Re: BUG: KCSAN: data-race in rtl8169_poll
From: Mirsad Todorovac <hidden>
Date: 2023-09-19 06:20:19
Also in:
lkml
Subsystem:
8169 10/100/1000 gigabit ethernet driver, networking drivers, the rest · Maintainers:
Heiner Kallweit, Andrew Lunn, "David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, Linus Torvalds
On 9/18/23 13:44, Marco Elver wrote:
On Mon, 18 Sept 2023 at 12:39, Eric Dumazet [off-list ref] wrote:quoted
On Mon, Sep 18, 2023 at 11:43 AM Mirsad Todorovac [off-list ref] wrote:quoted
On 9/18/23 09:41, Eric Dumazet wrote:quoted
On Mon, Sep 18, 2023 at 8:15 AM Mirsad Todorovac [off-list ref] wrote:quoted
Hi all, In the vanilla torvalds tree kernel on Ubuntu 22.04, commit 6.6.0-rc1-kcsan-00269-ge789286468a9, KCSAN discovered a data-race in rtl8169_poll(): [ 9591.740976] ================================================================== [ 9591.740990] BUG: KCSAN: data-race in rtl8169_poll (drivers/net/ethernet/realtek/r8169_main.c:4430 drivers/net/ethernet/realtek/r8169_main.c:4583) r8169 [ 9591.741060] race at unknown origin, with read to 0xffff888109773130 of 4 bytes by interrupt on cpu 21: [ 9591.741073] rtl8169_poll (drivers/net/ethernet/realtek/r8169_main.c:4430 drivers/net/ethernet/realtek/r8169_main.c:4583) r8169 [ 9591.741135] __napi_poll (net/core/dev.c:6527) [ 9591.741149] net_rx_action (net/core/dev.c:6596 net/core/dev.c:6727) [ 9591.741161] __do_softirq (kernel/softirq.c:553) [ 9591.741175] __irq_exit_rcu (kernel/softirq.c:427 kernel/softirq.c:632) [ 9591.741185] irq_exit_rcu (kernel/softirq.c:647) [ 9591.741194] common_interrupt (arch/x86/kernel/irq.c:247 (discriminator 14)) [ 9591.741206] asm_common_interrupt (./arch/x86/include/asm/idtentry.h:636) [ 9591.741217] cpuidle_enter_state (drivers/cpuidle/cpuidle.c:291) [ 9591.741227] cpuidle_enter (drivers/cpuidle/cpuidle.c:390) [ 9591.741237] call_cpuidle (kernel/sched/idle.c:135) [ 9591.741249] do_idle (kernel/sched/idle.c:219 kernel/sched/idle.c:282) [ 9591.741259] cpu_startup_entry (kernel/sched/idle.c:378 (discriminator 1)) [ 9591.741268] start_secondary (arch/x86/kernel/smpboot.c:210 arch/x86/kernel/smpboot.c:294) [ 9591.741281] secondary_startup_64_no_verify (arch/x86/kernel/head_64.S:433) [ 9591.741300] value changed: 0x80003fff -> 0x34044510 [ 9591.741314] Reported by Kernel Concurrency Sanitizer on: [ 9591.741322] CPU: 21 PID: 0 Comm: swapper/21 Tainted: G L 6.6.0-rc1-kcsan-00269-ge789286468a9-dirty #4 [ 9591.741334] Hardware name: ASRock X670E PG Lightning/X670E PG Lightning, BIOS 1.21 04/26/2023 [ 9591.741343] ================================================================== (The taint is not from the proprietary module, but triggered from the previous reported and unfixed bug.) Apparently, it is this code: static int rtl8169_poll(struct napi_struct *napi, int budget) { struct rtl8169_private *tp = container_of(napi, struct rtl8169_private, napi); struct net_device *dev = tp->dev; int work_done; rtl_tx(dev, tp, budget); → work_done = rtl_rx(dev, tp, budget); if (work_done < budget && napi_complete_done(napi, work_done)) rtl_irq_enable(tp); return