Regions will serve multiple purposes. First, with it we can decouple
ring/etc. object creation from registration / mapping of the memory they
will be placed in. We already have hacks that allow to put both SQ and
CQ into the same huge page, in the future we should be able to:
region = create_region(io_ring);
create_pbuf_ring(io_uring, region, offset=0);
create_pbuf_ring(io_uring, region, offset=N);
The second use case is efficiently passing parameters. The following
patch enables back on top of regions IORING_ENTER_EXT_ARG_REG, which
optimises wait arguments. It'll also be useful for request arguments
replacing iovecs, msghdr, etc. pointers. Eventually it would also be
handy for BPF as well if it comes to fruition.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/0798cf3a14fad19cfc96fc9feca5f3e11481691d.1731689588.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We've got a good number of mappings we share with the userspace, that
includes the main rings, provided buffer rings, upcoming rings for
zerocopy rx and more. All of them duplicate user argument parsing and
some internal details as well (page pinnning, huge page optimisations,
mmap'ing, etc.)
Introduce a notion of regions. For userspace for now it's just a new
structure called struct io_uring_region_desc which is supposed to
parameterise all such mapping / queue creations. A region either
represents a user provided chunk of memory, in which case the user_addr
field should point to it, or a request for the kernel to allocate the
memory, in which case the user would need to mmap it after using the
offset returned in the mmap_offset field. With a uniform userspace API
we can avoid additional boiler plate code and apply future optimisation
to all of them at once.
Internally, there is a new structure struct io_mapped_region holding all
relevant runtime information and some helpers to work with it. This
patch limits it to user provided regions.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/0e6fe25818dfbaebd1bd90b870a6cac503fe1a24.1731689588.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We're a bit too frivolous with types of nr_pages arguments, converting
it to long and back to int, passing an unsigned int pointer as an int
pointer and so on. Shouldn't cause any problem but should be carefully
reviewed, but until then let's add a WARN_ON_ONCE check to be more
confident callers don't pass poorely checked arguents.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/d48e0c097cbd90fb47acaddb6c247596510d8cfc.1731689588.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Use CLASS(fd) to get the file for sync message ring requests, rather
than open-code the file retrieval dance.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Link: https://lore.kernel.org/r/20241115034902.GP3387508@ZenIV
[axboe: make a more coherent commit message]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Clang warns (or errors with CONFIG_WERROR=y):
fs/ubifs/journal.c:986:20: error: variable 'err' is uninitialized when used here [-Werror,-Wuninitialized]
986 | ubifs_ro_mode(c, err);
| ^~~
Set err to -EPERM before the call to ubifs_ro_mode() and reuse it in the
return statement to resolve the warning.
Fixes: 957e1c4e17 ("ubifs: ubifs_jnl_write_inode: Only check once for the limitation of xattr count")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Richard Weinberger <richard@nod.at>
Here are the USB-serial updates for 6.13-rc1, including:
- improved support for quirky pl2303 hxd devices
- make sure ftdi_sio TIOCGSERIAL returns consistent data
Everything has been in linux-next with no reported issues.
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQQHbPq+cpGvN/peuzMLxc3C7H1lCAUCZzdTXQAKCRALxc3C7H1l
COSGAP0U6gSnhn9V6cSzz9WVJwfAAifKkBB6ZvPORmOX4f5nMwEAtEEQJQtdCXD6
d/9/9hEF1Xot5nhKfOTtFMbwWM+rtgI=
=gbhm
-----END PGP SIGNATURE-----
Merge tag 'usb-serial-6.13-rc1' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/johan/usb-serial into usb-next
Johan writes:
USB-serial updates for 6.13-rc1
Here are the USB-serial updates for 6.13-rc1, including:
- improved support for quirky pl2303 hxd devices
- make sure ftdi_sio TIOCGSERIAL returns consistent data
Everything has been in linux-next with no reported issues.
