| Age | Commit message (Collapse) | Author |
|---|
|
Neal Cardwell says:
====================
tcp: fix xmit timer rearming to avoid stalls
This patch series is a bug fix for a TCP loss recovery performance bug
reported independently in recent netdev threads:
(i) July 26, 2017: netdev thread "TCP fast retransmit issues"
(ii) July 26, 2017: netdev thread:
"[PATCH V2 net-next] TLP: Don't reschedule PTO when there's one
outstanding TLP retransmission"
Many thanks to Klavs Klavsen and Mao Wenan for the detailed reports,
traces, and packetdrill test cases, which enabled us to root-cause
this issue and verify the fix.
- v1 -> v2:
- In patch 2/3, changed an unclear comment in the pre-existing code
in tcp_schedule_loss_probe() to be more clear (thanks to Eric Dumazet
for suggesting we improve this).
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Fix a TCP loss recovery performance bug raised recently on the netdev
list, in two threads:
(i) July 26, 2017: netdev thread "TCP fast retransmit issues"
(ii) July 26, 2017: netdev thread:
"[PATCH V2 net-next] TLP: Don't reschedule PTO when there's one
outstanding TLP retransmission"
The basic problem is that incoming TCP packets that did not indicate
forward progress could cause the xmit timer (TLP or RTO) to be rearmed
and pushed back in time. In certain corner cases this could result in
the following problems noted in these threads:
- Repeated ACKs coming in with bogus SACKs corrupted by middleboxes
could cause TCP to repeatedly schedule TLPs forever. We kept
sending TLPs after every ~200ms, which elicited bogus SACKs, which
caused more TLPs, ad infinitum; we never fired an RTO to fill in
the holes.
- Incoming data segments could, in some cases, cause us to reschedule
our RTO or TLP timer further out in time, for no good reason. This
could cause repeated inbound data to result in stalls in outbound
data, in the presence of packet loss.
This commit fixes these bugs by changing the TLP and RTO ACK
processing to:
(a) Only reschedule the xmit timer once per ACK.
(b) Only reschedule the xmit timer if tcp_clean_rtx_queue() deems the
ACK indicates sufficient forward progress (a packet was
cumulatively ACKed, or we got a SACK for a packet that was sent
before the most recent retransmit of the write queue head).
This brings us back into closer compliance with the RFCs, since, as
the comment for tcp_rearm_rto() notes, we should only restart the RTO
timer after forward progress on the connection. Previously we were
restarting the xmit timer even in these cases where there was no
forward progress.
As a side benefit, this commit simplifies and speeds up the TCP timer
arming logic. We had been calling inet_csk_reset_xmit_timer() three
times on normal ACKs that cumulatively acknowledged some data:
1) Once near the top of tcp_ack() to switch from TLP timer to RTO:
if (icsk->icsk_pending == ICSK_TIME_LOSS_PROBE)
tcp_rearm_rto(sk);
2) Once in tcp_clean_rtx_queue(), to update the RTO:
if (flag & FLAG_ACKED) {
tcp_rearm_rto(sk);
3) Once in tcp_ack() after tcp_fastretrans_alert() to switch from RTO
to TLP:
if (icsk->icsk_pending == ICSK_TIME_RETRANS)
tcp_schedule_loss_probe(sk);
This commit, by only rescheduling the xmit timer once per ACK,
simplifies the code and reduces CPU overhead.
This commit was tested in an A/B test with Google web server
traffic. SNMP stats and request latency metrics were within noise
levels, substantiating that for normal web traffic patterns this is a
rare issue. This commit was also tested with packetdrill tests to
verify that it fixes the timer behavior in the corner cases discussed
in the netdev threads mentioned above.
This patch is a bug fix patch intended to be queued for -stable
relases.
Fixes: 6ba8a3b19e76 ("tcp: Tail loss probe (TLP)")
Reported-by: Klavs Klavsen <kl@vsen.dk>
Reported-by: Mao Wenan <maowenan@huawei.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Nandita Dukkipati <nanditad@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Have tcp_schedule_loss_probe() base the TLP scheduling decision based
on when the RTO *should* fire. This is to enable the upcoming xmit
timer fix in this series, where tcp_schedule_loss_probe() cannot
assume that the last timer installed was an RTO timer (because we are
no longer doing the "rearm RTO, rearm RTO, rearm TLP" dance on every
ACK). So tcp_schedule_loss_probe() must independently figure out when
an RTO would want to fire.
