From: Ben Hutchings <ben@decadent.org.uk>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: akpm@linux-foundation.org, "Helge Deller" <deller@gmx.de>,
"Mathieu Desnoyers" <mathieu.desnoyers@efficios.com>
Subject: [PATCH 3.2 33/70] parisc: Fix syscall restarts
Date: Mon, 18 Jan 2016 03:18:35 +0000 [thread overview]
Message-ID: <lsq.1453087115.853730823@decadent.org.uk> (raw)
In-Reply-To: <lsq.1453087114.713093519@decadent.org.uk>
3.2.76-rc1 review patch. If anyone has any objections, please let me know.
------------------
From: Helge Deller <deller@gmx.de>
commit 71a71fb5374a23be36a91981b5614590b9e722c3 upstream.
On parisc syscalls which are interrupted by signals sometimes failed to
restart and instead returned -ENOSYS which in the worst case lead to
userspace crashes.
A similiar problem existed on MIPS and was fixed by commit e967ef02
("MIPS: Fix restart of indirect syscalls").
On parisc the current syscall restart code assumes that all syscall
callers load the syscall number in the delay slot of the ble
instruction. That's how it is e.g. done in the unistd.h header file:
ble 0x100(%sr2, %r0)
ldi #syscall_nr, %r20
Because of that assumption the current code never restored %r20 before
returning to userspace.
This assumption is at least not true for code which uses the glibc
syscall() function, which instead uses this syntax:
ble 0x100(%sr2, %r0)
copy regX, %r20
where regX depend on how the compiler optimizes the code and register
usage.
This patch fixes this problem by adding code to analyze how the syscall
number is loaded in the delay branch and - if needed - copy the syscall
number to regX prior returning to userspace for the syscall restart.
Signed-off-by: Helge Deller <deller@gmx.de>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
[bwh: Backported to 3.2: adjust context]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
---
arch/parisc/kernel/signal.c | 64 ++++++++++++++++++++++++++++++++++++---------
1 file changed, 52 insertions(+), 12 deletions(-)
--- a/arch/parisc/kernel/signal.c
+++ b/arch/parisc/kernel/signal.c
@@ -468,6 +468,55 @@ handle_signal(unsigned long sig, siginfo
return 1;
}
+/*
+ * Check how the syscall number gets loaded into %r20 within
+ * the delay branch in userspace and adjust as needed.
+ */
+
+static void check_syscallno_in_delay_branch(struct pt_regs *regs)
+{
+ u32 opcode, source_reg;
+ u32 __user *uaddr;
+ int err;
+
+ /* Usually we don't have to restore %r20 (the system call number)
+ * because it gets loaded in the delay slot of the branch external
+ * instruction via the ldi instruction.
+ * In some cases a register-to-register copy instruction might have
+ * been used instead, in which case we need to copy the syscall
+ * number into the source register before returning to userspace.
+ */
+
+ /* A syscall is just a branch, so all we have to do is fiddle the
+ * return pointer so that the ble instruction gets executed again.
+ */
+ regs->gr[31] -= 8; /* delayed branching */
+
+ /* Get assembler opcode of code in delay branch */
+ uaddr = (unsigned int *) ((regs->gr[31] & ~3) + 4);
+ err = get_user(opcode, uaddr);
+ if (err)
+ return;
+
+ /* Check if delay branch uses "ldi int,%r20" */
+ if ((opcode & 0xffff0000) == 0x34140000)
+ return; /* everything ok, just return */
+
+ /* Check if delay branch uses "nop" */
+ if (opcode == INSN_NOP)
+ return;
+
+ /* Check if delay branch uses "copy %rX,%r20" */
+ if ((opcode & 0xffe0ffff) == 0x08000254) {
+ source_reg = (opcode >> 16) & 31;
+ regs->gr[source_reg] = regs->gr[20];
+ return;
+ }
+
+ pr_warn("syscall restart: %s (pid %d): unexpected opcode 0x%08x\n",
+ current->comm, task_pid_nr(current), opcode);
+}
+
static inline void
syscall_restart(struct pt_regs *regs, struct k_sigaction *ka)
{
@@ -489,10 +538,7 @@ syscall_restart(struct pt_regs *regs, st
}
/* fallthrough */
case -ERESTARTNOINTR:
- /* A syscall is just a branch, so all
- * we have to do is fiddle the return pointer.
- */
- regs->gr[31] -= 8; /* delayed branching */
+ check_syscallno_in_delay_branch(regs);
/* Preserve original r28. */
regs->gr[28] = regs->orig_r28;
break;
@@ -543,18 +589,12 @@ insert_restart_trampoline(struct pt_regs
}
case -ERESTARTNOHAND:
case -ERESTARTSYS:
- case -ERESTARTNOINTR: {
- /* Hooray for delayed branching. We don't
- * have to restore %r20 (the system call
- * number) because it gets loaded in the delay
- * slot of the branch external instruction.