* tag 'usb-serial-6.13-rc1' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/johan/usb-serial:
USB: serial: ftdi_sio: Fix atomicity violation in get_serial_info()
USB: serial: pl2303: account for deficits of clones
For unbound workqueue, pwqs usually map to just a few pools. Most of
the time, pwqs will be linked sequentially to wq->pwqs list by cpu
index. Usually, consecutive CPUs have the same workqueue attribute
(e.g. belong to the same NUMA node). This makes pwqs with the same
pool cluster together in the pwq list.
Only do lock/unlock if the pool has changed in flush_workqueue_prep_pwqs().
This reduces the number of expensive lock operations.
The performance data shows this change boosts FIO by 65x in some cases
when multiple concurrent threads write to xfs mount points with fsync.
FIO Benchmark Details
- FIO version: v3.35
- FIO Options: ioengine=libaio,iodepth=64,norandommap=1,rw=write,
size=128M,bs=4k,fsync=1
- FIO Job Configs: 64 jobs in total writing to 4 mount points (ramdisks
formatted as xfs file system).
- Kernel Codebase: v6.12-rc5
- Test Platform: Xeon 8380 (2 sockets)
Reviewed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Wangyang Guo <wangyang.guo@intel.com>
Reviewed-by: Lai Jiangshan <jiangshanlai@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
When CONFIG_CMDLINE_EXTEND is set, the core kernel command line handling
logic appends CONFIG_CMDLINE to the bootloader provided command line.
The EFI stub does the opposite, and parses the builtin one first.
The usual behavior of command line options is that the last one takes
precedence if it appears multiple times, unless there is a meaningful
way to combine them. In either case, parsing the builtin command line
first while the core kernel does it in the opposite order is likely to
produce inconsistent results in such cases.
Therefore, switch the order in the stub to match the core kernel.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Kexec bypasses EFI's switch to virtual mode. In exchange, it has its own
routine, kexec_enter_virtual_mode(), which replays the mappings made by
the original kernel. Unfortunately, that function fails to reinstate
EFI's memory attributes, which would've otherwise been set after
entering virtual mode. Remediate this by calling
efi_runtime_update_mappings() within kexec's routine.
Signed-off-by: Nicolas Saenz Julienne <nsaenz@amazon.com>
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Drop support for the EFI_PROPERTIES_TABLE. It was a failed, short-lived
experiment that broke the boot both on Linux and Windows, and was
replaced by the EFI_MEMORY_ATTRIBUTES_TABLE shortly after.
Suggested-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Nicolas Saenz Julienne <nsaenz@amazon.com>
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
When running bpf selftest (./test_progs -j), the following warnings
showed up:
$ ./test_progs -t arena_atomics
...
BUG: using smp_processor_id() in preemptible [00000000] code: kworker/u19:0/12501
caller is bpf_mem_free+0x128/0x330
...
Call Trace:
<TASK>
dump_stack_lvl
check_preemption_disabled
bpf_mem_free
range_tree_destroy
arena_map_free
bpf_map_free_deferred
process_scheduled_works
...
For selftests arena_htab and arena_list, similar smp_process_id() BUGs are
dumped, and the following are two stack trace:
<TASK>
dump_stack_lvl
check_preemption_disabled
bpf_mem_alloc
range_tree_set
arena_map_alloc
map_create
...
<TASK>
dump_stack_lvl
check_preemption_disabled
bpf_mem_alloc
range_tree_clear
arena_vm_fault
do_pte_missing
handle_mm_fault
do_user_addr_fault
...
Add migrate_{disable,enable}() around related bpf_mem_{alloc,free}()
calls to fix the issue.
Fixes: b795379757 ("bpf: Introduce range_tree data structure and use it in bpf arena")
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20241115060354.2832495-1-yonghong.song@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
If it fails we need to check what was the reason, what were the lines
that didn't match the expected format, so:
root@number:~# perf test -v "trace exit race"
--- start ---
test child forked, pid 2028724
Lines not matching the expected regexp: ' +[0-9]+\.[0-9]+ +true/[0-9]+ syscalls:sys_enter_exit_group\(\)$':
0.000 :2028750/2028750 syscalls:sys_enter_exit_group()
---- end(-1) ----
110: perf trace exit race : FAILED!
root@number:~#
In this case we're not resolving the process COMM for some reason and
fallback to printing just the pid/tid, this will be fixed in a followup
patch.
