| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
When creating a new object on the NFS server, we should not be sending
posix setacl requests unless the preceding posix_acl_create returned a
non-trivial acl. Doing so, causes Solaris servers in particular to
return an EINVAL.
Fixes: 013cdf1088d72 (nfs: use generic posix ACL infrastructure,,,)
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1132786
Cc: [email protected] # 3.14+
Signed-off-by: Trond Myklebust <[email protected]>
|
|
If we did an OPEN_DOWNGRADE, then the right thing to do on success, is
to apply the new open mode to the struct nfs4_state. Instead, we were
unconditionally clearing the state, making it appear to our state
machinery as if we had just performed a CLOSE.
Fixes: 226056c5c312b (NFSv4: Use correct locking when updating nfs4_state...)
Cc: [email protected] # 3.15+
Signed-off-by: Trond Myklebust <[email protected]>
|
|
In the presence of delegations, we can no longer assume that the
state->n_rdwr, state->n_rdonly, state->n_wronly reflect the open
stateid share mode, and so we need to calculate the initial value
for calldata->arg.fmode using the state->flags.
Reported-by: James Drews <[email protected]>
Fixes: 88069f77e1ac5 (NFSv41: Fix a potential state leakage when...)
Cc: [email protected] # 2.6.33+
Signed-off-by: Trond Myklebust <[email protected]>
|
|
Similarly to DRA7, OMAP5 has l3 and l4 clock rates incorrectly calculated.
Fixed by using proper divider clock types for the clock nodes.
Signed-off-by: Tero Kristo <[email protected]>
Reported-by: Tomi Valkeinen <[email protected]>
Tested-by: Tomi Valkeinen <[email protected]>
Signed-off-by: Tony Lindgren <[email protected]>
|
|
I have a j5 create (JUA210) USB 2 video device and adding it device id
to SIS USB video gets it to work.
Signed-off-by: Stephen Hemminger <[email protected]>
Cc: stable <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
Remove an unused local variable from head.S. It seems this was never
used even from the initial commit
9703d9d7f77ce129621f7d80a844822e2daa7008 (arm64: Kernel booting and
initialisation), and is a left over from a previous implementation
of __calc_phys_offset.
Signed-off-by: Geoff Levand <[email protected]>
Signed-off-by: Will Deacon <[email protected]>
|
|
A previous over-zealous factorisation of code means that we only treat
registers as volatile if they are readable. For most devices this is fine
since normally most registers can be read and volatility implies
readability but for format_write() devices where there is no readback from
the hardware and we use volatility to mean simply uncacheability this means
that we end up treating all registers as cacheble.
A bigger refactoring of the code to clarify this is in order but as a fix
make a minimal change and only check readability when checking volatility
if there is no format_write() operation defined for the device.
Signed-off-by: Mark Brown <[email protected]>
Tested-by: Lars-Peter Clausen <[email protected]>
Cc: [email protected]
|
|
This is a port of cedb655a3a7764c3fd946077944383c9e0e68dd4
to older asics. Fixes a possible divide by 0 if the harvest
register is invalid.
v2: drop some additional harvest munging.
Signed-off-by: Alex Deucher <[email protected]>
Cc: [email protected]
|
|
This fixes a problem with GPU resets and TLB flushes on SI/CIK.
Signed-off-by: Christian König <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
|
|
cfq_group_service_tree_add() is applying new_weight at the beginning of
the function via cfq_update_group_weight().
This actually allows weight to change between adding it to and subtracting
it from children_weight, and triggers WARN_ON_ONCE() in
cfq_group_service_tree_del(), or even causes oops by divide error during
vfr calculation in cfq_group_service_tree_add().
The detailed scenario is as follows:
1. Create blkio cgroups X and Y as a child of X.
Set X's weight to 500 and perform some I/O to apply new_weight.
This X's I/O completes before starting Y's I/O.
2. Y starts I/O and cfq_group_service_tree_add() is called with Y.
3. cfq_group_service_tree_add() walks up the tree during children_weight
calculation and adds parent X's weight (500) to children_weight of root.
children_weight becomes 500.
4. Set X's weight to 1000.
5. X starts I/O and cfq_group_service_tree_add() is called with X.
6. cfq_group_service_tree_add() applies its new_weight (1000).
