Skip to contentSkip to conversation

6.1.54-rc1 review

PATCH In review Greg Kroah-Hartman 19 participants 259 messages Sep 17, 2023 – Oct 9, 2023 Page 2 of 2

Subsystems (5)
lore
Share & tools
lore
author maintainer
#143 lore raw

PATCH 115/219 igc: Change IGC_MIN to allow set rx/tx value between 64 and 80

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Olga Zaborska <[email protected]>

[ Upstream commit 5aa48279712e1f134aac908acde4df798955a955 ]

Change the minimum value of RX/TX descriptors to 64 to enable setting the rx/tx value between 64 and 80. All igc devices can use as low as 64 descriptors. This change will unify igc with other drivers.
Based on commit 7b1be1987c1e ("e1000e: lower ring minimum size to 64")

Fixes 0507ef8a0372 ("igc: Add transmit and receive fastpath and interrupt handlers")
---
 drivers/net/ethernet/intel/igc/igc.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/intel/igc/igc.h +2 -2
index f83cbc4a1afa8..d3b17aa1d1a83 100644
@@ -354,11 +354,11 @@ static inline u32 igc_rss_type(const union igc_adv_rx_desc *rx_desc)
 /* TX/RX descriptor defines */
 #define IGC_DEFAULT_TXD		256
 #define IGC_DEFAULT_TX_WORK	128
-#define IGC_MIN_TXD		80
+#define IGC_MIN_TXD		64
 #define IGC_MAX_TXD		4096
 
 #define IGC_DEFAULT_RXD		256
-#define IGC_MIN_RXD		80
+#define IGC_MIN_RXD		64
 #define IGC_MAX_RXD		4096
 
 /* Supported Rx Buffer Sizes */
Signature · 1 line
2.40.1
author maintainer
#144 lore raw

PATCH 116/219 igbvf: Change IGBVF_MIN to allow set rx/tx value between 64 and 80

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Olga Zaborska <[email protected]>

[ Upstream commit 8360717524a24a421c36ef8eb512406dbd42160a ]

Change the minimum value of RX/TX descriptors to 64 to enable setting the rx/tx value between 64 and 80. All igbvf devices can use as low as 64 descriptors. This change will unify igbvf with other drivers.
Based on commit 7b1be1987c1e ("e1000e: lower ring minimum size to 64")

Fixes d4e0fe01a38a ("igbvf: add new driver to support 82576 virtual functions")
---
 drivers/net/ethernet/intel/igbvf/igbvf.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/intel/igbvf/igbvf.h +2 -2
index 57d39ee00b585..7b83678ba83a6 100644
@@ -39,11 +39,11 @@ enum latency_range {
 /* Tx/Rx descriptor defines */
 #define IGBVF_DEFAULT_TXD	256
 #define IGBVF_MAX_TXD		4096
-#define IGBVF_MIN_TXD		80
+#define IGBVF_MIN_TXD		64
 
 #define IGBVF_DEFAULT_RXD	256
 #define IGBVF_MAX_RXD		4096
-#define IGBVF_MIN_RXD		80
+#define IGBVF_MIN_RXD		64
 
 #define IGBVF_MIN_ITR_USECS	10 /* 100000 irq/sec */
 #define IGBVF_MAX_ITR_USECS	10000 /* 100    irq/sec */
Signature · 1 line
2.40.1
author maintainer
#145 lore raw

PATCH 117/219 igb: Change IGB_MIN to allow set rx/tx value between 64 and 80

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Olga Zaborska <[email protected]>

[ Upstream commit 6319685bdc8ad5310890add907b7c42f89302886 ]

Change the minimum value of RX/TX descriptors to 64 to enable setting the rx/tx value between 64 and 80. All igb devices can use as low as 64 descriptors. This change will unify igb with other drivers.
Based on commit 7b1be1987c1e ("e1000e: lower ring minimum size to 64")

Fixes 9d5c824399de ("igb: PCI-Express 82575 Gigabit Ethernet driver")
---
 drivers/net/ethernet/intel/igb/igb.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/intel/igb/igb.h +2 -2
index 015b781441149..a2b759531cb7b 100644
@@ -34,11 +34,11 @@ struct igb_adapter;
 /* TX/RX descriptor defines */
 #define IGB_DEFAULT_TXD		256
 #define IGB_DEFAULT_TX_WORK	128
-#define IGB_MIN_TXD		80
+#define IGB_MIN_TXD		64
 #define IGB_MAX_TXD		4096
 
 #define IGB_DEFAULT_RXD		256
-#define IGB_MIN_RXD		80
+#define IGB_MIN_RXD		64
 #define IGB_MAX_RXD		4096
 
 #define IGB_DEFAULT_ITR		3 /* dynamic */
Signature · 1 line
2.40.1
author maintainer
#146 lore raw

PATCH 118/219 s390/zcrypt: dont leak memory if dev_set_name() fails

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andy Shevchenko <[email protected]>

[ Upstream commit 6252f47b78031979ad919f971dc8468b893488bd ]

When dev_set_name() fails, zcdn_create() doesn't free the newly allocated resources. Do it.

Fixes 00fab2350e6b ("s390/zcrypt: multiple zcrypt device nodes support")
---
 drivers/s390/crypto/zcrypt_api.c | 1 +
 1 file changed, 1 insertion(+)
drivers/s390/crypto/zcrypt_api.c +1 -0
index f94b43ce9a658..28e34d155334b 100644
@@ -441,6 +441,7 @@ static int zcdn_create(const char *name)
 			 ZCRYPT_NAME "_%d", (int)MINOR(devt));
 	nodename[sizeof(nodename) - 1] = '\0';
 	if (dev_set_name(&zcdndev->device, nodename)) {
+		kfree(zcdndev);
 		rc = -EINVAL;
 		goto unlockout;
 	}
Signature · 1 line
2.40.1
author maintainer
#147 lore raw

PATCH 119/219 idr: fix param name in idr_alloc_cyclic() doc

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ariel Marcovitch <[email protected]>

[ Upstream commit 2a15de80dd0f7e04a823291aa9eb49c5294f56af ]

The relevant parameter is 'start' and not 'nextid'

Fixes 460488c58ca8 ("idr: Remove idr_alloc_ext")
---
 lib/idr.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
lib/idr.c +1 -1
index 7ecdfdb5309e7..13f2758c23773 100644
@@ -100,7 +100,7 @@ EXPORT_SYMBOL_GPL(idr_alloc);
  * @end: The maximum ID (exclusive).
  * @gfp: Memory allocation flags.
  *
- * Allocates an unused ID in the range specified by @nextid and @end.  If
+ * Allocates an unused ID in the range specified by @start and @end.  If
  * @end is <= 0, it is treated as one larger than %INT_MAX.  This allows
  * callers to use @start + N as @end as long as N is within integer range.
  * The search for an unused ID will start at the last ID allocated and will
Signature · 1 line
2.40.1
author maintainer
#148 lore raw

PATCH 120/219 ip_tunnels: use DEV_STATS_INC()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit 9b271ebaf9a2c5c566a54bc6cd915962e8241130 ]

syzbot/KCSAN reported data-races in iptunnel_xmit_stats() [1]

This can run from multiple cpus without mutual exclusion.

Adopt SMP safe DEV_STATS_INC() to update dev->stats fields.

[1]

BUG: KCSAN: data-race in iptunnel_xmit / iptunnel_xmit

read-write to 0xffff8881353df170 of 8 bytes by task 30263 on cpu 1:

iptunnel_xmit_stats include/net/ip_tunnels.h:493 [inline]
iptunnel_xmit+0x432/0x4a0 net/ipv4/ip_tunnel_core.c:87
ip_tunnel_xmit+0x1477/0x1750 net/ipv4/ip_tunnel.c:831
__gre_xmit net/ipv4/ip_gre.c:469 [inline]
ipgre_xmit+0x516/0x570 net/ipv4/ip_gre.c:662
__netdev_start_xmit include/linux/netdevice.h:4889 [inline]
netdev_start_xmit include/linux/netdevice.h:4903 [inline]
xmit_one net/core/dev.c:3544 [inline]
dev_hard_start_xmit+0x11b/0x3f0 net/core/dev.c:3560
__dev_queue_xmit+0xeee/0x1de0 net/core/dev.c:4340
dev_queue_xmit include/linux/netdevice.h:3082 [inline]
__bpf_tx_skb net/core/filter.c:2129 [inline]
__bpf_redirect_no_mac net/core/filter.c:2159 [inline]
__bpf_redirect+0x723/0x9c0 net/core/filter.c:2182
____bpf_clone_redirect net/core/filter.c:2453 [inline]
bpf_clone_redirect+0x16c/0x1d0 net/core/filter.c:2425
___bpf_prog_run+0xd7d/0x41e0 kernel/bpf/core.c:1954
__bpf_prog_run512+0x74/0xa0 kernel/bpf/core.c:2195
bpf_dispatcher_nop_func include/linux/bpf.h:1181 [inline]
__bpf_prog_run include/linux/filter.h:609 [inline]
bpf_prog_run include/linux/filter.h:616 [inline]
bpf_test_run+0x15d/0x3d0 net/bpf/test_run.c:423
bpf_prog_test_run_skb+0x77b/0xa00 net/bpf/test_run.c:1045
bpf_prog_test_run+0x265/0x3d0 kernel/bpf/syscall.c:3996
__sys_bpf+0x3af/0x780 kernel/bpf/syscall.c:5353
__do_sys_bpf kernel/bpf/syscall.c:5439 [inline]
__se_sys_bpf kernel/bpf/syscall.c:5437 [inline]
__x64_sys_bpf+0x43/0x50 kernel/bpf/syscall.c:5437
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

read-write to 0xffff8881353df170 of 8 bytes by task 30249 on cpu 0:

iptunnel_xmit_stats include/net/ip_tunnels.h:493 [inline]
iptunnel_xmit+0x432/0x4a0 net/ipv4/ip_tunnel_core.c:87
ip_tunnel_xmit+0x1477/0x1750 net/ipv4/ip_tunnel.c:831
__gre_xmit net/ipv4/ip_gre.c:469 [inline]
ipgre_xmit+0x516/0x570 net/ipv4/ip_gre.c:662
__netdev_start_xmit include/linux/netdevice.h:4889 [inline]
netdev_start_xmit include/linux/netdevice.h:4903 [inline]
xmit_one net/core/dev.c:3544 [inline]
dev_hard_start_xmit+0x11b/0x3f0 net/core/dev.c:3560
__dev_queue_xmit+0xeee/0x1de0 net/core/dev.c:4340
dev_queue_xmit include/linux/netdevice.h:3082 [inline]
__bpf_tx_skb net/core/filter.c:2129 [inline]
__bpf_redirect_no_mac net/core/filter.c:2159 [inline]
__bpf_redirect+0x723/0x9c0 net/core/filter.c:2182
____bpf_clone_redirect net/core/filter.c:2453 [inline]
bpf_clone_redirect+0x16c/0x1d0 net/core/filter.c:2425
___bpf_prog_run+0xd7d/0x41e0 kernel/bpf/core.c:1954
__bpf_prog_run512+0x74/0xa0 kernel/bpf/core.c:2195
bpf_dispatcher_nop_func include/linux/bpf.h:1181 [inline]
__bpf_prog_run include/linux/filter.h:609 [inline]
bpf_prog_run include/linux/filter.h:616 [inline]
bpf_test_run+0x15d/0x3d0 net/bpf/test_run.c:423
bpf_prog_test_run_skb+0x77b/0xa00 net/bpf/test_run.c:1045
bpf_prog_test_run+0x265/0x3d0 kernel/bpf/syscall.c:3996
__sys_bpf+0x3af/0x780 kernel/bpf/syscall.c:5353
__do_sys_bpf kernel/bpf/syscall.c:5439 [inline]
__se_sys_bpf kernel/bpf/syscall.c:5437 [inline]
__x64_sys_bpf+0x43/0x50 kernel/bpf/syscall.c:5437
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

value changed: 0x0000000000018830 -> 0x0000000000018831

Reported by Kernel Concurrency Sanitizer on:

CPU: 0 PID: 30249 Comm: syz-executor.4 Not tainted 6.5.0-syzkaller-11704-g3f86ed6ec0b3 #0
Fixes 039f50629b7f ("ip_tunnel: Move stats update to iptunnel_xmit()")
---
 include/net/ip_tunnels.h | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)
include/net/ip_tunnels.h +7 -8
index fca3576798166..bca80522f95c8 100644
@@ -473,15 +473,14 @@ static inline void iptunnel_xmit_stats(struct net_device *dev, int pkt_len)
 		u64_stats_inc(&tstats->tx_packets);
 		u64_stats_update_end(&tstats->syncp);
 		put_cpu_ptr(tstats);
+		return;
+	}
+
+	if (pkt_len < 0) {
+		DEV_STATS_INC(dev, tx_errors);
+		DEV_STATS_INC(dev, tx_aborted_errors);
 	} else {
-		struct net_device_stats *err_stats = &dev->stats;
-
-		if (pkt_len < 0) {
-			err_stats->tx_errors++;
-			err_stats->tx_aborted_errors++;
-		} else {
-			err_stats->tx_dropped++;
-		}
+		DEV_STATS_INC(dev, tx_dropped);
 	}
 }
 
Signature · 1 line
2.40.1
author maintainer
#149 lore raw

PATCH 121/219 net: dsa: sja1105: fix bandwidth discrepancy between tc-cbs software and offload

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vladimir Oltean <[email protected]>

[ Upstream commit 954ad9bf13c4f95a4958b5f8433301f2ab99e1f5 ]

More careful measurement of the tc-cbs bandwidth shows that the stream bandwidth (effectively idleslope) increases, there is a larger and larger discrepancy between the rate limit obtained by the software Qdisc, and the rate limit obtained by its offloaded counterpart.

The discrepancy becomes so large, that e.g. at an idleslope of 40000 (40Mbps), the offloaded cbs does not actually rate limit anything, and traffic will pass at line rate through a 100 Mbps port.

The reason for the discrepancy is that the hardware documentation I've been following is incorrect. UM11040.pdf (for SJA1105P/Q/R/S) states about IDLE_SLOPE that it is "the rate (in unit of bytes/sec) at which the credit counter is increased".

Cross-checking with UM10944.pdf (for SJA1105E/T) and UM11107.pdf (for SJA1110), the wording is different: "This field specifies the value, in bytes per second times link speed, by which the credit counter is increased".

So there's an extra scaling for link speed that the driver is currently not accounting for, and apparently (empirically), that link speed is expressed in Kbps.

I've pondered whether to pollute the sja1105_mac_link_up() implementation with CBS shaper reprogramming, but I don't think it is worth it. IMO, the UAPI exposed by tc-cbs requires user space to recalculate the sendslope anyway, since the formula for that depends on port_transmit_rate (see man tc-cbs), which is not an invariant from tc's perspective.

So we use the offload->sendslope and offload->idleslope to deduce the original port_transmit_rate from the CBS formula, and use that value to scale the offload->sendslope and offload->idleslope to values that the hardware understands.

Some numerical data points:

40Mbps stream, max interfering frame size 1500, port speed 100M
---------------------------------------------------------------

tc-cbs parameters:
idleslope 40000 sendslope -60000 locredit -900 hicredit 600

which result in hardware values:

 Before (doesn't work)           After (works)
 credit_hi    600                600
 credit_lo    900                900
 send_slope   7500000            75
 idle_slope   5000000            50

40Mbps stream, max interfering frame size 1500, port speed 1G
-------------------------------------------------------------

tc-cbs parameters:
idleslope 40000 sendslope -960000 locredit -1440 hicredit 60

which result in hardware values:

 Before (doesn't work)           After (works)
 credit_hi    60                 60
 credit_lo    1440               1440
 send_slope   120000000          120
 idle_slope   5000000            5

5.12Mbps stream, max interfering frame size 1522, port speed 100M
-----------------------------------------------------------------

tc-cbs parameters:
idleslope 5120 sendslope -94880 locredit -1444 hicredit 77

which result in hardware values:

 Before (doesn't work)           After (works)
 credit_hi    77                 77
 credit_lo    1444               1444
 send_slope   11860000           118
 idle_slope   640000             6

Tested on SJA1105T, SJA1105S and SJA1110A, at 1Gbps and 100Mbps.

Fixes 4d7525085a9b ("net: dsa: sja1105: offload the Credit-Based Shaper qdisc")
---
 drivers/net/dsa/sja1105/sja1105_main.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)
drivers/net/dsa/sja1105/sja1105_main.c +12 -3
index 947e8f7c09880..377f177502003 100644
@@ -2157,6 +2157,7 @@ static int sja1105_setup_tc_cbs(struct dsa_switch *ds, int port,
 {
 	struct sja1105_private *priv = ds->priv;
 	struct sja1105_cbs_entry *cbs;
+	s64 port_transmit_rate_kbps;
 	int index;
 
 	if (!offload->enable)
@@ -2174,9 +2175,17 @@ static int sja1105_setup_tc_cbs(struct dsa_switch *ds, int port,
 	 */
 	cbs->credit_hi = offload->hicredit;
 	cbs->credit_lo = abs(offload->locredit);
-	/* User space is in kbits/sec, hardware in bytes/sec */
-	cbs->idle_slope = offload->idleslope * BYTES_PER_KBIT;
-	cbs->send_slope = abs(offload->sendslope * BYTES_PER_KBIT);
+	/* User space is in kbits/sec, while the hardware in bytes/sec times
+	 * link speed. Since the given offload->sendslope is good only for the
+	 * current link speed anyway, and user space is likely to reprogram it
+	 * when that changes, don't even bother to track the port's link speed,
+	 * but deduce the port transmit rate from idleslope - sendslope.
+	 */
+	port_transmit_rate_kbps = offload->idleslope - offload->sendslope;
+	cbs->idle_slope = div_s64(offload->idleslope * BYTES_PER_KBIT,
+				  port_transmit_rate_kbps);
+	cbs->send_slope = div_s64(abs(offload->sendslope * BYTES_PER_KBIT),
+				  port_transmit_rate_kbps);
 	/* Convert the negative values from 64-bit 2's complement
 	 * to 32-bit 2's complement (for the case of 0x80000000 whose
 	 * negative is still negative).
Signature · 1 line
2.40.1
author maintainer
#150 lore raw

PATCH 122/219 net: dsa: sja1105: fix -ENOSPC when replacing the same tc-cbs too many times

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vladimir Oltean <[email protected]>

[ Upstream commit 894cafc5c62ccced758077bd4e970dc714c42637 ]

After running command [2] too many times in a row:

[1] $ tc qdisc add dev sw2p0 root handle 1: mqprio num_tc 8 \
	map 0 1 2 3 4 5 6 7 queues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7 hw 0
[2] $ tc qdisc replace dev sw2p0 parent 1:1 cbs offload 1 \
	idleslope 120000 sendslope -880000 locredit -1320 hicredit 180

(aka more than priv->info->num_cbs_shapers times)

we start seeing the following error message:

Error: Specified device failed to setup cbs hardware offload.

This comes from the fact that ndo_setup_tc(TC_SETUP_QDISC_CBS) presents the same API for the qdisc create and replace cases, and the sja1105 driver fails to distinguish between the 2. Thus, it always thinks that it must allocate the same shaper for a {port, queue} pair, when it may instead have to replace an existing one.

Fixes 4d7525085a9b ("net: dsa: sja1105: offload the Credit-Based Shaper qdisc")
---
 drivers/net/dsa/sja1105/sja1105_main.c | 23 ++++++++++++++++++++---
 1 file changed, 20 insertions(+), 3 deletions(-)
drivers/net/dsa/sja1105/sja1105_main.c +20 -3
index 377f177502003..9dd5cdcda2843 100644
@@ -2123,6 +2123,18 @@ static void sja1105_bridge_leave(struct dsa_switch *ds, int port,
 
 #define BYTES_PER_KBIT (1000LL / 8)
 
+static int sja1105_find_cbs_shaper(struct sja1105_private *priv,
+				   int port, int prio)
+{
+	int i;
+
+	for (i = 0; i < priv->info->num_cbs_shapers; i++)
+		if (priv->cbs[i].port == port && priv->cbs[i].prio == prio)
+			return i;
+
+	return -1;
+}
+
 static int sja1105_find_unused_cbs_shaper(struct sja1105_private *priv)
 {
 	int i;
@@ -2163,9 +2175,14 @@ static int sja1105_setup_tc_cbs(struct dsa_switch *ds, int port,
 	if (!offload->enable)
 		return sja1105_delete_cbs_shaper(priv, port, offload->queue);
 
-	index = sja1105_find_unused_cbs_shaper(priv);
-	if (index < 0)
-		return -ENOSPC;
+	/* The user may be replacing an existing shaper */
+	index = sja1105_find_cbs_shaper(priv, port, offload->queue);
+	if (index < 0) {
+		/* That isn't the case - see if we can allocate a new one */
+		index = sja1105_find_unused_cbs_shaper(priv);
+		if (index < 0)
+			return -ENOSPC;
+	}
 
 	cbs = &priv->cbs[index];
 	cbs->port = port;
Signature · 1 line
2.40.1
author maintainer
#151 lore raw

PATCH 123/219 net: dsa: sja1105: complete tc-cbs offload support on SJA1110

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vladimir Oltean <[email protected]>

[ Upstream commit 180a7419fe4adc8d9c8e0ef0fd17bcdd0cf78acd ]

The blamed commit left this delta behind:

struct sja1105_cbs_entry {
- u64 port;
- u64 prio;
+ u64 port; /* Not used for SJA1110 */
+ u64 prio; /* Not used for SJA1110 */
u64 credit_hi;
u64 credit_lo;
u64 send_slope;
u64 idle_slope;
};

but did not actually implement tc-cbs offload fully for the new switch. The offload is accepted, but it doesn't work.

The difference compared to earlier switch generations is that now, the table of CBS shapers is sparse, because there are many more shapers, so the mapping between a {port, prio} and a table index is static, rather than requiring us to store the port and prio into the sja1105_cbs_entry.

So, the problem is that the code programs the CBS shaper parameters at a dynamic table index which is incorrect.

All that needs to be done for SJA1110 CBS shapers to work is to bypass the logic which allocates shapers in a dense manner, as for SJA1105, and use the fixed mapping instead.

Fixes 3e77e59bf8cf ("net: dsa: sja1105: add support for the SJA1110 switch family")
---
 drivers/net/dsa/sja1105/sja1105.h      |  2 ++
 drivers/net/dsa/sja1105/sja1105_main.c | 13 +++++++++++++
 drivers/net/dsa/sja1105/sja1105_spi.c  |  4 ++++
 3 files changed, 19 insertions(+)
drivers/net/dsa/sja1105/sja1105.h +2 -0
index fb3cd4c78faa8..a831bb0a52074 100644
@@ -132,6 +132,8 @@ struct sja1105_info {
 	int max_frame_mem;
 	int num_ports;
 	bool multiple_cascade_ports;
+	/* Every {port, TXQ} has its own CBS shaper */
+	bool fixed_cbs_mapping;
 	enum dsa_tag_protocol tag_proto;
 	const struct sja1105_dynamic_table_ops *dyn_ops;
 	const struct sja1105_table_ops *static_ops;
drivers/net/dsa/sja1105/sja1105_main.c +13 -0
index 9dd5cdcda2843..ff94c5996fafb 100644
@@ -2122,12 +2122,22 @@ static void sja1105_bridge_leave(struct dsa_switch *ds, int port,
 }
 
 #define BYTES_PER_KBIT (1000LL / 8)
+/* Port 0 (the uC port) does not have CBS shapers */
+#define SJA1110_FIXED_CBS(port, prio) ((((port) - 1) * SJA1105_NUM_TC) + (prio))
 
 static int sja1105_find_cbs_shaper(struct sja1105_private *priv,
 				   int port, int prio)
 {
 	int i;
 
+	if (priv->info->fixed_cbs_mapping) {
+		i = SJA1110_FIXED_CBS(port, prio);
+		if (i >= 0 && i < priv->info->num_cbs_shapers)
+			return i;
+
+		return -1;
+	}
+
 	for (i = 0; i < priv->info->num_cbs_shapers; i++)
 		if (priv->cbs[i].port == port && priv->cbs[i].prio == prio)
 			return i;
@@ -2139,6 +2149,9 @@ static int sja1105_find_unused_cbs_shaper(struct sja1105_private *priv)
 {
 	int i;
 
+	if (priv->info->fixed_cbs_mapping)
+		return -1;
+
 	for (i = 0; i < priv->info->num_cbs_shapers; i++)
 		if (!priv->cbs[i].idle_slope && !priv->cbs[i].send_slope)
 			return i;
drivers/net/dsa/sja1105/sja1105_spi.c +4 -0
index d3c9ad6d39d46..e6b61aef4127c 100644
@@ -781,6 +781,7 @@ const struct sja1105_info sja1110a_info = {
 	.tag_proto		= DSA_TAG_PROTO_SJA1110,
 	.can_limit_mcast_flood	= true,
 	.multiple_cascade_ports	= true,
+	.fixed_cbs_mapping	= true,
 	.ptp_ts_bits		= 32,
 	.ptpegr_ts_bytes	= 8,
 	.max_frame_mem		= SJA1110_MAX_FRAME_MEMORY,
@@ -831,6 +832,7 @@ const struct sja1105_info sja1110b_info = {
 	.tag_proto		= DSA_TAG_PROTO_SJA1110,
 	.can_limit_mcast_flood	= true,
 	.multiple_cascade_ports	= true,
+	.fixed_cbs_mapping	= true,
 	.ptp_ts_bits		= 32,
 	.ptpegr_ts_bytes	= 8,
 	.max_frame_mem		= SJA1110_MAX_FRAME_MEMORY,
@@ -881,6 +883,7 @@ const struct sja1105_info sja1110c_info = {
 	.tag_proto		= DSA_TAG_PROTO_SJA1110,
 	.can_limit_mcast_flood	= true,
 	.multiple_cascade_ports	= true,
+	.fixed_cbs_mapping	= true,
 	.ptp_ts_bits		= 32,
 	.ptpegr_ts_bytes	= 8,
 	.max_frame_mem		= SJA1110_MAX_FRAME_MEMORY,
@@ -931,6 +934,7 @@ const struct sja1105_info sja1110d_info = {
 	.tag_proto		= DSA_TAG_PROTO_SJA1110,
 	.can_limit_mcast_flood	= true,
 	.multiple_cascade_ports	= true,
+	.fixed_cbs_mapping	= true,
 	.ptp_ts_bits		= 32,
 	.ptpegr_ts_bytes	= 8,
 	.max_frame_mem		= SJA1110_MAX_FRAME_MEMORY,
Signature · 1 line
2.40.1
author maintainer
#152 lore raw

PATCH 124/219 bpf: Remove prog->active check for bpf_lsm and bpf_iter

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Martin KaFai Lau <[email protected]>

[ Upstream commit 271de525e1d7f564e88a9d212c50998b49a54476 ]

The commit 64696c40d03c ("bpf: Add __bpf_prog_{enter,exit}_struct_ops for struct_ops trampoline") removed prog->active check for struct_ops prog. The bpf_lsm
and bpf_iter is also using trampoline. Like struct_ops, the bpf_lsm
and bpf_iter have fixed hooks for the prog to attach. The
kernel does not call the same hook in a recursive way.
This patch also removes the prog->active check for
bpf_lsm and bpf_iter.

A later patch has a test to reproduce the recursion issue for a sleepable bpf_lsm program.

This patch appends the '_recur' naming to the existing enter and exit functions that track the prog->active counter. New __bpf_prog_{enter,exit}[_sleepable] function are added to skip the prog->active tracking. The '_struct_ops' version is also removed.

It also moves the decision on picking the enter and exit function to the new bpf_trampoline_{enter,exit}(). It returns the '_recur' ones for all tracing progs to use. For bpf_lsm, bpf_iter, struct_ops (no prog->active tracking after 64696c40d03c), and bpf_lsm_cgroup (no prog->active tracking after 69fd337a975c7), it will return the functions that don't track the prog->active.

Stable-dep-of: 7645629f7dc8 ("bpf: Invoke __bpf_prog_exit_sleepable_recur() on recursion in kern_sys_bpf().")

---
 arch/arm64/net/bpf_jit_comp.c |  9 +---
 arch/x86/net/bpf_jit_comp.c   | 19 +--------
 include/linux/bpf.h           | 24 +++++------
 include/linux/bpf_verifier.h  | 13 ++++++
 kernel/bpf/syscall.c          |  5 ++-
 kernel/bpf/trampoline.c       | 80 +++++++++++++++++++++++++++++------
 6 files changed, 97 insertions(+), 53 deletions(-)
arch/arm64/net/bpf_jit_comp.c +2 -7
index 14134fd34ff79..0ce5f13eabb1b 100644
@@ -1655,13 +1655,8 @@ static void invoke_bpf_prog(struct jit_ctx *ctx, struct bpf_tramp_link *l,
 	struct bpf_prog *p = l->link.prog;
 	int cookie_off = offsetof(struct bpf_tramp_run_ctx, bpf_cookie);
 
-	if (p->aux->sleepable) {
-		enter_prog = (u64)__bpf_prog_enter_sleepable;
-		exit_prog = (u64)__bpf_prog_exit_sleepable;
-	} else {
-		enter_prog = (u64)__bpf_prog_enter;
-		exit_prog = (u64)__bpf_prog_exit;
-	}
+	enter_prog = (u64)bpf_trampoline_enter(p);
+	exit_prog = (u64)bpf_trampoline_exit(p);
 
 	if (l->cookie == 0) {
 		/* if cookie is zero, one instruction is enough to store it */
arch/x86/net/bpf_jit_comp.c +2 -17
index db6053a22e866..5e680e039d0e1 100644
@@ -1813,10 +1813,6 @@ static int invoke_bpf_prog(const struct btf_func_model *m, u8 **pprog,
 			   struct bpf_tramp_link *l, int stack_size,
 			   int run_ctx_off, bool save_ret)
 {
-	void (*exit)(struct bpf_prog *prog, u64 start,
-		     struct bpf_tramp_run_ctx *run_ctx) = __bpf_prog_exit;
-	u64 (*enter)(struct bpf_prog *prog,
-		     struct bpf_tramp_run_ctx *run_ctx) = __bpf_prog_enter;
 	u8 *prog = *pprog;
 	u8 *jmp_insn;
 	int ctx_cookie_off = offsetof(struct bpf_tramp_run_ctx, bpf_cookie);
@@ -1835,23 +1831,12 @@ static int invoke_bpf_prog(const struct btf_func_model *m, u8 **pprog,
 	 */
 	emit_stx(&prog, BPF_DW, BPF_REG_FP, BPF_REG_1, -run_ctx_off + ctx_cookie_off);
 
-	if (p->aux->sleepable) {
-		enter = __bpf_prog_enter_sleepable;
-		exit = __bpf_prog_exit_sleepable;
-	} else if (p->type == BPF_PROG_TYPE_STRUCT_OPS) {
-		enter = __bpf_prog_enter_struct_ops;
-		exit = __bpf_prog_exit_struct_ops;
-	} else if (p->expected_attach_type == BPF_LSM_CGROUP) {
-		enter = __bpf_prog_enter_lsm_cgroup;
-		exit = __bpf_prog_exit_lsm_cgroup;
-	}
-
 	/* arg1: mov rdi, progs[i] */
 	emit_mov_imm64(&prog, BPF_REG_1, (long) p >> 32, (u32) (long) p);
 	/* arg2: lea rsi, [rbp - ctx_cookie_off] */
 	EMIT4(0x48, 0x8D, 0x75, -run_ctx_off);
 
-	if (emit_call(&prog, enter, prog))
+	if (emit_call(&prog, bpf_trampoline_enter(p), prog))
 		return -EINVAL;
 	/* remember prog start time returned by __bpf_prog_enter */
 	emit_mov_reg(&prog, true, BPF_REG_6, BPF_REG_0);
@@ -1896,7 +1881,7 @@ static int invoke_bpf_prog(const struct btf_func_model *m, u8 **pprog,
 	emit_mov_reg(&prog, true, BPF_REG_2, BPF_REG_6);
 	/* arg3: lea rdx, [rbp - run_ctx_off] */
 	EMIT4(0x48, 0x8D, 0x55, -run_ctx_off);
-	if (emit_call(&prog, exit, prog))
+	if (emit_call(&prog, bpf_trampoline_exit(p), prog))
 		return -EINVAL;
 
 	*pprog = prog;
include/linux/bpf.h +10 -14
index 8cef9ec3a89c2..b3d3aa8437dce 100644
@@ -862,22 +862,18 @@ int arch_prepare_bpf_trampoline(struct bpf_tramp_image *tr, void *image, void *i
 				const struct btf_func_model *m, u32 flags,
 				struct bpf_tramp_links *tlinks,
 				void *orig_call);
-/* these two functions are called from generated trampoline */
-u64 notrace __bpf_prog_enter(struct bpf_prog *prog, struct bpf_tramp_run_ctx *run_ctx);
-void notrace __bpf_prog_exit(struct bpf_prog *prog, u64 start, struct bpf_tramp_run_ctx *run_ctx);
-u64 notrace __bpf_prog_enter_sleepable(struct bpf_prog *prog, struct bpf_tramp_run_ctx *run_ctx);
-void notrace __bpf_prog_exit_sleepable(struct bpf_prog *prog, u64 start,
-				       struct bpf_tramp_run_ctx *run_ctx);
-u64 notrace __bpf_prog_enter_lsm_cgroup(struct bpf_prog *prog,
-					struct bpf_tramp_run_ctx *run_ctx);
-void notrace __bpf_prog_exit_lsm_cgroup(struct bpf_prog *prog, u64 start,
-					struct bpf_tramp_run_ctx *run_ctx);
-u64 notrace __bpf_prog_enter_struct_ops(struct bpf_prog *prog,
-					struct bpf_tramp_run_ctx *run_ctx);
-void notrace __bpf_prog_exit_struct_ops(struct bpf_prog *prog, u64 start,
-					struct bpf_tramp_run_ctx *run_ctx);
+u64 notrace __bpf_prog_enter_sleepable_recur(struct bpf_prog *prog,
+					     struct bpf_tramp_run_ctx *run_ctx);
+void notrace __bpf_prog_exit_sleepable_recur(struct bpf_prog *prog, u64 start,
+					     struct bpf_tramp_run_ctx *run_ctx);
 void notrace __bpf_tramp_enter(struct bpf_tramp_image *tr);
 void notrace __bpf_tramp_exit(struct bpf_tramp_image *tr);
+typedef u64 (*bpf_trampoline_enter_t)(struct bpf_prog *prog,
+				      struct bpf_tramp_run_ctx *run_ctx);
+typedef void (*bpf_trampoline_exit_t)(struct bpf_prog *prog, u64 start,
+				      struct bpf_tramp_run_ctx *run_ctx);
+bpf_trampoline_enter_t bpf_trampoline_enter(const struct bpf_prog *prog);
+bpf_trampoline_exit_t bpf_trampoline_exit(const struct bpf_prog *prog);
 
 struct bpf_ksym {
 	unsigned long		 start;
include/linux/bpf_verifier.h +13 -0
index 0eb8f035b3d9f..1a32baa78ce26 100644
@@ -648,4 +648,17 @@ static inline enum bpf_prog_type resolve_prog_type(const struct bpf_prog *prog)
 		prog->aux->dst_prog->type : prog->type;
 }
 
+static inline bool bpf_prog_check_recur(const struct bpf_prog *prog)
+{
+	switch (resolve_prog_type(prog)) {
+	case BPF_PROG_TYPE_TRACING:
+		return prog->expected_attach_type != BPF_TRACE_ITER;
+	case BPF_PROG_TYPE_STRUCT_OPS:
+	case BPF_PROG_TYPE_LSM:
+		return false;
+	default:
+		return true;
+	}
+}
+
 #endif /* _LINUX_BPF_VERIFIER_H */
kernel/bpf/syscall.c +3 -2
index 0c44a716f0a24..7afec961c5728 100644
@@ -5136,13 +5136,14 @@ int kern_sys_bpf(int cmd, union bpf_attr *attr, unsigned int size)
 
 		run_ctx.bpf_cookie = 0;
 		run_ctx.saved_run_ctx = NULL;
-		if (!__bpf_prog_enter_sleepable(prog, &run_ctx)) {
+		if (!__bpf_prog_enter_sleepable_recur(prog, &run_ctx)) {
 			/* recursion detected */
 			bpf_prog_put(prog);
 			return -EBUSY;
 		}
 		attr->test.retval = bpf_prog_run(prog, (void *) (long) attr->test.ctx_in);
-		__bpf_prog_exit_sleepable(prog, 0 /* bpf_prog_run does runtime stats */, &run_ctx);
+		__bpf_prog_exit_sleepable_recur(prog, 0 /* bpf_prog_run does runtime stats */,
+						&run_ctx);
 		bpf_prog_put(prog);
 		return 0;
 #endif
kernel/bpf/trampoline.c +67 -13

134 lines not on this page · raw message

Signature · 1 line
2.40.1
author maintainer
#153 lore raw

PATCH 125/219 bpf: Invoke __bpf_prog_exit_sleepable_recur() on recursion in kern_sys_bpf().

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sebastian Andrzej Siewior <[email protected]>

[ Upstream commit 7645629f7dc88cd777f98970134bf1a54c8d77e3 ]

If __bpf_prog_enter_sleepable_recur() detects recursion then it returns 0 without undoing rcu_read_lock_trace(), migrate_disable() or decrementing the recursion counter. This is fine in the JIT case because the JIT code will jump in the 0 case to the end and invoke the matching exit trampoline (__bpf_prog_exit_sleepable_recur()).

This is not the case in kern_sys_bpf() which returns directly to the caller with an error code.

Add __bpf_prog_exit_sleepable_recur() as clean up in the recursion case.

Fixes b1d18a7574d0d ("bpf: Extend sys_bpf commands for bpf_syscall programs.")
---
 kernel/bpf/syscall.c | 1 +
 1 file changed, 1 insertion(+)
kernel/bpf/syscall.c +1 -0
index 7afec961c5728..76484137233a3 100644
@@ -5138,6 +5138,7 @@ int kern_sys_bpf(int cmd, union bpf_attr *attr, unsigned int size)
 		run_ctx.saved_run_ctx = NULL;
 		if (!__bpf_prog_enter_sleepable_recur(prog, &run_ctx)) {
 			/* recursion detected */
+			__bpf_prog_exit_sleepable_recur(prog, 0, &run_ctx);
 			bpf_prog_put(prog);
 			return -EBUSY;
 		}
Signature · 1 line
2.40.1
author maintainer
#154 lore raw

PATCH 126/219 bpf: Assign bpf_tramp_run_ctx::saved_run_ctx before recursion check.

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sebastian Andrzej Siewior <[email protected]>

[ Upstream commit 6764e767f4af1e35f87f3497e1182d945de37f93 ]

__bpf_prog_enter_recur() assigns bpf_tramp_run_ctx::saved_run_ctx before performing the recursion check which means in case of a recursion __bpf_prog_exit_recur() uses the previously set bpf_tramp_run_ctx::saved_run_ctx value.

__bpf_prog_enter_sleepable_recur() assigns bpf_tramp_run_ctx::saved_run_ctx after the recursion check which means in case of a recursion __bpf_prog_exit_sleepable_recur() uses an uninitialized value. This does not look right. If I read the entry trampoline code right, then bpf_tramp_run_ctx isn't initialized upfront.

Align __bpf_prog_enter_sleepable_recur() with __bpf_prog_enter_recur() and set bpf_tramp_run_ctx::saved_run_ctx before the recursion check is made. Remove the assignment of saved_run_ctx in kern_sys_bpf() since it happens a few cycles later.

Fixes e384c7b7b46d0 ("bpf, x86: Create bpf_tramp_run_ctx on the caller thread's stack")
---
 kernel/bpf/syscall.c    | 1 -
 kernel/bpf/trampoline.c | 5 ++---
 2 files changed, 2 insertions(+), 4 deletions(-)
kernel/bpf/syscall.c +0 -1
index 76484137233a3..0c8b7733573ee 100644
@@ -5135,7 +5135,6 @@ int kern_sys_bpf(int cmd, union bpf_attr *attr, unsigned int size)
 		}
 
 		run_ctx.bpf_cookie = 0;
-		run_ctx.saved_run_ctx = NULL;
 		if (!__bpf_prog_enter_sleepable_recur(prog, &run_ctx)) {
 			/* recursion detected */
 			__bpf_prog_exit_sleepable_recur(prog, 0, &run_ctx);
kernel/bpf/trampoline.c +2 -3
index 88841e352dcdf..c4381dfcd6b09 100644
@@ -955,13 +955,12 @@ u64 notrace __bpf_prog_enter_sleepable_recur(struct bpf_prog *prog,
 	migrate_disable();
 	might_fault();
 
+	run_ctx->saved_run_ctx = bpf_set_run_ctx(&run_ctx->run_ctx);
+
 	if (unlikely(this_cpu_inc_return(*(prog->active)) != 1)) {
 		bpf_prog_inc_misses_counter(prog);
 		return 0;
 	}
-
-	run_ctx->saved_run_ctx = bpf_set_run_ctx(&run_ctx->run_ctx);
-
 	return bpf_prog_start_time();
 }
 
Signature · 1 line
2.40.1
author maintainer
#155 lore raw

PATCH 127/219 netfilter: nftables: exthdr: fix 4-byte stack OOB write

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Florian Westphal <[email protected]>

[ Upstream commit fd94d9dadee58e09b49075240fe83423eb1dcd36 ]

If priv->len is a multiple of 4, then dst[len / 4] can write past the destination array which leads to stack corruption.

This construct is necessary to clean the remainder of the register in case ->len is NOT a multiple of the register size, so make it conditional just like nft_payload.c does.

The bug was added in 4.1 cycle and then copied/inherited when tcp/sctp and ip option support was added.

Bug reported by Zero Day Initiative project (ZDI-CAN-21950, ZDI-CAN-21951, ZDI-CAN-21961).

