Russell King [Tue, 25 Jul 2017 14:03:13 +0000 (15:03 +0100)]
phylink: add phylink infrastructure
The link between the ethernet MAC and its PHY has become more complex
as the interface evolves. This is especially true with serdes links,
where the part of the PHY is effectively integrated into the MAC.
Serdes links can be connected to a variety of devices, including SFF
modules soldered down onto the board with the MAC, a SFP cage with
a hotpluggable SFP module which may contain a PHY or directly modulate
the serdes signals onto optical media with or without a PHY, or even
a classical PHY connection.
Moreover, the negotiation information on serdes links comes in two
varieties - SGMII mode, where the PHY provides its speed/duplex/flow
control information to the MAC, and 1000base-X mode where both ends
exchange their abilities and each resolve the link capabilities.
This means we need a more flexible means to support these arrangements,
particularly with the hotpluggable nature of SFP, where the PHY can
be attached or detached after the network device has been brought up.
Ethtool information can come from multiple sources:
- we may have a PHY operating in either SGMII or 1000base-X mode, in
which case we take ethtool/mii data directly from the PHY.
- we may have a optical SFP module without a PHY, with the MAC
operating in 1000base-X mode - the ethtool/mii data needs to come
from the MAC.
- we may have a copper SFP module with a PHY whic can't be accessed,
which means we need to take ethtool/mii data from the MAC.
Phylink aims to solve this by providing an intermediary between the
MAC and PHY, providing a safe way for PHYs to be hotplugged, and
allowing a SFP driver to reconfigure the serdes connection.
Phylink also takes over support of fixed link connections, where the
speed/duplex/flow control are fixed, but link status may be controlled
by a GPIO signal. By avoiding the fixed-phy implementation, phylink
can provide a faster response to link events: fixed-phy has to wait for
phylib to operate its state machine, which can take several seconds.
In comparison, phylink takes milliseconds.
Signed-off-by: Russell King <[email protected]>
- remove sync status
- rework supported and advertisment handling
- add 1000base-x speed for fixed links
- use functionality exported from phy-core, reworking
__phylink_ethtool_ksettings_set for it Reviewed-by: Andrew Lunn <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Russell King [Tue, 25 Jul 2017 14:03:08 +0000 (15:03 +0100)]
net: phy: add I2C mdio bus
Add an I2C MDIO bus bridge library, to allow phylib to access PHYs which
are connected to an I2C bus instead of the more conventional MDIO bus.
Such PHYs can be found in SFP adapters and SFF modules.
Since PHYs appear at I2C bus address 0x40..0x5f, and 0x50/0x51 are
reserved for SFP EEPROMs/diagnostics, we must not allow the MDIO bus
to access these I2C addresses.
Russell King [Tue, 25 Jul 2017 14:03:03 +0000 (15:03 +0100)]
net: phy: export phy_start_machine() for phylink
phylink will need phy_start_machine exported, so lets export it as a
GPL symbol. Documentation/networking/phy.txt indicates that this
should be a PHY API function.
Russell King [Tue, 25 Jul 2017 14:02:58 +0000 (15:02 +0100)]
net: phy: provide a hook for link up/link down events
Sometimes, we need to do additional work between the PHY coming up and
marking the carrier present - for example, we may need to wait for the
PHY to MAC link to finish negotiation. This changes phylib to provide
a notification function pointer which avoids the built-in
netif_carrier_on() and netif_carrier_off() functions.
Standard ->adjust_link functionality is provided by hooking a helper
into the new ->phy_link_change method.
Russell King [Tue, 25 Jul 2017 14:02:52 +0000 (15:02 +0100)]
net: phy: add 1000Base-X to phy settings table
Add the missing 1000Base-X entry to the phy settings table. This was
not included because the original code could not cope with more than
32 bits of link mode mask.
Russell King [Tue, 25 Jul 2017 14:02:47 +0000 (15:02 +0100)]
net: phy: move phy_lookup_setting() and guts of phy_supported_speeds() to phy-core
phy_lookup_setting() provides useful functionality in ethtool code
outside phylib. Move it to phy-core and allow it to be re-used (eg,
in phylink) rather than duplicated elsewhere. Note that this supports
the larger linkmode space.
As we move the phy settings table, we also need to move the guts of
phy_supported_speeds() as well.
Russell King [Tue, 25 Jul 2017 14:02:37 +0000 (15:02 +0100)]
net: phy: allow settings table to support more than 32 link modes
Allow the phy settings table to support more than 32 link modes by
switching to the ethtool link mode bit number representation, rather
than storing the mask. This will allow phylink and other ethtool
code to share the settings table to look up settings.
The __ip_options_echo() function expect a valid dst entry in skb->dst;
as result we sometimes need to preserve the dst entry for the whole IP
RX path.
The current usage of skb->dst looks more a relic from ancient past that
a real functional constraint. This patchset tries to remove such usage,
and than drops some hacks currently in place in the IP code to keep
skb->dst around.
__ip_options_echo() uses of skb->dst for two different purposes: retrieving
the netns assicated with the skb, and modify the ingress packet LSRR address
list.
The first patch removes the code modifying the ingress packet, and the second
one provides an explicit netns argument to __ip_options_echo(). The following
patches cleanup the current code keeping arund skb->dst for __ip_options_echo's
sake.
Updating the __ip_options_echo() function has been previously discussed here:
Paolo Abeni [Thu, 3 Aug 2017 16:07:08 +0000 (18:07 +0200)]
udp: no need to preserve skb->dst
__ip_options_echo() does not need anymore skb->dst, so we can
avoid explicitly preserving it for its own sake.
This is almost a revert of commit 0ddf3fb2c43d ("udp: preserve
skb->dst if required for IP options processing") plus some
lifting to fit later changes.
Paolo Abeni [Thu, 3 Aug 2017 16:07:07 +0000 (18:07 +0200)]
Revert "ipv4: keep skb->dst around in presence of IP options"
ip_options_echo() does not use anymore the skb->dst and don't
need to keep the dst around for options's sake only.
This reverts commit 34b2cef20f19c87999fff3da4071e66937db9644.
Paolo Abeni [Thu, 3 Aug 2017 16:07:05 +0000 (18:07 +0200)]
IP: do not modify ingress packet IP option in ip_options_echo()
While computing the response option set for LSRR, ip_options_echo()
also changes the ingress packet LSRR addresses list, setting
the last one to the dst specific address for the ingress packet
- via memset(start[ ...
The only visible effect of such change - beyond possibly damaging
shared/cloned skbs - is modifying the data carried by ICMP replies
changing the header information for reported the ingress packet,
which violates RFC1122 3.2.2.6.
All the others call sites just ignore the ingress packet IP options
after calling ip_options_echo()
Note that the last element in the LSRR option address list for the
reply packet will be properly set later in the ip output path
via ip_options_build().
This buggy memset() predates git history and apparently was present
into the initial ip_options_echo() implementation in linux 1.3.30 but
still looks wrong.
The removal of the fib_compute_spec_dst() call will help
completely dropping the skb->dst usage by __ip_options_echo() with a
later patch.
