A mirror of https://github.com/libressl/openbsd.git https://git.lua4.win/openbsd/atom?h=OPENBSD_7_6 2024-07-14T14:32:45+00:00 2024-07-14T14:32:45+00:00 jsing 2024-07-14T14:32:45+00:00 urn:sha1:1ae23ffb91ba41cf24c39b874e0d6f164063cfe3 lhash_local.h was previously needed since conf/conf_api.c and objects/obj_dat.c were fiddling with lhash internals when deleting via a callback. Since we no longer need to do that, inline the structs in lhash.c and remove the header. ok tb@ 2024-06-30T14:13:08+00:00 jsing 2024-06-30T14:13:08+00:00 urn:sha1:483a1f1e260c531359332afefc57fdeaca0e43bb These are not exactly useful and we previously stopped exposing them. ok tb@ 2024-06-22T16:38:31+00:00 jsing 2024-06-22T16:38:31+00:00 urn:sha1:0887e34f971690ea1c39507a417ab9af2dffd5cb ok tb@ 2024-05-07T13:40:42+00:00 jsing 2024-05-07T13:40:42+00:00 urn:sha1:e81acc84558973196803c486679341ec923e5414 No functional change. 2024-05-06T14:38:20+00:00 jsing 2024-05-06T14:38:20+00:00 urn:sha1:fc602bd5ac38ee7034363f1bb0b799a0453146dd 2024-05-06T14:36:05+00:00 jsing 2024-05-06T14:36:05+00:00 urn:sha1:dbeee3adcf988a67916e3ed200288c03f120385f It is not safe to unconditionally call contract() - when called repeatedly it will shrink the bucket array to zero and then attempt to access that allocation on the next call. Use the same guard that is used in lh_delete(). Issue found when investigating haproxy crashes reported by wizard-it on GitHub. ok tb@ 2024-03-02T11:11:11+00:00 tb 2024-03-02T11:11:11+00:00 urn:sha1:26367add3db68a3e89bda58a3c85174507f8e71a This removes internals of these two special snowflakes and will allow further simplifications. Unfortunately, there are some pieces of software that actually use LHASH_OF() (looking at you, pound, Ruby, and openssl(1)), so we get to keep exposing this garbage, at least for now. Expose lh_error() as a symbol to replace a macro reaching into _LHASH. lh_down_load() is no longer available. _LHASH and _STACK are now opaque, LHASH_NODE becomes internal-only. from jsing 2024-03-02T11:04:52+00:00 tb 2024-03-02T11:04:52+00:00 urn:sha1:792421b3f220abcbd1405485edc9bf6b39485769 This could have been removed in an earlier bump. Now it's time for it to say goodbye. ok jsing 2024-01-24T14:02:52+00:00 jsing 2024-01-24T14:02:52+00:00 urn:sha1:43687ee924fc2437d5b22b75c6b497ef06c60e4a Currently, the callback cannot safely delete entries as it could lead to contraction of the hash table, which in turn could lead to doall skipping entries (and that typically leads to memory leaks). The recommended workaround is to reach in and fiddle with the hash table internals in order to prevent contraction, call the doall function and then restore the internals that were changed. Rather than just improving our documentation, actually make it safe to delete entries from an lhash doall callback by pausing contractions prior to starting the callback loop, then restoring the down load factor and triggering contraction once completed. This means that callers no longer need access to change hash table internals in order to achieve this same behaviour. ok tb@ 2023-07-07T13:40:44+00:00 beck 2023-07-07T13:40:44+00:00 urn:sha1:1355f5b75c135c6bbe53ddedd21c86e2952b9895 ok jsing@