linux-toradex.git/kernel/padata.c, branch v6.4-rc1

padata: use alignment when calculating the number of worker threads

2023-03-14T09:06:44+00:00

For multithreaded jobs the computed chunk size is rounded up by the
caller-specified alignment. However, the number of worker threads to
use is computed using the minimum chunk size without taking alignment
into account. A sufficiently large alignment value can result in too
many worker threads being allocated for the job.

Signed-off-by: Anthony Yznaga 
Acked-by: Daniel Jordan 
Signed-off-by: Herbert Xu

padata: Make kobj_type structure constant

2023-03-14T09:06:42+00:00

Since commit ee6d3dd4ed48 ("driver core: make kobj_type constant.")
the driver core allows the usage of const struct kobj_type.

Take advantage of this to constify the structure definition to prevent
modification at runtime.

Signed-off-by: Thomas Weißschuh 
Acked-by: Daniel Jordan 
Signed-off-by: Herbert Xu

Merge tag 'kbuild-v6.2' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild

2022-12-19T18:33:32+00:00

Pull Kbuild updates from Masahiro Yamada:

 - Support zstd-compressed debug info

 - Allow W=1 builds to detect objects shared among multiple modules

 - Add srcrpm-pkg target to generate a source RPM package

 - Make the -s option detection work for future GNU Make versions

 - Add -Werror to KBUILD_CPPFLAGS when CONFIG_WERROR=y

 - Allow W=1 builds to detect -Wundef warnings in any preprocessed files

 - Raise the minimum supported version of binutils to 2.25

 - Use $(intcmp ...) to compare integers if GNU Make >= 4.4 is used

 - Use $(file ...) to read a file if GNU Make >= 4.2 is used

 - Print error if GNU Make older than 3.82 is used

 - Allow modpost to detect section mismatches with Clang LTO

 - Include vmlinuz.efi into kernel tarballs for arm64 CONFIG_EFI_ZBOOT=y

* tag 'kbuild-v6.2' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild: (29 commits)
  buildtar: fix tarballs with EFI_ZBOOT enabled
  modpost: Include '.text.*' in TEXT_SECTIONS
  padata: Mark padata_work_init() as __ref
  kbuild: ensure Make >= 3.82 is used
  kbuild: refactor the prerequisites of the modpost rule
  kbuild: change module.order to list *.o instead of *.ko
  kbuild: use .NOTINTERMEDIATE for future GNU Make versions
  kconfig: refactor Makefile to reduce process forks
  kbuild: add read-file macro
  kbuild: do not sort after reading modules.order
  kbuild: add test-{ge,gt,le,lt} macros
  Documentation: raise minimum supported version of binutils to 2.25
  kbuild: add -Wundef to KBUILD_CPPFLAGS for W=1 builds
  kbuild: move -Werror from KBUILD_CFLAGS to KBUILD_CPPFLAGS
  kbuild: Port silent mode detection to future gnu make.
  init/version.c: remove #include 
  firmware_loader: remove #include 
  modpost: Mark uuid_le type to be suitable only for MEI
  kbuild: add ability to make source rpm buildable using koji
  kbuild: warn objects shared among multiple modules
  ...

padata: Mark padata_work_init() as __ref

2022-12-14T06:49:24+00:00

When building arm64 allmodconfig + ThinLTO with clang and a proposed
modpost update to account for -ffuncton-sections, the following warning
appears:

  WARNING: modpost: vmlinux.o: section mismatch in reference: padata_work_init (section: .text.padata_work_init) -> padata_mt_helper (section: .init.text)
  WARNING: modpost: vmlinux.o: section mismatch in reference: padata_work_init (section: .text.padata_work_init) -> padata_mt_helper (section: .init.text)

LLVM has optimized padata_work_init() to include the address of
padata_mt_helper() directly because it inlined the other call to
padata_work_init() with padata_parallel_worker(), meaning the remaining
uses of padata_work_init() use padata_mt_helper() as the work_fn
argument. This optimization causes modpost to complain since
padata_work_init() is not __init, whereas padata_mt_helper() is.

Since padata_work_init() is only called from __init code when
padata_mt_helper() is passed as the work_fn argument, mark
padata_work_init() as __ref, which makes it clear to modpost that this
scenario is okay.

Suggested-by: Daniel Jordan 
Signed-off-by: Nathan Chancellor 
Acked-by: Daniel Jordan 
Signed-off-by: Masahiro Yamada

padata: Fix list iterator in padata_do_serial()

2022-11-25T09:39:18+00:00

list_for_each_entry_reverse() assumes that the iterated list is nonempty
and that every list_head is embedded in the same type, but its use in
padata_do_serial() breaks both rules.

This doesn't cause any issues now because padata_priv and padata_list
happen to have their list fields at the same offset, but we really
shouldn't be relying on that.

Fixes: bfde23ce200e ("padata: unbind parallel jobs from specific CPUs")
Signed-off-by: Daniel Jordan 
Signed-off-by: Herbert Xu

padata: Always leave BHs disabled when running ->parallel()

2022-11-25T09:39:18+00:00

A deadlock can happen when an overloaded system runs ->parallel() in the
context of the current task:

    padata_do_parallel
      ->parallel()
        pcrypt_aead_enc/dec
          padata_do_serial
            spin_lock(&reorder->lock) // BHs still enabled
              
                ...
                  __do_softirq
                    ...
                      padata_do_serial
                        spin_lock(&reorder->lock)

It's a bug for BHs to be on in _do_serial as Steffen points out, so
ensure they're off in the "current task" case like they are in
padata_parallel_worker to avoid this situation.

Reported-by: syzbot+bc05445bc14148d51915@syzkaller.appspotmail.com
Fixes: 4611ce224688 ("padata: allocate work structures for parallel jobs from a pool")
Signed-off-by: Daniel Jordan 
Acked-by: Steffen Klassert 
Signed-off-by: Herbert Xu

padata: replace cpumask_weight with cpumask_empty in padata.c

2022-01-31T00:21:46+00:00

padata_do_parallel() calls cpumask_weight() to check if any bit of a
given cpumask is set. We can do it more efficiently with cpumask_empty()
because cpumask_empty() stops traversing the cpumask as soon as it finds
first set bit, while cpumask_weight() counts all bits unconditionally.

Signed-off-by: Yury Norov 
Signed-off-by: Herbert Xu

padata: Remove repeated verbose license text

2021-08-27T08:30:18+00:00

remove it because SPDX-License-Identifier is already used

Signed-off-by: Cai Huoqing 
Acked-by: Daniel Jordan 
Signed-off-by: Herbert Xu

padata: Replace deprecated CPU-hotplug functions.

2021-08-12T11:16:59+00:00

The functions get_online_cpus() and put_online_cpus() have been
deprecated during the CPU hotplug rework. They map directly to
cpus_read_lock() and cpus_read_unlock().

Replace deprecated CPU-hotplug functions with the official version.
The behavior remains unchanged.

Cc: Steffen Klassert 
Cc: Daniel Jordan 
Cc: linux-crypto@vger.kernel.org
Signed-off-by: Sebastian Andrzej Siewior 
Acked-by: Daniel Jordan 
Signed-off-by: Herbert Xu

padata: Convert from atomic_t to refcount_t on parallel_data->refcnt

2021-07-30T02:58:36+00:00

refcount_t type and corresponding API can protect refcounters from
accidental underflow and overflow and further use-after-free situations.

Signed-off-by: Xiyu Yang 
Signed-off-by: Xin Tan 
Acked-by: Daniel Jordan 
Signed-off-by: Herbert Xu