From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Chris Cole <chris@sageembedded.com>,
Russell King <rmk+kernel@armlinux.org.uk>,
Sasha Levin <sashal@kernel.org>
Subject: [PATCH AUTOSEL 4.19 51/73] ARM: 8814/1: mm: improve/fix ARM v7_dma_inv_range() unaligned address handling
Date: Wed, 12 Dec 2018 23:28:16 -0500 [thread overview]
Message-ID: <20181213042838.75160-51-sashal@kernel.org> (raw)
In-Reply-To: <20181213042838.75160-1-sashal@kernel.org>
From: Chris Cole <chris@sageembedded.com>
[ Upstream commit a1208f6a822ac29933e772ef1f637c5d67838da9 ]
This patch addresses possible memory corruption when
v7_dma_inv_range(start_address, end_address) address parameters are not
aligned to whole cache lines. This function issues "invalidate" cache
management operations to all cache lines from start_address (inclusive)
to end_address (exclusive). When start_address and/or end_address are
not aligned, the start and/or end cache lines are first issued "clean &
invalidate" operation. The assumption is this is done to ensure that any
dirty data addresses outside the address range (but part of the first or
last cache lines) are cleaned/flushed so that data is not lost, which
could happen if just an invalidate is issued.
The problem is that these first/last partial cache lines are issued
"clean & invalidate" and then "invalidate". This second "invalidate" is
not required and worse can cause "lost" writes to addresses outside the
address range but part of the cache line. If another component writes to
its part of the cache line between the "clean & invalidate" and
"invalidate" operations, the write can get lost. This fix is to remove
the extra "invalidate" operation when unaligned addressed are used.
A kernel module is available that has a stress test to reproduce the
issue and a unit test of the updated v7_dma_inv_range(). It can be
downloaded from
http://ftp.sageembedded.com/outgoing/linux/cache-test-20181107.tgz.
v7_dma_inv_range() is call by dmac_[un]map_area(addr, len, direction)
when the direction is DMA_FROM_DEVICE. One can (I believe) successfully
argue that DMA from a device to main memory should use buffers aligned
to cache line size, because the "clean & invalidate" might overwrite
data that the device just wrote using DMA. But if a driver does use
unaligned buffers, at least this fix will prevent memory corruption
outside the buffer.
Signed-off-by: Chris Cole <chris@sageembedded.com>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
arch/arm/mm/cache-v7.S | 8 +++++---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/arch/arm/mm/cache-v7.S b/arch/arm/mm/cache-v7.S
index 215df435bfb9..2149b47a0c5a 100644
--- a/arch/arm/mm/cache-v7.S
+++ b/arch/arm/mm/cache-v7.S
@@ -360,14 +360,16 @@ v7_dma_inv_range:
ALT_UP(W(nop))
#endif
mcrne p15, 0, r0, c7, c14, 1 @ clean & invalidate D / U line
+ addne r0, r0, r2
tst r1, r3
bic r1, r1, r3
mcrne p15, 0, r1, c7, c14, 1 @ clean & invalidate D / U line
-1:
- mcr p15, 0, r0, c7, c6, 1 @ invalidate D / U line
- add r0, r0, r2
cmp r0, r1
+1:
+ mcrlo p15, 0, r0, c7, c6, 1 @ invalidate D / U line
+ addlo r0, r0, r2
+ cmplo r0, r1
blo 1b
dsb st
ret lr
--
2.19.1
next prev parent reply other threads:[~2018-12-13 4:28 UTC|newest]
Thread overview: 79+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-12-13 4:27 [PATCH AUTOSEL 4.19 01/73] mac80211_hwsim: fix module init error paths for netlink Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 02/73] Input: hyper-v - fix wakeup from suspend-to-idle Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 03/73] i2c: rcar: check bus state before reinitializing Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 04/73] scsi: libiscsi: Fix NULL pointer dereference in iscsi_eh_session_reset Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 05/73] scsi: vmw_pscsi: Rearrange code to avoid multiple calls to free_irq during unload Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 06/73] tools/bpf: fix two test_btf unit test cases Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 07/73] tools/bpf: add addition type tests to test_btf Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 08/73] net: ethernet: ave: Replace NET_IP_ALIGN with AVE_FRAME_HEADROOM Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 09/73] net: phy: sfp: correct store of detected link modes Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 10/73] drm/amd/display: Fix 6x4K displays light-up on Vega20 (v2) Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 11/73] x86/earlyprintk/efi: Fix infinite loop on some screen widths Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 12/73] drm/msm: Fix task dump in gpu recovery Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 13/73] drm/msm/gpu: Fix a couple memory leaks in debugfs Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 14/73] drm/msm: fix handling of cmdstream offset Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 15/73] drm/msm/dsi: configure VCO rate for 10nm PLL driver Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 16/73] drm/msm: Grab a vblank reference when waiting for commit_done Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 17/73] drm/ttm: fix LRU handling in ttm_buffer_object_transfer Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 18/73] drm/amdgpu: wait for IB test on first device open Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 19/73] ARC: io.h: Implement reads{x}()/writes{x}() Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 20/73] net: stmmac: Move debugfs init/exit to ->probe()/->remove() Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 21/73] net: aquantia: fix rx checksum offload bits Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 22/73] bonding: fix 802.3ad state sent to partner when unbinding slave Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 23/73] bpf: Fix verifier log string check for bad alignment Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 24/73] liquidio: read