soc_compr_open() does rollback when failed (A),
but, it is almost same as soc_compr_free().
static int soc_compr_open(xxx)
{
...
if (ret < 0)
goto xxx_err;
...
return 0;
^ machine_err:
| ...
| out:
(A) ...
| pm_err:
| ...
v return ret;
}
The difference is
soc_compr_free() is for all dai/component/substream,
rollback is for succeeded part only.
This kind of duplicated code can be a hotbed of bugs,
thus, we want to share soc_compr_free() and rollback.
1) snd_soc_dai_compr_startup/shutdown()
=> 2) snd_soc_component_compr_open/free()
3) snd_soc_link_compr_startup/shutdown()
This patch is for 2) snd_soc_component_compr_open/free(),
and adds new cstream mark.
It will mark cstream when startup() was suceeded.
If rollback happen *after* that, it will check rollback flag
and marked cstream.
It cares *previous* startup() only now,
but we might want to check *whole* marked cstream in the future.
This patch is using macro so that it can be easily adjust to it.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87lfey5iwk.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
soc_compr_open() does rollback when failed (A),
but, it is almost same as soc_compr_free().
static int soc_compr_open(xxx)
{
...
if (ret < 0)
goto xxx_err;
...
return 0;
^ machine_err:
| ...
| out:
(A) ...
| pm_err:
| ...
v return ret;
}
The difference is
soc_compr_free() is for all dai/component/substream,
rollback is for succeeded part only.
This kind of duplicated code can be a hotbed of bugs,
thus, we want to share soc_compr_free() and rollback.
=> 1) snd_soc_dai_compr_startup/shutdown()
2) snd_soc_component_compr_open/free()
3) snd_soc_link_compr_startup/shutdown()
This patch is for 1) snd_soc_dai_compr_startup/shutdown(),
and adds new cstream mark.
It will mark cstream when startup() was suceeded.
If rollback happen *after* that, it will check rollback flag
and marked cstream.
It cares *previous* startup() only now,
but we might want to check *whole* marked cstream in the future.
This patch is using macro so that it can be easily adjust to it.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87mtze5iwp.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The SOF (Sound Open Firmware) tree contains a lot of references in
topology files to 'codec_slave'/'codec_master' terms, which in turn
come from alsa-lib and ALSA/ASoC topology support at the kernel
level. These terms are no longer compatible with the guidelines
adopted by the kernel community [1], standard organizations, and need
to change in backwards-compatible ways.
The main/secondary terms typically suggested in guidelines don't mean
anything for clocks, this patchset suggests instead the use of
'provider' and 'consumer' terms, with the 'codec' prefix kept to make
it clear that the codec is the reference. The CM/CS suffixes are also
replaced by CP/CC.
It can be argued that the change of suffix is invasive, but finding a
replacement that keeps the M and S shortcuts has proven difficult in
quite a few contexts.
The previous definitions are kept for backwards-compatibility so this
change should not have any functional impact. It is suggested that new
contributions only use the new terms but there is no requirement to
transition immediately to the new definitions for existing code. Intel
will however update all its past contributions related to bit
clock/frame sync configurations immediately.
This patchset contains the kernel changes only, the alsa-lib changes
were shared separately.
Feedback welcome
~Pierre
[1] https://lkml.org/lkml/2020/7/4/229
Pierre-Louis Bossart (4):
ASoC: topology: use inclusive language for bclk and fsync
ASoC: SOF: use inclusive language for bclk and fsync
ASoC: Intel: atom: use inclusive language for SSP bclk/fsync
ASoC: Intel: keembay: use inclusive language for bclk and fsync
include/sound/soc-dai.h | 32 +++++++++++++++---------
include/sound/sof/dai.h | 16 ++++++++----
include/uapi/sound/asoc.h | 22 ++++++++++------
sound/soc/intel/atom/sst-atom-controls.c | 12 ++++-----
sound/soc/intel/atom/sst-atom-controls.h | 4 +--
sound/soc/intel/keembay/kmb_platform.c | 22 ++++++++--------
sound/soc/intel/keembay/kmb_platform.h | 8 +++---
sound/soc/soc-topology.c | 24 +++++++++---------
sound/soc/sof/topology.c | 18 ++++++-------
9 files changed, 89 insertions(+), 69 deletions(-)
--
2.25.1
This commit exposes following functions which can be used by a sound
card driver based on generic audio graph driver. Idea is vendors can
have a thin driver and re-use common stuff from audio graph driver.
