From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Jim Quinlan <jim2101024@gmail.com>,
linux-mips@linux-mips.org, cernekee@gmail.com,
Ralf Baechle <ralf@linux-mips.org>,
John Ulvr <julvr@broadcom.com>
Subject: [PATCH 3.4 01/31] MIPS: DMA: For BMIPS5000 cores flush region just like non-coherent R10000
Date: Wed, 18 Dec 2013 13:08:09 -0800 [thread overview]
Message-ID: <20131218210522.833580254@linuxfoundation.org> (raw)
In-Reply-To: <20131218210522.790152625@linuxfoundation.org>
3.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jim Quinlan <jim2101024@gmail.com>
commit f86f55d3ad21b21b736bdeb29bee0f0937b77138 upstream.
The BMIPS5000 (Zephyr) processor utilizes instruction speculation. A
stale misprediction address in either the JTB or the CRS may trigger
a prefetch inside a region that is currently being used by a DMA engine,
which is not IO-coherent. This prefetch will fetch a line into the
scache, and that line will soon become stale (ie wrong) during/after the
DMA. Mayhem ensues.
In dma-default.c, the r10000 is handled as a special case in the same way
that we want to handle Zephyr. So we generalize the exception cases into
a function, and include Zephyr as one of the processors that needs this
special care.
Signed-off-by: Jim Quinlan <jim2101024@gmail.com>
Cc: linux-mips@linux-mips.org
Cc: cernekee@gmail.com
Patchwork: https://patchwork.linux-mips.org/patch/5776/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Cc: John Ulvr <julvr@broadcom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
arch/mips/mm/dma-default.c | 16 ++++++++++------
1 file changed, 10 insertions(+), 6 deletions(-)
--- a/arch/mips/mm/dma-default.c
+++ b/arch/mips/mm/dma-default.c
@@ -30,16 +30,20 @@ static inline struct page *dma_addr_to_p
}
/*
+ * The affected CPUs below in 'cpu_needs_post_dma_flush()' can
+ * speculatively fill random cachelines with stale data at any time,
+ * requiring an extra flush post-DMA.
+ *
* Warning on the terminology - Linux calls an uncached area coherent;
* MIPS terminology calls memory areas with hardware maintained coherency
* coherent.
*/
-
-static inline int cpu_is_noncoherent_r10000(struct device *dev)
+static inline int cpu_needs_post_dma_flush(struct device *dev)
{
return !plat_device_is_coherent(dev) &&
(current_cpu_type() == CPU_R10000 ||
- current_cpu_type() == CPU_R12000);
+ current_cpu_type() == CPU_R12000 ||
+ current_cpu_type() == CPU_BMIPS5000);
}
static gfp_t massage_gfp_flags(const struct device *dev, gfp_t gfp)
@@ -209,7 +213,7 @@ static inline void __dma_sync(struct pag
static void mips_dma_unmap_page(struct device *dev, dma_addr_t dma_addr,
size_t size, enum dma_data_direction direction, struct dma_attrs *attrs)
{
- if (cpu_is_noncoherent_r10000(dev))
+ if (cpu_needs_post_dma_flush(dev))
__dma_sync(dma_addr_to_page(dev, dma_addr),
dma_addr & ~PAGE_MASK, size, direction);
@@ -260,7 +264,7 @@ static void mips_dma_unmap_sg(struct dev
static void mips_dma_sync_single_for_cpu(struct device *dev,
dma_addr_t dma_handle, size_t size, enum dma_data_direction direction)
{
- if (cpu_is_noncoherent_r10000(dev))
+ if (cpu_needs_post_dma_flush(dev))
__dma_sync(dma_addr_to_page(dev, dma_handle),
dma_handle & ~PAGE_MASK, size, direction);
}
@@ -281,7 +285,7 @@ static void mips_dma_sync_sg_for_cpu(str
/* Make sure that gcc doesn't leave the empty loop body. */
for (i = 0; i < nelems; i++, sg++) {
- if (cpu_is_noncoherent_r10000(dev))
+ if (cpu_needs_post_dma_flush(dev))
__dma_sync(sg_page(sg), sg->offset, sg->length,
direction);
}
next prev parent reply other threads:[~2013-12-18 21:08 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-12-18 21:08 [PATCH 3.4 00/31] 3.4.75-stable review Greg Kroah-Hartman
2013-12-18 21:08 ` Greg Kroah-Hartman [this message]
2013-12-18 21:08 ` [PATCH 3.4 02/31] ALSA: memalloc.h - fix wrong truncation of dma_addr_t Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 03/31] ARM: pxa: tosa: fix keys mapping Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 04/31] ARM: OMAP3: hwmod data: Dont prevent RESET of USB Host module Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 05/31] ARM: 7912/1: check stack pointer in get_wchan Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 06/31] ARM: 7913/1: fix framepointer check in unwind_frame Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 07/31] KVM: Improve create VCPU parameter (CVE-2013-4587) Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 08/31] hwmon: (w83l786ng) Fix fan speed control mode setting and reporting Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 09/31] xfs: underflow bug in xfs_attrlist_by_handle() Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 10/31] futex: fix handling of read-only-mapped hugepages Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 11/31] usb: hub: Use correct reset for wedged USB3 devices that are NOTATTACHED Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 12/31] usb: dwc3: fix implementation of endpoint wedge Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 13/31] usb: gadget: composite: reset delayed_status on reset_config Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 14/31] USB: serial: option: blacklist interface 1 for Huawei E173s-6 Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 15/31] USB: option: support new huawei devices Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 16/31] Input: usbtouchscreen - separate report and transmit buffer size handling Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 17/31] drivers/rtc/rtc-at91rm9200.c: correct alarm over day/month wrap Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 18/31] selinux: handle TCP SYN-ACK packets correctly in selinux_ip_output() Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 19/31] selinux: handle TCP SYN-ACK packets correctly in selinux_ip_postroute() Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 20/31] mac80211: dont attempt to reorder multicast frames Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 21/31] drm/radeon: fixup bad vram size on SI Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 22/31] x86, efi: Dont use (U)EFI time services on 32 bit Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 23/31] dm bufio: initialize read-only module parameters Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 24/31] dm delay: fix a possible deadlock due to shared workqueue Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 25/31] dm table: fail dm_table_create on dm_round_up overflow Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 27/31] staging: comedi: pcmuio: fix possible NULL deref on detach Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 28/31] staging: comedi: ssv_dnp: use comedi_dio_update_state() Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 29/31] sc1200_wdt: Fix oops Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 30/31] hpfs: fix warnings when the filesystem fills up Greg Kroah-Hartman
2013-12-18 21:08 ` [PATCH 3.4 31/31] Revert "net: update consumers of MSG_MORE to recognize MSG_SENDPAGE_NOTLAST" Greg Kroah-Hartman
2013-12-19 2:02 ` [PATCH 3.4 00/31] 3.4.75-stable review Guenter Roeck
2013-12-19 3:40 ` Greg Kroah-Hartman
2013-12-19 19:54 ` Guenter Roeck
2013-12-19 20:13 ` Greg Kroah-Hartman
2013-12-19 20:46 ` Shuah Khan
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=20131218210522.833580254@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=cernekee@gmail.com \
--cc=jim2101024@gmail.com \
--cc=julvr@broadcom.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mips@linux-mips.org \
--cc=ralf@linux-mips.org \
--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).