427 commits

Author	SHA1	Message	Date
dmiller	538764bb6a	OpenSSL 4.0 compatibility. Fixes #3375	2026-06-16 13:58:41 +00:00
dmiller	b7e3657ec7	Make some more parameters const	2026-06-10 21:39:42 +00:00
dmiller	947f53c079	Use 1-byte alignment/packing for socks4. Fixes #2595	2026-05-26 01:52:17 +00:00
dmiller	082894dcad	Try to loosen OpenSSL security. Fixes #583	2026-05-12 21:53:50 +00:00
dmiller	c03bdbcc18	Use unsigned lengths to match libpcap	2026-04-30 22:01:59 +00:00
dmiller	7baee8efa9	Avoid leaking pending events in nsock_pool_delete	2026-04-15 03:21:56 +00:00
dmiller	209da5174d	Track and unload OpenSSL providers	2026-04-06 21:33:44 +00:00
dmiller	c693c60a54	Fix leak of canceled events	2026-04-06 21:33:44 +00:00
dmiller	61f7a8d3ca	Fix memory leak in proxychain parsing	2026-04-06 21:33:43 +00:00
dmiller	deb076224e	Reintegrate Nmap 7.99 release branch	2026-03-27 13:52:02 +00:00
dmiller	25a76d7876	Fix compilation with OpenSSL 1.0.1	2026-02-25 20:43:12 +00:00
dmiller	a16b2848ef	fix c99-ism in nsock test	2026-02-25 20:41:19 +00:00
dmiller	95d380b836	Bump copyright date to 2026 [skip ci]	2026-02-16 22:33:40 +00:00
dmiller	02b7df4d83	Fix build for OpenSSL <3.0	2026-02-16 19:47:13 +00:00
dmiller	f87c5e20b2	SSL data read before error results in successful read.	2026-02-11 22:27:51 +00:00
dmiller	1b3ca45054	Compatibility fixes and Github autobuilds. Closes #3214	2026-01-12 16:59:25 +00:00
dmiller	58874849d3	Reintegrate Nmap 7.98 release branch	2025-08-21 20:24:37 +00:00
dmiller	e048a3e91d	Fix incorrect assertions in Nsock poll engine	2025-08-09 02:52:55 +00:00
dmiller	f2028a2007	Nsock poll engine: notice HUP/ERR/NVAL on Windows. See #3138?	2025-06-25 00:47:44 +00:00
dmiller	40b5bda3ac	Consistent Windows build parameters	2025-04-15 23:58:46 +00:00
dmiller	5b52e7a3f2	Build fixes	2025-04-10 19:24:22 +00:00
dmiller	5f6bc69983	Account for VLAN header in pcap packets if needed.	2025-04-10 19:00:50 +00:00
dmiller	6db5c9cb85	Bump copyright date	2025-02-26 17:44:43 +00:00
dmiller	4f712b76a3	Avoid printing (null):-1 for STDIN in Nsock debug output	2024-12-09 20:03:48 +00:00
dmiller	8bb1334a87	Avoid printing junk timeouts in Nsock debug output	2024-12-06 20:44:28 +00:00
dmiller	92d5b6a310	Correctly process satisfied request from pending SSL data	2024-11-25 22:09:08 +00:00
dmiller	1264a6e472	Increment first, then decrement; avoid clearing desired bit	2024-11-25 22:09:07 +00:00
dmiller	3a0db5dc51	Try an SSL_read() immediately to get the correct ssl_desire	2024-11-21 20:11:38 +00:00
dmiller	38823b5cd1	Handle SSL_read()/SSL_write() returning 0 when not EOF. ... OpenSSL docs say, "Old documentation indicated a difference between 0 and -1, and that -1 was retryable. You should instead call SSL_get_error() to find out if it's retryable."	2024-11-21 20:11:38 +00:00
dmiller	ff1df7a80c	Zero IOD in nsock_iod_new2. Was reusing old peer info	2024-11-18 23:02:48 +00:00
dmiller	4efe456e0b	Nsock iocp: pass immediate errors on to process_event()	2024-11-18 21:19:46 +00:00
dmiller	b1f6f9b088	Track EOF at IOD level also	2024-11-18 20:26:28 +00:00
dmiller	1f6e8d9fe3	Handle EWOULDBLOCK as well as EAGAIN	2024-11-18 20:26:27 +00:00