Linus Torvalds [Sun, 6 Aug 2017 23:11:34 +0000 (16:11 -0700)]
Merge tag 'platform-drivers-x86-v4.13-4' of git://git.infradead.org/linux-platform-drivers-x86
Pull x86 platform driver fix from Darren Hart:
"Fix loop preventing some platforms from waking up via the power button
in s2idle:
- intel-vbtn: match power button on press rather than release"
* tag 'platform-drivers-x86-v4.13-4' of git://git.infradead.org/linux-platform-drivers-x86:
platform/x86: intel-vbtn: match power button on press rather than release
Linus Torvalds [Sun, 6 Aug 2017 19:31:17 +0000 (12:31 -0700)]
Merge tag 'ext4_for_linus_stable' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4
Pull ext4 fixes from Ted Ts'o:
"A large number of ext4 bug fixes and cleanups for v4.13"
* tag 'ext4_for_linus_stable' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
ext4: fix copy paste error in ext4_swap_extents()
ext4: fix overflow caused by missing cast in ext4_resize_fs()
ext4, project: expand inode extra size if possible
ext4: cleanup ext4_expand_extra_isize_ea()
ext4: restructure ext4_expand_extra_isize
ext4: fix forgetten xattr lock protection in ext4_expand_extra_isize
ext4: make xattr inode reads faster
ext4: inplace xattr block update fails to deduplicate blocks
ext4: remove unused mode parameter
ext4: fix warning about stack corruption
ext4: fix dir_nlink behaviour
ext4: silence array overflow warning
ext4: fix SEEK_HOLE/SEEK_DATA for blocksize < pagesize
ext4: release discard bio after sending discard commands
ext4: convert swap_inode_data() over to use swap() on most of the fields
ext4: error should be cleared if ea_inode isn't added to the cache
ext4: Don't clear SGID when inheriting ACLs
ext4: preserve i_mode if __ext4_set_acl() fails
ext4: remove unused metadata accounting variables
ext4: correct comment references to ext4_ext_direct_IO()
Linus Torvalds [Sun, 6 Aug 2017 18:52:01 +0000 (11:52 -0700)]
Merge branch 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus
Pull MIPS fixes from Ralf Baechle:
"This fixes two build issues for ralink platforms, both due to missing
#includes which used to be included indirectly via other headers"
* 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus:
MIPS: ralink: mt7620: Add missing header
MIPS: ralink: Fix build error due to missing header
Dmitry V. Levin [Sat, 5 Aug 2017 20:00:50 +0000 (23:00 +0300)]
Fix compat_sys_sigpending breakage
The latest change of compat_sys_sigpending in commit 8f13621abced
("sigpending(): move compat to native") has broken it in two ways.
First, it tries to write 4 bytes more than userspace expects:
sizeof(old_sigset_t) == sizeof(long) == 8 instead of
sizeof(compat_old_sigset_t) == sizeof(u32) == 4.
Second, on big endian architectures these bytes are being written in the
wrong order.
Jerry Lee [Sun, 6 Aug 2017 05:18:31 +0000 (01:18 -0400)]
ext4: fix overflow caused by missing cast in ext4_resize_fs()
On a 32-bit platform, the value of n_blcoks_count may be wrong during
the file system is resized to size larger than 2^32 blocks. This may
caused the superblock being corrupted with zero blocks count.
Miao Xie [Sun, 6 Aug 2017 05:00:49 +0000 (01:00 -0400)]
ext4, project: expand inode extra size if possible
When upgrading from old format, try to set project id
to old file first time, it will return EOVERFLOW, but if
that file is dirtied(touch etc), changing project id will
be allowed, this might be confusing for users, we could
try to expand @i_extra_isize here too.
Miao Xie [Sun, 6 Aug 2017 04:40:01 +0000 (00:40 -0400)]
ext4: restructure ext4_expand_extra_isize
Current ext4_expand_extra_isize just tries to expand extra isize, if
someone is holding xattr lock or some check fails, it will give up.
So rename its name to ext4_try_to_expand_extra_isize.
Besides that, we clean up unnecessary check and move some relative checks
into it.
Miao Xie [Sun, 6 Aug 2017 04:27:38 +0000 (00:27 -0400)]
ext4: fix forgetten xattr lock protection in ext4_expand_extra_isize
We should avoid the contention between the i_extra_isize update and
the inline data insertion, so move the xattr trylock in front of
i_extra_isize update.
Tahsin Erdogan [Sun, 6 Aug 2017 04:07:01 +0000 (00:07 -0400)]
ext4: make xattr inode reads faster
ext4_xattr_inode_read() currently reads each block sequentially while
waiting for io operation to complete before moving on to the next
block. This prevents request merging in block layer.
Add a ext4_bread_batch() function that starts reads for all blocks
then optionally waits for them to complete. A similar logic is used
in ext4_find_entry(), so update that code to use the new function.
Tahsin Erdogan [Sun, 6 Aug 2017 02:41:42 +0000 (22:41 -0400)]
ext4: inplace xattr block update fails to deduplicate blocks
When an xattr block has a single reference, block is updated inplace
and it is reinserted to the cache. Later, a cache lookup is performed
to see whether an existing block has the same contents. This cache
lookup will most of the time return the just inserted entry so
deduplication is not achieved.
Running the following test script will produce two xattr blocks which
can be observed in "File ACL: " line of debugfs output:
Arnd Bergmann [Sun, 6 Aug 2017 01:57:46 +0000 (21:57 -0400)]
ext4: fix warning about stack corruption
After commit 62d1034f53e3 ("fortify: use WARN instead of BUG for now"),
we get a warning about possible stack overflow from a memcpy that
was not strictly bounded to the size of the local variable:
inlined from 'ext4_mb_seq_groups_show' at fs/ext4/mballoc.c:2322:2:
include/linux/string.h:309:9: error: '__builtin_memcpy': writing between 161 and 1116 bytes into a region of size 160 overflows the destination [-Werror=stringop-overflow=]
We actually had a bug here that would have been found by the warning,
but it was already fixed last year in commit 30a9d7afe70e ("ext4: fix
stack memory corruption with 64k block size").
This replaces the fixed-length structure on the stack with a variable-length
structure, using the correct upper bound that tells the compiler that
everything is really fine here. I also change the loop count to check
for the same upper bound for consistency, but the existing code is
already correct here.
Note that while clang won't allow certain kinds of variable-length arrays
in structures, this particular instance is fine, as the array is at the
end of the structure, and the size is strictly bounded.
Andreas Dilger [Sat, 5 Aug 2017 23:47:34 +0000 (19:47 -0400)]
ext4: fix dir_nlink behaviour
The dir_nlink feature has been enabled by default for new ext4
filesystems since e2fsprogs-1.41 in 2008, and was automatically
enabled by the kernel for older ext4 filesystems since the
dir_nlink feature was added with ext4 in kernel 2.6.28+ when
the subdirectory count exceeded EXT4_LINK_MAX-1.
Automatically adding the file system features such as dir_nlink is
generally frowned upon, since it could cause the file system to not be
mountable on older kernel, thus preventing the administrator from
rolling back to an older kernel if necessary.
In this case, the administrator might also want to disable the feature
because glibc's fts_read() function does not correctly optimize
directory traversal for directories that use st_nlinks field of 1 to
indicate that the number of links in the directory are not tracked by
the file system, and could fail to traverse the full directory
hierarchy. Fortunately, in the past ten years very few users have
complained about incomplete file system traversal by glibc's
fts_read().
This commit also changes ext4_inc_count() to allow i_nlinks to reach
the full EXT4_LINK_MAX links on the parent directory (including "."
and "..") before changing i_links_count to be 1.
Jan Kara [Sat, 5 Aug 2017 21:43:24 +0000 (17:43 -0400)]
ext4: fix SEEK_HOLE/SEEK_DATA for blocksize < pagesize
ext4_find_unwritten_pgoff() does not properly handle a situation when
starting index is in the middle of a page and blocksize < pagesize. The
following command shows the bug on filesystem with 1k blocksize:
platform/x86: intel-vbtn: match power button on press rather than release
This fixes a problem where the system gets stuck in a loop
unable to wakeup via power button in s2idle.
