From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org,
Thomas Petazzoni <thomas.petazzoni@free-electrons.com>,
Gregory CLEMENT <gregory.clement@free-electrons.com>,
Jason Cooper <jason@lakedaemon.net>
Subject: [PATCH 3.17 33/73] ARM: mvebu: disable I/O coherency on non-SMP situations on Armada 370/375/38x/XP
Date: Tue, 6 Jan 2015 18:15:54 -0800 [thread overview]
Message-ID: <20150107021558.522207116@linuxfoundation.org> (raw)
In-Reply-To: <20150107021557.521276020@linuxfoundation.org>
3.17-stable review patch. If anyone has any objections, please let me know.
------------------
From: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
commit e55355453600a33bb5ca4f71f2d7214875f3b061 upstream.
Enabling the hardware I/O coherency on Armada 370, Armada 375, Armada
38x and Armada XP requires a certain number of conditions:
- On Armada 370, the cache policy must be set to write-allocate.
- On Armada 375, 38x and XP, the cache policy must be set to
write-allocate, the pages must be mapped with the shareable
attribute, and the SMP bit must be set
Currently, on Armada XP, when CONFIG_SMP is enabled, those conditions
are met. However, when Armada XP is used in a !CONFIG_SMP kernel, none
of these conditions are met. With Armada 370, the situation is worse:
since the processor is single core, regardless of whether CONFIG_SMP
or !CONFIG_SMP is used, the cache policy will be set to write-back by
the kernel and not write-allocate.
Since solving this problem turns out to be quite complicated, and we
don't want to let users with a mainline kernel known to have
infrequent but existing data corruptions, this commit proposes to
simply disable hardware I/O coherency in situations where it is known
not to work.
And basically, the is_smp() function of the kernel tells us whether it
is OK to enable hardware I/O coherency or not, so this commit slightly
refactors the coherency_type() function to return
COHERENCY_FABRIC_TYPE_NONE when is_smp() is false, or the appropriate
type of the coherency fabric in the other case.
Thanks to this, the I/O coherency fabric will no longer be used at all
in !CONFIG_SMP configurations. It will continue to be used in
CONFIG_SMP configurations on Armada XP, Armada 375 and Armada 38x
(which are multiple cores processors), but will no longer be used on
Armada 370 (which is a single core processor).
In the process, it simplifies the implementation of the
coherency_type() function, and adds a missing call to of_node_put().
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Fixes: e60304f8cb7bb545e79fe62d9b9762460c254ec2 ("arm: mvebu: Add hardware I/O Coherency support")
Acked-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Link: https://lkml.kernel.org/r/1415871540-20302-3-git-send-email-thomas.petazzoni@free-electrons.com
Signed-off-by: Jason Cooper <jason@lakedaemon.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
arch/arm/mach-mvebu/coherency.c | 46 ++++++++++++++++++++++++++--------------
1 file changed, 31 insertions(+), 15 deletions(-)
--- a/arch/arm/mach-mvebu/coherency.c
+++ b/arch/arm/mach-mvebu/coherency.c
@@ -361,25 +361,41 @@ static int coherency_type(void)
{
struct device_node *np;
const struct of_device_id *match;
+ int type;
+
+ /*
+ * The coherency fabric is needed:
+ * - For coherency between processors on Armada XP, so only
+ * when SMP is enabled.
+ * - For coherency between the processor and I/O devices, but
+ * this coherency requires many pre-requisites (write
+ * allocate cache policy, shareable pages, SMP bit set) that
+ * are only meant in SMP situations.
+ *
+ * Note that this means that on Armada 370, there is currently
+ * no way to use hardware I/O coherency, because even when
+ * CONFIG_SMP is enabled, is_smp() returns false due to the
+ * Armada 370 being a single-core processor. To lift this
+ * limitation, we would have to find a way to make the cache
+ * policy set to write-allocate (on all Armada SoCs), and to
+ * set the shareable attribute in page tables (on all Armada
+ * SoCs except the Armada 370). Unfortunately, such decisions
+ * are taken very early in the kernel boot process, at a point
+ * where we don't know yet on which SoC we are running.