- */
- regs->gr[31] -= 8;
+ case -ERESTARTNOINTR:
+ check_syscallno_in_delay_branch(regs);
/* Preserve original r28. */
regs->gr[28] = regs->orig_r28;
return;
- }
default:
break;
}
next prev parent reply other threads:[~2016-01-18 3:28 UTC|newest]
Thread overview: 77+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-01-18 3:18 [PATCH 3.2 00/70] 3.2.76-rc1 review Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 12/70] sh_eth: fix TX buffer byte-swapping Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 08/70] tty: Fix GPF in flush_to_ldisc() Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 43/70] KVM: x86: Reload pit counters for all channels when restoring state Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 05/70] rfkill: copy the name into the rfkill struct Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 54/70] ahci: Order SATA device IDs for codename Lewisburg Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 40/70] genirq: Prevent chip buslock deadlock Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 16/70] scripts: recordmcount: break hardlinks Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 31/70] USB: fix invalid memory access in hub_activate() Ben Hutchings
2016-01-18 3:18 ` Ben Hutchings [this message]
2016-01-18 3:18 ` [PATCH 3.2 69/70] cdrom: Random writing support for BD-RE media Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 26/70] xen/pciback: Return error on XEN_PCI_OP_enable_msix when device has MSI or MSI-X enabled Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 02/70] ipv6: sctp: fix lockdep splat in sctp_v6_get_dst() Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 35/70] ocfs2: fix BUG when calculate new backup super Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 42/70] net: possible use after free in dst_release Ben Hutchings
2016-01-18 3:49 ` Francesco Ruggeri
2016-01-18 11:50 ` Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 61/70] i2c: i801: Add device ID for Intel Wildcat Point PCH Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 29/70] xen/pciback: Don't allow MSI-X ops if PCI_COMMAND_MEMORY is not set Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 64/70] i2c: i801: Add DeviceIDs for SunrisePoint LP Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 17/70] net: fix warnings in 'make htmldocs' by moving macro definition out of field declaration Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 09/70] ALSA: tlv: compute TLV_*_ITEM lengths automatically Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 22/70] xen-netback: use RING_COPY_REQUEST() throughout Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 50/70] ahci: Remove Device ID for Intel Sunrise Point PCH Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 47/70] vmstat: allocate vmstat_wq before it is used Ben Hutchings
2016-01-18 22:26 ` Luis Henriques
2016-01-18 3:18 ` [PATCH 3.2 34/70] ipv6/addrlabel: fix ip6addrlbl_get() Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 23/70] xen-blkback: only read request operation from shared ring once Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 04/70] crypto: skcipher - Copy iv from desc even for 0-len walks Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 30/70] USB: ipaq.c: fix a timeout loop Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 70/70] HID: dragonrise: fix HID Descriptor for 0x0006 PID Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 32/70] KEYS: Fix race between read and revoke Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 51/70] ahci: Add Marvell 88se91a2 device id Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 63/70] i2c: i801: Add Device IDs for Intel Sunrise Point PCH Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 28/70] xen/pciback: For XEN_PCI_OP_disable_msi[|x] only disable if device has MSI(X) enabled Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 19/70] s390/dis: Fix handling of format specifiers Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 24/70] xen/pciback: Save xen_pci_op commands before processing it Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 52/70] ahci: add new Intel device IDs Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 37/70] MIPS: Fix restart of indirect syscalls Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 55/70] i2c: i801: SMBus patch for Intel Avoton DeviceIDs Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 06/70] ses: Fix problems with simple enclosures Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 58/70] i2c: i801: Add Device IDs for Intel Wildcat Point-LP PCH Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 59/70] i2c: i801: enable Intel BayTrail SMBUS Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 60/70] i2c: i801: Fix the alignment of the device table Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 10/70] ALSA: tlv: add DECLARE_TLV_DB_RANGE() Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 03/70] video: fbdev: fsl: Fix kernel crash when diu_ops is not implemented Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 53/70] ahci: Add Device ID for Intel Sunrise Point PCH Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 11/70] ALSA: usb-audio: Add a more accurate volume quirk for AudioQuest DragonFly Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 01/70] sctp: start t5 timer only when peer rwnd is 0 and local state is SHUTDOWN_PENDING Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 56/70] i2c: i801: Add Device IDs for Intel Wellsburg PCH Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 20/70] xen: Add RING_COPY_REQUEST() Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 62/70] i2c: i801: Add PCI ID for Intel Braswell Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 21/70] xen-netback: don't use last request to determine minimum Tx credit Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 41/70] ftrace/scripts: Fix incorrect use of sprintf in recordmcount Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 18/70] ftrace/scripts: Have recordmcount copy the object file Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 57/70] i2c: i801: SMBus patch for Intel Coleto Creek DeviceIDs Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 13/70] mISDN: fix a loop count Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 67/70] i2c: i801: Document Intel DNV and Broxton Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 25/70] xen/pciback: Return error on XEN_PCI_OP_enable_msi when device has MSI or MSI-X enabled Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 65/70] i2c: i801: Add support for Intel DNV Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 45/70] Revert "net: add length argument to skb_copy_and_csum_datagram_iovec" Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 68/70] i2c: i801: add Intel Lewisburg device IDs Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 07/70] ses: fix additional element traversal bug Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 49/70] ahci: Add JMicron 362 device IDs Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 27/70] xen/pciback: Do not install an IRQ handler for MSI interrupts Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 48/70] ahci: Add Device IDs for Intel Wellsburg PCH Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 15/70] spi: fix parent-device reference leak Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 14/70] ser_gigaset: fix deallocation of platform device structure Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 46/70] udp: properly support MSG_PEEK with truncated buffers Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 38/70] drm/radeon: fix hotplug race at startup Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 36/70] mm/memory_hotplug.c: check for missing sections in test_pages_in_a_zone() Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 66/70] i2c: i801: Add support for Intel Broxton Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 39/70] net/core: revert "net: fix __netdev_update_features return.." and add comment Ben Hutchings
2016-01-18 3:18 ` [PATCH 3.2 44/70] kvm: x86: only channel 0 of the i8254 is linked to the HPET Ben Hutchings
2016-01-18 3:45 ` [PATCH 3.2 00/70] 3.2.76-rc1 review Ben Hutchings
2016-01-18 9:12 ` Guenter Roeck
2016-01-18 11:50 ` 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.1453087115.853730823@decadent.org.uk \
--to=ben@decadent.org.uk \
--cc=akpm@linux-foundation.org \
--cc=deller@gmx.de \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=stable@vger.kernel.org \
/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