Howard Chu spotted a problem with single code surrounding a regexp, that
made the test always fail, but since there were some failures when I
tested (COMM not being resolved in some of the results) the end inverse
grep would show some lines and thus didn't notice the single quote
problem.
He also provided a patch to test if less than the number of expected
matches took place but all of them with the expected output, in which
case the inverse grep wouldn't show anything, confusing the tester.
Reviewed-by: Howard Chu <howardchu95@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Benjamin Peterson <benjamin@engflow.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/lkml/ZzdknoHqrJbojb6P@x1
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
- Fix a missing of_node_put() in RCPM
- Fix a missing error code on failure in CPM1 QMC
- Switch to using for_each_available_child_of_node_scoped() in CPM1 TSA
-----BEGIN PGP SIGNATURE-----
iJIEABYKADoWIQQQ/+b4s5DeF6zCYyNoqS/rAbjdeAUCZzYBFRwcY2hyaXN0b3Bo
ZS5sZXJveUBjc2dyb3VwLmV1AAoJEGipL+sBuN14vbcBALObPpE0CcTKWee4PMKu
iEMnznKztaety2HsU8EVmYVPAP9nYGA/3EkGUMiEwh2+Kc/0/pIIIXRveCt2xWq4
fBOACQ==
=WD3l
-----END PGP SIGNATURE-----
gpgsig -----BEGIN PGP SIGNATURE-----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=6Rkj
-----END PGP SIGNATURE-----
Merge tag 'soc_fsl-6.13-1' of https://github.com/chleroy/linux into soc/drivers
FSL SOC changes for 6.13:
- Fix a missing of_node_put() in RCPM
- Fix a missing error code on failure in CPM1 QMC
- Switch to using for_each_available_child_of_node_scoped() in CPM1 TSA
* tag 'soc_fsl-6.13-1' of https://github.com/chleroy/linux:
soc: fsl: cpm1: qmc: Set the ret error code on platform_get_irq() failure
soc: fsl: rcpm: fix missing of_node_put() in copy_ippdexpcr1_setting()
soc: fsl: cpm1: tsa: switch to for_each_available_child_of_node_scoped()
Link: https://lore.kernel.org/r/c3c4961b-fe2a-4fcc-a7a1-f8b5352e09a2@csgroup.eu
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
A previous commit changed how requests are linked in the plug structure,
but unlike the previous method, it uses a new type for it rather than
struct request. The latter is available even for !CONFIG_BLOCK, while
struct rq_list is now. Move it outside CONFIG_BLOCK.
Reported-by: Nathan Chancellor <nathan@kernel.org>
Fixes: a3396b9999 ("block: add a rq_list type")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The ordering in hdmi_codec_get_ch_alloc_table_idx() results in
wrong channel allocation for a number of cases, e.g. when ELD
reports FL|FR|LFE|FC|RL|RR or FL|FR|LFE|FC|RL|RR|RC|RLC|RRC:
ca_id 0x01 with speaker mask FL|FR|LFE is selected instead of
ca_id 0x03 with speaker mask FL|FR|LFE|FC for 4 channels
and
ca_id 0x04 with speaker mask FL|FR|RC gets selected instead of
ca_id 0x0b with speaker mask FL|FR|LFE|FC|RL|RR for 6 channels
Fix this by reordering the channel allocation list with most
specific speaker masks at the top.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Signed-off-by: Christian Hewitt <christianshewitt@gmail.com>
Link: https://patch.msgid.link/20241115044344.3510979-1-christianshewitt@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
If the 'realloc' fails, the thermal zones pointer is set to NULL. This
makes all thermal zones references which were previously successfully
initialized to be lost.
[dlezcano] : Fixed indentation
Signed-off-by: zhang jiao <zhangjiao2@cmss.chinamobile.com>
Link: https://lore.kernel.org/r/20241114084039.42149-1-zhangjiao2@cmss.chinamobile.com
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
One alternative to the fix Christian proposed in
https://lore.kernel.org/dri-devel/20241024124159.4519-3-christian.koenig@amd.com/
is to replace the rather complex open coded sorting loops with the kernel
standard sort followed by a context squashing pass.