In the new TLP implementation following in this series, we cannot
assume that icsk_timeout was set based on an RTO; after processing a
cumulative ACK the icsk_timeout we see can be from a previous TLP or
RTO. So we need to independently recalculate the RTO time (instead of
reading it out of icsk_timeout). Removing this dependency on the
nature of icsk_timeout makes things a little easier to reason about
anyway.
Note that the old and new code should be equivalent, since they are
both saying: "if the RTO is in the future, but at an earlier time than
the normal TLP time, then set the TLP timer to fire when the RTO would
have fired".
Fixes: 6ba8a3b19e76 ("tcp: Tail loss probe (TLP)")
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Nandita Dukkipati <nanditad@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Pure refactor. This helper will be required in the xmit timer fix
later in the patch series. (Because the TLP logic will want to make
this calculation.)
Fixes: 6ba8a3b19e76 ("tcp: Tail loss probe (TLP)")
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Nandita Dukkipati <nanditad@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Pull VFIO fixes from Alex Williamson:
- SPAPR/EEH config build fix (Murilo Opsfelder Araujo)
- Fix possible device lock deadlock (Alex Williamson)
- Correctly size integrated endpoint PCIe capabilities (Alex
Williamson)
* tag 'vfio-v4.13-rc4' of git://github.com/awilliam/linux-vfio:
vfio/pci: Fix handling of RC integrated endpoint PCIe capability size
vfio/pci: Use pci_try_reset_function() on initial open
include/linux/vfio.h: Guard powerpc-specific functions with CONFIG_VFIO_SPAPR_EEH
|
|
After commit c2ed1880fd61 ("net: ipv6: check route protocol when
deleting routes"), ipv6 route checks rt protocol when trying to
remove a rt entry.
It introduced a side effect causing 'ip -6 route flush cache' not
to work well. When flushing caches with iproute, all route caches
get dumped from kernel then removed one by one by sending DELROUTE
requests to kernel for each cache.
The thing is iproute sends the request with the cache whose proto
is set with RTPROT_REDIRECT by rt6_fill_node() when kernel dumps
it. But in kernel the rt_cache protocol is still 0, which causes
the cache not to be matched and removed.
So the real reason is rt6i_protocol in the route is not set when
it is allocated. As David Ahern's suggestion, this patch is to
set rt6i_protocol properly in the route when it is installed and
remove the codes setting rtm_protocol according to rt6i_flags in
rt6_fill_node.
This is also an improvement to keep rt6i_protocol consistent with
rtm_protocol.
Fixes: c2ed1880fd61 ("net: ipv6: check route protocol when deleting routes")
Reported-by: Jianlin Shi <jishi@redhat.com>
Suggested-by: David Ahern <dsahern@gmail.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Merge misc fixes from Andrew Morton:
"15 fixes"
[ This does not merge the "fortify: use WARN instead of BUG for now"
patch, which needs a bit of extra work to build cleanly with all
configurations. Arnd is on it. - Linus ]
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
ocfs2: don't clear SGID when inheriting ACLs
mm: allow page_cache_get_speculative in interrupt context
userfaultfd: non-cooperative: flush event_wqh at release time
ipc: add missing container_of()s for randstruct
cpuset: fix a deadlock due to incomplete patching of cpusets_enabled()
userfaultfd_zeropage: return -ENOSPC in case mm has gone
mm: take memory hotplug lock within numa_zonelist_order_handler()
mm/page_io.c: fix oops during block io poll in swapin path
zram: do not free pool->size_class
kthread: fix documentation build warning
kasan: avoid -Wmaybe-uninitialized warning
userfaultfd: non-cooperative: notify about unmap of destination during mremap
mm, mprotect: flush TLB if potentially racing with a parallel reclaim leaving stale TLB entries
pid: kill pidhash_size in pidhash_init()
mm/hugetlb.c: __get_user_pages ignores certain follow_hugetlb_page errors
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull ACPI fixes from Rafael Wysocki:
"These fix two issues in the ACPI SoC drivers (Intel LPSS and AMD APD),
a crash in the PCC mailbox initialization code and a WDAT watchdog
initialization failure.
Specifics:
- Fix a device ID of Hisilicon Hip07/08 in the ACPI APD (AMD SoC)
driver (Hanjun Guo).
- Fix list corruption (introduced during the 4.11 cycle) in the ACPI
LPSS (Intel SoC) driver (Hans de Goede).
- Fix PCC mailbox handling code crash during initialization when PCCT
is not present and PCC channel 0 is requested (Hoan Tran).