7. I/O of Y completes and cfq_group_service_tree_del() is called with Y.
8. I/O of X completes and cfq_group_service_tree_del() is called with X.
9. cfq_group_service_tree_del() subtracts X's weight (1000) from
children_weight of root. children_weight becomes -500.
This triggers WARN_ON_ONCE().
10. Set X's weight to 500.
11. X starts I/O and cfq_group_service_tree_add() is called with X.
12. cfq_group_service_tree_add() applies its new_weight (500) and adds it
to children_weight of root. children_weight becomes 0. Calcularion of
vfr triggers oops by divide error.
weight should be updated right before adding it to children_weight.
Reported-by: Ruki Sekiya <[email protected]>
Signed-off-by: Toshiaki Makita <[email protected]>
Acked-by: Tejun Heo <[email protected]>
Cc: [email protected]
Signed-off-by: Jens Axboe <[email protected]>
|
|
The original version of the driver did not read the SATA calibration
fuse to remove the dependency to the fuse driver. The fuse driver
is now merged, so add this functionality.
The calibration fuse contains a 2-bit value used to pick a set
of calibration values for the SATA pad.
Signed-off-by: Mikko Perttunen <[email protected]>
Acked-by: Thierry Reding <[email protected]>
Signed-off-by: Tejun Heo <[email protected]>
|
|
Add the missing unlock before return from function msm_fbdev_create()
in the error handling case.
Signed-off-by: Wei Yongjun <[email protected]>
Signed-off-by: Rob Clark <[email protected]>
|
|
Signed-off-by: Rob Clark <[email protected]>
|
|
This avoids a problem seen with weston (for example) where the display
gets stuck in "black screen" if starting weston first thing after boot.
Possibly mdp5 needs something similar. The downstream android fbdev
driver always requests DMA_E (or DMA_P) when display is active, rather
than only enabling it on-demand as the drm driver does, which I believe
has the same end result.
Signed-off-by: Rob Clark <[email protected]>
|
|
87e956e9 changed the fault handler to return -ENOSYS, which causes the
iommu driver to print out a huge splat. Which wouldn't be quite so bad
if nothing ever faulted. But seems like some EXA composite operations
generate quite a lot of (seemingly harmless) faults. That is probably a
userspace problem, but the huge increase in verbosity from iommu fault
dumps makes things kind of unusable.
We probably should actually log *some* message (not conditional on
drm.debug). But ratelimit it.
Signed-off-by: Rob Clark <[email protected]>
|
|
Before commit 2cada584b200 ("block: cleanup error handling in sg_io"),
we had ret = 0 before entering the last big if block of sg_io.
Since 2cada584b200, ret = -EFAULT, which breaks hdparm:
/dev/sda:
setting Advanced Power Management level to 0xc8 (200)
HDIO_DRIVE_CMD failed: Bad address
APM_level = 128
Signed-off-by: Sabrina Dubroca <[email protected]>
Fixes: 2cada584b200 ("block: cleanup error handling in sg_io")
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
|
|
When looking at this driver for a client, I noticed the code that
configures the HSYNC pulse clobbers the display width in the same
register. It only preserves the MS part of the width in bit 3 and zeros
the LS part of the width in bits 9 to 4. This doesn't matter during
initialization as the width is configured afterwards, but subsequent use
of the FBIPUT_HSYNC ioctl would clobber the width.
Preserve bits 9 to 0 of LCD_RASTER_TIMING_0_REG when configuring the
horizontal sync.
Signed-off-by: Ian Abbott <[email protected]>
Signed-off-by: Tomi Valkeinen <[email protected]>
|
|
Add missing dependency on RESET_CONTROLLER in order to fix
the following build error.
drivers/gpu/drm/sti/sti_hdmi.c: In function 'sti_hdmi_probe'
drivers/gpu/drm/sti/sti_hdmi.c:780:2: error: implicit declaration of function 'devm_reset_control_get'
[-Werror=implicit-function-declaration]
Benjamin Gaignard remark:
I have change "depends on" to "select" but keep the original author name.
Signed-off-by: Jingoo Han <[email protected]>
|
|
Make of_device_id array const, because all OF functions handle it as const.
Signed-off-by: Kiran Padwal <[email protected]>
|
|
In case of error, the function platform_device_register_resndata()
returns ERR_PTR() and never returns NULL. The NULL test in the return
value check should be replaced with IS_ERR().
