Skip to content

L
linux
Switch branch/tag
  1. 13 Aug, 2021 16 commits
  3. 12 Aug, 2021 24 commits
    • Alex Elder's avatar
      dt-bindings: net: qcom,ipa: make imem interconnect optional · b769cf44
      Alex Elder authored 
      On some newer SoCs, the interconnect between IPA and SoC internal
memory (imem) is not used.  Update the binding to indicate that
having just the memory and config interconnects is another allowed
configuration.
Signed-off-by: default avatarAlex Elder <elder@linaro.org>
Link: https://lore.kernel.org/r/20210811141802.2635424-1-elder@linaro.orgSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      b769cf44
    • Alex Elder's avatar
      net: ipa: always inline ipa_aggr_granularity_val() · 676eec8e
      Alex Elder authored 
      It isn't required, but all callers of ipa_aggr_granularity_val()
pass a constant value (IPA_AGGR_GRANULARITY) as the usec argument.
Two of those callers are in ipa_validate_build(), with the result
being passed to BUILD_BUG_ON().

Evidently the "sparc64-linux-gcc" compiler (at least) doesn't always
inline ipa_aggr_granularity_val(), so the result of the function is
not constant at compile time, and that leads to build errors.

Define the function with the __always_inline attribute to avoid the
errors.  We can see by inspection that the value passed is never
zero, so we can just remove its WARN_ON() call.

Fixes: 5bc55884 ("net: ipa: use WARN_ON() rather than assertions")
Reported-by: default avatarkernel test robot <lkp@intel.com>
Signed-off-by: default avatarAlex Elder <elder@linaro.org>
Link: https://lore.kernel.org/r/20210811135948.2634264-1-elder@linaro.orgSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      676eec8e
    • Dave Airlie's avatar
      Merge tag 'drm-misc-fixes-2021-08-12' of git://anongit.freedesktop.org/drm/drm-misc into drm-fixes · a1fa7268
      Dave Airlie authored 
      Short summary of fixes pull:

 * meson: Fix colors when booting with HDR
Signed-off-by: default avatarDave Airlie <airlied@redhat.com>

From: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/YRTb+qUuBYWjJDVg@linux-uq9g.fritz.box
      a1fa7268
    • Dave Airlie's avatar
      Merge tag 'drm-intel-fixes-2021-08-12' of... · 3e234e9f
      Dave Airlie authored 
      Merge tag 'drm-intel-fixes-2021-08-12' of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes

- GVT fix for Windows VM hang.
- Display fix of 12 BPC bits for display 12 and newer.
- Don't try to access some media register for fused off domains.
- Fix kerneldoc build warnings.
Signed-off-by: default avatarDave Airlie <airlied@redhat.com>

From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/YRU/hnQ1sNr+j37x@intel.com
      3e234e9f
    • Jakub Kicinski's avatar
      Merge tag 'ieee802154-for-davem-2021-08-12' of... · a9a50701
      Jakub Kicinski authored 
      Merge tag 'ieee802154-for-davem-2021-08-12' of git://git.kernel.org/pub/scm/linux/kernel/git/sschmidt/wpan

Stefan Schmidt says:

====================
ieee802154 for net 2021-08-12

Mostly fixes coming from bot reports. Dongliang Mu tackled some syzkaller
reports in hwsim again and Takeshi Misawa a memory leak  in  ieee802154 raw.