- Fix a WDAT watchdog initialization issue causing platform device
creation to fail due to partially overlapping address ranges in
resources (Ryan Kennedy)"
* tag 'acpi-4.13-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
ACPI: APD: Fix HID for Hisilicon Hip07/08
mailbox: pcc: Fix crash when request PCC channel 0
ACPI / watchdog: Fix init failure with overlapping register regions
ACPI / LPSS: Only call pwm_add_table() for the first PWM controller
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull power management fixes from Rafael Wysocki:
"These fix two cpufreq issues, one introduced recently and one related
to recent changes, fix cpufreq documentation, fix up recently added
code in the Thunderbolt driver and update runtime PM framework
documentation.
Specifics:
- Fix the handling of the scaling_cur_freq cpufreq policy attribute
on x86 systems with the MPERF/APERF registers present to make it
behave more as expected after recent changes (Rafael Wysocki).
- Drop a leftover callback from the intel_pstate driver which also
prevents the cpuinfo_cur_freq cpufreq policy attribute from being
incorrectly exposed when intel_pstate works in the active mode
(Rafael Wysocki).
- Add a missing piece describing the cpuinfo_cur_freq policy
attribute to cpufreq documentation (Rafael Wysocki).
- Fix up a recently added part of the Thunderbolt driver to avoid
aborting system suspends if its mailbox commands time out (Rafael
Wysocki).
- Update device runtime PM framework documentation to reflect the
current behavior of the code (Johan Hovold)"
* tag 'pm-4.13-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
thunderbolt: icm: Ignore mailbox errors in icm_suspend()
cpufreq: x86: Make scaling_cur_freq behave more as expected
PM / runtime: Document new pm_runtime_set_suspended() constraint
cpufreq: docs: Add missing cpuinfo_cur_freq description
cpufreq: intel_pstate: Drop ->get from intel_pstate structure
|
|
This WA is required when decoupled frequencies for slice and unslice
are enabled. This disables DOP clock gating for skl.
v2: enable the WA for all gen9 platforms (not just for SKL GT4 where
the hang issue is originally reported) to avoid rare hangs (David)
v3: as per WaDatabase, enable it only for SKL (Rodrigo)
Cc: David Weinehall <david.weinehall@linux.intel.com>
Reviewed-by: David Weinehall <david.weinehall@linux.intel.com>
Signed-off-by: Praveen Paneri <praveen.paneri@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1501781530-8186-1-git-send-email-praveen.paneri@intel.com
|
|
No functional change.
KBP was based on SPT and spec wasn't clear about the full name.
There was the initial point of the "Point" confusion.
Later the split with Coffee Lake and Cannon Lake both using CNP
and also some uncertainty from the specs we had at that time
made us to propagated the mistake along.
So, let's fix this now and avoid propagating these wrong
"points".
Cc: Anusha Srivatsa <anusha.srivatsa@intel.com>
Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Anusha Srivatsa <anusha.srivatsa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170731185220.758-1-rodrigo.vivi@intel.com
|
|
* acpi-soc:
ACPI: APD: Fix HID for Hisilicon Hip07/08
ACPI / LPSS: Only call pwm_add_table() for the first PWM controller
* acpi-wdat:
ACPI / watchdog: Fix init failure with overlapping register regions
* acpi-cppc:
mailbox: pcc: Fix crash when request PCC channel 0
|
|
* pm-core:
PM / runtime: Document new pm_runtime_set_suspended() constraint
* pm-misc:
thunderbolt: icm: Ignore mailbox errors in icm_suspend()
|
|
* pm-cpufreq-x86:
cpufreq: x86: Make scaling_cur_freq behave more as expected
* pm-cpufreq-docs:
cpufreq: docs: Add missing cpuinfo_cur_freq description
* intel_pstate:
cpufreq: intel_pstate: Drop ->get from intel_pstate structure
|
|
This adds a new drm_setup_crtcs_fb() function to handle the parts of
drm_setup_crtcs() that touch fb_helper->fb and fb_helper->fbdev. When
drm_setup_crtcs() is called during initialization, these fields are NULL
because they have not been allocated yet.
There is currently a hack at the end of drm_fb_helper_single_fb_probe()
that sets fb_helper->fb, so it is moved to the new drm_setup_crtcs_fb()
function.
This is also done in preparation for addition setup that requires access
to fb_helper->fbdev.