Signed-off-by: Wei Yongjun <[email protected]>
|
|
In case of error, the function devm_ioremap_nocache() returns NULL
pointer not ERR_PTR(). The IS_ERR() test in the return value check
should be replaced with NULL test.
Signed-off-by: Wei Yongjun <[email protected]>
|
|
In case of error, the function devm_ioremap_nocache() returns NULL
pointer not ERR_PTR(). The IS_ERR() test in the return value check
should be replaced with NULL test.
Signed-off-by: Wei Yongjun <[email protected]>
|
|
In case of error, the function devm_ioremap_nocache() returns NULL
pointer not ERR_PTR(). The IS_ERR() test in the return value check
should be replaced with NULL test.
Signed-off-by: Wei Yongjun <[email protected]>
|
|
Somehow the intel_ddi_set_vc_payload_alloc(false) call has ended up
in ironlake_crtc_disable() rather than haswell_crtc_disable(). Move it
to the correct place.
intel_ddi_disable_transcoder_func() already disables the vc payload
allocation so this doesn't actually do anything more. The spec
says we should wait for some kind of ack after frobbing the bit. We
don't appear to do that currently, but if and when someone decides
that we should do it, intel_ddi_set_vc_payload_alloc() would appear
to be be the right place for it. So having the function call in
haswell_crtc_disable() seems like the right thing for the future
even if it does nothing currently.
Cc: Dave Airlie <[email protected]>
Signed-off-by: Ville Syrjälä <[email protected]>
Signed-off-by: Jani Nikula <[email protected]>
|
|
If we're runtime suspended and try to use the plane interfaces, we
will get a lot of WARNs saying we did the wrong thing.
We need to get runtime PM references to pin the objects, and to
change the fences. The pin functions are the ideal places for
this, but intel_crtc_cursor_set_obj() doesn't call them, so we also
have to add get/put calls inside it. There is no problem if we runtime
suspend right after these functions are finished, because the
registers written are forwarded to system memory.
Note: for a complete fix of the cursor-dpms test case, we also need
the patch named "drm/i915: Don't try to enable cursor from setplane
when crtc is disabled".
v2: - Narrow the put/get calls on intel_crtc_cursor_set_obj() (Daniel)
v3: - Make get/put also surround the fence and unpin calls (Daniel and
Ville).
- Merge all the plane changes into a single patch since they're
the same fix.
- Add the comment requested by Daniel.
v4: - Remove spurious whitespace (Ville).
v5: - Remove intel_crtc_update_cursor() chunk since Ville did an
equivalent fix in another patch (Ville).
v6: - Remove unpin chunk: it will be on a separate patch (Ville,
Chris, Daniel).
v7: - Same thing, new color.
Testcase: igt/pm_rpm/cursor
Testcase: igt/pm_rpm/cursor-dpms
Testcase: igt/pm_rpm/legacy-planes
Testcase: igt/pm_rpm/legacy-planes-dpms
Testcase: igt/pm_rpm/universal-planes
Testcase: igt/pm_rpm/universal-planes-dpms
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=81645
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=82603
Cc: [email protected]
Signed-off-by: Paulo Zanoni <[email protected]>
Reviewed-by: Ville Syrjälä <[email protected]>
Signed-off-by: Jani Nikula <[email protected]>
|
|
commit c675949ec58ca50d5a3ae3c757892f1560f6e896
Author: Jani Nikula <[email protected]>
Date: Wed Apr 9 11:31:37 2014 +0300
drm/i915: do not setup backlight if not available according to VBT
prevents backlight setup on the Acer C720 (Core i3 4005U CPU), which has a
misconfigured VBT. Apply quirk to ignore the VBT backlight presence check
during backlight setup.
Signed-off-by: Scot Doyle <[email protected]>
Tested-by: Tyler Cleveland <[email protected]>
Cc: Jani Nikula <[email protected]>
Cc: Daniel Vetter <[email protected]>
Cc: [email protected] (3.15+)
Signed-off-by: Jani Nikula <[email protected]>
|
|
The display_timings_release() function frees "disp" and we free it
again on the next line.
Signed-off-by: Dan Carpenter <[email protected]>
Signed-off-by: Tomi Valkeinen <[email protected]>
|
|
Originally found by cppcheck:
[arch/arm64/crypto/sha2-ce-glue.c:153]: (warning) Assignment of
function parameter has no effect outside the function. Did you
forget dereferencing it?
Updating data by blocks * SHA256_BLOCK_SIZE at the end of
sha2_finup is redundant code and can be removed.