* tag 'ieee802154-for-davem-2021-08-12' of git://git.kernel.org/pub/scm/linux/kernel/git/sschmidt/wpan:
  net: Fix memory leak in ieee802154_raw_deliver
  ieee802154: hwsim: fix GPF in hwsim_new_edge_nl
  ieee802154: hwsim: fix GPF in hwsim_set_edge_lqi
====================

Link: https://lore.kernel.org/r/20210812183912.1663996-1-stefan@datenfreihafen.orgSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      a9a50701
    • Longpeng(Mike)'s avatar
      vsock/virtio: avoid potential deadlock when vsock device remove · 49b0b6ff
      Longpeng(Mike) authored 
      There's a potential deadlock case when remove the vsock device or
process the RESET event:

  vsock_for_each_connected_socket:
      spin_lock_bh(&vsock_table_lock) ----------- (1)
      ...
          virtio_vsock_reset_sock:
              lock_sock(sk) --------------------- (2)
      ...
      spin_unlock_bh(&vsock_table_lock)

lock_sock() may do initiative schedule when the 'sk' is owned by
other thread at the same time, we would receivce a warning message
that "scheduling while atomic".

Even worse, if the next task (selected by the scheduler) try to
release a 'sk', it need to request vsock_table_lock and the deadlock
occur, cause the system into softlockup state.
  Call trace:
   queued_spin_lock_slowpath
   vsock_remove_bound
   vsock_remove_sock
   virtio_transport_release
   __vsock_release
   vsock_release
   __sock_release
   sock_close
   __fput
   ____fput

So we should not require sk_lock in this case, just like the behavior
in vhost_vsock or vmci.

Fixes: 0ea9e1d3 ("VSOCK: Introduce virtio_transport.ko")
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: default avatarLongpeng(Mike) <longpeng2@huawei.com>
Reviewed-by: default avatarStefano Garzarella <sgarzare@redhat.com>
Link: https://lore.kernel.org/r/20210812053056.1699-1-longpeng2@huawei.comSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      49b0b6ff
    • Linus Torvalds's avatar
      Merge branch 'for-v5.14' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace · f8fbb47c
      Linus Torvalds authored 
      Pull ucounts fix from Eric Biederman:
 "This fixes the ucount sysctls on big endian architectures.

  The counts were expanded to be longs instead of ints, and the sysctl
  code was overlooked, so only the low 32bit were being processed. On
  litte endian just processing the low 32bits is fine, but on 64bit big
  endian processing just the low 32bits results in the high order bits
  instead of the low order bits being processed and nothing works
  proper.

  This change took a little bit to mature as we have the SYSCTL_ZERO,
  and SYSCTL_INT_MAX macros that are only usable for sysctls operating
  on ints, but unfortunately are not obviously broken. Which resulted in
  the versions of this change working on big endian and not on little
  endian, because the int SYSCTL_ZERO when extended 64bit wound up being
  0x100000000. So we only allowed values greater than 0x100000000 and
  less than 0faff. Which unfortunately broken everything that tried to
  set the sysctls. (First reported with the windows subsystem for
  linux).

  I have tested this on x86_64 64bit after first reproducing the
  problems with the earlier version of this change, and then verifying
  the problems do not exist when we use appropriate long min and max
  values for extra1 and extra2"

* 'for-v5.14' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace:
  ucounts: add missing data type changes
      f8fbb47c
    • Linus Torvalds's avatar
      Merge tag 'sound-5.14-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound · 59cd4f43
      Linus Torvalds authored 
      Pull sound fixes from Takashi Iwai:
 "This seems to be a usual bump in the middle, containing lots of
  pending ASoC fixes:

   - Yet another PCM mmap regression fix

   - Fix for ASoC DAPM prefix handling

   - Various cs42l42 codec fixes

   - PCM buffer reference fixes in a few ASoC drivers

   - Fixes for ASoC SOF, AMD, tlv320, WM

   - HD-audio quirks"