dmiller	cffc94e845	Consolidate event list management to nevent_unref() ... Removes duplicate logic for PCAP_BSD_SELECT_HACK. May address accounting problems that led to issues like #187 (macOS) and #2912 (Windows).	2024-11-13 19:15:12 +00:00
dmiller	222add87a6	Paranoid gh_list check: node is in list before removal	2024-11-13 19:15:11 +00:00
dmiller	3f64103cf0	Move some prototypes to nsock_internal.h	2024-11-13 19:15:10 +00:00
dmiller	379b56b582	Include timeout info in debug output for all nsock reads/connects	2024-11-12 20:05:56 +00:00
dmiller	1e8c1ab3cd	Let all nsock engines use STDIN on Windows!	2024-11-06 16:48:29 +00:00
dmiller	635cc50011	Nsock iocp: fix IPv6 connect	2024-11-06 16:06:32 +00:00
dmiller	411b186576	Simplify nevent_delete for PCAP_BSD_SELECT_HACK mode	2024-11-01 18:40:09 +00:00
dmiller	2ca95ad6b4	Nsock select engine: don't process if sd is -1 (nonselectable)	2024-11-01 18:40:08 +00:00
dmiller	931c13cd21	Fixes for IOCP engine build	2024-10-15 20:37:11 +00:00
dmiller	7703b045e5	Nsock poll engine: efficient use of pollfds array ... Using socket descriptor as an index works okay for UNIX, though it wastes the first several indices which are assigned to STDIN/STDOUT, files, etc. However, for Windows it is really bad, since descriptors are large, nonconsecutive numbers. Using a little overhead to track each iod's index and the next empty space is worth it.	2024-10-11 20:59:35 +00:00
dmiller	b94ae785ba	Fix handling of non-select()-able pcap handles ... 1. pcap_get_selectable_fd() may return -1 for some devices, even if the platform as a whole supports select() on these handles. Check for this condition throughout. 2. The various backend system calls (kevent, poll, epoll, etc.) do not sleep for the timeout period if no valid handles are registered, unlike select on UNIX. This leads to busy wait, looping continuously. Instead, we call usleep() in those cases.	2024-10-11 20:59:34 +00:00
dmiller	6186767cd1	Move a useful function to nsock_core.c	2024-10-11 20:59:33 +00:00
dmiller	532938f49c	Do a non-blocking check for events when pcap data is available ... For platforms without selectable pcap handles (e.g. Windows), the arrival of data for a pcap read would previously skip checking for any triggered non-pcap events in that loop iteration. This is not usually a problem because the next loop will be triggered immediately, picking up the non-pcap events before any further pcap data arrives. However, excessive pcap data on a handle in immediate mode might prevent the engine loop from checking for non-pcap events for long enough to result in timeouts. Instead, do a non-blocking check for triggered events in this case and handle those in the same loop iteration.	2024-10-09 17:40:10 +00:00
dmiller	167fafe65c	Correctly handle all cases for canceling IO in iocp engine	2024-10-08 21:14:43 +00:00
dmiller	f765af4070	IOCP engine: further checks against duplicating eovs	2024-10-08 21:14:42 +00:00
dmiller	30bc184e0f	Fix iocp engine when calling nsock_read on unconnected socket ... When the IOD isn't connected, iod->peerlen is 0, which means WSARecvFrom returns WSAEFAULT because the lpFrom parameter is not NULL, and 0 bytes is not enough space to fit any sockaddr.	2024-10-08 21:14:42 +00:00
dmiller	d42dd7eed9	Additional tests and checks for gh_heap validity	2024-10-08 21:14:41 +00:00