Fixes 49499c3e6e18 ("netfilter: nf_tables: switch registers to 32 bit addressing")
Fixes 935b7f643018 ("netfilter: nft_exthdr: add TCP option matching")
Fixes 133dc203d77d ("netfilter: nft_exthdr: Support SCTP chunks")
Fixes dbb5281a1f84 ("netfilter: nf_tables: add support for matching IPv4 options")
---
 net/netfilter/nft_exthdr.c | 22 ++++++++++++++--------
 1 file changed, 14 insertions(+), 8 deletions(-)
net/netfilter/nft_exthdr.c +14 -8
index c307c57a93e57..efb50c2b41f32 100644
@@ -35,6 +35,14 @@ static unsigned int optlen(const u8 *opt, unsigned int offset)
 		return opt[offset + 1];
 }
 
+static int nft_skb_copy_to_reg(const struct sk_buff *skb, int offset, u32 *dest, unsigned int len)
+{
+	if (len % NFT_REG32_SIZE)
+		dest[len / NFT_REG32_SIZE] = 0;
+
+	return skb_copy_bits(skb, offset, dest, len);
+}
+
 static void nft_exthdr_ipv6_eval(const struct nft_expr *expr,
 				 struct nft_regs *regs,
 				 const struct nft_pktinfo *pkt)
@@ -56,8 +64,7 @@ static void nft_exthdr_ipv6_eval(const struct nft_expr *expr,
 	}
 	offset += priv->offset;
 
-	dest[priv->len / NFT_REG32_SIZE] = 0;
-	if (skb_copy_bits(pkt->skb, offset, dest, priv->len) < 0)
+	if (nft_skb_copy_to_reg(pkt->skb, offset, dest, priv->len) < 0)
 		goto err;
 	return;
 err:
@@ -153,8 +160,7 @@ static void nft_exthdr_ipv4_eval(const struct nft_expr *expr,
 	}
 	offset += priv->offset;
 
-	dest[priv->len / NFT_REG32_SIZE] = 0;
-	if (skb_copy_bits(pkt->skb, offset, dest, priv->len) < 0)
+	if (nft_skb_copy_to_reg(pkt->skb, offset, dest, priv->len) < 0)
 		goto err;
 	return;
 err:
@@ -210,7 +216,8 @@ static void nft_exthdr_tcp_eval(const struct nft_expr *expr,
 		if (priv->flags & NFT_EXTHDR_F_PRESENT) {
 			*dest = 1;
 		} else {
-			dest[priv->len / NFT_REG32_SIZE] = 0;
+			if (priv->len % NFT_REG32_SIZE)
+				dest[priv->len / NFT_REG32_SIZE] = 0;
 			memcpy(dest, opt + offset, priv->len);
 		}
 
@@ -388,9 +395,8 @@ static void nft_exthdr_sctp_eval(const struct nft_expr *expr,
 			    offset + ntohs(sch->length) > pkt->skb->len)
 				break;
 
-			dest[priv->len / NFT_REG32_SIZE] = 0;
-			if (skb_copy_bits(pkt->skb, offset + priv->offset,
-					  dest, priv->len) < 0)
+			if (nft_skb_copy_to_reg(pkt->skb, offset + priv->offset,
+						dest, priv->len) < 0)
 				break;
 			return;
 		}
Signature · 1 line
2.40.1
author maintainer
#156 lore raw

PATCH 128/219 netfilter: nfnetlink_osf: avoid OOB read

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wander Lairson Costa <[email protected]>

[ Upstream commit f4f8a7803119005e87b716874bec07c751efafec ]

The opt_num field is controlled by user mode and is not currently validated inside the kernel. An attacker can take advantage of this to trigger an OOB read and potentially leak information.

BUG: KASAN: slab-out-of-bounds in nf_osf_match_one+0xbed/0xd10 net/netfilter/nfnetlink_osf.c:88
Read of size 2 at addr ffff88804bc64272 by task poc/6431

CPU: 1 PID: 6431 Comm: poc Not tainted 6.0.0-rc4 #1
Call Trace:
 nf_osf_match_one+0xbed/0xd10 net/netfilter/nfnetlink_osf.c:88
 nf_osf_find+0x186/0x2f0 net/netfilter/nfnetlink_osf.c:281
 nft_osf_eval+0x37f/0x590 net/netfilter/nft_osf.c:47
 expr_call_ops_eval net/netfilter/nf_tables_core.c:214
 nft_do_chain+0x2b0/0x1490 net/netfilter/nf_tables_core.c:264
 nft_do_chain_ipv4+0x17c/0x1f0 net/netfilter/nft_chain_filter.c:23
 [..]

Also add validation to genre, subtype and version fields.

Fixes 11eeef41d5f6 ("netfilter: passive OS fingerprint xtables match")
---
 net/netfilter/nfnetlink_osf.c | 8 ++++++++
 1 file changed, 8 insertions(+)
net/netfilter/nfnetlink_osf.c +8 -0
index 8f1bfa6ccc2d9..50723ba082890 100644
@@ -315,6 +315,14 @@ static int nfnl_osf_add_callback(struct sk_buff *skb,
 
 	f = nla_data(osf_attrs[OSF_ATTR_FINGER]);
 
+	if (f->opt_num > ARRAY_SIZE(f->opt))
+		return -EINVAL;
+
+	if (!memchr(f->genre, 0, MAXGENRELEN) ||
+	    !memchr(f->subtype, 0, MAXGENRELEN) ||
+	    !memchr(f->version, 0, MAXGENRELEN))
+		return -EINVAL;
+
 	kf = kmalloc(sizeof(struct nf_osf_finger), GFP_KERNEL);
 	if (!kf)
 		return -ENOMEM;
Signature · 1 line
2.40.1
author maintainer
#157 lore raw

PATCH 129/219 net: hns3: fix tx timeout issue

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jian Shen <[email protected]>

[ Upstream commit 61a1deacc3d4fd3d57d7fda4d935f7f7503e8440 ]

Currently, the driver knocks the ring doorbell before updating the ring->last_to_use in tx flow. if the hardware transmiting packet and napi poll scheduling are fast enough, it may get the old ring->last_to_use in drivers' napi poll.
In this case, the driver will think the tx is not completed, and return directly without clear the flag __QUEUE_STATE_STACK_XOFF, which may cause tx timeout.

Fixes 20d06ca2679c ("net: hns3: optimize the tx clean process")
---
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)
drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +12 -5
index 61f833d61f583..9942b21cd6193 100644
@@ -2102,8 +2102,12 @@ static void hns3_tx_doorbell(struct hns3_enet_ring *ring, int num,
 	 */
 	if (test_bit(HNS3_NIC_STATE_TX_PUSH_ENABLE, &priv->state) && num &&
 	    !ring->pending_buf && num <= HNS3_MAX_PUSH_BD_NUM && doorbell) {
+		/* This smp_store_release() pairs with smp_load_aquire() in
+		 * hns3_nic_reclaim_desc(). Ensure that the BD valid bit
+		 * is updated.
+		 */
+		smp_store_release(&ring->last_to_use, ring->next_to_use);
 		hns3_tx_push_bd(ring, num);
-		WRITE_ONCE(ring->last_to_use, ring->next_to_use);
 		return;
 	}
 
@@ -2114,6 +2118,11 @@ static void hns3_tx_doorbell(struct hns3_enet_ring *ring, int num,
 		return;
 	}
 
+	/* This smp_store_release() pairs with smp_load_aquire() in
+	 * hns3_nic_reclaim_desc(). Ensure that the BD valid bit is updated.
+	 */
+	smp_store_release(&ring->last_to_use, ring->next_to_use);
+
 	if (ring->tqp->mem_base)
 		hns3_tx_mem_doorbell(ring);
 	else
@@ -2121,7 +2130,6 @@ static void hns3_tx_doorbell(struct hns3_enet_ring *ring, int num,
 		       ring->tqp->io_base + HNS3_RING_TX_RING_TAIL_REG);
 
 	ring->pending_buf = 0;
-	WRITE_ONCE(ring->last_to_use, ring->next_to_use);
 }
 
 static void hns3_tsyn(struct net_device *netdev, struct sk_buff *skb,
@@ -3562,9 +3570,8 @@ static void hns3_reuse_buffer(struct hns3_enet_ring *ring, int i)
 static bool hns3_nic_reclaim_desc(struct hns3_enet_ring *ring,
 				  int *bytes, int *pkts, int budget)
 {
-	/* pair with ring->last_to_use update in hns3_tx_doorbell(),
-	 * smp_store_release() is not used in hns3_tx_doorbell() because
-	 * the doorbell operation already have the needed barrier operation.
+	/* This smp_load_acquire() pairs with smp_store_release() in
+	 * hns3_tx_doorbell().
 	 */
 	int ltu = smp_load_acquire(&ring->last_to_use);
 	int ntc = ring->next_to_clean;
Signature · 1 line
2.40.1
author maintainer
#158 lore raw

PATCH 130/219 net: hns3: fix byte order conversion issue in hclge_dbg_fd_tcam_read()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hao Chen <[email protected]>

[ Upstream commit efccf655e99b6907ca07a466924e91805892e7d3 ]

req1->tcam_data is defined as "u8 tcam_data[8]", and we convert it as (u32 *) without considerring byte order conversion,
it may result in printing wrong data for tcam_data.

Convert tcam_data to (__le32 *) first to fix it.

Fixes b5a0b70d77b9 ("net: hns3: refactor dump fd tcam of debugfs")
---
 .../ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c +7 -7
index 5cb8f1818e51c..a1c59f4aae988 100644
@@ -1517,7 +1517,7 @@ static int hclge_dbg_fd_tcam_read(struct hclge_dev *hdev, bool sel_x,
 	struct hclge_desc desc[3];
 	int pos = 0;
 	int ret, i;
-	u32 *req;
+	__le32 *req;
 
 	hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_FD_TCAM_OP, true);
 	desc[0].flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_NEXT);
@@ -1542,22 +1542,22 @@ static int hclge_dbg_fd_tcam_read(struct hclge_dev *hdev, bool sel_x,
 			 tcam_msg.loc);
 
 	/* tcam_data0 ~ tcam_data1 */
-	req = (u32 *)req1->tcam_data;
+	req = (__le32 *)req1->tcam_data;
 	for (i = 0; i < 2; i++)
 		pos += scnprintf(tcam_buf + pos, HCLGE_DBG_TCAM_BUF_SIZE - pos,
-				 "%08x\n", *req++);
+				 "%08x\n", le32_to_cpu(*req++));
 
 	/* tcam_data2 ~ tcam_data7 */
-	req = (u32 *)req2->tcam_data;
+	req = (__le32 *)req2->tcam_data;
 	for (i = 0; i < 6; i++)
 		pos += scnprintf(tcam_buf + pos, HCLGE_DBG_TCAM_BUF_SIZE - pos,
-				 "%08x\n", *req++);
+				 "%08x\n", le32_to_cpu(*req++));
 
 	/* tcam_data8 ~ tcam_data12 */
-	req = (u32 *)req3->tcam_data;
+	req = (__le32 *)req3->tcam_data;
 	for (i = 0; i < 5; i++)
 		pos += scnprintf(tcam_buf + pos, HCLGE_DBG_TCAM_BUF_SIZE - pos,
-				 "%08x\n", *req++);
+				 "%08x\n", le32_to_cpu(*req++));
 
 	return ret;
 }
Signature · 1 line
2.40.1
author maintainer
#159 lore raw

PATCH 131/219 net: hns3: fix debugfs concurrency issue between kfree buffer and read

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hao Chen <[email protected]>

[ Upstream commit c295160b1d95e885f1af4586a221cb221d232d10 ]

Now in hns3_dbg_uninit(), there may be concurrency between kfree buffer and read, it may result in memory error.

Moving debugfs_remove_recursive() in front of kfree buffer to ensure they don't happen at the same time.

Fixes 5e69ea7ee2a6 ("net: hns3: refactor the debugfs process")
---
 drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c +4 -3
index 69d1549e63a98..00eed9835cb55 100644
@@ -1406,9 +1406,9 @@ int hns3_dbg_init(struct hnae3_handle *handle)
 	return 0;
 
 out:
-	mutex_destroy(&handle->dbgfs_lock);
 	debugfs_remove_recursive(handle->hnae3_dbgfs);
 	handle->hnae3_dbgfs = NULL;
+	mutex_destroy(&handle->dbgfs_lock);
 	return ret;
 }
 
@@ -1416,6 +1416,9 @@ void hns3_dbg_uninit(struct hnae3_handle *handle)
 {
 	u32 i;
 
+	debugfs_remove_recursive(handle->hnae3_dbgfs);
+	handle->hnae3_dbgfs = NULL;
+
 	for (i = 0; i < ARRAY_SIZE(hns3_dbg_cmd); i++)
 		if (handle->dbgfs_buf[i]) {
 			kvfree(handle->dbgfs_buf[i]);
@@ -1423,8 +1426,6 @@ void hns3_dbg_uninit(struct hnae3_handle *handle)
 		}
 
 	mutex_destroy(&handle->dbgfs_lock);
-	debugfs_remove_recursive(handle->hnae3_dbgfs);
-	handle->hnae3_dbgfs = NULL;
 }
 
 void hns3_dbg_register_debugfs(const char *debugfs_dir_name)
Signature · 1 line
2.40.1
author maintainer
#160 lore raw

PATCH 132/219 net: hns3: fix invalid mutex between tc qdisc and dcb ets command issue

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jijie Shao <[email protected]>

[ Upstream commit fa5564945f7d15ae2390b00c08b6abaef0165cda ]

We hope that tc qdisc and dcb ets commands can not be used crosswise. If we want to use any of the commands to configure tc,
We must use the other command to clear the existing configuration.

However, when we configure a single tc with tc qdisc, we can still configure it with dcb ets.
Because we use mqprio_active as the tag of tc qdisc configuration, but with dcb ets, we do not check mqprio_active.

This patch fix this issue by check mqprio_active before executing the dcb ets command. and add dcb_ets_active to replace HCLGE_FLAG_DCB_ENABLE and HCLGE_FLAG_MQPRIO_ENABLE at the hclge layer,

Fixes cacde272dd00 ("net: hns3: Add hclge_dcb module for the support of DCB feature")
---
 drivers/net/ethernet/hisilicon/hns3/hnae3.h   |  1 +
 .../hisilicon/hns3/hns3pf/hclge_dcb.c         | 20 +++++--------------
 .../hisilicon/hns3/hns3pf/hclge_main.c        |  5 +++--
 .../hisilicon/hns3/hns3pf/hclge_main.h        |  2 --
 4 files changed, 9 insertions(+), 19 deletions(-)
drivers/net/ethernet/hisilicon/hns3/hnae3.h +1 -0
index fcb8b6dc5ab92..c693bb701ba3e 100644
@@ -797,6 +797,7 @@ struct hnae3_tc_info {
 	u8 max_tc; /* Total number of TCs */
 	u8 num_tc; /* Total number of enabled TCs */
 	bool mqprio_active;
+	bool dcb_ets_active;
 };
 
 #define HNAE3_MAX_DSCP			64
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_dcb.c +5 -15
index 09362823140d5..2740f0d703e4f 100644
@@ -251,7 +251,7 @@ static int hclge_ieee_setets(struct hnae3_handle *h, struct ieee_ets *ets)
 	int ret;
 
 	if (!(hdev->dcbx_cap & DCB_CAP_DCBX_VER_IEEE) ||
-	    hdev->flag & HCLGE_FLAG_MQPRIO_ENABLE)
+	    h->kinfo.tc_info.mqprio_active)
 		return -EINVAL;
 
 	ret = hclge_ets_validate(hdev, ets, &num_tc, &map_changed);
@@ -267,10 +267,7 @@ static int hclge_ieee_setets(struct hnae3_handle *h, struct ieee_ets *ets)
 	}
 
 	hclge_tm_schd_info_update(hdev, num_tc);
-	if (num_tc > 1)
-		hdev->flag |= HCLGE_FLAG_DCB_ENABLE;
-	else
-		hdev->flag &= ~HCLGE_FLAG_DCB_ENABLE;
+	h->kinfo.tc_info.dcb_ets_active = num_tc > 1;
 
 	ret = hclge_ieee_ets_to_tm_info(hdev, ets);
 	if (ret)
@@ -463,7 +460,7 @@ static u8 hclge_getdcbx(struct hnae3_handle *h)
 	struct hclge_vport *vport = hclge_get_vport(h);
 	struct hclge_dev *hdev = vport->back;
 
-	if (hdev->flag & HCLGE_FLAG_MQPRIO_ENABLE)
+	if (h->kinfo.tc_info.mqprio_active)
 		return 0;
 
 	return hdev->dcbx_cap;
@@ -587,7 +584,8 @@ static int hclge_setup_tc(struct hnae3_handle *h,
 	if (!test_bit(HCLGE_STATE_NIC_REGISTERED, &hdev->state))
 		return -EBUSY;
 
-	if (hdev->flag & HCLGE_FLAG_DCB_ENABLE)
+	kinfo = &vport->nic.kinfo;
+	if (kinfo->tc_info.dcb_ets_active)
 		return -EINVAL;
 
 	ret = hclge_mqprio_qopt_check(hdev, mqprio_qopt);
@@ -601,7 +599,6 @@ static int hclge_setup_tc(struct hnae3_handle *h,
 	if (ret)
 		return ret;
 
-	kinfo = &vport->nic.kinfo;
 	memcpy(&old_tc_info, &kinfo->tc_info, sizeof(old_tc_info));
 	hclge_sync_mqprio_qopt(&kinfo->tc_info, mqprio_qopt);
 	kinfo->tc_info.mqprio_active = tc > 0;
@@ -610,13 +607,6 @@ static int hclge_setup_tc(struct hnae3_handle *h,
 	if (ret)
 		goto err_out;
 
-	hdev->flag &= ~HCLGE_FLAG_DCB_ENABLE;
-
-	if (tc > 1)
-		hdev->flag |= HCLGE_FLAG_MQPRIO_ENABLE;
-	else
-		hdev->flag &= ~HCLGE_FLAG_MQPRIO_ENABLE;
-
 	return hclge_notify_init_up(hdev);
 
 err_out:
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +3 -2
index 84ecd8b9be48c..884e45fb6b72e 100644
@@ -11132,6 +11132,7 @@ static void hclge_get_mdix_mode(struct hnae3_handle *handle,
 
 static void hclge_info_show(struct hclge_dev *hdev)
 {
+	struct hnae3_handle *handle = &hdev->vport->nic;
 	struct device *dev = &hdev->pdev->dev;
 
 	dev_info(dev, "PF info begin:\n");
@@ -11148,9 +11149,9 @@ static void hclge_info_show(struct hclge_dev *hdev)
 	dev_info(dev, "This is %s PF\n",
 		 hdev->flag & HCLGE_FLAG_MAIN ? "main" : "not main");
 	dev_info(dev, "DCB %s\n",
-		 hdev->flag & HCLGE_FLAG_DCB_ENABLE ? "enable" : "disable");
+		 handle->kinfo.tc_info.dcb_ets_active ? "enable" : "disable");
 	dev_info(dev, "MQPRIO %s\n",
-		 hdev->flag & HCLGE_FLAG_MQPRIO_ENABLE ? "enable" : "disable");
+		 handle->kinfo.tc_info.mqprio_active ? "enable" : "disable");
 	dev_info(dev, "Default tx spare buffer size: %u\n",
 		 hdev->tx_spare_buf_size);
 
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h +0 -2
index 13f23d606e77b..f6fef790e16c1 100644
@@ -916,8 +916,6 @@ struct hclge_dev {
 
 #define HCLGE_FLAG_MAIN			BIT(0)
 #define HCLGE_FLAG_DCB_CAPABLE		BIT(1)
-#define HCLGE_FLAG_DCB_ENABLE		BIT(2)
-#define HCLGE_FLAG_MQPRIO_ENABLE	BIT(3)
 	u32 flag;
 
 	u32 pkt_buf_size; /* Total pf buf size for tx/rx */
Signature · 1 line
2.40.1
author maintainer
#161 lore raw

PATCH 133/219 net: hns3: fix the port information display when sfp is absent

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yisen Zhuang <[email protected]>

[ Upstream commit 674d9591a32d01df75d6b5fffed4ef942a294376 ]

When sfp is absent or unidentified, the port type should be displayed as PORT_OTHERS, rather than PORT_FIBRE.

Fixes 88d10bd6f730 ("net: hns3: add support for multiple media type")
---
 drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c +3 -1
index cdf76fb58d45e..e22835ae8a941 100644
@@ -776,7 +776,9 @@ static int hns3_get_link_ksettings(struct net_device *netdev,
 		hns3_get_ksettings(h, cmd);
 		break;
 	case HNAE3_MEDIA_TYPE_FIBER:
-		if (module_type == HNAE3_MODULE_TYPE_CR)
+		if (module_type == HNAE3_MODULE_TYPE_UNKNOWN)
+			cmd->base.port = PORT_OTHER;
+		else if (module_type == HNAE3_MODULE_TYPE_CR)
 			cmd->base.port = PORT_DA;
 		else
 			cmd->base.port = PORT_FIBRE;
Signature · 1 line
2.40.1
author maintainer
#162 lore raw

PATCH 134/219 net: hns3: remove GSO partial feature bit

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jie Wang <[email protected]>

[ Upstream commit 60326634f6c54528778de18bfef1e8a7a93b3771 ]

HNS3 NIC does not support GSO partial packets segmentation. Actually tunnel packets for example NvGRE packets segment offload and checksum offload is already supported. There is no need to keep gso partial feature bit. So this patch removes it.

Fixes 76ad4f0ee747 ("net: hns3: Add support of HNS3 Ethernet Driver for hip08 SoC")
---
 drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 2 --
 1 file changed, 2 deletions(-)
drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +0 -2
index 9942b21cd6193..8aae179554a81 100644
@@ -3315,8 +3315,6 @@ static void hns3_set_default_feature(struct net_device *netdev)
 
 	netdev->priv_flags |= IFF_UNICAST_FLT;
 
-	netdev->gso_partial_features |= NETIF_F_GSO_GRE_CSUM;
-
 	netdev->features |= NETIF_F_HW_VLAN_CTAG_FILTER |
 		NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX |
 		NETIF_F_RXCSUM | NETIF_F_SG | NETIF_F_GSO |
Signature · 1 line
2.40.1
author maintainer
#163 lore raw

PATCH 135/219 sh: boards: Fix CEU buffer size passed to dma_declare_coherent_memory()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Petr Tesarik <[email protected]>

[ Upstream commit fb60211f377b69acffead3147578f86d0092a7a5 ]

In all these cases, the last argument to dma_declare_coherent_memory() is the buffer end address, but the expected value should be the size of the reserved region.

Fixes 39fb993038e1 ("media: arch: sh: ap325rxa: Use new renesas-ceu camera driver")
Fixes c2f9b05fd5c1 ("media: arch: sh: ecovec: Use new renesas-ceu camera driver")
Fixes f3590dc32974 ("media: arch: sh: kfr2r09: Use new renesas-ceu camera driver")
Fixes 186c446f4b84 ("media: arch: sh: migor: Use new renesas-ceu camera driver")
Fixes 1a3c230b4151 ("media: arch: sh: ms7724se: Use new renesas-ceu camera driver")
---
 arch/sh/boards/mach-ap325rxa/setup.c | 2 +-
 arch/sh/boards/mach-ecovec24/setup.c | 6 ++----
 arch/sh/boards/mach-kfr2r09/setup.c  | 2 +-
 arch/sh/boards/mach-migor/setup.c    | 2 +-
 arch/sh/boards/mach-se/7724/setup.c  | 6 ++----
 5 files changed, 7 insertions(+), 11 deletions(-)
arch/sh/boards/mach-ap325rxa/setup.c +1 -1
index c77b5f00a66a3..d8f8dca4d7968 100644
@@ -530,7 +530,7 @@ static int __init ap325rxa_devices_setup(void)
 	device_initialize(&ap325rxa_ceu_device.dev);
 	dma_declare_coherent_memory(&ap325rxa_ceu_device.dev,
 			ceu_dma_membase, ceu_dma_membase,
-			ceu_dma_membase + CEU_BUFFER_MEMORY_SIZE - 1);
+			CEU_BUFFER_MEMORY_SIZE);
 
 	platform_device_add(&ap325rxa_ceu_device);
 
arch/sh/boards/mach-ecovec24/setup.c +2 -4
index 674da7ebd8b7f..7ec03d4a4edf0 100644
@@ -1454,15 +1454,13 @@ static int __init arch_setup(void)
 	device_initialize(&ecovec_ceu_devices[0]->dev);
 	dma_declare_coherent_memory(&ecovec_ceu_devices[0]->dev,
 				    ceu0_dma_membase, ceu0_dma_membase,
-				    ceu0_dma_membase +
-				    CEU_BUFFER_MEMORY_SIZE - 1);
+				    CEU_BUFFER_MEMORY_SIZE);
 	platform_device_add(ecovec_ceu_devices[0]);
 
 	device_initialize(&ecovec_ceu_devices[1]->dev);
 	dma_declare_coherent_memory(&ecovec_ceu_devices[1]->dev,
 				    ceu1_dma_membase, ceu1_dma_membase,
-				    ceu1_dma_membase +
-				    CEU_BUFFER_MEMORY_SIZE - 1);
+				    CEU_BUFFER_MEMORY_SIZE);
 	platform_device_add(ecovec_ceu_devices[1]);
 
 	gpiod_add_lookup_table(&cn12_power_gpiod_table);
arch/sh/boards/mach-kfr2r09/setup.c +1 -1
index 20f4db778ed6a..c6d556dfbbbe6 100644
@@ -603,7 +603,7 @@ static int __init kfr2r09_devices_setup(void)
 	device_initialize(&kfr2r09_ceu_device.dev);
 	dma_declare_coherent_memory(&kfr2r09_ceu_device.dev,
 			ceu_dma_membase, ceu_dma_membase,
-			ceu_dma_membase + CEU_BUFFER_MEMORY_SIZE - 1);
+			CEU_BUFFER_MEMORY_SIZE);
 
 	platform_device_add(&kfr2r09_ceu_device);
 
arch/sh/boards/mach-migor/setup.c +1 -1
index f60061283c482..773ee767d0c4e 100644
@@ -604,7 +604,7 @@ static int __init migor_devices_setup(void)
 	device_initialize(&migor_ceu_device.dev);
 	dma_declare_coherent_memory(&migor_ceu_device.dev,
 			ceu_dma_membase, ceu_dma_membase,
-			ceu_dma_membase + CEU_BUFFER_MEMORY_SIZE - 1);
+			CEU_BUFFER_MEMORY_SIZE);
 
 	platform_device_add(&migor_ceu_device);
 
arch/sh/boards/mach-se/7724/setup.c +2 -4
index b60a2626e18b2..6495f93540654 100644
@@ -940,15 +940,13 @@ static int __init devices_setup(void)
 	device_initialize(&ms7724se_ceu_devices[0]->dev);
 	dma_declare_coherent_memory(&ms7724se_ceu_devices[0]->dev,
 				    ceu0_dma_membase, ceu0_dma_membase,
-				    ceu0_dma_membase +
-				    CEU_BUFFER_MEMORY_SIZE - 1);
+				    CEU_BUFFER_MEMORY_SIZE);
 	platform_device_add(ms7724se_ceu_devices[0]);
 
 	device_initialize(&ms7724se_ceu_devices[1]->dev);
 	dma_declare_coherent_memory(&ms7724se_ceu_devices[1]->dev,
 				    ceu1_dma_membase, ceu1_dma_membase,
-				    ceu1_dma_membase +
-				    CEU_BUFFER_MEMORY_SIZE - 1);
+				    CEU_BUFFER_MEMORY_SIZE);
 	platform_device_add(ms7724se_ceu_devices[1]);
 
 	return platform_add_devices(ms7724se_devices,
Signature · 1 line
2.40.1
author maintainer
#164 lore raw

PATCH 136/219 Multi-gen LRU: avoid race in inc_min_seq()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kalesh Singh <[email protected]>

commit bb5e7f234eacf34b65be67ebb3613e3b8cf11b87 upstream.

inc_max_seq() will try to inc_min_seq() if nr_gens == MAX_NR_GENS. This is because the generations are reused (the last oldest now empty generation will become the next youngest generation).

inc_min_seq() is retried until successful, dropping the lru_lock and yielding the CPU on each failure, and retaking the lock before trying again:

        while (!inc_min_seq(lruvec, type, can_swap)) {
                spin_unlock_irq(&lruvec->lru_lock);
                cond_resched();
                spin_lock_irq(&lruvec->lru_lock);
        }

However, the initial condition that required incrementing the min_seq (nr_gens == MAX_NR_GENS) is not retested. This can change by another call to inc_max_seq() from run_aging() with force_scan=true from the debugfs interface.

Since the eviction stalls when the nr_gens == MIN_NR_GENS, avoid unnecessarily incrementing the min_seq by rechecking the number of generations before each attempt.

This issue was uncovered in previous discussion on the list by Yu Zhao and Aneesh Kumar [1].

[1] https://lore.kernel.org/linux-mm/CAOUHufbO7CaVm=xjEb1avDhHVvnC8pJmGyKcFf2iY_dpf+zR3w@mail.gmail.com/

Fixes d6c3af7d8a2b ("mm: multi-gen LRU: debugfs interface")
Cc Yu Zhao <[email protected]>
Cc Aneesh Kumar K V <[email protected]>
Cc Barry Song <[email protected]>
Cc Brian Geffon <[email protected]>
Cc Jan Alexander Steffens (heftig) <[email protected]>
Cc Lecopzer Chen <[email protected]>
Cc Matthias Brugger <[email protected]>
Cc Oleksandr Natalenko <[email protected]>
Cc Qi Zheng <[email protected]>
Cc Steven Barrett <[email protected]>
Cc Suleiman Souhlal <[email protected]>
Cc Suren Baghdasaryan <[email protected]>
---
 mm/vmscan.c |   12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)
mm/vmscan.c +7 -5
@@ -4331,6 +4331,7 @@ static void inc_max_seq(struct lruvec *l
 	int type, zone;
 	struct lru_gen_struct *lrugen = &lruvec->lrugen;
 
+restart:
 	spin_lock_irq(&lruvec->lru_lock);
 
 	VM_WARN_ON_ONCE(!seq_is_valid(lruvec));
@@ -4341,11 +4342,12 @@ static void inc_max_seq(struct lruvec *l
 
 		VM_WARN_ON_ONCE(!force_scan && (type == LRU_GEN_FILE || can_swap));
 
-		while (!inc_min_seq(lruvec, type, can_swap)) {
-			spin_unlock_irq(&lruvec->lru_lock);
-			cond_resched();
-			spin_lock_irq(&lruvec->lru_lock);
-		}
+		if (inc_min_seq(lruvec, type, can_swap))
+			continue;
+
+		spin_unlock_irq(&lruvec->lru_lock);
+		cond_resched();
+		goto restart;
 	}
 
 	/*
author maintainer
#165 lore raw

PATCH 137/219 net/mlx5: Free IRQ rmap and notifier on kernel shutdown

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Saeed Mahameed <[email protected]>

commit 314ded538e5f22e7610b1bf621402024a180ec80 upstream.

The kernel IRQ system needs the irq affinity notifier to be clear before attempting to free the irq, see WARN_ON log below.

On a normal driver unload we don't have this issue since we do the complete cleanup of the irq resources.

To fix this, put the important resources cleanup in a helper function and use it in both normal driver unload and shutdown flows.

[ 4497.498434] ------------[ cut here ]------------
[ 4497.498726] WARNING: CPU: 0 PID: 9 at kernel/irq/manage.c:2034 free_irq+0x295/0x340
[ 4497.499193] Modules linked in:
[ 4497.499386] CPU: 0 PID: 9 Comm: kworker/0:1 Tainted: G        W          6.4.0-rc4+ #10
[ 4497.499876] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.2-1.fc38 04/01/2014
[ 4497.500518] Workqueue: events do_poweroff
[ 4497.500849] RIP: 0010:free_irq+0x295/0x340
[ 4497.501132] Code: 85 c0 0f 84 1d ff ff ff 48 89 ef ff d0 0f 1f 00 e9 10 ff ff ff 0f 0b e9 72 ff ff ff 49 8d 7f 28 ff d0 0f 1f 00 e9 df fd ff ff <0f> 0b 48 c7 80 c0 008
[ 4497.502269] RSP: 0018:ffffc90000053da0 EFLAGS: 00010282
[ 4497.502589] RAX: ffff888100949600 RBX: ffff88810330b948 RCX: 0000000000000000
[ 4497.503035] RDX: ffff888100949600 RSI: ffff888100400490 RDI: 0000000000000023
[ 4497.503472] RBP: ffff88810330c7e0 R08: ffff8881004005d0 R09: ffffffff8273a260
[ 4497.503923] R10: 0000000000000000 R11: 0000000000000000 R12: ffff8881009ae000
[ 4497.504359] R13: ffff8881009ae148 R14: 0000000000000000 R15: ffff888100949600
[ 4497.504804] FS:  0000000000000000(0000) GS:ffff88813bc00000(0000) knlGS:0000000000000000
[ 4497.505302] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 4497.505671] CR2: 00007fce98806298 CR3: 000000000262e005 CR4: 0000000000370ef0
[ 4497.506104] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 4497.506540] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 4497.507002] Call Trace:
[ 4497.507158]  <TASK>
[ 4497.507299]  ? free_irq+0x295/0x340
[ 4497.507522]  ? __warn+0x7c/0x130
[ 4497.507740]  ? free_irq+0x295/0x340
[ 4497.507963]  ? report_bug+0x171/0x1a0
[ 4497.508197]  ? handle_bug+0x3c/0x70
[ 4497.508417]  ? exc_invalid_op+0x17/0x70
[ 4497.508662]  ? asm_exc_invalid_op+0x1a/0x20
[ 4497.508926]  ? free_irq+0x295/0x340
[ 4497.509146]  mlx5_irq_pool_free_irqs+0x48/0x90
[ 4497.509421]  mlx5_irq_table_free_irqs+0x38/0x50
[ 4497.509714]  mlx5_core_eq_free_irqs+0x27/0x40
[ 4497.509984]  shutdown+0x7b/0x100
[ 4497.510184]  pci_device_shutdown+0x30/0x60
[ 4497.510440]  device_shutdown+0x14d/0x240
[ 4497.510698]  kernel_power_off+0x30/0x70
[ 4497.510938]  process_one_work+0x1e6/0x3e0
[ 4497.511183]  worker_thread+0x49/0x3b0
[ 4497.511407]  ? __pfx_worker_thread+0x10/0x10
[ 4497.511679]  kthread+0xe0/0x110
[ 4497.511879]  ? __pfx_kthread+0x10/0x10
[ 4497.512114]  ret_from_fork+0x29/0x50
[ 4497.512342]  </TASK>
Fixes 9c2d08010963 ("net/mlx5: Free irqs only on shutdown callback")
---
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c |   26 ++++++++++++++++------
 1 file changed, 20 insertions(+), 6 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c +20 -6
@@ -123,18 +123,32 @@ out:
 	return ret;
 }
 
-static void irq_release(struct mlx5_irq *irq)
+/* mlx5_system_free_irq - Free an IRQ
+ * @irq: IRQ to free
+ *
+ * Free the IRQ and other resources such as rmap from the system.
+ * BUT doesn't free or remove reference from mlx5.
+ * This function is very important for the shutdown flow, where we need to
+ * cleanup system resoruces but keep mlx5 objects alive,
+ * see mlx5_irq_table_free_irqs().
+ */
+static void mlx5_system_free_irq(struct mlx5_irq *irq)
 {
-	struct mlx5_irq_pool *pool = irq->pool;
-
-	xa_erase(&pool->irqs, irq->index);
 	/* free_irq requires that affinity_hint and rmap will be cleared
 	 * before calling it. This is why there is asymmetry with set_rmap
 	 * which should be called after alloc_irq but before request_irq.
 	 */
 	irq_update_affinity_hint(irq->irqn, NULL);
-	free_cpumask_var(irq->mask);
 	free_irq(irq->irqn, &irq->nh);
+}
+
+static void irq_release(struct mlx5_irq *irq)
+{
+	struct mlx5_irq_pool *pool = irq->pool;
+
+	xa_erase(&pool->irqs, irq->index);
+	mlx5_system_free_irq(irq);
+	free_cpumask_var(irq->mask);
 	kfree(irq);
 }
 
@@ -597,7 +611,7 @@ static void mlx5_irq_pool_free_irqs(stru
 	unsigned long index;
 
 	xa_for_each(&pool->irqs, index, irq)
-		free_irq(irq->irqn, &irq->nh);
+		mlx5_system_free_irq(irq);
 }
 
 static void mlx5_irq_pools_free_irqs(struct mlx5_irq_table *table)
author maintainer
#166 lore raw

PATCH 138/219 ARC: atomics: Add compiler barrier to atomic operations...

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pavel Kozlov <[email protected]>

commit 42f51fb24fd39cc547c086ab3d8a314cc603a91c upstream.

... to avoid unwanted gcc optimizations

SMP kernels fail to boot with commit 596ff4a09b89 ("cpumask: re-introduce constant-sized cpumask optimizations").

|
| percpu: BUG: failure at mm/percpu.c:2981/pcpu_build_alloc_info()!
|

The write operation performed by the SCOND instruction in the atomic inline asm code is not properly passed to the compiler. The compiler cannot correctly optimize a nested loop that runs through the cpumask in the pcpu_build_alloc_info() function.

Fix this by add a compiler barrier (memory clobber in inline asm).

Apparently atomic ops used to have memory clobber implicitly via surrounding smp_mb(). However commit b64be6836993c431e ("ARC: atomics: implement relaxed variants") removed the smp_mb() for the relaxed variants, but failed to add the explicit compiler barrier.

Cc <[email protected]> # v6.3+
Fixes b64be6836993c43 ("ARC: atomics: implement relaxed variants")
[vgupta: tweaked the changelog and added Fixes tag]
---
 arch/arc/include/asm/atomic-llsc.h    |    6 +++---
 arch/arc/include/asm/atomic64-arcv2.h |    6 +++---
 2 files changed, 6 insertions(+), 6 deletions(-)
arch/arc/include/asm/atomic-llsc.h +3 -3
@@ -18,7 +18,7 @@ static inline void arch_atomic_##op(int
 	: [val]	"=&r"	(val) /* Early clobber to prevent reg reuse */	\
 	: [ctr]	"r"	(&v->counter), /* Not "m": llock only supports reg direct addr mode */	\
 	  [i]	"ir"	(i)						\
-	: "cc");							\
+	: "cc", "memory");						\
 }									\
 
 #define ATOMIC_OP_RETURN(op, asm_op)				\
@@ -34,7 +34,7 @@ static inline int arch_atomic_##op##_ret
 	: [val]	"=&r"	(val)						\
 	: [ctr]	"r"	(&v->counter),					\
 	  [i]	"ir"	(i)						\
-	: "cc");							\
+	: "cc", "memory");						\
 									\
 	return val;							\
 }
@@ -56,7 +56,7 @@ static inline int arch_atomic_fetch_##op
 	  [orig] "=&r" (orig)						\
 	: [ctr]	"r"	(&v->counter),					\
 	  [i]	"ir"	(i)						\
-	: "cc");							\
+	: "cc", "memory");						\
 									\
 	return orig;							\
 }
arch/arc/include/asm/atomic64-arcv2.h +3 -3
@@ -60,7 +60,7 @@ static inline void arch_atomic64_##op(s6
 	"	bnz     1b		\n"				\
 	: "=&r"(val)							\
 	: "r"(&v->counter), "ir"(a)					\
-	: "cc");							\
+	: "cc", "memory");						\
 }									\
 
 #define ATOMIC64_OP_RETURN(op, op1, op2)		        	\
@@ -77,7 +77,7 @@ static inline s64 arch_atomic64_##op##_r
 	"	bnz     1b		\n"				\
 	: [val] "=&r"(val)						\
 	: "r"(&v->counter), "ir"(a)					\
-	: "cc");	/* memory clobber comes from smp_mb() */	\
+	: "cc", "memory");						\
 									\
 	return val;							\
 }
@@ -99,7 +99,7 @@ static inline s64 arch_atomic64_fetch_##
 	"	bnz     1b		\n"				\
 	: "=&r"(orig), "=&r"(val)					\
 	: "r"(&v->counter), "ir"(a)					\
-	: "cc");	/* memory clobber comes from smp_mb() */	\
+	: "cc", "memory");						\
 									\
 	return orig;							\
 }
author maintainer
#167 lore raw

PATCH 139/219 clocksource/drivers/arm_arch_timer: Disable timer before programming CVAL

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Walter Chang <[email protected]>

commit e7d65e40ab5a5940785c5922f317602d0268caaf upstream.

Due to the fact that the use of writeq_relaxed() to program CVAL is not guaranteed to be atomic, it is necessary to disable the timer before programming CVAL.

However, if the MMIO timer is already enabled and has not yet expired, there is a possibility of unexpected behavior occurring: when the CPU enters the idle state during this period, and if the CPU's local event is earlier than the broadcast event, the following process occurs:

tick_broadcast_enter()
tick_broadcast_oneshot_control(TICK_BROADCAST_ENTER)
__tick_broadcast_oneshot_control()
___tick_broadcast_oneshot_control()
tick_broadcast_set_event()
clockevents_program_event()
set_next_event_mem()

During this process, the MMIO timer remains enabled while programming CVAL. To prevent such behavior, disable timer explicitly prior to programming CVAL.

---
 drivers/clocksource/arm_arch_timer.c |    7 +++++++
 1 file changed, 7 insertions(+)
drivers/clocksource/arm_arch_timer.c +7 -0
@@ -773,6 +773,13 @@ static __always_inline void set_next_eve
 	u64 cnt;
 
 	ctrl = arch_timer_reg_read(access, ARCH_TIMER_REG_CTRL, clk);
+
+	/* Timer must be disabled before programming CVAL */
+	if (ctrl & ARCH_TIMER_CTRL_ENABLE) {
+		ctrl &= ~ARCH_TIMER_CTRL_ENABLE;
+		arch_timer_reg_write(access, ARCH_TIMER_REG_CTRL, ctrl, clk);
+	}
+
 	ctrl |= ARCH_TIMER_CTRL_ENABLE;
 	ctrl &= ~ARCH_TIMER_CTRL_IT_MASK;
 
author maintainer
#168 lore raw

PATCH 140/219 dmaengine: sh: rz-dmac: Fix destination and source data size setting

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hien Huynh <[email protected]>

commit c6ec8c83a29fb3aec3efa6fabbf5344498f57c7f upstream.