Acked-by: Ard Biesheuvel <[email protected]>
Signed-off-by: Colin Ian King <[email protected]>
Signed-off-by: Will Deacon <[email protected]>
|
|
Convert a zero return value on error to a negative one, as returned
elsewhere in the function.
A simplified version of the semantic match that finds this problem is as
follows: (http://coccinelle.lip6.fr/)
// <smpl>
(
if@p1 (\(ret < 0\|ret != 0\))
{ ... return ret; }
|
ret@p1 = 0
)
... when != ret = e1
when != &ret
*if(...)
{
... when != ret = e2
when forall
return ret;
}
// </smpl>
Signed-off-by: Julia Lawall <[email protected]>
Acked-by: Nicolas Ferre <[email protected]>
Signed-off-by: Tomi Valkeinen <[email protected]>
|
|
If the device-tree specifies a max-memory-bandwidth property then the
CLCD driver uses that to calculate the bits-per-pixel supported,
however, this calculation is faulty for two reasons.
1. It doesn't ensure that the result is a sane value, i.e. a power of 2
and <= 32 as the rest of the code assumes.
2. It uses the displayed resolution and calculates the average bandwidth
across the whole frame. It should instead calculate the peak
bandwidth based on the pixel clock.
This patch fixes both the above.
Signed-off-by: Jon Medhurst <[email protected]>
Acked-by: Pawel Moll <[email protected]>
Signed-off-by: Tomi Valkeinen <[email protected]>
|
|
Fix build failure caused as follows:
The function chipsfb_pci_init() references
the function __init chips_hw_init().
This is often because chipsfb_pci_init lacks a __init
annotation or the annotation of chips_hw_init is wrong.
make: *** [drivers] Error 2
by removing the __init annotation from chips_hw_init(). The other thing that
could have been done was annotating chipsfb_pci_init(). But that cannot be done
since chipsfb_pci_init() is called from non __init functions.
Signed-off-by: Pranith Kumar <[email protected]>
Signed-off-by: Tomi Valkeinen <[email protected]>
|
|
The edma_setup_from_hw() should know about the CC number when parsing the
CCCFG register - when it reads the register to be precise. The base
addresses for CCs stored in an array and we need to provide the correct id
to edma_read() in order to read the correct register.
Cc: <[email protected]> # 3.16
Signed-off-by: Peter Ujfalusi <[email protected]>
Signed-off-by: Sekhar Nori <[email protected]>
|
|
Checking adev == NULL is not sufficient as
acpi_bus_get_device() might not touch the value of this
parameter in an error case, so check the return value
directly.
Fixes: ed40356b5fcf1ce28e026ab39c5b2b6939068b50
Cc: David Woodhouse <[email protected]>
Signed-off-by: Joerg Roedel <[email protected]>
Reviewed-by: Alex Williamson <[email protected]>
|
|
When a non-PCI device is passed to that function it might
pass group == NULL to iommu_group_add_device() which then
dereferences it and cause a crash this way. Fix it by
just returning an error for non-PCI devices.
Fixes: 104a1c13ac66e40cf8c6ae74d76ff14ff24b9b01
Cc: Alex Williamson <[email protected]>
Acked-by: Alex Williamson <[email protected]>
Signed-off-by: Joerg Roedel <[email protected]>
|
|
The "RX active" string is too long, so the columns get
shifted. Change it to just "RX" to avoid this.
Signed-off-by: Johannes Berg <[email protected]>
|
|
sta->last_seq_ctrl is the seq_ctrl field from the last header
seen, need to shift it 4 bits to extract the sequence number.
Otherwise the ieee80211_sn_less() check at the top of
ieee80211_sta_manage_reorder_buf drops frames until the sequence
number catches up.
Cc: Michal Kazior <[email protected]>
Signed-off-by: Denton Gentry <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
The 802.11 standard says when processing a plink confirm
frame:
"If the peerLinkID in the mesh peering instance has not been
set, the Local Link ID field of the Mesh Peering Confirm
request shall be copied into the peerLinkID in the mesh
peering instance."
We were only doing this when receiving an open peering frame,
but it could happen that the open frame gets lost and so we
should handle this case rather than rejecting the confirm and
failing the whole peering process.
