Re: [PATCH] drivers/char/ipmi_si: prevent null deref during module exit

From: Corey Minyard
Date: Thu Nov 09 2017 - 13:54:15 EST


On 11/09/2017 12:38 PM, Corey Minyard wrote:
On 11/09/2017 12:02 PM, Andrew Banman wrote:
On 11/8/17 2:00 PM, Corey Minyard wrote:
On 11/08/2017 11:11 AM, Andrew Banman wrote:
On 11/8/17 11:06 AM, Andrew Banman wrote:
If there are uninitialized SMIs in the smi_infos list, i.e. with no
handlers set, then disable_si_irq() in cleanup_smi_one() will hit a null
pointer dereference when the former attempts to start the check enables
transaction. Thus, we panic during module exit.
I think this points to a broader problem of holding uninitialized smi_info
structs in smi_infos list. There are many places where handlers and other struct
members are assumed. Maybe a better design would be to remove SMIs from the list
if we have no intention of initializing them?

This begs the question: How did you produce this? From what I can tell,
there is no way you can get to this code if you don't have a working and
initialized smi_info structure, and that's not the only place this would
have to be fixed if it wasn't. So it's not what you assume, I don't think
it's an uninitialized smi_info structure on the list.
My kernel is missing 0944d889a2, wherein the dmi handling moved to a platform
device. Without this commit the driver discovers both ACPI and SMBIOS records
but only initializes the former; this is visible in the tracebacks below.

Is there a case for pushing a similar patch to stable releases predating 0944d889a2
to prevent this crash?

It looks like you are using a Redhat 3.10 kernel. Is that right?

Actually, never mind, I misread something. The following change should fix your issue:

index b89078b..d6d9258 100644
--- a/drivers/char/ipmi/ipmi_si_intf.c
+++ b/drivers/char/ipmi/ipmi_si_intf.c
@@ -1805,8 +1805,10 @@ static struct smi_info *smi_info_alloc(void)
Â{
ÂÂÂÂ struct smi_info *info = kzalloc(sizeof(*info), GFP_KERNEL);

-ÂÂÂ if (info)
+ÂÂÂ if (info) {
ÂÂÂÂ ÂÂÂ spin_lock_init(&info->si_lock);
+ÂÂÂ ÂÂÂ info->interrupt_disabled = true;
+ÂÂÂ }
ÂÂÂÂ return info;
Â}

@@ -3603,7 +3605,6 @@ static int try_smi_init(struct smi_info *new_smi)
ÂÂÂÂ for (i = 0; i < SI_NUM_STATS; i++)
ÂÂÂÂ ÂÂÂ atomic_set(&new_smi->stats[i], 0);

-ÂÂÂ new_smi->interrupt_disabled = true;
ÂÂÂÂ atomic_set(&new_smi->need_watch, 0);

ÂÂÂÂ rv = try_enable_event_buffer(new_smi);