Before setting DDS and SDS values, we need to clear its value first otherwise, we get incorrect results when we change/update the DMA bus width several times due to the 'OR' expression.

---
 drivers/dma/sh/rz-dmac.c |   11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)
drivers/dma/sh/rz-dmac.c +7 -4
@@ -9,6 +9,7 @@
  * Copyright 2012 Javier Martin, Vista Silicon <[email protected]>
  */
 
+#include <linux/bitfield.h>
 #include <linux/dma-mapping.h>
 #include <linux/dmaengine.h>
 #include <linux/interrupt.h>
@@ -145,8 +146,8 @@ struct rz_dmac {
 #define CHCFG_REQD			BIT(3)
 #define CHCFG_SEL(bits)			((bits) & 0x07)
 #define CHCFG_MEM_COPY			(0x80400008)
-#define CHCFG_FILL_DDS(a)		(((a) << 16) & GENMASK(19, 16))
-#define CHCFG_FILL_SDS(a)		(((a) << 12) & GENMASK(15, 12))
+#define CHCFG_FILL_DDS_MASK		GENMASK(19, 16)
+#define CHCFG_FILL_SDS_MASK		GENMASK(15, 12)
 #define CHCFG_FILL_TM(a)		(((a) & BIT(5)) << 22)
 #define CHCFG_FILL_AM(a)		(((a) & GENMASK(4, 2)) << 6)
 #define CHCFG_FILL_LVL(a)		(((a) & BIT(1)) << 5)
@@ -609,13 +610,15 @@ static int rz_dmac_config(struct dma_cha
 	if (val == CHCFG_DS_INVALID)
 		return -EINVAL;
 
-	channel->chcfg |= CHCFG_FILL_DDS(val);
+	channel->chcfg &= ~CHCFG_FILL_DDS_MASK;
+	channel->chcfg |= FIELD_PREP(CHCFG_FILL_DDS_MASK, val);
 
 	val = rz_dmac_ds_to_val_mapping(config->src_addr_width);
 	if (val == CHCFG_DS_INVALID)
 		return -EINVAL;
 
-	channel->chcfg |= CHCFG_FILL_SDS(val);
+	channel->chcfg &= ~CHCFG_FILL_SDS_MASK;
+	channel->chcfg |= FIELD_PREP(CHCFG_FILL_SDS_MASK, val);
 
 	return 0;
 }
author maintainer
#169 lore raw

PATCH 141/219 jbd2: fix checkpoint cleanup performance regression

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhang Yi <[email protected]>

commit 373ac521799d9e97061515aca6ec6621789036bb upstream.

journal_clean_one_cp_list() has been merged into journal_shrink_one_cp_list(), but do chekpoint buffer cleanup from the committing process is just a best effort, it should stop scan once it meet a busy buffer, or else it will cause a lot of invalid buffer scan and checks. We catch a performance regression when doing fs_mark tests below.

Test cmd:
./fs_mark -d scratch -s 1024 -n 10000 -t 1 -D 100 -N 100

Before merging checkpoint buffer cleanup:

 FSUse%        Count         Size    Files/sec     App Overhead
     95        10000         1024       8304.9            49033

After merging checkpoint buffer cleanup:

 FSUse%        Count         Size    Files/sec     App Overhead
     95        10000         1024       7649.0            50012
 FSUse%        Count         Size    Files/sec     App Overhead
     95        10000         1024       2107.1            50871

After merging checkpoint buffer cleanup, the total loop count in journal_shrink_one_cp_list() could be up to 6,261,600+ (50,000+ ~ 100,000+ in general), most of them are invalid. This patch fix it through passing 'shrink_type' into journal_shrink_one_cp_list() and add a new 'SHRINK_BUSY_STOP' to indicate it should stop once meet a busy buffer. After fix, the loop count descending back to 10,000+.

After this fix:

 FSUse%        Count         Size    Files/sec     App Overhead
     95        10000         1024       8558.4            49109
---
 fs/jbd2/checkpoint.c |   20 ++++++++++++++------
 1 file changed, 14 insertions(+), 6 deletions(-)
fs/jbd2/checkpoint.c +14 -6
@@ -349,6 +349,8 @@ int jbd2_cleanup_journal_tail(journal_t
 
 /* Checkpoint list management */
 
+enum shrink_type {SHRINK_DESTROY, SHRINK_BUSY_STOP, SHRINK_BUSY_SKIP};
+
 /*
  * journal_shrink_one_cp_list
  *
@@ -360,7 +362,8 @@ int jbd2_cleanup_journal_tail(journal_t
  * Called with j_list_lock held.
  */
 static unsigned long journal_shrink_one_cp_list(struct journal_head *jh,
-						bool destroy, bool *released)
+						enum shrink_type type,
+						bool *released)
 {
 	struct journal_head *last_jh;
 	struct journal_head *next_jh = jh;
@@ -376,12 +379,15 @@ static unsigned long journal_shrink_one_
 		jh = next_jh;
 		next_jh = jh->b_cpnext;
 
-		if (destroy) {
+		if (type == SHRINK_DESTROY) {
 			ret = __jbd2_journal_remove_checkpoint(jh);
 		} else {
 			ret = jbd2_journal_try_remove_checkpoint(jh);
-			if (ret < 0)
-				continue;
+			if (ret < 0) {
+				if (type == SHRINK_BUSY_SKIP)
+					continue;
+				break;
+			}
 		}
 
 		nr_freed++;
@@ -445,7 +451,7 @@ again:
 		tid = transaction->t_tid;
 
 		freed = journal_shrink_one_cp_list(transaction->t_checkpoint_list,
-						   false, &released);
+						   SHRINK_BUSY_SKIP, &released);
 		nr_freed += freed;
 		(*nr_to_scan) -= min(*nr_to_scan, freed);
 		if (*nr_to_scan == 0)
@@ -485,19 +491,21 @@ out:
 void __jbd2_journal_clean_checkpoint_list(journal_t *journal, bool destroy)
 {
 	transaction_t *transaction, *last_transaction, *next_transaction;
+	enum shrink_type type;
 	bool released;
 
 	transaction = journal->j_checkpoint_transactions;
 	if (!transaction)
 		return;
 
+	type = destroy ? SHRINK_DESTROY : SHRINK_BUSY_STOP;
 	last_transaction = transaction->t_cpprev;
 	next_transaction = transaction;
 	do {
 		transaction = next_transaction;
 		next_transaction = transaction->t_cpnext;
 		journal_shrink_one_cp_list(transaction->t_checkpoint_list,
-					   destroy, &released);
+					   type, &released);
 		/*
 		 * This function only frees up some memory if possible so we
 		 * dont have an obligation to finish processing. Bail out if
author maintainer
#170 lore raw

PATCH 142/219 jbd2: check jh->b_transaction before removing it from checkpoint

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhihao Cheng <[email protected]>

commit 590a809ff743e7bd890ba5fb36bc38e20a36de53 upstream.

Following process will corrupt ext4 image:
Step 1:
jbd2_journal_commit_transaction
__jbd2_journal_insert_checkpoint(jh, commit_transaction) // Put jh into trans1->t_checkpoint_list journal->j_checkpoint_transactions = commit_transaction // Put trans1 into journal->j_checkpoint_transactions

Step 2:
do_get_write_access
test_clear_buffer_dirty(bh) // clear buffer dirty,set jbd dirty __jbd2_journal_file_buffer(jh, transaction) // jh belongs to trans2

Step 3:
drop_cache
journal_shrink_one_cp_list
jbd2_journal_try_remove_checkpoint
if (!trylock_buffer(bh)) // lock bh, true
if (buffer_dirty(bh)) // buffer is not dirty
__jbd2_journal_remove_checkpoint(jh)
// remove jh from trans1->t_checkpoint_list

Step 4:
jbd2_log_do_checkpoint
trans1 = journal->j_checkpoint_transactions // jh is not in trans1->t_checkpoint_list jbd2_cleanup_journal_tail(journal) // trans1 is done

Step 5: Power cut, trans2 is not committed, jh is lost in next mounting.

Fix it by checking 'jh->b_transaction' before remove it from checkpoint.

---
 fs/jbd2/checkpoint.c |    2 ++
 1 file changed, 2 insertions(+)
fs/jbd2/checkpoint.c +2 -0
@@ -639,6 +639,8 @@ int jbd2_journal_try_remove_checkpoint(s
 {
 	struct buffer_head *bh = jh2bh(jh);
 
+	if (jh->b_transaction)
+		return -EBUSY;
 	if (!trylock_buffer(bh))
 		return -EBUSY;
 	if (buffer_dirty(bh)) {
author maintainer
#171 lore raw

PATCH 143/219 jbd2: correct the end of the journal recovery scan range

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhang Yi <[email protected]>

commit 2dfba3bb40ad8536b9fa802364f2d40da31aa88e upstream.

We got a filesystem inconsistency issue below while running generic/475 I/O failure pressure test with fast_commit feature enabled.

Symlink /p3/d3/d1c/d6c/dd6/dce/l101 (inode #132605) is invalid.

If fast_commit feature is enabled, a special fast_commit journal area is appended to the end of the normal journal area. The journal->j_last point to the first unused block behind the normal journal area instead of the whole log area, and the journal->j_fc_last point to the first unused block behind the fast_commit journal area. While doing journal recovery, do_one_pass(PASS_SCAN) should first scan the normal journal area and turn around to the first block once it meet journal->j_last, but the wrap() macro misuse the journal->j_fc_last, so the recovering could not read the next magic block (commit block perhaps) and would end early mistakenly and missing tN and every transaction after it in the following example. Finally, it could lead to filesystem inconsistency.

 | normal journal area                             | fast commit area |
 +-------------------------------------------------+------------------+
 | tN(rere) | tN+1 |~| tN-x |...| tN-1 | tN(front) |       ....       |
 +-------------------------------------------------+------------------+
                     /                             /                  /
                start               journal->j_last journal->j_fc_last

This patch fix it by use the correct ending journal->j_last.

---
 fs/jbd2/recovery.c |   12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)
fs/jbd2/recovery.c +3 -9
@@ -229,12 +229,8 @@ static int count_tags(journal_t *journal
 /* Make sure we wrap around the log correctly! */
 #define wrap(journal, var)						\
 do {									\
-	unsigned long _wrap_last =					\
-		jbd2_has_feature_fast_commit(journal) ?			\
-			(journal)->j_fc_last : (journal)->j_last;	\
-									\
-	if (var >= _wrap_last)						\
-		var -= (_wrap_last - (journal)->j_first);		\
+	if (var >= (journal)->j_last)					\
+		var -= ((journal)->j_last - (journal)->j_first);	\
 } while (0)
 
 static int fc_do_one_pass(journal_t *journal,
@@ -517,9 +513,7 @@ static int do_one_pass(journal_t *journa
 				break;
 
 		jbd2_debug(2, "Scanning for sequence ID %u at %lu/%lu\n",
-			  next_commit_ID, next_log_block,
-			  jbd2_has_feature_fast_commit(journal) ?
-			  journal->j_fc_last : journal->j_last);
+			  next_commit_ID, next_log_block, journal->j_last);
 
 		/* Skip over each chunk of the transaction looking
 		 * either the next descriptor block or the final commit
author maintainer
#172 lore raw

PATCH 144/219 ext4: add correct group descriptors and reserved GDT blocks to system zone

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wang Jianjian <[email protected]>

commit 68228da51c9a436872a4ef4b5a7692e29f7e5bc7 upstream.

When setup_system_zone, flex_bg is not initialized so it is always 1. Use a new helper function, ext4_num_base_meta_blocks() which does not depend on sbi->s_log_groups_per_flex being initialized.

[ Squashed two patches in the Link URL's below together into a single
commit, which is simpler to review/understand. Also fix checkpatch warnings. --TYT ]

---
 fs/ext4/balloc.c         |   15 +++++++++++----
 fs/ext4/block_validity.c |    8 ++++----
 fs/ext4/ext4.h           |    2 ++
 3 files changed, 17 insertions(+), 8 deletions(-)
fs/ext4/balloc.c +11 -4
@@ -910,11 +910,11 @@ unsigned long ext4_bg_num_gdb(struct sup
 }
 
 /*
- * This function returns the number of file system metadata clusters at
+ * This function returns the number of file system metadata blocks at
  * the beginning of a block group, including the reserved gdt blocks.
  */
-static unsigned ext4_num_base_meta_clusters(struct super_block *sb,
-				     ext4_group_t block_group)
+unsigned int ext4_num_base_meta_blocks(struct super_block *sb,
+				       ext4_group_t block_group)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	unsigned num;
@@ -932,8 +932,15 @@ static unsigned ext4_num_base_meta_clust
 	} else { /* For META_BG_BLOCK_GROUPS */
 		num += ext4_bg_num_gdb(sb, block_group);
 	}
-	return EXT4_NUM_B2C(sbi, num);
+	return num;
 }
+
+static unsigned int ext4_num_base_meta_clusters(struct super_block *sb,
+						ext4_group_t block_group)
+{
+	return EXT4_NUM_B2C(EXT4_SB(sb), ext4_num_base_meta_blocks(sb, block_group));
+}
+
 /**
  *	ext4_inode_to_goal_block - return a hint for block allocation
  *	@inode: inode for block allocation
fs/ext4/block_validity.c +4 -4
@@ -215,7 +215,6 @@ int ext4_setup_system_zone(struct super_
 	struct ext4_system_blocks *system_blks;
 	struct ext4_group_desc *gdp;
 	ext4_group_t i;
-	int flex_size = ext4_flex_bg_size(sbi);
 	int ret;
 
 	system_blks = kzalloc(sizeof(*system_blks), GFP_KERNEL);
@@ -223,12 +222,13 @@ int ext4_setup_system_zone(struct super_
 		return -ENOMEM;
 
 	for (i=0; i < ngroups; i++) {
+		unsigned int meta_blks = ext4_num_base_meta_blocks(sb, i);
+
 		cond_resched();
-		if (ext4_bg_has_super(sb, i) &&
-		    ((i < 5) || ((i % flex_size) == 0))) {
+		if (meta_blks != 0) {
 			ret = add_system_zone(system_blks,
 					ext4_group_first_block_no(sb, i),
-					ext4_bg_num_gdb(sb, i) + 1, 0);
+					meta_blks, 0);
 			if (ret)
 				goto err;
 		}
fs/ext4/ext4.h +2 -0
@@ -3096,6 +3096,8 @@ extern const char *ext4_decode_error(str
 extern void ext4_mark_group_bitmap_corrupted(struct super_block *sb,
 					     ext4_group_t block_group,
 					     unsigned int flags);
+extern unsigned int ext4_num_base_meta_blocks(struct super_block *sb,
+					      ext4_group_t block_group);
 
 extern __printf(7, 8)
 void __ext4_error(struct super_block *, const char *, unsigned int, bool,
author maintainer
#173 lore raw

PATCH 145/219 ext4: fix memory leaks in ext4_fname_{setup_filename,prepare_lookup}

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Luís Henriques <[email protected]>

commit 7ca4b085f430f3774c3838b3da569ceccd6a0177 upstream.

If the filename casefolding fails, we'll be leaking memory from the fscrypt_name struct, namely from the 'crypto_buf.name' member.

Make sure we free it in the error path on both ext4_fname_setup_filename() and ext4_fname_prepare_lookup() functions.

---
 fs/ext4/crypto.c |    4 ++++
 1 file changed, 4 insertions(+)
fs/ext4/crypto.c +4 -0
@@ -33,6 +33,8 @@ int ext4_fname_setup_filename(struct ino
 
 #if IS_ENABLED(CONFIG_UNICODE)
 	err = ext4_fname_setup_ci_filename(dir, iname, fname);
+	if (err)
+		ext4_fname_free_filename(fname);
 #endif
 	return err;
 }
@@ -51,6 +53,8 @@ int ext4_fname_prepare_lookup(struct ino
 
 #if IS_ENABLED(CONFIG_UNICODE)
 	err = ext4_fname_setup_ci_filename(dir, &dentry->d_name, fname);
+	if (err)
+		ext4_fname_free_filename(fname);
 #endif
 	return err;
 }
author maintainer
#174 lore raw

PATCH 146/219 f2fs: flush inode if atomic file is aborted

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jaegeuk Kim <[email protected]>

commit a3ab55746612247ce3dcaac6de66f5ffc055b9df upstream.

Let's flush the inode being aborted atomic operation to avoid stale dirty inode during eviction in this call stack:

  f2fs_mark_inode_dirty_sync+0x22/0x40 [f2fs]
  f2fs_abort_atomic_write+0xc4/0xf0 [f2fs]
  f2fs_evict_inode+0x3f/0x690 [f2fs]
  ? sugov_start+0x140/0x140
  evict+0xc3/0x1c0
  evict_inodes+0x17b/0x210
  generic_shutdown_super+0x32/0x120
  kill_block_super+0x21/0x50
  deactivate_locked_super+0x31/0x90
  cleanup_mnt+0x100/0x160
  task_work_run+0x59/0x90
  do_exit+0x33b/0xa50
  do_group_exit+0x2d/0x80
  __x64_sys_exit_group+0x14/0x20
  do_syscall_64+0x3b/0x90
  entry_SYSCALL_64_after_hwframe+0x63/0xcd

This triggers f2fs_bug_on() in f2fs_evict_inode:
f2fs_bug_on(sbi, is_inode_flag_set(inode, FI_DIRTY_INODE));

This fixes the syzbot report:

loop0: detected capacity change from 0 to 131072 F2FS-fs (loop0): invalid crc value
F2FS-fs (loop0): Found nat_bits in checkpoint F2FS-fs (loop0): Mounted with checkpoint version = 48b305e4

------------[ cut here ]------------
kernel BUG at fs/f2fs/inode.c:869!
invalid opcode: 0000 [#1] PREEMPT SMP KASAN
CPU: 0 PID: 5014 Comm: syz-executor220 Not tainted 6.4.0-syzkaller-11479-g6cd06ab12d1a #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/27/2023
RIP: 0010:f2fs_evict_inode+0x172d/0x1e00 fs/f2fs/inode.c:869
Code: ff df 48 c1 ea 03 80 3c 02 00 0f 85 6a 06 00 00 8b 75 40 ba 01 00 00 00 4c 89 e7 e8 6d ce 06 00 e9 aa fc ff ff e8 63 22 e2 fd <0f> 0b e8 5c 22 e2 fd 48 c7 c0 a8 3a 18 8d 48 ba 00 00 00 00 00 fc
RSP: 0018:ffffc90003a6fa00 EFLAGS: 00010293
RAX: 0000000000000000 RBX: 0000000000000001 RCX: 0000000000000000
RDX: ffff8880273b8000 RSI: ffffffff83a2bd0d RDI: 0000000000000007
RBP: ffff888077db91b0 R08: 0000000000000007 R09: 0000000000000000
R10: 0000000000000001 R11: 0000000000000001 R12: ffff888029a3c000
R13: ffff888077db9660 R14: ffff888029a3c0b8 R15: ffff888077db9c50
FS:  0000000000000000(0000) GS:ffff8880b9800000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f1909bb9000 CR3: 00000000276a9000 CR4: 0000000000350ef0
Call Trace:
 <TASK>
 evict+0x2ed/0x6b0 fs/inode.c:665
 dispose_list+0x117/0x1e0 fs/inode.c:698
 evict_inodes+0x345/0x440 fs/inode.c:748
 generic_shutdown_super+0xaf/0x480 fs/super.c:478
 kill_block_super+0x64/0xb0 fs/super.c:1417
 kill_f2fs_super+0x2af/0x3c0 fs/f2fs/super.c:4704
 deactivate_locked_super+0x98/0x160 fs/super.c:330
 deactivate_super+0xb1/0xd0 fs/super.c:361
 cleanup_mnt+0x2ae/0x3d0 fs/namespace.c:1254
 task_work_run+0x16f/0x270 kernel/task_work.c:179
 exit_task_work include/linux/task_work.h:38 [inline]
 do_exit+0xa9a/0x29a0 kernel/exit.c:874
 do_group_exit+0xd4/0x2a0 kernel/exit.c:1024
 __do_sys_exit_group kernel/exit.c:1035 [inline]
 __se_sys_exit_group kernel/exit.c:1033 [inline]
 __x64_sys_exit_group+0x3e/0x50 kernel/exit.c:1033
 do_syscall_x64 arch/x86/entry/common.c:50 [inline]
 do_syscall_64+0x39/0xb0 arch/x86/entry/common.c:80
 entry_SYSCALL_64_after_hwframe+0x63/0xcd
RIP: 0033:0x7f309be71a09
Code: Unable to access opcode bytes at 0x7f309be719df.
RSP: 002b:00007fff171df518 EFLAGS: 00000246 ORIG_RAX: 00000000000000e7
RAX: ffffffffffffffda RBX: 00007f309bef7330 RCX: 00007f309be71a09
RDX: 000000000000003c RSI: 00000000000000e7 RDI: 0000000000000001
RBP: 0000000000000001 R08: ffffffffffffffc0 R09: 00007f309bef1e40
R10: 0000000000010600 R11: 0000000000000246 R12: 00007f309bef7330
R13: 0000000000000001 R14: 0000000000000000 R15: 0000000000000001
 </TASK>
Modules linked in:
---[ end trace 0000000000000000 ]---
RIP: 0010:f2fs_evict_inode+0x172d/0x1e00 fs/f2fs/inode.c:869
Code: ff df 48 c1 ea 03 80 3c 02 00 0f 85 6a 06 00 00 8b 75 40 ba 01 00 00 00 4c 89 e7 e8 6d ce 06 00 e9 aa fc ff ff e8 63 22 e2 fd <0f> 0b e8 5c 22 e2 fd 48 c7 c0 a8 3a 18 8d 48 ba 00 00 00 00 00 fc
RSP: 0018:ffffc90003a6fa00 EFLAGS: 00010293
RAX: 0000000000000000 RBX: 0000000000000001 RCX: 0000000000000000
RDX: ffff8880273b8000 RSI: ffffffff83a2bd0d RDI: 0000000000000007
RBP: ffff888077db91b0 R08: 0000000000000007 R09: 0000000000000000
R10: 0000000000000001 R11: 0000000000000001 R12: ffff888029a3c000
R13: ffff888077db9660 R14: ffff888029a3c0b8 R15: ffff888077db9c50
FS:  0000000000000000(0000) GS:ffff8880b9800000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f1909bb9000 CR3: 00000000276a9000 CR4: 0000000000350ef0
---
 fs/f2fs/segment.c |    2 ++
 1 file changed, 2 insertions(+)
fs/f2fs/segment.c +2 -0
@@ -204,6 +204,8 @@ void f2fs_abort_atomic_write(struct inod
 		f2fs_i_size_write(inode, fi->original_i_size);
 		fi->original_i_size = 0;
 	}
+	/* avoid stale dirty inode during eviction */
+	sync_inode_metadata(inode, 0);
 }
 
 static int __replace_atomic_write_block(struct inode *inode, pgoff_t index,
author maintainer
#175 lore raw

PATCH 147/219 f2fs: avoid false alarm of circular locking

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jaegeuk Kim <[email protected]>

commit 5c13e2388bf3426fd69a89eb46e50469e9624e56 upstream.

======================================================
WARNING: possible circular locking dependency detected 6.5.0-rc5-syzkaller-00353-gae545c3283dc #0 Not tainted
------------------------------------------------------
syz-executor273/5027 is trying to acquire lock:
ffff888077fe1fb0 (&fi->i_sem){+.+.}-{3:3}, at: f2fs_down_write fs/f2fs/f2fs.h:2133 [inline] ffff888077fe1fb0 (&fi->i_sem){+.+.}-{3:3}, at: f2fs_add_inline_entry+0x300/0x6f0 fs/f2fs/inline.c:644

but task is already holding lock:
ffff888077fe07c8 (&fi->i_xattr_sem){.+.+}-{3:3}, at: f2fs_down_read fs/f2fs/f2fs.h:2108 [inline] ffff888077fe07c8 (&fi->i_xattr_sem){.+.+}-{3:3}, at: f2fs_add_dentry+0x92/0x230 fs/f2fs/dir.c:783

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #1 (&fi->i_xattr_sem){.+.+}-{3:3}:

       down_read+0x9c/0x470 kernel/locking/rwsem.c:1520
       f2fs_down_read fs/f2fs/f2fs.h:2108 [inline]
       f2fs_getxattr+0xb1e/0x12c0 fs/f2fs/xattr.c:532
       __f2fs_get_acl+0x5a/0x900 fs/f2fs/acl.c:179
       f2fs_acl_create fs/f2fs/acl.c:377 [inline]
       f2fs_init_acl+0x15c/0xb30 fs/f2fs/acl.c:420
       f2fs_init_inode_metadata+0x159/0x1290 fs/f2fs/dir.c:558
       f2fs_add_regular_entry+0x79e/0xb90 fs/f2fs/dir.c:740
       f2fs_add_dentry+0x1de/0x230 fs/f2fs/dir.c:788
       f2fs_do_add_link+0x190/0x280 fs/f2fs/dir.c:827
       f2fs_add_link fs/f2fs/f2fs.h:3554 [inline]
       f2fs_mkdir+0x377/0x620 fs/f2fs/namei.c:781
       vfs_mkdir+0x532/0x7e0 fs/namei.c:4117
       do_mkdirat+0x2a9/0x330 fs/namei.c:4140
       __do_sys_mkdir fs/namei.c:4160 [inline]
       __se_sys_mkdir fs/namei.c:4158 [inline]
       __x64_sys_mkdir+0xf2/0x140 fs/namei.c:4158
       do_syscall_x64 arch/x86/entry/common.c:50 [inline]
       do_syscall_64+0x38/0xb0 arch/x86/entry/common.c:80
       entry_SYSCALL_64_after_hwframe+0x63/0xcd

-> #0 (&fi->i_sem){+.+.}-{3:3}:

       check_prev_add kernel/locking/lockdep.c:3142 [inline]
       check_prevs_add kernel/locking/lockdep.c:3261 [inline]
       validate_chain kernel/locking/lockdep.c:3876 [inline]
       __lock_acquire+0x2e3d/0x5de0 kernel/locking/lockdep.c:5144
       lock_acquire kernel/locking/lockdep.c:5761 [inline]
       lock_acquire+0x1ae/0x510 kernel/locking/lockdep.c:5726
       down_write+0x93/0x200 kernel/locking/rwsem.c:1573
       f2fs_down_write fs/f2fs/f2fs.h:2133 [inline]
       f2fs_add_inline_entry+0x300/0x6f0 fs/f2fs/inline.c:644
       f2fs_add_dentry+0xa6/0x230 fs/f2fs/dir.c:784
       f2fs_do_add_link+0x190/0x280 fs/f2fs/dir.c:827
       f2fs_add_link fs/f2fs/f2fs.h:3554 [inline]
       f2fs_mkdir+0x377/0x620 fs/f2fs/namei.c:781
       vfs_mkdir+0x532/0x7e0 fs/namei.c:4117
       ovl_do_mkdir fs/overlayfs/overlayfs.h:196 [inline]
       ovl_mkdir_real+0xb5/0x370 fs/overlayfs/dir.c:146
       ovl_workdir_create+0x3de/0x820 fs/overlayfs/super.c:309
       ovl_make_workdir fs/overlayfs/super.c:711 [inline]
       ovl_get_workdir fs/overlayfs/super.c:864 [inline]
       ovl_fill_super+0xdab/0x6180 fs/overlayfs/super.c:1400
       vfs_get_super+0xf9/0x290 fs/super.c:1152
       vfs_get_tree+0x88/0x350 fs/super.c:1519
       do_new_mount fs/namespace.c:3335 [inline]
       path_mount+0x1492/0x1ed0 fs/namespace.c:3662
       do_mount fs/namespace.c:3675 [inline]
       __do_sys_mount fs/namespace.c:3884 [inline]
       __se_sys_mount fs/namespace.c:3861 [inline]
       __x64_sys_mount+0x293/0x310 fs/namespace.c:3861
       do_syscall_x64 arch/x86/entry/common.c:50 [inline]
       do_syscall_64+0x38/0xb0 arch/x86/entry/common.c:80
       entry_SYSCALL_64_after_hwframe+0x63/0xcd

other info that might help us debug this:

 Possible unsafe locking scenario:
       CPU0                    CPU1
       ----                    ----
  rlock(&fi->i_xattr_sem);
                               lock(&fi->i_sem);
                               lock(&fi->i_xattr_sem);

lock(&fi->i_sem);

Reported-and-tested-by [email protected]
Fixes 5eda1ad1aaff "f2fs: fix deadlock in i_xattr_sem and inode page lock"
---
 fs/f2fs/f2fs.h   |   24 +++++++++++++++---------
 fs/f2fs/inline.c |    3 ++-
 2 files changed, 17 insertions(+), 10 deletions(-)
fs/f2fs/f2fs.h +15 -9
@@ -2160,15 +2160,6 @@ static inline int f2fs_down_read_trylock
 	return down_read_trylock(&sem->internal_rwsem);
 }
 
-#ifdef CONFIG_DEBUG_LOCK_ALLOC
-static inline void f2fs_down_read_nested(struct f2fs_rwsem *sem, int subclass)
-{
-	down_read_nested(&sem->internal_rwsem, subclass);
-}
-#else
-#define f2fs_down_read_nested(sem, subclass) f2fs_down_read(sem)
-#endif
-
 static inline void f2fs_up_read(struct f2fs_rwsem *sem)
 {
 	up_read(&sem->internal_rwsem);
@@ -2179,6 +2170,21 @@ static inline void f2fs_down_write(struc
 	down_write(&sem->internal_rwsem);
 }
 
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+static inline void f2fs_down_read_nested(struct f2fs_rwsem *sem, int subclass)
+{
+	down_read_nested(&sem->internal_rwsem, subclass);
+}
+
+static inline void f2fs_down_write_nested(struct f2fs_rwsem *sem, int subclass)
+{
+	down_write_nested(&sem->internal_rwsem, subclass);
+}
+#else
+#define f2fs_down_read_nested(sem, subclass) f2fs_down_read(sem)
+#define f2fs_down_write_nested(sem, subclass) f2fs_down_write(sem)
+#endif
+
 static inline int f2fs_down_write_trylock(struct f2fs_rwsem *sem)
 {
 	return down_write_trylock(&sem->internal_rwsem);
fs/f2fs/inline.c +2 -1
@@ -642,7 +642,8 @@ int f2fs_add_inline_entry(struct inode *
 	}
 
 	if (inode) {
-		f2fs_down_write(&F2FS_I(inode)->i_sem);
+		f2fs_down_write_nested(&F2FS_I(inode)->i_sem,
+						SINGLE_DEPTH_NESTING);
 		page = f2fs_init_inode_metadata(inode, dir, fname, ipage);
 		if (IS_ERR(page)) {
 			err = PTR_ERR(page);
author maintainer
#176 lore raw

PATCH 148/219 lib: test_scanf: Add explicit type cast to result initialization in test_number_prefix()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nathan Chancellor <[email protected]>

commit 92382d744176f230101d54f5c017bccd62770f01 upstream.

A recent change in clang allows it to consider more expressions as compile time constants, which causes it to point out an implicit conversion in the scanf tests:

  lib/test_scanf.c:661:2: warning: implicit conversion from 'int' to 'unsigned char' changes value from -168 to 88 [-Wconstant-conversion]
    661 |         test_number_prefix(unsigned char,       "0xA7", "%2hhx%hhx", 0, 0xa7, 2, check_uchar);
        |         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  lib/test_scanf.c:609:29: note: expanded from macro 'test_number_prefix'
    609 |         T result[2] = {~expect[0], ~expect[1]};                                 \
        |                       ~            ^~~~~~~~~~
  1 warning generated.

The result of the bitwise negation is the type of the operand after going through the integer promotion rules, so this truncation is expected but harmless, as the initial values in the result array get overwritten by _test() anyways. Add an explicit cast to the expected type in test_number_prefix() to silence the warning. There is no functional change, as all the tests still pass with GCC 13.1.0 and clang 18.0.0.

---
 lib/test_scanf.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
lib/test_scanf.c +1 -1
@@ -606,7 +606,7 @@ static void __init numbers_slice(void)
 #define test_number_prefix(T, str, scan_fmt, expect0, expect1, n_args, fn)	\
 do {										\
 	const T expect[2] = { expect0, expect1 };				\
-	T result[2] = {~expect[0], ~expect[1]};					\
+	T result[2] = { (T)~expect[0], (T)~expect[1] };				\
 										\
 	_test(fn, &expect, str, scan_fmt, n_args, &result[0], &result[1]);	\
 } while (0)
author maintainer
#177 lore raw

PATCH 149/219 hwspinlock: qcom: add missing regmap config for SFPB MMIO implementation

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christian Marangi <[email protected]>

commit 23316be8a9d450f33a21f1efe7d89570becbec58 upstream.

Commit 5d4753f741d8 ("hwspinlock: qcom: add support for MMIO on older SoCs") introduced and made regmap_config mandatory in the of_data struct but didn't add the regmap_config for sfpb based devices.

SFPB based devices can both use the legacy syscon way to probe or the new MMIO way and currently device that use the MMIO way are broken as they lack the definition of the now required regmap_config and always return -EINVAL (and indirectly makes fail probing everything that depends on it, smem, nandc with smem-parser...)

Fix this by correctly adding the missing regmap_config and restore function of hwspinlock on SFPB based devices with MMIO implementation.

Fixes 5d4753f741d8 ("hwspinlock: qcom: add support for MMIO on older SoCs")
---
 drivers/hwspinlock/qcom_hwspinlock.c |    9 +++++++++
 1 file changed, 9 insertions(+)
drivers/hwspinlock/qcom_hwspinlock.c +9 -0
@@ -69,9 +69,18 @@ static const struct hwspinlock_ops qcom_
 	.unlock		= qcom_hwspinlock_unlock,
 };
 
+static const struct regmap_config sfpb_mutex_config = {
+	.reg_bits		= 32,
+	.reg_stride		= 4,
+	.val_bits		= 32,
+	.max_register		= 0x100,
+	.fast_io		= true,
+};
+
 static const struct qcom_hwspinlock_of_data of_sfpb_mutex = {
 	.offset = 0x4,
 	.stride = 0x4,
+	.regmap_config = &sfpb_mutex_config,
 };
 
 static const struct regmap_config tcsr_msm8226_mutex_config = {
author maintainer
#178 lore raw

PATCH 150/219 ata: ahci: Add Elkhart Lake AHCI controller

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Werner Fischer <[email protected]>

commit 2a2df98ec592667927b5c1351afa6493ea125c9f upstream.

Elkhart Lake is the successor of Apollo Lake and Gemini Lake. These CPUs and their PCHs are used in mobile and embedded environments.

With this patch I suggest that Elkhart Lake SATA controllers [1] should use the default LPM policy for mobile chipsets.
The disadvantage of missing hot-plug support with this setting should not be an issue, as those CPUs are used in embedded environments and not in servers with hot-plug backplanes.

We discovered that the Elkhart Lake SATA controllers have been missing in ahci.c after a customer reported the throttling of his SATA SSD after a short period of higher I/O. We determined the high temperature of the SSD controller in idle mode as the root cause for that.

Depending on the used SSD, we have seen up to 1.8 Watt lower system idle power usage and up to 30°C lower SSD controller temperatures in our tests, when we set med_power_with_dipm manually. I have provided a table showing seven different SATA SSDs from ATP, Intel/Solidigm and Samsung [2].

Intel lists a total of 3 SATA controller IDs (4B60, 4B62, 4B63) in [1] for those mobile PCHs.
This commit just adds 0x4b63 as I do not have test systems with 0x4b60 and 0x4b62 SATA controllers.
I have tested this patch with a system which uses 0x4b63 as SATA controller.

[1] https://sata-io.org/product/8803
[2] https://www.thomas-krenn.com/en/wiki/SATA_Link_Power_Management#Example_LES_v4

---
 drivers/ata/ahci.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/ata/ahci.c +2 -0
@@ -422,6 +422,8 @@ static const struct pci_device_id ahci_p
 	{ PCI_VDEVICE(INTEL, 0x34d3), board_ahci_low_power }, /* Ice Lake LP AHCI */
 	{ PCI_VDEVICE(INTEL, 0x02d3), board_ahci_low_power }, /* Comet Lake PCH-U AHCI */
 	{ PCI_VDEVICE(INTEL, 0x02d7), board_ahci_low_power }, /* Comet Lake PCH RAID */
+	/* Elkhart Lake IDs 0x4b60 & 0x4b62 https://sata-io.org/product/8803 not tested yet */
+	{ PCI_VDEVICE(INTEL, 0x4b63), board_ahci_low_power }, /* Elkhart Lake AHCI */
 
 	/* JMicron 360/1/3/5/6, match class to avoid IDE function */
 	{ PCI_VENDOR_ID_JMICRON, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
author maintainer
#179 lore raw

PATCH 151/219 ata: pata_falcon: fix IO base selection for Q40

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Michael Schmitz <[email protected]>

commit 8a1f00b753ecfdb117dc1a07e68c46d80e7923ea upstream.

With commit 44b1fbc0f5f3 ("m68k/q40: Replace q40ide driver with pata_falcon and falconide"), the Q40 IDE driver was replaced by pata_falcon.c.

Both IO and memory resources were defined for the Q40 IDE platform device, but definition of the IDE register addresses was modeled after the Falcon case, both in use of the memory resources and in including register shift and byte vs. word offset in the address.

This was correct for the Falcon case, which does not apply any address translation to the register addresses. In the Q40 case, all of device base address, byte access offset and register shift is included in the platform specific ISA access translation (in asm/mm_io.h).

As a consequence, such address translation gets applied twice, and register addresses are mangled.

Use the device base address from the platform IO resource for Q40 (the IO address translation will then add the correct ISA window base address and byte access offset), with register shift 1. Use MMIO base address and register shift 2 as before for Falcon.

Encode PIO_OFFSET into IO port addresses for all registers for Q40 except the data transfer register. Encode the MMIO offset there (pata_falcon_data_xfer() directly uses raw IO with no address translation).

---
 drivers/ata/pata_falcon.c |   50 ++++++++++++++++++++++++++--------------------
 1 file changed, 29 insertions(+), 21 deletions(-)
drivers/ata/pata_falcon.c +29 -21
@@ -123,8 +123,8 @@ static int __init pata_falcon_init_one(s
 	struct resource *base_res, *ctl_res, *irq_res;
 	struct ata_host *host;
 	struct ata_port *ap;
-	void __iomem *base;
-	int irq = 0;
+	void __iomem *base, *ctl_base;
+	int irq = 0, io_offset = 1, reg_shift = 2; /* Falcon defaults */
 
 	dev_info(&pdev->dev, "Atari Falcon and Q40/Q60 PATA controller\n");
 
@@ -165,26 +165,34 @@ static int __init pata_falcon_init_one(s
 	ap->pio_mask = ATA_PIO4;
 	ap->flags |= ATA_FLAG_SLAVE_POSS | ATA_FLAG_NO_IORDY;
 
-	base = (void __iomem *)base_mem_res->start;
 	/* N.B. this assumes data_addr will be used for word-sized I/O only */
-	ap->ioaddr.data_addr		= base + 0 + 0 * 4;
-	ap->ioaddr.error_addr		= base + 1 + 1 * 4;
-	ap->ioaddr.feature_addr		= base + 1 + 1 * 4;
-	ap->ioaddr.nsect_addr		= base + 1 + 2 * 4;
-	ap->ioaddr.lbal_addr		= base + 1 + 3 * 4;
-	ap->ioaddr.lbam_addr		= base + 1 + 4 * 4;
-	ap->ioaddr.lbah_addr		= base + 1 + 5 * 4;
-	ap->ioaddr.device_addr		= base + 1 + 6 * 4;
-	ap->ioaddr.status_addr		= base + 1 + 7 * 4;
-	ap->ioaddr.command_addr		= base + 1 + 7 * 4;
-
-	base = (void __iomem *)ctl_mem_res->start;
-	ap->ioaddr.altstatus_addr	= base + 1;
-	ap->ioaddr.ctl_addr		= base + 1;
-
-	ata_port_desc(ap, "cmd 0x%lx ctl 0x%lx",
-		      (unsigned long)base_mem_res->start,
-		      (unsigned long)ctl_mem_res->start);
+	ap->ioaddr.data_addr = (void __iomem *)base_mem_res->start;
+
+	if (base_res) {		/* only Q40 has IO resources */
+		io_offset = 0x10000;
+		reg_shift = 0;
+		base = (void __iomem *)base_res->start;
+		ctl_base = (void __iomem *)ctl_res->start;
+	} else {
+		base = (void __iomem *)base_mem_res->start;
+		ctl_base = (void __iomem *)ctl_mem_res->start;
+	}
+
+	ap->ioaddr.error_addr	= base + io_offset + (1 << reg_shift);
+	ap->ioaddr.feature_addr	= base + io_offset + (1 << reg_shift);
+	ap->ioaddr.nsect_addr	= base + io_offset + (2 << reg_shift);
+	ap->ioaddr.lbal_addr	= base + io_offset + (3 << reg_shift);
+	ap->ioaddr.lbam_addr	= base + io_offset + (4 << reg_shift);
+	ap->ioaddr.lbah_addr	= base + io_offset + (5 << reg_shift);
+	ap->ioaddr.device_addr	= base + io_offset + (6 << reg_shift);
+	ap->ioaddr.status_addr	= base + io_offset + (7 << reg_shift);
+	ap->ioaddr.command_addr	= base + io_offset + (7 << reg_shift);
+
+	ap->ioaddr.altstatus_addr	= ctl_base + io_offset;
+	ap->ioaddr.ctl_addr		= ctl_base + io_offset;
+
+	ata_port_desc(ap, "cmd %px ctl %px data %px",
+		      base, ctl_base, ap->ioaddr.data_addr);
 
 	irq_res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
 	if (irq_res && irq_res->start > 0) {
author maintainer
#180 lore raw

PATCH 152/219 ata: sata_gemini: Add missing MODULE_DESCRIPTION

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Damien Le Moal <[email protected]>

commit 8566572bf3b4d6e416a4bf2110dbb4817d11ba59 upstream.

