linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Ben Hutchings <ben@decadent.org.uk>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: akpm@linux-foundation.org, Denis Kirjanov <kda@linux-powerpc.org>,
	"Dmitry Safonov" <dsafonov@virtuozzo.com>,
	"Oleg Nesterov" <oleg@redhat.com>,
	"Srikar Dronamraju" <srikar@linux.vnet.ibm.com>,
	"Masami Hiramatsu" <mhiramat@kernel.org>,
	"Sebastian Mayr" <me@sam.st>,
	"Thomas Gleixner" <tglx@linutronix.de>
Subject: [PATCH 3.16 50/83] uprobes/x86: Fix detection of 32-bit user mode
Date: Wed, 20 Nov 2019 15:38:00 +0000	[thread overview]
Message-ID: <lsq.1574264230.178025366@decadent.org.uk> (raw)
In-Reply-To: <lsq.1574264230.280218497@decadent.org.uk>

3.16.78-rc1 review patch.  If anyone has any objections, please let me know.

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

From: Sebastian Mayr <me@sam.st>

commit 9212ec7d8357ea630031e89d0d399c761421c83b upstream.

32-bit processes running on a 64-bit kernel are not always detected
correctly, causing the process to crash when uretprobes are installed.

The reason for the crash is that in_ia32_syscall() is used to determine the
process's mode, which only works correctly when called from a syscall.

In the case of uretprobes, however, the function is called from a exception
and always returns 'false' on a 64-bit kernel. In consequence this leads to
corruption of the process's return address.

Fix this by using user_64bit_mode() instead of in_ia32_syscall(), which
is correct in any situation.

[ tglx: Add a comment and the following historical info ]

This should have been detected by the rename which happened in commit

  abfb9498ee13 ("x86/entry: Rename is_{ia32,x32}_task() to in_{ia32,x32}_syscall()")

which states in the changelog:

    The is_ia32_task()/is_x32_task() function names are a big misnomer: they
    suggests that the compat-ness of a system call is a task property, which
    is not true, the compatness of a system call purely depends on how it
    was invoked through the system call layer.
    .....

and then it went and blindly renamed every call site.

Sadly enough this was already mentioned here:

   8faaed1b9f50 ("uprobes/x86: Introduce sizeof_long(), cleanup adjust_ret_addr() and
arch_uretprobe_hijack_return_addr()")

where the changelog says:

    TODO: is_ia32_task() is not what we actually want, TS_COMPAT does
    not necessarily mean 32bit. Fortunately syscall-like insns can't be
    probed so it actually works, but it would be better to rename and
    use is_ia32_frame().

and goes all the way back to:

    0326f5a94dde ("uprobes/core: Handle breakpoint and singlestep exceptions")

Oh well. 7+ years until someone actually tried a uretprobe on a 32bit
process on a 64bit kernel....

Fixes: 0326f5a94dde ("uprobes/core: Handle breakpoint and singlestep exceptions")
Signed-off-by: Sebastian Mayr <me@sam.st>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Dmitry Safonov <dsafonov@virtuozzo.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Link: https://lkml.kernel.org/r/20190728152617.7308-1-me@sam.st
[bwh: Backported to 3.16: adjust context]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
---
 arch/x86/kernel/uprobes.c | 17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)

--- a/arch/x86/kernel/uprobes.c
+++ b/arch/x86/kernel/uprobes.c
@@ -455,9 +455,12 @@ struct uprobe_xol_ops {
 	void	(*abort)(struct arch_uprobe *, struct pt_regs *);
 };
 
-static inline int sizeof_long(void)
+static inline int sizeof_long(struct pt_regs *regs)
 {
-	return is_ia32_task() ? 4 : 8;
+	/*
+	 * Check registers for mode as in_xxx_syscall() does not apply here.
+	 */
+	return user_64bit_mode(regs) ? 8 : 4;
 }
 
 static int default_pre_xol_op(struct arch_uprobe *auprobe, struct pt_regs *regs)
@@ -468,9 +471,9 @@ static int default_pre_xol_op(struct arc
 
 static int push_ret_address(struct pt_regs *regs, unsigned long ip)
 {
-	unsigned long new_sp = regs->sp - sizeof_long();
+	unsigned long new_sp = regs->sp - sizeof_long(regs);
 