- graph_card_probe()
- graph_parse_of()
In doing so a new header file is added for above. The graph_probe()
function is simplified by moving more common stuff to graph_parse_of().
Signed-off-by: Sameer Pujar <spujar@nvidia.com>
Cc: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/1604329814-24779-8-git-send-email-spujar@nvidia.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Add new members in struct 'asoc_simple_priv'. Idea is to leverage
simple or graph card driver as much as possible and vendor can
maintain a thin driver to control the behavior by populating these
newly exposed members.
Following are the members added in 'asoc_simple_priv':
- 'ops' struct: In some cases SoC vendor drivers may want to
implement 'snd_soc_ops' callbacks differently. In such cases
custom callbacks would be used.
- 'force_dpcm' flag: Right now simple or graph card drivers
detect DAI links as DPCM links if:
* The dpcm_selectable is set AND
* Codec is connected to multiple CPU endpoints or aconvert
property is used for rate/channels.
So there is no way to directly specify usage of DPCM alone. So a
flag is exposed to mark all links as DPCM. Vendor driver can
set this if required.
- 'dpcm_selectable': Currently simple or audio graph drivers
provide a way to enable this for specific compatibles. However
vendor driver may want to define some additional info. Thus
expose this variable where vendor drivers can set this if
required.
Audio graph driver is updated to consider above flags or callbacks.
Subsequent patches in the series illustrate usage for above.
Signed-off-by: Sameer Pujar <spujar@nvidia.com>
Cc: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/1604329814-24779-7-git-send-email-spujar@nvidia.com
Signed-off-by: Mark Brown <broonie@kernel.org>
dpcm_end_walk_at_be() stops the graph walk when first BE is found for
the given FE component. In a component model we may want to connect
multiple DAIs from different components. A new flag is introduced in
'snd_soc_card', which when set allows DAI/component chaining. Later
PCM operations can be called for all these listed components for a
valid DAPM path.
Signed-off-by: Sameer Pujar <spujar@nvidia.com>
Link: https://lore.kernel.org/r/1604329814-24779-3-git-send-email-spujar@nvidia.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Hi Mark
soc_pcm_hw_params() does rollback when failed (A),
but, it is almost same as soc_pcm_hw_free().
static int soc_pcm_hw_params(xxx)
{
...
if (ret < 0)
goto xxx_err;
...
return ret;
^ component_err:
| ...
| interface_err:
(A) ...
| codec_err:
| ...
v return ret;
}
This kind of duplicated code can be a hotbed of bugs,
thus, this patch-set share soc_pcm_hw_free() and rollback.
Kuninori Morimoto (6):
ASoC: soc.h: remove for_each_rtd_dais_rollback()
ASoC: soc-pcm: move soc_pcm_hw_free() next to soc_pcm_hw_params()
ASoC: soc-link: add mark for snd_soc_link_hw_params/free()
ASoC: soc-component: add mark for snd_soc_pcm_component_hw_params/free()
ASoC: soc-dai: add mark for snd_soc_dai_hw_params/free()
ASoC: soc-pcm: add soc_pcm_hw_clean() and call it from soc_pcm_hw_params/free()
include/sound/soc-component.h | 6 +-
include/sound/soc-dai.h | 4 +-
include/sound/soc-link.h | 3 +-
include/sound/soc.h | 7 +-
sound/soc/soc-component.c | 19 ++---
sound/soc/soc-dai.c | 13 +++-
sound/soc/soc-dapm.c | 4 +-
sound/soc/soc-link.c | 12 +++-
sound/soc/soc-pcm.c | 131 ++++++++++++++--------------------
9 files changed, 97 insertions(+), 102 deletions(-)
--
2.25.1
The change removes the platform_data include/definition. It only contains
some values for the MICBIAS.