Reported-by: Yu Niiro <[email protected]>
Signed-off-by: Bob Copeland <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
In ieee80211_sta_ps_deliver_wakeup, sdata->smps_mode is checked. This is
initialized only for the base AP interface, not the individual VLANs.
Signed-off-by: Felix Fietkau <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
When bringing down the AP, a WARN_ON is hit because the bss config chandef
is empty here.
Since AP_VLAN channel settings do not matter for anything chanctx related
(always inherits the settings from the AP interface), let's just ignore
it here.
Signed-off-by: Felix Fietkau <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
Commit 6cfec04bcc05 ("regmap: Separate regmap dev initialization") moved the
regmap debugfs initialization after regcache initialization. This means
that the regmap debugfs directory is not created yet when the cache
initialization runs and so any debugfs files registered by the regcache are
created in the debugfs root directory rather than the debugfs directory of
the regmap instance. Fix this by adding a separate callback for the
regcache debugfs initialization which will be called after the parent
debugfs entry has been created.
Fixes: 6cfec04bcc05 (regmap: Separate regmap dev initialization)
Signed-off-by: Lars-Peter Clausen <[email protected]>
Signed-off-by: Mark Brown <[email protected]>
Cc: [email protected]
|
|
This reverts commit 24aa11ab8ae03292d38ec0dbd9bc2ac49fe8a6dd.
That commit was wrong since it uses data that hasn't even been set
up yet, but might be a hold-over from a previous connection.
Additionally, it seems like a driver-specific workaround that
shouldn't have been in mac80211 to start with.
Cc: [email protected]
Fixes: 24aa11ab8ae0 ("mac80211: disable uAPSD if all ACs are under ACM")
Reviewed-by: Luciano Coelho <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
|
|
drivers/net/ethernet/marvell/mvneta.c: In function 'mvneta_skb_tx_csum':
drivers/net/ethernet/marvell/mvneta.c:1374:3: error: implicit declaration of function 'vlan_get_protocol' [-Werror=implicit-function-declaration]
__be16 l3_proto = vlan_get_protocol(skb);
^
Reporeted-by: Stephen Rothwell <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
There were several issues (of varying degree of importance) pointed out
with this code late in the review cycle, yet the code was still merged.
Let's rip it out for now and look at resubmitting at a later time.
This reverts most of commit 4fbe66d9903425156c193ae44c81c0f7557755c4.
Signed-off-by: Brian Norris <[email protected]>
Signed-off-by: Olof Johansson <[email protected]>
|
|
Interrupt is enabled when bind_interdomain_evtchn_to_irqhandler returns.
If there's interrupt pending interrupt handler is invoked.
NAPI needs to be initialised before binding interrupt otherwise the
interrupt handler will try to scheduling a NAPI instance that is not
initialised yet, resulting in kernel OOPS.
This fixes a regression introduced in ea2c5e13 ("xen-netback: move NAPI
add/remove calls").
Ideally function calls to create kthreads should also be moved before
binding but I intent to fix this regression with minimal changes and
refactor the code with another patch.
Reported-by: Thomas Leonard <[email protected]>
Signed-off-by: Wei Liu <[email protected]>
Cc: Ian Campbell <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Vladislav Yasevich says:
====================
Fix TSO and checksum issues with non-accelerated vlan traffic.
I've recently ran across something rather interesting when testing vlans
from inside VMs. In some scenarios I was getting awfull thruput.
Some debugging uncovered a very scary packet corruption. I was
seeing packets that had original TSO length as IP total length
and their ip checksum was 0. This was with e1000e driver.
A bit more debugging uncovered an assumption made by that driver
that skb->protocol will contain l3 protocol information. This
was not the case in my setup since I manually turned off vlan
tx acceleration for the device. This caused the driver to not
initialize the tso information correctly and resulted in
corrupt TSO frames on the wire.
I decided to do some auditing of the usage of skb->protocols
in the drivers. Out of all the drivers, the included 8 appear
to be effected. They all allow user to control vlan acceleration
settings, all support TSO on vlan devices, and all use
skb->protocol to decide how to encode TSO information. Some
also have similar problems when initializing hw checksum information.
On such device, it is simple enough to reproduce the issue.
Simply turn off TX VLAN acceleration on the device, create a vlan,
and run you favorite network performance tool.
There is 1 driver I ran across that I belive will trigger a BUG
in the system when used with vlans. That driver is tile/tilepro.c
I have not changed it in this patch set and would hope that
the maintainer has time to look at it.