Proposed advantage of this would be readability but one concern Christian
raised was that there could be many fences, that they are typically mostly
sorted, and so the kernel's heap sort would be much worse by the proposed
algorithm.
I had a look running some games and vkcube to see what are the typical
number of input fences. Tested scenarios:
1) Hogwarts Legacy under Gamescope
450 calls per second to __dma_fence_unwrap_merge.
Percentages per number of fences buckets, before and after checking for
signalled status, sorting and flattening:
N Before After
0 0.91%
1 69.40%
2-3 28.72% 9.4% (90.6% resolved to one fence)
4-5 0.93%
6-9 0.03%
10+
2) Cyberpunk 2077 under Gamescope
1050 calls per second, amounting to 0.01% CPU time according to perf top.
N Before After
0 1.13%
1 52.30%
2-3 40.34% 55.57%
4-5 1.46% 0.50%
6-9 2.44%
10+ 2.34%
3) vkcube under Plasma
90 calls per second.
N Before After
0
1
2-3 100% 0% (Ie. all resolved to a single fence)
4-5
6-9
10+
In the case of vkcube all invocations in the 2-3 bucket were actually
just two input fences.
From these numbers it looks like the heap sort should not be a
disadvantage, given how the dominant case is <= 2 input fences which heap
sort solves with just one compare and swap. (And for the case of one input
fence we have a fast path in the previous patch.)
A complementary possibility is to implement a different sorting algorithm
under the same API as the kernel's sort() and so keep the simplicity,
potentially moving the new sort under lib/ if it would be found more
widely useful.
v2:
* Hold on to fence references and reduce commentary. (Christian)
* Record and use latest signaled timestamp in the 2nd loop too.
* Consolidate zero or one fences fast paths.
v3:
* Reverse the seqno sort order for a simpler squashing pass. (Christian)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Fixes: 245a4a7b53 ("dma-buf: generalize dma_fence unwrap & merging v3")
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3617
Cc: Christian König <christian.koenig@amd.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Cc: Gustavo Padovan <gustavo@padovan.org>
Cc: Friedrich Vock <friedrich.vock@gmx.de>
Cc: linux-media@vger.kernel.org
Cc: dri-devel@lists.freedesktop.org
Cc: linaro-mm-sig@lists.linaro.org
Cc: <stable@vger.kernel.org> # v6.0+
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20241115102153.1980-3-tursulin@igalia.com
Release all fence references if the output dma-fence-array could not be
allocated.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Fixes: 245a4a7b53 ("dma-buf: generalize dma_fence unwrap & merging v3")
Cc: Christian König <christian.koenig@amd.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Cc: Gustavo Padovan <gustavo@padovan.org>
Cc: Friedrich Vock <friedrich.vock@gmx.de>
Cc: linux-media@vger.kernel.org
Cc: dri-devel@lists.freedesktop.org
Cc: linaro-mm-sig@lists.linaro.org
Cc: <stable@vger.kernel.org> # v6.0+
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20241115102153.1980-2-tursulin@igalia.com
In cesa/cipher.c most declarations of struct mv_cesa_op_ctx are uninitialized.
This causes one of the values in the struct to be left unitialized in later
usages.
This patch fixes it by adding initializations in the same way it is done in
cesa/hash.c.
Fixes errors discovered in coverity: 1600942, 1600939, 1600935, 1600934, 1600929, 1600927,
1600925, 1600921, 1600920, 1600919, 1600915, 1600914
Signed-off-by: Karol Przybylski <karprzy7@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
If do_cpt_init() fails, a previous dma_alloc_coherent() call needs to be
undone.
Add the needed dma_free_coherent() before returning.
Fixes: 9e2c7d9994 ("crypto: cavium - Add Support for Octeon-tx CPT Engine")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
This patch reverts commit 0fbafd06bd
("crypto: aesni - fix failing setkey for rfc4106-gcm-aesni") by
moving the aesni init function back to module_init from late_initcall.