Add the missing MODULE_DESCRIPTION() to avoid warnings such as:

WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/ata/sata_gemini.o

when compiling with W=1.

Fixes be4e456ed3a5 ("ata: Add driver for Faraday Technology FTIDE010")
---
 drivers/ata/sata_gemini.c |    1 +
 1 file changed, 1 insertion(+)
drivers/ata/sata_gemini.c +1 -0
@@ -428,6 +428,7 @@ static struct platform_driver gemini_sat
 };
 module_platform_driver(gemini_sata_driver);
 
+MODULE_DESCRIPTION("low level driver for Cortina Systems Gemini SATA bridge");
 MODULE_AUTHOR("Linus Walleij <[email protected]>");
 MODULE_LICENSE("GPL");
 MODULE_ALIAS("platform:" DRV_NAME);
author maintainer
#181 lore raw

PATCH 153/219 ata: pata_ftide010: Add missing MODULE_DESCRIPTION

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Damien Le Moal <[email protected]>

commit 7274eef5729037300f29d14edeb334a47a098f65 upstream.

Add the missing MODULE_DESCRIPTION() to avoid warnings such as:

WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/ata/pata_ftide010.o

when compiling with W=1.

Fixes be4e456ed3a5 ("ata: Add driver for Faraday Technology FTIDE010")
---
 drivers/ata/pata_ftide010.c |    1 +
 1 file changed, 1 insertion(+)
drivers/ata/pata_ftide010.c +1 -0
@@ -567,6 +567,7 @@ static struct platform_driver pata_ftide
 };
 module_platform_driver(pata_ftide010_driver);
 
+MODULE_DESCRIPTION("low level driver for Faraday Technology FTIDE010");
 MODULE_AUTHOR("Linus Walleij <[email protected]>");
 MODULE_LICENSE("GPL");
 MODULE_ALIAS("platform:" DRV_NAME);
author maintainer
#182 lore raw

PATCH 154/219 fuse: nlookup missing decrement in fuse_direntplus_link

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: ruanmeisi <[email protected]>

commit b8bd342d50cbf606666488488f9fea374aceb2d5 upstream.

During our debugging of glusterfs, we found an Assertion failed error:
inode_lookup >= nlookup, which was caused by the nlookup value in the kernel being greater than that in the FUSE file system.

The issue was introduced by fuse_direntplus_link, where in the function, fuse_iget increments nlookup, and if d_splice_alias returns failure, fuse_direntplus_link returns failure without decrementing nlookup
https://github.com/gluster/glusterfs/pull/4081

Fixes 0b05b18381ee ("fuse: implement NFS-like readdirplus support")
Cc <[email protected]> # v3.9
---
 fs/fuse/readdir.c |   10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)
fs/fuse/readdir.c +9 -1
@@ -243,8 +243,16 @@ retry:
 			dput(dentry);
 			dentry = alias;
 		}
-		if (IS_ERR(dentry))
+		if (IS_ERR(dentry)) {
+			if (!IS_ERR(inode)) {
+				struct fuse_inode *fi = get_fuse_inode(inode);
+
+				spin_lock(&fi->lock);
+				fi->nlookup--;
+				spin_unlock(&fi->lock);
+			}
 			return PTR_ERR(dentry);
+		}
 	}
 	if (fc->readdirplus_auto)
 		set_bit(FUSE_I_INIT_RDPLUS, &get_fuse_inode(inode)->state);
author maintainer
#183 lore raw

PATCH 155/219 btrfs: zoned: do not zone finish data relocation block group

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Naohiro Aota <[email protected]>

commit 332581bde2a419d5f12a93a1cdc2856af649a3cc upstream.

When multiple writes happen at once, we may need to sacrifice a currently active block group to be zone finished for a new allocation. We choose a block group with the least free space left, and zone finish it.

To do the finishing, we need to send IOs for already allocated region and wait for them and on-going IOs. Otherwise, these IOs fail because the zone is already finished at the time the IO reach a device.

However, if a block group dedicated to the data relocation is zone finished, there is a chance that finishing it before an ongoing write IO reaches the device. That is because there is timing gap between an allocation is done (block_group->reservations == 0, as pre-allocation is done) and an ordered extent is created when the relocation IO starts. Thus, if we finish the zone between them, we can fail the IOs.

We cannot simply use "fs_info->data_reloc_bg == block_group->start" to avoid the zone finishing. Because, the data_reloc_bg may already switch to a new block group, while there are still ongoing write IOs to the old data_reloc_bg.

So, this patch reworks the BLOCK_GROUP_FLAG_ZONED_DATA_RELOC bit to indicate there is a data relocation allocation and/or ongoing write to the block group. The bit is set on allocation and cleared in end_io function of the last IO for the currently allocated region.

To change the timing of the bit setting also solves the issue that the bit being left even after there is no IO going on. With the current code, if the data_reloc_bg switches after the last IO to the current data_reloc_bg, the bit is set at this timing and there is no one clearing that bit. As a result, that block group is kept unallocatable for anything.

Fixes 343d8a30851c ("btrfs: zoned: prevent allocation from previous data relocation BG")
Fixes 74e91b12b115 ("btrfs: zoned: zone finish unused block group")
---
 fs/btrfs/extent-tree.c |   43 +++++++++++++++++++++++--------------------
 fs/btrfs/zoned.c       |   16 +++++++++++++---
 2 files changed, 36 insertions(+), 23 deletions(-)
fs/btrfs/extent-tree.c +23 -20
@@ -3810,7 +3810,8 @@ static int do_allocation_zoned(struct bt
 	       fs_info->data_reloc_bg == 0);
 
 	if (block_group->ro ||
-	    test_bit(BLOCK_GROUP_FLAG_ZONED_DATA_RELOC, &block_group->runtime_flags)) {
+	    (!ffe_ctl->for_data_reloc &&
+	     test_bit(BLOCK_GROUP_FLAG_ZONED_DATA_RELOC, &block_group->runtime_flags))) {
 		ret = 1;
 		goto out;
 	}
@@ -3853,8 +3854,26 @@ static int do_allocation_zoned(struct bt
 	if (ffe_ctl->for_treelog && !fs_info->treelog_bg)
 		fs_info->treelog_bg = block_group->start;
 
-	if (ffe_ctl->for_data_reloc && !fs_info->data_reloc_bg)
-		fs_info->data_reloc_bg = block_group->start;
+	if (ffe_ctl->for_data_reloc) {
+		if (!fs_info->data_reloc_bg)
+			fs_info->data_reloc_bg = block_group->start;
+		/*
+		 * Do not allow allocations from this block group, unless it is
+		 * for data relocation. Compared to increasing the ->ro, setting
+		 * the ->zoned_data_reloc_ongoing flag still allows nocow
+		 * writers to come in. See btrfs_inc_nocow_writers().
+		 *
+		 * We need to disable an allocation to avoid an allocation of
+		 * regular (non-relocation data) extent. With mix of relocation
+		 * extents and regular extents, we can dispatch WRITE commands
+		 * (for relocation extents) and ZONE APPEND commands (for
+		 * regular extents) at the same time to the same zone, which
+		 * easily break the write pointer.
+		 *
+		 * Also, this flag avoids this block group to be zone finished.
+		 */
+		set_bit(BLOCK_GROUP_FLAG_ZONED_DATA_RELOC, &block_group->runtime_flags);
+	}
 
 	ffe_ctl->found_offset = start + block_group->alloc_offset;
 	block_group->alloc_offset += num_bytes;
@@ -3872,24 +3891,8 @@ static int do_allocation_zoned(struct bt
 out:
 	if (ret && ffe_ctl->for_treelog)
 		fs_info->treelog_bg = 0;
-	if (ret && ffe_ctl->for_data_reloc &&
-	    fs_info->data_reloc_bg == block_group->start) {
-		/*
-		 * Do not allow further allocations from this block group.
-		 * Compared to increasing the ->ro, setting the
-		 * ->zoned_data_reloc_ongoing flag still allows nocow
-		 *  writers to come in. See btrfs_inc_nocow_writers().
-		 *
-		 * We need to disable an allocation to avoid an allocation of
-		 * regular (non-relocation data) extent. With mix of relocation
-		 * extents and regular extents, we can dispatch WRITE commands
-		 * (for relocation extents) and ZONE APPEND commands (for
-		 * regular extents) at the same time to the same zone, which
-		 * easily break the write pointer.
-		 */
-		set_bit(BLOCK_GROUP_FLAG_ZONED_DATA_RELOC, &block_group->runtime_flags);
+	if (ret && ffe_ctl->for_data_reloc)
 		fs_info->data_reloc_bg = 0;
-	}
 	spin_unlock(&fs_info->relocation_bg_lock);
 	spin_unlock(&fs_info->treelog_bg_lock);
 	spin_unlock(&block_group->lock);
fs/btrfs/zoned.c +13 -3
@@ -2009,6 +2009,10 @@ static int do_zone_finish(struct btrfs_b
 	 * and block_group->meta_write_pointer for metadata.
 	 */
 	if (!fully_written) {
+		if (test_bit(BLOCK_GROUP_FLAG_ZONED_DATA_RELOC, &block_group->runtime_flags)) {
+			spin_unlock(&block_group->lock);
+			return -EAGAIN;
+		}
 		spin_unlock(&block_group->lock);
 
 		ret = btrfs_inc_block_group_ro(block_group, false);
@@ -2037,7 +2041,9 @@ static int do_zone_finish(struct btrfs_b
 			return 0;
 		}
 
-		if (block_group->reserved) {
+		if (block_group->reserved ||
+		    test_bit(BLOCK_GROUP_FLAG_ZONED_DATA_RELOC,
+			     &block_group->runtime_flags)) {
 			spin_unlock(&block_group->lock);
 			btrfs_dec_block_group_ro(block_group);
 			return -EAGAIN;
@@ -2268,7 +2274,10 @@ void btrfs_zoned_release_data_reloc_bg(s
 
 	/* All relocation extents are written. */
 	if (block_group->start + block_group->alloc_offset == logical + length) {
-		/* Now, release this block group for further allocations. */
+		/*
+		 * Now, release this block group for further allocations and
+		 * zone finish.
+		 */
 		clear_bit(BLOCK_GROUP_FLAG_ZONED_DATA_RELOC,
 			  &block_group->runtime_flags);
 	}
@@ -2292,7 +2301,8 @@ int btrfs_zone_finish_one_bg(struct btrf
 
 		spin_lock(&block_group->lock);
 		if (block_group->reserved || block_group->alloc_offset == 0 ||
-		    (block_group->flags & BTRFS_BLOCK_GROUP_SYSTEM)) {
+		    (block_group->flags & BTRFS_BLOCK_GROUP_SYSTEM) ||
+		    test_bit(BLOCK_GROUP_FLAG_ZONED_DATA_RELOC, &block_group->runtime_flags)) {
 			spin_unlock(&block_group->lock);
 			continue;
 		}
author maintainer
#184 lore raw

PATCH 156/219 btrfs: fix start transaction qgroup rsv double free

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Boris Burkov <[email protected]>

commit a6496849671a5bc9218ecec25a983253b34351b1 upstream.

btrfs_start_transaction reserves metadata space of the PERTRANS type before it identifies a transaction to start/join. This allows flushing when reserving that space without a deadlock. However, it results in a race which temporarily breaks qgroup rsv accounting.

T1 T2
start_transaction
do_stuff
start_transaction
qgroup_reserve_meta_pertrans
commit_transaction
qgroup_free_meta_all_pertrans
hit an error starting txn goto reserve_fail qgroup_free_meta_pertrans (already freed!)

The basic issue is that there is nothing preventing another commit from committing before start_transaction finishes (in fact sometimes we intentionally wait for it) so any error path that frees the reserve is at risk of this race.

While this exact space was getting freed anyway, and it's not a huge deal to double free it (just a warning, the free code catches this), it can result in incorrectly freeing some other pertrans reservation in this same reservation, which could then lead to spuriously granting reservations we might not have the space for. Therefore, I do believe it is worth fixing.

To fix it, use the existing prealloc->pertrans conversion mechanism. When we first reserve the space, we reserve prealloc space and only when we are sure we have a transaction do we convert it to pertrans. This way any racing commits do not blow away our reservation, but we still get a pertrans reservation that is freed when _this_ transaction gets committed.

This issue can be reproduced by running generic/269 with either qgroups or squotas enabled via mkfs on the scratch device.

---
 fs/btrfs/transaction.c |   19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)
fs/btrfs/transaction.c +16 -3
@@ -580,8 +580,13 @@ start_transaction(struct btrfs_root *roo
 		u64 delayed_refs_bytes = 0;
 
 		qgroup_reserved = num_items * fs_info->nodesize;
-		ret = btrfs_qgroup_reserve_meta_pertrans(root, qgroup_reserved,
-				enforce_qgroups);
+		/*
+		 * Use prealloc for now, as there might be a currently running
+		 * transaction that could free this reserved space prematurely
+		 * by committing.
+		 */
+		ret = btrfs_qgroup_reserve_meta_prealloc(root, qgroup_reserved,
+							 enforce_qgroups, false);
 		if (ret)
 			return ERR_PTR(ret);
 
@@ -693,6 +698,14 @@ again:
 		h->reloc_reserved = reloc_reserved;
 	}
 
+	/*
+	 * Now that we have found a transaction to be a part of, convert the
+	 * qgroup reservation from prealloc to pertrans. A different transaction
+	 * can't race in and free our pertrans out from under us.
+	 */
+	if (qgroup_reserved)
+		btrfs_qgroup_convert_reserved_meta(root, qgroup_reserved);
+
 got_it:
 	if (!current->journal_info)
 		current->journal_info = h;
@@ -740,7 +753,7 @@ alloc_fail:
 		btrfs_block_rsv_release(fs_info, &fs_info->trans_block_rsv,
 					num_bytes, NULL);
 reserve_fail:
-	btrfs_qgroup_free_meta_pertrans(root, qgroup_reserved);
+	btrfs_qgroup_free_meta_prealloc(root, qgroup_reserved);
 	return ERR_PTR(ret);
 }
 
author maintainer
#185 lore raw

PATCH 157/219 btrfs: free qgroup rsv on io failure

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Boris Burkov <[email protected]>

commit e28b02118b94e42be3355458a2406c6861e2dd32 upstream.

If we do a write whose bio suffers an error, we will never reclaim the qgroup reserved space for it. We allocate the space in the write_iter codepath, then release the reservation as we allocate the ordered extent, but we only create a delayed ref if the ordered extent finishes. If it has an error, we simply leak the rsv. This is apparent in running any error injecting (dmerror) fstests like btrfs/146 or btrfs/160. Such tests fail due to dmesg on umount complaining about the leaked qgroup data space.

When we clean up other aspects of space on failed ordered_extents, also free the qgroup rsv.

---
 fs/btrfs/inode.c |    7 +++++++
 1 file changed, 7 insertions(+)
fs/btrfs/inode.c +7 -0
@@ -3393,6 +3393,13 @@ out:
 			btrfs_free_reserved_extent(fs_info,
 					ordered_extent->disk_bytenr,
 					ordered_extent->disk_num_bytes, 1);
+			/*
+			 * Actually free the qgroup rsv which was released when
+			 * the ordered extent was created.
+			 */
+			btrfs_qgroup_free_refroot(fs_info, inode->root->root_key.objectid,
+						  ordered_extent->qgroup_rsv,
+						  BTRFS_QGROUP_RSV_DATA);
 		}
 	}
 
author maintainer
#186 lore raw

PATCH 158/219 btrfs: dont start transaction when joining with TRANS_JOIN_NOSTART

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Filipe Manana <[email protected]>

commit 4490e803e1fe9fab8db5025e44e23b55df54078b upstream.

When joining a transaction with TRANS_JOIN_NOSTART, if we don't find a running transaction we end up creating one. This goes against the purpose of TRANS_JOIN_NOSTART which is to join a running transaction if its state is at or below the state TRANS_STATE_COMMIT_START, otherwise return an -ENOENT error and don't start a new transaction. So fix this to not create a new transaction if there's no running transaction at or below that state.

Fixes a6d155d2e363 ("Btrfs: fix deadlock between fiemap and transaction commits")
---
 fs/btrfs/transaction.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
fs/btrfs/transaction.c +4 -3
@@ -279,10 +279,11 @@ loop:
 	spin_unlock(&fs_info->trans_lock);
 
 	/*
-	 * If we are ATTACH, we just want to catch the current transaction,
-	 * and commit it. If there is no transaction, just return ENOENT.
+	 * If we are ATTACH or TRANS_JOIN_NOSTART, we just want to catch the
+	 * current transaction, and commit it. If there is no transaction, just
+	 * return ENOENT.
 	 */
-	if (type == TRANS_ATTACH)
+	if (type == TRANS_ATTACH || type == TRANS_JOIN_NOSTART)
 		return -ENOENT;
 
 	/*
author maintainer
#187 lore raw

PATCH 159/219 btrfs: set page extent mapped after read_folio in relocate_one_page

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Josef Bacik <[email protected]>

commit e7f1326cc24e22b38afc3acd328480a1183f9e79 upstream.

One of the CI runs triggered the following panic

assertion failed: PagePrivate(page) && page->private, in fs/btrfs/subpage.c:229

  ------------[ cut here ]------------
  kernel BUG at fs/btrfs/subpage.c:229!
  Internal error: Oops - BUG: 00000000f2000800 [#1] SMP
  CPU: 0 PID: 923660 Comm: btrfs Not tainted 6.5.0-rc3+ #1
  pstate: 61400005 (nZCv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--)
  pc : btrfs_subpage_assert+0xbc/0xf0
  lr : btrfs_subpage_assert+0xbc/0xf0
  sp : ffff800093213720
  x29: ffff800093213720 x28: ffff8000932138b4 x27: 000000000c280000
  x26: 00000001b5d00000 x25: 000000000c281000 x24: 000000000c281fff
  x23: 0000000000001000 x22: 0000000000000000 x21: ffffff42b95bf880
  x20: ffff42b9528e0000 x19: 0000000000001000 x18: ffffffffffffffff
  x17: 667274622f736620 x16: 6e69202c65746176 x15: 0000000000000028
  x14: 0000000000000003 x13: 00000000002672d7 x12: 0000000000000000
  x11: ffffcd3f0ccd9204 x10: ffffcd3f0554ae50 x9 : ffffcd3f0379528c
  x8 : ffff800093213428 x7 : 0000000000000000 x6 : ffffcd3f091771e8
  x5 : ffff42b97f333948 x4 : 0000000000000000 x3 : 0000000000000000
  x2 : 0000000000000000 x1 : ffff42b9556cde80 x0 : 000000000000004f
  Call trace:
   btrfs_subpage_assert+0xbc/0xf0
   btrfs_subpage_set_dirty+0x38/0xa0
   btrfs_page_set_dirty+0x58/0x88
   relocate_one_page+0x204/0x5f0
   relocate_file_extent_cluster+0x11c/0x180
   relocate_data_extent+0xd0/0xf8
   relocate_block_group+0x3d0/0x4e8
   btrfs_relocate_block_group+0x2d8/0x490
   btrfs_relocate_chunk+0x54/0x1a8
   btrfs_balance+0x7f4/0x1150
   btrfs_ioctl+0x10f0/0x20b8
   __arm64_sys_ioctl+0x120/0x11d8
   invoke_syscall.constprop.0+0x80/0xd8
   do_el0_svc+0x6c/0x158
   el0_svc+0x50/0x1b0
   el0t_64_sync_handler+0x120/0x130
   el0t_64_sync+0x194/0x198
  Code: 91098021 b0007fa0 91346000 97e9c6d2 (d4210000)

This is the same problem outlined in 17b17fcd6d44 ("btrfs:
set_page_extent_mapped after read_folio in btrfs_cont_expand") , and the fix is the same. I originally looked for the same pattern elsewhere in our code, but mistakenly skipped over this code because I saw the page cache readahead before we set_page_extent_mapped, not realizing that this was only in the !page case, that we can still end up with a !uptodate page and then do the btrfs_read_folio further down.

The fix here is the same as the above mentioned patch, move the set_page_extent_mapped call to after the btrfs_read_folio() block to make sure that we have the subpage blocksize stuff setup properly before using the page.

---
 fs/btrfs/relocation.c |   12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)
fs/btrfs/relocation.c +9 -3
@@ -2985,9 +2985,6 @@ static int relocate_one_page(struct inod
 		if (!page)
 			return -ENOMEM;
 	}
-	ret = set_page_extent_mapped(page);
-	if (ret < 0)
-		goto release_page;
 
 	if (PageReadahead(page))
 		page_cache_async_readahead(inode->i_mapping, ra, NULL,
@@ -3003,6 +3000,15 @@ static int relocate_one_page(struct inod
 		}
 	}
 
+	/*
+	 * We could have lost page private when we dropped the lock to read the
+	 * page above, make sure we set_page_extent_mapped here so we have any
+	 * of the subpage blocksize stuff we need in place.
+	 */
+	ret = set_page_extent_mapped(page);
+	if (ret < 0)
+		goto release_page;
+
 	page_start = page_offset(page);
 	page_end = page_start + PAGE_SIZE - 1;
 
author maintainer
#188 lore raw

PATCH 160/219 btrfs: zoned: re-enable metadata over-commit for zoned mode

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Naohiro Aota <[email protected]>

commit 5b135b382a360f4c87cf8896d1465b0b07f10cb0 upstream.

Now that, we can re-enable metadata over-commit. As we moved the activation from the reservation time to the write time, we no longer need to ensure all the reserved bytes is properly activated.

Without the metadata over-commit, it suffers from lower performance because it needs to flush the delalloc items more often and allocate more block groups. Re-enabling metadata over-commit will solve the issue.

Fixes 79417d040f4f ("btrfs: zoned: disable metadata overcommit for zoned")
---
 fs/btrfs/space-info.c |    6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)
fs/btrfs/space-info.c +1 -5
@@ -404,11 +404,7 @@ int btrfs_can_overcommit(struct btrfs_fs
 		return 0;
 
 	used = btrfs_space_info_used(space_info, true);
-	if (test_bit(BTRFS_FS_ACTIVE_ZONE_TRACKING, &fs_info->flags) &&
-	    (space_info->flags & BTRFS_BLOCK_GROUP_METADATA))
-		avail = 0;
-	else
-		avail = calc_available_free_space(fs_info, space_info, flush);
+	avail = calc_available_free_space(fs_info, space_info, flush);
 
 	if (used + bytes < writable_total_bytes(fs_info, space_info) + avail)
 		return 1;
author maintainer
#189 lore raw

PATCH 161/219 btrfs: use the correct superblock to compare fsid in btrfs_validate_super

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Anand Jain <[email protected]>

commit d167aa76dc0683828588c25767da07fb549e4f48 upstream.

The function btrfs_validate_super() should verify the fsid in the provided superblock argument. Because, all its callers expect it to do that.

Such as in the following stack:

   write_all_supers()
       sb = fs_info->super_for_commit;
       btrfs_validate_write_super(.., sb)
         btrfs_validate_super(.., sb, ..)

scrub_one_super()

	btrfs_validate_super(.., sb, ..)

And
check_dev_super()

	btrfs_validate_super(.., sb, ..)

However, it currently verifies the fs_info::super_copy::fsid instead, which is not correct. Fix this using the correct fsid in the superblock argument.

---
 fs/btrfs/disk-io.c |    5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
fs/btrfs/disk-io.c +2 -3
@@ -2721,11 +2721,10 @@ int btrfs_validate_super(struct btrfs_fs
 		ret = -EINVAL;
 	}
 
-	if (memcmp(fs_info->fs_devices->fsid, fs_info->super_copy->fsid,
-		   BTRFS_FSID_SIZE)) {
+	if (memcmp(fs_info->fs_devices->fsid, sb->fsid, BTRFS_FSID_SIZE) != 0) {
 		btrfs_err(fs_info,
 		"superblock fsid doesn't match fsid of fs_devices: %pU != %pU",
-			fs_info->super_copy->fsid, fs_info->fs_devices->fsid);
+			  sb->fsid, fs_info->fs_devices->fsid);
 		ret = -EINVAL;
 	}
 
author maintainer
#190 lore raw

PATCH 162/219 drm/mxsfb: Disable overlay plane in mxsfb_plane_overlay_atomic_disable()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Liu Ying <[email protected]>

commit aa656d48e871a1b062e1bbf9474d8b831c35074c upstream.

When disabling overlay plane in mxsfb_plane_overlay_atomic_update(), overlay plane's framebuffer pointer is NULL. So, dereferencing it would cause a kernel Oops(NULL pointer dereferencing). Fix the issue by disabling overlay plane in mxsfb_plane_overlay_atomic_disable() instead.

Fixes cb285a5348e7 ("drm: mxsfb: Replace mxsfb_get_fb_paddr() with drm_fb_cma_get_gem_addr()")
---
 drivers/gpu/drm/mxsfb/mxsfb_kms.c |    9 +++++++++
 1 file changed, 9 insertions(+)
drivers/gpu/drm/mxsfb/mxsfb_kms.c +9 -0
@@ -611,6 +611,14 @@ static void mxsfb_plane_overlay_atomic_u
 	writel(ctrl, mxsfb->base + LCDC_AS_CTRL);
 }
 
+static void mxsfb_plane_overlay_atomic_disable(struct drm_plane *plane,
+					       struct drm_atomic_state *state)
+{
+	struct mxsfb_drm_private *mxsfb = to_mxsfb_drm_private(plane->dev);
+
+	writel(0, mxsfb->base + LCDC_AS_CTRL);
+}
+
 static bool mxsfb_format_mod_supported(struct drm_plane *plane,
 				       uint32_t format,
 				       uint64_t modifier)
@@ -626,6 +634,7 @@ static const struct drm_plane_helper_fun
 static const struct drm_plane_helper_funcs mxsfb_plane_overlay_helper_funcs = {
 	.atomic_check = mxsfb_plane_atomic_check,
 	.atomic_update = mxsfb_plane_overlay_atomic_update,
+	.atomic_disable = mxsfb_plane_overlay_atomic_disable,
 };
 
 static const struct drm_plane_funcs mxsfb_plane_funcs = {
author maintainer
#191 lore raw

PATCH 163/219 mtd: rawnand: brcmnand: Fix crash during the panic_write

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: William Zhang <[email protected]>

commit e66dd317194daae0475fe9e5577c80aa97f16cb9 upstream.

When executing a NAND command within the panic write path, wait for any pending command instead of calling BUG_ON to avoid crashing while already crashing.

---
 drivers/mtd/nand/raw/brcmnand/brcmnand.c |   12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)
drivers/mtd/nand/raw/brcmnand/brcmnand.c +11 -1
@@ -1592,7 +1592,17 @@ static void brcmnand_send_cmd(struct brc
 
 	dev_dbg(ctrl->dev, "send native cmd %d addr 0x%llx\n", cmd, cmd_addr);
 
-	BUG_ON(ctrl->cmd_pending != 0);
+	/*
+	 * If we came here through _panic_write and there is a pending
+	 * command, try to wait for it. If it times out, rather than
+	 * hitting BUG_ON, just return so we don't crash while crashing.
+	 */
+	if (oops_in_progress) {
+		if (ctrl->cmd_pending &&
+			bcmnand_ctrl_poll_status(ctrl, NAND_CTRL_RDY, NAND_CTRL_RDY, 0))
+			return;
+	} else
+		BUG_ON(ctrl->cmd_pending != 0);
 	ctrl->cmd_pending = cmd;
 
 	ret = bcmnand_ctrl_poll_status(ctrl, NAND_CTRL_RDY, NAND_CTRL_RDY, 0);
author maintainer
#192 lore raw

PATCH 164/219 mtd: rawnand: brcmnand: Fix potential out-of-bounds access in oob write

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: William Zhang <[email protected]>

commit 5d53244186c9ac58cb88d76a0958ca55b83a15cd upstream.

When the oob buffer length is not in multiple of words, the oob write function does out-of-bounds read on the oob source buffer at the last iteration. Fix that by always checking length limit on the oob buffer read and fill with 0xff when reaching the end of the buffer to the oob registers.

Fixes 27c5b17cd1b1 ("mtd: nand: add NAND driver "library" for Broadcom STB NAND controller")
---
 drivers/mtd/nand/raw/brcmnand/brcmnand.c |   18 ++++++++++++++++--
 1 file changed, 16 insertions(+), 2 deletions(-)
drivers/mtd/nand/raw/brcmnand/brcmnand.c +16 -2
@@ -1461,19 +1461,33 @@ static int write_oob_to_regs(struct brcm
 			     const u8 *oob, int sas, int sector_1k)
 {
 	int tbytes = sas << sector_1k;
-	int j;
+	int j, k = 0;
+	u32 last = 0xffffffff;
+	u8 *plast = (u8 *)&last;
 
 	/* Adjust OOB values for 1K sector size */
 	if (sector_1k && (i & 0x01))
 		tbytes = max(0, tbytes - (int)ctrl->max_oob);
 	tbytes = min_t(int, tbytes, ctrl->max_oob);
 
-	for (j = 0; j < tbytes; j += 4)
+	/*
+	 * tbytes may not be multiple of words. Make sure we don't read out of
+	 * the boundary and stop at last word.
+	 */
+	for (j = 0; (j + 3) < tbytes; j += 4)
 		oob_reg_write(ctrl, j,
 				(oob[j + 0] << 24) |
 				(oob[j + 1] << 16) |
 				(oob[j + 2] <<  8) |
 				(oob[j + 3] <<  0));
+
+	/* handle the remaing bytes */
+	while (j < tbytes)
+		plast[k++] = oob[j++];
+
+	if (tbytes & 0x3)
+		oob_reg_write(ctrl, (tbytes & ~0x3), (__force u32)cpu_to_be32(last));
+
 	return tbytes;
 }
 
author maintainer
#193 lore raw

PATCH 165/219 mtd: spi-nor: Correct flags for Winbond w25q128

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Linus Walleij <[email protected]>

commit 83e824a4a595132f9bd7ac4f5afff857bfc5991e upstream.

The Winbond "w25q128" (actual vendor name W25Q128JV) has exactly the same flags as the sibling device "w25q128jv". The devices both require unlocking to enable write access.

The actual product naming between devices vs the Linux strings in winbond.c:

0xef4018: "w25q128" W25Q128JV-IN/IQ/JQ
0xef7018: "w25q128jv" W25Q128JV-IM/JM

The latter device, "w25q128jv" supports features named DTQ and QPI, otherwise it is the same.

Not having the right flags has the annoying side effect that write access does not work.

After this patch I can write to the flash on the Inteno XG6846 router.

The flash memory also supports dual and quad SPI modes. This does not currently manifest, but by turning on SFDP parsing, the right SPI modes are emitted in /sys/kernel/debug/spi-nor/spi1.0/capabilities for this chip, so we also turn on this.

Since we now have determined that SFDP parsing works on the device, we also detect the geometry using SFDP.

After this dmesg and sysfs says:

[    1.062401] spi-nor spi1.0: w25q128 (16384 Kbytes)

cat erasesize
65536
(16384*1024)/65536 = 256 sectors

spi-nor sysfs:
cat jedec_id
ef4018
cat manufacturer
winbond
cat partname
w25q128
hexdump -v -C sfdp

00000000  53 46 44 50 05 01 00 ff  00 05 01 10 80 00 00 ff
00000010  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff
00000020  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff
00000030  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff
00000040  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff
00000050  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff
00000060  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff
00000070  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff
00000080  e5 20 f9 ff ff ff ff 07  44 eb 08 6b 08 3b 42 bb
00000090  fe ff ff ff ff ff 00 00  ff ff 40 eb 0c 20 0f 52
000000a0  10 d8 00 00 36 02 a6 00  82 ea 14 c9 e9 63 76 33
000000b0  7a 75 7a 75 f7 a2 d5 5c  19 f7 4d ff e9 30 f8 80
---
 drivers/mtd/spi-nor/winbond.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/mtd/spi-nor/winbond.c +3 -2
@@ -120,8 +120,9 @@ static const struct flash_info winbond_n
 		NO_SFDP_FLAGS(SECT_4K) },
 	{ "w25q80bl", INFO(0xef4014, 0, 64 * 1024,  16)
 		NO_SFDP_FLAGS(SECT_4K) },
-	{ "w25q128", INFO(0xef4018, 0, 64 * 1024, 256)
-		NO_SFDP_FLAGS(SECT_4K) },
+	{ "w25q128", INFO(0xef4018, 0, 0, 0)
+		PARSE_SFDP
+		FLAGS(SPI_NOR_HAS_LOCK | SPI_NOR_HAS_TB) },
 	{ "w25q256", INFO(0xef4019, 0, 64 * 1024, 512)
 		NO_SFDP_FLAGS(SECT_4K | SPI_NOR_DUAL_READ | SPI_NOR_QUAD_READ)
 		.fixups = &w25q256_fixups },
author maintainer
#194 lore raw

PATCH 166/219 mtd: rawnand: brcmnand: Fix potential false time out warning

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: William Zhang <[email protected]>

commit 9cc0a598b944816f2968baf2631757f22721b996 upstream.

If system is busy during the command status polling function, the driver may not get the chance to poll the status register till the end of time out and return the premature status. Do a final check after time out happens to ensure reading the correct status.

Fixes 9d2ee0a60b8b ("mtd: nand: brcmnand: Check flash #WP pin status before nand erase/program")
---
 drivers/mtd/nand/raw/brcmnand/brcmnand.c |    8 ++++++++
 1 file changed, 8 insertions(+)
drivers/mtd/nand/raw/brcmnand/brcmnand.c +8 -0
@@ -1072,6 +1072,14 @@ static int bcmnand_ctrl_poll_status(stru
 		cpu_relax();
 	} while (time_after(limit, jiffies));
 
+	/*
+	 * do a final check after time out in case the CPU was busy and the driver
+	 * did not get enough time to perform the polling to avoid false alarms
+	 */
+	val = brcmnand_read_reg(ctrl, BRCMNAND_INTFC_STATUS);
+	if ((val & mask) == expected_val)
+		return 0;
+
 	dev_warn(ctrl->dev, "timeout on status poll (expected %x got %x)\n",
 		 expected_val, val & mask);
 
author maintainer
#195 lore raw

PATCH 167/219 mtd: rawnand: brcmnand: Fix ECC level field setting for v7.2 controller

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: William Zhang <[email protected]>

commit 2ec2839a9062db8a592525a3fdabd42dcd9a3a9b upstream.

v7.2 controller has different ECC level field size and shift in the acc control register than its predecessor and successor controller. It needs to be set specifically.

---
 drivers/mtd/nand/raw/brcmnand/brcmnand.c |   74 +++++++++++++++++--------------
 1 file changed, 41 insertions(+), 33 deletions(-)
drivers/mtd/nand/raw/brcmnand/brcmnand.c +41 -33
@@ -272,6 +272,7 @@ struct brcmnand_controller {
 	const unsigned int	*page_sizes;
 	unsigned int		page_size_shift;
 	unsigned int		max_oob;
+	u32			ecc_level_shift;
 	u32			features;
 
 	/* for low-power standby/resume only */
@@ -596,6 +597,34 @@ enum {
 	INTFC_CTLR_READY		= BIT(31),
 };
 
+/***********************************************************************
+ * NAND ACC CONTROL bitfield
+ *
+ * Some bits have remained constant throughout hardware revision, while
+ * others have shifted around.
+ ***********************************************************************/
+
+/* Constant for all versions (where supported) */
+enum {
+	/* See BRCMNAND_HAS_CACHE_MODE */
+	ACC_CONTROL_CACHE_MODE				= BIT(22),
+
+	/* See BRCMNAND_HAS_PREFETCH */
+	ACC_CONTROL_PREFETCH				= BIT(23),
+
+	ACC_CONTROL_PAGE_HIT				= BIT(24),
+	ACC_CONTROL_WR_PREEMPT				= BIT(25),
+	ACC_CONTROL_PARTIAL_PAGE			= BIT(26),
+	ACC_CONTROL_RD_ERASED				= BIT(27),
+	ACC_CONTROL_FAST_PGM_RDIN			= BIT(28),
+	ACC_CONTROL_WR_ECC				= BIT(30),
+	ACC_CONTROL_RD_ECC				= BIT(31),
+};
+
+#define	ACC_CONTROL_ECC_SHIFT			16
+/* Only for v7.2 */
+#define	ACC_CONTROL_ECC_EXT_SHIFT		13
+
 static inline bool brcmnand_non_mmio_ops(struct brcmnand_controller *ctrl)
 {
 #if IS_ENABLED(CONFIG_MTD_NAND_BRCMNAND_BCMA)
@@ -737,6 +766,12 @@ static int brcmnand_revision_init(struct
 	else if (of_property_read_bool(ctrl->dev->of_node, "brcm,nand-has-wp"))
 		ctrl->features |= BRCMNAND_HAS_WP;
 
+	/* v7.2 has different ecc level shift in the acc register */
+	if (ctrl->nand_version == 0x0702)
+		ctrl->ecc_level_shift = ACC_CONTROL_ECC_EXT_SHIFT;
+	else
+		ctrl->ecc_level_shift = ACC_CONTROL_ECC_SHIFT;
+
 	return 0;
 }
 
@@ -931,30 +966,6 @@ static inline int brcmnand_cmd_shift(str
 	return 0;
 }
 
-/***********************************************************************
- * NAND ACC CONTROL bitfield
- *
- * Some bits have remained constant throughout hardware revision, while
- * others have shifted around.
- ***********************************************************************/
-
-/* Constant for all versions (where supported) */
-enum {
-	/* See BRCMNAND_HAS_CACHE_MODE */
-	ACC_CONTROL_CACHE_MODE				= BIT(22),
-
-	/* See BRCMNAND_HAS_PREFETCH */
-	ACC_CONTROL_PREFETCH				= BIT(23),
-
-	ACC_CONTROL_PAGE_HIT				= BIT(24),
-	ACC_CONTROL_WR_PREEMPT				= BIT(25),
-	ACC_CONTROL_PARTIAL_PAGE			= BIT(26),
-	ACC_CONTROL_RD_ERASED				= BIT(27),
-	ACC_CONTROL_FAST_PGM_RDIN			= BIT(28),
-	ACC_CONTROL_WR_ECC				= BIT(30),
-	ACC_CONTROL_RD_ECC				= BIT(31),
-};
-
 static inline u32 brcmnand_spare_area_mask(struct brcmnand_controller *ctrl)
 {
 	if (ctrl->nand_version == 0x0702)
@@ -967,18 +978,15 @@ static inline u32 brcmnand_spare_area_ma
 		return GENMASK(4, 0);
 }
 
-#define NAND_ACC_CONTROL_ECC_SHIFT	16
-#define NAND_ACC_CONTROL_ECC_EXT_SHIFT	13
-
 static inline u32 brcmnand_ecc_level_mask(struct brcmnand_controller *ctrl)
 {
 	u32 mask = (ctrl->nand_version >= 0x0600) ? 0x1f : 0x0f;
 
-	mask <<= NAND_ACC_CONTROL_ECC_SHIFT;
+	mask <<= ACC_CONTROL_ECC_SHIFT;
 
 	/* v7.2 includes additional ECC levels */
-	if (ctrl->nand_version >= 0x0702)
-		mask |= 0x7 << NAND_ACC_CONTROL_ECC_EXT_SHIFT;
+	if (ctrl->nand_version == 0x0702)
+		mask |= 0x7 << ACC_CONTROL_ECC_EXT_SHIFT;
 
 	return mask;
 }
@@ -992,8 +1000,8 @@ static void brcmnand_set_ecc_enabled(str
 
 	if (en) {
 		acc_control |= ecc_flags; /* enable RD/WR ECC */
-		acc_control |= host->hwcfg.ecc_level
-			       << NAND_ACC_CONTROL_ECC_SHIFT;
+		acc_control &= ~brcmnand_ecc_level_mask(ctrl);
+		acc_control |= host->hwcfg.ecc_level << ctrl->ecc_level_shift;
 	} else {
 		acc_control &= ~ecc_flags; /* disable RD/WR ECC */
 		acc_control &= ~brcmnand_ecc_level_mask(ctrl);
@@ -2593,7 +2601,7 @@ static int brcmnand_set_cfg(struct brcmn
 	tmp &= ~brcmnand_ecc_level_mask(ctrl);
 	tmp &= ~brcmnand_spare_area_mask(ctrl);
 	if (ctrl->nand_version >= 0x0302) {
-		tmp |= cfg->ecc_level << NAND_ACC_CONTROL_ECC_SHIFT;
+		tmp |= cfg->ecc_level << ctrl->ecc_level_shift;
 		tmp |= cfg->spare_area_size;
 	}
 	nand_writereg(ctrl, acc_control_offs, tmp);
author maintainer
#196 lore raw

PATCH 168/219 drm/amd/display: enable cursor degamma for DCN3+ DRM legacy gamma

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Melissa Wen <[email protected]>

commit 57a943ebfcdb4a97fbb409640234bdb44bfa1953 upstream.

For DRM legacy gamma, AMD display manager applies implicit sRGB degamma using a pre-defined sRGB transfer function. It works fine for DCN2 family where degamma ROM and custom curves go to the same color block. But, on DCN3+, degamma is split into two blocks: degamma ROM for pre-defined TFs and gamma correction for user/custom curves and degamma ROM settings doesn't apply to cursor plane. To get DRM legacy gamma working as expected, enable cursor degamma ROM for implict sRGB degamma on HW with this configuration.