Signed-off-by: David Lechner <david@lechnology.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/1501777149-8310-2-git-send-email-david@lechnology.com
|
|
The motivation behind this new interface is expose at runtime the
creation of new OA configs which can be used as part of the i915 perf
open interface. This will enable the kernel to learn new configs which
may be experimental, or otherwise not part of the core set currently
available through the i915 perf interface.
v2: Drop DRM_ERROR for userspace errors (Matthew)
Add padding to userspace structure (Matthew)
s/guid/uuid/ (Matthew)
v3: Use u32 instead of int to iterate through registers (Matthew)
v4: Lock access to dynamic config list (Lionel)
v5: by Matthew:
Fix uninitialized error values
Fix incorrect unwiding when opening perf stream
Use kmalloc_array() to store register
Use uuid_is_valid() to valid config uuids
Declare ioctls as write only
Check padding members are set to 0
by Lionel:
Return ENOENT rather than EINVAL when trying to remove non
existing config
v6: by Chris:
Use ref counts for OA configs
Store UUID in drm_i915_perf_oa_config rather then using pointer
Shuffle fields of drm_i915_perf_oa_config to avoid padding
v7: by Chris
Rename uapi pointers fields to end with '_ptr'
v8: by Andrzej, Marek, Sebastian
Update register whitelisting
by Lionel
Add more register names for documentation
Allow configuration programming in non-paranoid mode
Add support for value filter for a couple of registers already
programmed in other part of the kernel
v9: Documentation fix (Lionel)
Allow writing WAIT_FOR_RC6_EXIT only on Gen8+ (Andrzej)
v10: Perform read access_ok() on register pointers (Lionel)
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Signed-off-by: Andrzej Datczuk <andrzej.datczuk@intel.com>
Reviewed-by: Andrzej Datczuk <andrzej.datczuk@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170803165812.2373-2-lionel.g.landwerlin@intel.com
|
|
It makes things easier to read when implementing whitelisting in the
following patches.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170803165812.2373-6-lionel.g.landwerlin@intel.com
|
|
We already do it on Haswell and the documentation says it saves power.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170803165812.2373-5-lionel.g.landwerlin@intel.com
|
|
There will be a need for userspaces configurations to set this
register. We can apply the same model inside the kernel for test
configs.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170803165812.2373-4-lionel.g.landwerlin@intel.com
|
|
In the following commit we'll introduce loadable userspace
configs. This change reworks how configurations are handled in the
perf driver and retains only the test configurations in kernel space.
We now store the test config in dev_priv and resolve the id only once
when opening the perf stream. The OA config is then handled through a
pointer to the structure holding the configuration details.
v2: Rework how test configs are handled (Lionel)
v3: Use u32 to hold number of register (Matthew)
v4: Removed unused dev_priv->perf.oa.current_config variable (Matthew)
v5: Lock device when accessing exclusive_stream (Lionel)
v6: Ensure OACTXCONTROL is always reprogrammed (Lionel)
v7: Switch a couple of index variable from int to u32 (Matthew)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170803165812.2373-3-lionel.g.landwerlin@intel.com
|
|
We were reserving fewer dwords in the ring than necessary. Indeed
we're always writing all registers once, so discard the actual number
of registers given by the user and just program the whitelisted ones
once.
Fixes: 19f81df2859e ("drm/i915/perf: Add OA unit support for Gen 8+")
Reported-by: Matthew Auld <matthew.william.auld@gmail.com>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Cc: <stable@vger.kernel.org> # v4.12+
Link: https://patchwork.freedesktop.org/patch/msgid/20170803165812.2373-6-lionel.g.landwerlin@intel.com
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/johan/usb-serial into usb-linus
Johan writes:
USB-serial fixes for v4.13-rc4
Here are some new device ids for v4.13-rc4.
All have been in linux-next with no reported issues.