These are moved into 'dt-bindings/sound/adi,adau1977.h' so that they can be
used inside device-trees. When moving then, they need to be converted to
pre-compiler defines, so that the DT compiler can understand them.
The driver then, also needs to include the new
'dt-bindings/sound/adi,adau1977.h' file.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Link: https://lore.kernel.org/r/20201019105313.24862-1-alexandru.ardelean@analog.com
Signed-off-by: Mark Brown <broonie@kernel.org>
soc_pcm_hw_params() does rollback when failed (A),
but, it is almost same as soc_pcm_hw_free().
static int soc_pcm_hw_params(xxx)
{
...
if (ret < 0)
goto xxx_err;
...
return ret;
^ component_err:
| ...
| interface_err:
(A) ...
| codec_err:
| ...
v return ret;
}
The difference is
soc_pcm_hw_free() is for all dai/component/substream,
rollback is for succeeded part only.
This kind of duplicated code can be a hotbed of bugs,
thus, we want to share soc_pcm_hw_free() and rollback.
Now, soc_pcm_hw_params/free() are handling
1) snd_soc_link_hw_params/free()
2) snd_soc_pcm_component_hw_params/free()
3) snd_soc_dai_hw_params/free()
Now, 1) to 3) are handled.
This patch adds new soc_pcm_hw_clean() and call it from
soc_pcm_hw_params() as rollback, and from soc_pcm_hw_free() as
normal close handler.
Other difference is that soc_pcm_hw_free() handles digital mute
if it was last user. Rollback also handles it by this patch.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87h7rhgqab.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
soc_pcm_hw_params() does rollback when failed (A),
but, it is almost same as soc_pcm_hw_free().
static int soc_pcm_hw_params(xxx)
{
...
if (ret < 0)
goto xxx_err;
...
return ret;
^ component_err:
| ...
| interface_err:
(A) ...
| codec_err:
| ...
v return ret;
}
The difference is
soc_pcm_hw_free() is for all dai/component/substream,
rollback is for succeeded part only.
This kind of duplicated code can be a hotbed of bugs,
thus, we want to share soc_pcm_hw_free() and rollback.
Now, soc_pcm_hw_params/free() are handling
1) snd_soc_link_hw_params/free()
2) snd_soc_pcm_component_hw_params/free()
=> 3) snd_soc_dai_hw_params/free()
This patch is for 3) snd_soc_dai_hw_params/free().
The idea of having bit-flag or counter is not enough for this purpose.
For example if one DAI is used for 2xPlaybacks for some reasons,
and if 1st Playback was succeeded but 2nd Playback was failed,
2nd Playback rollback doesn't need to call shutdown.
But it has succeeded bit-flag or counter via 1st Playback,
thus, 2nd Playback rollback will call unneeded shutdown.
And 1st Playback's necessary shutdown will not be called,
because bit-flag or counter was cleared by wrong 2nd Playback rollback.
To avoid such case, this patch marks substream pointer when hw_params() was
succeeded. If rollback needed, it will check rollback flag and marked
substream pointer.
One note here is that it cares *previous* hw_params() only now,
but we might want to check *whole* marked substream in the future.
This patch is using macro named "push/pop", so that it can be easily
update.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87imbxgqai.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
soc_pcm_hw_params() does rollback when failed (A),
but, it is almost same as soc_pcm_hw_free().
static int soc_pcm_hw_params(xxx)
{
...
if (ret < 0)
goto xxx_err;
...
return ret;
^ component_err:
| ...
| interface_err:
(A) ...
| codec_err:
| ...
v return ret;
}
The difference is
soc_pcm_hw_free() is for all dai/component/substream,
rollback is for succeeded part only.
This kind of duplicated code can be a hotbed of bugs,
thus, we want to share soc_pcm_hw_free() and rollback.
Now, soc_pcm_hw_params/free() are handling
1) snd_soc_link_hw_params/free()
=> 2) snd_soc_pcm_component_hw_params/free()
3) snd_soc_dai_hw_params/free()
This patch is for 2) snd_soc_pcm_component_hw_params/free().
The idea of having bit-flag or counter is not enough for this purpose.
