All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: "Greg Kroah-Hartman" <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, "Vince Weaver" <vincent.weaver@maine.edu>,
	"Robert Święcki" <robert@swiecki.net>,
	"Thomas Gleixner" <tglx@linutronix.de>,
	"Ingo Molnar" <mingo@kernel.org>,
	"Andy Lutomirski" <luto@amacapital.net>,
	"Alexander Shishkin" <alexander.shishkin@linux.intel.com>,
	"Linus Torvalds" <torvalds@linux-foundation.org>,
	"Vince Weaver" <vince@deater.net>,
	"Jiri Kosina" <jkosina@suze.cz>,
	"Hugh Dickins" <hughd@google.com>
Subject: [PATCH 4.4 12/67] kaiser: fix intel_bts perf crashes
Date: Fri,  2 Feb 2018 17:57:41 +0100	[thread overview]
Message-ID: <20180202140816.605663601@linuxfoundation.org> (raw)
In-Reply-To: <20180202140815.091718203@linuxfoundation.org>

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

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

From: Hugh Dickins <hughd@google.com>

Vince reported perf_fuzzer quickly locks up on 4.15-rc7 with PTI;
Robert reported Bad RIP with KPTI and Intel BTS also on 4.15-rc7:
honggfuzz -f /tmp/somedirectorywithatleastonefile \
          --linux_perf_bts_edge -s -- /bin/true
(honggfuzz from https://github.com/google/honggfuzz) crashed with
BUG: unable to handle kernel paging request at ffff9d3215100000
(then narrowed it down to
perf record --per-thread -e intel_bts//u -- /bin/ls).

The intel_bts driver does not use the 'normal' BTS buffer which is
exposed through kaiser_add_mapping(), but instead uses the memory
allocated for the perf AUX buffer.

This obviously comes apart when using PTI, because then the kernel
mapping, which includes that AUX buffer memory, disappears while
switched to user page tables.

Easily fixed in old-Kaiser backports, by applying kaiser_add_mapping()
to those pages; perhaps not so easy for upstream, where 4.15-rc8 commit
99a9dc98ba52 ("x86,perf: Disable intel_bts when PTI") disables for now.

Slightly reorganized surrounding code in bts_buffer_setup_aux(),
so it can better match bts_buffer_free_aux(): free_aux with an #ifdef
to avoid the loop when PTI is off, but setup_aux needs to loop anyway
(and kaiser_add_mapping() is cheap when PTI config is off or "pti=off").

Reported-by: Vince Weaver <vincent.weaver@maine.edu>
Reported-by: Robert Święcki <robert@swiecki.net>
Analyzed-by: Peter Zijlstra <peterz@infradead.org>
Analyzed-by: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Vince Weaver <vince@deater.net>
Cc: stable@vger.kernel.org
Cc: Jiri Kosina <jkosina@suze.cz>
Signed-off-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 arch/x86/kernel/cpu/perf_event_intel_bts.c |   44 +++++++++++++++++++++--------
 1 file changed, 33 insertions(+), 11 deletions(-)

--- a/arch/x86/kernel/cpu/perf_event_intel_bts.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_bts.c
@@ -22,6 +22,7 @@
 #include <linux/debugfs.h>
 #include <linux/device.h>
 #include <linux/coredump.h>
+#include <linux/kaiser.h>
 
 #include <asm-generic/sizes.h>
 #include <asm/perf_event.h>
@@ -67,6 +68,23 @@ static size_t buf_size(struct page *page
 	return 1 << (PAGE_SHIFT + page_private(page));
 }
 