The problem happens because:
- press power button:
- system emits 0xc0 (power press), event ignored
- system emits 0xc1 (power release), event processed,
emited as KEY_POWER
- set wakeup_mode to true
- system goes to s2idle
- press power button
- system emits 0xc0 (power press), wakeup_mode is true,
system wakes
- system emits 0xc1 (power release), event processed,
emited as KEY_POWER
- system goes to s2idle again
To avoid this situation, process the presses (which matches what
intel-hid does too).
Linus Torvalds [Sat, 5 Aug 2017 21:09:26 +0000 (14:09 -0700)]
Merge tag 'media/v4.13-2' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media
Pull media fixes from Mauro Carvalho Chehab:
"This series is larger than I would like to submit for -rc4. My
original intent were to sent it to either -rc2 or -rc3. Unfortunately,
due to my vacations, I got a lot of pending stuff after my return, and
had to do some biz trips, with prevented me to send this earlier.
Several fixes:
- some fixes at atomisp staging driver
- several gcc 7 warning fixes
- cleanup media SVG files, in order to fix PDF build on some distros
- fix random Kconfig build of venus driver
- some fixes for the venus driver
- some changes from semaphone to mutex in ngene's driver
- some locking fixes at dib0700 driver
- several fixes on ngene's driver and frontends to make it properly
support some new boards added on Kernel 4.13
- some fixes to CEC drivers
- omap_vout: vrfb: convert to dmaengine
- docs-rst: document EBUSY for VIDIOC_S_FMT
Please notice that the big diffstat changes here are at the SVG files.
Visually, the images look the same, but the file size is now a lot
smaller than before, and they don't use some XML tags that would cause
them to be badly parsed by some ImageMagick versions, or to require a
lot of memory by TeTex, with would break PDF output on some
distributions"
* tag 'media/v4.13-2' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media: (68 commits)
media: atomisp2: array underflow in imx_enum_frame_size()
media: atomisp2: array underflow in ap1302_enum_frame_size()
media: atomisp2: Array underflow in atomisp_enum_input()
media: platform: davinci: drop VPFE_CMD_S_CCDC_RAW_PARAMS
media: platform: davinci: return -EINVAL for VPFE_CMD_S_CCDC_RAW_PARAMS ioctl
media: venus: don't abuse dma_alloc for non-DMA allocations
media: venus: hfi: fix error handling in hfi_sys_init_done()
media: venus: fix compile-test build on non-qcom ARM platform
media: venus: mark PM functions as __maybe_unused
media: cec-notifier: small improvements
media: pulse8-cec: persistent_config should be off by default
media: cec: cec_transmit_attempt_done: ignore CEC_TX_STATUS_MAX_RETRIES
media: staging: atomisp: array underflow in ioctl
media: lirc: LIRC_GET_REC_RESOLUTION should return microseconds
media: svg: avoid too long lines
media: svg files: simplify files
media: selection.svg: simplify the SVG file
media: vimc: set id_table for platform drivers
media: staging: atomisp: disable warnings with cc-disable-warning
media: davinci: variable 'common' set but not used
...
Daeho Jeong [Sat, 5 Aug 2017 17:11:57 +0000 (13:11 -0400)]
ext4: release discard bio after sending discard commands
We've changed the discard command handling into parallel manner.
But, in this change, I forgot decreasing the usage count of the bio
which was used to send discard request. I'm sorry about that.
Fixes: a015434480dc ("ext4: send parallel discards on commit completions") Signed-off-by: Daeho Jeong <[email protected]> Signed-off-by: Theodore Ts'o <[email protected]> Reviewed-by: Jan Kara <[email protected]>
Linus Torvalds [Sat, 5 Aug 2017 13:55:13 +0000 (06:55 -0700)]
Merge tag 'gpio-v4.13-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio
Pull GPIO fixes from Linus Walleij:
- LP87565: set the proper output level for direction_output.
- stm32: fix the kernel build by selecting the hierarchical irqdomain
symbol properly - this happens to be done in the pin control
framework but whatever, it had dependencies to GPIO so we need to
apply it here.
- Select the hierarchical IRQ domain also for Xgene.
- Fix wakeups to work on MXC.
- Fix up the device tree binding on Exar that went astray, also add the
right bindings.
- Fix the unwanted events for edges from the library.
- Fix the unbalanced chanined IRQ on the Tegra.
* tag 'gpio-v4.13-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio:
gpio: tegra: fix unbalanced chained_irq_enter/exit
gpiolib: skip unwanted events, don't convert them to opposite edge
gpio: exar: Use correct property prefix and document bindings
gpio: gpio-mxc: Fix: higher 16 GPIOs usable as wake source
gpio: xgene-sb: select IRQ_DOMAIN_HIERARCHY
pinctrl: stm32: select IRQ_DOMAIN_HIERARCHY instead of depends on
gpio: lp87565: Set proper output level and direction for direction_output
MAINTAINERS: Add entry for Whiskey Cove PMIC GPIO driver
Linus Torvalds [Fri, 4 Aug 2017 23:45:29 +0000 (16:45 -0700)]
Merge tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux
Pull clk fixes from Stephen Boyd:
"A handful of critical fixes for changes introduce this merge window.
- The TI sci_clk_get() API was pretty broken and nobody noticed.
- There were some CPUfreq crashes on C.H.I.P devices because we
failed to propagate rates up the clk tree.
- Also, the Intel Atom PMC clk driver needs to mark a clk critical if
the firmware has it enabled already so that audio doesn't get
killed on Baytrail.
- Gemini devices have a dead serial console because the reset control
usage in the serial driver assume one method of reset that gemini
doesn't support (this will be fixed in the next version in the
reset framework so this is the small fix for -rc series).
- Finally we have two rate calculation fixes, one for Exynos and one
for Meson SoCs, that fix rate inconsistencies"
* tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
clk: keystone: sci-clk: Fix sci_clk_get
clk: meson: mpll: fix mpll0 fractional part ignored
clk: samsung: exynos5420: The EPLL rate table corrections
clk: sunxi-ng: sun5i: Add clk_set_rate_parent to the CPU clock
clk: x86: Do not gate clocks enabled by the firmware
clk: gemini: Fix reset regression
Linus Torvalds [Fri, 4 Aug 2017 22:18:27 +0000 (15:18 -0700)]
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull KVM fixes from Radim Krčmář:
"ARM:
- Yet another race with VM destruction plugged
- A set of small vgic fixes
x86:
- Preserve pending INIT
- RCU fixes in paravirtual async pf, VM teardown, and VMXOFF
emulation
- nVMX interrupt injection and dirty tracking fixes
- initialize to make UBSAN happy"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
KVM: arm/arm64: vgic: Use READ_ONCE fo cmpxchg
KVM: nVMX: Fix interrupt window request with "Acknowledge interrupt on exit"
KVM: nVMX: mark vmcs12 pages dirty on L2 exit
kvm: nVMX: don't flush VMCS12 during VMXOFF or VCPU teardown
KVM: nVMX: do not pin the VMCS12
KVM: avoid using rcu_dereference_protected
KVM: X86: init irq->level in kvm_pv_kick_cpu_op
KVM: X86: Fix loss of pending INIT due to race
KVM: async_pf: make rcu irq exit if not triggered from idle task
KVM: nVMX: fixes to nested virt interrupt injection
KVM: nVMX: do not fill vm_exit_intr_error_code in prepare_vmcs12
KVM: arm/arm64: Handle hva aging while destroying the vm
KVM: arm/arm64: PMU: Fix overflow interrupt injection
KVM: arm/arm64: Fix bug in advertising KVM_CAP_MSI_DEVID capability
Linus Torvalds [Fri, 4 Aug 2017 22:16:09 +0000 (15:16 -0700)]
Merge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fix from Thomas Gleixner:
"The recent irq core changes unearthed API abuse in the HPET code,
which manifested itself in a suspend/resume regression.