sc->iq_no before release sc Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 25/73] nfs: don't dirty kernel pages read by direct-io Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 26/73] SUNRPC: Fix leak of krb5p encode pages Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 27/73] SUNRPC: Fix a potential race in xprt_connect() Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 28/73] sbus: char: add of_node_put() Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 29/73] drivers/sbus/char: " Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 30/73] drivers/tty: add missing of_node_put() Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 31/73] ide: pmac: add of_node_put() Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 32/73] drm/msm/hdmi: Enable HPD after HDMI IRQ is set up Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 33/73] drm/msm: dpu: Don't set legacy plane->crtc pointer Sasha Levin
2018-12-13 4:27 ` [PATCH AUTOSEL 4.19 34/73] drm/msm: dpu: Fix "WARNING: invalid free of devm_ allocated data" Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 35/73] drm/msm: Fix error return checking Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 36/73] drm/amd/powerplay: issue pre-display settings for display change event Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 37/73] clk: mvebu: Off by one bugs in cp110_of_clk_get() Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 38/73] clk: mmp: Off by one in mmp_clk_add() Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 39/73] Input: synaptics - enable SMBus for HP 15-ay000 Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 40/73] Input: omap-keypad - fix keyboard debounce configuration Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 41/73] libata: whitelist all SAMSUNG MZ7KM* solid-state disks Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 42/73] net: phy: don't allow __set_phy_supported to add unsupported modes Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 43/73] net: 8139cp: fix a BUG triggered by changing mtu with network traffic Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 44/73] net: phy: Fix not to call phy_resume() if PHY is not attached Sasha Levin
2018-12-13 4:48 ` Yoshihiro Shimoda
2018-12-19 13:43 ` Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 45/73] net: phy: micrel: add toggling phy reset " Sasha Levin
2018-12-13 4:44 ` Yoshihiro Shimoda
2018-12-19 13:43 ` Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 46/73] macvlan: return correct error value Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 47/73] mv88e6060: disable hardware level MAC learning Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 48/73] net/mlx4_en: Change min MTU size to ETH_MIN_MTU Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 49/73] net/mlx4_en: Fix build break when CONFIG_INET is off Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 50/73] bpf: check pending signals while verifying programs Sasha Levin
2018-12-13 4:28 ` Sasha Levin [this message]
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 52/73] ARM: 8815/1: V7M: align v7m_dma_inv_range() with v7 counterpart Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 53/73] ARM: 8816/1: dma-mapping: fix potential uninitialized return Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 54/73] ethernet: fman: fix wrong of_node_put() in probe function Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 55/73] thermal: armada: fix legacy validity test sense Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 56/73] net: mvpp2: fix detection of 10G SFP modules Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 57/73] net: mvpp2: fix phylink handling of invalid PHY modes Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 58/73] x86/build: Fix compiler support check for CONFIG_RETPOLINE Sasha Levin
2018-12-18 13:26 ` Gi-Oh Kim
2018-12-18 15:50 ` Greg KH
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 59/73] drm/amdgpu/vcn: Update vcn.cur_state during suspend Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 60/73] tools/testing/nvdimm: Align test resources to 128M Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 61/73] acpi/nfit: Fix user-initiated ARS to be "ARS-long" rather than "ARS-short" Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 62/73] drm/ast: Fix connector leak during driver unload Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 63/73] cifs: In Kconfig CONFIG_CIFS_POSIX needs depends on legacy (insecure cifs) Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 64/73] vhost/vsock: fix reset orphans race with close timeout Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 65/73] mlxsw: spectrum_switchdev: Fix VLAN device deletion via ioctl Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 66/73] i2c: axxia: properly handle master timeout Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 67/73] i2c: scmi: Fix probe error on devices with an empty SMB0001 ACPI device node Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 68/73] i2c: uniphier: fix violation of tLOW requirement for Fast-mode Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 69/73] i2c: uniphier-f: " Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 70/73] nvme: validate controller state before rescheduling keep alive Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 71/73] nvmet-rdma: fix response use after free Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 72/73] Revert "net/ibm/emac: wrong bit is used for STA control" Sasha Levin
2018-12-13 4:28 ` [PATCH AUTOSEL 4.19 73/73] net/mlx4_core: Correctly set PFC param if global pause is turned off Sasha Levin
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=20181213042838.75160-51-sashal@kernel.org \
--to=sashal@kernel.org \
--cc=chris@sageembedded.com \
--cc=linux-kernel@vger.kernel.org \
--cc=rmk+kernel@armlinux.org.uk \
--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;
as well as URLs for NNTP newsgroup(s).