+static void bts_buffer_free_aux(void *data)
+{
+#ifdef CONFIG_PAGE_TABLE_ISOLATION
+	struct bts_buffer *buf = data;
+	int nbuf;
+
+	for (nbuf = 0; nbuf < buf->nr_bufs; nbuf++) {
+		struct page *page = buf->buf[nbuf].page;
+		void *kaddr = page_address(page);
+		size_t page_size = buf_size(page);
+
+		kaiser_remove_mapping((unsigned long)kaddr, page_size);
+	}
+#endif
+	kfree(data);
+}
+
 static void *
 bts_buffer_setup_aux(int cpu, void **pages, int nr_pages, bool overwrite)
 {
@@ -103,29 +121,33 @@ bts_buffer_setup_aux(int cpu, void **pag
 	buf->real_size = size - size % BTS_RECORD_SIZE;
 
 	for (pg = 0, nbuf = 0, offset = 0, pad = 0; nbuf < buf->nr_bufs; nbuf++) {
-		unsigned int __nr_pages;
+		void *kaddr = pages[pg];
+		size_t page_size;
+
+		page = virt_to_page(kaddr);
+		page_size = buf_size(page);
+
+		if (kaiser_add_mapping((unsigned long)kaddr,
+					page_size, __PAGE_KERNEL) < 0) {
+			buf->nr_bufs = nbuf;
+			bts_buffer_free_aux(buf);
+			return NULL;
+		}
 
-		page = virt_to_page(pages[pg]);
-		__nr_pages = PagePrivate(page) ? 1 << page_private(page) : 1;
 		buf->buf[nbuf].page = page;
 		buf->buf[nbuf].offset = offset;
 		buf->buf[nbuf].displacement = (pad ? BTS_RECORD_SIZE - pad : 0);
-		buf->buf[nbuf].size = buf_size(page) - buf->buf[nbuf].displacement;
+		buf->buf[nbuf].size = page_size - buf->buf[nbuf].displacement;
 		pad = buf->buf[nbuf].size % BTS_RECORD_SIZE;
 		buf->buf[nbuf].size -= pad;
 
-		pg += __nr_pages;
-		offset += __nr_pages << PAGE_SHIFT;
+		pg += page_size >> PAGE_SHIFT;
+		offset += page_size;
 	}
 
 	return buf;
 }
 
-static void bts_buffer_free_aux(void *data)
-{
-	kfree(data);
-}
-
 static unsigned long bts_buffer_offset(struct bts_buffer *buf, unsigned int idx)
 {
 	return buf->buf[idx].offset + buf->buf[idx].displacement;

  parent reply	other threads:[~2018-02-02 16:57 UTC|newest]