The fix replaces the cruft with the proper function calls and cures
the regression"
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/hpet: Cure interface abuse in the resume path
Linus Torvalds [Fri, 4 Aug 2017 22:12:15 +0000 (15:12 -0700)]
Merge tag 'armsoc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
Pull ARM SoC fixes from Arnd Bergmann:
"This comes a bit later than I planned, and as a consequence is a
larger than it should be.
Most of the changes are devicetree fixes, across lots of platforms:
Renesas, Samsung Exynos, Marvell EBU, TI OMAP, Rockchips, Amlogic
Meson, Sigma Desings Tango, Allwinner SUNxi and TI Davinci.
Also across many platforms, I applied an older series of simple
randconfig build fixes. This includes making the CONFIG_MTD_XIP option
compile again, which had been broken for many years and probably has
not been missed, but it felt wrong to just remove it completely.
The only other changes are:
- We enable HWSPINLOCK in defconfig to get some Qualcomm boards to
work out of the box.
- A few regression fixes for Texas Instruments OMAP2+.
- A boot regression fix for the Renesas regulator quirk.
- A suspend/resume fix for Uniphier SoCs, fixing the resume of the
system bus"
* tag 'armsoc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc: (43 commits)
ARM: dts: tango4: Request RGMII RX and TX clock delays
bus: uniphier-system-bus: set up registers when resuming
ARM64: dts: marvell: armada-37xx: Fix the number of GPIO on south bridge
ARM: shmobile: rcar-gen2: Fix deadlock in regulator quirk
arm64: defconfig: enable missing HWSPINLOCK
ARM: pxa: select both FB and FB_W100 for eseries
ARM: ixp4xx: fix ioport_unmap definition
ARM: ep93xx: use ARM_PATCH_PHYS_VIRT correctly
ARM: mmp: mark usb_dma_mask as __maybe_unused
ARM: omap2: mark unused functions as __maybe_unused
ARM: omap1: avoid unused variable warning
ARM: sirf: mark sirfsoc_init_late as __maybe_unused
ARM: ixp4xx: use normal prototype for {read,write}s{b,w,l}
ARM: omap1/ams-delta: warn about failed regulator enable
ARM: rpc: rename RAM_SIZE macro
ARM: w90x900: normalize clk API
ARM: ep93xx: normalize clk API
ARM: dts: sun8i: a83t: Switch to CCU device tree binding macros
arm64: allwinner: sun50i-a64: Correct emac register size
ARM: dts: sunxi: h3/h5: Correct emac register size
...
Linus Torvalds [Fri, 4 Aug 2017 19:11:48 +0000 (12:11 -0700)]
Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull arm64 fixes from Will Deacon:
"Here are some more arm64 fixes for 4.13. The main one is the PTE race
with the hardware walker, but there are a couple of other things too.
- Report correct timer frequency to userspace when trapping
CNTFRQ_EL0
- Fix race with hardware page table updates when updating access
flags
- Silence clang overflow warning in VA_START and PAGE_OFFSET
calculations"
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64: avoid overflow in VA_START and PAGE_OFFSET
arm64: Fix potential race with hardware DBM in ptep_set_access_flags()
arm64: Use arch_timer_get_rate when trapping CNTFRQ_EL0
====================
net: sched: summer cleanup part 1, mainly in exts area
This patchset is one of the couple cleanup patchsets I have in queue.
The motivation aside the obvious need to "make things nicer" is also
to prepare for shared filter blocks introduction. That requires tp->q
removal, and therefore removal of all tp->q users.
Patch 1 is just some small thing I spotted on the way
Patch 2 removes one user of tp->q, namely tcf_em_tree_change
Patches 3-8 do preparations for exts->nr_actions removal
Patches 9-10 do simple renames of functions in cls*
Patches 11-19 remove unnecessary calls of tcf_exts_change helper
The last patch changes tcf_exts_change to don't take lock
Tested by tools/testing/selftests/tc-testing
v1->v2:
- removed conversion of action array to list as noted by Cong
- added the past patch instead
- small rebases of patches 11-19
====================
Jiri Pirko [Fri, 4 Aug 2017 12:29:14 +0000 (14:29 +0200)]
net: sched: cls_u32: no need to call tcf_exts_change for newly allocated struct
As the n struct was allocated right before u32_set_parms call,
no need to use tcf_exts_change to do atomic change, and we can just
fill-up the unused exts struct directly by tcf_exts_validate.
Jiri Pirko [Fri, 4 Aug 2017 12:29:13 +0000 (14:29 +0200)]
net: sched: cls_route: no need to call tcf_exts_change for newly allocated struct
As the f struct was allocated right before route4_set_parms call,
no need to use tcf_exts_change to do atomic change, and we can just
fill-up the unused exts struct directly by tcf_exts_validate.
Jiri Pirko [Fri, 4 Aug 2017 12:29:12 +0000 (14:29 +0200)]
net: sched: cls_flow: no need to call tcf_exts_change for newly allocated struct
As the fnew struct just was allocated, so no need to use tcf_exts_change
to do atomic change, and we can just fill-up the unused exts struct
directly by tcf_exts_validate.
Jiri Pirko [Fri, 4 Aug 2017 12:29:11 +0000 (14:29 +0200)]
net: sched: cls_cgroup: no need to call tcf_exts_change for newly allocated struct
As the new struct just was allocated, so no need to use tcf_exts_change
to do atomic change, and we can just fill-up the unused exts struct
directly by tcf_exts_validate.
Jiri Pirko [Fri, 4 Aug 2017 12:29:10 +0000 (14:29 +0200)]
net: sched: cls_bpf: no need to call tcf_exts_change for newly allocated struct
As the prog struct was allocated right before cls_bpf_set_parms call,
no need to use tcf_exts_change to do atomic change, and we can just
fill-up the unused exts struct directly by tcf_exts_validate.
Jiri Pirko [Fri, 4 Aug 2017 12:29:09 +0000 (14:29 +0200)]
net: sched: cls_basic: no need to call tcf_exts_change for newly allocated struct
As the f struct was allocated right before basic_set_parms call, no need
to use tcf_exts_change to do atomic change, and we can just fill-up
the unused exts struct directly by tcf_exts_validate.
Jiri Pirko [Fri, 4 Aug 2017 12:29:08 +0000 (14:29 +0200)]
net: sched: cls_matchall: no need to call tcf_exts_change for newly allocated struct
As the head struct was allocated right before mall_set_parms call,
no need to use tcf_exts_change to do atomic change, and we can just
fill-up the unused exts struct directly by tcf_exts_validate.
Jiri Pirko [Fri, 4 Aug 2017 12:29:07 +0000 (14:29 +0200)]
net: sched: cls_fw: no need to call tcf_exts_change for newly allocated struct
As the f struct was allocated right before fw_set_parms call, no need
to use tcf_exts_change to do atomic change, and we can just fill-up
the unused exts struct directly by tcf_exts_validate.
Jiri Pirko [Fri, 4 Aug 2017 12:29:06 +0000 (14:29 +0200)]
net: sched: cls_flower: no need to call tcf_exts_change for newly allocated struct
As the f struct was allocated right before fl_set_parms call, no need
to use tcf_exts_change to do atomic change, and we can just fill-up
the unused exts struct directly by tcf_exts_validate.
Lin Yun Sheng [Fri, 4 Aug 2017 09:24:59 +0000 (17:24 +0800)]
net: hns: Fix for __udivdi3 compiler error
This patch fixes the __udivdi3 undefined error reported by
test robot.