Signed-off-by: Johan Hovold <johan@kernel.org>
|
|
syzkaller was able to trigger a divide by 0 in TCP stack [1]
Issue here is that keepalive timer needs to be updated to not attempt
to send a probe if the connection setup was deferred using
TCP_FASTOPEN_CONNECT socket option added in linux-4.11
[1]
divide error: 0000 [#1] SMP
CPU: 18 PID: 0 Comm: swapper/18 Not tainted
task: ffff986f62f4b040 ti: ffff986f62fa2000 task.ti: ffff986f62fa2000
RIP: 0010:[<ffffffff8409cc0d>] [<ffffffff8409cc0d>] __tcp_select_window+0x8d/0x160
Call Trace:
<IRQ>
[<ffffffff8409d951>] tcp_transmit_skb+0x11/0x20
[<ffffffff8409da21>] tcp_xmit_probe_skb+0xc1/0xe0
[<ffffffff840a0ee8>] tcp_write_wakeup+0x68/0x160
[<ffffffff840a151b>] tcp_keepalive_timer+0x17b/0x230
[<ffffffff83b3f799>] call_timer_fn+0x39/0xf0
[<ffffffff83b40797>] run_timer_softirq+0x1d7/0x280
[<ffffffff83a04ddb>] __do_softirq+0xcb/0x257
[<ffffffff83ae03ac>] irq_exit+0x9c/0xb0
[<ffffffff83a04c1a>] smp_apic_timer_interrupt+0x6a/0x80
[<ffffffff83a03eaf>] apic_timer_interrupt+0x7f/0x90
<EOI>
[<ffffffff83fed2ea>] ? cpuidle_enter_state+0x13a/0x3b0
[<ffffffff83fed2cd>] ? cpuidle_enter_state+0x11d/0x3b0
Tested:
Following packetdrill no longer crashes the kernel
`echo 0 >/proc/sys/net/ipv4/tcp_timestamps`
// Cache warmup: send a Fast Open cookie request
0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
+0 fcntl(3, F_SETFL, O_RDWR|O_NONBLOCK) = 0
+0 setsockopt(3, SOL_TCP, TCP_FASTOPEN_CONNECT, [1], 4) = 0
+0 connect(3, ..., ...) = -1 EINPROGRESS (Operation is now in progress)
+0 > S 0:0(0) <mss 1460,nop,nop,sackOK,nop,wscale 8,FO,nop,nop>
+.01 < S. 123:123(0) ack 1 win 14600 <mss 1460,nop,nop,sackOK,nop,wscale 6,FO abcd1234,nop,nop>
+0 > . 1:1(0) ack 1
+0 close(3) = 0
+0 > F. 1:1(0) ack 1
+0 < F. 1:1(0) ack 2 win 92
+0 > . 2:2(0) ack 2
+0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 4
+0 fcntl(4, F_SETFL, O_RDWR|O_NONBLOCK) = 0
+0 setsockopt(4, SOL_TCP, TCP_FASTOPEN_CONNECT, [1], 4) = 0
+0 setsockopt(4, SOL_SOCKET, SO_KEEPALIVE, [1], 4) = 0
+.01 connect(4, ..., ...) = 0
+0 setsockopt(4, SOL_TCP, TCP_KEEPIDLE, [5], 4) = 0
+10 close(4) = 0
`echo 1 >/proc/sys/net/ipv4/tcp_timestamps`
Fixes: 19f6d3f3c842 ("net/tcp-fastopen: Add new API support")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Cc: Wei Wang <weiwan@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Simon Wunderlich says:
====================
Here is a batman-adv bugfix:
- fix TT sync flag inconsistency problems, which can lead to excess packets,
by Linus Luessing
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/balbi/usb into usb-linus
Felipe writes:
usb: fixes for v4.13-rc4
Another fix for isochronous transfers on dwc3. This time around, we're
making sure that we use correct PIDs in all transfer sizes.
MSM PHY driver got a fix for the use of devm_regulator_bulk_get() API
which will avoid kernel crashes.
Renesas DRD driver got 3 fixes: a fix on giveback, a fix for proper
controller programming and the removal of set-but-never-used variable.
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm
KVM/ARM Fixes for v4.13-rc4
- Yet another race with VM destruction plugged
- A set of small vgic fixes
|
|
Commit 8fba54aebbdf ("fuse: direct-io: don't dirty ITER_BVEC pages") fixes
the ITER_BVEC page deadlock for direct io in fuse by checking in
fuse_direct_io(), whether the page is a bvec page or not, before locking
it. However, this check is missed when the "async_dio" mount option is
enabled. In this case, set_page_dirty_lock() is called from the req->end
callback in request_end(), when the fuse thread is returning from userspace
to respond to the read request. This will cause the same deadlock because
the bvec condition is not checked in this path.
Here is the stack of the deadlocked thread, while returning from userspace:
[13706.656686] INFO: task glusterfs:3006 blocked for more than 120 seconds.
[13706.657808] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables
this message.