work_done; } and static int rtl_rx(struct net_device *dev, struct rtl8169_private *tp, int budget) { struct device *d = tp_to_dev(tp); int count; for (count = 0; count < budget; count++, tp->cur_rx++) { unsigned int pkt_size, entry = tp->cur_rx % NUM_RX_DESC; struct RxDesc *desc = tp->RxDescArray + entry; struct sk_buff *skb; const void *rx_buf; dma_addr_t addr; u32 status; → status = le32_to_cpu(desc->opts1); if (status & DescOwn) break; /* This barrier is needed to keep us from reading * any other fields out of the Rx descriptor until * we know the status of DescOwn */ dma_rmb(); if (unlikely(status & RxRES)) { . . . The reason isn't obvious, so it might be interesting if this is a valid report and whether it caused spurious corruption of the network data on Realtek 8169 compatible cards ...I think this is pretty much expected. Driver reads a piece of memory that the hardware can modify. Adding data_race() annotations could avoid these false positives.quoted
Hope this helps. Best regards, Mirsad TodorovacWell, another approach was this quick fix that eliminated all those rtl8169_poll() KCSAN warnings. If READ_ONCE(desc->opts1) fixed it, then maybe there is more to this than meets the eye? ------------------------------------------------- drivers/net/ethernet/realtek/r8169_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index 6351a2dc13bc..051551ee2a15 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c@@ -4427,7 +4427,7 @@ static int rtl_rx(struct net_device *dev, struct rtl8169_private *tp, int budget dma_addr_t addr; u32 status; - status = le32_to_cpu(desc->opts1); + status = le32_to_cpu(READ_ONCE(desc->opts1)); if (status & DescOwn) break;This is also working, but in this case we already have barriers ( dma_rmb() here) to synchronize host side and hardware (when flipping DescOwn) bit.READ_ONCE() does not imply any (strong) barriers (it does imply address-dependency barriers, i.e. ordering dependent reads/writes, but if that can be relied upon if the concurrent writer is a device and not CPU I don't know). So in this case pairing READ_ONCE() with dma_rmb() is perfectly reasonable: writes to desc->opts1 can happen concurrently, and the READ_ONCE() ensures the compiler doesn't mess up that access; later reads must be ordered by dma_rmb(). The data race here is a consequence of a concurrent write with the read of desc->opts1. The dma_rmb() does not prevent that from happening, and therefore we still have to mark the racing access.
Hi,
As I understood, this patch stops the KCSAN data-race warning for this line, so adding data_race() annotation
is not needed?
Should we proceed to a formal patch?
1. BTW, there is a similar data-race in the rtl8169_poll, as I already have your attention:
[ 707.765194] ==================================================================
[ 707.765207] BUG: KCSAN: data-race in rtl8169_poll (/home/marvin/linux/kernel/torvalds2/drivers/net/ethernet/realtek/r8169_main.c:4368 /home/marvin/linux/kernel/torvalds2/drivers/net/ethernet/realtek/r8169_main.c:4581) r8169