* tag 'sound-5.14-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: (32 commits)
  ALSA: hda/realtek: fix mute/micmute LEDs for HP ProBook 650 G8 Notebook PC
  ALSA: pcm: Fix mmap breakage without explicit buffer setup
  ALSA: hda: Add quirk for ASUS Flow x13
  ASoC: cs42l42: Fix mono playback
  ASoC: cs42l42: Constrain sample rate to prevent illegal SCLK
  ASoC: cs42l42: Fix LRCLK frame start edge
  ASoC: cs42l42: PLL must be running when changing MCLK_SRC_SEL
  ASoC: cs42l42: Remove duplicate control for WNF filter frequency
  ASoC: cs42l42: Fix inversion of ADC Notch Switch control
  ASoC: SOF: Intel: hda-ipc: fix reply size checking
  ASoC: SOF: Intel: Kconfig: fix SoundWire dependencies
  ASoC: amd: Fix reference to PCM buffer address
  ASoC: nau8824: Fix open coded prefix handling
  ASoC: kirkwood: Fix reference to PCM buffer address
  ASoC: uniphier: Fix reference to PCM buffer address
  ASoC: xilinx: Fix reference to PCM buffer address
  ASoC: intel: atom: Fix reference to PCM buffer address
  ASoC: cs42l42: Fix bclk calculation for mono
  ASoC: cs42l42: Don't allow SND_SOC_DAIFMT_LEFT_J
  ASoC: cs42l42: Correct definition of ADC Volume control
  ...
      59cd4f43
    • Andy Shevchenko's avatar
      wwan: core: Avoid returning NULL from wwan_create_dev() · d9d5b896
      Andy Shevchenko authored 
      Make wwan_create_dev() to return either valid or error pointer,
In some cases it may return NULL. Prevent this by converting
it to the respective error pointer.

Fixes: 9a44c1cc ("net: Add a WWAN subsystem")
Signed-off-by: default avatarAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: default avatarSergey Ryazanov <ryazanov.s.a@gmail.com>
Reviewed-by: default avatarLoic Poulain <loic.poulain@linaro.org>
Link: https://lore.kernel.org/r/20210811124845.10955-1-andriy.shevchenko@linux.intel.comSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      d9d5b896
    • David S. Miller's avatar
      Merge tag 'mlx5-updates-2021-08-11' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux · bed5a942
      David S. Miller authored 
      Saeed Mahameed says:

====================
mlx5 updates 2021-08-11

This series provides misc updates to mlx5.
For more information please see tag log below.

Please pull and let me know if there is any problem.

mlx5-updates-2021-08-11

Misc. cleanup for mlx5.

1) Typos and use of netdev_warn()
2) smatch cleanup
3) Minor fix to inner TTC table creation
4) Dynamic capability cache allocation
====================
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      bed5a942
    • David S. Miller's avatar
      Merge branch 'dsa-cross-chip-notifiers' · dea80774
      David S. Miller authored 
      Vladimir Oltean says:

====================
Improvements to the DSA tag_8021q cross-chip notifiers

This series improves cross-chip notifier error messages and addresses a
benign error message seen during reboot on a system with disjoint DSA
trees.
====================
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      dea80774
    • Vladimir Oltean's avatar
      net: dsa: tag_8021q: don't broadcast during setup/teardown · 724395f4
      Vladimir Oltean authored 
      Currently, on my board with multiple sja1105 switches in disjoint trees
described in commit f66a6a69 ("net: dsa: permit cross-chip bridging
between all trees in the system"), rebooting the board triggers the
following benign warnings:

[   12.345566] sja1105 spi2.0: port 0 failed to notify tag_8021q VLAN 1088 deletion: -ENOENT
[   12.353804] sja1105 spi2.0: port 0 failed to notify tag_8021q VLAN 2112 deletion: -ENOENT
[   12.362019] sja1105 spi2.0: port 1 failed to notify tag_8021q VLAN 1089 deletion: -ENOENT
[   12.370246] sja1105 spi2.0: port 1 failed to notify tag_8021q VLAN 2113 deletion: -ENOENT
[   12.378466] sja1105 spi2.0: port 2 failed to notify tag_8021q VLAN 1090 deletion: -ENOENT
[   12.386683] sja1105 spi2.0: port 2 failed to notify tag_8021q VLAN 2114 deletion: -ENOENT

Basically switch 1 calls dsa_tag_8021q_unregister, and switch 1's TX and
RX VLANs cannot be found on switch 2's CPU port.