Fixes 96b020e2163f ("drm/amd/display: check attr flag before set cursor degamma on DCN3+")
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c |    7 +++++++
 1 file changed, 7 insertions(+)
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c +7 -0
@@ -1269,6 +1269,13 @@ void handle_cursor_update(struct drm_pla
 	attributes.rotation_angle    = 0;
 	attributes.attribute_flags.value = 0;
 
+	/* Enable cursor degamma ROM on DCN3+ for implicit sRGB degamma in DRM
+	 * legacy gamma setup.
+	 */
+	if (crtc_state->cm_is_degamma_srgb &&
+	    adev->dm.dc->caps.color.dpp.gamma_corr)
+		attributes.attribute_flags.bits.ENABLE_CURSOR_DEGAMMA = 1;
+
 	attributes.pitch = afb->base.pitches[0] / afb->base.format->cpp[0];
 
 	if (crtc_state->stream) {
author maintainer
#197 lore raw

PATCH 169/219 drm/amd/display: prevent potential division by zero errors

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hamza Mahfooz <[email protected]>

commit 07e388aab042774f284a2ad75a70a194517cdad4 upstream.

There are two places in apply_below_the_range() where it's possible for a divide by zero error to occur. So, to fix this make sure the divisor is non-zero before attempting the computation in both cases.

Fixes a463b263032f ("drm/amd/display: Fix frames_to_insert math")
Fixes ded6119e825a ("drm/amd/display: Reinstate LFC optimization")
---
 drivers/gpu/drm/amd/display/modules/freesync/freesync.c |    9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
drivers/gpu/drm/amd/display/modules/freesync/freesync.c +6 -3
@@ -338,7 +338,9 @@ static void apply_below_the_range(struct
 		 *  - Delta for CEIL: delta_from_mid_point_in_us_1
 		 *  - Delta for FLOOR: delta_from_mid_point_in_us_2
 		 */
-		if ((last_render_time_in_us / mid_point_frames_ceil) < in_out_vrr->min_duration_in_us) {
+		if (mid_point_frames_ceil &&
+		    (last_render_time_in_us / mid_point_frames_ceil) <
+		    in_out_vrr->min_duration_in_us) {
 			/* Check for out of range.
 			 * If using CEIL produces a value that is out of range,
 			 * then we are forced to use FLOOR.
@@ -385,8 +387,9 @@ static void apply_below_the_range(struct
 		/* Either we've calculated the number of frames to insert,
 		 * or we need to insert min duration frames
 		 */
-		if (last_render_time_in_us / frames_to_insert <
-				in_out_vrr->min_duration_in_us){
+		if (frames_to_insert &&
+		    (last_render_time_in_us / frames_to_insert) <
+		    in_out_vrr->min_duration_in_us){
 			frames_to_insert -= (frames_to_insert > 1) ?
 					1 : 0;
 		}
author maintainer
#198 lore raw

PATCH 170/219 KVM: SVM: Take and hold ir_list_lock when updating vCPUs Physical ID entry

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sean Christopherson <[email protected]>

commit 4c08e737f056fec930b416a2bd37ed266d724f95 upstream.

Hoist the acquisition of ir_list_lock from avic_update_iommu_vcpu_affinity() to its two callers, avic_vcpu_load() and avic_vcpu_put(), specifically to encapsulate the write to the vCPU's entry in the AVIC Physical ID table. This will allow a future fix to pull information from the Physical ID entry when updating the IRTE, without potentially consuming stale information, i.e. without racing with the vCPU being (un)loaded.

Add a comment to call out that ir_list_lock does NOT protect against multiple writers, specifically that reading the Physical ID entry in avic_vcpu_put() outside of the lock is safe.

To preserve some semblance of independence from ir_list_lock, keep the READ_ONCE() in avic_vcpu_load() even though acuiring the spinlock effectively ensures the load(s) will be generated after acquiring the lock.

---
 arch/x86/kvm/svm/avic.c |   31 +++++++++++++++++++++++--------
 1 file changed, 23 insertions(+), 8 deletions(-)
arch/x86/kvm/svm/avic.c +23 -8
@@ -1022,10 +1022,11 @@ static inline int
 avic_update_iommu_vcpu_affinity(struct kvm_vcpu *vcpu, int cpu, bool r)
 {
 	int ret = 0;
-	unsigned long flags;
 	struct amd_svm_iommu_ir *ir;
 	struct vcpu_svm *svm = to_svm(vcpu);
 
+	lockdep_assert_held(&svm->ir_list_lock);
+
 	if (!kvm_arch_has_assigned_device(vcpu->kvm))
 		return 0;
 
@@ -1033,19 +1034,15 @@ avic_update_iommu_vcpu_affinity(struct k
 	 * Here, we go through the per-vcpu ir_list to update all existing
 	 * interrupt remapping table entry targeting this vcpu.
 	 */
-	spin_lock_irqsave(&svm->ir_list_lock, flags);
-
 	if (list_empty(&svm->ir_list))
-		goto out;
+		return 0;
 
 	list_for_each_entry(ir, &svm->ir_list, node) {
 		ret = amd_iommu_update_ga(cpu, r, ir->data);
 		if (ret)
-			break;
+			return ret;
 	}
-out:
-	spin_unlock_irqrestore(&svm->ir_list_lock, flags);
-	return ret;
+	return 0;
 }
 
 void avic_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
@@ -1053,6 +1050,7 @@ void avic_vcpu_load(struct kvm_vcpu *vcp
 	u64 entry;
 	int h_physical_id = kvm_cpu_get_apicid(cpu);
 	struct vcpu_svm *svm = to_svm(vcpu);
+	unsigned long flags;
 
 	lockdep_assert_preemption_disabled();
 
@@ -1069,6 +1067,8 @@ void avic_vcpu_load(struct kvm_vcpu *vcp
 	if (kvm_vcpu_is_blocking(vcpu))
 		return;
 
+	spin_lock_irqsave(&svm->ir_list_lock, flags);
+
 	entry = READ_ONCE(*(svm->avic_physical_id_cache));
 
 	entry &= ~AVIC_PHYSICAL_ID_ENTRY_HOST_PHYSICAL_ID_MASK;
@@ -1077,25 +1077,40 @@ void avic_vcpu_load(struct kvm_vcpu *vcp
 
 	WRITE_ONCE(*(svm->avic_physical_id_cache), entry);
 	avic_update_iommu_vcpu_affinity(vcpu, h_physical_id, true);
+
+	spin_unlock_irqrestore(&svm->ir_list_lock, flags);
 }
 
 void avic_vcpu_put(struct kvm_vcpu *vcpu)
 {
 	u64 entry;
 	struct vcpu_svm *svm = to_svm(vcpu);
+	unsigned long flags;
 
 	lockdep_assert_preemption_disabled();
 
+	/*
+	 * Note, reading the Physical ID entry outside of ir_list_lock is safe
+	 * as only the pCPU that has loaded (or is loading) the vCPU is allowed
+	 * to modify the entry, and preemption is disabled.  I.e. the vCPU
+	 * can't be scheduled out and thus avic_vcpu_{put,load}() can't run
+	 * recursively.
+	 */
 	entry = READ_ONCE(*(svm->avic_physical_id_cache));
 
 	/* Nothing to do if IsRunning == '0' due to vCPU blocking. */
 	if (!(entry & AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK))
 		return;
 
+	spin_lock_irqsave(&svm->ir_list_lock, flags);
+
 	avic_update_iommu_vcpu_affinity(vcpu, -1, 0);
 
 	entry &= ~AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK;
 	WRITE_ONCE(*(svm->avic_physical_id_cache), entry);
+
+	spin_unlock_irqrestore(&svm->ir_list_lock, flags);
+
 }
 
 void avic_refresh_virtual_apic_mode(struct kvm_vcpu *vcpu)
author maintainer
#199 lore raw

PATCH 171/219 KVM: SVM: Dont inject #UD if KVM attempts to skip SEV guest insn

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sean Christopherson <[email protected]>

commit cb49631ad111570f1bad37702c11c2ae07fa2e3c upstream.

Don't inject a #UD if KVM attempts to "emulate" to skip an instruction for an SEV guest, and instead resume the guest and hope that it can make forward progress. When commit 04c40f344def ("KVM: SVM: Inject #UD on attempted emulation for SEV guest w/o insn buffer") added the completely arbitrary #UD behavior, there were no known scenarios where a well-behaved guest would induce a VM-Exit that triggered emulation, i.e. it was thought that injecting #UD would be helpful.

However, now that KVM (correctly) attempts to re-inject INT3/INTO, e.g. if a #NPF is encountered when attempting to deliver the INT3/INTO, an SEV guest can trigger emulation without a buffer, through no fault of its own. Resuming the guest and retrying the INT3/INTO is architecturally wrong, e.g. the vCPU will incorrectly re-hit code #DBs, but for SEV guests there is literally no other option that has a chance of making forward progress.

Drop the #UD injection for all "skip" emulation, not just those related to INT3/INTO, even though that means that the guest will likely end up in an infinite loop instead of getting a #UD (the vCPU may also crash, e.g. if KVM emulated everything about an instruction except for advancing RIP). There's no evidence that suggests that an unexpected #UD is actually better than hanging the vCPU, e.g. a soft-hung vCPU can still respond to IRQs and NMIs to generate a backtrace.

---
 arch/x86/kvm/svm/svm.c |   35 +++++++++++++++++++++++++++--------
 1 file changed, 27 insertions(+), 8 deletions(-)
arch/x86/kvm/svm/svm.c +27 -8
@@ -366,6 +366,8 @@ static void svm_set_interrupt_shadow(str
 		svm->vmcb->control.int_state |= SVM_INTERRUPT_SHADOW_MASK;
 
 }
+static bool svm_can_emulate_instruction(struct kvm_vcpu *vcpu, int emul_type,
+					void *insn, int insn_len);
 
 static int __svm_skip_emulated_instruction(struct kvm_vcpu *vcpu,
 					   bool commit_side_effects)
@@ -386,6 +388,14 @@ static int __svm_skip_emulated_instructi
 	}
 
 	if (!svm->next_rip) {
+		/*
+		 * FIXME: Drop this when kvm_emulate_instruction() does the
+		 * right thing and treats "can't emulate" as outright failure
+		 * for EMULTYPE_SKIP.
+		 */
+		if (!svm_can_emulate_instruction(vcpu, EMULTYPE_SKIP, NULL, 0))
+			return 0;
+
 		if (unlikely(!commit_side_effects))
 			old_rflags = svm->vmcb->save.rflags;
 
@@ -4592,16 +4602,25 @@ static bool svm_can_emulate_instruction(
 	 * and cannot be decrypted by KVM, i.e. KVM would read cyphertext and
 	 * decode garbage.
 	 *
-	 * Inject #UD if KVM reached this point without an instruction buffer.
-	 * In practice, this path should never be hit by a well-behaved guest,
-	 * e.g. KVM doesn't intercept #UD or #GP for SEV guests, but this path
-	 * is still theoretically reachable, e.g. via unaccelerated fault-like
-	 * AVIC access, and needs to be handled by KVM to avoid putting the
-	 * guest into an infinite loop.   Injecting #UD is somewhat arbitrary,
-	 * but its the least awful option given lack of insight into the guest.
+	 * If KVM is NOT trying to simply skip an instruction, inject #UD if
+	 * KVM reached this point without an instruction buffer.  In practice,
+	 * this path should never be hit by a well-behaved guest, e.g. KVM
+	 * doesn't intercept #UD or #GP for SEV guests, but this path is still
+	 * theoretically reachable, e.g. via unaccelerated fault-like AVIC
+	 * access, and needs to be handled by KVM to avoid putting the guest
+	 * into an infinite loop.   Injecting #UD is somewhat arbitrary, but
+	 * its the least awful option given lack of insight into the guest.
+	 *
+	 * If KVM is trying to skip an instruction, simply resume the guest.
+	 * If a #NPF occurs while the guest is vectoring an INT3/INTO, then KVM
+	 * will attempt to re-inject the INT3/INTO and skip the instruction.
+	 * In that scenario, retrying the INT3/INTO and hoping the guest will
+	 * make forward progress is the only option that has a chance of
+	 * success (and in practice it will work the vast majority of the time).
 	 */
 	if (unlikely(!insn)) {
-		kvm_queue_exception(vcpu, UD_VECTOR);
+		if (!(emul_type & EMULTYPE_SKIP))
+			kvm_queue_exception(vcpu, UD_VECTOR);
 		return false;
 	}
 
author maintainer
#200 lore raw

PATCH 172/219 KVM: SVM: Get source vCPUs from source VM for SEV-ES intrahost migration

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sean Christopherson <[email protected]>

commit f1187ef24eb8f36e8ad8106d22615ceddeea6097 upstream.

Fix a goof where KVM tries to grab source vCPUs from the destination VM when doing intrahost migration. Grabbing the wrong vCPU not only hoses the guest, it also crashes the host due to the VMSA pointer being left NULL.

  BUG: unable to handle page fault for address: ffffe38687000000
  #PF: supervisor read access in kernel mode
  #PF: error_code(0x0000) - not-present page
  PGD 0 P4D 0
  Oops: 0000 [#1] SMP NOPTI
  CPU: 39 PID: 17143 Comm: sev_migrate_tes Tainted: GO       6.5.0-smp--fff2e47e6c3b-next #151
  Hardware name: Google, Inc. Arcadia_IT_80/Arcadia_IT_80, BIOS 34.28.0 07/10/2023
  RIP: 0010:__free_pages+0x15/0xd0
  RSP: 0018:ffff923fcf6e3c78 EFLAGS: 00010246
  RAX: 0000000000000000 RBX: ffffe38687000000 RCX: 0000000000000100
  RDX: 0000000000000100 RSI: 0000000000000000 RDI: ffffe38687000000
  RBP: ffff923fcf6e3c88 R08: ffff923fcafb0000 R09: 0000000000000000
  R10: 0000000000000000 R11: ffffffff83619b90 R12: ffff923fa9540000
  R13: 0000000000080007 R14: ffff923f6d35d000 R15: 0000000000000000
  FS:  0000000000000000(0000) GS:ffff929d0d7c0000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: ffffe38687000000 CR3: 0000005224c34005 CR4: 0000000000770ee0
  PKRU: 55555554
  Call Trace:
   <TASK>
   sev_free_vcpu+0xcb/0x110 [kvm_amd]
   svm_vcpu_free+0x75/0xf0 [kvm_amd]
   kvm_arch_vcpu_destroy+0x36/0x140 [kvm]
   kvm_destroy_vcpus+0x67/0x100 [kvm]
   kvm_arch_destroy_vm+0x161/0x1d0 [kvm]
   kvm_put_kvm+0x276/0x560 [kvm]
   kvm_vm_release+0x25/0x30 [kvm]
   __fput+0x106/0x280
   ____fput+0x12/0x20
   task_work_run+0x86/0xb0
   do_exit+0x2e3/0x9c0
   do_group_exit+0xb1/0xc0
   __x64_sys_exit_group+0x1b/0x20
   do_syscall_64+0x41/0x90
   entry_SYSCALL_64_after_hwframe+0x63/0xcd
   </TASK>
  CR2: ffffe38687000000
---
 arch/x86/kvm/svm/sev.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/kvm/svm/sev.c +1 -1
@@ -1723,7 +1723,7 @@ static void sev_migrate_from(struct kvm
 		 * Note, the source is not required to have the same number of
 		 * vCPUs as the destination when migrating a vanilla SEV VM.
 		 */
-		src_vcpu = kvm_get_vcpu(dst_kvm, i);
+		src_vcpu = kvm_get_vcpu(src_kvm, i);
 		src_svm = to_svm(src_vcpu);
 
 		/*
author maintainer
#201 lore raw

PATCH 173/219 KVM: nSVM: Check instead of asserting on nested TSC scaling support

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sean Christopherson <[email protected]>

commit 7cafe9b8e22bb3d77f130c461aedf6868c4aaf58 upstream.

Check for nested TSC scaling support on nested SVM VMRUN instead of asserting that TSC scaling is exposed to L1 if L1's MSR_AMD64_TSC_RATIO has diverged from KVM's default. Userspace can trigger the WARN at will by writing the MSR and then updating guest CPUID to hide the feature (modifying guest CPUID is allowed anytime before KVM_RUN). E.g. hacking KVM's state_test selftest to do

		vcpu_set_msr(vcpu, MSR_AMD64_TSC_RATIO, 0);
		vcpu_clear_cpuid_feature(vcpu, X86_FEATURE_TSCRATEMSR);

after restoring state in a new VM+vCPU yields an endless supply of:

  ------------[ cut here ]------------
  WARNING: CPU: 164 PID: 62565 at arch/x86/kvm/svm/nested.c:699
           nested_vmcb02_prepare_control+0x3d6/0x3f0 [kvm_amd]
  Call Trace:
   <TASK>
   enter_svm_guest_mode+0x114/0x560 [kvm_amd]
   nested_svm_vmrun+0x260/0x330 [kvm_amd]
   vmrun_interception+0x29/0x30 [kvm_amd]
   svm_invoke_exit_handler+0x35/0x100 [kvm_amd]
   svm_handle_exit+0xe7/0x180 [kvm_amd]
   kvm_arch_vcpu_ioctl_run+0x1eab/0x2570 [kvm]
   kvm_vcpu_ioctl+0x4c9/0x5b0 [kvm]
   __se_sys_ioctl+0x7a/0xc0
   __x64_sys_ioctl+0x21/0x30
   do_syscall_64+0x41/0x90
   entry_SYSCALL_64_after_hwframe+0x63/0xcd
  RIP: 0033:0x45ca1b

Note, the nested #VMEXIT path has the same flaw, but needs a different fix and will be handled separately.

Fixes 5228eb96a487 ("KVM: x86: nSVM: implement nested TSC scaling")
Cc Maxim Levitsky <[email protected]>
---
 arch/x86/kvm/svm/nested.c |    5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
arch/x86/kvm/svm/nested.c +2 -3
@@ -660,10 +660,9 @@ static void nested_vmcb02_prepare_contro
 
 	vmcb02->control.tsc_offset = vcpu->arch.tsc_offset;
 
-	if (svm->tsc_ratio_msr != kvm_caps.default_tsc_scaling_ratio) {
-		WARN_ON(!svm->tsc_scaling_enabled);
+	if (svm->tsc_scaling_enabled &&
+	    svm->tsc_ratio_msr != kvm_caps.default_tsc_scaling_ratio)
 		nested_svm_update_tsc_ratio_msr(vcpu);
-	}
 
 	vmcb02->control.int_ctl             =
 		(svm->nested.ctl.int_ctl & int_ctl_vmcb12_bits) |
author maintainer
#202 lore raw

PATCH 174/219 KVM: nSVM: Load L1s TSC multiplier based on L1 state, not L2 state

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sean Christopherson <[email protected]>

commit 0c94e2468491cbf0754f49a5136ab51294a96b69 upstream.

When emulating nested VM-Exit, load L1's TSC multiplier if L1's desired ratio doesn't match the current ratio, not if the ratio L1 is using for L2 diverges from the default. Functionally, the end result is the same as KVM will run L2 with L1's multiplier if L2's multiplier is the default, i.e. checking that L1's multiplier is loaded is equivalent to checking if L2 has a non-default multiplier.

However, the assertion that TSC scaling is exposed to L1 is flawed, as userspace can trigger the WARN at will by writing the MSR and then updating guest CPUID to hide the feature (modifying guest CPUID is allowed anytime before KVM_RUN). E.g. hacking KVM's state_test selftest to do

                vcpu_set_msr(vcpu, MSR_AMD64_TSC_RATIO, 0);
                vcpu_clear_cpuid_feature(vcpu, X86_FEATURE_TSCRATEMSR);

after restoring state in a new VM+vCPU yields an endless supply of:

  ------------[ cut here ]------------
  WARNING: CPU: 10 PID: 206939 at arch/x86/kvm/svm/nested.c:1105
           nested_svm_vmexit+0x6af/0x720 [kvm_amd]
  Call Trace:
   nested_svm_exit_handled+0x102/0x1f0 [kvm_amd]
   svm_handle_exit+0xb9/0x180 [kvm_amd]
   kvm_arch_vcpu_ioctl_run+0x1eab/0x2570 [kvm]
   kvm_vcpu_ioctl+0x4c9/0x5b0 [kvm]
   ? trace_hardirqs_off+0x4d/0xa0
   __se_sys_ioctl+0x7a/0xc0
   __x64_sys_ioctl+0x21/0x30
   do_syscall_64+0x41/0x90
   entry_SYSCALL_64_after_hwframe+0x63/0xcd

Unlike the nested VMRUN path, hoisting the svm->tsc_scaling_enabled check into the if-statement is wrong as KVM needs to ensure L1's multiplier is
loaded in the above scenario. Alternatively, the WARN_ON() could simply
be deleted, but that would make KVM's behavior even more subtle, e.g. it's not immediately obvious why it's safe to write MSR_AMD64_TSC_RATIO when checking only tsc_ratio_msr.

Fixes 5228eb96a487 ("KVM: x86: nSVM: implement nested TSC scaling")
Cc Maxim Levitsky <[email protected]>
---
 arch/x86/kvm/svm/nested.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/x86/kvm/svm/nested.c +2 -2
@@ -1021,8 +1021,8 @@ int nested_svm_vmexit(struct vcpu_svm *s
 		vmcb_mark_dirty(vmcb01, VMCB_INTERCEPTS);
 	}
 
-	if (svm->tsc_ratio_msr != kvm_caps.default_tsc_scaling_ratio) {
-		WARN_ON(!svm->tsc_scaling_enabled);
+	if (kvm_caps.has_tsc_control &&
+	    vcpu->arch.tsc_scaling_ratio != vcpu->arch.l1_tsc_scaling_ratio) {
 		vcpu->arch.tsc_scaling_ratio = vcpu->arch.l1_tsc_scaling_ratio;
 		__svm_write_tsc_multiplier(vcpu->arch.tsc_scaling_ratio);
 	}
author maintainer
#203 lore raw

PATCH 175/219 KVM: SVM: Set target pCPU during IRTE update if target vCPU is running

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sean Christopherson <[email protected]>

commit f3cebc75e7425d6949d726bb8e937095b0aef025 upstream.

Update the target pCPU for IOMMU doorbells when updating IRTE routing if KVM is actively running the associated vCPU. KVM currently only updates the pCPU when loading the vCPU (via avic_vcpu_load()), and so doorbell events will be delayed until the vCPU goes through a put+load cycle (which might very well "never" happen for the lifetime of the VM).

To avoid inserting a stale pCPU, e.g. due to racing between updating IRTE routing and vCPU load/put, get the pCPU information from the vCPU's Physical APIC ID table entry (a.k.a. avic_physical_id_cache in KVM) and update the IRTE while holding ir_list_lock. Add comments with --verbose enabled to explain exactly what is and isn't protected by ir_list_lock.

Fixes 411b44ba80ab ("svm: Implements update_pi_irte hook to setup posted interrupt")
Cc Alejandro Jimenez <[email protected]>
Cc Joao Martins <[email protected]>
Cc Maxim Levitsky <[email protected]>
Cc Suravee Suthikulpanit <[email protected]>
---
 arch/x86/kvm/svm/avic.c |   28 ++++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)
arch/x86/kvm/svm/avic.c +28 -0
@@ -810,6 +810,7 @@ static int svm_ir_list_add(struct vcpu_s
 	int ret = 0;
 	unsigned long flags;
 	struct amd_svm_iommu_ir *ir;
+	u64 entry;
 
 	/**
 	 * In some cases, the existing irte is updated and re-set,
@@ -843,6 +844,18 @@ static int svm_ir_list_add(struct vcpu_s
 	ir->data = pi->ir_data;
 
 	spin_lock_irqsave(&svm->ir_list_lock, flags);
+
+	/*
+	 * Update the target pCPU for IOMMU doorbells if the vCPU is running.
+	 * If the vCPU is NOT running, i.e. is blocking or scheduled out, KVM
+	 * will update the pCPU info when the vCPU awkened and/or scheduled in.
+	 * See also avic_vcpu_load().
+	 */
+	entry = READ_ONCE(*(svm->avic_physical_id_cache));
+	if (entry & AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK)
+		amd_iommu_update_ga(entry & AVIC_PHYSICAL_ID_ENTRY_HOST_PHYSICAL_ID_MASK,
+				    true, pi->ir_data);
+
 	list_add(&ir->node, &svm->ir_list);
 	spin_unlock_irqrestore(&svm->ir_list_lock, flags);
 out:
@@ -1067,6 +1080,13 @@ void avic_vcpu_load(struct kvm_vcpu *vcp
 	if (kvm_vcpu_is_blocking(vcpu))
 		return;
 
+	/*
+	 * Grab the per-vCPU interrupt remapping lock even if the VM doesn't
+	 * _currently_ have assigned devices, as that can change.  Holding
+	 * ir_list_lock ensures that either svm_ir_list_add() will consume
+	 * up-to-date entry information, or that this task will wait until
+	 * svm_ir_list_add() completes to set the new target pCPU.
+	 */
 	spin_lock_irqsave(&svm->ir_list_lock, flags);
 
 	entry = READ_ONCE(*(svm->avic_physical_id_cache));
@@ -1102,6 +1122,14 @@ void avic_vcpu_put(struct kvm_vcpu *vcpu
 	if (!(entry & AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK))
 		return;
 
+	/*
+	 * Take and hold the per-vCPU interrupt remapping lock while updating
+	 * the Physical ID entry even though the lock doesn't protect against
+	 * multiple writers (see above).  Holding ir_list_lock ensures that
+	 * either svm_ir_list_add() will consume up-to-date entry information,
+	 * or that this task will wait until svm_ir_list_add() completes to
+	 * mark the vCPU as not running.
+	 */
 	spin_lock_irqsave(&svm->ir_list_lock, flags);
 
 	avic_update_iommu_vcpu_affinity(vcpu, -1, 0);
author maintainer
#204 lore raw

PATCH 176/219 KVM: SVM: Skip VMSA init in sev_es_init_vmcb() if pointer is NULL

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sean Christopherson <[email protected]>

commit 1952e74da96fb3e48b72a2d0ece78c688a5848c1 upstream.

Skip initializing the VMSA physical address in the VMCB if the VMSA is NULL, which occurs during intrahost migration as KVM initializes the VMCB before copying over state from the source to the destination (including the VMSA and its physical address).

In normal builds, __pa() is just math, so the bug isn't fatal, but with CONFIG_DEBUG_VIRTUAL=y, the validity of the virtual address is verified and passing in NULL will make the kernel unhappy.

---
 arch/x86/kvm/svm/sev.c |    7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
arch/x86/kvm/svm/sev.c +5 -2
@@ -2951,9 +2951,12 @@ static void sev_es_init_vmcb(struct vcpu
 	/*
 	 * An SEV-ES guest requires a VMSA area that is a separate from the
 	 * VMCB page. Do not include the encryption mask on the VMSA physical
-	 * address since hardware will access it using the guest key.
+	 * address since hardware will access it using the guest key.  Note,
+	 * the VMSA will be NULL if this vCPU is the destination for intrahost
+	 * migration, and will be copied later.
 	 */
-	svm->vmcb->control.vmsa_pa = __pa(svm->sev_es.vmsa);
+	if (svm->sev_es.vmsa)
+		svm->vmcb->control.vmsa_pa = __pa(svm->sev_es.vmsa);
 
 	/* Can't intercept CR register access, HV can't modify CR registers */
 	svm_clr_intercept(svm, INTERCEPT_CR0_READ);
author maintainer
#205 lore raw

PATCH 177/219 MIPS: Fix CONFIG_CPU_DADDI_WORKAROUNDS `modules_install regression

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Maciej W. Rozycki <[email protected]>

commit a79a404e6c2241ebc528b9ebf4c0832457b498c3 upstream.

Remove a build-time check for the presence of the GCC `-msym32' option. This option has been there since GCC 4.1.0, which is below the minimum required as at commit 805b2e1d427a ("kbuild: include Makefile.compiler only when compiler is needed"), when an error message:

arch/mips/Makefile:306: *** CONFIG_CPU_DADDI_WORKAROUNDS unsupported without -msym32. Stop.

started to trigger for the modules_install' target with configurations such as decstation_64_defconfig' that set CONFIG_CPU_DADDI_WORKAROUNDS, because said commit has made `cc-option-yn' an undefined function for non-build targets.

Fixes 805b2e1d427a ("kbuild: include Makefile.compiler only when compiler is needed")
Cc [email protected] # v5.13+
---
 arch/mips/Makefile |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/mips/Makefile +2 -2
@@ -308,8 +308,8 @@ ifdef CONFIG_64BIT
     endif
   endif
 
-  ifeq ($(KBUILD_SYM32)$(call cc-option-yn,-msym32), yy)
-    cflags-y += -msym32 -DKBUILD_64BIT_SYM32
+  ifeq ($(KBUILD_SYM32), y)
+    cflags-$(KBUILD_SYM32) += -msym32 -DKBUILD_64BIT_SYM32
   else
     ifeq ($(CONFIG_CPU_DADDI_WORKAROUNDS), y)
       $(error CONFIG_CPU_DADDI_WORKAROUNDS unsupported without -msym32)
author maintainer
#206 lore raw

PATCH 178/219 perf hists browser: Fix hierarchy mode header

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Namhyung Kim <[email protected]>

commit e2cabf2a44791f01c21f8d5189b946926e34142e upstream.

The commit ef9ff6017e3c4593 ("perf ui browser: Move the extra title lines from the hists browser") introduced ui_browser__gotorc_title() to help moving non-title lines easily. But it missed to update the title for the hierarchy mode so it won't print the header line on TUI at all.

  $ perf report --hierarchy
Fixes ef9ff6017e3c4593 ("perf ui browser: Move the extra title lines from the hists browser")
Cc Adrian Hunter <[email protected]>
Cc Ian Rogers <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Jiri Olsa <[email protected]>
Cc Peter Zijlstra <[email protected]>
---
 tools/perf/ui/browsers/hists.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/ui/browsers/hists.c +1 -1
@@ -1779,7 +1779,7 @@ static void hists_browser__hierarchy_hea
 	hists_browser__scnprintf_hierarchy_headers(browser, headers,
 						   sizeof(headers));
 
-	ui_browser__gotorc(&browser->b, 0, 0);
+	ui_browser__gotorc_title(&browser->b, 0, 0);
 	ui_browser__set_color(&browser->b, HE_COLORSET_ROOT);
 	ui_browser__write_nstring(&browser->b, headers, browser->b.width + 1);
 }
author maintainer
#207 lore raw

PATCH 179/219 perf test shell stat_bpf_counters: Fix test on Intel

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Namhyung Kim <[email protected]>

commit 68ca249c964f520af7f8763e22f12bd26b57b870 upstream.

As of now, bpf counters (bperf) don't support event groups. But the default perf stat includes topdown metrics if supported (on recent Intel machines) which require groups. That makes perf stat exiting.

  $ sudo perf stat --bpf-counter true
  bpf managed perf events do not yet support groups.

Actually the test explicitly uses cycles event only, but it missed to pass the option when it checks the availability of the command.

Fixes 2c0cb9f56020d2ea ("perf test: Add a shell test for 'perf stat --bpf-counters' new option")
Cc Adrian Hunter <[email protected]>
Cc Ian Rogers <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Jiri Olsa <[email protected]>
Cc Peter Zijlstra <[email protected]>
---
 tools/perf/tests/shell/stat_bpf_counters.sh |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
tools/perf/tests/shell/stat_bpf_counters.sh +2 -2
@@ -22,10 +22,10 @@ compare_number()
 }
 
 # skip if --bpf-counters is not supported
-if ! perf stat --bpf-counters true > /dev/null 2>&1; then
+if ! perf stat -e cycles --bpf-counters true > /dev/null 2>&1; then
 	if [ "$1" = "-v" ]; then
 		echo "Skipping: --bpf-counters not supported"
-		perf --no-pager stat --bpf-counters true || true
+		perf --no-pager stat -e cycles --bpf-counters true || true
 	fi
 	exit 2
 fi
author maintainer
#208 lore raw

PATCH 180/219 perf tools: Handle old data in PERF_RECORD_ATTR

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Namhyung Kim <[email protected]>

commit 9bf63282ea77a531ea58acb42fb3f40d2d1e4497 upstream.

The PERF_RECORD_ATTR is used for a pipe mode to describe an event with attribute and IDs. The ID table comes after the attr and it calculate size of the table using the total record size and the attr size.

n_ids = (total_record_size - end_of_the_attr_field) / sizeof(u64)

This is fine for most use cases, but sometimes it saves the pipe output in a file and then process it later. And it becomes a problem if there is a change in attr size between the record and report.

  $ perf record -o- > perf-pipe.data  # old version
  $ perf report -i- < perf-pipe.data  # new version

For example, if the attr size is 128 and it has 4 IDs, then it would save them in 168 byte like below:

8 byte: perf event header { .type = PERF_RECORD_ATTR, .size = 168 },
128 byte: perf event attr { .size = 128, ... },
32 byte: event IDs [] = { 1234, 1235, 1236, 1237 },

But when report later, it thinks the attr size is 136 then it only read the last 3 entries as ID.

8 byte: perf event header { .type = PERF_RECORD_ATTR, .size = 168 },
136 byte: perf event attr { .size = 136, ... },
24 byte: event IDs [] = { 1235, 1236, 1237 }, // 1234 is missing

So it should use the recorded version of the attr. The attr has the size field already then it should honor the size when reading data.

Fixes 2c46dbb517a10b18 ("perf: Convert perf header attrs into attr events")
Cc Adrian Hunter <[email protected]>
Cc Ian Rogers <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Jiri Olsa <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Tom Zanussi <[email protected]>
---
 tools/perf/util/header.c |   11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)
tools/perf/util/header.c +6 -5
@@ -4331,7 +4331,8 @@ int perf_event__process_attr(struct perf
 			     union perf_event *event,
 			     struct evlist **pevlist)
 {
-	u32 i, ids, n_ids;
+	u32 i, n_ids;
+	u64 *ids;
 	struct evsel *evsel;
 	struct evlist *evlist = *pevlist;
 
@@ -4347,9 +4348,8 @@ int perf_event__process_attr(struct perf
 
 	evlist__add(evlist, evsel);
 
-	ids = event->header.size;
-	ids -= (void *)&event->attr.id - (void *)event;
-	n_ids = ids / sizeof(u64);
+	n_ids = event->header.size - sizeof(event->header) - event->attr.attr.size;
+	n_ids = n_ids / sizeof(u64);
 	/*
 	 * We don't have the cpu and thread maps on the header, so
 	 * for allocating the perf_sample_id table we fake 1 cpu and
@@ -4358,8 +4358,9 @@ int perf_event__process_attr(struct perf
 	if (perf_evsel__alloc_id(&evsel->core, 1, n_ids))
 		return -ENOMEM;
 
+	ids = (void *)&event->attr.attr + event->attr.attr.size;
 	for (i = 0; i < n_ids; i++) {
-		perf_evlist__id_add(&evlist->core, &evsel->core, 0, i, event->attr.id[i]);
+		perf_evlist__id_add(&evlist->core, &evsel->core, 0, i, ids[i]);
 	}
 
 	return 0;
author maintainer
#209 lore raw

PATCH 181/219 perf hists browser: Fix the number of entries for e key

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Namhyung Kim <[email protected]>

commit f6b8436bede3e80226e8b2100279c4450c73806a upstream.

The 'e' key is to toggle expand/collapse the selected entry only. But the current code has a bug that it only increases the number of entries by 1 in the hierarchy mode so users cannot move under the current entry after the key stroke. This is due to a wrong assumption in the hist_entry__set_folding().

The commit b33f922651011eff ("perf hists browser: Put hist_entry folding logic into single function") factored out the code, but actually it should be handled separately. The hist_browser__set_folding() is to update fold state for each entry so it needs to traverse all (child) entries regardless of the current fold state. So it increases the number of entries by 1.

But the hist_entry__set_folding() only cares the currently selected entry and its all children. So it should count all unfolded child entries. This code is implemented in hist_browser__toggle_fold() already so we can just call it.

Fixes b33f922651011eff ("perf hists browser: Put hist_entry folding logic into single function")
Cc Adrian Hunter <[email protected]>
Cc Ian Rogers <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Jiri Olsa <[email protected]>
Cc Peter Zijlstra <[email protected]>
---
 tools/perf/ui/browsers/hists.c |   58 ++++++++++++++++-------------------------
 1 file changed, 24 insertions(+), 34 deletions(-)
tools/perf/ui/browsers/hists.c +24 -34
@@ -407,11 +407,6 @@ static bool hist_browser__selection_has_
 	return container_of(ms, struct callchain_list, ms)->has_children;
 }
 
-static bool hist_browser__he_selection_unfolded(struct hist_browser *browser)
-{
-	return browser->he_selection ? browser->he_selection->unfolded : false;
-}
-
 static bool hist_browser__selection_unfolded(struct hist_browser *browser)
 {
 	struct hist_entry *he = browser->he_selection;
@@ -584,8 +579,8 @@ static int hierarchy_set_folding(struct
 	return n;
 }
 
-static void __hist_entry__set_folding(struct hist_entry *he,
-				      struct hist_browser *hb, bool unfold)
+static void hist_entry__set_folding(struct hist_entry *he,
+				    struct hist_browser *hb, bool unfold)
 {
 	hist_entry__init_have_children(he);
 	he->unfolded = unfold ? he->has_children : false;
@@ -603,34 +598,12 @@ static void __hist_entry__set_folding(st
 		he->nr_rows = 0;
 }
 
-static void hist_entry__set_folding(struct hist_entry *he,
-				    struct hist_browser *browser, bool unfold)
-{
-	double percent;
-
-	percent = hist_entry__get_percent_limit(he);
-	if (he->filtered || percent < browser->min_pcnt)
-		return;
-
-	__hist_entry__set_folding(he, browser, unfold);
-
-	if (!he->depth || unfold)
-		browser->nr_hierarchy_entries++;
-	if (he->leaf)
-		browser->nr_callchain_rows += he->nr_rows;
-	else if (unfold && !hist_entry__has_hierarchy_children(he, browser->min_pcnt)) {
-		browser->nr_hierarchy_entries++;
-		he->has_no_entry = true;
-		he->nr_rows = 1;
-	} else
-		he->has_no_entry = false;
-}
-
 static void
 __hist_browser__set_folding(struct hist_browser *browser, bool unfold)
 {
 	struct rb_node *nd;
 	struct hist_entry *he;
+	double percent;
 
 	nd = rb_first_cached(&browser->hists->entries);
 	while (nd) {
@@ -640,6 +613,21 @@ __hist_browser__set_folding(struct hist_
 		nd = __rb_hierarchy_next(nd, HMD_FORCE_CHILD);
 
 		hist_entry__set_folding(he, browser, unfold);
+
+		percent = hist_entry__get_percent_limit(he);
+		if (he->filtered || percent < browser->min_pcnt)
+			continue;
+
+		if (!he->depth || unfold)
+			browser->nr_hierarchy_entries++;
+		if (he->leaf)
+			browser->nr_callchain_rows += he->nr_rows;
+		else if (unfold && !hist_entry__has_hierarchy_children(he, browser->min_pcnt)) {
+			browser->nr_hierarchy_entries++;
+			he->has_no_entry = true;
+			he->nr_rows = 1;
+		} else
+			he->has_no_entry = false;
 	}
 }
 
@@ -659,8 +647,10 @@ static void hist_browser__set_folding_se
 	if (!browser->he_selection)
 		return;
 
-	hist_entry__set_folding(browser->he_selection, browser, unfold);
-	browser->b.nr_entries = hist_browser__nr_entries(browser);
+	if (unfold == browser->he_selection->unfolded)
+		return;
+
+	hist_browser__toggle_fold(browser);
 }
 
 static void ui_browser__warn_lost_events(struct ui_browser *browser)
@@ -732,8 +722,8 @@ static int hist_browser__handle_hotkey(s
 		hist_browser__set_folding(browser, true);
 		break;
 	case 'e':
-		/* Expand the selected entry. */
-		hist_browser__set_folding_selected(browser, !hist_browser__he_selection_unfolded(browser));
+		/* Toggle expand/collapse the selected entry. */
+		hist_browser__toggle_fold(browser);
 		break;
 	case 'H':
 		browser->show_headers = !browser->show_headers;
author maintainer
#210 lore raw

PATCH 182/219 drm/amd/display: always switch off ODM before committing more streams

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wenjing Liu <[email protected]>

commit 49a30c3d1a2258fc93cfe6eea8e4951dabadc824 upstream.

ODM power optimization is only supported with single stream. When ODM power optimization is enabled, we might not have enough free pipes for enabling other stream. So when we are committing more than 1 stream we should first switch off ODM power optimization to make room for new stream and then allocating pipe resource for the new stream.

Fixes 59de751e3845 ("drm/amd/display: add ODM case when looking for first split pipe")
---
 drivers/gpu/drm/amd/display/dc/core/dc.c |   10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
drivers/gpu/drm/amd/display/dc/core/dc.c +5 -5
@@ -1977,12 +1977,12 @@ enum dc_status dc_commit_streams(struct
 		}
 	}
 
-	/* Check for case where we are going from odm 2:1 to max
-	 *  pipe scenario.  For these cases, we will call
-	 *  commit_minimal_transition_state() to exit out of odm 2:1
-	 *  first before processing new streams
+	/* ODM Combine 2:1 power optimization is only applied for single stream
+	 * scenario, it uses extra pipes than needed to reduce power consumption
+	 * We need to switch off this feature to make room for new streams.
 	 */
-	if (stream_count == dc->res_pool->pipe_count) {
+	if (stream_count > dc->current_state->stream_count &&
+			dc->current_state->stream_count == 1) {
 		for (i = 0; i < dc->res_pool->pipe_count; i++) {
 			pipe = &dc->current_state->res_ctx.pipe_ctx[i];
 			if (pipe->next_odm_pipe)
author maintainer
#211 lore raw

PATCH 183/219 drm/amd/display: Remove wait while locked

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Gabe Teeger <[email protected]>

commit 5a3ccb1400339268c5e3dc1fa044a7f6c7f59a02 upstream.

[Why]
We wait for mpc idle while in a locked state, leading to potential deadlock.

[What]
Move the wait_for_idle call to outside of HW lock. This and a call to wait_drr_doublebuffer_pending_clear are moved added to a new static helper function called wait_for_outstanding_hw_updates, to make the interface clearer.