[ 707.765278] race at unknown origin, with read to 0xffff888144acf5e0 of 4 bytes by interrupt on cpu 21:
[ 707.765290] rtl8169_poll (/home/marvin/linux/kernel/torvalds2/drivers/net/ethernet/realtek/r8169_main.c:4368 /home/marvin/linux/kernel/torvalds2/drivers/net/ethernet/realtek/r8169_main.c:4581) r8169
[ 707.765353] __napi_poll (/home/marvin/linux/kernel/torvalds2/net/core/dev.c:6527)
[ 707.765367] net_rx_action (/home/marvin/linux/kernel/torvalds2/net/core/dev.c:6596 /home/marvin/linux/kernel/torvalds2/net/core/dev.c:6727)
[ 707.765380] __do_softirq (/home/marvin/linux/kernel/torvalds2/kernel/softirq.c:553)
[ 707.765393] do_softirq.part.0 (/home/marvin/linux/kernel/torvalds2/kernel/softirq.c:454)
[ 707.765403] __local_bh_enable_ip (/home/marvin/linux/kernel/torvalds2/./arch/x86/include/asm/preempt.h:27 /home/marvin/linux/kernel/torvalds2/kernel/softirq.c:381)
[ 707.765413] __dev_queue_xmit (/home/marvin/linux/kernel/torvalds2/net/core/dev.c:4368)
[ 707.765425] ip_finish_output2 (/home/marvin/linux/kernel/torvalds2/./include/linux/netdevice.h:3082 /home/marvin/linux/kernel/torvalds2/./include/net/neighbour.h:526 /home/marvin/linux/kernel/torvalds2/./include/net/neighbour.h:540 /home/marvin/linux/kernel/torvalds2/net/ipv4/ip_output.c:233)
[ 707.765436] __ip_finish_output (/home/marvin/linux/kernel/torvalds2/net/ipv4/ip_output.c:311 /home/marvin/linux/kernel/torvalds2/net/ipv4/ip_output.c:293)
[ 707.765447] ip_finish_output (/home/marvin/linux/kernel/torvalds2/net/ipv4/ip_output.c:328)
[ 707.765458] ip_output (/home/marvin/linux/kernel/torvalds2/net/ipv4/ip_output.c:435)
[ 707.765468] ip_send_skb (/home/marvin/linux/kernel/torvalds2/./include/net/dst.h:458 /home/marvin/linux/kernel/torvalds2/net/ipv4/ip_output.c:127 /home/marvin/linux/kernel/torvalds2/net/ipv4/ip_output.c:1486)
[ 707.765478] udp_send_skb (/home/marvin/linux/kernel/torvalds2/net/ipv4/udp.c:963)
[ 707.765491] udp_sendmsg (/home/marvin/linux/kernel/torvalds2/net/ipv4/udp.c:1246)
[ 707.765501] inet_sendmsg (/home/marvin/linux/kernel/torvalds2/net/ipv4/af_inet.c:840 (discriminator 4))
[ 707.765511] sock_sendmsg (/home/marvin/linux/kernel/torvalds2/net/socket.c:730 /home/marvin/linux/kernel/torvalds2/net/socket.c:753)
[ 707.765525] __sys_sendto (/home/marvin/linux/kernel/torvalds2/net/socket.c:2177)
[ 707.765535] __x64_sys_sendto (/home/marvin/linux/kernel/torvalds2/net/socket.c:2185)
[ 707.765545] do_syscall_64 (/home/marvin/linux/kernel/torvalds2/arch/x86/entry/common.c:50 /home/marvin/linux/kernel/torvalds2/arch/x86/entry/common.c:80)
[ 707.765558] entry_SYSCALL_64_after_hwframe (/home/marvin/linux/kernel/torvalds2/arch/x86/entry/entry_64.S:120)
[ 707.765574] value changed: 0xb000004a -> 0x00000000
[ 707.765588] Reported by Kernel Concurrency Sanitizer on:
[ 707.765596] CPU: 21 PID: 3816 Comm: Chrome_ChildIOT Not tainted 6.6.0-rc2-kcsan-00003-g16819584c239-dirty #11
[ 707.765608] Hardware name: ASRock X670E PG Lightning/X670E PG Lightning, BIOS 1.21 04/26/2023
[ 707.765615] ==================================================================
This one does not seem so easy to patch:
drivers/net/ethernet/realtek/r8169_main.c:4363-4380:
----------------------------------------------------
while (READ_ONCE(tp->cur_tx) != dirty_tx) {
unsigned int entry = dirty_tx % NUM_TX_DESC;
u32 status;
status = le32_to_cpu(tp->TxDescArray[entry].opts1);
→ if (status & DescOwn)
break;
skb = tp->tx_skb[entry].skb;
rtl8169_unmap_tx_skb(tp, entry);
if (skb) {
pkts_compl++;
bytes_compl += skb->len;
napi_consume_skb(skb, budget);
}
dirty_tx++;
}
------------------------------------------------------
drivers/net/ethernet/realtek/r8169_main.c:4363-4380:
------------------------------------------------------
static int rtl8169_poll(struct napi_struct *napi, int budget)
{
struct rtl8169_private *tp = container_of(napi, struct rtl8169_private, napi);
struct net_device *dev = tp->dev;
int work_done;
→ rtl_tx(dev, tp, budget);
work_done = rtl_rx(dev, tp, budget);
if (work_done < budget && napi_complete_done(napi, work_done))
rtl_irq_enable(tp);
return work_done;
}
------------------------------------------------------
2. The similar data race is here:
[ 8953.703581] ==================================================================
[ 8953.703653] BUG: KCSAN: data-race in rtl8169_poll [r8169] / rtl8169_start_xmit [r8169]
[ 8953.703778] write (marked) to 0xffff88811a274b74 of 4 bytes by task 4011 on cpu 0:
[ 8953.703789] rtl8169_start_xmit (/home/marvin/linux/kernel/torvalds2/drivers/net/ethernet/realtek/r8169_main.c:4254) r8169
[ 8953.703852] dev_hard_start_xmit (/home/marvin/linux/kernel/torvalds2/./include/linux/netdevice.h:4889 /home/marvin/linux/kernel/torvalds2/./include/linux/netdevice.h:4903 /home/marvin/linux/kernel/torvalds2/net/core/dev.c:3544 /home/marvin/linux/kernel/torvalds2/net/core/dev.c:3560)
[ 8953.703866] sch_direct_xmit (/home/marvin/linux/kernel/torvalds2/net/sched/sch_generic.c:342)
[ 8953.703879] __dev_queue_xmit (/home/marvin/linux/kernel/torvalds2/net/core/dev.c:3817 /home/marvin/linux/kernel/torvalds2/net/core/dev.c:4306)
[ 8953.703891] ip_finish_output2 (/home/marvin/linux/kernel/torvalds2/./include/linux/netdevice.h:3082 /home/marvin/linux/kernel/torvalds2/./include/net/neighbour.h:526 /home/marvin/linux/kernel/torvalds2/./include/net/neighbour.h:540 /home/marvin/linux/kernel/torvalds2/net/ipv4/ip_output.c:233)
[ 8953.703903] __ip_finish_output (/home/marvin/linux/kernel/torvalds2/net/ipv4/ip_output.c:311 /home/marvin/linux/kernel/torvalds2/net/ipv4/ip_output.c:293)
[ 8953.703914] ip_finish_output (/home/marvin/linux/kernel/torvalds2/net/ipv4/ip_output.c:328)
[ 8953.703925] ip_output (/home/marvin/linux/kernel/torvalds2/net/ipv4/ip_output.c:435)
[ 8953.703935] ip_send_skb (/home/marvin/linux/kernel/torvalds2/./include/net/dst.h:458 /home/marvin/linux/kernel/torvalds2/net/ipv4/ip_output.c:127 /home/marvin/linux/kernel/torvalds2/net/ipv4/ip_output.c:1486)
[ 8953.703945] udp_send_skb (/home/marvin/linux/kernel/torvalds2/net/ipv4/udp.c:963)
[ 8953.703958] udp_sendmsg (/home/marvin/linux/kernel/torvalds2/net/ipv4/udp.c:1246)
[ 8953.703968] inet_sendmsg (/home/marvin/linux/kernel/torvalds2/net/ipv4/af_inet.c:840 (discriminator 4))
