- 19 Jan, 2017 11 commits
-
-
Rolf Eike Beer authored
commit 3659f98b upstream. Nothing in this minimal script seems to require bash. We often run these tests on embedded devices where the only shell available is the busybox ash. Use sh instead. Signed-off-by:
Rolf Eike Beer <eb@emlix.com> Signed-off-by:
Shuah Khan <shuahkh@osg.samsung.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Dan Carpenter authored
commit f7741aa7 upstream. A recent cleanup changed the kmalloc() + copy_from_user() to memdup_user() but the error handling wasn't updated so we might call kfree(-EFAULT) and crash. Fixes: a6e3918b ('GPU-DRM-Savage: Use memdup_user() rather than duplicating') Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161012062227.GU12841@mwandaSigned-off-by:
-
Dan Carpenter authored
commit b2cdeb19 upstream. If the allocation fails the current code returns success. If copy_from_user() fails it returns the number of bytes remaining instead of -EFAULT. Fixes: d5b1a78a ("drm/vc4: Add support for drawing 3D frames.") Signed-off-by:
Eric Anholt <eric@anholt.net> Signed-off-by:
-
Christophe Jaillet authored
commit 9376cad2 upstream. The devm_pinctrl_register() function returns an error pointer or a valid handle. So checking for NULL here is pointless and can never trigger. Check the returned value with IS_ERR instead and propagate this value as done in the other functions which call devm_pinctrl_register(). Fixes: 0751bb5c ("drm/tegra: dpaux: Add pinctrl support") Signed-off-by:
Christophe JAILLET <christophe.jaillet@wanadoo.fr> Acked-by:
Jon Hunter <jonathanh@nvidia.com> Signed-off-by:
Thierry Reding <treding@nvidia.com> Signed-off-by:
-
Chen-Yu Tsai authored
commit 618c8089 upstream. The maximum supported voltage for ldo_io# is 3.3V, but on cold boot the selector comes up at 0x1f, which maps to 3.8V. This was previously corrected by Allwinner's U-boot, which set all regulators on the PMICs to some pre-configured voltage. With recent progress in U-boot SPL support, this is no longer the case. In any case we should handle this quirk in the kernel driver as well. This invalid setting causes _regulator_get_voltage() to fail with -EINVAL which causes regulator registration to fail when constrains are used: [ 1.054181] vcc-pg: failed to get the current voltage(-22) [ 1.059670] axp20x-regulator axp20x-regulator.0: Failed to register ldo_io0 [ 1.069749] axp20x-regulator: probe of axp20x-regulator.0 failed with error -22 This commits makes the axp20x regulator driver accept the 0x1f register value, fixing this. The datasheet does not guarantee reliable operation above 3.3V, so on boards where this regulator is used the regulator-max-microvolt setting must be 3.3V or less. This is essentially the same as the commit f40d4896 ("regulator: axp20x: Fix axp22x ldo_io registration error on cold boot") for AXP22x PMICs. Fixes: a51f9f46 ("regulator: axp20x: support AXP809 variant") Signed-off-by:
Chen-Yu Tsai <wens@csie.org> Signed-off-by:
Mark Brown <broonie@kernel.org> Signed-off-by:
-
Andrew F. Davis authored
commit d8ca5bd1 upstream. The BUCK regulators 3, 4, and 5 also have a 10mV step mode, adjust the tables and logic to reflect the data-sheet for these regulators. fixes: d2a2e729 ("regulator: tps65086: Add regulator driver for the TPS65086 PMIC") Signed-off-by:
Andrew F. Davis <afd@ti.com> Signed-off-by:
-
Niklas Söderlund authored
commit c314c9f1 upstream. On some SoC there are no simple mapping of pins to bias register bits and a lookup table is needed. This logic is already implemented in some SoC specific drivers that could benefit from a generic implementation. Add helpers to deal with the lookup which later can be used by the SoC specific drivers. The logic used to lookup are different from the one it aims to replace, this is intentional. This new method reduces the memory consumption at the cost of increased CPU usage and fix a bug where a WARN() would incorrectly be triggered if the register offset is 0. Signed-off-by:
Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se> Reviewed-by:
Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by:
Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by:
-
Niklas Söderlund authored
commit d3b861bc upstream. There is a bug in the r8a7795 bias code where a WARN() is trigged anytime a pin from PUEN0/PUD0 is accessed. # cat /sys/kernel/debug/pinctrl/e6060000.pfc/pinconf-pins WARNING: CPU: 2 PID: 2391 at drivers/pinctrl/sh-pfc/pfc-r8a7795.c:5364 r8a7795_pinmux_get_bias+0xbc/0xc8 [..] Call trace: [<ffff0000083c442c>] r8a7795_pinmux_get_bias+0xbc/0xc8 [<ffff0000083c37f4>] sh_pfc_pinconf_get+0x194/0x270 [<ffff0000083b0768>] pin_config_get_for_pin+0x20/0x30 [<ffff0000083b11e8>] pinconf_generic_dump_one+0x168/0x188 [<ffff0000083b144c>] pinconf_generic_dump_pins+0x5c/0x98 [<ffff0000083b0628>] pinconf_pins_show+0xc8/0x128 [<ffff0000081fe3bc>] seq_read+0x16c/0x420 [<ffff00000831a110>] full_proxy_read+0x58/0x88 [<ffff0000081d7ad4>] __vfs_read+0x1c/0xf8 [<ffff0000081d8874>] vfs_read+0x84/0x148 [<ffff0000081d9d64>] SyS_read+0x44/0xa0 [<ffff000008082f4c>] __sys_trace_return+0x0/0x4 This is due to the WARN() check if the reg field of the pullups struct is zero, and this should be 0 for pins controlled by the PUEN0/PUD0 registers since PU0 is defined as 0. Change the data structure and use the generic sh_pfc_pin_to_bias_info() function to get the register offset and bit information. Fixes: 56065524 ("pinctrl: sh-pfc: r8a7795: Add bias pinconf support") Signed-off-by:
-
Gary Bisson authored
commit 8f5983ad upstream. Fixes: 6e408ed8 ("pinctrl: imx: fix initialization of imx_pinctrl_desc") Reviewed-by:
Vladimir Zapolskiy <vladimir_zapolskiy@mentor.com> Reviewed-by:
Peng Fan <peng.fan@nxp.com> Signed-off-by:
Gary Bisson <gary.bisson@boundarydevices.com> Signed-off-by:
Linus Walleij <linus.walleij@linaro.org>
-
Marcos Paulo de Souza authored
commit 41c567a5 upstream. Avoid AUX loopback in Pegatron C15B touchpad, so input subsystem is able to recognize a Synaptics touchpad in the AUX port. Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=93791 (Touchpad is not detected on DNS 0801480 notebook (PEGATRON C15B)) Suggested-by:
Dmitry Torokhov <dmitry.torokhov@gmail.com> Signed-off-by:
Marcos Paulo de Souza <marcos.souza.org@gmail.com> Signed-off-by:
-
Pavel Rojtberg authored
commit b6fc513d upstream. currently the controllers get the same product id as the wireless receiver. However the controllers actually have their own product id. The patch makes the driver expose the same product id as the windows driver. This improves compatibility when running applications with WINE. see https://github.com/paroj/xpad/issues/54Signed-off-by:
Pavel Rojtberg <rojtberg@gmail.com> Signed-off-by:
-
- 15 Jan, 2017 29 commits
-
-
Greg Kroah-Hartman authored
-
Larry Finger authored
commit 60f59ce0 upstream. These drivers need to be able to reference "struct ieee80211_hw" from the driver's private data, and vice versa. The USB driver failed to store the address of ieee80211_hw in the private data. Although this bug has been present for a long time, it was not exposed until commit ba9f93f8 ("rtlwifi: Fix enter/exit power_save"). Fixes: ba9f93f8 ("rtlwifi: Fix enter/exit power_save") Signed-off-by:
Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by:
Kalle Valo <kvalo@codeaurora.org> Signed-off-by:
-
Larry Finger authored
commit ba9f93f8 upstream. In commit a5ffbe0a ("rtlwifi: Fix scheduling while atomic bug") and commit a269913c ("rtlwifi: Rework rtl_lps_leave() and rtl_lps_enter() to use work queue"), an error was introduced in the power-save routines due to the fact that leaving PS was delayed by the use of a work queue. This problem is fixed by detecting if the enter or leave routines are in interrupt mode. If so, the workqueue is used to place the request. If in normal mode, the enter or leave routines are called directly. Fixes: a269913c ("rtlwifi: Rework rtl_lps_leave() and rtl_lps_enter() to use work queue") Reported-by:
Ping-Ke Shih <pkshih@realtek.com> Signed-off-by:
-
Imre Deak authored
commit 2c7d0602 upstream. commit 848496e5 Author: Ville Syrjälä <ville.syrjala@linux.intel.com> Date: Wed Jul 13 16:32:03 2016 +0300 drm/i915: Wait up to 3ms for the pcu to ack the cdclk change request on SKL increased the timeout to match the spec, but we still see a timeout on at least one SKL. A CDCLK change request following the failed one will succeed nevertheless. I could reproduce this problem easily by running kms_pipe_crc_basic in a loop. In all failure cases _wait_for() was pre-empted for >3ms and so in the worst case - when the pre-emption happened right after calculating timeout__ in _wait_for() - we called skl_cdclk_wait_for_pcu_ready() only once which failed and so _wait_for() timed out. As opposed to this the spec says to keep retrying the request for at most a 3ms period. To fix this send the first request explicitly to guarantee that there is 3ms between the first and last request. Though this matches the spec, I noticed that in rare cases this can still time out if we sent only a few requests (in the worst case 2) _and_ PCODE is busy for some reason even after a previous request and a 3ms delay. To work around this retry the polling with pre-emption disabled to maximize the number of requests. Also increase the timeout to 10ms to account for interrupts that could reduce the number of requests. With this change I couldn't trigger the problem. v2: - Use 1ms poll period instead of 10us. (Chris) v3: - Poll with pre-emption disabled to increase the number of request attempts. (Ville, Chris) - Factor out a helper to poll, it's also needed by the next patch. v4: - Pass reply_mask, reply to skl_pcode_request(), instead of assuming the reply is generic. (Ville) v5: - List the request specific timeout values as code comment. (Ville) v6: - Try the poll first with preemption enabled. - Add code comment about first request being queued by PCODE. (Art) - Add timeout_base_ms argument. (Ville) v7: - Clarify code comment about first queued request. (Chris) Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Art Runyan <arthur.j.runyan@intel.com> Cc: <stable@vger.kernel.org> # v4.2- : 3b2c1710 : drm/i915: Wait up to 3ms Cc: <stable@vger.kernel.org> # v4.2- Fixes: 5d96d8af ("drm/i915/skl: Deinit/init the display at suspend/resume") Reference: https://bugs.freedesktop.org/show_bug.cgi?id=97929 Testcase: igt/kms_pipe_crc_basic/suspend-read-crc-pipe-B Signed-off-by:
Imre Deak <imre.deak@intel.com> Reviewed-by:
Chris Wilson <chris@chris-wilson.co.uk> Link: http://patchwork.freedesktop.org/patch/msgid/1480955258-26311-1-git-send-email-imre.deak@intel.com (cherry picked from commit a0b8a1fe) Signed-off-by:
Jani Nikula <jani.nikula@intel.com> Signed-off-by:
-
Dennis Kadioglu authored
commit 2e40795c upstream. Plantronics BT600 does not support reading the sample rate which leads to many lines of "cannot get freq at ep 0x1" and "cannot get freq at ep 0x82". This patch adds the USB ID of the BT600 to quirks.c and avoids those error messages. Signed-off-by:
Dennis Kadioglu <denk@post.com> Signed-off-by:
Takashi Iwai <tiwai@suse.de> Signed-off-by:
-
Uwe Kleine-König authored
commit 7243e0b2 upstream. The calculation of SPR and SPPR doesn't round correctly at several places which might result in baud rates that are too big. For example with tclk_hz = 250000001 and target rate 25000000 it determined a divider of 10 which is wrong. Instead of fixing all the corner cases replace the calculation by an algorithm without a loop which should even be quicker to execute apart from being correct. Fixes: df59fa7f ("spi: orion: support armada extended baud rates") Signed-off-by:
Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Signed-off-by:
-
Grygorii Strashko authored