Fixes 8f0d304d21b3 ("drm/amd/display: Do not commit pipe when updating DRR")
---
 drivers/gpu/drm/amd/display/dc/Makefile            |    1 
 drivers/gpu/drm/amd/display/dc/core/dc.c           |   58 ++++++++++++++-------
 drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c |   11 ---
 3 files changed, 42 insertions(+), 28 deletions(-)
drivers/gpu/drm/amd/display/dc/Makefile +1 -0
@@ -82,3 +82,4 @@ DC_EDID += dc_edid_parser.o
 AMD_DISPLAY_DMUB = $(addprefix $(AMDDALPATH)/dc/,$(DC_DMUB))
 AMD_DISPLAY_EDID = $(addprefix $(AMDDALPATH)/dc/,$(DC_EDID))
 AMD_DISPLAY_FILES += $(AMD_DISPLAY_DMUB) $(AMD_DISPLAY_EDID)
+
drivers/gpu/drm/amd/display/dc/core/dc.c +41 -17

71 lines not on this page · raw message

drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +0 -11
@@ -1515,17 +1515,6 @@ static void dcn20_update_dchubp_dpp(
 			|| plane_state->update_flags.bits.global_alpha_change
 			|| plane_state->update_flags.bits.per_pixel_alpha_change) {
 		// MPCC inst is equal to pipe index in practice
-		int mpcc_inst = hubp->inst;
-		int opp_inst;
-		int opp_count = dc->res_pool->pipe_count;
-
-		for (opp_inst = 0; opp_inst < opp_count; opp_inst++) {
-			if (dc->res_pool->opps[opp_inst]->mpcc_disconnect_pending[mpcc_inst]) {
-				dc->res_pool->mpc->funcs->wait_for_idle(dc->res_pool->mpc, mpcc_inst);
-				dc->res_pool->opps[opp_inst]->mpcc_disconnect_pending[mpcc_inst] = false;
-				break;
-			}
-		}
 		hws->funcs.update_mpcc(dc, pipe_ctx);
 	}
 
author maintainer
#212 lore raw

PATCH 184/219 drm/amdgpu: register a dirty framebuffer callback for fbcon

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hamza Mahfooz <[email protected]>

commit 0a611560f53bfd489e33f4a718c915f1a6123d03 upstream.

fbcon requires that we implement &drm_framebuffer_funcs.dirty. Otherwise, the framebuffer might take a while to flush (which would manifest as noticeable lag). However, we can't enable this callback for non-fbcon cases since it may cause too many atomic commits to be made at once. So, implement amdgpu_dirtyfb() and only enable it for fbcon framebuffers (we can use the "struct drm_file file" parameter in the callback to check for this since it is only NULL when called by fbcon, at least in the mainline kernel) on devices that support atomic KMS.

---
 drivers/gpu/drm/amd/amdgpu/amdgpu_display.c |   26 +++++++++++++++++++++++++-
 1 file changed, 25 insertions(+), 1 deletion(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +25 -1
@@ -38,6 +38,8 @@
 #include <linux/pci.h>
 #include <linux/pm_runtime.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_damage_helper.h>
+#include <drm/drm_drv.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_fb_helper.h>
@@ -493,11 +495,29 @@ bool amdgpu_display_ddc_probe(struct amd
 	return true;
 }
 
+static int amdgpu_dirtyfb(struct drm_framebuffer *fb, struct drm_file *file,
+			  unsigned int flags, unsigned int color,
+			  struct drm_clip_rect *clips, unsigned int num_clips)
+{
+
+	if (file)
+		return -ENOSYS;
+
+	return drm_atomic_helper_dirtyfb(fb, file, flags, color, clips,
+					 num_clips);
+}
+
 static const struct drm_framebuffer_funcs amdgpu_fb_funcs = {
 	.destroy = drm_gem_fb_destroy,
 	.create_handle = drm_gem_fb_create_handle,
 };
 
+static const struct drm_framebuffer_funcs amdgpu_fb_funcs_atomic = {
+	.destroy = drm_gem_fb_destroy,
+	.create_handle = drm_gem_fb_create_handle,
+	.dirty = amdgpu_dirtyfb
+};
+
 uint32_t amdgpu_display_supported_domains(struct amdgpu_device *adev,
 					  uint64_t bo_flags)
 {
@@ -1100,7 +1120,11 @@ static int amdgpu_display_gem_fb_verify_
 	if (ret)
 		goto err;
 
-	ret = drm_framebuffer_init(dev, &rfb->base, &amdgpu_fb_funcs);
+	if (drm_drv_uses_atomic_modeset(dev))
+		ret = drm_framebuffer_init(dev, &rfb->base,
+					   &amdgpu_fb_funcs_atomic);
+	else
+		ret = drm_framebuffer_init(dev, &rfb->base, &amdgpu_fb_funcs);
 
 	if (ret)
 		goto err;
author maintainer
#213 lore raw

PATCH 185/219 kunit: Fix wild-memory-access bug in kunit_free_suite_set()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jinjie Ruan <[email protected]>

[ Upstream commit 2810c1e99867a811e631dd24e63e6c1e3b78a59d ]

Inject fault while probing kunit-example-test.ko, if kstrdup() fails in mod_sysfs_setup() in load_module(), the mod->state will switch from MODULE_STATE_COMING to MODULE_STATE_GOING instead of from MODULE_STATE_LIVE to MODULE_STATE_GOING, so only kunit_module_exit() will be called without kunit_module_init(), and the mod->kunit_suites is no set correctly and the free in kunit_free_suite_set() will cause below wild-memory-access bug.

The mod->state state machine when load_module() succeeds:

MODULE_STATE_UNFORMED ---> MODULE_STATE_COMING ---> MODULE_STATE_LIVE

	 ^						|
	 |						| delete_module
	 +---------------- MODULE_STATE_GOING <---------+

The mod->state state machine when load_module() fails at mod_sysfs_setup():

MODULE_STATE_UNFORMED ---> MODULE_STATE_COMING ---> MODULE_STATE_GOING

	^						|
	|						|
	+-----------------------------------------------+

Call kunit_module_init() at MODULE_STATE_COMING state to fix the issue because MODULE_STATE_LIVE is transformed from it.

 Unable to handle kernel paging request at virtual address ffffff341e942a88
 KASAN: maybe wild-memory-access in range [0x0003f9a0f4a15440-0x0003f9a0f4a15447]
 Mem abort info:
   ESR = 0x0000000096000004
   EC = 0x25: DABT (current EL), IL = 32 bits
   SET = 0, FnV = 0
   EA = 0, S1PTW = 0
   FSC = 0x04: level 0 translation fault
 Data abort info:
   ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000
   CM = 0, WnR = 0, TnD = 0, TagAccess = 0
   GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
 swapper pgtable: 4k pages, 48-bit VAs, pgdp=00000000441ea000
 [ffffff341e942a88] pgd=0000000000000000, p4d=0000000000000000
 Internal error: Oops: 0000000096000004 [#1] PREEMPT SMP
 Modules linked in: kunit_example_test(-) cfg80211 rfkill 8021q garp mrp stp llc ipv6 [last unloaded: kunit_example_test]
 CPU: 3 PID: 2035 Comm: modprobe Tainted: G        W        N 6.5.0-next-20230828+ #136
 Hardware name: linux,dummy-virt (DT)
 pstate: a0000005 (NzCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
 pc : kfree+0x2c/0x70
 lr : kunit_free_suite_set+0xcc/0x13c
 sp : ffff8000829b75b0
 x29: ffff8000829b75b0 x28: ffff8000829b7b90 x27: 0000000000000000
 x26: dfff800000000000 x25: ffffcd07c82a7280 x24: ffffcd07a50ab300
 x23: ffffcd07a50ab2e8 x22: 1ffff00010536ec0 x21: dfff800000000000
 x20: ffffcd07a50ab2f0 x19: ffffcd07a50ab2f0 x18: 0000000000000000
 x17: 0000000000000000 x16: 0000000000000000 x15: ffffcd07c24b6764
 x14: ffffcd07c24b63c0 x13: ffffcd07c4cebb94 x12: ffff700010536ec7
 x11: 1ffff00010536ec6 x10: ffff700010536ec6 x9 : dfff800000000000
 x8 : 00008fffefac913a x7 : 0000000041b58ab3 x6 : 0000000000000000
 x5 : 1ffff00010536ec5 x4 : ffff8000829b7628 x3 : dfff800000000000
 x2 : ffffff341e942a80 x1 : ffffcd07a50aa000 x0 : fffffc0000000000
 Call trace:
  kfree+0x2c/0x70
  kunit_free_suite_set+0xcc/0x13c
  kunit_module_notify+0xd8/0x360
  blocking_notifier_call_chain+0xc4/0x128
  load_module+0x382c/0x44a4
  init_module_from_file+0xd4/0x128
  idempotent_init_module+0x2c8/0x524
  __arm64_sys_finit_module+0xac/0x100
  invoke_syscall+0x6c/0x258
  el0_svc_common.constprop.0+0x160/0x22c
  do_el0_svc+0x44/0x5c
  el0_svc+0x38/0x78
  el0t_64_sync_handler+0x13c/0x158
  el0t_64_sync+0x190/0x194
 Code: aa0003e1 b25657e0 d34cfc42 8b021802 (f9400440)
 ---[ end trace 0000000000000000 ]---
 Kernel panic - not syncing: Oops: Fatal exception
 SMP: stopping secondary CPUs
 Kernel Offset: 0x4d0742200000 from 0xffff800080000000
 PHYS_OFFSET: 0xffffee43c0000000
 CPU features: 0x88000203,3c020000,1000421b
 Memory Limit: none
 Rebooting in 1 seconds..
Fixes 3d6e44623841 ("kunit: unify module and builtin suite definitions")
---
 lib/kunit/test.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
lib/kunit/test.c +2 -1
index 184df6f701b48..a90bd265d73db 100644
@@ -667,12 +667,13 @@ static int kunit_module_notify(struct notifier_block *nb, unsigned long val,
 
 	switch (val) {
 	case MODULE_STATE_LIVE:
-		kunit_module_init(mod);
 		break;
 	case MODULE_STATE_GOING:
 		kunit_module_exit(mod);
 		break;
 	case MODULE_STATE_COMING:
+		kunit_module_init(mod);
+		break;
 	case MODULE_STATE_UNFORMED:
 		break;
 	}
Signature · 1 line
2.40.1
author maintainer
#214 lore raw

PATCH 186/219 net: ipv4: fix one memleak in __inet_del_ifa()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Liu Jian <[email protected]>

[ Upstream commit ac28b1ec6135649b5d78b028e47264cb3ebca5ea ]

I got the below warning when do fuzzing test:
unregister_netdevice: waiting for bond0 to become free. Usage count = 2

It can be repoduced via:

ip link add bond0 type bond
sysctl -w net.ipv4.conf.bond0.promote_secondaries=1 ip addr add 4.117.174.103/0 scope 0x40 dev bond0
ip addr add 192.168.100.111/255.255.255.254 scope 0 dev bond0 ip addr add 0.0.0.4/0 scope 0x40 secondary dev bond0 ip addr del 4.117.174.103/0 scope 0x40 dev bond0
ip link delete bond0 type bond

In this reproduction test case, an incorrect 'last_prim' is found in __inet_del_ifa(), as a result, the secondary address(0.0.0.4/0 scope 0x40) is lost. The memory of the secondary address is leaked and the reference of in_device and net_device is leaked.

Fix this problem:
Look for 'last_prim' starting at location of the deleted IP and inserting the promoted IP into the location of 'last_prim'.

Fixes 0ff60a45678e ("[IPV4]: Fix secondary IP addresses after promotion")
---
 net/ipv4/devinet.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
net/ipv4/devinet.c +5 -5
index e8b9a9202fecd..35d6e74be8406 100644
@@ -354,14 +354,14 @@ static void __inet_del_ifa(struct in_device *in_dev,
 {
 	struct in_ifaddr *promote = NULL;
 	struct in_ifaddr *ifa, *ifa1;
-	struct in_ifaddr *last_prim;
+	struct in_ifaddr __rcu **last_prim;
 	struct in_ifaddr *prev_prom = NULL;
 	int do_promote = IN_DEV_PROMOTE_SECONDARIES(in_dev);
 
 	ASSERT_RTNL();
 
 	ifa1 = rtnl_dereference(*ifap);
-	last_prim = rtnl_dereference(in_dev->ifa_list);
+	last_prim = ifap;
 	if (in_dev->dead)
 		goto no_promotions;
 
@@ -375,7 +375,7 @@ static void __inet_del_ifa(struct in_device *in_dev,
 		while ((ifa = rtnl_dereference(*ifap1)) != NULL) {
 			if (!(ifa->ifa_flags & IFA_F_SECONDARY) &&
 			    ifa1->ifa_scope <= ifa->ifa_scope)
-				last_prim = ifa;
+				last_prim = &ifa->ifa_next;
 
 			if (!(ifa->ifa_flags & IFA_F_SECONDARY) ||
 			    ifa1->ifa_mask != ifa->ifa_mask ||
@@ -439,9 +439,9 @@ static void __inet_del_ifa(struct in_device *in_dev,
 
 			rcu_assign_pointer(prev_prom->ifa_next, next_sec);
 
-			last_sec = rtnl_dereference(last_prim->ifa_next);
+			last_sec = rtnl_dereference(*last_prim);
 			rcu_assign_pointer(promote->ifa_next, last_sec);
-			rcu_assign_pointer(last_prim->ifa_next, promote);
+			rcu_assign_pointer(*last_prim, promote);
 		}
 
 		promote->ifa_flags &= ~IFA_F_SECONDARY;
Signature · 1 line
2.40.1
author maintainer
#215 lore raw

PATCH 187/219 kselftest/runner.sh: Propagate SIGTERM to runner child

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Björn Töpel <[email protected]>

[ Upstream commit 9616cb34b08ec86642b162eae75c5a7ca8debe3c ]

Timeouts in kselftest are done using the "timeout" command with the "--foreground" option. Without the "foreground" option, it is not possible for a user to cancel the runner using SIGINT, because the signal is not propagated to timeout which is running in a different process group. The "forground" options places the timeout in the same process group as its parent, but only sends the SIGTERM (on timeout) signal to the forked process. Unfortunately, this does not play nice with all kselftests, e.g. "net:fcnal-test.sh", where the child processes will linger because timeout does not send SIGTERM to the group.

Some users have noted these hangs [1].

Fix this by nesting the timeout with an additional timeout without the foreground option.

Fixes 651e0d881461 ("kselftest/runner: allow to properly deliver signals to tests")
---
 tools/testing/selftests/kselftest/runner.sh | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
tools/testing/selftests/kselftest/runner.sh +2 -1
index 294619ade49fe..1333ab1eda708 100644
@@ -35,7 +35,8 @@ tap_timeout()
 {
 	# Make sure tests will time out if utility is available.
 	if [ -x /usr/bin/timeout ] ; then
-		/usr/bin/timeout --foreground "$kselftest_timeout" $1
+		/usr/bin/timeout --foreground "$kselftest_timeout" \
+			/usr/bin/timeout "$kselftest_timeout" $1
 	else
 		$1
 	fi
Signature · 1 line
2.40.1
author maintainer
#216 lore raw

PATCH 188/219 selftests: Keep symlinks, when possible

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Björn Töpel <[email protected]>

[ Upstream commit 3f3f384139ed147c71e1d770accf610133d5309b ]

When kselftest is built/installed with the 'gen_tar' target, rsync is used for the installation step to copy files. Extra care is needed for tests that have symlinks. Commit ae108c48b5d2 ("selftests: net: Fix cross-tree inclusion of scripts") added '-L' (transform symlink into referent file/dir) to rsync, to fix dangling links. However, that broke some tests where the symlink (being a symlink) is part of the test (e.g. exec:execveat).

Use rsync's '--copy-unsafe-links' that does right thing.

Fixes ae108c48b5d2 ("selftests: net: Fix cross-tree inclusion of scripts")
---
 tools/testing/selftests/lib.mk | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
tools/testing/selftests/lib.mk +2 -2
index 05400462c7799..aa646e0661f36 100644
@@ -72,7 +72,7 @@ endef
 run_tests: all
 ifdef building_out_of_srctree
 	@if [ "X$(TEST_PROGS)$(TEST_PROGS_EXTENDED)$(TEST_FILES)" != "X" ]; then \
-		rsync -aLq $(TEST_PROGS) $(TEST_PROGS_EXTENDED) $(TEST_FILES) $(OUTPUT); \
+		rsync -aq --copy-unsafe-links $(TEST_PROGS) $(TEST_PROGS_EXTENDED) $(TEST_FILES) $(OUTPUT); \
 	fi
 	@if [ "X$(TEST_PROGS)" != "X" ]; then \
 		$(call RUN_TESTS, $(TEST_GEN_PROGS) $(TEST_CUSTOM_PROGS) \
@@ -86,7 +86,7 @@ endif
 
 define INSTALL_SINGLE_RULE
 	$(if $(INSTALL_LIST),@mkdir -p $(INSTALL_PATH))
-	$(if $(INSTALL_LIST),rsync -aL $(INSTALL_LIST) $(INSTALL_PATH)/)
+	$(if $(INSTALL_LIST),rsync -a --copy-unsafe-links $(INSTALL_LIST) $(INSTALL_PATH)/)
 endef
 
 define INSTALL_RULE
Signature · 1 line
2.40.1
author maintainer
#217 lore raw

PATCH 189/219 net/smc: use smc_lgr_list.lock to protect smc_lgr_list.list iterate in smcr_port_add

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Guangguan Wang <[email protected]>

[ Upstream commit f5146e3ef0a9eea405874b36178c19a4863b8989 ]

While doing smcr_port_add, there maybe linkgroup add into or delete from smc_lgr_list.list at the same time, which may result kernel crash. So, use smc_lgr_list.lock to protect smc_lgr_list.list iterate in smcr_port_add.

The crash calltrace show below:

BUG: kernel NULL pointer dereference, address: 0000000000000000
PGD 0 P4D 0
Oops: 0000 [#1] SMP NOPTI
CPU: 0 PID: 559726 Comm: kworker/0:92 Kdump: loaded Tainted: G
Hardware name: Alibaba Cloud Alibaba Cloud ECS, BIOS 449e491 04/01/2014
Workqueue: events smc_ib_port_event_work [smc]
RIP: 0010:smcr_port_add+0xa6/0xf0 [smc]
RSP: 0000:ffffa5a2c8f67de0 EFLAGS: 00010297
RAX: 0000000000000001 RBX: ffff9935e0650000 RCX: 0000000000000000
RDX: 0000000000000010 RSI: ffff9935e0654290 RDI: ffff9935c8560000
RBP: 0000000000000000 R08: 0000000000000000 R09: ffff9934c0401918
R10: 0000000000000000 R11: ffffffffb4a5c278 R12: ffff99364029aae4
R13: ffff99364029aa00 R14: 00000000ffffffed R15: ffff99364029ab08
FS:  0000000000000000(0000) GS:ffff994380600000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000000 CR3: 0000000f06a10003 CR4: 0000000002770ef0
PKRU: 55555554
Call Trace:
 smc_ib_port_event_work+0x18f/0x380 [smc]
 process_one_work+0x19b/0x340
 worker_thread+0x30/0x370
 ? process_one_work+0x340/0x340
 kthread+0x114/0x130
 ? __kthread_cancel_work+0x50/0x50
 ret_from_fork+0x1f/0x30
Fixes 1f90a05d9ff9 ("net/smc: add smcr_port_add() and smcr_link_up() processing")
---
 net/smc/smc_core.c | 2 ++
 1 file changed, 2 insertions(+)
net/smc/smc_core.c +2 -0
index c676d92af7b7d..64b6dd439938e 100644
@@ -1650,6 +1650,7 @@ void smcr_port_add(struct smc_ib_device *smcibdev, u8 ibport)
 {
 	struct smc_link_group *lgr, *n;
 
+	spin_lock_bh(&smc_lgr_list.lock);
 	list_for_each_entry_safe(lgr, n, &smc_lgr_list.list, list) {
 		struct smc_link *link;
 
@@ -1665,6 +1666,7 @@ void smcr_port_add(struct smc_ib_device *smcibdev, u8 ibport)
 		if (link)
 			smc_llc_add_link_local(link);
 	}
+	spin_unlock_bh(&smc_lgr_list.lock);
 }
 
 /* link is down - switch connections to alternate link,
Signature · 1 line
2.40.1
author maintainer
#218 lore raw

PATCH 190/219 net: stmmac: fix handling of zero coalescing tx-usecs

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vincent Whitchurch <[email protected]>

[ Upstream commit fa60b8163816f194786f3ee334c9a458da7699c6 ]

Setting ethtool -C eth0 tx-usecs 0 is supposed to disable the use of the coalescing timer but currently it gets programmed with zero delay instead.

Disable the use of the coalescing timer if tx-usecs is zero by preventing it from being restarted. Note that to keep things simple we don't start/stop the timer when the coalescing settings are changed, but just let that happen on the next transmit or timer expiry.

Fixes 8fce33317023 ("net: stmmac: Rework coalesce timer and fix multi-queue races")
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +6 -4
index a07bcb2f5d2e2..1559a4dafd413 100644
@@ -2692,9 +2692,7 @@ static int stmmac_tx_clean(struct stmmac_priv *priv, int budget, u32 queue)
 
 	/* We still have pending packets, let's call for a new scheduling */
 	if (tx_q->dirty_tx != tx_q->cur_tx)
-		hrtimer_start(&tx_q->txtimer,
-			      STMMAC_COAL_TIMER(priv->tx_coal_timer[queue]),
-			      HRTIMER_MODE_REL);
+		stmmac_tx_timer_arm(priv, queue);
 
 	__netif_tx_unlock_bh(netdev_get_tx_queue(priv->dev, queue));
 
@@ -2975,9 +2973,13 @@ static int stmmac_init_dma_engine(struct stmmac_priv *priv)
 static void stmmac_tx_timer_arm(struct stmmac_priv *priv, u32 queue)
 {
 	struct stmmac_tx_queue *tx_q = &priv->dma_conf.tx_queue[queue];
+	u32 tx_coal_timer = priv->tx_coal_timer[queue];
+
+	if (!tx_coal_timer)
+		return;
 
 	hrtimer_start(&tx_q->txtimer,
-		      STMMAC_COAL_TIMER(priv->tx_coal_timer[queue]),
+		      STMMAC_COAL_TIMER(tx_coal_timer),
 		      HRTIMER_MODE_REL);
 }
 
Signature · 1 line
2.40.1
author maintainer
#219 lore raw

PATCH 191/219 net: ethernet: mvpp2_main: fix possible OOB write in mvpp2_ethtool_get_rxnfc()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hangyu Hua <[email protected]>

[ Upstream commit 51fe0a470543f345e3c62b6798929de3ddcedc1d ]

rules is allocated in ethtool_get_rxnfc and the size is determined by rule_cnt from user space. So rule_cnt needs to be check before using rules to avoid OOB writing or NULL pointer dereference.

Fixes 90b509b39ac9 ("net: mvpp2: cls: Add Classification offload support")
---
 drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c +5 -0
index b399bdb1ca362..f936640cca4e6 100644
@@ -5578,6 +5578,11 @@ static int mvpp2_ethtool_get_rxnfc(struct net_device *dev,
 		break;
 	case ETHTOOL_GRXCLSRLALL:
 		for (i = 0; i < MVPP2_N_RFS_ENTRIES_PER_FLOW; i++) {
+			if (loc == info->rule_cnt) {
+				ret = -EMSGSIZE;
+				break;
+			}
+
 			if (port->rfs_rules[i])
 				rules[loc++] = i;
 		}
Signature · 1 line
2.40.1
author maintainer
#220 lore raw

PATCH 192/219 net: ethernet: mtk_eth_soc: fix possible NULL pointer dereference in mtk_hwlro_get_fdir_all()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hangyu Hua <[email protected]>

[ Upstream commit e4c79810755f66c9a933ca810da2724133b1165a ]

rule_locs is allocated in ethtool_get_rxnfc and the size is determined by rule_cnt from user space. So rule_cnt needs to be check before using rule_locs to avoid NULL pointer dereference.

Fixes 7aab747e5563 ("net: ethernet: mediatek: add ethtool functions to configure RX flows of HW LRO")
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/net/ethernet/mediatek/mtk_eth_soc.c +3 -0
index 7e318133423a9..0ac5ae16308f6 100644
@@ -2698,6 +2698,9 @@ static int mtk_hwlro_get_fdir_all(struct net_device *dev,
 	int i;
 
 	for (i = 0; i < MTK_MAX_LRO_IP_CNT; i++) {
+		if (cnt == cmd->rule_cnt)
+			return -EMSGSIZE;
+
 		if (mac->hwlro_ip[i]) {
 			rule_locs[cnt] = i;
 			cnt++;
Signature · 1 line
2.40.1
author maintainer
#221 lore raw

PATCH 193/219 hsr: Fix uninit-value access in fill_frame_info()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ziyang Xuan <[email protected]>

[ Upstream commit 484b4833c604c0adcf19eac1ca14b60b757355b5 ]

Syzbot reports the following uninit-value access problem.

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

BUG: KMSAN: uninit-value in fill_frame_info net/hsr/hsr_forward.c:601 [inline]
BUG: KMSAN: uninit-value in hsr_forward_skb+0x9bd/0x30f0 net/hsr/hsr_forward.c:616
 fill_frame_info net/hsr/hsr_forward.c:601 [inline]
 hsr_forward_skb+0x9bd/0x30f0 net/hsr/hsr_forward.c:616
 hsr_dev_xmit+0x192/0x330 net/hsr/hsr_device.c:223
 __netdev_start_xmit include/linux/netdevice.h:4889 [inline]
 netdev_start_xmit include/linux/netdevice.h:4903 [inline]
 xmit_one net/core/dev.c:3544 [inline]
 dev_hard_start_xmit+0x247/0xa10 net/core/dev.c:3560
 __dev_queue_xmit+0x34d0/0x52a0 net/core/dev.c:4340
 dev_queue_xmit include/linux/netdevice.h:3082 [inline]
 packet_xmit+0x9c/0x6b0 net/packet/af_packet.c:276
 packet_snd net/packet/af_packet.c:3087 [inline]
 packet_sendmsg+0x8b1d/0x9f30 net/packet/af_packet.c:3119
 sock_sendmsg_nosec net/socket.c:730 [inline]
 sock_sendmsg net/socket.c:753 [inline]
 __sys_sendto+0x781/0xa30 net/socket.c:2176
 __do_sys_sendto net/socket.c:2188 [inline]
 __se_sys_sendto net/socket.c:2184 [inline]
 __ia32_sys_sendto+0x11f/0x1c0 net/socket.c:2184
 do_syscall_32_irqs_on arch/x86/entry/common.c:112 [inline]
 __do_fast_syscall_32+0xa2/0x100 arch/x86/entry/common.c:178
 do_fast_syscall_32+0x37/0x80 arch/x86/entry/common.c:203
 do_SYSENTER_32+0x1f/0x30 arch/x86/entry/common.c:246
 entry_SYSENTER_compat_after_hwframe+0x70/0x82

Uninit was created at:

 slab_post_alloc_hook+0x12f/0xb70 mm/slab.h:767
 slab_alloc_node mm/slub.c:3478 [inline]
 kmem_cache_alloc_node+0x577/0xa80 mm/slub.c:3523
 kmalloc_reserve+0x148/0x470 net/core/skbuff.c:559
 __alloc_skb+0x318/0x740 net/core/skbuff.c:644
 alloc_skb include/linux/skbuff.h:1286 [inline]
 alloc_skb_with_frags+0xc8/0xbd0 net/core/skbuff.c:6299
 sock_alloc_send_pskb+0xa80/0xbf0 net/core/sock.c:2794
 packet_alloc_skb net/packet/af_packet.c:2936 [inline]
 packet_snd net/packet/af_packet.c:3030 [inline]
 packet_sendmsg+0x70e8/0x9f30 net/packet/af_packet.c:3119
 sock_sendmsg_nosec net/socket.c:730 [inline]
 sock_sendmsg net/socket.c:753 [inline]
 __sys_sendto+0x781/0xa30 net/socket.c:2176
 __do_sys_sendto net/socket.c:2188 [inline]
 __se_sys_sendto net/socket.c:2184 [inline]
 __ia32_sys_sendto+0x11f/0x1c0 net/socket.c:2184
 do_syscall_32_irqs_on arch/x86/entry/common.c:112 [inline]
 __do_fast_syscall_32+0xa2/0x100 arch/x86/entry/common.c:178
 do_fast_syscall_32+0x37/0x80 arch/x86/entry/common.c:203
 do_SYSENTER_32+0x1f/0x30 arch/x86/entry/common.c:246
 entry_SYSENTER_compat_after_hwframe+0x70/0x82

It is because VLAN not yet supported in hsr driver. Return error when protocol is ETH_P_8021Q in fill_frame_info() now to fix it.

Fixes 451d8123f897 ("net: prp: add packet handling support")
Reported-by [email protected]
---
 net/hsr/hsr_forward.c | 1 +
 1 file changed, 1 insertion(+)
net/hsr/hsr_forward.c +1 -0
index 629daacc96071..b71dab630a873 100644
@@ -594,6 +594,7 @@ static int fill_frame_info(struct hsr_frame_info *frame,
 		proto = vlan_hdr->vlanhdr.h_vlan_encapsulated_proto;
 		/* FIXME: */
 		netdev_warn_once(skb->dev, "VLAN not yet supported");
+		return -EINVAL;
 	}
 
 	frame->is_from_san = false;
Signature · 1 line
2.40.1
author maintainer
#222 lore raw

PATCH 194/219 net: ethernet: adi: adin1110: use eth_broadcast_addr() to assign broadcast address

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yang Yingliang <[email protected]>

[ Upstream commit 54024dbec95585243391caeb9f04a2620e630765 ]

Use eth_broadcast_addr() to assign broadcast address instead of memset().

Stable-dep-of: 32530dba1bd4 ("net:ethernet:adi:adin1110: Fix forwarding offload")

---
 drivers/net/ethernet/adi/adin1110.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/net/ethernet/adi/adin1110.c +4 -4
index ecce5f7a549f2..cc026780ee0e8 100644
@@ -740,7 +740,7 @@ static int adin1110_broadcasts_filter(struct adin1110_port_priv *port_priv,
 	u32 port_rules = 0;
 	u8 mask[ETH_ALEN];
 
-	memset(mask, 0xFF, ETH_ALEN);
+	eth_broadcast_addr(mask);
 
 	if (accept_broadcast && port_priv->state == BR_STATE_FORWARDING)
 		port_rules = adin1110_port_rules(port_priv, true, true);
@@ -761,7 +761,7 @@ static int adin1110_set_mac_address(struct net_device *netdev,
 		return -EADDRNOTAVAIL;
 
 	eth_hw_addr_set(netdev, dev_addr);
-	memset(mask, 0xFF, ETH_ALEN);
+	eth_broadcast_addr(mask);
 
 	mac_slot = (!port_priv->nr) ?  ADIN_MAC_P1_ADDR_SLOT : ADIN_MAC_P2_ADDR_SLOT;
 	port_rules = adin1110_port_rules(port_priv, true, false);
@@ -1251,7 +1251,7 @@ static int adin1110_port_set_blocking_state(struct adin1110_port_priv *port_priv
 		goto out;
 
 	/* Allow only BPDUs to be passed to the CPU */
-	memset(mask, 0xFF, ETH_ALEN);
+	eth_broadcast_addr(mask);
 	port_rules = adin1110_port_rules(port_priv, true, false);
 	ret = adin1110_write_mac_address(port_priv, mac_slot, mac,
 					 mask, port_rules);
@@ -1366,7 +1366,7 @@ static int adin1110_fdb_add(struct adin1110_port_priv *port_priv,
 
 	other_port = priv->ports[!port_priv->nr];
 	port_rules = adin1110_port_rules(port_priv, false, true);
-	memset(mask, 0xFF, ETH_ALEN);
+	eth_broadcast_addr(mask);
 
 	return adin1110_write_mac_address(other_port, mac_nr, (u8 *)fdb->addr,
 					  mask, port_rules);
Signature · 1 line
2.40.1
author maintainer
#223 lore raw

PATCH 195/219 net:ethernet:adi:adin1110: Fix forwarding offload

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ciprian Regus <[email protected]>

[ Upstream commit 32530dba1bd48da4437d18d9a8dbc9d2826938a6 ]

Currently, when a new fdb entry is added (with both ports of the ADIN2111 bridged), the driver configures the MAC filters for the wrong port, which results in the forwarding being done by the host, and not actually hardware offloaded.

The ADIN2111 offloads the forwarding by setting filters on the destination MAC address of incoming frames. Based on these, they may be routed to the other port. Thus, if a frame has to be forwarded from port 1 to port 2, the required configuration for the ADDR_FILT_UPRn register should set the APPLY2PORT1 bit (instead of APPLY2PORT2, as it's currently the case).

Fixes bc93e19d088b ("net: ethernet: adi: Add ADIN1110 support")
---
 drivers/net/ethernet/adi/adin1110.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/adi/adin1110.c +1 -1
index cc026780ee0e8..ed2863ed6a5bb 100644
@@ -1365,7 +1365,7 @@ static int adin1110_fdb_add(struct adin1110_port_priv *port_priv,
 		return -ENOMEM;
 
 	other_port = priv->ports[!port_priv->nr];
-	port_rules = adin1110_port_rules(port_priv, false, true);
+	port_rules = adin1110_port_rules(other_port, false, true);
 	eth_broadcast_addr(mask);
 
 	return adin1110_write_mac_address(other_port, mac_nr, (u8 *)fdb->addr,
Signature · 1 line
2.40.1
author maintainer
#224 lore raw

PATCH 196/219 net: dsa: sja1105: hide all multicast addresses from "bridge fdb show"

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vladimir Oltean <[email protected]>

[ Upstream commit 02c652f5465011126152bbd93b6a582a1d0c32f1 ]

Commit 4d9423549501 ("net: dsa: sja1105: offload bridge port flags to device") has partially hidden some multicast entries from showing up in the "bridge fdb show" output, but it wasn't enough. Addresses which are added through "bridge mdb add" still show up. Hide them all.

Fixes 291d1e72b756 ("net: dsa: sja1105: Add support for FDB and MDB management")
---
 drivers/net/dsa/sja1105/sja1105_main.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)
drivers/net/dsa/sja1105/sja1105_main.c +6 -5
index ff94c5996fafb..f8228a81234fe 100644
@@ -1875,13 +1875,14 @@ static int sja1105_fdb_dump(struct dsa_switch *ds, int port,
 		if (!(l2_lookup.destports & BIT(port)))
 			continue;
 
-		/* We need to hide the FDB entry for unknown multicast */
-		if (l2_lookup.macaddr == SJA1105_UNKNOWN_MULTICAST &&
-		    l2_lookup.mask_macaddr == SJA1105_UNKNOWN_MULTICAST)
-			continue;
-
 		u64_to_ether_addr(l2_lookup.macaddr, macaddr);
 
+		/* Hardware FDB is shared for fdb and mdb, "bridge fdb show"
+		 * only wants to see unicast
+		 */
+		if (is_multicast_ether_addr(macaddr))
+			continue;
+
 		/* We need to hide the dsa_8021q VLANs from the user. */
 		if (vid_is_dsa_8021q(l2_lookup.vlanid))
 			l2_lookup.vlanid = 0;
Signature · 1 line
2.40.1
author maintainer
#225 lore raw

PATCH 197/219 net: dsa: sja1105: propagate exact error code from sja1105_dynamic_config_poll_valid()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vladimir Oltean <[email protected]>

[ Upstream commit c956798062b5a308db96e75157747291197f0378 ]

Currently, sja1105_dynamic_config_wait_complete() returns either 0 or -ETIMEDOUT, because it just looks at the read_poll_timeout() return code.

There will be future changes which move some more checks to sja1105_dynamic_config_poll_valid(). It is important that we propagate their exact return code (-ENOENT, -EINVAL), because callers of sja1105_dynamic_config_read() depend on them.

Stable-dep-of: 7cef293b9a63 ("net: dsa: sja1105: fix multicast forwarding working only for last added mdb entry")

---
 drivers/net/dsa/sja1105/sja1105_dynamic_config.c | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)
drivers/net/dsa/sja1105/sja1105_dynamic_config.c +8 -7
index 7729d3f8b7f50..93d47dab8d3e9 100644
@@ -1211,13 +1211,14 @@ sja1105_dynamic_config_wait_complete(struct sja1105_private *priv,
 				     struct sja1105_dyn_cmd *cmd,
 				     const struct sja1105_dynamic_table_ops *ops)
 {
-	int rc;
-
-	return read_poll_timeout(sja1105_dynamic_config_poll_valid,
-				 rc, rc != -EAGAIN,
-				 SJA1105_DYNAMIC_CONFIG_SLEEP_US,
-				 SJA1105_DYNAMIC_CONFIG_TIMEOUT_US,
-				 false, priv, cmd, ops);
+	int err, rc;
+
+	err = read_poll_timeout(sja1105_dynamic_config_poll_valid,
+				rc, rc != -EAGAIN,
+				SJA1105_DYNAMIC_CONFIG_SLEEP_US,
+				SJA1105_DYNAMIC_CONFIG_TIMEOUT_US,
+				false, priv, cmd, ops);
+	return err < 0 ? err : rc;
 }
 
 /* Provides read access to the settings through the dynamic interface
Signature · 1 line
2.40.1
author maintainer
#226 lore raw

PATCH 198/219 net: dsa: sja1105: fix multicast forwarding working only for last added mdb entry

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vladimir Oltean <[email protected]>

[ Upstream commit 7cef293b9a634a05fcce9e1df4aee3aeed023345 ]

The commit cited in Fixes: did 2 things: it refactored the read-back polling from sja1105_dynamic_config_read() into a new function, sja1105_dynamic_config_wait_complete(), and it called that from sja1105_dynamic_config_write() too.

What is problematic is the refactoring.

The refactored code from sja1105_dynamic_config_poll_valid() works like the previous one, but the problem is that it uses another packed_buf[] SPI buffer, and there was code at the end of sja1105_dynamic_config_read() which was relying on the read-back packed_buf[]:

	/* Don't dereference possibly NULL pointer - maybe caller
	 * only wanted to see whether the entry existed or not.
	 */
	if (entry)
		ops->entry_packing(packed_buf, entry, UNPACK);

After the change, the packed_buf[] that this code sees is no longer the entry read back from hardware, but the original entry that the caller passed to the sja1105_dynamic_config_read(), packed into this buffer.

This difference is the most notable with the SJA1105_SEARCH uses from sja1105pqrs_fdb_add() - used for both fdb and mdb. There, we have logic added by commit 728db843df88 ("net: dsa: sja1105: ignore the FDB entry for unknown multicast when adding a new address") to figure out whether the address we're trying to add matches on any existing hardware entry, with the exception of the catch-all multicast address.

That logic was broken, because with sja1105_dynamic_config_read() not working properly, it doesn't return us the entry read back from hardware, but the entry that we passed to it. And, since for multicast, a match will always exist, it will tell us that any mdb entry already exists at index=0 L2 Address Lookup table. It is index=0 because the caller doesn't know the index - it wants to find it out, and sja1105_dynamic_config_read() does:

	if (index < 0) { // SJA1105_SEARCH
		/* Avoid copying a signed negative number to an u64 */
		cmd.index = 0; // <- this
		cmd.search = true;
	} else {
		cmd.index = index;
		cmd.search = false;
	}

So, to the caller of sja1105_dynamic_config_read(), the returned info looks entirely legit, and it will add all mdb entries to FDB index 0. There, they will always overwrite each other (not to mention, potentially they can also overwrite a pre-existing bridge fdb entry), and the user-visible impact will be that only the last mdb entry will be forwarded as it should. The others won't (will be flooded or dropped, depending on the egress flood settings).

Fixing is a bit more complicated, and involves either passing the same packed_buf[] to sja1105_dynamic_config_wait_complete(), or moving all the extra processing on the packed_buf[] to sja1105_dynamic_config_wait_complete(). I've opted for the latter, because it makes sja1105_dynamic_config_wait_complete() a bit more self-contained.