-	if (copy_to_user((void __user *)new_sp, &ip, sizeof_long()))
+	if (copy_to_user((void __user *)new_sp, &ip, sizeof_long(regs)))
 		return -EFAULT;
 
 	regs->sp = new_sp;
@@ -503,7 +506,7 @@ static int default_post_xol_op(struct ar
 		long correction = utask->vaddr - utask->xol_vaddr;
 		regs->ip += correction;
 	} else if (auprobe->defparam.fixups & UPROBE_FIX_CALL) {
-		regs->sp += sizeof_long(); /* Pop incorrect return address */
+		regs->sp += sizeof_long(regs); /* Pop incorrect return address */
 		if (push_ret_address(regs, utask->vaddr + auprobe->defparam.ilen))
 			return -ERESTART;
 	}
@@ -612,7 +615,7 @@ static int branch_post_xol_op(struct arc
 	 * "call" insn was executed out-of-line. Just restore ->sp and restart.
 	 * We could also restore ->ip and try to call branch_emulate_op() again.
 	 */
-	regs->sp += sizeof_long();
+	regs->sp += sizeof_long(regs);
 	return -ERESTART;
 }
 
@@ -903,7 +906,7 @@ bool arch_uprobe_skip_sstep(struct arch_
 unsigned long
 arch_uretprobe_hijack_return_addr(unsigned long trampoline_vaddr, struct pt_regs *regs)
 {
-	int rasize = sizeof_long(), nleft;
+	int rasize = sizeof_long(regs), nleft;
 	unsigned long orig_ret_vaddr = 0; /* clear high bits for 32-bit apps */
 
 	if (copy_from_user(&orig_ret_vaddr, (void __user *)regs->sp, rasize))


  parent reply	other threads:[~2019-11-20 15:43 UTC|newest]