[13706.658788] glusterfs D ffffffff816c80f0 0 3006 1
0x00000080
[13706.658797] ffff8800d6713a58 0000000000000086 ffff8800d9ad7000
ffff8800d9ad5400
[13706.658799] ffff88011ffd5cc0 ffff8800d6710008 ffff88011fd176c0
7fffffffffffffff
[13706.658801] 0000000000000002 ffffffff816c80f0 ffff8800d6713a78
ffffffff816c790e
[13706.658803] Call Trace:
[13706.658809] [<ffffffff816c80f0>] ? bit_wait_io_timeout+0x80/0x80
[13706.658811] [<ffffffff816c790e>] schedule+0x3e/0x90
[13706.658813] [<ffffffff816ca7e5>] schedule_timeout+0x1b5/0x210
[13706.658816] [<ffffffff81073ffb>] ? gup_pud_range+0x1db/0x1f0
[13706.658817] [<ffffffff810668fe>] ? kvm_clock_read+0x1e/0x20
[13706.658819] [<ffffffff81066909>] ? kvm_clock_get_cycles+0x9/0x10
[13706.658822] [<ffffffff810f5792>] ? ktime_get+0x52/0xc0
[13706.658824] [<ffffffff816c6f04>] io_schedule_timeout+0xa4/0x110
[13706.658826] [<ffffffff816c8126>] bit_wait_io+0x36/0x50
[13706.658828] [<ffffffff816c7d06>] __wait_on_bit_lock+0x76/0xb0
[13706.658831] [<ffffffffa0545636>] ? lock_request+0x46/0x70 [fuse]
[13706.658834] [<ffffffff8118800a>] __lock_page+0xaa/0xb0
[13706.658836] [<ffffffff810c8500>] ? wake_atomic_t_function+0x40/0x40
[13706.658838] [<ffffffff81194d08>] set_page_dirty_lock+0x58/0x60
[13706.658841] [<ffffffffa054d968>] fuse_release_user_pages+0x58/0x70 [fuse]
[13706.658844] [<ffffffffa0551430>] ? fuse_aio_complete+0x190/0x190 [fuse]
[13706.658847] [<ffffffffa0551459>] fuse_aio_complete_req+0x29/0x90 [fuse]
[13706.658849] [<ffffffffa05471e9>] request_end+0xd9/0x190 [fuse]
[13706.658852] [<ffffffffa0549126>] fuse_dev_do_write+0x336/0x490 [fuse]
[13706.658854] [<ffffffffa054963e>] fuse_dev_write+0x6e/0xa0 [fuse]
[13706.658857] [<ffffffff812a9ef3>] ? security_file_permission+0x23/0x90
[13706.658859] [<ffffffff81205300>] do_iter_readv_writev+0x60/0x90
[13706.658862] [<ffffffffa05495d0>] ? fuse_dev_splice_write+0x350/0x350
[fuse]
[13706.658863] [<ffffffff812062a1>] do_readv_writev+0x171/0x1f0
[13706.658866] [<ffffffff810b3d00>] ? try_to_wake_up+0x210/0x210
[13706.658868] [<ffffffff81206361>] vfs_writev+0x41/0x50
[13706.658870] [<ffffffff81206496>] SyS_writev+0x56/0xf0
[13706.658872] [<ffffffff810257a1>] ? syscall_trace_leave+0xf1/0x160
[13706.658874] [<ffffffff816cbb2e>] system_call_fastpath+0x12/0x71
Fix this by making should_dirty a fuse_io_priv parameter that can be
checked in fuse_aio_complete_req().
Reported-by: Tiger Yang <tiger.yang@oracle.com>
Signed-off-by: Ashish Samant <ashish.samant@oracle.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
There is a small chance that the compiler could generate separate loads
for the dist->propbaser which could be modified from another CPU. As we
want to make sure we atomically update the entire value, and don't race
with other updates, guarantee that the cmpxchg operation compares
against the original value.
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
|
|
drm_*_reference() and drm_*_unreference() functions are just
compatibility alias for drm_*_get() and drm_*_put() and should not be
used by new code. So convert all users of compatibility functions to use
the new APIs.
Signed-off-by: Cihangir Akturk <cakturk@gmail.com>
Acked-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1501761585-11757-6-git-send-email-cakturk@gmail.com
|
|
drm_plane_helper_funcs and drm_plane_funcsare not supposed to change
at runtime. All functions working with drm_plane_helper_funcs and
drm_plane_funcs work with const. So mark the non-const structs as const.