Fixes df405910ab9f ("net: dsa: sja1105: wait for dynamic config command completion on writes too")
---
 .../net/dsa/sja1105/sja1105_dynamic_config.c  | 80 +++++++++----------
 1 file changed, 37 insertions(+), 43 deletions(-)
drivers/net/dsa/sja1105/sja1105_dynamic_config.c +37 -43
index 93d47dab8d3e9..984c0e604e8de 100644
@@ -1175,18 +1175,15 @@ const struct sja1105_dynamic_table_ops sja1110_dyn_ops[BLK_IDX_MAX_DYN] = {
 
 static int
 sja1105_dynamic_config_poll_valid(struct sja1105_private *priv,
-				  struct sja1105_dyn_cmd *cmd,
-				  const struct sja1105_dynamic_table_ops *ops)
+				  const struct sja1105_dynamic_table_ops *ops,
+				  void *entry, bool check_valident,
+				  bool check_errors)
 {
 	u8 packed_buf[SJA1105_MAX_DYN_CMD_SIZE] = {};
+	struct sja1105_dyn_cmd cmd = {};
 	int rc;
 
-	/* We don't _need_ to read the full entry, just the command area which
-	 * is a fixed SJA1105_SIZE_DYN_CMD. But our cmd_packing() API expects a
-	 * buffer that contains the full entry too. Additionally, our API
-	 * doesn't really know how many bytes into the buffer does the command
-	 * area really begin. So just read back the whole entry.
-	 */
+	/* Read back the whole entry + command structure. */
 	rc = sja1105_xfer_buf(priv, SPI_READ, ops->addr, packed_buf,
 			      ops->packed_size);
 	if (rc)
@@ -1195,11 +1192,25 @@ sja1105_dynamic_config_poll_valid(struct sja1105_private *priv,
 	/* Unpack the command structure, and return it to the caller in case it
 	 * needs to perform further checks on it (VALIDENT).
 	 */
-	memset(cmd, 0, sizeof(*cmd));
-	ops->cmd_packing(packed_buf, cmd, UNPACK);
+	ops->cmd_packing(packed_buf, &cmd, UNPACK);
 
 	/* Hardware hasn't cleared VALID => still working on it */
-	return cmd->valid ? -EAGAIN : 0;
+	if (cmd.valid)
+		return -EAGAIN;
+
+	if (check_valident && !cmd.valident && !(ops->access & OP_VALID_ANYWAY))
+		return -ENOENT;
+
+	if (check_errors && cmd.errors)
+		return -EINVAL;
+
+	/* Don't dereference possibly NULL pointer - maybe caller
+	 * only wanted to see whether the entry existed or not.
+	 */
+	if (entry)
+		ops->entry_packing(packed_buf, entry, UNPACK);
+
+	return 0;
 }
 
 /* Poll the dynamic config entry's control area until the hardware has
@@ -1208,8 +1219,9 @@ sja1105_dynamic_config_poll_valid(struct sja1105_private *priv,
  */
 static int
 sja1105_dynamic_config_wait_complete(struct sja1105_private *priv,
-				     struct sja1105_dyn_cmd *cmd,
-				     const struct sja1105_dynamic_table_ops *ops)
+				     const struct sja1105_dynamic_table_ops *ops,
+				     void *entry, bool check_valident,
+				     bool check_errors)
 {
 	int err, rc;
 
@@ -1217,7 +1229,8 @@ sja1105_dynamic_config_wait_complete(struct sja1105_private *priv,
 				rc, rc != -EAGAIN,
 				SJA1105_DYNAMIC_CONFIG_SLEEP_US,
 				SJA1105_DYNAMIC_CONFIG_TIMEOUT_US,
-				false, priv, cmd, ops);
+				false, priv, ops, entry, check_valident,
+				check_errors);
 	return err < 0 ? err : rc;
 }
 
@@ -1287,25 +1300,14 @@ int sja1105_dynamic_config_read(struct sja1105_private *priv,
 	mutex_lock(&priv->dynamic_config_lock);
 	rc = sja1105_xfer_buf(priv, SPI_WRITE, ops->addr, packed_buf,
 			      ops->packed_size);
-	if (rc < 0) {
-		mutex_unlock(&priv->dynamic_config_lock);
-		return rc;
-	}
-
-	rc = sja1105_dynamic_config_wait_complete(priv, &cmd, ops);
-	mutex_unlock(&priv->dynamic_config_lock);
 	if (rc < 0)
-		return rc;
+		goto out;
 
-	if (!cmd.valident && !(ops->access & OP_VALID_ANYWAY))
-		return -ENOENT;
+	rc = sja1105_dynamic_config_wait_complete(priv, ops, entry, true, false);
+out:
+	mutex_unlock(&priv->dynamic_config_lock);
 
-	/* Don't dereference possibly NULL pointer - maybe caller
-	 * only wanted to see whether the entry existed or not.
-	 */
-	if (entry)
-		ops->entry_packing(packed_buf, entry, UNPACK);
-	return 0;
+	return rc;
 }
 
 int sja1105_dynamic_config_write(struct sja1105_private *priv,
@@ -1357,22 +1359,14 @@ int sja1105_dynamic_config_write(struct sja1105_private *priv,
 	mutex_lock(&priv->dynamic_config_lock);
 	rc = sja1105_xfer_buf(priv, SPI_WRITE, ops->addr, packed_buf,
 			      ops->packed_size);
-	if (rc < 0) {
-		mutex_unlock(&priv->dynamic_config_lock);
-		return rc;
-	}
-
-	rc = sja1105_dynamic_config_wait_complete(priv, &cmd, ops);
-	mutex_unlock(&priv->dynamic_config_lock);
 	if (rc < 0)
-		return rc;
+		goto out;
 
-	cmd = (struct sja1105_dyn_cmd) {0};
-	ops->cmd_packing(packed_buf, &cmd, UNPACK);
-	if (cmd.errors)
-		return -EINVAL;
+	rc = sja1105_dynamic_config_wait_complete(priv, ops, NULL, false, true);
+out:
+	mutex_unlock(&priv->dynamic_config_lock);
 
-	return 0;
+	return rc;
 }
 
 static u8 sja1105_crc8_add(u8 crc, u8 byte, u8 poly)
Signature · 1 line
2.40.1
author maintainer
#227 lore raw

PATCH 199/219 net: dsa: sja1105: serialize sja1105_port_mcast_flood() with other FDB accesses

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vladimir Oltean <[email protected]>

[ Upstream commit ea32690daf4fa525dc5a4d164bd00ed8c756e1c6 ]

sja1105_fdb_add() runs from the dsa_owq, and sja1105_port_mcast_flood() runs from switchdev_deferred_process_work(). Prior to the blamed commit, they used to be indirectly serialized through the rtnl_lock(), which no longer holds true because dsa_owq dropped that.

So, it is now possible that we traverse the static config BLK_IDX_L2_LOOKUP elements concurrently compared to when we change them, in sja1105_static_fdb_change(). That is not ideal, since it might result in data corruption.

Introduce a mutex which serializes accesses to the hardware FDB and to the static config elements for the L2 Address Lookup table.

I can't find a good reason to add locking around sja1105_fdb_dump(). I'll add it later if needed.

Fixes 0faf890fc519 ("net: dsa: drop rtnl_lock from dsa_slave_switchdev_event_work")
---
 drivers/net/dsa/sja1105/sja1105.h      |  2 +
 drivers/net/dsa/sja1105/sja1105_main.c | 56 ++++++++++++++++++++------
 2 files changed, 45 insertions(+), 13 deletions(-)
drivers/net/dsa/sja1105/sja1105.h +2 -0
index a831bb0a52074..06e0ebfe652a4 100644
@@ -264,6 +264,8 @@ struct sja1105_private {
 	 * the switch doesn't confuse them with one another.
 	 */
 	struct mutex mgmt_lock;
+	/* Serializes accesses to the FDB */
+	struct mutex fdb_lock;
 	/* PTP two-step TX timestamp ID, and its serialization lock */
 	spinlock_t ts_id_lock;
 	u8 ts_id;
drivers/net/dsa/sja1105/sja1105_main.c +43 -13

124 lines not on this page · raw message

Signature · 1 line
2.40.1
author maintainer
#228 lore raw

PATCH 200/219 net: dsa: sja1105: block FDB accesses that are concurrent with a switch reset

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vladimir Oltean <[email protected]>

[ Upstream commit 86899e9e1e29e854b5f6dcc24ba4f75f792c89aa ]

Currently, when we add the first sja1105 port to a bridge with vlan_filtering 1, then we sometimes see this output:

sja1105 spi2.2: port 4 failed to read back entry for be:79:b4:9e:9e:96 vid 3088: -ENOENT sja1105 spi2.2: Reset switch and programmed static config. Reason: VLAN filtering sja1105 spi2.2: port 0 failed to add be:79:b4:9e:9e:96 vid 0 to fdb: -2

It is because sja1105_fdb_add() runs from the dsa_owq which is no longer serialized with switch resets since it dropped the rtnl_lock() in the blamed commit.

Either performing the FDB accesses before the reset, or after the reset, is equally fine, because sja1105_static_fdb_change() backs up those changes in the static config, but FDB access during reset isn't ok.

Make sja1105_static_config_reload() take the fdb_lock to fix that.

Fixes 0faf890fc519 ("net: dsa: drop rtnl_lock from dsa_slave_switchdev_event_work")
---
 drivers/net/dsa/sja1105/sja1105_main.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/net/dsa/sja1105/sja1105_main.c +2 -0
index c6c74e302e4dd..f1f1368e8146f 100644
@@ -2304,6 +2304,7 @@ int sja1105_static_config_reload(struct sja1105_private *priv,
 	int rc, i;
 	s64 now;
 
+	mutex_lock(&priv->fdb_lock);
 	mutex_lock(&priv->mgmt_lock);
 
 	mac = priv->static_config.tables[BLK_IDX_MAC_CONFIG].entries;
@@ -2418,6 +2419,7 @@ int sja1105_static_config_reload(struct sja1105_private *priv,
 		goto out;
 out:
 	mutex_unlock(&priv->mgmt_lock);
+	mutex_unlock(&priv->fdb_lock);
 
 	return rc;
 }
Signature · 1 line
2.40.1
author maintainer
#229 lore raw

PATCH 201/219 r8152: check budget for r8152_poll()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hayes Wang <[email protected]>

[ Upstream commit a7b8d60b37237680009dd0b025fe8c067aba0ee3 ]

According to the document of napi, there is no rx process when the budget is 0. Therefore, r8152_poll() has to return 0 directly when the budget is equal to 0.

Fixes d2187f8e4454 ("r8152: divide the tx and rx bottom functions")
---
 drivers/net/usb/r8152.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/net/usb/r8152.c +3 -0
index 059d610901d84..fc1458f96e170 100644
@@ -2628,6 +2628,9 @@ static int r8152_poll(struct napi_struct *napi, int budget)
 	struct r8152 *tp = container_of(napi, struct r8152, napi);
 	int work_done;
 
+	if (!budget)
+		return 0;
+
 	work_done = rx_bottom(tp, budget);
 
 	if (work_done < budget) {
Signature · 1 line
2.40.1
author maintainer
#230 lore raw

PATCH 202/219 kcm: Fix memory leak in error path of kcm_sendmsg()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Shigeru Yoshida <[email protected]>

[ Upstream commit c821a88bd720b0046433173185fd841a100d44ad ]

syzbot reported a memory leak like below:

BUG: memory leak
unreferenced object 0xffff88810b088c00 (size 240):
  comm "syz-executor186", pid 5012, jiffies 4294943306 (age 13.680s)
  hex dump (first 32 bytes):
    00 89 08 0b 81 88 ff ff 00 00 00 00 00 00 00 00  ................
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
    [<ffffffff83e5d5ff>] __alloc_skb+0x1ef/0x230 net/core/skbuff.c:634
    [<ffffffff84606e59>] alloc_skb include/linux/skbuff.h:1289 [inline]
    [<ffffffff84606e59>] kcm_sendmsg+0x269/0x1050 net/kcm/kcmsock.c:815
    [<ffffffff83e479c6>] sock_sendmsg_nosec net/socket.c:725 [inline]
    [<ffffffff83e479c6>] sock_sendmsg+0x56/0xb0 net/socket.c:748
    [<ffffffff83e47f55>] ____sys_sendmsg+0x365/0x470 net/socket.c:2494
    [<ffffffff83e4c389>] ___sys_sendmsg+0xc9/0x130 net/socket.c:2548
    [<ffffffff83e4c536>] __sys_sendmsg+0xa6/0x120 net/socket.c:2577
    [<ffffffff84ad7bb8>] do_syscall_x64 arch/x86/entry/common.c:50 [inline]
    [<ffffffff84ad7bb8>] do_syscall_64+0x38/0xb0 arch/x86/entry/common.c:80
    [<ffffffff84c0008b>] entry_SYSCALL_64_after_hwframe+0x63/0xcd

In kcm_sendmsg(), kcm_tx_msg(head)->last_skb is used as a cursor to append newly allocated skbs to 'head'. If some bytes are copied, an error occurred, and jumped to out_error label, 'last_skb' is left unmodified. A later kcm_sendmsg() will use an obsoleted 'last_skb' reference, corrupting the 'head' frag_list and causing the leak.

This patch fixes this issue by properly updating the last allocated skb in 'last_skb'.

Fixes ab7ac4eb9832 ("kcm: Kernel Connection Multiplexor module")
Reported-and-tested-by [email protected]
---
 net/kcm/kcmsock.c | 2 ++
 1 file changed, 2 insertions(+)
net/kcm/kcmsock.c +2 -0
index 6a97662d7548e..dd929a8350740 100644
@@ -1074,6 +1074,8 @@ static int kcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
 
 	if (head != kcm->seq_skb)
 		kfree_skb(head);
+	else if (copied)
+		kcm_tx_msg(head)->last_skb = skb;
 
 	err = sk_stream_error(sk, msg->msg_flags, err);
 
Signature · 1 line
2.40.1
author maintainer
#231 lore raw

PATCH 203/219 platform/mellanox: mlxbf-tmfifo: Drop the Rx packet if no more descriptors

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Liming Sun <[email protected]>

[ Upstream commit 78034cbece79c2d730ad0770b3b7f23eedbbecf5 ]

This commit fixes tmfifo console stuck issue when the virtual networking interface is in down state. In such case, the network Rx descriptors runs out and causes the Rx network packet staying in the head of the tmfifo thus blocking the console packets. The fix is to drop the Rx network packet when no more Rx descriptors. Function name mlxbf_tmfifo_release_pending_pkt() is also renamed to mlxbf_tmfifo_release_pkt() to be more approperiate.

---
 drivers/platform/mellanox/mlxbf-tmfifo.c | 66 ++++++++++++++++++------
 1 file changed, 49 insertions(+), 17 deletions(-)
drivers/platform/mellanox/mlxbf-tmfifo.c +49 -17
index d31fe7eed38df..ee8648a271fda 100644
@@ -56,6 +56,7 @@ struct mlxbf_tmfifo;
  * @vq: pointer to the virtio virtqueue
  * @desc: current descriptor of the pending packet
  * @desc_head: head descriptor of the pending packet
+ * @drop_desc: dummy desc for packet dropping
  * @cur_len: processed length of the current descriptor
  * @rem_len: remaining length of the pending packet
  * @pkt_len: total length of the pending packet
@@ -72,6 +73,7 @@ struct mlxbf_tmfifo_vring {
 	struct virtqueue *vq;
 	struct vring_desc *desc;
 	struct vring_desc *desc_head;
+	struct vring_desc drop_desc;
 	int cur_len;
 	int rem_len;
 	u32 pkt_len;
@@ -83,6 +85,14 @@ struct mlxbf_tmfifo_vring {
 	struct mlxbf_tmfifo *fifo;
 };
 
+/* Check whether vring is in drop mode. */
+#define IS_VRING_DROP(_r) ({ \
+	typeof(_r) (r) = (_r); \
+	(r->desc_head == &r->drop_desc ? true : false); })
+
+/* A stub length to drop maximum length packet. */
+#define VRING_DROP_DESC_MAX_LEN		GENMASK(15, 0)
+
 /* Interrupt types. */
 enum {
 	MLXBF_TM_RX_LWM_IRQ,
@@ -243,6 +253,7 @@ static int mlxbf_tmfifo_alloc_vrings(struct mlxbf_tmfifo *fifo,
 		vring->align = SMP_CACHE_BYTES;
 		vring->index = i;
 		vring->vdev_id = tm_vdev->vdev.id.device;
+		vring->drop_desc.len = VRING_DROP_DESC_MAX_LEN;
 		dev = &tm_vdev->vdev.dev;
 
 		size = vring_size(vring->num, vring->align);
@@ -348,7 +359,7 @@ static u32 mlxbf_tmfifo_get_pkt_len(struct mlxbf_tmfifo_vring *vring,
 	return len;
 }
 
-static void mlxbf_tmfifo_release_pending_pkt(struct mlxbf_tmfifo_vring *vring)
+static void mlxbf_tmfifo_release_pkt(struct mlxbf_tmfifo_vring *vring)
 {
 	struct vring_desc *desc_head;
 	u32 len = 0;
@@ -577,19 +588,25 @@ static void mlxbf_tmfifo_rxtx_word(struct mlxbf_tmfifo_vring *vring,
 
 	if (vring->cur_len + sizeof(u64) <= len) {
 		/* The whole word. */
-		if (is_rx)
-			memcpy(addr + vring->cur_len, &data, sizeof(u64));
-		else
-			memcpy(&data, addr + vring->cur_len, sizeof(u64));
+		if (!IS_VRING_DROP(vring)) {
+			if (is_rx)
+				memcpy(addr + vring->cur_len, &data,
+				       sizeof(u64));
+			else
+				memcpy(&data, addr + vring->cur_len,
+				       sizeof(u64));
+		}
 		vring->cur_len += sizeof(u64);
 	} else {
 		/* Leftover bytes. */
-		if (is_rx)
-			memcpy(addr + vring->cur_len, &data,
-			       len - vring->cur_len);
-		else
-			memcpy(&data, addr + vring->cur_len,
-			       len - vring->cur_len);
+		if (!IS_VRING_DROP(vring)) {
+			if (is_rx)
+				memcpy(addr + vring->cur_len, &data,
+				       len - vring->cur_len);
+			else
+				memcpy(&data, addr + vring->cur_len,
+				       len - vring->cur_len);
+		}
 		vring->cur_len = len;
 	}
 
@@ -690,8 +707,16 @@ static bool mlxbf_tmfifo_rxtx_one_desc(struct mlxbf_tmfifo_vring *vring,
 	/* Get the descriptor of the next packet. */
 	if (!vring->desc) {
 		desc = mlxbf_tmfifo_get_next_pkt(vring, is_rx);
-		if (!desc)
-			return false;
+		if (!desc) {
+			/* Drop next Rx packet to avoid stuck. */
+			if (is_rx) {
+				desc = &vring->drop_desc;
+				vring->desc_head = desc;
+				vring->desc = desc;
+			} else {
+				return false;
+			}
+		}
 	} else {
 		desc = vring->desc;
 	}
@@ -724,17 +749,24 @@ static bool mlxbf_tmfifo_rxtx_one_desc(struct mlxbf_tmfifo_vring *vring,
 		vring->rem_len -= len;
 
 		/* Get the next desc on the chain. */
-		if (vring->rem_len > 0 &&
+		if (!IS_VRING_DROP(vring) && vring->rem_len > 0 &&
 		    (virtio16_to_cpu(vdev, desc->flags) & VRING_DESC_F_NEXT)) {
 			idx = virtio16_to_cpu(vdev, desc->next);
 			desc = &vr->desc[idx];
 			goto mlxbf_tmfifo_desc_done;
 		}
 
-		/* Done and release the pending packet. */
-		mlxbf_tmfifo_release_pending_pkt(vring);
+		/* Done and release the packet. */
 		desc = NULL;
 		fifo->vring[is_rx] = NULL;
+		if (!IS_VRING_DROP(vring)) {
+			mlxbf_tmfifo_release_pkt(vring);
+		} else {
+			vring->pkt_len = 0;
+			vring->desc_head = NULL;
+			vring->desc = NULL;
+			return false;
+		}
 
 		/*
 		 * Make sure the load/store are in order before
@@ -914,7 +946,7 @@ static void mlxbf_tmfifo_virtio_del_vqs(struct virtio_device *vdev)
 
 		/* Release the pending packet. */
 		if (vring->desc)
-			mlxbf_tmfifo_release_pending_pkt(vring);
+			mlxbf_tmfifo_release_pkt(vring);
 		vq = vring->vq;
 		if (vq) {
 			vring->vq = NULL;
Signature · 1 line
2.40.1
author maintainer
#232 lore raw

PATCH 204/219 platform/mellanox: mlxbf-tmfifo: Drop jumbo frames

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Liming Sun <[email protected]>

[ Upstream commit fc4c655821546239abb3cf4274d66b9747aa87dd ]

This commit drops over-sized network packets to avoid tmfifo queue stuck.

---
 drivers/platform/mellanox/mlxbf-tmfifo.c | 24 +++++++++++++++++-------
 1 file changed, 17 insertions(+), 7 deletions(-)
drivers/platform/mellanox/mlxbf-tmfifo.c +17 -7
index ee8648a271fda..a04ff89a7ec44 100644
@@ -205,7 +205,7 @@ static u8 mlxbf_tmfifo_net_default_mac[ETH_ALEN] = {
 static efi_char16_t mlxbf_tmfifo_efi_name[] = L"RshimMacAddr";
 
 /* Maximum L2 header length. */
-#define MLXBF_TMFIFO_NET_L2_OVERHEAD	36
+#define MLXBF_TMFIFO_NET_L2_OVERHEAD	(ETH_HLEN + VLAN_HLEN)
 
 /* Supported virtio-net features. */
 #define MLXBF_TMFIFO_NET_FEATURES \
@@ -623,13 +623,14 @@ static void mlxbf_tmfifo_rxtx_word(struct mlxbf_tmfifo_vring *vring,
  * flag is set.
  */
 static void mlxbf_tmfifo_rxtx_header(struct mlxbf_tmfifo_vring *vring,
-				     struct vring_desc *desc,
+				     struct vring_desc **desc,
 				     bool is_rx, bool *vring_change)
 {
 	struct mlxbf_tmfifo *fifo = vring->fifo;
 	struct virtio_net_config *config;
 	struct mlxbf_tmfifo_msg_hdr hdr;
 	int vdev_id, hdr_len;
+	bool drop_rx = false;
 
 	/* Read/Write packet header. */
 	if (is_rx) {
@@ -649,8 +650,8 @@ static void mlxbf_tmfifo_rxtx_header(struct mlxbf_tmfifo_vring *vring,
 			if (ntohs(hdr.len) >
 			    __virtio16_to_cpu(virtio_legacy_is_little_endian(),
 					      config->mtu) +
-			    MLXBF_TMFIFO_NET_L2_OVERHEAD)
-				return;
+					      MLXBF_TMFIFO_NET_L2_OVERHEAD)
+				drop_rx = true;
 		} else {
 			vdev_id = VIRTIO_ID_CONSOLE;
 			hdr_len = 0;
@@ -665,16 +666,25 @@ static void mlxbf_tmfifo_rxtx_header(struct mlxbf_tmfifo_vring *vring,
 
 			if (!tm_dev2)
 				return;
-			vring->desc = desc;
+			vring->desc = *desc;
 			vring = &tm_dev2->vrings[MLXBF_TMFIFO_VRING_RX];
 			*vring_change = true;
 		}
+
+		if (drop_rx && !IS_VRING_DROP(vring)) {
+			if (vring->desc_head)
+				mlxbf_tmfifo_release_pkt(vring);
+			*desc = &vring->drop_desc;
+			vring->desc_head = *desc;
+			vring->desc = *desc;
+		}
+
 		vring->pkt_len = ntohs(hdr.len) + hdr_len;
 	} else {
 		/* Network virtio has an extra header. */
 		hdr_len = (vring->vdev_id == VIRTIO_ID_NET) ?
 			   sizeof(struct virtio_net_hdr) : 0;
-		vring->pkt_len = mlxbf_tmfifo_get_pkt_len(vring, desc);
+		vring->pkt_len = mlxbf_tmfifo_get_pkt_len(vring, *desc);
 		hdr.type = (vring->vdev_id == VIRTIO_ID_NET) ?
 			    VIRTIO_ID_NET : VIRTIO_ID_CONSOLE;
 		hdr.len = htons(vring->pkt_len - hdr_len);
@@ -723,7 +733,7 @@ static bool mlxbf_tmfifo_rxtx_one_desc(struct mlxbf_tmfifo_vring *vring,
 
 	/* Beginning of a packet. Start to Rx/Tx packet header. */
 	if (vring->pkt_len == 0) {
-		mlxbf_tmfifo_rxtx_header(vring, desc, is_rx, &vring_change);
+		mlxbf_tmfifo_rxtx_header(vring, &desc, is_rx, &vring_change);
 		(*avail)--;
 
 		/* Return if new packet is for another ring. */
Signature · 1 line
2.40.1
author maintainer
#233 lore raw

PATCH 205/219 platform/mellanox: mlxbf-pmc: Fix potential buffer overflows

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Shravan Kumar Ramani <[email protected]>

[ Upstream commit 80ccd40568bcd3655b0fd0be1e9b3379fd6e1056 ]

Replace sprintf with sysfs_emit where possible.
Size check in mlxbf_pmc_event_list_show should account for "\0".

---
 drivers/platform/mellanox/mlxbf-pmc.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)
drivers/platform/mellanox/mlxbf-pmc.c +7 -7
index be967d797c28e..95afcae7b9fa9 100644
@@ -1008,7 +1008,7 @@ static ssize_t mlxbf_pmc_counter_show(struct device *dev,
 	} else
 		return -EINVAL;
 
-	return sprintf(buf, "0x%llx\n", value);
+	return sysfs_emit(buf, "0x%llx\n", value);
 }
 
 /* Store function for "counter" sysfs files */
@@ -1078,13 +1078,13 @@ static ssize_t mlxbf_pmc_event_show(struct device *dev,
 
 	err = mlxbf_pmc_read_event(blk_num, cnt_num, is_l3, &evt_num);
 	if (err)
-		return sprintf(buf, "No event being monitored\n");
+		return sysfs_emit(buf, "No event being monitored\n");
 
 	evt_name = mlxbf_pmc_get_event_name(pmc->block_name[blk_num], evt_num);
 	if (!evt_name)
 		return -EINVAL;
 
-	return sprintf(buf, "0x%llx: %s\n", evt_num, evt_name);
+	return sysfs_emit(buf, "0x%llx: %s\n", evt_num, evt_name);
 }
 
 /* Store function for "event" sysfs files */
@@ -1139,9 +1139,9 @@ static ssize_t mlxbf_pmc_event_list_show(struct device *dev,
 		return -EINVAL;
 
 	for (i = 0, buf[0] = '\0'; i < size; ++i) {
-		len += sprintf(e_info, "0x%x: %s\n", events[i].evt_num,
-			       events[i].evt_name);
-		if (len > PAGE_SIZE)
+		len += snprintf(e_info, sizeof(e_info), "0x%x: %s\n",
+				events[i].evt_num, events[i].evt_name);
+		if (len >= PAGE_SIZE)
 			break;
 		strcat(buf, e_info);
 		ret = len;
@@ -1168,7 +1168,7 @@ static ssize_t mlxbf_pmc_enable_show(struct device *dev,
 
 	value = FIELD_GET(MLXBF_PMC_L3C_PERF_CNT_CFG_EN, perfcnt_cfg);
 
-	return sprintf(buf, "%d\n", value);
+	return sysfs_emit(buf, "%d\n", value);
 }
 
 /* Store function for "enable" sysfs files - only for l3cache */
Signature · 1 line
2.40.1
author maintainer
#234 lore raw

PATCH 206/219 platform/mellanox: mlxbf-pmc: Fix reading of unprogrammed events

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Shravan Kumar Ramani <[email protected]>

[ Upstream commit 0f5969452e162efc50bdc98968fb62b424a9874b ]

This fix involves 2 changes:
- All event regs have a reset value of 0, which is not a valid event_number as per the event_list for most blocks and hence seen as an error. Add a "disable" event with event_number 0 for all blocks.

- The enable bit for each counter need not be checked before reading the event info, and hence removed.

---
 drivers/platform/mellanox/mlxbf-pmc.c | 27 +++++++--------------------
 1 file changed, 7 insertions(+), 20 deletions(-)
drivers/platform/mellanox/mlxbf-pmc.c +7 -20
index 95afcae7b9fa9..2d4bbe99959ef 100644
@@ -191,6 +191,7 @@ static const struct mlxbf_pmc_events mlxbf_pmc_smgen_events[] = {
 };
 
 static const struct mlxbf_pmc_events mlxbf_pmc_trio_events_1[] = {
+	{ 0x0, "DISABLE" },
 	{ 0xa0, "TPIO_DATA_BEAT" },
 	{ 0xa1, "TDMA_DATA_BEAT" },
 	{ 0xa2, "MAP_DATA_BEAT" },
@@ -214,6 +215,7 @@ static const struct mlxbf_pmc_events mlxbf_pmc_trio_events_1[] = {
 };
 
 static const struct mlxbf_pmc_events mlxbf_pmc_trio_events_2[] = {
+	{ 0x0, "DISABLE" },
 	{ 0xa0, "TPIO_DATA_BEAT" },
 	{ 0xa1, "TDMA_DATA_BEAT" },
 	{ 0xa2, "MAP_DATA_BEAT" },
@@ -246,6 +248,7 @@ static const struct mlxbf_pmc_events mlxbf_pmc_trio_events_2[] = {
 };
 
 static const struct mlxbf_pmc_events mlxbf_pmc_ecc_events[] = {
+	{ 0x0, "DISABLE" },
 	{ 0x100, "ECC_SINGLE_ERROR_CNT" },
 	{ 0x104, "ECC_DOUBLE_ERROR_CNT" },
 	{ 0x114, "SERR_INJ" },
@@ -258,6 +261,7 @@ static const struct mlxbf_pmc_events mlxbf_pmc_ecc_events[] = {
 };
 
 static const struct mlxbf_pmc_events mlxbf_pmc_mss_events[] = {
+	{ 0x0, "DISABLE" },
 	{ 0xc0, "RXREQ_MSS" },
 	{ 0xc1, "RXDAT_MSS" },
 	{ 0xc2, "TXRSP_MSS" },
@@ -265,6 +269,7 @@ static const struct mlxbf_pmc_events mlxbf_pmc_mss_events[] = {
 };
 
 static const struct mlxbf_pmc_events mlxbf_pmc_hnf_events[] = {
+	{ 0x0, "DISABLE" },
 	{ 0x45, "HNF_REQUESTS" },
 	{ 0x46, "HNF_REJECTS" },
 	{ 0x47, "ALL_BUSY" },
@@ -323,6 +328,7 @@ static const struct mlxbf_pmc_events mlxbf_pmc_hnf_events[] = {
 };
 
 static const struct mlxbf_pmc_events mlxbf_pmc_hnfnet_events[] = {
+	{ 0x0, "DISABLE" },
 	{ 0x12, "CDN_REQ" },
 	{ 0x13, "DDN_REQ" },
 	{ 0x14, "NDN_REQ" },
@@ -892,7 +898,7 @@ static int mlxbf_pmc_read_event(int blk_num, uint32_t cnt_num, bool is_l3,
 				uint64_t *result)
 {
 	uint32_t perfcfg_offset, perfval_offset;
-	uint64_t perfmon_cfg, perfevt, perfctl;
+	uint64_t perfmon_cfg, perfevt;
 
 	if (cnt_num >= pmc->block[blk_num].counters)
 		return -EINVAL;
@@ -904,25 +910,6 @@ static int mlxbf_pmc_read_event(int blk_num, uint32_t cnt_num, bool is_l3,
 	perfval_offset = perfcfg_offset +
 			 pmc->block[blk_num].counters * MLXBF_PMC_REG_SIZE;
 
-	/* Set counter in "read" mode */
-	perfmon_cfg = FIELD_PREP(MLXBF_PMC_PERFMON_CONFIG_ADDR,
-				 MLXBF_PMC_PERFCTL);
-	perfmon_cfg |= FIELD_PREP(MLXBF_PMC_PERFMON_CONFIG_STROBE, 1);
-	perfmon_cfg |= FIELD_PREP(MLXBF_PMC_PERFMON_CONFIG_WR_R_B, 0);
-
-	if (mlxbf_pmc_write(pmc->block[blk_num].mmio_base + perfcfg_offset,
-			    MLXBF_PMC_WRITE_REG_64, perfmon_cfg))
-		return -EFAULT;
-
-	/* Check if the counter is enabled */
-
-	if (mlxbf_pmc_read(pmc->block[blk_num].mmio_base + perfval_offset,
-			   MLXBF_PMC_READ_REG_64, &perfctl))
-		return -EFAULT;
-
-	if (!FIELD_GET(MLXBF_PMC_PERFCTL_EN0, perfctl))
-		return -EINVAL;
-
 	/* Set counter in "read" mode */
 	perfmon_cfg = FIELD_PREP(MLXBF_PMC_PERFMON_CONFIG_ADDR,
 				 MLXBF_PMC_PERFEVT);
Signature · 1 line
2.40.1
author maintainer
#235 lore raw

PATCH 207/219 platform/mellanox: NVSW_SN2201 should depend on ACPI

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Geert Uytterhoeven <[email protected]>

[ Upstream commit 0a138f1670bd1af13ba6949c48ea86ddd4bf557e ]

The only probing method supported by the Nvidia SN2201 platform driver is probing through an ACPI match table. Hence add a dependency on ACPI, to prevent asking the user about this driver when configuring a kernel without ACPI support.

---
 drivers/platform/mellanox/Kconfig | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/platform/mellanox/Kconfig +2 -2
index 382793e73a60a..30b50920b278c 100644
@@ -80,8 +80,8 @@ config MLXBF_PMC
 
 config NVSW_SN2201
 	tristate "Nvidia SN2201 platform driver support"
-	depends on HWMON
-	depends on I2C
+	depends on HWMON && I2C
+	depends on ACPI || COMPILE_TEST
 	select REGMAP_I2C
 	help
 	  This driver provides support for the Nvidia SN2201 platform.
Signature · 1 line
2.40.1
author maintainer
#236 lore raw

PATCH 208/219 net/tls: do not free tls_rec on async operation in bpf_exec_tx_verdict()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Liu Jian <[email protected]>

[ Upstream commit cfaa80c91f6f99b9342b6557f0f0e1143e434066 ]

I got the below warning when do fuzzing test:

BUG: KASAN: null-ptr-deref in scatterwalk_copychunks+0x320/0x470
Read of size 4 at addr 0000000000000008 by task kworker/u8:1/9

CPU: 0 PID: 9 Comm: kworker/u8:1 Tainted: G           OE
Hardware name: linux,dummy-virt (DT)
Workqueue: pencrypt_parallel padata_parallel_worker
Call trace:
 dump_backtrace+0x0/0x420
 show_stack+0x34/0x44
 dump_stack+0x1d0/0x248
 __kasan_report+0x138/0x140
 kasan_report+0x44/0x6c
 __asan_load4+0x94/0xd0
 scatterwalk_copychunks+0x320/0x470
 skcipher_next_slow+0x14c/0x290
 skcipher_walk_next+0x2fc/0x480
 skcipher_walk_first+0x9c/0x110
 skcipher_walk_aead_common+0x380/0x440
 skcipher_walk_aead_encrypt+0x54/0x70
 ccm_encrypt+0x13c/0x4d0
 crypto_aead_encrypt+0x7c/0xfc
 pcrypt_aead_enc+0x28/0x84
 padata_parallel_worker+0xd0/0x2dc
 process_one_work+0x49c/0xbdc
 worker_thread+0x124/0x880
 kthread+0x210/0x260
 ret_from_fork+0x10/0x18

This is because the value of rec_seq of tls_crypto_info configured by the user program is too large, for example, 0xffffffffffffff. In addition, TLS is asynchronously accelerated. When tls_do_encryption() returns -EINPROGRESS and sk->sk_err is set to EBADMSG due to rec_seq overflow, skmsg is released before the asynchronous encryption process ends. As a result, the UAF problem occurs during the asynchronous processing of the encryption module.

If the operation is asynchronous and the encryption module returns EINPROGRESS, do not free the record information.

Fixes 635d93981786 ("net/tls: free record only on encryption error")
---
 net/tls/tls_sw.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
net/tls/tls_sw.c +2 -2
index 96b4545ea700f..9be00ebbb2341 100644
@@ -802,7 +802,7 @@ static int bpf_exec_tx_verdict(struct sk_msg *msg, struct sock *sk,
 	psock = sk_psock_get(sk);
 	if (!psock || !policy) {
 		err = tls_push_record(sk, flags, record_type);
-		if (err && sk->sk_err == EBADMSG) {
+		if (err && err != -EINPROGRESS && sk->sk_err == EBADMSG) {
 			*copied -= sk_msg_free(sk, msg);
 			tls_free_open_rec(sk);
 			err = -sk->sk_err;
@@ -831,7 +831,7 @@ static int bpf_exec_tx_verdict(struct sk_msg *msg, struct sock *sk,
 	switch (psock->eval) {
 	case __SK_PASS:
 		err = tls_push_record(sk, flags, record_type);
-		if (err && sk->sk_err == EBADMSG) {
+		if (err && err != -EINPROGRESS && sk->sk_err == EBADMSG) {
 			*copied -= sk_msg_free(sk, msg);
 			tls_free_open_rec(sk);
 			err = -sk->sk_err;
Signature · 1 line
2.40.1
author maintainer
#237 lore raw

PATCH 209/219 net: macb: Enable PTP unicast

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Harini Katakam <[email protected]>

[ Upstream commit ee4e92c26c60b7344b7261035683a37da5a6119b ]

Enable transmission and reception of PTP unicast packets by updating PTP unicast config bit and setting current HW mac address as allowed address in PTP unicast filter registers.

Stable-dep-of: 403f0e771457 ("net: macb: fix sleep inside spinlock")

---
 drivers/net/ethernet/cadence/macb.h      |  4 ++++
 drivers/net/ethernet/cadence/macb_main.c | 13 +++++++++++--
 2 files changed, 15 insertions(+), 2 deletions(-)
drivers/net/ethernet/cadence/macb.h +4 -0
index 9c410f93a1039..1aa578c1ca4ad 100644
@@ -95,6 +95,8 @@
 #define GEM_SA4B		0x00A0 /* Specific4 Bottom */
 #define GEM_SA4T		0x00A4 /* Specific4 Top */
 #define GEM_WOL			0x00b8 /* Wake on LAN */
+#define GEM_RXPTPUNI		0x00D4 /* PTP RX Unicast address */
+#define GEM_TXPTPUNI		0x00D8 /* PTP TX Unicast address */
 #define GEM_EFTSH		0x00e8 /* PTP Event Frame Transmitted Seconds Register 47:32 */
 #define GEM_EFRSH		0x00ec /* PTP Event Frame Received Seconds Register 47:32 */
 #define GEM_PEFTSH		0x00f0 /* PTP Peer Event Frame Transmitted Seconds Register 47:32 */
@@ -245,6 +247,8 @@
 #define MACB_TZQ_OFFSET		12 /* Transmit zero quantum pause frame */
 #define MACB_TZQ_SIZE		1
 #define MACB_SRTSM_OFFSET	15 /* Store Receive Timestamp to Memory */
+#define MACB_PTPUNI_OFFSET	20 /* PTP Unicast packet enable */
+#define MACB_PTPUNI_SIZE	1
 #define MACB_OSSMODE_OFFSET	24 /* Enable One Step Synchro Mode */
 #define MACB_OSSMODE_SIZE	1
 #define MACB_MIIONRGMII_OFFSET	28 /* MII Usage on RGMII Interface */
drivers/net/ethernet/cadence/macb_main.c +11 -2
index 5fb991835078a..9470e895591e5 100644
@@ -288,6 +288,11 @@ static void macb_set_hwaddr(struct macb *bp)
 	top = cpu_to_le16(*((u16 *)(bp->dev->dev_addr + 4)));
 	macb_or_gem_writel(bp, SA1T, top);
 
+	if (gem_has_ptp(bp)) {
+		gem_writel(bp, RXPTPUNI, bottom);
+		gem_writel(bp, TXPTPUNI, bottom);
+	}
+
 	/* Clear unused address register sets */
 	macb_or_gem_writel(bp, SA2B, 0);
 	macb_or_gem_writel(bp, SA2T, 0);
@@ -721,8 +726,12 @@ static void macb_mac_link_up(struct phylink_config *config,
 
 	spin_unlock_irqrestore(&bp->lock, flags);
 
-	/* Enable Rx and Tx */
-	macb_writel(bp, NCR, macb_readl(bp, NCR) | MACB_BIT(RE) | MACB_BIT(TE));
+	/* Enable Rx and Tx; Enable PTP unicast */
+	ctrl = macb_readl(bp, NCR);
+	if (gem_has_ptp(bp))
+		ctrl |= MACB_BIT(PTPUNI);
+
+	macb_writel(bp, NCR, ctrl | MACB_BIT(RE) | MACB_BIT(TE));
 
 	netif_tx_wake_all_queues(ndev);
 }
Signature · 1 line
2.40.1
author maintainer
#238 lore raw

PATCH 210/219 net: macb: fix sleep inside spinlock

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sascha Hauer <[email protected]>

[ Upstream commit 403f0e771457e2b8811dc280719d11b9bacf10f4 ]

macb_set_tx_clk() is called under a spinlock but itself calls clk_set_rate() which can sleep. This results in:

| BUG: sleeping function called from invalid context at kernel/locking/mutex.c:580
| pps pps1: new PPS source ptp1
| in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 40, name: kworker/u4:3
| preempt_count: 1, expected: 0
| RCU nest depth: 0, expected: 0
| 4 locks held by kworker/u4:3/40:
|  #0: ffff000003409148
| macb ff0c0000.ethernet: gem-ptp-timer ptp clock registered.
|  ((wq_completion)events_power_efficient){+.+.}-{0:0}, at: process_one_work+0x14c/0x51c
|  #1: ffff8000833cbdd8 ((work_completion)(&pl->resolve)){+.+.}-{0:0}, at: process_one_work+0x14c/0x51c
|  #2: ffff000004f01578 (&pl->state_mutex){+.+.}-{4:4}, at: phylink_resolve+0x44/0x4e8
|  #3: ffff000004f06f50 (&bp->lock){....}-{3:3}, at: macb_mac_link_up+0x40/0x2ac
| irq event stamp: 113998
| hardirqs last  enabled at (113997): [<ffff800080e8503c>] _raw_spin_unlock_irq+0x30/0x64
| hardirqs last disabled at (113998): [<ffff800080e84478>] _raw_spin_lock_irqsave+0xac/0xc8
| softirqs last  enabled at (113608): [<ffff800080010630>] __do_softirq+0x430/0x4e4
| softirqs last disabled at (113597): [<ffff80008001614c>] ____do_softirq+0x10/0x1c
| CPU: 0 PID: 40 Comm: kworker/u4:3 Not tainted 6.5.0-11717-g9355ce8b2f50-dirty #368
| Hardware name: ... ZynqMP ... (DT)
| Workqueue: events_power_efficient phylink_resolve
| Call trace:
|  dump_backtrace+0x98/0xf0
|  show_stack+0x18/0x24
|  dump_stack_lvl+0x60/0xac
|  dump_stack+0x18/0x24
|  __might_resched+0x144/0x24c
|  __might_sleep+0x48/0x98
|  __mutex_lock+0x58/0x7b0
|  mutex_lock_nested+0x24/0x30
|  clk_prepare_lock+0x4c/0xa8
|  clk_set_rate+0x24/0x8c
|  macb_mac_link_up+0x25c/0x2ac
|  phylink_resolve+0x178/0x4e8
|  process_one_work+0x1ec/0x51c
|  worker_thread+0x1ec/0x3e4
|  kthread+0x120/0x124
|  ret_from_fork+0x10/0x20

The obvious fix is to move the call to macb_set_tx_clk() out of the protected area. This seems safe as rx and tx are both disabled anyway at this point.
It is however not entirely clear what the spinlock shall protect. It could be the read-modify-write access to the NCFGR register, but this is accessed in macb_set_rx_mode() and macb_set_rxcsum_feature() as well without holding the spinlock. It could also be the register accesses done in mog_init_rings() or macb_init_buffers(), but again these functions are called without holding the spinlock in macb_hresp_error_task(). The locking seems fishy in this driver and it might deserve another look before this patch is applied.

