| Message ID | 20180618101919.51973-1-mark.rutland@arm.com |
|---|---|
| Headers | show |
| Series | atomics: API cleanups | expand |
On Mon, Jun 18, 2018 at 11:19:01AM +0100, Mark Rutland wrote: > This series contains a few cleanups of the atomic API, fixing > inconsistencies between atomic_* and atomic64_*, and minimizing > repetition in arch code. This is nicer for arch code, and the improved > regularity will help when generating the atomic headers in future. Apart from the Alpha patch: Reviewed-by: Will Deacon <will.deacon@arm.com> I also tried to compare disassembly before/after, but the changes to add_unless made that quite fiddly for anything using it as a backend. Do you plan to move arm64 over to atomic-instrumented.h at some point? Will
On Mon, Jun 18, 2018 at 05:38:06PM +0100, Will Deacon wrote: > On Mon, Jun 18, 2018 at 11:19:01AM +0100, Mark Rutland wrote: > > This series contains a few cleanups of the atomic API, fixing > > inconsistencies between atomic_* and atomic64_*, and minimizing > > repetition in arch code. This is nicer for arch code, and the improved > > regularity will help when generating the atomic headers in future. > > Apart from the Alpha patch: > > Reviewed-by: Will Deacon <will.deacon@arm.com> Cheers! I assume that also holds with patch 7 fixes up to use s64. > I also tried to compare disassembly before/after, but the changes to > add_unless made that quite fiddly for anything using it as a backend. > > Do you plan to move arm64 over to atomic-instrumented.h at some point? That's the plan -- there's still a way to go with the generated atomic headers for that to work. Thanks, Mark.
On Mon, Jun 18, 2018 at 08:21:27PM +0100, Mark Rutland wrote: > On Mon, Jun 18, 2018 at 05:38:06PM +0100, Will Deacon wrote: > > On Mon, Jun 18, 2018 at 11:19:01AM +0100, Mark Rutland wrote: > > > This series contains a few cleanups of the atomic API, fixing > > > inconsistencies between atomic_* and atomic64_*, and minimizing > > > repetition in arch code. This is nicer for arch code, and the improved > > > regularity will help when generating the atomic headers in future. > > > > Apart from the Alpha patch: > > > > Reviewed-by: Will Deacon <will.deacon@arm.com> > > Cheers! I assume that also holds with patch 7 fixes up to use s64. I've pushed out the series with those fixes and your Reviewed-by tags. Given the whole series has your Reviewed-By and Peter's Acked-by, I assume that you're both happy for this to be queued? What's your prefered way for that to happen? Should I send a v4 with those fixes, a pull request, or are you happy to fetch that in a little while regardless? Thanks, Mark.
On Tue, Jun 19, 2018 at 11:18:01AM +0100, Mark Rutland wrote: > On Mon, Jun 18, 2018 at 08:21:27PM +0100, Mark Rutland wrote: > > On Mon, Jun 18, 2018 at 05:38:06PM +0100, Will Deacon wrote: > > > On Mon, Jun 18, 2018 at 11:19:01AM +0100, Mark Rutland wrote: > > > > This series contains a few cleanups of the atomic API, fixing > > > > inconsistencies between atomic_* and atomic64_*, and minimizing > > > > repetition in arch code. This is nicer for arch code, and the improved > > > > regularity will help when generating the atomic headers in future. > > > > > > Apart from the Alpha patch: > > > > > > Reviewed-by: Will Deacon <will.deacon@arm.com> > > > > Cheers! I assume that also holds with patch 7 fixes up to use s64. > > I've pushed out the series with those fixes and your Reviewed-by tags. > > Given the whole series has your Reviewed-By and Peter's Acked-by, I > assume that you're both happy for this to be queued? > > What's your prefered way for that to happen? Should I send a v4 with > those fixes, a pull request, or are you happy to fetch that in a little > while regardless? Probably best to send a v4, then Ingo can take it all via -tip. Before you do that, can you also spell-check your commit messages please? I spotted a bunch of silly typos, and it will save Ingo from having to fix them up if you do it first. Will
On Tue, Jun 19, 2018 at 11:20:49AM +0100, Will Deacon wrote: > On Tue, Jun 19, 2018 at 11:18:01AM +0100, Mark Rutland wrote: > > On Mon, Jun 18, 2018 at 08:21:27PM +0100, Mark Rutland wrote: > > > On Mon, Jun 18, 2018 at 05:38:06PM +0100, Will Deacon wrote: > > > > On Mon, Jun 18, 2018 at 11:19:01AM +0100, Mark Rutland wrote: > > > > > This series contains a few cleanups of the atomic API, fixing > > > > > inconsistencies between atomic_* and atomic64_*, and minimizing > > > > > repetition in arch code. This is nicer for arch code, and the improved > > > > > regularity will help when generating the atomic headers in future. > > > > > > > > Apart from the Alpha patch: > > > > > > > > Reviewed-by: Will Deacon <will.deacon@arm.com> > > > > > > Cheers! I assume that also holds with patch 7 fixes up to use s64. > > > > I've pushed out the series with those fixes and your Reviewed-by tags. > > > > Given the whole series has your Reviewed-By and Peter's Acked-by, I > > assume that you're both happy for this to be queued? > > > > What's your prefered way for that to happen? Should I send a v4 with > > those fixes, a pull request, or are you happy to fetch that in a little > > while regardless? > > Probably best to send a v4, then Ingo can take it all via -tip. Ok. > Before you do that, can you also spell-check your commit messages > please? I spotted a bunch of silly typos, and it will save Ingo from > having to fix them up if you do it first. Sorry about that. I've gone through those now, and fixed what I have spotted. I'll give it another pass before I post v4, just in case. :) Thanks, Mark.