[ 8953.703978] sock_sendmsg (/home/marvin/linux/kernel/torvalds2/net/socket.c:730 /home/marvin/linux/kernel/torvalds2/net/socket.c:753)
[ 8953.703992] __sys_sendto (/home/marvin/linux/kernel/torvalds2/net/socket.c:2177)
[ 8953.704002] __x64_sys_sendto (/home/marvin/linux/kernel/torvalds2/net/socket.c:2185)
[ 8953.704012] do_syscall_64 (/home/marvin/linux/kernel/torvalds2/arch/x86/entry/common.c:50 /home/marvin/linux/kernel/torvalds2/arch/x86/entry/common.c:80)
[ 8953.704025] entry_SYSCALL_64_after_hwframe (/home/marvin/linux/kernel/torvalds2/arch/x86/entry/entry_64.S:120)
[ 8953.704042] read to 0xffff88811a274b74 of 4 bytes by interrupt on cpu 21:
[ 8953.704053] rtl8169_poll (/home/marvin/linux/kernel/torvalds2/drivers/net/ethernet/realtek/r8169_main.c:4397 /home/marvin/linux/kernel/torvalds2/drivers/net/ethernet/realtek/r8169_main.c:4581) r8169
[ 8953.704114] __napi_poll (/home/marvin/linux/kernel/torvalds2/net/core/dev.c:6527)
[ 8953.704127] net_rx_action (/home/marvin/linux/kernel/torvalds2/net/core/dev.c:6596 /home/marvin/linux/kernel/torvalds2/net/core/dev.c:6727)
[ 8953.704140] __do_softirq (/home/marvin/linux/kernel/torvalds2/kernel/softirq.c:553)
[ 8953.704152] __irq_exit_rcu (/home/marvin/linux/kernel/torvalds2/kernel/softirq.c:427 /home/marvin/linux/kernel/torvalds2/kernel/softirq.c:632)
[ 8953.704163] irq_exit_rcu (/home/marvin/linux/kernel/torvalds2/kernel/softirq.c:647)
[ 8953.704172] sysvec_apic_timer_interrupt (/home/marvin/linux/kernel/torvalds2/arch/x86/kernel/apic/apic.c:1074 (discriminator 14))
[ 8953.704186] asm_sysvec_apic_timer_interrupt (/home/marvin/linux/kernel/torvalds2/./arch/x86/include/asm/idtentry.h:645)
[ 8953.704197] cpuidle_enter_state (/home/marvin/linux/kernel/torvalds2/drivers/cpuidle/cpuidle.c:291)
[ 8953.704207] cpuidle_enter (/home/marvin/linux/kernel/torvalds2/drivers/cpuidle/cpuidle.c:390)
[ 8953.704217] call_cpuidle (/home/marvin/linux/kernel/torvalds2/kernel/sched/idle.c:135)
[ 8953.704229] do_idle (/home/marvin/linux/kernel/torvalds2/kernel/sched/idle.c:219 /home/marvin/linux/kernel/torvalds2/kernel/sched/idle.c:282)
[ 8953.704238] cpu_startup_entry (/home/marvin/linux/kernel/torvalds2/kernel/sched/idle.c:378 (discriminator 1))
[ 8953.704247] start_secondary (/home/marvin/linux/kernel/torvalds2/arch/x86/kernel/smpboot.c:210 /home/marvin/linux/kernel/torvalds2/arch/x86/kernel/smpboot.c:294)
[ 8953.704260] secondary_startup_64_no_verify (/home/marvin/linux/kernel/torvalds2/arch/x86/kernel/head_64.S:433)
[ 8953.704279] value changed: 0x0002f84c -> 0x0002f84d
[ 8953.704293] Reported by Kernel Concurrency Sanitizer on:
[ 8953.704301] CPU: 21 PID: 0 Comm: swapper/21 Tainted: G L 6.6.0-rc2-kcsan-00003-g16819584c239-dirty #16
[ 8953.704313] Hardware name: ASRock X670E PG Lightning/X670E PG Lightning, BIOS 1.21 04/26/2023
[ 8953.704320] ==================================================================
I am unable to say if this could spuriously corrupt data and/or cause dropped packets which are probably
fixed by the upper level protocols?
