeout_atomic(ocelot_mact_read_macaccess, ocelot, val, (val & ANA_TABLES_MACACCESS_MAC_TABLE_CMD_M) == MACACCESS_CMD_IDLE, --
Quoted 2 lines
2.17.1
Regards,
-Vladimir
David Miller
maintainerFrom: Xulin Sun <[email protected]>
Date: Fri, 15 May 2020 11:18:13 +0800
Quoted 8 lines
BUG: sleeping function called from invalid context at drivers/net/ethernet/mscc/ocelot.c:59 in_atomic(): 1, irqs_disabled(): 0, pid: 3778, name: ifconfig INFO: lockdep is turned off. Preemption disabled at: [<ffff2b163c83b78c>] dev_set_rx_mode+0x24/0x40 Hardware name: LS1028A RDB Board (DT) Call trace: dump_backtrace+0x0/0x140
Quoted 8 lines
show_stack+0x24/0x30 dump_stack+0xc4/0x10c ___might_sleep+0x194/0x230 __might_sleep+0x58/0x90 ocelot_mact_forget+0x74/0xf8 ocelot_mc_unsync+0x2c/0x38 __hw_addr_sync_dev+0x6c/0x130 ocelot_set_rx_mode+0x8c/0xa0
Vladimir states that this call chain is not possible in mainline.
I'm not applying this.
Quoted 21 lines · David Miller
From: Xulin Sun <[email protected]>
Date: Fri, 15 May 2020 11:18:13 +0800Quoted 8 lines
BUG: sleeping function called from invalid context at drivers/net/ethernet/mscc/ocelot.c:59 in_atomic(): 1, irqs_disabled(): 0, pid: 3778, name: ifconfig INFO: lockdep is turned off. Preemption disabled at: [<ffff2b163c83b78c>] dev_set_rx_mode+0x24/0x40 Hardware name: LS1028A RDB Board (DT) Call trace: dump_backtrace+0x0/0x140Quoted 8 lines
show_stack+0x24/0x30 dump_stack+0xc4/0x10c ___might_sleep+0x194/0x230 __might_sleep+0x58/0x90 ocelot_mact_forget+0x74/0xf8 ocelot_mc_unsync+0x2c/0x38 __hw_addr_sync_dev+0x6c/0x130 ocelot_set_rx_mode+0x8c/0xa0
Quoted 3 lines
Vladimir states that this call chain is not possible in mainline.
I'm not applying this.
(but the essence of the problem is legitimate though)
There are 2 specific things I don't like:
- The problem is claimed to reproduce on "LS1028A RDB Board (DT)" which does not call ocelot_set_rx_mode. So it claims to fix a problem for which only Xulin has the ability to decide whether it is the right solution or not.
- On ocelot, it _looks_ like it is indeed a problem which was introduced in 639c1b2625af ("net: mscc: ocelot: Register poll timeout should be wall time not attempts"). But there was no attempt to bring it up with the author of that patch, who very clearly expressed that he is working on hardware where the polling timeout is in the order of milliseconds, and the timeout for the driver is currently set at 100 ms. I'm not very sure that it is desirable to spin in atomic context for 100 ms.
Thanks,
-Vladimir