File size before:
text data bss dec hex filename
6072 596 0 6668 1a0c atmel_hlcdc_plane.o
File size After adding 'const':
text data bss dec hex filename
6218 436 0 6654 19fe atmel_hlcdc_plane.o
Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Link: https://patchwork.freedesktop.org/patch/msgid/646415a3b2e62182f85254115e8491e5caf4b2c7.1499098826.git.arvind.yadav.cs@gmail.com
|
|
------------[ cut here ]------------
WARNING: CPU: 5 PID: 2288 at arch/x86/kvm/vmx.c:11124 nested_vmx_vmexit+0xd64/0xd70 [kvm_intel]
CPU: 5 PID: 2288 Comm: qemu-system-x86 Not tainted 4.13.0-rc2+ #7
RIP: 0010:nested_vmx_vmexit+0xd64/0xd70 [kvm_intel]
Call Trace:
vmx_check_nested_events+0x131/0x1f0 [kvm_intel]
? vmx_check_nested_events+0x131/0x1f0 [kvm_intel]
kvm_arch_vcpu_ioctl_run+0x5dd/0x1be0 [kvm]
? vmx_vcpu_load+0x1be/0x220 [kvm_intel]
? kvm_arch_vcpu_load+0x62/0x230 [kvm]
kvm_vcpu_ioctl+0x340/0x700 [kvm]
? kvm_vcpu_ioctl+0x340/0x700 [kvm]
? __fget+0xfc/0x210
do_vfs_ioctl+0xa4/0x6a0
? __fget+0x11d/0x210
SyS_ioctl+0x79/0x90
do_syscall_64+0x8f/0x750
? trace_hardirqs_on_thunk+0x1a/0x1c
entry_SYSCALL64_slow_path+0x25/0x25
This can be reproduced by booting L1 guest w/ 'noapic' grub parameter, which
means that tells the kernel to not make use of any IOAPICs that may be present
in the system.
Actually external_intr variable in nested_vmx_vmexit() is the req_int_win
variable passed from vcpu_enter_guest() which means that the L0's userspace
requests an irq window. I observed the scenario (!kvm_cpu_has_interrupt(vcpu) &&
L0's userspace reqeusts an irq window) is true, so there is no interrupt which
L1 requires to inject to L2, we should not attempt to emualte "Acknowledge
interrupt on exit" for the irq window requirement in this scenario.
This patch fixes it by not attempt to emulate "Acknowledge interrupt on exit"
if there is no L1 requirement to inject an interrupt to L2.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
[Added code comment to make it obvious that the behavior is not correct.
We should do a userspace exit with open interrupt window instead of the
nested VM exit. This patch still improves the behavior, so it was
accepted as a (temporary) workaround.]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
|
|
for_each_obj_in_state is about to be removed, so use the correct new
iterator macros.
Also look at new_plane_state instead of plane->state when looking up
the hw planes in use. They should be the same except when reallocating,
(in which case this code is skipped) and we should really stop looking
at obj->state whenever possible.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
|
|
The driver recently switched from handling page flip completion in the
DU vertical blanking handler to the VSP frame end handler to fix a race
condition. This unfortunately resulted in incorrect timestamps in the
vertical blanking events sent to userspace as vertical blanking is now
handled after sending the event.
To fix this we must reverse the order of the two operations. The easiest
way is to handle vertical blanking in the VSP frame end handler before
sending the event. The VSP frame end interrupt occurs approximately 50µs
earlier than the DU frame end interrupt, but this should not cause any
undue harm.
As we need to handle vertical blanking even when page flip completion is
delayed, the VSP driver now needs to call the frame end completion
callback unconditionally, with a new argument to report whether page
flip has completed.
With this new scheme the DU vertical blanking interrupt isn't needed
anymore, so we can stop enabling it.
Fixes: d503a43ac06a ("drm: rcar-du: Register a completion callback with VSP1")
Signed-off-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Acked-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
|
|
When stopping the CRTC the driver must disable all planes and wait for
the change to take effect at the next vblank. Merely calling
drm_crtc_wait_one_vblank() is not enough, as the function doesn't
include any mechanism to handle the race with vblank interrupts.
Replace the drm_crtc_wait_one_vblank() call with a manual mechanism that
handles the vblank interrupt race.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
|
|
Page flips can take more than one vertical blanking to complete if
arming the page flips races with the vertical blanking interrupt.
Waiting for one vblank to complete the atomic commit in the commit tail
handler is thus incorrect, and can lead to framebuffers being released
while still being scanned out.
Fix this by waiting for flip completion instead, using the
drm_atomic_helper_wait_for_flip_done() helper.
Fixes: 0d230422d256 ("drm: rcar-du: Register a completion callback with VSP1")
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
|
|
When implementing support for interlaced modes, the driver switched from
reporting vblank events on the vertical blanking (VBK) interrupt to the
frame end interrupt (FRM). This incorrectly divided the reported refresh
rate by two. Fix it by moving back to the VBK interrupt.