4355 static void rtl_tx(struct net_device *dev, struct rtl8169_private *tp,
4356 int budget)
4357 {
4358 unsigned int dirty_tx, bytes_compl = 0, pkts_compl = 0;
4359 struct sk_buff *skb;
4360
4361 dirty_tx = tp->dirty_tx;
4362
4363 while (READ_ONCE(tp->cur_tx) != dirty_tx) {
4364 unsigned int entry = dirty_tx % NUM_TX_DESC;
4365 u32 status;
4366
→ 4367 status = le32_to_cpu(READ_ONCE(tp->TxDescArray[entry].opts1));
4368 if (status & DescOwn)
4369 break;
4370
4371 skb = tp->tx_skb[entry].skb;
4372 rtl8169_unmap_tx_skb(tp, entry);
4373
4374 if (skb) {
4375 pkts_compl++;
4376 bytes_compl += skb->len;
4377 napi_consume_skb(skb, budget);
4378 }
4379 dirty_tx++;
4380 }
4381
4382 if (tp->dirty_tx != dirty_tx) {
4383 dev_sw_netstats_tx_add(dev, pkts_compl, bytes_compl);
4384 WRITE_ONCE(tp->dirty_tx, dirty_tx);
4385
4386 netif_subqueue_completed_wake(dev, 0, pkts_compl, bytes_compl,
4387 rtl_tx_slots_avail(tp),
4388 R8169_TX_START_THRS);
4389 /*
4390 * 8168 hack: TxPoll requests are lost when the Tx packets are
4391 * too close. Let's kick an extra TxPoll request when a burst
4392 * of start_xmit activity is detected (if it is not detected,
4393 * it is slow enough). -- FR
4394 * If skb is NULL then we come here again once a tx irq is
4395 * triggered after the last fragment is marked transmitted.
4396 */
→ 4397 if (tp->cur_tx != dirty_tx && skb)
4398 rtl8169_doorbell(tp);
4399 }
4400 }
The conflicting code is:
4251 /* rtl_tx needs to see descriptor changes before updated tp->cur_tx */
4252 smp_wmb();
4253
→ 4254 WRITE_ONCE(tp->cur_tx, tp->cur_tx + frags + 1);
4255
All of those are being patched with READ_ONCE() around the variable that is accessed concurrently:
------------------------------------------------diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 6351a2dc13bc..361b90007148 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c@@ -4364,7 +4364,7 @@ static void rtl_tx(struct net_device *dev, struct rtl8169_private *tp, unsigned int entry = dirty_tx % NUM_TX_DESC; u32 status; - status = le32_to_cpu(tp->TxDescArray[entry].opts1); + status = le32_to_cpu(READ_ONCE(tp->TxDescArray[entry].opts1)); if (status & DescOwn) break;
@@ -4394,7 +4394,7 @@ static void rtl_tx(struct net_device *dev, struct rtl8169_private *tp, * If skb is NULL then we come here again once a tx irq is * triggered after the last fragment is marked transmitted. */ - if (tp->cur_tx != dirty_tx && skb) + if (READ_ONCE(tp->cur_tx) != dirty_tx && skb) rtl8169_doorbell(tp); } }
@@ -4427,7 +4427,7 @@ static int rtl_rx(struct net_device *dev, struct rtl8169_private *tp, int budget dma_addr_t addr; u32 status; - status = le32_to_cpu(desc->opts1); + status = le32_to_cpu(READ_ONCE(desc->opts1)); if (status & DescOwn) break; ---
Just like in line 4363, tp->cur_tx in line 4397 is prone to data-race unless enclosed in READ_ONCE(). There are no additional data-races reported by the KCSAN for rtl8169 driver ATM. Best regards, Mirsad Todorovac