The original patch was needed because tests were synchronous. This
is no longer the case so there is no need to postpone the registration.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
This function is part of the exposed API and should be exported.
Otherwise a modular user would fail to build, e.g., crypto/rsa.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
A hwcap feature bit is passed to cpu_has_feature, resulting in testing
for CPU_FTR_MMCRA instead of the 3.1 platform revision.
Fixes: c954b252de ("crypto: powerpc/p10-aes-gcm - Register modules as SIMD")
Reported-by: Nicolai Stange <nstange@suse.com>
Signed-off-by: Michal Suchanek <msuchanek@suse.de>
Acked-by: Michael Ellerman <mpe@ellerman.id.au> (powerpc)
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Commit 62f8f307c8 ("powerpc/64: Remove maple platform") removes the
PPC_MAPLE config as a consequence of the platform’s removal.
The config definition of HW_RANDOM_AMD refers to this removed config option
in its dependencies.
Remove the reference to the removed config option.
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@redhat.com>
Acked-by: Michael Ellerman <mpe@ellerman.id.au> (powerpc)
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The CRC-T10DIF algorithm produces a 16-bit CRC, and this is reflected in
the folding coefficients, which are also only 16 bits wide.
This means that the polynomial multiplications involving these
coefficients can be performed using 8-bit long polynomial multiplication
(8x8 -> 16) in only a few steps, and this is an instruction that is part
of the base NEON ISA, which is all most real ARMv7 cores implement. (The
64-bit PMULL instruction is part of the crypto extensions, which are
only implemented by 64-bit cores)
The final reduction is a bit more involved, but we can delegate that to
the generic CRC-T10DIF implementation after folding the entire input
into a 16 byte vector.
This results in a speedup of around 6.6x on Cortex-A72 running in 32-bit
mode. On Cortex-A8 (BeagleBone White), the results are substantially
better than that, but not sufficiently reproducible (with tcrypt) to
quote a number here.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
To allow an alternative version to be created of the PMULL based
CRC-T10DIF algorithm, turn the bulk of it into a macro, except for the
final reduction, which will only be used by the existing version.
Reviewed-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The only remaining user of the fallback implementation of 64x64
polynomial multiplication using 8x8 PMULL instructions is the final
reduction from a 16 byte vector to a 16-bit CRC.
The fallback code is complicated and messy, and this reduction has
little impact on the overall performance, so instead, let's calculate
the final CRC by passing the 16 byte vector to the generic CRC-T10DIF
implementation when running the fallback version.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The CRC-T10DIF implementation for arm64 has a version that uses 8x8
polynomial multiplication, for cores that lack the crypto extensions,
which cover the 64x64 polynomial multiplication instruction that the
algorithm was built around.
This fallback version rather naively adopted the 64x64 polynomial
multiplication algorithm that I ported from ARM for the GHASH driver,
which needs 8 PMULL8 instructions to implement one PMULL64. This is
reasonable, given that each 8-bit vector element needs to be multiplied
with each element in the other vector, producing 8 vectors with partial
results that need to be combined to yield the correct result.
However, most PMULL64 invocations in the CRC-T10DIF code involve
multiplication by a pair of 16-bit folding coefficients, and so all the
partial results from higher order bytes will be zero, and there is no
need to calculate them to begin with.
Then, the CRC-T10DIF algorithm always XORs the output values of the
PMULL64 instructions being issued in pairs, and so there is no need to
faithfully implement each individual PMULL64 instruction, as long as
XORing the results pairwise produces the expected result.
Implementing these improvements results in a speedup of 3.3x on low-end
platforms such as Raspberry Pi 4 (Cortex-A72)
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
This is a partial revert of commit fc754c024a, which moved the logic
into C code which ensures that kernel mode NEON code does not hog the
CPU for too long.
This is no longer needed now that kernel mode NEON no longer disables
preemption, so we can drop this.
Reviewed-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The ahash_init functions may return fails. The ahash_hmac_init should
not return ok when ahash_init returns error. For an example, ahash_init
will return -ENOMEM when allocation memory is error.