commit f86a2c87 upstream. The commit 55ee7017 ("arm: omap2: board-generic: use omap4_local_timer_init for AM437x") unintentionally changes the clocksource devices for AM437x from OMAP GP Timer to SyncTimer32K. Unfortunately, the SyncTimer32K is starving from frequency deviation as mentioned in commit 5b5c0135 ("ARM: OMAP2+: AM43x: Use gptimer as clocksource") and, as reported by Franklin [1], even its monotonic nature is under question (most probably there is a HW issue, but it's still under investigation). Taking into account above facts It's reasonable to rollback to the use of omap3_gptimer_timer_init(). [1] http://www.spinics.net/lists/linux-omap/msg127425.html Fixes: 55ee7017 ("arm: omap2: board-generic: use omap4_local_timer_init for AM437x") Reported-by:
Cooper Jr., Franklin <fcooper@ti.com> Signed-off-by:
Grygorii Strashko <grygorii.strashko@ti.com> Signed-off-by:
Lokesh Vutla <lokeshvutla@ti.com> Signed-off-by:
Keerthy <j-keerthy@ti.com> Signed-off-by:
Tony Lindgren <tony@atomide.com> Signed-off-by:
-
Geert Uytterhoeven authored
commit 9388093d upstream. Unlike clk_register_clkdev(), clk_hw_register_clkdev() doesn't check for passed error objects from a previous registration call. Hence the caller of clk_hw_register_*() has to check for errors before calling clk_hw_register_clkdev*(). Make clk_hw_register_clkdev() more similar to clk_register_clkdev() by adding this error check, removing the burden from callers that do mass registration. Fixes: e4f1b49b ("clkdev: Add clk_hw based registration APIs") Fixes: 944b9a41 ("clk: ls1x: Migrate to clk_hw based OF and registration APIs") Fixes: 44ce9a9a ("MIPS: TXx9: Convert to Common Clock Framework") Fixes: f48d947a ("clk: clps711x: Migrate to clk_hw based OF and registration APIs") Fixes: b4626a7f ("CLK: Add Loongson1C clock support") Signed-off-by:
Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by:
Russell King <rmk+kernel@arm.linux.org.uk> Signed-off-by:
-
Tony Lindgren authored
commit cbf26428 upstream. We don't want to fall through to a bunch of errors for retention if PM_OMAP4_CPU_OSWR_DISABLE is not configured for a SoC. Fixes: 6099dd37 ("ARM: OMAP5 / DRA7: Enable CPU RET on suspend") Acked-by:
Santosh Shilimkar <ssantosh@kernel.org> Signed-off-by:
-
Tony Lindgren authored
commit da6d5993 upstream. It's CONFIG_SOC_OMAP5, not CONFIG_ARCH_OMAP5. Looks like make randconfig builds have not hit this one yet. Fixes: b3bf289c ("ARM: OMAP2+: Fix build with CONFIG_SMP and CONFIG_PM is not set") Acked-by:
-
Tony Lindgren authored
commit 8a8be46a upstream. We need to properly initialize mpuss also on omap5 like we do on omap4. Otherwise we run into similar kexec problems like we had on omap4 when trying to kexec from a kernel with PM initialized. Fixes: 0573b957 ("ARM: OMAP4+: Prevent CPU1 related hang with kexec") Acked-by:
-
Thomas Gleixner authored
commit 26242b33 upstream. In case the driver registration fails, the hotplug callback is leaked. Not fatal, because it's never invoked as there are no instances registered, but wrong nevertheless. Fixes: fdc15a36 ("bus/arm-ccn: Convert to hotplug statemachine") Signed-off-by:
Thomas Gleixner <tglx@linutronix.de> Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Pawel Moll <pawel.moll@arm.com> Cc: Suzuki K Poulose <suzuki.poulose@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Will Deacon <will.deacon@arm.com> Signed-off-by:
-
Chuck Lever authored
commit 1b9f700b upstream. Logic copied from xs_setup_bc_tcp(). Fixes: 39a9beab ('rpc: share one xps between all backchannels') Signed-off-by:
Chuck Lever <chuck.lever@oracle.com> Signed-off-by:
J. Bruce Fields <bfields@redhat.com> Signed-off-by:
-
Neil Armstrong authored
commit 20678773 upstream. Correct prefix is MDM instead of MSM. Fixes: 8aa788d3 ("ARM: configs: qualcomm: Add MDM9615 missing defconfigs") Signed-off-by:
Neil Armstrong <narmstrong@baylibre.com> Reviewed-by:
Stephen Boyd <sboyd@codeaurora.org> Signed-off-by:
Andy Gross <andy.gross@linaro.org> Signed-off-by:
-
Kyle Roeschley authored
commit 7a3cc2a7 upstream. On Zynq, we haven't been reserving the correct amount of DMA-incapable RAM to keep DMA away from it (per the Zynq TRM Section 4.1, it should be the first 512k). In older kernels, this was masked by the memblock_reserve call in arm_memblock_init(). Now, reserve the correct amount excplicitly rather than relying on swapper_pg_dir, which is an address and not a size anyway. Fixes: 46f5b960 ("ARM: zynq: Reserve not DMAable space in front of the kernel") Signed-off-by:
Kyle Roeschley <kyle.roeschley@ni.com> Tested-by:
Nathan Rossi <nathan@nathanrossi.com> Signed-off-by:
Michal Simek <michal.simek@xilinx.com> Signed-off-by:
-
Robert Jarzmik authored
commit e413bd33 upstream. In the device-tree case, the root interrupt controller cannot be accessed through the 6th coprocessor, contrary to pxa27x and pxa3xx architectures. Fix it to behave as in non-devicetree builds. Fixes: 32f17997 ("ARM: pxa: remove irq init from dt machines") Signed-off-by:
Robert Jarzmik <robert.jarzmik@free.fr> Signed-off-by:
-
Andreas Färber authored
commit 4f24450c upstream. bcm2837-rpi-3-b.dts, its only in-tree user, was overriding it as "brcm,bcm2837" already. Fixes: 9d56c22a ("ARM: bcm2835: Add devicetree for the Raspberry Pi 3.") Cc: Stephen Warren <swarren@wwwdotorg.org> Signed-off-by:
Andreas Färber <afaerber@suse.de> Signed-off-by:
-
Andrea Merello authored
commit a44e87b4 upstream. We are incorrectly defining the pwr LED, attaching it to a gpio line that is wired to the Wi-Fi SDIO module (which fails due to this). The actual power LED is connected to the GPIO expander, which we don't expose currently. Fixes: 9d56c22a ("ARM: bcm2835: Add devicetree for the Raspberry Pi 3.") Thanks-to: Eric Anholt <eric@anholt.net> [for clarifying we can't control the LED] Signed-off-by:
Andrea Merello <andrea.merello@gmail.com> Signed-off-by:
-
Matthias Brugger authored
commit a3207d64 upstream. The devicetree node for mt8173-auxadc lacks the clock and io-channel-cells property. This leads to a non-working driver. mt6577-auxadc 11001000.auxadc: failed to get auxadc clock mt6577-auxadc: probe of 11001000.auxadc failed with error -2 Fix these fields to get the device up and running. Fixes: 748c7d4d ("ARM64: dts: mt8173: Add thermal/auxadc device nodes") Signed-off-by:
Matthias Brugger <matthias.bgg@gmail.com> Signed-off-by:
-
Mark Rutland authored
commit 5da889c7 upstream. The virtio tools implementation of READ_ONCE() has a single parameter called 'var', but erroneously refers to 'val' for its cast, and thus won't work unless there's a variable of the correct type that happens to be called 'var'. Fix this with s/var/val/, making READ_ONCE() work as expected regardless. Fixes: a7c49033 ("tools/virtio: use virt_xxx barriers") Signed-off-by:
Mark Rutland <mark.rutland@arm.com> Cc: Jason Wang <jasowang@redhat.com> Cc: Michael S. Tsirkin <mst@redhat.com> Cc: linux-kernel@vger.kernel.org Cc: virtualization@lists.linux-foundation.org Signed-off-by:
Michael S. Tsirkin <mst@redhat.com> Reviewed-by:
Cornelia Huck <cornelia.huck@de.ibm.com> Reviewed-by:
Jason Wang <jasowang@redhat.com> Signed-off-by:
-
Larry Finger authored
commit 8ae679c4 upstream. I am getting the following warning when I build kernel 4.9-git on my PowerBook G4 with a 32-bit PPC processor: AS arch/powerpc/kernel/misc_32.o arch/powerpc/kernel/misc_32.S:299:7: warning: "CONFIG_FSL_BOOKE" is not defined [-Wundef] This problem is evident after commit 989cea5c ("kbuild: prevent lib-ksyms.o rebuilds"); however, this change in kbuild only exposes an error that has been in the code since 2005 when this source file was created. That was with commit 9994a338 ("powerpc: Introduce entry_{32,64}.S, misc_{32,64}.S, systbl.S"). The offending line does not make a lot of sense. This error does not seem to cause any errors in the executable, thus I am not recommending that it be applied to any stable versions. Thanks to Nicholas Piggin for suggesting this solution. Fixes: 9994a338 ("powerpc: Introduce entry_{32,64}.S, misc_{32,64}.S, systbl.S") Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