But why would switch 2 even attempt to delete switch 1's TX and RX
tag_8021q VLANs from its CPU port? Well, because we use dsa_broadcast,
and it is supposed that it had added those VLANs in the first place
(because in dsa_port_tag_8021q_vlan_match, all CPU ports match
regardless of their tree index or switch index).

The two trees probe asynchronously, and when switch 1 probed, it called
dsa_broadcast which did not notify the tree of switch 2, because that
didn't probe yet. But during unbind, switch 2's tree _is_ probed, so it
_is_ notified of the deletion.

Before jumping to introduce a synchronization mechanism between the
probing across disjoint switch trees, let's take a step back and see
whether we _need_ to do that in the first place.

The RX and TX VLANs of switch 1 would be needed on switch 2's CPU port
only if switch 1 and 2 were part of a cross-chip bridge. And
dsa_tag_8021q_bridge_join takes care precisely of that (but if probing
was synchronous, the bridge_join would just end up bumping the VLANs'
refcount, because they are already installed by the setup path).

Since by the time the ports are bridged, all DSA trees are already set
up, and we don't need the tag_8021q VLANs of one switch installed on the
other switches during probe time, the answer is that we don't need to
fix the synchronization issue.

So make the setup and teardown code paths call dsa_port_notify, which
notifies only the local tree, and the bridge code paths call
dsa_broadcast, which let the other trees know as well.
Signed-off-by: default avatarVladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: default avatarFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      724395f4
    • Vladimir Oltean's avatar
      net: dsa: print more information when a cross-chip notifier fails · ab97462b
      Vladimir Oltean authored 
      Currently this error message does not say a lot:

[   32.693498] DSA: failed to notify tag_8021q VLAN deletion: -ENOENT
[   32.699725] DSA: failed to notify tag_8021q VLAN deletion: -ENOENT
[   32.705931] DSA: failed to notify tag_8021q VLAN deletion: -ENOENT
[   32.712139] DSA: failed to notify tag_8021q VLAN deletion: -ENOENT
[   32.718347] DSA: failed to notify tag_8021q VLAN deletion: -ENOENT
[   32.724554] DSA: failed to notify tag_8021q VLAN deletion: -ENOENT

but in this form, it is immediately obvious (at least to me) what the
problem is, even without further looking at the code:

[   12.345566] sja1105 spi2.0: port 0 failed to notify tag_8021q VLAN 1088 deletion: -ENOENT
[   12.353804] sja1105 spi2.0: port 0 failed to notify tag_8021q VLAN 2112 deletion: -ENOENT
[   12.362019] sja1105 spi2.0: port 1 failed to notify tag_8021q VLAN 1089 deletion: -ENOENT
[   12.370246] sja1105 spi2.0: port 1 failed to notify tag_8021q VLAN 2113 deletion: -ENOENT
[   12.378466] sja1105 spi2.0: port 2 failed to notify tag_8021q VLAN 1090 deletion: -ENOENT
[   12.386683] sja1105 spi2.0: port 2 failed to notify tag_8021q VLAN 2114 deletion: -ENOENT
Signed-off-by: default avatarVladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: default avatarFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      ab97462b
    • Daniel Vetter's avatar
      drm/doc/rfc: drop lmem uapi section · ffd5caa2
      Daniel Vetter authored 
      We still have quite a bit more work to do with overall reworking of
the ttm-based dg1 code, but the uapi stuff is now finalized with the
latest pull. So remove that.