Fixes: 9d12ba86f8 ("crypto: brcm - Add Broadcom SPU driver")
Signed-off-by: Chen Ridong <chenridong@huawei.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The caam_rsa_set_priv_key_form did not check for memory allocation errors.
Add the checks to the caam_rsa_set_priv_key_form functions.
Fixes: 52e26d77b8 ("crypto: caam - add support for RSA key form 2")
Signed-off-by: Chen Ridong <chenridong@huawei.com>
Reviewed-by: Gaurav Jain <gaurav.jain@nxp.com>
Reviewed-by: Horia Geantă <horia.geanta@nxp.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Hitherto, these operations have been converted in user space to
mask-and-xor operations on one register and two immediate values, and it
is the latter which have been evaluated by the kernel. We add support
for evaluating these operations directly in kernel space on one register
and either an immediate value or a second register.
Pablo made a few changes to the original patch:
- EINVAL if NFTA_BITWISE_SREG2 is used with fast version.
- Allow _AND,_OR,_XOR with _DATA != sizeof(u32)
- Dump _SREG2 or _DATA with _AND,_OR,_XOR
Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
There are reports [0] of cases where a corrupt EFI Memory Attributes
Table leads to out of memory issues at boot because the descriptor size
and entry count in the table header are still used to reserve the entire
table in memory, even though the resulting region is gigabytes in size.
Given that the EFI Memory Attributes Table is supposed to carry up to 3
entries for each EfiRuntimeServicesCode region in the EFI memory map,
and given that there is no reason for the descriptor size used in the
table to exceed the one used in the EFI memory map, 3x the size of the
entire EFI memory map is a reasonable upper bound for the size of this
table. This means that sizes exceeding that are highly likely to be
based on corrupted data, and the table should just be ignored instead.
[0] https://bugzilla.suse.com/show_bug.cgi?id=1231465
Cc: Gregory Price <gourry@gourry.net>
Cc: Usama Arif <usamaarif642@gmail.com>
Acked-by: Jiri Slaby <jirislaby@kernel.org>
Acked-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/all/20240912155159.1951792-2-ardb+git@google.com/
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Eric Sandeen <sandeen@redhat.com> says:
This is lightly tested with the kernel tests present in ecryptfs-utils,
but it could certainly use a bit more testing and review, particularly
with invalid mount option sets.
This one is a little unique compared to other filesystems in that I
allocate both an fs context and the *sbi in .init_fs_context; the *sbi
is long-lived, and the context is only present during the initial mount.
Allocating sbi with the filesystem context means we can set options
into it directly, rather than needing to do it after parsing. And it's
particularly simple to do it this way given that there is no remount.
* patches from https://lore.kernel.org/r/20241028143359.605061-1-sandeen@redhat.com:
ecryptfs: Convert ecryptfs to use the new mount API
ecryptfs: Factor out mount option validation
Link: https://lore.kernel.org/r/20241028143359.605061-1-sandeen@redhat.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
Under the new mount API, mount options are parsed one at a time.
Any validation that examines multiple options must be done after parsing
is complete, so factor out a ecryptfs_validate_options() which can be
called separately.
To facilitate this, temporarily move the local variables that tracked
whether various options have been set in the parsing function, into the
ecryptfs_mount_crypt_stat structure so that they can be examined later.
These will be moved to a more ephemeral struct in the mount api conversion
patch to follow.
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Link: https://lore.kernel.org/r/20241028143359.605061-2-sandeen@redhat.com
Acked-by: Tyler Hicks <code@tyhicks.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
Amir Goldstein <amir73il@gmail.com> says:
These patches bring the NFS connectable file handles feature to
userspace servers.
They rely on Christian's and Aleksa's changes recently merged to v6.12.
The API I chose for encoding conenctable file handles is pretty
conventional (AT_HANDLE_CONNECTABLE).
open_by_handle_at(2) does not have AT_ flags argument, but also, I find
it more useful API that encoding a connectable file handle can mandate
the resolving of a connected fd, without having to opt-in for a
connected fd independently.
I chose to implemnent this by using upper bits in the handle type field
It may be that out-of-tree filesystems return a handle type with upper
bits set, but AFAIK, no in-tree filesystem does that.