-corey

Thank you for your time,

Andrew

As usual with these sorts of things, please send tracebacks and reproduction
procedures.
# dmesg | grep -i ipmi
[ÂÂ 14.883554] ipmi message handler version 39.2
[ÂÂ 14.905543] ipmi device interface
[ÂÂ 14.931149] ipmi_si IPI0001:00: ipmi_si: probing via ACPI
[ 14.937254] ipmi_si IPI0001:00: [io 0x0ce4-0x0ce6] regsize 1 spacing 1 irq 6
[ÂÂ 14.946767] ipmi_si: Adding ACPI-specified bt state machine
[ÂÂ 14.954621] IPMI System Interface driver.
[ÂÂ 14.975257] ipmi_si: probing via SMBIOS
[ÂÂ 14.980446] ipmi_si: SMBIOS: mem 0xce4 regsize 1 spacing 1 irq 6
[ÂÂ 14.987163] ipmi_si: Adding SMBIOS-specified bt state machine
[ÂÂ 14.996706] ipmi_si: probing via SPMI
[ÂÂ 15.002351] ipmi_si: SPMI: io 0xce4 regsize 1 spacing 1 irq 6
[ÂÂ 15.010320] ipmi_si: Adding SPMI-specified bt state machine duplicate interface
[ÂÂ 15.020056] ipmi_si: Trying ACPI-specified bt state machine at i/o address 0xce4, slave address 0x0, irq 6
[ÂÂ 15.066306] ipmi_si IPI0001:00: Using irq 6
[ÂÂ 15.077855] IPMI BT: req2rsp=2 secs retries=1
[ÂÂ 15.130045] ipmi_si IPI0001:00: Found new BMC (man_id: 0x000000, prod_id: 0x0101, dev_id: 0x20)
[ÂÂ 15.140472] ipmi_si IPI0001:00: IPMI bt interface initialized
# rmmod ipmi_si
[ÂÂ 85.492578] BUG: unable to handle kernel NULL pointer dereference at 0000000000000010
[ÂÂ 85.501346] IP: [<ffffffffc027840e>] start_check_enables+0x3e/0x80 [ipmi_si]
[ÂÂ 85.509233] PGD 7fb3875067 PUD 7fbc222067 PMD 0
[ÂÂ 85.514421] Oops: 0000 [#1] SMP
[ÂÂ 85.518056] Modules linked in: xt_CHECKSUM iptable_mangle ipt_MASQUERADE nf_nat_masquerade_ipv4 iptable_nat nf_nat_ipv4 nf_nat nf_conntrack_ipv4 nf_defrag_ipv4 xt_conntrack nf_conntrack ipt_REJECT nf_reject_ipv4 tun bridge stp llc ebtable_filter ebtables ip6table_filter ip6_tables iptable_filter dm_mirror dm_region_hash dm_log dm_mod iTCO_wdt iTCO_vendor_support intel_powerclamp coretemp vfat intel_rapl fat iosf_mbi kvm_intel kvm irqbypass crc32_pclmul ghash_clmulni_intel aesni_intel lrw i40e gf128mul glue_helper ablk_helper cryptd ptp pcspkr pps_core joydev sg i2c_i801 shpchp lpc_ich ipmi_si(-) wmi ipmi_devintf ipmi_msghandler nfit libnvdimm acpi_cpufreq nfsd auth_rpcgss nfs_acl lockd numatools(OE) grace hwperf(OE) binfmt_misc sunrpc ip_tables sr_mod cdrom xfs libcrc32c sd_mod crc_t10dif crct10dif_generic mgag200 i2c_algo_bit drm_kms_helper syscopyarea sysfillrect crct10dif_pclmul crct10dif_common sysimgblt crc32c_intel fb_sys_fops ahci ttm libahci uas drm libata usb_storage i2c_core
[ÂÂ 85.616799] CPU: 241 PID: 5743 Comm: rmmod Tainted: GÂÂÂÂÂÂÂÂÂÂ OEÂ ------------ÂÂ 3.10.0-693.el7.x86_64 #1
[ÂÂ 85.627674] Hardware name: HPE Superdome Flex/Superdome Flex, BIOS IP147.006.000.137.000.1711022001 11/02/2017
[ÂÂ 85.638843] task: ffff887fb8ac6eb0 ti: ffff887ffd30c000 task.ti: ffff887ffd30c000
[ÂÂ 85.647200] RIP: 0010:[<ffffffffc027840e>] [<ffffffffc027840e>] start_check_enables+0x3e/0x80 [ipmi_si]
[ÂÂ 85.657794] RSP: 0018:ffff887ffd30fea8Â EFLAGS: 00010246
[ÂÂ 85.663723] RAX: 0000000000000000 RBX: ffff88016930b000 RCX: 0000000000000006
[ÂÂ 85.671688] RDX: 0000000000000002 RSI: ffff887ffd30feae RDI: 0000000000000000
[ÂÂ 85.679655] RBP: ffff887ffd30fec0 R08: ffff88016930b1a0 R09: 00000001820001fb
[ÂÂ 85.687623] R10: 00000000beb72001 R11: ffffea013efadc80 R12: ffffffffc02822c0
[ÂÂ 85.695589] R13: 0000000000000800 R14: 0000000001f222b0 R15: 0000000001f22010
[ÂÂ 85.703555] FS:Â 00007fb9a5c10740(0000) GS:ffff887fbea40000(0000) knlGS:0000000000000000
[ÂÂ 85.712587] CS:Â 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ÂÂ 85.719002] CR2: 0000000000000010 CR3: 0000007ffd2a5000 CR4: 00000000003407e0
[ÂÂ 85.726968] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ÂÂ 85.734936] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ÂÂ 85.742904] Stack:
[ÂÂ 85.745150]Â 2f18887ffd30fec0 000000007706aaa0 ffff88016930b000 ffff887ffd30fed8
[ÂÂ 85.753456]Â ffffffffc027ae19 ffffffffc0281ff0 ffff887ffd30fef0 ffffffffc027b9cb
[ÂÂ 85.761762]Â fffffffffffffff5 ffff887ffd30ff78 ffffffff810fe55b 0000000000000000
[ÂÂ 85.770063] Call Trace:
[ÂÂ 85.772797]Â [<ffffffffc027ae19>] cleanup_one_si+0x149/0x180 [ipmi_si]
[ÂÂ 85.780090]Â [<ffffffffc027b9cb>] cleanup_ipmi_si+0x6b/0xc0 [ipmi_si]
[ÂÂ 85.787293]Â [<ffffffff810fe55b>] SyS_delete_module+0x19b/0x300
[ÂÂ 85.793913]Â [<ffffffff816b4fc9>] system_call_fastpath+0x16/0x1b
[ÂÂ 85.800621] Code: ee 18 c6 45 ef 2f 65 48 8b 04 25 28 00 00 00 48 89 45 f0 31 c0 40 84 f6 75 33 48 8b 47 18 ba 02 00 00 00 48 8b 7f 10 48 8d 75 ee <ff> 50 10 48 8b 45 f0 65 48 33 04 25 28 00 00 00 c7 43 38 05 00
[ÂÂ 85.822638] RIPÂ [<ffffffffc027840e>] start_check_enables+0x3e/0x80 [ipmi_si]
[ÂÂ 85.830611]Â RSP <ffff887ffd30fea8>
[ÂÂ 85.834504] CR2: 0000000000000010


