haiwan/ck - ck - Gitea: Git with a cup of tea

Commit Graph

Author	SHA1	Message	Date
Samy Al Bahra	9893f12bcf	ck_ht: ck_hm not making it to next release, remove warning.	9 years ago
Samy Al Bahra	2e75aefc4a	ck_epoch: Allow for forward progress in concurrent epoch sections. This work is derived directly from the work of John Esmet and Paul Khuong ({jesmet,pkhuong}@appnexus.com) and PR34.	9 years ago
Samy Al Bahra	0d6d384f3f	ck_epoch: Cache associated epoch state machine in record. This ends up removing arguments (and vector for error) for several functions on the common path of ck_epoch.	9 years ago
Paul Khuong	f1302c4920	ck_ring: explicit casts to clarify the semantics of generic (pointer) dequeuing	9 years ago
Samy Al Bahra	52f42c4392	ck_pr: Add ck_pr_fence_acqrel. Provides both acquire and release semantics.	9 years ago
Olivier Houchard	27fb1bc00f	ck: Reimplement a few libc headers, to make CK build when compiling in the FreeBSD kernel.	9 years ago
Samy Al Bahra	4677de8473	ck_ring: [whitespace] Add comment to PROTOTYPE family.	9 years ago
Samy Al Bahra	fee4e7187b	ck_ring: Clean-up internal implementations. Break out internal implementations to _mp and _sc variants from which public interface is built on. Do not rely on macro. Adopt CK_CC_RESTRICT instead of using restrict directly.	9 years ago
Samy Al Bahra	414ba224ca	ck_cc: Add FORCE_INLINE attribute and remove stale comments.	9 years ago
Samy Al Bahra	cc600e7d4f	ck_ring: Add MPMC and MPSC variants.	9 years ago
Samy Al Bahra	e88c91cdbd	ck_elide: Use CK_CC_INLINE instead of inline.	9 years ago
Olivier Houchard	37d94f4f4e	ck_pr/arm : Add load, store and CAS for double.	9 years ago
Michael Haberler	118cc8eae8	configure: detect git SHA, export as string CK_GIT_SHA in ck_md.h this helps correlating tests by other people	9 years ago
Michael Haberler	ce7d26aad4	ck_mg.h: provide printable version tag as CK_VERSION	9 years ago
Olivier Houchard	c0c0b19716	configure: Add --platform and --use-cc-builtins --platform let you set the platform, instead of relying on uname -m --use-cc-builtins force the usage of gcc atomic builtins, instead of using the one provided by CK.	9 years ago
Michael Haberler	abc8e871af	arm/c++: cast arguments using CK_CPP_CAST	9 years ago
Michael Haberler	9911e46c7d	x86_64/c++: cast arguments using CK_CPP_CAST	9 years ago
Michael Haberler	a9005e55f1	c++: introduce CK_CPP_CAST to static_cast arguments	9 years ago
Olivier Houchard	d2ee99b7f4	ck_pr/arm : Tell the compiler to which "1" he should jump to.	9 years ago
Olivier Houchard	576e65fc74	ck_ht: Fallback to use 32bits values when 64bits load/store isn't supported. That way we can use ck_ht on i386.	9 years ago
Samy Al Bahra	f6f02e8585	Merge pull request #54 from mhaberler/limits ck_limits.h: assure UINT*_MAX ff are available in-kernel	9 years ago
Samy Al Bahra	35f9ff1ede	Merge pull request #53 from mhaberler/typeof CK_PR_ACCESS: use __typeof__ instead of typeof	9 years ago
Samy Al Bahra	b039de2f2f	ck_cc: Add parenthesis to make default easier.	9 years ago
Michael Haberler	383eb139ff	ck_limits.h: assure UINT*_MAX ff are available in-kernel this should work back to at least 3.5 kernels	9 years ago
Michael Haberler	ac1c170b29	CK_PR_ACCESS: use __typeof__ instead of typeof this makes ck_pr.h safe to include in c++ with -std=c++11	9 years ago
Olivier Houchard	1c3501a524	ck_pr/arm : Don't try to build the 64bits version of CAS for armv6 either.	9 years ago
Olivier Houchard	005c06f9e2	ck_pr/arm: There's no need to force the use of r4, r5 and r6 anymore, this should fix issue #50	9 years ago
Olivier Houchard	b002ed5f11	ck_pr/arm: Don't use ldrexd/strexd if we're building for armv6.	9 years ago
Olivier Houchard	b202a8224a	ck_pr/arm: Rename ck_pr_[load\|store]_64 to ck_pr_md_[load\|store]_64, as it should have been done.	9 years ago