For example if one DAI is used for 2xPlaybacks for some reasons,
and if 1st Playback was succeeded but 2nd Playback was failed,
2nd Playback rollback doesn't need to call shutdown.
But it has succeeded bit-flag or counter via 1st Playback,
thus, 2nd Playback rollback will call unneeded shutdown.
And 1st Playback's necessary shutdown will not be called,
because bit-flag or counter was cleared by wrong 2nd Playback rollback.
To avoid such case, this patch marks substream pointer when hw_params() was
succeeded. If rollback needed, it will check rollback flag and marked
substream pointer.
One note here is that it cares *previous* hw_params() only now,
but we might want to check *whole* marked substream in the future.
This patch is using macro named "push/pop", so that it can be easily
update.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87k0wdgqav.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
soc_pcm_hw_params() does rollback when failed (A),
but, it is almost same as soc_pcm_hw_free().
static int soc_pcm_hw_params(xxx)
{
...
if (ret < 0)
goto xxx_err;
...
return ret;
^ component_err:
| ...
| interface_err:
(A) ...
| codec_err:
| ...
v return ret;
}
The difference is
soc_pcm_hw_free() is for all dai/component/substream,
rollback is for succeeded part only.
This kind of duplicated code can be a hotbed of bugs,
thus, we want to share soc_pcm_hw_free() and rollback.
Now, soc_pcm_hw_params/free() are handling
=> 1) snd_soc_link_hw_params/free()
2) snd_soc_pcm_component_hw_params/free()
3) snd_soc_dai_hw_params/free()
This patch is for 1) snd_soc_link_hw_params/free().
The idea of having bit-flag or counter is not enough for this purpose.
For example if one DAI is used for 2xPlaybacks for some reasons,
and if 1st Playback was succeeded but 2nd Playback was failed,
2nd Playback rollback doesn't need to call shutdown.
But it has succeeded bit-flag or counter via 1st Playback,
thus, 2nd Playback rollback will call unneeded shutdown.
And 1st Playback's necessary shutdown will not be called,
because bit-flag or counter was cleared by wrong 2nd Playback rollback.
To avoid such case, this patch marks substream pointer when hw_params() was
succeeded. If rollback needed, it will check rollback flag and marked
substream pointer.
One note here ist that it cares *previous* hw_params() only now,
but we might want to check *whole* marked substream in the future.
This patch is using macro named "push/pop", so that it can be easily
update.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87lfgtgqba.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Commit 453431a549 ("mm, treewide: rename kzfree() to
kfree_sensitive()") renamed kzfree() to kfree_sensitive(),
but it left a compatibility definition of kzfree() to avoid
being too disruptive.
Since then a few more instances of kzfree() have slipped in.
Just get rid of them and remove the compatibility definition
once and for all.
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull perf fix from Thomas Gleixner:
"A single fix to compute the field offset of the SNOOPX bit in the data
source bitmask of perf events correctly"
* tag 'perf-urgent-2020-10-25' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf: correct SNOOPX field offset
Pull locking fix from Thomas Gleixner:
"Just a trivial fix for kernel-doc warnings"
* tag 'locking-urgent-2020-10-25' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
locking/seqlocks: Fix kernel-doc warnings
Pull more parisc updates from Helge Deller:
- During this merge window O_NONBLOCK was changed to become 000200000,
but we missed that the syscalls timerfd_create(), signalfd4(),
eventfd2(), pipe2(), inotify_init1() and userfaultfd() do a strict
bit-wise check of the flags parameter.
To provide backward compatibility with existing userspace we
introduce parisc specific wrappers for those syscalls which filter
out the old O_NONBLOCK value and replaces it with the new one.