This also fixes kerneldoc build warnings because we've included the
same headers in two places, resulting in sphinx complaining about
duplicated symbols. This regression has been created when we moved the
uapi definitions to the real include/uapi/ folder in 727ecd99
("drm/doc/rfc: drop the i915_gem_lmem.h header")

v2: Fix a few references that I missed, the htmldocs build took
forever.
Acked-by: default avatarJason Ekstrand <jason@jlekstrand.net>
Acked-by: default avatarMaarten Lankhorst <maarten.lankhorst@linux.intel.com>
Tested-by Stephen Rothwell <sfr@canb.auug.org.au> (v1)
References: https://lore.kernel.org/dri-devel/20210603193242.1ce99344@canb.auug.org.au/Reported-by: default avatarStephen Rothwell <sfr@canb.auug.org.au>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Fixes: 727ecd99 ("drm/doc/rfc: drop the i915_gem_lmem.h header")
Cc: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: default avatarDaniel Vetter <daniel.vetter@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210810142748.1983271-1-daniel.vetter@ffwll.ch
(cherry picked from commit dae2d288)
Signed-off-by: default avatarRodrigo Vivi <rodrigo.vivi@intel.com>
      ffd5caa2
    • Matt Roper's avatar
      drm/i915: Only access SFC_DONE when media domain is not fused off · 24d032e2
      Matt Roper authored 
      The SFC_DONE register lives within the corresponding VD0/VD2/VD4/VD6
forcewake domain and is not accessible if the vdbox in that domain is
fused off and the forcewake is not initialized.

This mistake went unnoticed because until recently we were using the
wrong register offset for the SFC_DONE register; once the register
offset was corrected, we started hitting errors like

  <4> [544.989065] i915 0000:cc:00.0: Uninitialized forcewake domain(s) 0x80 accessed at 0x1ce000

on parts with fused-off vdbox engines.

Fixes: e50dbdbf ("drm/i915/tgl: Add SFC instdone to error state")
Fixes: 9c9c6d0a ("drm/i915: Correct SFC_DONE register offset")
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Signed-off-by: default avatarMatt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210806174130.1058960-1-matthew.d.roper@intel.comReviewed-by: default avatarJosé Roberto de Souza <jose.souza@intel.com>
(cherry picked from commit c5589bb5)
Signed-off-by: default avatarRodrigo Vivi <rodrigo.vivi@intel.com>
[Changed Fixes tag to match the cherry-picked 82929a21]
      24d032e2
    • Andy Shevchenko's avatar
      wwan: core: Unshadow error code returned by ida_alloc_range() · 0de6fd5f
      Andy Shevchenko authored 
      ida_alloc_range() may return other than -ENOMEM error code.
Unshadow it in the wwan_create_port().
Signed-off-by: default avatarAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: default avatarSergey Ryazanov <ryazanov.s.a@gmail.com>
Reviewed-by: default avatarLoic Poulain <loic.poulain@linaro.org>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      0de6fd5f
    • Ankit Nautiyal's avatar
      drm/i915/display: Fix the 12 BPC bits for PIPE_MISC reg · abd9d66a
      Ankit Nautiyal authored 
      Till DISPLAY12 the PIPE_MISC bits 5-7 are used to set the
Dithering BPC, with valid values of 6, 8, 10 BPC.
For ADLP+ these bits are used to set the PORT OUTPUT BPC, with valid
values of: 6, 8, 10, 12 BPC, and need to be programmed whether
dithering is enabled or not.

This patch:
-corrects the bits 5-7 for PIPE MISC register for 12 BPC.
-renames the bits and mask to have generic names for these bits for
dithering bpc and port output bpc.

v3: Added a note for MIPI DSI which uses the PIPE_MISC for readout
for pipe_bpp. (Uma Shankar)

v2: Added 'display' to the subject and fixes tag. (Uma Shankar)

Fixes: 756f85cf ("drm/i915/bdw: Broadwell has PIPEMISC")
Cc: Paulo Zanoni <paulo.r.zanoni@intel.com> (v1)
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: intel-gfx@lists.freedesktop.org
Cc: <stable@vger.kernel.org> # v3.13+
Signed-off-by: default avatarAnkit Nautiyal <ankit.k.nautiyal@intel.com>
Reviewed-by: default avatarUma Shankar <uma.shankar@intel.com>
Signed-off-by: default avatarUma Shankar <uma.shankar@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210811051857.109723-1-ankit.k.nautiyal@intel.com
(cherry picked from commit 70418a68)
Signed-off-by: default avatarRodrigo Vivi <rodrigo.vivi@intel.com>
      abd9d66a
    • Vladimir Oltean's avatar
      net: dsa: sja1105: unregister the MDIO buses during teardown · 700fa08d
      Vladimir Oltean authored 
      The call to sja1105_mdiobus_unregister is present in the error path but
absent from the main driver unbind path.