Thread overview: 88+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-11-20 15:37 [PATCH 3.16 00/83] 3.16.78-rc1 review Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 01/83] hwmon: (nct6775) Fix register address and added missed tolerance for nct6106 Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 02/83] x86/sysfb_efi: Add quirks for some devices with swapped width and height Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 03/83] mmc: mmc_spi: Enable stable writes Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 04/83] ALSA: compress: Fix regression on compressed capture streams Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 05/83] can: peak_usb: fix potential double kfree_skb() Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 06/83] usb: pci-quirks: Correct AMD PLL quirk detection Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 07/83] usb: wusbcore: fix unbalanced get/put cluster_id Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 08/83] x86/speculation/mds: Apply more accurate check on hypervisor platform Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 09/83] hpet: Fix division by zero in hpet_time_div() Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 10/83] sched/fair: Don't free p->numa_faults with concurrent readers Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 11/83] tty/ldsem, locking/rwsem: Add missing ACQUIRE to read_failed sleep loop Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 12/83] bnx2x: Disable multi-cos feature Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 13/83] arm64: compat: Allow single-byte watchpoints on all addresses Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 14/83] net: sched: Fix a possible null-pointer dereference in dequeue_func() Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 15/83] net: fix ifindex collision during namespace removal Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 16/83] libata: zpodd: Fix small read overflow in zpodd_get_mech_type() Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 17/83] selinux: fix memory leak in policydb_init() Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 18/83] net: bridge: mcast: don't delete permanent entries when fast leave is enabled Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 19/83] xen/swiotlb: fix condition for calling xen_destroy_contiguous_region() Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 20/83] s390/dasd: fix endless loop after read unit address configuration Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 21/83] can: peak_usb: pcan_usb_pro: Fix info-leaks to USB devices Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 22/83] asm-generic: fix -Wtype-limits compiler warnings Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 23/83] NFSv4: Fix a potential sleep while atomic in nfs4_do_reclaim() Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 24/83] USB: serial: option: Add support for ZTE MF871A Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 25/83] usb: yurex: Fix use-after-free in yurex_delete Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 26/83] SMB3: Fix deadlock in validate negotiate hits reconnect Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 27/83] smb3: send CAP_DFS capability during session setup Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 28/83] sound: fix a memory leak bug Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 29/83] ALSA: firewire: " Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 30/83] ALSA: hda - Fix " Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 31/83] sh: kernel: hw_breakpoint: Fix missing break in switch statement Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 32/83] net: tc35815: Explicitly check NET_IP_ALIGN is not zero in tc35815_rx Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 33/83] staging: comedi: dt3000: Fix signed integer overflow 'divider * base' Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 34/83] staging: comedi: dt3000: Fix rounding up of timer divisor Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 35/83] USB: core: Fix races in character device registration and deregistraion Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 36/83] netfilter: conntrack: Use consistent ct id hash calculation Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 37/83] sctp: fix the transport error_count check Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 38/83] USB: serial: option: Add Motorola modem UARTs Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 39/83] usb: cdc-acm: make sure a refcount is taken early enough Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 40/83] net/packet: fix race in tpacket_snd() Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 41/83] Revert "cfg80211: fix processing world regdomain when non modular" Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 42/83] usb-storage: Add new JMS567 revision to unusual_devs Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 43/83] dm btree: fix order of block initialization in btree_split_beneath Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 44/83] dm space map metadata: fix missing store of apply_bops() return value Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 45/83] dm table: fix invalid memory accesses with too high sector number Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 46/83] x86/retpoline: Don't clobber RFLAGS during CALL_NOSPEC on i386 Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 47/83] batman-adv: Only read OGM tvlv_len after buffer len check Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 48/83] ALSA: seq: Fix potential concurrent access to the deleted pool Ben Hutchings
2019-11-20 15:37 ` [PATCH 3.16 49/83] ptrace,x86: Make user_64bit_mode() available to 32-bit builds Ben Hutchings
2019-11-20 15:38 ` Ben Hutchings [this message]
2019-11-20 15:38 ` [PATCH 3.16 51/83] x86/apic: Do not initialize LDR and DFR for bigsmp Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 52/83] x86/apic: Drop logical_smp_processor_id() inline Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 53/83] x86/apic/32: Avoid bogus LDR warnings Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 54/83] usb: host: ohci: fix a race condition between shutdown and irq Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 55/83] USB: storage: ums-realtek: Update module parameter description for auto_delink_en Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 56/83] USB: storage: ums-realtek: Whitelist auto-delink support Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 57/83] USB: cdc-wdm: fix race between write and disconnect due to flag abuse Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 58/83] VMCI: Release resource if the work is already queued Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 59/83] mld: fix memory leak in mld_del_delrec() Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 60/83] ALSA: hda - Fix potential endless loop at applying quirks Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 61/83] mmc: core: Fix init of SD cards reporting an invalid VDD range Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 62/83] net: seeq: Fix the function used to release some memory in an error handling path Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 63/83] sched/fair: Don't assign runtime for throttled cfs_rq Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 64/83] vhost/test: fix build for vhost test Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 65/83] sctp: use transport pf_retrans in sctp_do_8_2_transport_strike Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 66/83] genirq: Prevent NULL pointer dereference in resend_irqs() Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 67/83] keys: Fix missing null pointer check in request_key_auth_describe() Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 68/83] sch_hhf: ensure quantum and hhf_non_hh_weight are non-zero Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 69/83] tcp: fix tcp_ecn_withdraw_cwr() to clear TCP_ECN_QUEUE_CWR Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 70/83] tun: fix use-after-free when register netdev failed Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 71/83] ipv6: Fix the link time qualifier of 'ping_v6_proc_exit_net()' Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 72/83] net: Fix null de-reference of device refcount Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 73/83] sctp: Fix the link time qualifier of 'sctp_ctrlsock_exit()' Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 74/83] KVM: nVMX: handle page fault in vmread Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 75/83] KVM: x86: work around leak of uninitialized stack contents Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 76/83] PCI: tegra: Enable Relaxed Ordering only for Tegra20 & Tegra30 Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 77/83] alarmtimer: Use EOPNOTSUPP instead of ENOTSUPP Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 78/83] cifs: use cifsInodeInfo->open_file_lock while iterating to avoid a panic Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 79/83] CIFS: Fix use after free of file info structures Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 80/83] md/raid: raid5 preserve the writeback action after the parity check Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 81/83] btrfs: partially apply b8b93addde Ben Hutchings
2019-11-20 15:56   ` Hans van Kranenburg
2019-11-21 21:42     ` Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 82/83] btrfs: volumes: Cleanup stripe size calculation Ben Hutchings
2019-11-20 15:38 ` [PATCH 3.16 83/83] btrfs: alloc_chunk: fix more DUP stripe size handling Ben Hutchings
2019-11-20 15:46 ` [PATCH 3.16 00/83] 3.16.78-rc1 review Guenter Roeck
2019-11-20 15:50   ` Ben Hutchings

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=lsq.1574264230.178025366@decadent.org.uk \
    --to=ben@decadent.org.uk \
    --cc=akpm@linux-foundation.org \
    --cc=dsafonov@virtuozzo.com \
    --cc=kda@linux-powerpc.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=me@sam.st \
    --cc=mhiramat@kernel.org \
    --cc=oleg@redhat.com \
    --cc=srikar@linux.vnet.ibm.com \
    --cc=stable@vger.kernel.org \
    --cc=tglx@linutronix.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).