- Prevent HIL bus driver to get stuck when keyboard or mouse isn't
attached
- Improve error return codes when setting rtc time
- Minor documentation fix in pata_ns87415.c
* 'parisc-5.10-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
ata: pata_ns87415.c: Document support on parisc with superio chip
parisc: Add wrapper syscalls to fix O_NONBLOCK flag usage
hil/parisc: Disable HIL driver when it gets stuck
parisc: Improve error return codes when setting rtc time
Pull more xen updates from Juergen Gross:
- a series for the Xen pv block drivers adding module parameters for
better control of resource usge
- a cleanup series for the Xen event driver
* tag 'for-linus-5.10b-rc1c-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
Documentation: add xen.fifo_events kernel parameter description
xen/events: unmask a fifo event channel only if it was masked
xen/events: only register debug interrupt for 2-level events
xen/events: make struct irq_info private to events_base.c
xen: remove no longer used functions
xen-blkfront: Apply changed parameter name to the document
xen-blkfront: add a parameter for disabling of persistent grants
xen-blkback: add a parameter for disabling of persistent grants
Pull random32 updates from Willy Tarreau:
"Make prandom_u32() less predictable.
This is the cleanup of the latest series of prandom_u32
experimentations consisting in using SipHash instead of Tausworthe to
produce the randoms used by the network stack.
The changes to the files were kept minimal, and the controversial
commit that used to take noise from the fast_pool (f227e3ec3b) was
reverted. Instead, a dedicated "net_rand_noise" per_cpu variable is
fed from various sources of activities (networking, scheduling) to
perturb the SipHash state using fast, non-trivially predictable data,
instead of keeping it fully deterministic. The goal is essentially to
make any occasional memory leakage or brute-force attempt useless.
The resulting code was verified to be very slightly faster on x86_64
than what is was with the controversial commit above, though this
remains barely above measurement noise. It was also tested on i386 and
arm, and build- tested only on arm64"
Link: https://lore.kernel.org/netdev/20200808152628.GA27941@SDF.ORG/
* tag '20201024-v4-5.10' of git://git.kernel.org/pub/scm/linux/kernel/git/wtarreau/prandom:
random32: add a selftest for the prandom32 code
random32: add noise from network and scheduling activity
random32: make prandom_u32() output unpredictable
Pull block fixes from Jens Axboe:
- NVMe pull request from Christoph
- rdma error handling fixes (Chao Leng)
- fc error handling and reconnect fixes (James Smart)
- fix the qid displace when tracing ioctl command (Keith Busch)
- don't use BLK_MQ_REQ_NOWAIT for passthru (Chaitanya Kulkarni)
- fix MTDT for passthru (Logan Gunthorpe)
- blacklist Write Same on more devices (Kai-Heng Feng)
- fix an uninitialized work struct (zhenwei pi)"
- lightnvm out-of-bounds fix (Colin)
- SG allocation leak fix (Doug)
- rnbd fixes (Gioh, Guoqing, Jack)
- zone error translation fixes (Keith)
- kerneldoc markup fix (Mauro)
- zram lockdep fix (Peter)
- Kill unused io_context members (Yufen)
- NUMA memory allocation cleanup (Xianting)
- NBD config wakeup fix (Xiubo)
* tag 'block-5.10-2020-10-24' of git://git.kernel.dk/linux-block: (27 commits)
block: blk-mq: fix a kernel-doc markup
nvme-fc: shorten reconnect delay if possible for FC
nvme-fc: wait for queues to freeze before calling update_hr_hw_queues
nvme-fc: fix error loop in create_hw_io_queues
nvme-fc: fix io timeout to abort I/O
null_blk: use zone status for max active/open
nvmet: don't use BLK_MQ_REQ_NOWAIT for passthru
nvmet: cleanup nvmet_passthru_map_sg()
nvmet: limit passthru MTDS by BIO_MAX_PAGES
nvmet: fix uninitialized work for zero kato
nvme-pci: disable Write Zeroes on Sandisk Skyhawk
nvme: use queuedata for nvme_req_qid
nvme-rdma: fix crash due to incorrect cqe
nvme-rdma: fix crash when connect rejected
block: remove unused members for io_context
blk-mq: remove the calling of local_memory_node()
zram: Fix __zram_bvec_{read,write}() locking order
skd_main: remove unused including <linux/version.h>
sgl_alloc_order: fix memory leak
lightnvm: fix out-of-bounds write to array devices->info[]
...
Pull io_uring fixes from Jens Axboe:
- fsize was missed in previous unification of work flags
- Few fixes cleaning up the flags unification creds cases (Pavel)
- Fix NUMA affinities for completely unplugged/replugged node for io-wq
- Two fallout fixes from the set_fs changes. One local to io_uring, one
for the splice entry point that io_uring uses.