Emilio G. Cota	3f48bc39fb	ck_sequence: relax sequence increment from atomic to regular store The atomicity of the sequence number's increment is unnecessary, since there should be only one writer at any given time. Fix it by changing it for a regular increment + store. Signed-off-by: Emilio G. Cota <cota@braap.org>	10 years ago
Samy Al Bahra	3971541783	ppc: lwsync is the right thing for WB-memory. We will expose a flag to enable isync.	10 years ago
Samy Al Bahra	41c1080f43	ck_bytelock: Fall-back to store_8 if FAS is unavailable.	10 years ago
Samy Al Bahra	c2ce635080	spinlock/ticket: Remove dead variable.	10 years ago
Samy Al Bahra	a5b4ca9ffc	ck_pr/ppc64: Use unsafe for ugly aliasing hack. Once we migrate to using the wrappers, this ugliness won't be necessary. Last argument can be macro-expanded instead.	10 years ago
Samy Al Bahra	e693cb9afe	ck_*lock: Migrate all locking primitives to lock / unlock. This only affects RMO. This adds stricter semantics for critical section serialization. In addition to this, asymmetric synchronization primitives will now provide load ordering with respect to readers. This also modifies locked operations to have acquire semantics (they're there for elision predicates, and this doesn't impact them in any way). There are several performance improvements included in this as well (redundant fence was removed from days of wanting to support Alpha).	10 years ago
Samy Al Bahra	1537c8091d	ck_pr: Introduce ck_pr_fence_lock and fence_unlock. These primitives are meant to be used in lock implementations where control dependency ordering is sufficient to enforce ordering of critical section. At the moment, this only affects PPC. Currently, we rely on lwsync for entry into critical sections which is insufficient. sync is rather heavy-weight, and assuming we aren't falling victim into compiler re-ordering, isync should be sufficient. There is follow-up work to be done in ARM, as we may have cheaper (but target-specialized) ISB-tricks for load-load ordering.	10 years ago
Samy Al Bahra	787cffe9c9	ck_pr/arm: store_load requires DMB. Emit full synchronize for store_load.	10 years ago
Samy Al Bahra	b2407aa3f0	whitespace/mcs: Wrap to 80 columns.	10 years ago
Samy Al Bahra	060ef37e5e	whitespace/ck_bitmap: Fix minor typo in comments. Urban Dictionary reveals a particular interpretation of exclusing.	10 years ago
Samy Al Bahra	0e8d445049	ck_epoch: Specialize begin according to memory model. Previously, we were specializing for x86.	10 years ago
Samy Al Bahra	a26c891f1a	ck_hp_fifo: Migrate to ck_hp_set_fence. On Pentium M, this yields 30%-40% reduction in latency.	10 years ago
Samy Al Bahra	419ddca3ad	ck_hp_stack: Switch to ck_hp_set_fence.	10 years ago
Samy Al Bahra	9efb74da4b	ck_hp: Implement ck_hp_set_fence. On TSO architectures, this relies on atomic ordering guarantees rather than a full barrier. On Pentium M, this results in approximately 30% improvement in latency for stack.	10 years ago
Olivier Houchard	8b0bdb1901	ck_rhs: Introduce ck_rhs_set_load_factor() to change the load factor. The default value is still 50, but that may be revisited later. Also, pre-calculate the max number of entries before growing, toi avoid having to do it at each insert.	10 years ago
William Light	e9d96eb9f6	ck_ring: make a few arguments const	10 years ago
Olivier Houchard	477b37b249	gcc/ck_pr: I made up my mind, ck_pr_stale() has to provide a compiler barrier.	10 years ago
Olivier Houchard	a30c54f64a	gcc/ck_pr: Revert previous commit	10 years ago
Olivier Houchard	6dc4d2664f	gcc/ck_pr: ck_pr_stall() should act as a compiler barrier too.	10 years ago
Samy Al Bahra	63d6c00c3d	gcc/ck_pr: GCC-backed primitives should serve as compiler barriers. Olivier pointed out that he was seeing some load / store operations get optimized away and / or re-ordered otherwise.	10 years ago
Samy Al Bahra	8ee26212f9	ck_pr: Reference MD-implementation for CAS wrapper.	10 years ago

1 2 3 4 5 ...

600 Commits (0.5.0)