Fixes: 906eff7fcada ("drm: rcar-du: Implement support for interlaced modes")
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
|
|
Update the device description with the HDMI output.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
|
|
"params" can't be NULL here. The next lines assume that we either
hit the break statement of "params->mpixelclock == ~0UL". The
inconsistent NULL checking makes static checkers complain. I've just
removed the test.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
|
|
Commit 52055bafa1ff ("drm: rcar-du: Move plane commit code from CRTC
start to CRTC resume") changed the order of the plane commit and CRTC
enable operations to accommodate the runtime PM requirements. However,
this introduced corruption in the first displayed frame, as the CRTC is
now enabled without any plane configured. On Gen2 hardware the first
frame will be black and likely unnoticed, but on Gen3 hardware we end up
starting the display before the VSP compositor, which is more
noticeable.
To fix this, revert the order of the commit operations back, and handle
runtime PM requirements in the CRTC .atomic_begin() and .atomic_enable()
helper operation handlers.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
|
|
On Gen3 SoCs DPAD0 routing is configured through the last CRTC group,
unlike on Gen2 where it is configured through the first CRTC group. Fix
the driver accordingly.
Fixes: 2427b3037710 ("drm: rcar-du: Add R8A7795 device support")
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
|
|
The H3 ES1.x exhibits dot clock duty cycle stability issues. We can work
around them by configuring the DPLL to twice the desired frequency,
coupled with a /2 post-divider. This isn't needed on other SoCs and
breaks HDMI output on M3-W for a currently unknown reason, so restrict
the workaround to H3 ES1.x.
From an implementation point of view, move work around handling outside
of the rcar_du_dpll_divider() function by requesting a x2 DPLL output
frequency explicitly. The existing post-divider calculation mechanism
will then take care of dividing the clock by two automatically.
While at it, print a more useful debugging message to ease debugging
clock rate issues.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
|
|
On R-Car H3 ES2.0, DU channels 0 and 3 are served by two separate
pipelines from the same VSP. Support this in the DU driver.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
|
|
To avoid mixing comment styles when new comments complying with the
kernel coding style are introduced, fix all multiline comments in one
go.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Acked-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
|
|
Now, we can use of_graph_get_remote_endpoint(). Let's use it.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
|
|
The VSP supports both header and headerless display lists. The latter is
easier to use when the VSP feeds data directly to the DU in continuous
mode, and the driver thus uses headerless display lists for DU operation
and header display lists otherwise.
Headerless display lists are only available on WPF.0. This has never
been an issue so far, as only WPF.0 is connected to the DU. However, on
H3 ES2.0, the VSP-DL instance has both WPF.0 and WPF.1 connected to the
DU. We thus can't use headerless display lists unconditionally for DU
operation.
Implement support for continuous mode with header display lists, and use
it for DU operation on WPF outputs that don't support headerless mode.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Acked-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
|
|
The R-Car H3 ES2.0 VSP-DL instance has two LIF entities and can drive
two display pipelines at the same time. Refactor the VSP DRM code to
support that by introducing a vsp_drm_pipeline object that models one
display pipeline.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Acked-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
|
|
The VSP2-DL instance (present in the H3 ES2.0 and M3-N SoCs) has two LIF
instances. Adapt the driver infrastructure to support multiple LIFs.
Support for multiple display pipelines will be added separately.
The change to the entity routing table removes the ability to connect
the LIF output to the HGO or HGT histogram generators. This feature is
only available on Gen2 hardware, isn't supported by the rest of the
driver, and has no known use case, so this isn't an issue.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Acked-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
|
|
New Gen3 SoCs come with two new VSP2 variants names VSP2-BS and VSP2-DL,
as well as a new VSP2-D variant on V3M and V3H SoCs. Add new entries for
them in the VSP device info table.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Acked-by: Mauro Carvalho Chehab <mchehab@s-opensource.com>
|
|
Add const to bin_attribute structures as they are only passed to the
functions sysfs_{remove/create}_bin_file or
device_{remove/create}_bin_file. The corresponding arguments are of
type const, so declare the structures to be const.
Signed-off-by: Bhumika Goyal <bhumirks@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/1501694447-14356-1-git-send-email-bhumirks@gmail.com
|
|
fb_crtc is used before a null check, so move the use after the null check.
This was just identified by inspection. I haven't actually observed a crash
here, so it is possible that the null check could be unnecessary.
Signed-off-by: David Lechner <david@lechnology.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/1501696813-8807-1-git-send-email-david@lechnology.com
|