-
Takashi Sakamoto authored
commit 6a2a2f45 upstream. This module has a bug not to return error code in a case that data structure for transmitted packets fails to be initialized. This commit fixes the bug. Fixes: 35efa5c4 ("ALSA: firewire-tascam: add streaming functionality") Signed-off-by:
Takashi Sakamoto <o-takashi@sakamocchi.jp> Signed-off-by:
-
Greg Kroah-Hartman authored
commit 1ebb7114 upstream. Make sure we have enough of a report structure to validate before looking at it. Reported-by:
Benoit Camredon <benoit.camredon@airbus.com> Tested-by:
Jiri Kosina <jkosina@suse.cz>
-
David Ahern authored
[ Upstream commit 24c63bbc ] Frank reported that vrf devices can be created with a table id of 0. This breaks many of the run time table id checks and should not be allowed. Detect this condition at create time and fail with EINVAL. Fixes: 193125db ("net: Introduce VRF device driver") Reported-by:
Frank Kellermann <frank.kellermann@atos.net> Signed-off-by:
David Ahern <dsa@cumulusnetworks.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
-
David Ahern authored
[ Upstream commit 7a18c5b9 ] fib_select_path does not call fib_select_multipath if oif is set in the flow struct. For VRF use cases oif is always set, so multipath route selection is bypassed. Use the FLOWI_FLAG_SKIP_NH_OIF to skip the oif check similar to what is done in fib_table_lookup. Add saddr and proto to the flow struct for the fib lookup done by the VRF driver to better match hash computation for a flow. Fixes: 613d09b3 ("net: Use VRF device index for lookups on TX") Signed-off-by:
-
Gil Rockah authored
[ Upstream commit 0bbcc0a8 ] When trying to do interface down or changing interface configuration under heavy traffic, some of the adaptive moderation corner cases can occur and leave a WARN_ONCE call trace in the kernel log. Those WARN_ONCE are meant for debug only, and should have been inserted only under debug. We avoid such call traces by removing those WARN_ONCE. Fixes: cb3c7fd4 ("net/mlx5e: Support adaptive RX coalescing") Signed-off-by:
Gil Rockah <gilr@mellanox.com> Signed-off-by:
Saeed Mahameed <saeedm@mellanox.com> Signed-off-by:
-
Herbert Xu authored
[ Upstream commit 57ea52a8 ] The GRO fast path caches the frag0 address. This address becomes invalid if frag0 is modified by pskb_may_pull or its variants. So whenever that happens we must disable the frag0 optimization. This is usually done through the combination of gro_header_hard and gro_header_slow, however, the IPv6 extension header path did the pulling directly and would continue to use the GRO fast path incorrectly. This patch fixes it by disabling the fast path when we enter the IPv6 extension header path. Fixes: 78a478d0 ("gro: Inline skb_gro_header and cache frag0 virtual address") Reported-by:
Slava Shwartsman <slavash@mellanox.com> Signed-off-by:
Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by:
Eric Dumazet <edumazet@google.com> Signed-off-by:
-
Eric Dumazet authored
[ Upstream commit 7cfd5fd5 ] On 32bit arches, (skb->end - skb->data) is not 'unsigned int', so we shall use min_t() instead of min() to avoid a compiler error. Fixes: 1272ce87 ("gro: Enter slow-path if there is no tailroom") Reported-by:
kernel test robot <fengguang.wu@intel.com> Signed-off-by:
-
Herbert Xu authored
[ Upstream commit 1272ce87 ] The GRO path has a fast-path where we avoid calling pskb_may_pull and pskb_expand by directly accessing frag0. However, this should only be done if we have enough tailroom in the skb as otherwise we'll have to expand it later anyway. This patch adds the check by capping frag0_len with the skb tailroom. Fixes: cb18978c ("gro: Open-code final pskb_may_pull") Reported-by:
-