V2: Fix i40ev using the wrong function name. Full build.
====================
Signed-off-by: David S. Miller <[email protected]>
|
|
This device claims TSO support for vlans. It also allows a user to
control vlan acceleration offloading. As such, it is possible to turn
off vlan acceleration and configure a vlan which will continue to send
TSO traffic.
In such situation the packet passed down the the device will contain
a vlan header and skb->protocol will be set to ETH_P_8021Q.
The device assumes that skb->protocol contains network protocol
value and uses that value to set up TSO information.
This results in corrupted frames sent on the wire.
This patch extracts the protocol value correctly by using a
vlan_get_protocol() helper and corrects corrupt TSO frames.
CC: Shahed Shaikh <[email protected]>
CC: Jitendra Kalsaria <[email protected]>
CC: Ron Mercer <[email protected]>
CC: [email protected]
Signed-off-by: Vladislav Yasevich <[email protected]>
Acked-by: Shahed Shaikh <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
This driver doesn't appear to support vlan acceleration at
all. However, it does claim to support TSO and IP checksums
for vlan devices. Thus any configured vlan device would
end up passing down partial checksums or TSO frames.
The driver also uses the value from skb->protocol to
determine TSO and checksum offload information, but assumes
that skb->protocol holds the l3 protocol information.
As a result, vlan traffic with partial checksums or TSO
will fail those checks and TSO will not happen.
Fix this by using vlan_get_protocol() helper.
CC: Thomas Petazzoni <[email protected]>
Signed-off-by: Vladislav Yasevich <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
This device claims TSO and checksum support for vlans. It also
allows a user to control vlan acceleration offloading. As such,
it is possible to turn off vlan acceleration and configure a vlan
which will continue to support TSO and hw checksums.
In such situation the packet passed down the the device will contain
a vlan header and skb->protocol will be set to ETH_P_8021Q.
The device assumes that skb->protocol contains network protocol
value and uses that value to set up TSO and checksum information.
This results in corrupted frames sent on the wire.
This patch extract the protocol value correctly and corrects TSO
and checksums for non-accelerated traffic.
Fix this by using vlan_get_protocol() helper.
CC: Jeff Kirsher <[email protected]>
CC: Jesse Brandeburg <[email protected]>
CC: Bruce Allan <[email protected]>
CC: Carolyn Wyborny <[email protected]>
CC: Don Skidmore <[email protected]>
CC: Greg Rose <[email protected]>
CC: Alex Duyck <[email protected]>
CC: John Ronciak <[email protected]>
CC: Mitch Williams <[email protected]>
CC: Linux NICS <[email protected]>
CC: [email protected]
Signed-off-by: Vladislav Yasevich <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
This device claims TSO and checksum support for vlans. It also
allows a user to control vlan acceleration offloading. As such,
it is possible to turn off vlan acceleration and configure a vlan
which will continue to support TSO and hw checksums.
In such situation the packet passed down the the device will contain
a vlan header and skb->protocol will be set to ETH_P_8021Q.
The device assumes that skb->protocol contains network protocol
value and uses that value to set up TSO and checksum information.
This results in corrupted frames sent on the wire.
This patch extract the protocol value correctly and corrects TSO
and checksums for non-accelerated traffic.
Fix this by using vlan_get_protocol() helper.
CC: Jeff Kirsher <[email protected]>
CC: Jesse Brandeburg <[email protected]>
CC: Bruce Allan <[email protected]>
CC: Carolyn Wyborny <[email protected]>
CC: Don Skidmore <[email protected]>
CC: Greg Rose <[email protected]>
CC: Alex Duyck <[email protected]>
CC: John Ronciak <[email protected]>
CC: Mitch Williams <[email protected]>
CC: Linux NICS <[email protected]>
CC: [email protected]
Signed-off-by: Vladislav Yasevich <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
The driver claims that it can do TSO and IP checksums on vlan
devices and also allows user to control vlan acceleration offloading.
This makes it possible to push traffic to this driver that has TSO or
partial checksums set, but also have a non-accelearted vlan
header. In this case, the driver will fail to correctly
identify such traffic and will not correctly perform
segmentation and checksum calculation.
Fix this by using vlan_get_protocol() helper instead of
assuming skb->protocol always has this information.
CC: Thadeu Lima de Souza Cascardo <[email protected]>
Signed-off-by: Vladislav Yasevich <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|