Fixes: b8c17f708831 ("net: hns: Add self-adaptive interrupt coalesce support in hns driver") Signed-off-by: Yunsheng Lin <[email protected]> Signed-off-by: David S. Miller <[email protected]>
- block interrupts properly across the entire MMU context change (both
the hw MMU context change and the TSB table change) so that we don't
get a perf event interrupt in the middle. From Rob Gardner.
- be sure to register hugepages early enough, from Nitin Gupta.
- UltraSPARC-III user copy exception handling would return garbage for
the copied length in some circumstances.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc:
sparc64: Fix exception handling in UltraSPARC-III memcpy.
sbus: Convert to using %pOF instead of full_name
sparc: defconfig: Cleanup from old Kconfig options
sparc64: Register hugepages during arch init
sparc64: Prevent perf from running during super critical sections
Linus Torvalds [Fri, 4 Aug 2017 17:15:11 +0000 (10:15 -0700)]
Merge tag 'ceph-for-4.13-rc4' of git://github.com/ceph/ceph-client
Pull ceph fixes from Ilya Dryomov:
"A bunch of fixes and follow-ups for -rc1 Luminous patches: issues with
->reencode_message() and last minute RADOS semantic changes in
v12.1.2"
* tag 'ceph-for-4.13-rc4' of git://github.com/ceph/ceph-client:
libceph: make RECOVERY_DELETES feature create a new interval
libceph: upmap semantic changes
crush: assume weight_set != null imples weight_set_size > 0
libceph: fallback for when there isn't a pool-specific choose_arg
libceph: don't call ->reencode_message() more than once per message
libceph: make encode_request_*() work with r_mempool requests
Linus Torvalds [Fri, 4 Aug 2017 17:11:13 +0000 (10:11 -0700)]
Merge tag 'sound-4.13-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Pull sound fixes from Takashi Iwai:
"Now we hit the usual ASoC-fix-flood in the middle of release.
Most of the changes are trivial and device-specific, while one
significant change is the fix for unbalanced of_graph_*() refcounts.
This involved a change in the graph API itself that had been a bit
messy"
* tag 'sound-4.13-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ALSA: hda - Fix speaker output from VAIO VPCL14M1R
device property: Fix usecount for of_graph_get_port_parent()
ASoC: rt5665: fix wrong register for bclk ratio control
ASoC: Intel: Use MCLK instead of BLCK as the sysclock for RT5514 codec on kabylake platform
ASoC: Intel: Enabling ASRC for RT5663 codec on kabylake platform
ASoC: codecs: msm8916-analog: fix DIG_CLK_CTL_RXD3_CLK_EN define
ASoC: Intel: Skylake: Fix missing sentinels in sst_acpi_mach
ASoC: sh: hac: add missing "int ret"
ASoC: samsung: odroid: Fix EPLL frequency values
ASoC: sgtl5000: Use snd_soc_kcontrol_codec()
ASoC: rt5665: fix GPIO6 pin function define
ASoC: ux500: Restore platform DAI assignments
ASoC: fix pcm-creation regression
ASoC: do not close shared backend dailink
ASoC: pxa: SND_PXA2XX_SOC should depend on HAS_DMA
ASoC: Intel: Skylake: Fix default dma_buffer_size
ASoC: rt5663: Update the HW default values based on the shipping version
ASoC: imx-ssi: add check on platform_get_irq return value
Linus Torvalds [Fri, 4 Aug 2017 17:05:29 +0000 (10:05 -0700)]
Merge tag 'iommu-fixes-v4.13-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu
Pull IOMMU fixes from Joerg Roedel:
- fix a scheduling-while-atomic bug in the AMD IOMMU driver. It was
found after the checker was enabled earlier.
- a fix for the virtual APIC code in the AMD IOMMU driver which
delivers device interrupts directly into KVM guests for assigned
devices.
- fixes for the recently merged lock-less page-table code for ARM. The
redundant TLB syncs got reverted and locks added again around the TLB
sync code.
- fix for error handling in arm_smmu_add_device()
- address sanitization fix for arm io-pgtable code
* tag 'iommu-fixes-v4.13-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
iommu/amd: Fix schedule-while-atomic BUG in initialization code
iommu/amd: Enable ga_log_intr when enabling guest_mode
iommu/io-pgtable: Sanitise map/unmap addresses
iommu/arm-smmu: Fix the error path in arm_smmu_add_device
Revert "iommu/io-pgtable: Avoid redundant TLB syncs"
iommu/mtk: Avoid redundant TLB syncs locally
iommu/arm-smmu: Reintroduce locking around TLB sync operations
Linus Torvalds [Fri, 4 Aug 2017 17:02:56 +0000 (10:02 -0700)]
Merge tag 'mmc-v4.13-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc
Pull MMC fixes from Ulf Hansson:
"A couple of mmc fixes intended for v4.13-rc4.
MMC core:
- Fix NULL pointer dereference for block I/O during hotplug
MMC host:
- sdhci-of-at91: Fix card detect for non-removable cards"
* tag 'mmc-v4.13-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc:
mmc: block: bypass the queue even if usage is present for hotplug
mmc: sdhci-of-at91: force card detect value for non removable devices
Linus Torvalds [Fri, 4 Aug 2017 16:59:24 +0000 (09:59 -0700)]
Merge tag 'drm-fixes-for-v4.13-rc4' of git://people.freedesktop.org/~airlied/linux
Pull drm fixes from Dave Airlie:
"Either my email ate everything or everyone is on holidays, either way
all I can find is some lonely AMD fixes"
[ Europe might be on vacation, and the Pacific NW is too hot for work. ]
* tag 'drm-fixes-for-v4.13-rc4' of git://people.freedesktop.org/~airlied/linux:
drm/amdgpu: Use list_del_init in amdgpu_mn_unregister
drm/amdgpu: Fix undue fallthroughs in golden registers initialization
drm/amdgpu: fix header on gfx9 clear state
Linus Torvalds [Fri, 4 Aug 2017 16:56:54 +0000 (09:56 -0700)]
Merge tag 'powerpc-4.13-5' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
Pull powerpc fixes from Michael Ellerman:
"Fixes for recently merged code:
- a fix for the _PAGE_DEVMAP support, which was breaking KVM on
Power9 radix
- avoid a (harmless) lockdep warning in the early SMP code
- return failure for some uses of dma_set_mask() rather than falling
back to 32-bits
- fix stack setup in watchdog soft_nmi_common() to use emergency
stack
- fix of_irq_to_resource() error check in of_fsl_spi_probe()
Two fixes going to stable:
- fix saving of Transactional Memory SPRs in core dump
- fix __check_irq_replay missing decrementer interrupt
And two misc:
- fix 64-bit boot wrapper build with non-biarch compiler
- work around a POWER9 PMU hang after state-loss idle
Thanks to: Alistair Popple, Aneesh Kumar K.V, Cyril Bur, Gustavo
Romero, Jose Ricardo Ziviani, Laurent Vivier, Nicholas Piggin, Oliver
O'Halloran, Sergei Shtylyov, Suraj Jitindar Singh, Thomas Gleixner"
* tag 'powerpc-4.13-5' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
powerpc/64: Fix __check_irq_replay missing decrementer interrupt
powerpc/perf: POWER9 PMU stops after idle workaround
powerpc/83xx/mpc832x_rdb: fix of_irq_to_resource() error check
powerpc/64s: Fix stack setup in watchdog soft_nmi_common()
powerpc/powernv/pci: Return failure for some uses of dma_set_mask()
powerpc/boot: Fix 64-bit boot wrapper build with non-biarch compiler
powerpc/smp: Call smp_ops->setup_cpu() directly on the boot CPU
powerpc/tm: Fix saving of TM SPRs in core dump
powerpc/mm: Fix pmd/pte_devmap() on non-leaf entries
David S. Miller [Fri, 4 Aug 2017 16:47:52 +0000 (09:47 -0700)]
sparc64: Fix exception handling in UltraSPARC-III memcpy.