+
+ */
+ if (!is_smp())
+ return COHERENCY_FABRIC_TYPE_NONE;
np = of_find_matching_node_and_match(NULL, of_coherency_table, &match);
- if (np) {
- int type = (int) match->data;
+ if (!np)
+ return COHERENCY_FABRIC_TYPE_NONE;
+
+ type = (int) match->data;
- /* Armada 370/XP coherency works in both UP and SMP */
- if (type == COHERENCY_FABRIC_TYPE_ARMADA_370_XP)
- return type;
-
- /* Armada 375 coherency works only on SMP */
- else if (type == COHERENCY_FABRIC_TYPE_ARMADA_375 && is_smp())
- return type;
-
- /* Armada 380 coherency works only on SMP */
- else if (type == COHERENCY_FABRIC_TYPE_ARMADA_380 && is_smp())
- return type;
- }
+ of_node_put(np);
- return COHERENCY_FABRIC_TYPE_NONE;
+ return type;
}
int coherency_available(void)
next prev parent reply other threads:[~2015-01-07 2:35 UTC|newest]
Thread overview: 75+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-01-07 2:15 [PATCH 3.17 00/73] 3.17.8-stable review Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 01/73] isofs: Fix infinite looping over CE entries Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 02/73] x86/tls: Validate TLS entries to protect espfix Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 03/73] x86/tls: Disallow unusual TLS segments Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 04/73] x86_64, switch_to(): Load TLS descriptors before switching DS and ES Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 05/73] x86, kvm: Clear paravirt_enabled on KVM guests for espfix32s benefit Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 06/73] md/bitmap: always wait for writes on unplug Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 07/73] mfd: twl4030-power: Fix regression with missing compatible flag Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 08/73] mfd: tc6393xb: Fail ohci suspend if full state restore is required Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 09/73] mmc: dw_mmc: avoid write to CDTHRCTL on older versions Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 10/73] mmc: omap_hsmmc: Fix UHS card with DDR50 support Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 11/73] mmc: block: add newline to sysfs display of force_ro Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 12/73] mmc: sdhci-pci-o2micro: Fix Dell E5440 issue Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 13/73] megaraid_sas: corrected return of wait_event from abort frame path Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 14/73] regulator: anatop: Set default voltage selector for vddpu Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 15/73] scsi: correct return values for .eh_abort_handler implementations Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 16/73] f2fs: fix possible data corruption in f2fs_write_begin() Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 17/73] nfs41: fix nfs4_proc_layoutget error handling Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 18/73] dcache: fix kmemcheck warning in switch_names Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 19/73] dm bufio: fix memleak when using a dm_buffers inline bio Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 20/73] dm crypt: use memzero_explicit for on-stack buffer Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 21/73] dm cache: only use overwrite optimisation for promotion when in writeback mode Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 22/73] dm cache: dirty flag was mistakenly being cleared when promoting via overwrite Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 23/73] dm cache: fix spurious cell_defer when dealing with partial block at end of device Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 24/73] dm space map metadata: fix sm_bootstrap_get_nr_blocks() Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 25/73] dm thin: fix inability to discard blocks when in out-of-data-space mode Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 26/73] dm thin: fix missing out-of-data-space to write mode transition if blocks are released Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 27/73] dm thin: fix a race in thin_dtr Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 28/73] brcmfmac: Fix bitmap malloc bug in msgbuf Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 29/73] arm64: Add COMPAT_HWCAP_LPAE Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 30/73] drm/tegra: gem: dumb: pitch and size are outputs Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 31/73] ARM: tegra: Re-add removed SoC id macro to tegra_resume() Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 32/73] ARM: mvebu: make the coherency_ll.S functions work with no coherency fabric Greg Kroah-Hartman
2015-01-07 2:15 ` Greg Kroah-Hartman [this message]
2015-01-07 2:15 ` [PATCH 3.17 35/73] x86/asm/traps: Disable tracing and kprobes in fixup_bad_iret and sync_regs Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 36/73] x86/tls: Dont validate lm in set_thread_area() after all Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 37/73] isofs: Fix unchecked printing of ER records Greg Kroah-Hartman
2015-01-07 2:15 ` [PATCH 3.17 38/73] KEYS: Fix stale key registration at error path Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 40/73] mac80211: free management frame keys when removing station Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 41/73] mnt: Fix a memory stomp in umount Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 42/73] thermal: Fix error path in thermal_init() Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 43/73] mnt: Implicitly add MNT_NODEV on remount when it was implicitly added by mount Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 44/73] mnt: Update unprivileged remount test Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 45/73] umount: Disallow unprivileged mount force Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 46/73] groups: Consolidate the setgroups permission checks Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 47/73] userns: Document what the invariant required for safe unprivileged mappings Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 48/73] userns: Dont allow setgroups until a gid mapping has been setablished Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 49/73] userns: Dont allow unprivileged creation of gid mappings Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 50/73] userns: Check euid no fsuid when establishing an unprivileged uid mapping Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 51/73] userns: Only allow the creator of the userns unprivileged mappings Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 52/73] userns: Rename id_map_mutex to userns_state_mutex Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 53/73] userns: Add a knob to disable setgroups on a per user namespace basis Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 54/73] userns: Allow setting gid_maps without privilege when setgroups is disabled Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 55/73] userns: Unbreak the unprivileged remount tests Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 56/73] audit: use supplied gfp_mask from audit_buffer in kauditd_send_multicast_skb Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 57/73] audit: dont attempt to lookup PIDs when changing PID filtering audit rules Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 58/73] audit: restore AUDIT_LOGINUID unset ABI Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 59/73] crypto: af_alg - fix backlog handling Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 60/73] ncpfs: return proper error from NCP_IOC_SETROOT ioctl Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 61/73] exit: pidns: alloc_pid() leaks pid_namespace if child_reaper is exiting Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 62/73] udf: Check path length when reading symlink Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 63/73] udf: Verify i_size when loading inode Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 64/73] udf: Verify symlink size before loading it Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 65/73] udf: Check component length before reading it Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 66/73] eCryptfs: Force RO mount when encrypted view is enabled Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 67/73] eCryptfs: Remove buggy and unnecessary write in file name decode routine Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 68/73] Btrfs: make sure we wait on logged extents when fsycning two subvols Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 69/73] btrfs: fix wrong accounting of raid1 data profile in statfs Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 70/73] Btrfs: do not move em to modified list when unpinning Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 71/73] Btrfs: make sure logged extents complete in the current transaction V3 Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 72/73] Btrfs: fix loop writing of async reclaim Greg Kroah-Hartman
2015-01-07 2:16 ` [PATCH 3.17 73/73] Btrfs: fix fs corruption on transaction abort if device supports discard Greg Kroah-Hartman
2015-01-07 13:46 ` [PATCH 3.17 00/73] 3.17.8-stable review Guenter Roeck
2015-01-07 23:46 ` Greg Kroah-Hartman
2015-01-07 23:34 ` 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=20150107021558.522207116@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=gregory.clement@free-electrons.com \
--cc=jason@lakedaemon.net \
--cc=linux-kernel@vger.kernel.org \
--cc=stable@vger.kernel.org \
--cc=thomas.petazzoni@free-electrons.com \
/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
Powered by JetHome