And with debug printks printing out some contents of the struct at the start of
module exit:

# rmmod ipmi_si; dmesg -c
[ 1083.527776]ÂÂÂÂÂÂÂÂÂÂÂÂ smi_info ffff880fbb38f600
[ 1083.527779]ÂÂÂÂÂÂÂÂÂÂÂÂ intf_num 0
[ 1083.527779]ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ intf ffff880fbb800000
[ 1083.527780]ÂÂÂÂÂÂÂÂÂÂÂÂÂÂ *si_sm ffff880fbb144400
[ 1083.527780]ÂÂÂÂÂÂÂÂÂÂÂ *handlers ffffffffa051bac0
[ 1083.527780]ÂÂÂÂÂÂÂÂÂÂÂÂÂ si_type 2

[ 1083.527782]ÂÂÂÂÂÂÂÂÂÂÂÂ smi_info ffff880fbb38f200
[ 1083.527783]ÂÂÂÂÂÂÂÂÂÂÂÂ intf_num 0
[ 1083.527783]ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ intfÂÂÂÂÂÂÂÂÂÂ (null)
[ 1083.527784]ÂÂÂÂÂÂÂÂÂÂÂÂÂÂ *si_smÂÂÂÂÂÂÂÂÂÂ (null)
[ 1083.527785]ÂÂÂÂÂÂÂÂÂÂÂ *handlersÂÂÂÂÂÂÂÂÂÂ (null)
[ 1083.527786]ÂÂÂÂÂÂÂÂÂÂÂÂÂ si_type 2

@@ -3961,6 +3985,15 @@ static void cleanup_ipmi_si(void)
ÂÂÂÂÂ if (!initialized)
ÂÂÂÂÂÂÂÂÂ return;
 + list_for_each_entry_safe(e, tmp_e, &smi_infos, link) {
+ÂÂÂÂÂÂÂ pr_crit("%20s %p\n", "smi_info",ÂÂÂ e);
+ÂÂÂÂÂÂÂ pr_crit("%20s %d\n", "intf_num",ÂÂÂ e->intf_num);
+ÂÂÂÂÂÂÂ pr_crit("%20s %p\n", "intf",ÂÂÂÂÂÂÂ e->intf);
+ÂÂÂÂÂÂÂ pr_crit("%20s %p\n", "*si_sm",ÂÂÂÂÂÂÂ e->si_sm);
+ÂÂÂÂÂÂÂ pr_crit("%20s %p\n", "*handlers",ÂÂÂ e->handlers);
+ÂÂÂÂÂÂÂ pr_crit("%20s %d\n\n", "si_type",ÂÂÂ e->si_type);
+ÂÂÂ }
+
 #ifdef CONFIG_PCI
ÂÂÂÂÂ if (pci_registered)
ÂÂÂÂÂÂÂÂÂ pci_unregister_driver(&ipmi_pci_driver);
If you are removing the module and this happens, there may be a race
conditions, but this is the wrong fix. More likely that the structure gets
cleaned up and this function is called afterwards.

-corey

Andrew

Avoid panicking when there are uninitialized SMIs by checking for a handler
pointer before starting the check enables transaction.

Signed-off-by: Andrew Banman <abanman@xxxxxxx>
---
ÂÂÂdrivers/char/ipmi/ipmi_si_intf.c | 2 +-
ÂÂÂ1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c
index cb5719e..6c0b1b3 100644
--- a/drivers/char/ipmi/ipmi_si_intf.c
+++ b/drivers/char/ipmi/ipmi_si_intf.c
@@ -442,7 +442,7 @@ static void start_check_enables(struct smi_info
*smi_info, bool start_timer)

ÂÂÂÂÂÂ if (start_timer)
ÂÂÂÂÂÂÂÂÂÂ start_new_msg(smi_info, msg, 2);
-ÂÂÂ else
+ÂÂÂ else if (smi_info->handlers)
smi_info->handlers->start_transaction(smi_info->si_sm, msg, 2);
ÂÂÂÂÂÂ smi_info->si_state = SI_CHECKING_ENABLES;
ÂÂÂ}