Mikael Pettersson reported that some test programs in the strace-4.18
testsuite cause an OOPS.
After some debugging it turns out that garbage values are returned
when an exception occurs, causing the fixup memset() to be run with
bogus arguments.
The problem is that two of the exception handler stubs write the
successfully copied length into the wrong register.
Fixes: ee841d0aff64 ("sparc64: Convert U3copy_{from,to}_user to accurate exception reporting.") Reported-by: Mikael Pettersson <[email protected]> Tested-by: Mikael Pettersson <[email protected]> Reviewed-by: Sam Ravnborg <[email protected]> Signed-off-by: David S. Miller <[email protected]>
It would be preferrable to use GENMASK_ULL() instead, but it's not set
up to be used from assembly (the UL() macro token pastes UL suffixes
when not included in assembly sources).
arm64: Fix potential race with hardware DBM in ptep_set_access_flags()
In a system with DBM (dirty bit management) capable agents there is a
possible race between a CPU executing ptep_set_access_flags() (maybe
non-DBM capable) and a hardware update of the dirty state (clearing of
PTE_RDONLY). The scenario:
a) the pte is writable (PTE_WRITE set), clean (PTE_RDONLY set) and old
(PTE_AF clear)
b) ptep_set_access_flags() is called as a result of a read access and it
needs to set the pte to writable, clean and young (PTE_AF set)
c) a DBM-capable agent, as a result of a different write access, is
marking the entry as young (setting PTE_AF) and dirty (clearing
PTE_RDONLY)
The current ptep_set_access_flags() implementation would set the
PTE_RDONLY bit in the resulting value overriding the DBM update and
losing the dirty state.
This patch fixes such race by setting PTE_RDONLY to the most permissive
(lowest value) of the current entry and the new one.
Arnd Bergmann [Fri, 4 Aug 2017 11:22:33 +0000 (13:22 +0200)]
Merge tag 'davinci-fixes-for-v4.13' of git://git.kernel.org/pub/scm/linux/kernel/git/nsekhar/linux-davinci into fixes
Pull "DaVinci fixes for v4.13" from Sekhar Nori:
Drop unused VPIF endpoints from device-tree.
They should be used only when an actual
remote-endpoint is connected.
* tag 'davinci-fixes-for-v4.13' of git://git.kernel.org/pub/scm/linux/kernel/git/nsekhar/linux-davinci:
ARM: dts: da850-lcdk: drop unused VPIF endpoints
ARM: dts: da850-evm: drop unused VPIF endpoints
Arnd Bergmann [Fri, 4 Aug 2017 11:04:42 +0000 (13:04 +0200)]
Merge tag 'sunxi-fixes-for-4.13' of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux into fixes
Pull "Allwinner fixes for 4.13" from Chen-Yu Tsai:
Two fixes to correct the EMAC blocks memory region size to match the
datasheet. One that converts raw A83T clock indices to macros from the
clk dt-binding header, completing the A83T sunxi-ng clk driver.
* tag 'sunxi-fixes-for-4.13' of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux:
ARM: dts: sun8i: a83t: Switch to CCU device tree binding macros
arm64: allwinner: sun50i-a64: Correct emac register size
ARM: dts: sunxi: h3/h5: Correct emac register size
Arnd Bergmann [Fri, 4 Aug 2017 11:03:24 +0000 (13:03 +0200)]
Merge tag 'qcom-arm64-defconfig-fixes-for-4.13-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/agross/linux into fixes
Pull "Qualcomm ARM64 based defconfig Fixes for v4.13-rc2" from Andy Gross:
* Enable missing HWSPINLOCK
* tag 'qcom-arm64-defconfig-fixes-for-4.13-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/agross/linux:
arm64: defconfig: enable missing HWSPINLOCK
Arnd Bergmann [Fri, 4 Aug 2017 10:54:41 +0000 (12:54 +0200)]
Merge tag 'renesas-fixes3-for-v4.13' of https://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas into fixes
Pull "Third Round of Renesas ARM Based SoC Fixes for v4.13" from Simon Horman:
Fix deadlock in regulator quirk for R-Car Gen 2 SoCs
The da9063/da9210 regulator quirk for R-Car Gen2 boards uses a bus
notifier, and unregisters the notifier when it is no longer needed.
However, a notifier must not be unregistered from within the call chain.
This bug went unnoticed, as blocking_notifier_chain_unregister() didn't
take the semaphore during early boot. This is no longer the case as of
upstream commit 1c3c5eab171590f8 ("sched/core: Enable might_sleep() and
smp_processor_id() checks early") and a deadlock occurs.
* tag 'renesas-fixes3-for-v4.13' of https://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas:
ARM: shmobile: rcar-gen2: Fix deadlock in regulator quirk
Arnd Bergmann [Fri, 4 Aug 2017 10:53:21 +0000 (12:53 +0200)]
Merge tag 'mvebu-fixes-4.13-2' of git://git.infradead.org/linux-mvebu into fixes
Pull "mvebu fixes for 4.13 (part 2)" from Gregory CLEMENT:
All the fixes are for ARM64 mvebu:
- Fix the RTC interrupt on A7K/A8K which was missed when switching
from GIC to ICU
- Mark the A7K/A8K crypto engine as dma coherent
- Fix the number of GPIO on south bridge on Armada 3700
* tag 'mvebu-fixes-4.13-2' of git://git.infradead.org/linux-mvebu:
ARM64: dts: marvell: armada-37xx: Fix the number of GPIO on south bridge
arm64: dts: marvell: mark the cp110 crypto engine as dma coherent
arm64: dts: marvell: use ICU for the CP110 slave RTC
Arnd Bergmann [Fri, 4 Aug 2017 10:50:52 +0000 (12:50 +0200)]
Merge tag 'amlogic-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/khilman/linux-amlogic into fixes
Pull "Amlogic fixes for v4.13-rc" from Kevin Hilman:
- 2 minor DT fixes
* tag 'amlogic-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/khilman/linux-amlogic:
ARM64: dts: meson-gxl-s905x-libretech-cc: fixup board definition
ARM64: dts: meson-gx: use specific compatible for the AO pwms
Arnd Bergmann [Fri, 4 Aug 2017 10:48:46 +0000 (12:48 +0200)]
Merge tag 'v4.13-rockchip-dts32fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip into fixes
Pull "Rockchip dts32 fixes for 4.13" from Heiko Stübner:
Fix for the recently added mali dt support. The example
showed a wrong value, so fix it before it gets copy-pasted
to much.
* tag 'v4.13-rockchip-dts32fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip:
ARM: dts: rockchip: fix mali gpu node on rk3288
dt-bindings: gpu: drop wrong compatible from midgard binding example
David S. Miller [Fri, 4 Aug 2017 04:37:30 +0000 (21:37 -0700)]
Merge branch 'socket-sendmsg-zerocopy'
Willem de Bruijn says:
====================
socket sendmsg MSG_ZEROCOPY
Introduce zerocopy socket send flag MSG_ZEROCOPY. This extends the
shared page support (SKBTX_SHARED_FRAG) from sendpage to sendmsg.
Implement the feature for TCP initially, as large writes benefit
most.
On a send call with MSG_ZEROCOPY, the kernel pins user pages and
links these directly into the skbuff frags[] array.