Fixes: 5a8f0974 ("net: dsa: sja1105: register the MDIO buses for 100base-T1 and 100base-TX")
Signed-off-by: default avatarVladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: default avatarFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      700fa08d
    • DENG Qingfang's avatar
      net: dsa: mt7530: fix VLAN traffic leaks again · 7428022b
      DENG Qingfang authored 
      When a port leaves a VLAN-aware bridge, the current code does not clear
other ports' matrix field bit. If the bridge is later set to VLAN-unaware
mode, traffic in the bridge may leak to that port.

Remove the VLAN filtering check in mt7530_port_bridge_leave.

Fixes: 474a2dda ("net: dsa: mt7530: fix VLAN traffic leaks")
Fixes: 83163f7d ("net: dsa: mediatek: add VLAN support for MT7530")
Signed-off-by: default avatarDENG Qingfang <dqfext@gmail.com>
Reviewed-by: default avatarVladimir Oltean <olteanv@gmail.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      7428022b
    • Oleksij Rempel's avatar
      net: phy: nxp-tja11xx: log critical health state · e0ba6050
      Oleksij Rempel authored 
      TJA1102 provides interrupt notification for the critical health states
like overtemperature and undervoltage.

The overtemperature bit is set if package temperature is beyond 155C°.
This functionality was tested by heating the package up to 200C°

The undervoltage bit is set if supply voltage drops beyond some critical
threshold. Currently not tested.

In a typical use case, both of this events should be logged and stored
(or send to some remote system) for further investigations.
Signed-off-by: default avatarOleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: default avatarAndrew Lunn <andrew@lunn.ch>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      e0ba6050
    • David S. Miller's avatar
      Merge branch 'pktgen-imix' · 445af0d2
      David S. Miller authored 
      Nick Richardson says:

====================
pktgen: Add IMIX mode

Adds internet mix (IMIX) mode to pktgen. Internet mix is
included in many user-space network perf testing tools. It allows
for the user to specify a distribution of discrete packet sizes to be
generated. This type of test is common among vendors when perf testing
their devices.
[RFC link: https://datatracker.ietf.org/doc/html/rfc2544#section-9.1]

This allows users to get a
more complete picture of how their device will perform in the
real-world.

This feature adds a command that allows users to specify an imix
distribution in the following format:
  imix_weights size_1,weight_1 size_2,weight_2 ... size_n,weight_n

The distribution of packets with size_i will be
(weight_i / total_weights) where
total_weights = weight_1 + weight_2 + ... + weight_n

For example:
  imix_weights 40,7 576,4 1500,1

The pkt_size "40" will account for 7 / (7 + 4 + 1) = ~58% of the total
packets sent.

This patch was tested with the following:
1. imix_weights = 40,7 576,4 1500,1
2. imix_weights = 0,7 576,4 1500,1
  - Packet size of 0 is resized to the minimum, 42
3. imix_weights = 40,7 576,4 1500,1 count = 0
  - Zero count.
  - Runs until user stops pktgen.
Invalid Configurations
1. clone_skb = 200 imix_weights = 40,7 576,4 1500,1
    - Returns error code -524 (-ENOTSUPP) when setting imix_weights
2. len(imix_weights) > MAX_IMIX_ENTRIES
    - Returns -7 (-E2BIG)

This patch is split into three parts, each provide different aspects of
required functionality:
  1. Parse internet mix input.
  2. Add IMIX Distribution representation.
  3. Process and output IMIX results.