Thread overview: 77+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-02-02 16:57 [PATCH 4.4 00/67] 4.4.115-stable review Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 01/67] loop: fix concurrent lo_open/lo_release Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 02/67] bpf: fix branch pruning logic Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 03/67] x86: bpf_jit: small optimization in emit_bpf_tail_call() Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 04/67] bpf: fix bpf_tail_call() x64 JIT Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 05/67] bpf: introduce BPF_JIT_ALWAYS_ON config Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 06/67] bpf: arsh is not supported in 32 bit alu thus reject it Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 07/67] bpf: avoid false sharing of map refcount with max_entries Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 08/67] bpf: fix divides by zero Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 09/67] bpf: fix 32-bit divide " Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 10/67] bpf: reject stores into ctx via st and xadd Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 11/67] x86/pti: Make unpoison of pgd for trusted boot work for real Greg Kroah-Hartman
2018-02-02 16:57 ` Greg Kroah-Hartman [this message]
2018-02-02 16:57 ` [PATCH 4.4 13/67] ALSA: seq: Make ioctls race-free Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 14/67] crypto: aesni - handle zero length dst buffer Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 15/67] crypto: af_alg - whitelist mask and type Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 16/67] power: reset: zx-reboot: add missing MODULE_DESCRIPTION/AUTHOR/LICENSE Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 17/67] gpio: iop: " Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 18/67] gpio: ath79: add missing MODULE_DESCRIPTION/LICENSE Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 19/67] mtd: nand: denali_pci: add missing MODULE_DESCRIPTION/AUTHOR/LICENSE Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 20/67] igb: Free IRQs when device is hotplugged Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 21/67] KVM: x86: emulator: Return to user-mode on L1 CPL=0 emulation failure Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 22/67] KVM: x86: Dont re-execute instruction when not passing CR2 value Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 23/67] KVM: X86: Fix operand/address-size during instruction decoding Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 24/67] KVM: x86: ioapic: Fix level-triggered EOI and IOAPIC reconfigure race Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 25/67] KVM: x86: ioapic: Clear Remote IRR when entry is switched to edge-triggered Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 26/67] KVM: x86: ioapic: Preserve read-only values in the redirection table Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 27/67] ACPI / bus: Leave modalias empty for devices which are not present Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 28/67] cpufreq: Add Loongson machine dependencies Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 29/67] bcache: check return value of register_shrinker Greg Kroah-Hartman
2018-02-02 16:57 ` [PATCH 4.4 30/67] drm/amdgpu: Fix SDMA load/unload sequence on HWS disabled mode Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 31/67] drm/amdkfd: Fix SDMA ring buffer size calculation Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 32/67] drm/amdkfd: Fix SDMA oversubsription handling Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 33/67] openvswitch: fix the incorrect flow action alloc size Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 34/67] mac80211: fix the update of path metric for RANN frame Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 35/67] btrfs: fix deadlock when writing out space cache Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 36/67] KVM: VMX: Fix rflags cache during vCPU reset Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 37/67] xen-netfront: remove warning when unloading module Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 38/67] nfsd: CLOSE SHOULD return the invalid special stateid for NFSv4.x (x>0) Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 39/67] nfsd: Ensure we check stateid validity in the seqid operation checks Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 40/67] grace: replace BUG_ON by WARN_ONCE in exit_net hook Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 41/67] nfsd: check for use of the closed special stateid Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 42/67] lockd: fix "list_add double add" caused by legacy signal interface Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 43/67] hwmon: (pmbus) Use 64bit math for DIRECT format values Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 44/67] net: ethernet: xilinx: Mark XILINX_LL_TEMAC broken on 64-bit Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 45/67] quota: Check for register_shrinker() failure Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 46/67] SUNRPC: Allow connect to return EHOSTUNREACH Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 47/67] kmemleak: add scheduling point to kmemleak_scan() Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 48/67] drm/omap: Fix error handling path in omap_dmm_probe() Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 49/67] xfs: ubsan fixes Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 50/67] scsi: aacraid: Prevent crash in case of free interrupt during scsi EH path Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 51/67] scsi: ufs: ufshcd: fix potential NULL pointer dereference in ufshcd_config_vreg Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 52/67] media: usbtv: add a new usbid Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 53/67] usb: gadget: dont dereference g until after it has been null checked Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 54/67] staging: rtl8188eu: Fix incorrect response to SIOCGIWESSID Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 55/67] usb: option: Add support for FS040U modem Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 56/67] USB: serial: pl2303: new device id for Chilitag Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 57/67] USB: cdc-acm: Do not log urb submission errors on disconnect Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 58/67] CDC-ACM: apply quirk for card reader Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 59/67] USB: serial: io_edgeport: fix possible sleep-in-atomic Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 60/67] usbip: prevent bind loops on devices attached to vhci_hcd Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 61/67] usbip: list: dont list " Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 62/67] USB: serial: simple: add Motorola Tetra driver Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 63/67] usb: f_fs: Prevent gadget unbind if it is already unbound Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 64/67] usb: uas: unconditionally bring back host after reset Greg Kroah-Hartman
2018-03-03  0:19   ` Ben Hutchings
2018-03-03  9:14     ` Hans de Goede
2018-02-02 16:58 ` [PATCH 4.4 65/67] selinux: general protection fault in sock_has_perm Greg Kroah-Hartman
2018-02-02 16:58   ` Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 66/67] serial: imx: Only wakeup via RTSDEN bit if the system has RTS/CTS Greg Kroah-Hartman
2018-02-02 16:58 ` [PATCH 4.4 67/67] spi: imx: do not access registers while clocks disabled Greg Kroah-Hartman
2018-02-02 18:40 ` [PATCH 4.4 00/67] 4.4.115-stable review Nathan Chancellor
2018-02-03  5:18   ` Greg Kroah-Hartman
2018-02-02 22:20 ` Shuah Khan
2018-02-02 22:33 ` Dan Rue
2018-02-03 15:28 ` Guenter Roeck
2018-02-03 15:44   ` Greg Kroah-Hartman

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=20180202140816.605663601@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=hughd@google.com \
    --cc=jkosina@suze.cz \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luto@amacapital.net \
    --cc=mingo@kernel.org \
    --cc=robert@swiecki.net \
    --cc=stable@vger.kernel.org \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.org \
    --cc=vince@deater.net \
    --cc=vincent.weaver@maine.edu \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.