Each send call with MSG_ZEROCOPY that transmits data will eventually
queue a completion notification on the error queue: a per-socket u32
incremented on each such call. A request may have to revert to copy
to succeed, for instance when a device cannot support scatter-gather
IO. In that case a flag is passed along to notify that the operation
succeeded without zerocopy optimization.
The implementation extends the existing zerocopy infra for tuntap,
vhost and xen with features needed for TCP, notably reference
counting to handle cloning on retransmit and GSO.
For more details, see also the netdev 2.1 paper and presentation at
https://netdevconf.org/2.1/session.html?debruijn
Changelog:
v3 -> v4:
- dropped UDP, RAW and PF_PACKET for now
Without loopback support, datagrams are usually smaller than
the ~8KB size threshold needed to benefit from zerocopy.
- style: a few reverse chrismas tree
- minor: SO_ZEROCOPY returns ENOTSUPP on unsupported protocols
- minor: squashed SO_EE_CODE_ZEROCOPY_COPIED patch
- minor: rebased on top of net-next with kmap_atomic fix
v1 -> v2:
- fix (kbuild-bot): do not remove uarg until patch 5
- fix (kbuild-bot): move zerocopy_sg_from_iter doc with function
- fix: remove unused extern in header file
RFCv2 -> v1:
- patch 2
- review comment: in skb_copy_ubufs, always allocate order-0
page, also when replacing compound source pages.
- patch 3
- fix: always queue completion notification on MSG_ZEROCOPY,
also if revert to copy.
- fix: on syscall abort, correctly revert notification state
- minor: skip queue notification on SOCK_DEAD
- minor: replace BUG_ON with WARN_ON in recoverable error
- patch 4
- new: add socket option SOCK_ZEROCOPY.
only honor MSG_ZEROCOPY if set, ignore for legacy apps.
- patch 5
- fix: clear zerocopy state on skb_linearize
- patch 6
- fix: only coalesce if prev errqueue elem is zerocopy
- minor: try coalescing with list tail instead of head
- minor: merge bytelen limit patch
- patch 7
- new: signal when data had to be copied
- patch 8 (tcp)
- optimize: avoid setting PSH bit when exceeding max frags.
that limits GRO on the client. do not goto new_segment.
- fix: fail on MSG_ZEROCOPY | MSG_FASTOPEN
- minor: do not wait for memory: does not work for optmem
- minor: simplify alloc
- patch 9 (udp)
- new: add PF_INET6
- fix: attach zerocopy notification even if revert to copy
- minor: simplify alloc size arithmetic
- patch 10 (raw hdrinc)
- new: add PF_INET6
- patch 11 (pf_packet)
- minor: simplify slightly
- patch 12
- new msg_zerocopy regression test: use veth pair to test
all protocols: ipv4/ipv6/packet, tcp/udp/raw, cork
all relevant ethtool settings: rx off, sg off
all relevant packet lengths: 0, <MAX_HEADER, max size
RFC -> RFCv2:
- review comment: do not loop skb with zerocopy frags onto rx:
add skb_orphan_frags_rx to orphan even refcounted frags
call this in __netif_receive_skb_core, deliver_skb and tun:
same as commit 1080e512d44d ("net: orphan frags on receive")
- fix: hold an explicit sk reference on each notification skb.
previously relied on the reference (or wmem) held by the
data skb that would trigger notification, but this breaks
on skb_orphan.
- fix: when aborting a send, do not inc the zerocopy counter
this caused gaps in the notification chain
- fix: in packet with SOCK_DGRAM, pull ll headers before calling
zerocopy_sg_from_iter
- fix: if sock_zerocopy_realloc does not allow coalescing,
do not fail, just allocate a new ubuf
- fix: in tcp, check return value of second allocation attempt
- chg: allocate notification skbs from optmem
to avoid affecting tcp write queue accounting (TSQ)
- chg: limit #locked pages (ulimit) per user instead of per process
- chg: grow notification ids from 16 to 32 bit
- pass range [lo, hi] through 32 bit fields ee_info and ee_data
- chg: rebased to davem-net-next on top of v4.10-rc7
- add: limit notification coalescing
sharing ubufs limits overhead, but delays notification until
the last packet is released, possibly unbounded. Add a cap.
- tests: add snd_zerocopy_lo pf_packet test
- tests: two bugfixes (add do_flush_tcp, ++sent not only in debug)
Limitations / Known Issues:
- TCP may build slightly smaller than max TSO packets due to
exceeding MAX_SKB_FRAGS frags when zerocopy pages are unaligned.
- All SKBTX_SHARED_FRAG may require additional __skb_linearize or
skb_copy_ubufs calls in u32, skb_find_text, similar to
skb_checksum_help.
Notification skbuffs are allocated from optmem. For sockets that
cannot effectively coalesce notifications, the optmem max may need
to be increased to avoid hitting -ENOBUFS:
In application load, copy avoidance shows a roughly 5% systemwide
reduction in cycles when streaming large flows and a 4-8% reduction in
wall clock time on early tensorflow test workloads.
For the single-machine veth tests to succeed, loopback support has to
be temporarily enabled by making skb_orphan_frags_rx map to
skb_orphan_frags.
* Performance
The below table shows cycles reported by perf for a netperf process
sending a single 10 Gbps TCP_STREAM. The first three columns show
Mcycles spent in the netperf process context. The second three columns
show time spent systemwide (-a -C A,B) on the two cpus that run the
process and interrupt handler. Reported is the median of at least 3
runs. std is a standard netperf, zc uses zerocopy and % is the ratio.
Netperf is pinned to cpu 2, network interrupts to cpu3, rps and rfs
are disabled and the kernel is booted with idle=halt.
The number of pages that can be pinned on behalf of a user with
MSG_ZEROCOPY is bound by the locked memory ulimit.
While the kernel holds process memory pinned, a process cannot safely
reuse those pages for other purposes. Packets looped onto the receive
stack and queued to a socket can be held indefinitely. Avoid unbounded
notification latency by restricting user pages to egress paths only.
skb_orphan_frags_rx() will create a private copy of pages even for
refcounted packets when these are looped, as did skb_orphan_frags for
the original tun zerocopy implementation.
Pages are not remapped read-only. Processes can modify packet contents
while packets are in flight in the kernel path. Bytes on which kernel
control flow depends (headers) are copied to avoid TOCTTOU attacks.
Datapath integrity does not otherwise depend on payload, with three
exceptions: checksums, optional sk_filter/tc u32/.. and device +
driver logic. The effect of wrong checksums is limited to the
misbehaving process. TC filters that access contents may have to be
excluded by adding an skb_orphan_frags_rx.
Processes can also safely avoid OOM conditions by bounding the number
of bytes passed with MSG_ZEROCOPY and by removing shared pages after
transmission from their own memory map.
====================
Introduce regression test for msg_zerocopy feature. Send traffic from
one process to another with and without zerocopy.
Evaluate tcp, udp, raw and packet sockets, including variants
- udp: corking and corking with mixed copy/zerocopy calls
- raw: with and without hdrincl
- packet: at both raw and dgram level
Test on both ipv4 and ipv6, optionally with ethtool changes to
disable scatter-gather, tx checksum or tso offload. All of these
can affect zerocopy behavior.
The regression test can be run on a single machine if over a veth
pair. Then skb_orphan_frags_rx must be modified to be identical to
skb_orphan_frags to allow forwarding zerocopy locally.
The msg_zerocopy.sh script will setup the veth pair in network
namespaces and run all tests.
Enable support for MSG_ZEROCOPY to the TCP stack. TSO and GSO are
both supported. Only data sent to remote destinations is sent without
copying. Packets looped onto a local destination have their payload
copied to avoid unbounded latency.