Changes in v2:
* Remove __ prefix outside of uAPI.
* Use seq_puts instead of seq_printf where necessary.
* Reorder variable declaration.
* Return -EINVAL instead of -ENOTSUPP when using IMIX with clone_skb > 0
====================
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      445af0d2
    • Nick Richardson's avatar
      pktgen: Add output for imix results · 769afb3f
      Nick Richardson authored 
      The bps for imix mode is calculated by:
sum(imix_entry.size) / time_elapsed

The actual counts of each imix_entry are displayed under the
"Current:" section of the interface output in the following format:
imix_size_counts: size_1,count_1 size_2,count_2 ... size_n,count_n

Example (count = 200000):
imix_weights: 256,1 859,3 205,2
imix_size_counts: 256,32082 859,99796 205,68122
Result: OK: 17992362(c17964678+d27684) usec, 200000 (859byte,0frags)
  11115pps 47Mb/sec (47977140bps) errors: 0

Summary of changes:
Calculate bps based on imix counters when in IMIX mode.
Add output for IMIX counters.
Signed-off-by: default avatarNick Richardson <richardsonnick@google.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      769afb3f
    • Nick Richardson's avatar
      pktgen: Add imix distribution bins · 90149031
      Nick Richardson authored 
      In order to represent the distribution of imix packet sizes, a
pre-computed data structure is used. It features 100 (IMIX_PRECISION)
"bins". Contiguous ranges of these bins represent the respective
packet size of each imix entry. This is done to avoid the overhead of
selecting the correct imix packet size based on the corresponding weights.

Example:
imix_weights 40,7 576,4 1500,1
total_weight = 7 + 4 + 1 = 12

pkt_size 40 occurs 7/total_weight = 58% of the time
pkt_size 576 occurs 4/total_weight = 33% of the time
pkt_size 1500 occurs 1/total_weight = 9% of the time

We generate a random number between 0-100 and select the corresponding
packet size based on the specified weights.
Eg. random number = 358723895 % 100 = 65
Selects the packet size corresponding to index:65 in the pre-computed
imix_distribution array.
An example of the  pre-computed array is below:

The imix_distribution will look like the following:
0        ->  0 (index of imix_entry.size == 40)
1        ->  0 (index of imix_entry.size == 40)
2        ->  0 (index of imix_entry.size == 40)
[...]    ->  0 (index of imix_entry.size == 40)
57       ->  0 (index of imix_entry.size == 40)
58       ->  1 (index of imix_entry.size == 576)
[...]    ->  1 (index of imix_entry.size == 576)
90       ->  1 (index of imix_entry.size == 576)
91       ->  2 (index of imix_entry.size == 1500)
[...]    ->  2 (index of imix_entry.size == 1500)
99       ->  2 (index of imix_entry.size == 1500)

Create and use "bin" representation of the imix distribution.
Signed-off-by: default avatarNick Richardson <richardsonnick@google.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      90149031
    • Nick Richardson's avatar
      pktgen: Parse internet mix (imix) input · 52a62f86
      Nick Richardson authored 
      Adds "imix_weights" command for specifying internet mix distribution.

The command is in this format:
"imix_weights size_1,weight_1 size_2,weight_2 ... size_n,weight_n"
where the probability that packet size_i is picked is:
weight_i / (weight_1 + weight_2 + .. + weight_n)

The user may provide up to 100 imix entries (size_i,weight_i) in this
command.

The user specified imix entries will be displayed in the "Params"
section of the interface output.

Values for clone_skb > 0 is not supported in IMIX mode.

Summary of changes:
Add flag for enabling internet mix mode.
Add command (imix_weights) for internet mix input.
Return -ENOTSUPP when clone_skb > 0 in IMIX mode.
Display imix_weights in Params.
Create data structures to store imix entries and distribution.
Signed-off-by: default avatarNick Richardson <richardsonnick@google.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      52a62f86
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7