This series contains a few cleanups of the atomic API, fixing inconsistencies between atomic_* and atomic64_*, and minimizing repetition in arch code. This is nicer for arch code, and the improved regularity will help when generating the atomic headers in future. Since v1 [1]: * Add missing inc_not_zero #define for x86_32 * Remove newly redundant test op definitions for riscv * Remove atomic_inc_not_zero_hint() * Make inc/dec ops optional Since v2 [2] * Make test ops return bool * Clean up ifdeffery for andnot opes * Update comments for clarity * Rebase to v4.18-rc1 * Accumulate acks * Fix typos The bulk of the patches reorganise things so architectures consistently provide <atomic>_fetch_add_unless(), with atomic_fetch_add_unless() provided as a wrapper by core code. A generic fallback is provided for <atomic>_fetch_add_unless(), based on <atomic>_read() and <atomic>_try_cmpxchg(). Other patches in the series add common fallbacks for: * atomic64_inc_not_zero() * <atomic>_inc_and_test() * <atomic>_dec_and_test() * <atomic>_sub_and_test() * <atomic>add_negative() * <atomic>_*inc*() * <atomic>_*dec*() ... as almost all architectures provide (near-)identical implementation of these today. Where an architecture provides a non-trivial definition, it is updated to provide a matching preprocessor symbol, and the instrumented atomics are updated correspondingly. The end result is a strongly negative diffstat, though <linux/atomic.h> grows by a reasonable amount. We could halve this in future by templating the various fallbacks for atomic{,64}_t with code generation. I've pushed this out to my atomics/api-cleanup branch [3] on kernel.org. Thanks, Mark. [1] https://lkml.kernel.org/r/20180523133533.1076-1-mark.rutland@arm.com [2] https://lkml.kernel.org/r/20180529154346.3168-1-mark.rutland@arm.com [3] git://git.kernel.org/pub/scm/linux/kernel/git/mark/linux.git atomics/api-cleanup Mark Rutland (18): atomics/treewide: s/__atomic_add_unless/atomic_fetch_add_unless/ atomics/treewide: remove redundant atomic_inc_not_zero() definitions atomics/treewide: remove atomic_inc_not_zero_hint() atomics: make conditional ops return bool atomics/treewide: make atomic64_inc_not_zero() optional atomics/treewide: make atomic_fetch_add_unless() optional atomics: prepare for atomic64_fetch_add_unless() atomics/generic: define atomic64_fetch_add_unless() atomics/alpha: define atomic64_fetch_add_unless() atomics/arc: define atomic64_fetch_add_unless() atomics/arm: define atomic64_fetch_add_unless() atomics/powerpc: define atomic64_fetch_add_unless() atomics/riscv: define atomic64_fetch_add_unless() atomics/treewide: make atomic64_fetch_add_unless() optional atomics/treewide: make test ops optional atomics/treewide: make unconditional inc/dec ops optional atomics/treewide: make conditional inc/dec ops optional atomics/treewide: clean up andnot ifdeffery arch/alpha/include/asm/atomic.h | 56 ++-- arch/arc/include/asm/atomic.h | 86 ++----- arch/arm/include/asm/atomic.h | 55 +--- arch/arm64/include/asm/atomic.h | 47 +--- arch/h8300/include/asm/atomic.h | 15 +- arch/hexagon/include/asm/atomic.h | 18 +- arch/ia64/include/asm/atomic.h | 81 ------ arch/m68k/include/asm/atomic.h | 24 +- arch/mips/include/asm/atomic.h | 172 ------------- arch/openrisc/include/asm/atomic.h | 4 +- arch/parisc/include/asm/atomic.h | 107 -------- arch/powerpc/include/asm/atomic.h | 52 ++-- arch/riscv/include/asm/atomic.h | 149 +---------- arch/s390/include/asm/atomic.h | 65 ----- arch/sh/include/asm/atomic.h | 35 --- arch/sparc/include/asm/atomic_32.h | 24 +- arch/sparc/include/asm/atomic_64.h | 65 +---- arch/sparc/lib/atomic32.c | 4 +- arch/x86/include/asm/atomic.h | 30 +-- arch/x86/include/asm/atomic64_32.h | 61 +---- arch/x86/include/asm/atomic64_64.h | 48 +--- arch/xtensa/include/asm/atomic.h | 98 ------- drivers/block/rbd.c | 2 +- drivers/infiniband/core/rdma_core.c | 2 +- fs/afs/rxrpc.c | 2 +- include/asm-generic/atomic-instrumented.h | 71 ++++- include/asm-generic/atomic.h | 33 --- include/asm-generic/atomic64.h | 15 +- include/linux/atomic.h | 415 +++++++++++++++++++++++------- kernel/bpf/syscall.c | 4 +- lib/atomic64.c | 12 +- net/atm/pppoatm.c | 2 +- net/rxrpc/call_object.c | 2 +- net/rxrpc/conn_object.c | 4 +- net/rxrpc/local_object.c | 2 +- net/rxrpc/peer_object.c | 2 +- 36 files changed, 511 insertions(+), 1353 deletions(-) -- 2.11.0