Tested:
A 10x TCP_STREAM between two hosts showed a reduction in netserver
process cycles by up to 70%, depending on packet size. Systemwide,
savings are of course much less pronounced, at up to 20% best case.
msg_zerocopy.sh 4 tcp:
without zerocopy
tx=121792 (7600 MB) txc=0 zc=n
rx=60458 (7600 MB)
with zerocopy
tx=286257 (17863 MB) txc=286257 zc=y
rx=140022 (17863 MB)
This test opens a pair of sockets over veth, one one calls send with
64KB and optionally MSG_ZEROCOPY and on the other reads the initial
bytes. The receiver truncates, so this is strictly an upper bound on
what is achievable. It is more representative of sending data out of
a physical NIC (when payload is not touched, either).
In the simple case, each sendmsg() call generates data and eventually
a zerocopy ready notification N, where N indicates the Nth successful
invocation of sendmsg() with the MSG_ZEROCOPY flag on this socket.
TCP and corked sockets can cause send() calls to append new data to an
existing sk_buff and, thus, ubuf_info. In that case the notification
must hold a range. odify ubuf_info to store a inclusive range [N..N+m]
and add skb_zerocopy_realloc() to optionally extend an existing range.
Also coalesce notifications in this common case: if a notification
[1, 1] is about to be queued while [0, 0] is the queue tail, just modify
the head of the queue to read [0, 1].
Coalescing is limited to a few TSO frames worth of data to bound
notification latency.
Prepare the datapath for refcounted ubuf_info. Clone ubuf_info with
skb_zerocopy_clone() wherever needed due to skb split, merge, resize
or clone.
Split skb_orphan_frags into two variants. The split, merge, .. paths
support reference counted zerocopy buffers, so do not do a deep copy.
Add skb_orphan_frags_rx for paths that may loop packets to receive
sockets. That is not allowed, as it may cause unbounded latency.
Deep copy all zerocopy copy buffers, ref-counted or not, in this path.
The exact locations to modify were chosen by exhaustively searching
through all code that might modify skb_frag references and/or the
the SKBTX_DEV_ZEROCOPY tx_flags bit.
The changes err on the safe side, in two ways.
(1) legacy ubuf_info paths virtio and tap are not modified. They keep
a 1:1 ubuf_info to sk_buff relationship. Calls to skb_orphan_frags
still call skb_copy_ubufs and thus copy frags in this case.
(2) not all copies deep in the stack are addressed yet. skb_shift,
skb_split and skb_try_coalesce can be refined to avoid copying.
These are not in the hot path and this patch is hairy enough as
is, so that is left for future refinement.
The send call ignores unknown flags. Legacy applications may already
unwittingly pass MSG_ZEROCOPY. Continue to ignore this flag unless a
socket opts in to zerocopy.
Introduce socket option SO_ZEROCOPY to enable MSG_ZEROCOPY processing.
Processes can also query this socket option to detect kernel support
for the feature. Older kernels will return ENOPROTOOPT.
The kernel supports zerocopy sendmsg in virtio and tap. Expand the
infrastructure to support other socket types. Introduce a completion
notification channel over the socket error queue. Notifications are
returned with ee_origin SO_EE_ORIGIN_ZEROCOPY. ee_errno is 0 to avoid
blocking the send/recv path on receiving notifications.
Add reference counting, to support the skb split, merge, resize and
clone operations possible with SOCK_STREAM and other socket types.
Refine skb_copy_ubufs to support compound pages. With upcoming TCP
zerocopy sendmsg, such fragments may appear.
The existing code replaces each page one for one. Splitting each
compound page into an independent number of regular pages can result
in exceeding limit MAX_SKB_FRAGS if data is not exactly page aligned.
Instead, fill all destination pages but the last to PAGE_SIZE.
Split the existing alloc + copy loop into separate stages:
1. compute bytelength and minimum number of pages to store this.
2. allocate
3. copy, filling each page except the last to PAGE_SIZE bytes
4. update skb frag array
Add sock_omalloc and sock_ofree to be able to allocate control skbs,
for instance for looping errors onto sk_error_queue.
The transmit budget (sk_wmem_alloc) is involved in transmit skb
shaping, most notably in TCP Small Queues. Using this budget for
control packets would impact transmission.
If the decrementer wraps again and de-asserts the decrementer
exception while hard-disabled, __check_irq_replay() has a test to
notice the wrap when interrupts are re-enabled.
The decrementer check must be done when clearing the PACA_IRQ_HARD_DIS
flag, not when the PACA_IRQ_DEC flag is tested. Previously this worked
because the decrementer interrupt was always the first one checked
after clearing the hard disable flag, but HMI check was moved ahead of
that, which introduced this bug.
This can cause a missed decrementer interrupt if we soft-disable
interrupts then take an HMI which is recorded in irq_happened, then
hard-disable interrupts for > 4s to wrap the decrementer.
Dave Airlie [Fri, 4 Aug 2017 01:43:14 +0000 (11:43 +1000)]
Merge branch 'drm-fixes-4.13' of git://people.freedesktop.org/~agd5f/linux into drm-fixes
Just a few small fixes for 4.13.
* 'drm-fixes-4.13' of git://people.freedesktop.org/~agd5f/linux:
drm/amdgpu: Use list_del_init in amdgpu_mn_unregister
drm/amdgpu: Fix undue fallthroughs in golden registers initialization
drm/amdgpu: fix header on gfx9 clear state
David S. Miller [Thu, 3 Aug 2017 22:36:01 +0000 (15:36 -0700)]
Merge branch 'mlxsw-Support-for-IPv6-UC-router'
Jiri Pirko says:
====================
mlxsw: Support for IPv6 UC router
Ido says:
This set adds support for IPv6 unicast routes offload. The first four
patches make the FIB notification chain generic so that it could be used
by address families other than IPv4. This is done by having each address
family register its callbacks with the common code, so that its FIB tables
and rules could be dumped upon registration to the chain, while ensuring
the integrity of the dump. The exact mechanics are explained in detail in
the first patch.
The next six patches build upon this work and add the necessary callbacks
in IPv6 code. This allows listeners of the chain to receive notifications
about IPv6 routes addition, deletion and replacement as well as FIB rules
notifications.
Unlike user space notifications for IPv6 multipath routes, the FIB
notification chain notifies these on a per-nexthop basis. This allows
us to keep the common code lean and is also unnecessary, as notifications
are serialized by each table's lock whereas applications maintaining
netlink caches may suffer from concurrent dumps and deletions / additions
of routes.
The next five patches audit the different code paths reading the route's
reference count (rt6i_ref) and remove assumptions regarding its meaning.
This is needed since non-FIB users need to be able to hold a reference on
the route and a non-zero reference count no longer means the route is in
the FIB.
The last six patches enable the mlxsw driver to offload IPv6 unicast
routes to the Spectrum ASIC. Without resorting to ACLs, lookup is done
solely based on the destination IP, so the abort mechanism is invoked
upon the addition of source-specific routes.
Follow-up patch sets will increase the scale of gatewayed routes by
consolidating identical nexthop groups to one adjacency entry in the
device's adjacency table (as in IPv4), as well as add support for
NH_{ADD,DEL} events which enable support for the
'ignore_routes_with_linkdown' sysctl.
Changes in v2:
* Provide offload indication for individual nexthops (David Ahern).
* Use existing route reference count instead of adding another one.
This resulted in several new patches to remove assumptions regarding
current semantics of the existing reference count (David Ahern).
* Add helpers to allow non-FIB users to take a reference on route.
* Remove use of tb6_lock in mlxsw (David Ahern).
* Add IPv6 dependency to mlxsw.
====================
projects / linux.git / log