- Linked timeout fixes (Pavel)
- Removal of ->flush() ->files work-around that we don't need anymore
with referenced files (Pavel)
- Various cleanups (Pavel)
* tag 'io_uring-5.10-2020-10-24' of git://git.kernel.dk/linux-block:
splice: change exported internal do_splice() helper to take kernel offset
io_uring: make loop_rw_iter() use original user supplied pointers
io_uring: remove req cancel in ->flush()
io-wq: re-set NUMA node affinities if CPUs come online
io_uring: don't reuse linked_timeout
io_uring: unify fsize with def->work_flags
io_uring: fix racy REQ_F_LINK_TIMEOUT clearing
io_uring: do poll's hash_node init in common code
io_uring: inline io_poll_task_handler()
io_uring: remove extra ->file check in poll prep
io_uring: make cached_cq_overflow non atomic_t
io_uring: inline io_fail_links()
io_uring: kill ref get/drop in personality init
io_uring: flags-based creds init in queue
Pull misc vfs updates from Al Viro:
"Assorted stuff all over the place (the largest group here is
Christoph's stat cleanups)"
* 'work.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
fs: remove KSTAT_QUERY_FLAGS
fs: remove vfs_stat_set_lookup_flags
fs: move vfs_fstatat out of line
fs: implement vfs_stat and vfs_lstat in terms of vfs_fstatat
fs: remove vfs_statx_fd
fs: omfs: use kmemdup() rather than kmalloc+memcpy
[PATCH] reduce boilerplate in fsid handling
fs: Remove duplicated flag O_NDELAY occurring twice in VALID_OPEN_FLAGS
selftests: mount: add nosymfollow tests
Add a "nosymfollow" mount option.
Pull dma-mapping fixes from Christoph Hellwig:
- document the new dma_{alloc,free}_pages() API
- two fixups for the dma-mapping.h split
* tag 'dma-mapping-5.10-1' of git://git.infradead.org/users/hch/dma-mapping:
dma-mapping: document dma_{alloc,free}_pages
dma-mapping: move more functions to dma-map-ops.h
ARM/sa1111: add a missing include of dma-map-ops.h
With the removal of the interrupt perturbations in previous random32
change (random32: make prandom_u32() output unpredictable), the PRNG
has become 100% deterministic again. While SipHash is expected to be
way more robust against brute force than the previous Tausworthe LFSR,
there's still the risk that whoever has even one temporary access to
the PRNG's internal state is able to predict all subsequent draws till
the next reseed (roughly every minute). This may happen through a side
channel attack or any data leak.
This patch restores the spirit of commit f227e3ec3b ("random32: update
the net random state on interrupt and activity") in that it will perturb
the internal PRNG's statee using externally collected noise, except that
it will not pick that noise from the random pool's bits nor upon
interrupt, but will rather combine a few elements along the Tx path
that are collectively hard to predict, such as dev, skb and txq
pointers, packet length and jiffies values. These ones are combined
using a single round of SipHash into a single long variable that is
mixed with the net_rand_state upon each invocation.
The operation was inlined because it produces very small and efficient
code, typically 3 xor, 2 add and 2 rol. The performance was measured
to be the same (even very slightly better) than before the switch to
SipHash; on a 6-core 12-thread Core i7-8700k equipped with a 40G NIC
(i40e), the connection rate dropped from 556k/s to 555k/s while the
SYN cookie rate grew from 5.38 Mpps to 5.45 Mpps.
Link: https://lore.kernel.org/netdev/20200808152628.GA27941@SDF.ORG/
Cc: George Spelvin <lkml@sdf.org>
Cc: Amit Klein <aksecurity@gmail.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: "Jason A. Donenfeld" <Jason@zx2c4.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: tytso@mit.edu
Cc: Florian Westphal <fw@strlen.de>
Cc: Marc Plumb <lkml.mplumb@gmail.com>
Tested-by: Sedat Dilek <sedat.dilek@gmail.com>
Signed-off-by: Willy Tarreau <w@1wt.eu>