I added some warnings just in case we encouter that.
I have written an fstest [1] and a man page draft [2] for the feature.
[1] https://github.com/amir73il/xfstests/commits/connectable-fh/
[2] https://github.com/amir73il/man-pages/commits/connectable-fh/
* patches from https://lore.kernel.org/r/20241011090023.655623-1-amir73il@gmail.com:
fs: open_by_handle_at() support for decoding "explicit connectable" file handles
fs: name_to_handle_at() support for "explicit connectable" file handles
fs: prepare for "explicit connectable" file handles
Link: https://lore.kernel.org/r/20241011090023.655623-1-amir73il@gmail.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
Teach open_by_handle_at(2) about the type format of "explicit connectable"
file handles that were created using the AT_HANDLE_CONNECTABLE flag to
name_to_handle_at(2).
When decoding an "explicit connectable" file handles, name_to_handle_at(2)
should fail if it cannot open a "connected" fd with known path, which is
accessible (to capable user) from mount fd path.
Note that this does not check if the path is accessible to the calling
user, just that it is accessible wrt the mount namesapce, so if there
is no "connected" alias, or if parts of the path are hidden in the
mount namespace, open_by_handle_at(2) will return -ESTALE.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Link: https://lore.kernel.org/r/20241011090023.655623-4-amir73il@gmail.com
Fixes: 570df4e9c2 ("ceph: snapshot nfs re-export")
Acked-by:
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
nfsd encodes "connectable" file handles for the subtree_check feature,
which can be resolved to an open file with a connected path.
So far, userspace nfs server could not make use of this functionality.
Introduce a new flag AT_HANDLE_CONNECTABLE to name_to_handle_at(2).
When used, the encoded file handle is "explicitly connectable".
The "explicitly connectable" file handle sets bits in the high 16bit of
the handle_type field, so open_by_handle_at(2) will know that it needs
to open a file with a connected path.
old kernels will now recognize the handle_type with high bits set,
so "explicitly connectable" file handles cannot be decoded by
open_by_handle_at(2) on old kernels.
The flag AT_HANDLE_CONNECTABLE is not allowed together with either
AT_HANDLE_FID or AT_EMPTY_PATH.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Link: https://lore.kernel.org/r/20241011090023.655623-3-amir73il@gmail.com
Fixes: 570df4e9c2 ("ceph: snapshot nfs re-export")
Acked-by:
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
We would like to use the high 16bit of the handle_type field to encode
file handle traits, such as "connectable".
In preparation for this change, make sure that filesystems do not return
a handle_type value with upper bits set and that the open_by_handle_at(2)
syscall rejects these handle types.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Link: https://lore.kernel.org/r/20241011090023.655623-2-amir73il@gmail.com
Fixes: 570df4e9c2 ("ceph: snapshot nfs re-export")
Acked-by:
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Christian Brauner <brauner@kernel.org>
In the next patch we add support for doing AND, OR and XOR operations
directly in the kernel, so rename some functions and an enum constant
related to mask-and-xor boolean operations.
Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Use ip4h_dscp() instead of reading iph->tos directly.
ip4h_dscp() returns a dscp_t value which is temporarily converted back
to __u8 with inet_dscp_to_dsfield(). When converting ->flowi4_tos to
dscp_t in the future, we'll only have to remove that
inet_dscp_to_dsfield() call.
Signed-off-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Use ip4h_dscp() instead of reading iph->tos directly.
ip4h_dscp() returns a dscp_t value which is temporarily converted back
to __u8 with inet_dscp_to_dsfield(). When converting ->flowi4_tos to
dscp_t in the future, we'll only have to remove that
inet_dscp_to_dsfield() call.
Signed-off-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Use ip4h_dscp() instead of reading iph->tos directly.
ip4h_dscp() returns a dscp_t value which is temporarily converted back
to __u8 with inet_dscp_to_dsfield(). When converting ->flowi4_tos to
dscp_t in the future, we'll only have to remove that
inet_dscp_to_dsfield() call.
Signed-off-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