Fixes 633e98a711ac0 ("net: macb: use resolved link config in mac_link_up()")
---
 drivers/net/ethernet/cadence/macb_main.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/net/ethernet/cadence/macb_main.c +3 -2
index 9470e895591e5..54b032a46b48a 100644
@@ -705,8 +705,6 @@ static void macb_mac_link_up(struct phylink_config *config,
 		if (rx_pause)
 			ctrl |= MACB_BIT(PAE);
 
-		macb_set_tx_clk(bp, speed);
-
 		/* Initialize rings & buffers as clearing MACB_BIT(TE) in link down
 		 * cleared the pipeline and control registers.
 		 */
@@ -726,6 +724,9 @@ static void macb_mac_link_up(struct phylink_config *config,
 
 	spin_unlock_irqrestore(&bp->lock, flags);
 
+	if (!(bp->caps & MACB_CAPS_MACB_IS_EMAC))
+		macb_set_tx_clk(bp, speed);
+
 	/* Enable Rx and Tx; Enable PTP unicast */
 	ctrl = macb_readl(bp, NCR);
 	if (gem_has_ptp(bp))
Signature · 1 line
2.40.1
author maintainer
#239 lore raw

PATCH 211/219 ipv6: fix ip6_sock_set_addr_preferences() typo

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit 8cdd9f1aaedf823006449faa4e540026c692ac43 ]

ip6_sock_set_addr_preferences() second argument should be an integer.

SUNRPC attempts to set IPV6_PREFER_SRC_PUBLIC were translated to IPV6_PREFER_SRC_TMP

Fixes 18d5ad623275 ("ipv6: add ip6_sock_set_addr_preferences")
Cc Christoph Hellwig <[email protected]>
Cc Chuck Lever <[email protected]>
---
 include/net/ipv6.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/net/ipv6.h +1 -1
index e4ceef687c1c2..8ced112167599 100644
@@ -1322,7 +1322,7 @@ static inline int __ip6_sock_set_addr_preferences(struct sock *sk, int val)
 	return 0;
 }
 
-static inline int ip6_sock_set_addr_preferences(struct sock *sk, bool val)
+static inline int ip6_sock_set_addr_preferences(struct sock *sk, int val)
 {
 	int ret;
 
Signature · 1 line
2.40.1
author maintainer
#240 lore raw

PATCH 212/219 ipv6: Remove in6addr_any alternatives.

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kuniyuki Iwashima <[email protected]>

[ Upstream commit 8cdc3223e78c43e1b60ea1c536a103e32fdca3c5 ]

Some code defines the IPv6 wildcard address as a local variable and use it with memcmp() or ipv6_addr_equal().

Let's use in6addr_any and ipv6_addr_any() instead.

Stable-dep-of: aa99e5f87bd5 ("tcp: Fix bind() regression for v4-mapped-v6 wildcard address.")

---
 .../net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c |  5 ++---
 include/net/ip6_fib.h                                 |  9 +++------
 include/trace/events/fib.h                            |  5 ++---
 include/trace/events/fib6.h                           |  5 +----
 net/ethtool/ioctl.c                                   | 10 +++++-----
 net/ipv4/inet_hashtables.c                            | 11 ++++-------
 6 files changed, 17 insertions(+), 28 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c +2 -3
index 72b61f66df37a..cd15d36b1507e 100644
@@ -97,7 +97,6 @@ int mlx5e_tc_set_attr_rx_tun(struct mlx5e_tc_flow *flow,
 #if IS_ENABLED(CONFIG_INET) && IS_ENABLED(CONFIG_IPV6)
 	else if (ip_version == 6) {
 		int ipv6_size = MLX5_FLD_SZ_BYTES(ipv6_layout, ipv6);
-		struct in6_addr zerov6 = {};
 
 		daddr = MLX5_ADDR_OF(fte_match_param, spec->match_value,
 				     outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6);
@@ -105,8 +104,8 @@ int mlx5e_tc_set_attr_rx_tun(struct mlx5e_tc_flow *flow,
 				     outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6);
 		memcpy(&tun_attr->dst_ip.v6, daddr, ipv6_size);
 		memcpy(&tun_attr->src_ip.v6, saddr, ipv6_size);
-		if (!memcmp(&tun_attr->dst_ip.v6, &zerov6, sizeof(zerov6)) ||
-		    !memcmp(&tun_attr->src_ip.v6, &zerov6, sizeof(zerov6)))
+		if (ipv6_addr_any(&tun_attr->dst_ip.v6) ||
+		    ipv6_addr_any(&tun_attr->src_ip.v6))
 			return 0;
 	}
 #endif
include/net/ip6_fib.h +3 -6
index a92f6eb853068..fa4e6af382e2a 100644
@@ -472,13 +472,10 @@ void rt6_get_prefsrc(const struct rt6_info *rt, struct in6_addr *addr)
 	rcu_read_lock();
 
 	from = rcu_dereference(rt->from);
-	if (from) {
+	if (from)
 		*addr = from->fib6_prefsrc.addr;
-	} else {
-		struct in6_addr in6_zero = {};
-
-		*addr = in6_zero;
-	}
+	else
+		*addr = in6addr_any;
 
 	rcu_read_unlock();
 }
include/trace/events/fib.h +2 -3
index c2300c407f583..76297ecd4935c 100644
@@ -36,7 +36,6 @@ TRACE_EVENT(fib_table_lookup,
 	),
 
 	TP_fast_assign(
-		struct in6_addr in6_zero = {};
 		struct net_device *dev;
 		struct in6_addr *in6;
 		__be32 *p32;
@@ -74,7 +73,7 @@ TRACE_EVENT(fib_table_lookup,
 				*p32 = nhc->nhc_gw.ipv4;
 
 				in6 = (struct in6_addr *)__entry->gw6;
-				*in6 = in6_zero;
+				*in6 = in6addr_any;
 			} else if (nhc->nhc_gw_family == AF_INET6) {
 				p32 = (__be32 *) __entry->gw4;
 				*p32 = 0;
@@ -87,7 +86,7 @@ TRACE_EVENT(fib_table_lookup,
 			*p32 = 0;
 
 			in6 = (struct in6_addr *)__entry->gw6;
-			*in6 = in6_zero;
+			*in6 = in6addr_any;
 		}
 	),
 
include/trace/events/fib6.h +1 -4
index 6e821eb794503..4d3e607b3cdec 100644
@@ -68,11 +68,8 @@ TRACE_EVENT(fib6_table_lookup,
 			strcpy(__entry->name, "-");
 		}
 		if (res->f6i == net->ipv6.fib6_null_entry) {
-			struct in6_addr in6_zero = {};
-
 			in6 = (struct in6_addr *)__entry->gw;
-			*in6 = in6_zero;
-
+			*in6 = in6addr_any;
 		} else if (res->nh) {
 			in6 = (struct in6_addr *)__entry->gw;
 			*in6 = res->nh->fib_nh_gw6;
net/ethtool/ioctl.c +5 -5
index 940c0e27be735..e31d1247b9f08 100644
@@ -27,6 +27,7 @@
 #include <linux/net.h>
 #include <linux/pm_runtime.h>
 #include <net/devlink.h>
+#include <net/ipv6.h>
 #include <net/xdp_sock_drv.h>
 #include <net/flow_offload.h>
 #include <linux/ethtool_netlink.h>
@@ -3090,7 +3091,6 @@ struct ethtool_rx_flow_rule *
 ethtool_rx_flow_rule_create(const struct ethtool_rx_flow_spec_input *input)
 {
 	const struct ethtool_rx_flow_spec *fs = input->fs;
-	static struct in6_addr zero_addr = {};
 	struct ethtool_rx_flow_match *match;
 	struct ethtool_rx_flow_rule *flow;
 	struct flow_action_entry *act;
@@ -3196,20 +3196,20 @@ ethtool_rx_flow_rule_create(const struct ethtool_rx_flow_spec_input *input)
 
 		v6_spec = &fs->h_u.tcp_ip6_spec;
 		v6_m_spec = &fs->m_u.tcp_ip6_spec;
-		if (memcmp(v6_m_spec->ip6src, &zero_addr, sizeof(zero_addr))) {
+		if (!ipv6_addr_any((struct in6_addr *)v6_m_spec->ip6src)) {
 			memcpy(&match->key.ipv6.src, v6_spec->ip6src,
 			       sizeof(match->key.ipv6.src));
 			memcpy(&match->mask.ipv6.src, v6_m_spec->ip6src,
 			       sizeof(match->mask.ipv6.src));
 		}
-		if (memcmp(v6_m_spec->ip6dst, &zero_addr, sizeof(zero_addr))) {
+		if (!ipv6_addr_any((struct in6_addr *)v6_m_spec->ip6dst)) {
 			memcpy(&match->key.ipv6.dst, v6_spec->ip6dst,
 			       sizeof(match->key.ipv6.dst));
 			memcpy(&match->mask.ipv6.dst, v6_m_spec->ip6dst,
 			       sizeof(match->mask.ipv6.dst));
 		}
-		if (memcmp(v6_m_spec->ip6src, &zero_addr, sizeof(zero_addr)) ||
-		    memcmp(v6_m_spec->ip6dst, &zero_addr, sizeof(zero_addr))) {
+		if (!ipv6_addr_any((struct in6_addr *)v6_m_spec->ip6src) ||
+		    !ipv6_addr_any((struct in6_addr *)v6_m_spec->ip6dst)) {
 			match->dissector.used_keys |=
 				BIT(FLOW_DISSECTOR_KEY_IPV6_ADDRS);
 			match->dissector.offset[FLOW_DISSECTOR_KEY_IPV6_ADDRS] =
net/ipv4/inet_hashtables.c +4 -7
index c19b462662ad0..38a92c8c66863 100644
@@ -813,13 +813,11 @@ bool inet_bind2_bucket_match_addr_any(const struct inet_bind2_bucket *tb, const
 				      unsigned short port, int l3mdev, const struct sock *sk)
 {
 #if IS_ENABLED(CONFIG_IPV6)
-	struct in6_addr addr_any = {};
-
 	if (sk->sk_family != tb->family) {
 		if (sk->sk_family == AF_INET)
 			return net_eq(ib2_net(tb), net) && tb->port == port &&
 				tb->l3mdev == l3mdev &&
-				ipv6_addr_equal(&tb->v6_rcv_saddr, &addr_any);
+				ipv6_addr_any(&tb->v6_rcv_saddr);
 
 		return false;
 	}
@@ -827,7 +825,7 @@ bool inet_bind2_bucket_match_addr_any(const struct inet_bind2_bucket *tb, const
 	if (sk->sk_family == AF_INET6)
 		return net_eq(ib2_net(tb), net) && tb->port == port &&
 			tb->l3mdev == l3mdev &&
-			ipv6_addr_equal(&tb->v6_rcv_saddr, &addr_any);
+			ipv6_addr_any(&tb->v6_rcv_saddr);
 	else
 #endif
 		return net_eq(ib2_net(tb), net) && tb->port == port &&
@@ -853,11 +851,10 @@ inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, in
 {
 	struct inet_hashinfo *hinfo = tcp_or_dccp_get_hashinfo(sk);
 	u32 hash;
-#if IS_ENABLED(CONFIG_IPV6)
-	struct in6_addr addr_any = {};
 
+#if IS_ENABLED(CONFIG_IPV6)
 	if (sk->sk_family == AF_INET6)
-		hash = ipv6_portaddr_hash(net, &addr_any, port);
+		hash = ipv6_portaddr_hash(net, &in6addr_any, port);
 	else
 #endif
 		hash = ipv4_portaddr_hash(net, 0, port);
Signature · 1 line
2.40.1
author maintainer
#241 lore raw

PATCH 213/219 tcp: Factorise sk_family-independent comparison in inet_bind2_bucket_match(_addr_any).

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kuniyuki Iwashima <[email protected]>

[ Upstream commit c6d277064b1da7f9015b575a562734de87a7e463 ]

This is a prep patch to make the following patches cleaner that touch inet_bind2_bucket_match() and inet_bind2_bucket_match_addr_any().

Both functions have duplicated comparison for netns, port, and l3mdev. Let's factorise them.

Stable-dep-of: aa99e5f87bd5 ("tcp: Fix bind() regression for v4-mapped-v6 wildcard address.")

---
 net/ipv4/inet_hashtables.c | 28 +++++++++++++---------------
 1 file changed, 13 insertions(+), 15 deletions(-)
net/ipv4/inet_hashtables.c +13 -15
index 38a92c8c66863..71d77b3b6536f 100644
@@ -795,41 +795,39 @@ static bool inet_bind2_bucket_match(const struct inet_bind2_bucket *tb,
 				    const struct net *net, unsigned short port,
 				    int l3mdev, const struct sock *sk)
 {
+	if (!net_eq(ib2_net(tb), net) || tb->port != port ||
+	    tb->l3mdev != l3mdev)
+		return false;
+
 #if IS_ENABLED(CONFIG_IPV6)
 	if (sk->sk_family != tb->family)
 		return false;
 
 	if (sk->sk_family == AF_INET6)
-		return net_eq(ib2_net(tb), net) && tb->port == port &&
-			tb->l3mdev == l3mdev &&
-			ipv6_addr_equal(&tb->v6_rcv_saddr, &sk->sk_v6_rcv_saddr);
-	else
+		return ipv6_addr_equal(&tb->v6_rcv_saddr, &sk->sk_v6_rcv_saddr);
 #endif
-		return net_eq(ib2_net(tb), net) && tb->port == port &&
-			tb->l3mdev == l3mdev && tb->rcv_saddr == sk->sk_rcv_saddr;
+	return tb->rcv_saddr == sk->sk_rcv_saddr;
 }
 
 bool inet_bind2_bucket_match_addr_any(const struct inet_bind2_bucket *tb, const struct net *net,
 				      unsigned short port, int l3mdev, const struct sock *sk)
 {
+	if (!net_eq(ib2_net(tb), net) || tb->port != port ||
+	    tb->l3mdev != l3mdev)
+		return false;
+
 #if IS_ENABLED(CONFIG_IPV6)
 	if (sk->sk_family != tb->family) {
 		if (sk->sk_family == AF_INET)
-			return net_eq(ib2_net(tb), net) && tb->port == port &&
-				tb->l3mdev == l3mdev &&
-				ipv6_addr_any(&tb->v6_rcv_saddr);
+			return ipv6_addr_any(&tb->v6_rcv_saddr);
 
 		return false;
 	}
 
 	if (sk->sk_family == AF_INET6)
-		return net_eq(ib2_net(tb), net) && tb->port == port &&
-			tb->l3mdev == l3mdev &&
-			ipv6_addr_any(&tb->v6_rcv_saddr);
-	else
+		return ipv6_addr_any(&tb->v6_rcv_saddr);
 #endif
-		return net_eq(ib2_net(tb), net) && tb->port == port &&
-			tb->l3mdev == l3mdev && tb->rcv_saddr == 0;
+	return tb->rcv_saddr == 0;
 }
 
 /* The socket's bhash2 hashbucket spinlock must be held when this is called */
Signature · 1 line
2.40.1
author maintainer
#242 lore raw

PATCH 214/219 tcp: Fix bind() regression for v4-mapped-v6 wildcard address.

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kuniyuki Iwashima <[email protected]>

[ Upstream commit aa99e5f87bd54db55dd37cb130bd5eb55933027f ]

Andrei Vagin reported bind() regression with strace logs.

If we bind() a TCPv6 socket to ::FFFF:0.0.0.0 and then bind() a TCPv4 socket to 127.0.0.1, the 2nd bind() should fail but now succeeds.

from socket import *

s1 = socket(AF_INET6, SOCK_STREAM)
s1.bind(('::ffff:0.0.0.0', 0))

s2 = socket(AF_INET, SOCK_STREAM)
s2.bind(('127.0.0.1', s1.getsockname()[1]))

During the 2nd bind(), if tb->family is AF_INET6 and sk->sk_family is AF_INET in inet_bind2_bucket_match_addr_any(), we still need to check if tb has the v4-mapped-v6 wildcard address.

The example above does not work after commit 5456262d2baa ("net: Fix incorrect address comparison when searching for a bind2 bucket"), but the blamed change is not the commit.

Before the commit, the leading zeros of ::FFFF:0.0.0.0 were treated as 0.0.0.0, and the sequence above worked by chance. Technically, this case has been broken since bhash2 was introduced.

Note that if we bind() two sockets to 127.0.0.1 and then ::FFFF:0.0.0.0, the 2nd bind() fails properly because we fall back to using bhash to detect conflicts for the v4-mapped-v6 address.

---
 include/net/ipv6.h         | 5 +++++
 net/ipv4/inet_hashtables.c | 3 ++-
 2 files changed, 7 insertions(+), 1 deletion(-)
include/net/ipv6.h +5 -0
index 8ced112167599..517bdae78614b 100644
@@ -750,6 +750,11 @@ static inline bool ipv6_addr_v4mapped(const struct in6_addr *a)
 					cpu_to_be32(0x0000ffff))) == 0UL;
 }
 
+static inline bool ipv6_addr_v4mapped_any(const struct in6_addr *a)
+{
+	return ipv6_addr_v4mapped(a) && ipv4_is_zeronet(a->s6_addr32[3]);
+}
+
 static inline bool ipv6_addr_v4mapped_loopback(const struct in6_addr *a)
 {
 	return ipv6_addr_v4mapped(a) && ipv4_is_loopback(a->s6_addr32[3]);
net/ipv4/inet_hashtables.c +2 -1
index 71d77b3b6536f..eb522c0374d59 100644
@@ -819,7 +819,8 @@ bool inet_bind2_bucket_match_addr_any(const struct inet_bind2_bucket *tb, const
 #if IS_ENABLED(CONFIG_IPV6)
 	if (sk->sk_family != tb->family) {
 		if (sk->sk_family == AF_INET)
-			return ipv6_addr_any(&tb->v6_rcv_saddr);
+			return ipv6_addr_any(&tb->v6_rcv_saddr) ||
+				ipv6_addr_v4mapped_any(&tb->v6_rcv_saddr);
 
 		return false;
 	}
Signature · 1 line
2.40.1
author maintainer
#243 lore raw

PATCH 215/219 tcp: Fix bind() regression for v4-mapped-v6 non-wildcard address.

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kuniyuki Iwashima <[email protected]>

[ Upstream commit c48ef9c4aed3632566b57ba66cec6ec78624d4cb ]

Since bhash2 was introduced, the example below does not work as expected. These two bind() should conflict, but the 2nd bind() now succeeds.

from socket import *

s1 = socket(AF_INET6, SOCK_STREAM)
s1.bind(('::ffff:127.0.0.1', 0))

s2 = socket(AF_INET, SOCK_STREAM)
s2.bind(('127.0.0.1', s1.getsockname()[1]))

During the 2nd bind() in inet_csk_get_port(), inet_bind2_bucket_find() fails to find the 1st socket's tb2, so inet_bind2_bucket_create() allocates a new tb2 for the 2nd socket. Then, we call inet_csk_bind_conflict() that checks conflicts in the new tb2 by inet_bhash2_conflict(). However, the new tb2 does not include the 1st socket, thus the bind() finally succeeds.

In this case, inet_bind2_bucket_match() must check if AF_INET6 tb2 has the conflicting v4-mapped-v6 address so that inet_bind2_bucket_find() returns the 1st socket's tb2.

Note that if we bind two sockets to 127.0.0.1 and then ::FFFF:127.0.0.1, the 2nd bind() fails properly for the same reason mentinoed in the previous commit.

Fixes 28044fc1d495 ("net: Add a bhash2 table hashed by port and address")
---
 net/ipv4/inet_hashtables.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
net/ipv4/inet_hashtables.c +6 -1
index eb522c0374d59..d79de4b95186b 100644
@@ -800,8 +800,13 @@ static bool inet_bind2_bucket_match(const struct inet_bind2_bucket *tb,
 		return false;
 
 #if IS_ENABLED(CONFIG_IPV6)
-	if (sk->sk_family != tb->family)
+	if (sk->sk_family != tb->family) {
+		if (sk->sk_family == AF_INET)
+			return ipv6_addr_v4mapped(&tb->v6_rcv_saddr) &&
+				tb->v6_rcv_saddr.s6_addr32[3] == sk->sk_rcv_saddr;
+
 		return false;
+	}
 
 	if (sk->sk_family == AF_INET6)
 		return ipv6_addr_equal(&tb->v6_rcv_saddr, &sk->sk_v6_rcv_saddr);
Signature · 1 line
2.40.1
author maintainer
#244 lore raw

PATCH 216/219 ixgbe: fix timestamp configuration code

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vadim Fedorenko <[email protected]>

[ Upstream commit 3c44191dd76cf9c0cc49adaf34384cbd42ef8ad2 ]

The commit in fixes introduced flags to control the status of hardware configuration while processing packets. At the same time another structure is used to provide configuration of timestamper to user-space applications. The way it was coded makes this structures go out of sync easily. The repro is easy for 82599 chips:

[root@hostname ~]# hwstamp_ctl -i eth0 -r 12 -t 1
current settings:
tx_type 0
rx_filter 0
new settings:
tx_type 1
rx_filter 12

The eth0 device is properly configured to timestamp any PTPv2 events.

[root@hostname ~]# hwstamp_ctl -i eth0 -r 1 -t 1 current settings:
tx_type 1
rx_filter 12
SIOCSHWTSTAMP failed: Numerical result out of range
The requested time stamping mode is not supported by the hardware.

The error is properly returned because HW doesn't support all packets timestamping. But the adapter->flags is cleared of timestamp flags even though no HW configuration was done. From that point no RX timestamps are received by user-space application. But configuration shows good values:

[root@hostname ~]# hwstamp_ctl -i eth0
current settings:
tx_type 1
rx_filter 12

Fix the issue by applying new flags only when the HW was actually configured.

---
 drivers/net/ethernet/intel/ixgbe/ixgbe_ptp.c | 28 +++++++++++---------
 1 file changed, 15 insertions(+), 13 deletions(-)
drivers/net/ethernet/intel/ixgbe/ixgbe_ptp.c +15 -13
index f8605f57bd067..75e1383263c1e 100644
@@ -995,6 +995,7 @@ static int ixgbe_ptp_set_timestamp_mode(struct ixgbe_adapter *adapter,
 	u32 tsync_tx_ctl = IXGBE_TSYNCTXCTL_ENABLED;
 	u32 tsync_rx_ctl = IXGBE_TSYNCRXCTL_ENABLED;
 	u32 tsync_rx_mtrl = PTP_EV_PORT << 16;
+	u32 aflags = adapter->flags;
 	bool is_l2 = false;
 	u32 regval;
 
@@ -1012,20 +1013,20 @@ static int ixgbe_ptp_set_timestamp_mode(struct ixgbe_adapter *adapter,
 	case HWTSTAMP_FILTER_NONE:
 		tsync_rx_ctl = 0;
 		tsync_rx_mtrl = 0;
-		adapter->flags &= ~(IXGBE_FLAG_RX_HWTSTAMP_ENABLED |
-				    IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER);
+		aflags &= ~(IXGBE_FLAG_RX_HWTSTAMP_ENABLED |
+			    IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER);
 		break;
 	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
 		tsync_rx_ctl |= IXGBE_TSYNCRXCTL_TYPE_L4_V1;
 		tsync_rx_mtrl |= IXGBE_RXMTRL_V1_SYNC_MSG;
-		adapter->flags |= (IXGBE_FLAG_RX_HWTSTAMP_ENABLED |
-				   IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER);
+		aflags |= (IXGBE_FLAG_RX_HWTSTAMP_ENABLED |
+			   IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER);
 		break;
 	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
 		tsync_rx_ctl |= IXGBE_TSYNCRXCTL_TYPE_L4_V1;
 		tsync_rx_mtrl |= IXGBE_RXMTRL_V1_DELAY_REQ_MSG;
-		adapter->flags |= (IXGBE_FLAG_RX_HWTSTAMP_ENABLED |
-				   IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER);
+		aflags |= (IXGBE_FLAG_RX_HWTSTAMP_ENABLED |
+			   IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER);
 		break;
 	case HWTSTAMP_FILTER_PTP_V2_EVENT:
 	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
@@ -1039,8 +1040,8 @@ static int ixgbe_ptp_set_timestamp_mode(struct ixgbe_adapter *adapter,
 		tsync_rx_ctl |= IXGBE_TSYNCRXCTL_TYPE_EVENT_V2;
 		is_l2 = true;
 		config->rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
-		adapter->flags |= (IXGBE_FLAG_RX_HWTSTAMP_ENABLED |
-				   IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER);
+		aflags |= (IXGBE_FLAG_RX_HWTSTAMP_ENABLED |
+			   IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER);
 		break;
 	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
 	case HWTSTAMP_FILTER_NTP_ALL:
@@ -1051,7 +1052,7 @@ static int ixgbe_ptp_set_timestamp_mode(struct ixgbe_adapter *adapter,
 		if (hw->mac.type >= ixgbe_mac_X550) {
 			tsync_rx_ctl |= IXGBE_TSYNCRXCTL_TYPE_ALL;
 			config->rx_filter = HWTSTAMP_FILTER_ALL;
-			adapter->flags |= IXGBE_FLAG_RX_HWTSTAMP_ENABLED;
+			aflags |= IXGBE_FLAG_RX_HWTSTAMP_ENABLED;
 			break;
 		}
 		fallthrough;
@@ -1062,8 +1063,6 @@ static int ixgbe_ptp_set_timestamp_mode(struct ixgbe_adapter *adapter,
 		 * Delay_Req messages and hardware does not support
 		 * timestamping all packets => return error
 		 */
-		adapter->flags &= ~(IXGBE_FLAG_RX_HWTSTAMP_ENABLED |
-				    IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER);
 		config->rx_filter = HWTSTAMP_FILTER_NONE;
 		return -ERANGE;
 	}
@@ -1095,8 +1094,8 @@ static int ixgbe_ptp_set_timestamp_mode(struct ixgbe_adapter *adapter,
 			       IXGBE_TSYNCRXCTL_TYPE_ALL |
 			       IXGBE_TSYNCRXCTL_TSIP_UT_EN;
 		config->rx_filter = HWTSTAMP_FILTER_ALL;
-		adapter->flags |= IXGBE_FLAG_RX_HWTSTAMP_ENABLED;
-		adapter->flags &= ~IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER;
+		aflags |= IXGBE_FLAG_RX_HWTSTAMP_ENABLED;
+		aflags &= ~IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER;
 		is_l2 = true;
 		break;
 	default:
@@ -1129,6 +1128,9 @@ static int ixgbe_ptp_set_timestamp_mode(struct ixgbe_adapter *adapter,
 
 	IXGBE_WRITE_FLUSH(hw);
 
+	/* configure adapter flags only when HW is actually configured */
+	adapter->flags = aflags;
+
 	/* clear TX/RX time stamp registers, just to be sure */
 	ixgbe_ptp_clear_tx_timestamp(adapter);
 	IXGBE_READ_REG(hw, IXGBE_RXSTMPH);
Signature · 1 line
2.40.1
author maintainer
#245 lore raw

PATCH 217/219 kcm: Fix error handling for SOCK_DGRAM in kcm_sendmsg().

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kuniyuki Iwashima <[email protected]>

[ Upstream commit a22730b1b4bf437c6bbfdeff5feddf54be4aeada ]

syzkaller found a memory leak in kcm_sendmsg(), and commit c821a88bd720 ("kcm: Fix memory leak in error path of kcm_sendmsg()") suppressed it by updating kcm_tx_msg(head)->last_skb if partial data is copied so that the following sendmsg() will resume from the skb.

However, we cannot know how many bytes were copied when we get the error. Thus, we could mess up the MSG_MORE queue.

When kcm_sendmsg() fails for SOCK_DGRAM, we should purge the queue as we do so for UDP by udp_flush_pending_frames().

Even without this change, when the error occurred, the following sendmsg() resumed from a wrong skb and the queue was messed up. However, we have yet to get such a report, and only syzkaller stumbled on it. So, this can be changed safely.

Note this does not change SOCK_SEQPACKET behaviour.

Fixes c821a88bd720 ("kcm: Fix memory leak in error path of kcm_sendmsg()")
Fixes ab7ac4eb9832 ("kcm: Kernel Connection Multiplexor module")
---
 net/kcm/kcmsock.c | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)
net/kcm/kcmsock.c +8 -7
index dd929a8350740..65845c59c0655 100644
@@ -1065,17 +1065,18 @@ static int kcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
 out_error:
 	kcm_push(kcm);
 
-	if (copied && sock->type == SOCK_SEQPACKET) {
+	if (sock->type == SOCK_SEQPACKET) {
 		/* Wrote some bytes before encountering an
 		 * error, return partial success.
 		 */
-		goto partial_message;
-	}
-
-	if (head != kcm->seq_skb)
+		if (copied)
+			goto partial_message;
+		if (head != kcm->seq_skb)
+			kfree_skb(head);
+	} else {
 		kfree_skb(head);
-	else if (copied)
-		kcm_tx_msg(head)->last_skb = skb;
+		kcm->seq_skb = NULL;
+	}
 
 	err = sk_stream_error(sk, msg->msg_flags, err);
 
Signature · 1 line
2.40.1
author maintainer
#246 lore raw

PATCH 218/219 MIPS: Only fiddle with CHECKFLAGS if `need-compiler

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Maciej W. Rozycki <[email protected]>

commit 4fe4a6374c4db9ae2b849b61e84b58685dca565a upstream.

We have originally guarded fiddling with CHECKFLAGS in our arch Makefile by checking for the CONFIG_MIPS variable, not set for targets such as distclean', etc. that neither include .config' nor use the compiler.

Starting from commit 805b2e1d427a ("kbuild: include Makefile.compiler only when compiler is needed") we have had a generic `need-compiler' variable explicitly telling us if the compiler will be used and thus its capabilities need to be checked and expressed in the form of compilation flags. If this variable is not set, then make' functions such as cc-option' are undefined, causing all kinds of weirdness to happen if we expect specific results to be returned, most recently:

cc1: error: '-mloongson-mmi' must be used with '-mhard-float'

messages with configurations such as fuloong2e_defconfig' and the modules_install' target, which does include `.config' and yet does not use the compiler.

Replace the check for CONFIG_MIPS with one for `need-compiler' instead, so as to prevent the compiler from being ever called for CHECKFLAGS when not needed.

---
 arch/mips/Makefile |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/mips/Makefile +1 -1
@@ -350,7 +350,7 @@ KBUILD_CFLAGS += -fno-asynchronous-unwin
 
 KBUILD_LDFLAGS		+= -m $(ld-emul)
 
-ifdef CONFIG_MIPS
+ifdef need-compiler
 CHECKFLAGS += $(shell $(CC) $(KBUILD_CFLAGS) -dM -E -x c /dev/null | \
 	egrep -vw '__GNUC_(MINOR_|PATCHLEVEL_)?_' | \
 	sed -e "s/^\#define /-D'/" -e "s/ /'='/" -e "s/$$/'/" -e 's/\$$/&&/g')
author maintainer
#247 lore raw

PATCH 219/219 drm/amd/display: Fix a bug when searching for insert_above_mpcc

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wesley Chalmers <[email protected]>

commit 3d028d5d60d516c536de1ddd3ebf3d55f3f8983b upstream.

[WHY]
Currently, when insert_plane is called with insert_above_mpcc parameter that is equal to tree->opp_list, the function returns NULL.

[HOW]
Instead, the function should insert the plane at the top of the tree.

---
 drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c +3 -2
@@ -212,8 +212,9 @@ struct mpcc *mpc1_insert_plane(
 		/* check insert_above_mpcc exist in tree->opp_list */
 		struct mpcc *temp_mpcc = tree->opp_list;
 
-		while (temp_mpcc && temp_mpcc->mpcc_bot != insert_above_mpcc)
-			temp_mpcc = temp_mpcc->mpcc_bot;
+		if (temp_mpcc != insert_above_mpcc)
+			while (temp_mpcc && temp_mpcc->mpcc_bot != insert_above_mpcc)
+				temp_mpcc = temp_mpcc->mpcc_bot;
 		if (temp_mpcc == NULL)
 			return NULL;
 	}
maintainer Tested-by on the v1 series
#248 lore raw
In reply to Greg Kroah-Hartman

Hello,

Quoted 5 lines · On Sun, 17 Sep 2023 21:12:07 +0200 Greg Kroah-Hartman…

This is the start of the stable review cycle for the 6.1.54 release. There are 219 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Quoted 8 lines

Responses should be made by Tue, 19 Sep 2023 19:10:04 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

	https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.54-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

This rc kernel passes DAMON functionality test[1] on my test machine. Attaching the test results summary below. Please note that I retrieved the kernel from linux-stable-rc tree[2].

[1] https://github.com/awslabs/damon-tests/tree/next/corr
[2] 89fc7c511aa5 ("Linux 6.1.54-rc1")

Thanks,
SJ

[...]

---

# .config:1408:warning: override: reassigning to symbol CGROUPS ok 15 selftests: damon-tests: build_nomemcg.sh

# kselftest dir '/home/sjpark/damon-tests-cont/linux/tools/testing/selftests/damon-tests' is in dirty state.
# the log is at '/home/sjpark/log'.
 [32m
ok 1 selftests: damon: debugfs_attrs.sh
ok 2 selftests: damon: debugfs_schemes.sh
ok 3 selftests: damon: debugfs_target_ids.sh
ok 4 selftests: damon: debugfs_empty_targets.sh
ok 5 selftests: damon: debugfs_huge_count_read_write.sh
ok 6 selftests: damon: debugfs_duplicate_context_creation.sh
ok 7 selftests: damon: sysfs.sh
ok 1 selftests: damon-tests: kunit.sh
ok 2 selftests: damon-tests: huge_count_read_write.sh
ok 3 selftests: damon-tests: buffer_overflow.sh
ok 4 selftests: damon-tests: rm_contexts.sh
ok 5 selftests: damon-tests: record_null_deref.sh
ok 6 selftests: damon-tests: dbgfs_target_ids_read_before_terminate_race.sh
ok 7 selftests: damon-tests: dbgfs_target_ids_pid_leak.sh
ok 8 selftests: damon-tests: damo_tests.sh
ok 9 selftests: damon-tests: masim-record.sh
ok 10 selftests: damon-tests: build_i386.sh
ok 11 selftests: damon-tests: build_m68k.sh
ok 12 selftests: damon-tests: build_arm64.sh
ok 13 selftests: damon-tests: build_i386_idle_flag.sh
ok 14 selftests: damon-tests: build_i386_highpte.sh
ok 15 selftests: damon-tests: build_nomemcg.sh
 [33m
 [92mPASS [39m
_remote_run_corr.sh SUCCESS
Tested-by on the v1 series
#249 lore raw
In reply to Greg Kroah-Hartman

Hi Greg

Quoted 10 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 6.1.54 release. There are 219 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Tue, 19 Sep 2023 19:10:04 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 8 lines
        https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.54-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

thanks,

greg k-h

6.1.54-rc1 tested.

Build successfully completed.
Boot successfully completed.
No dmesg regressions.
Video output normal.
Sound output normal.

Lenovo ThinkPad X1 Carbon Gen10(Intel i7-1260P(x86_64) arch linux)

Thanks

Tested-by on the v1 series
#250 lore raw
In reply to Greg Kroah-Hartman
Quoted 5 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 6.1.54 release. There are 219 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Successfully compiled and installed bindeb-pkgs on my computer (Acer Aspire E15, Intel Core i3 Haswell). No noticeable regressions.

Signature · 1 line
An old man doll... just what I always wanted! - Clara
Tested-by on the v1 series
#252 lore raw
In reply to Greg Kroah-Hartman
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 6.1.54 release. There are 219 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Tue, 19 Sep 2023 19:10:04 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 8 lines
	https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.54-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

thanks,

greg k-h

Built and booted successfully on RISC-V RV64 (HiFive Unmatched).

maintainer Tested-by on the v1 series
#253 lore raw
In reply to Greg Kroah-Hartman
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 6.1.54 release. There are 219 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Tue, 19 Sep 2023 19:10:04 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 8 lines
	https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.54-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

thanks,

greg k-h

All tests passing for Tegra ...

Test results for stable-v6.1:

    11 builds:	11 pass, 0 fail
    28 boots:	28 pass, 0 fail
    130 tests:	130 pass, 0 fail

Linux version: 6.1.54-rc1-g89fc7c511aa5
Boards tested: tegra124-jetson-tk1, tegra186-p2771-0000,

                tegra194-p2972-0000, tegra194-p3509-0000+p3668-0000,
                tegra20-ventana, tegra210-p2371-2180,
                tegra210-p3450-0000, tegra30-cardhu-a04

Jon

Tested-by on the v1 series
#254 lore raw
In reply to Greg Kroah-Hartman
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 6.1.54 release. There are 219 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Tue, 19 Sep 2023 19:10:04 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 8 lines
	https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.54-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

thanks,

greg k-h

On ARCH_BRCMSTB using 32-bit and 64-bit ARM kernels, build tested on BMIPS_GENERIC:

Signature · 1 line
Florian
Tested-by on the v1 series
#255 lore raw
In reply to Greg Kroah-Hartman
Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 6.1.54 release. There are 219 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Tue, 19 Sep 2023 19:10:04 +0000. Anything received after that time might be too late.

Build results:

	total: 157 pass: 157 fail: 0
Qemu test results:
	total: 529 pass: 529 fail: 0

Guenter

Tested-by Linux Kernel Functional Testing on the v1 series
#256 lore raw
In reply to Greg Kroah-Hartman
Quoted 10 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 6.1.54 release. There are 219 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Tue, 19 Sep 2023 19:10:04 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 8 lines
        https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.54-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

thanks,

greg k-h

Results from Linaro’s test farm.
No regressions on arm64, arm, x86_64, and i386.

## Build
* kernel: 6.1.54-rc1
* git: https://gitlab.com/Linaro/lkft/mirrors/stable/linux-stable-rc
* git branch: linux-6.1.y
* git commit: 89fc7c511aa5cd0b21e82ec42611db04d9e3b7c2
* git describe: v6.1.52-813-g89fc7c511aa5
* test details:
https://qa-reports.linaro.org/lkft/linux-stable-rc-linux-6.1.y/build/v6.1.52-813-g89fc7c511aa5

## Test Regressions (compared to v6.1.52)

## Metric Regressions (compared to v6.1.52)

## Test Fixes (compared to v6.1.52)

## Metric Fixes (compared to v6.1.52)

## Test result summary
total: 206086, pass: 176646, fail: 2859, skip: 26303, xfail: 278

## Build Summary
* arc: 10 total, 10 passed, 0 failed
* arm: 284 total, 283 passed, 1 failed
* arm64: 89 total, 87 passed, 2 failed
* i386: 67 total, 65 passed, 2 failed
* mips: 56 total, 54 passed, 2 failed
* parisc: 7 total, 7 passed, 0 failed
* powerpc: 70 total, 68 passed, 2 failed
* riscv: 28 total, 26 passed, 2 failed
* s390: 28 total, 27 passed, 1 failed
* sh: 26 total, 24 passed, 2 failed
* sparc: 14 total, 14 passed, 0 failed
* x86_64: 76 total, 72 passed, 4 failed

## Test suites summary
* boot
* kselftest-android
* kselftest-arm64
* kselftest-breakpoints
* kselftest-capabilities
* kselftest-clone3
* kselftest-core
* kselftest-cpu-hotplug
* kselftest-drivers-dma-buf
* kselftest-exec
* kselftest-fpu
* kselftest-ftrace
* kselftest-futex
* kselftest-gpio
* kselftest-intel_pstate
* kselftest-ipc
* kselftest-ir
* kselftest-kcmp
* kselftest-kexec
* kselftest-lib
* kselftest-membarrier
* kselftest-memfd
* kselftest-memory-hotplug
* kselftest-mincore
* kselftest-mount
* kselftest-mqueue
* kselftest-net
* kselftest-net-forwarding
* kselftest-net-mptcp
* kselftest-netfilter
* kselftest-nsfs
* kselftest-openat2
* kselftest-pid_namespace
* kselftest-pidfd
* kselftest-proc
* kselftest-pstore
* kselftest-ptrace
* kselftest-rseq
* kselftest-rtc
* kselftest-seccomp
* kselftest-sigaltstack
* kselftest-size
* kselftest-splice
* kselftest-static_keys
* kselftest-sync
* kselftest-sysctl
* kselftest-tc-testing
* kselftest-timens
* kselftest-timers
* kselftest-tmpfs
* kselftest-tpm2
* kselftest-user
* kselftest-user_events
* kselftest-vDSO
* kselftest-vm
* kselftest-watchdog
* kselftest-x86
* kunit
* kvm-unit-tests
* libgpiod
* log-parser-boot
* log-parser-test
* ltp-cap_bounds
* ltp-commands
* ltp-containers
* ltp-controllers
* ltp-cpuhotplug
* ltp-crypto
* ltp-cve
* ltp-dio
* ltp-fcntl-locktests
* ltp-filecaps
* ltp-fs
* ltp-fs_bind
* ltp-fs_perms_simple
* ltp-fsx
* ltp-hugetlb
* ltp-io
* ltp-ipc
* ltp-math
* ltp-mm
* ltp-nptl
* ltp-pty
* ltp-sched
* ltp-securebits
* ltp-smoke
* ltp-syscalls
* ltp-tracing
* perf
* rcutorture

--
Linaro LKFT
https://lkft.linaro.org

Tested-by on the v1 series
#257 lore raw
In reply to Greg Kroah-Hartman
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 6.1.54 release. There are 219 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Tue, 19 Sep 2023 19:10:04 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 9 lines
	https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.54-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

thanks,

greg k-h

Compiled and booted on my test system. No dmesg regressions.

thanks,
-- Shuah