Age	Commit message (Collapse)	Author	Lines
24 hours	Share one lock-file helper between the caches	Bryce Kwon	-29/+45
25 hours	Keep the cache's trailing slash helper in cache.c	Bryce Kwon	-3/+12
25 hours	Let go of a slot lock that cannot take its key	Bryce Kwon	-0/+3
	A failed truncate or key write returned with the descriptor still open, so the lock was held while the page rendered uncached and the file was left behind. It is removed before its last close, as every holder does.
25 hours	Keep lock contention out of the error log	Bryce Kwon	-11/+25
	process_slot logged every failure to lock, so each request that found its slot being filled left a line, while refresh_slot logged none, so an expired slot in a directory the server cannot write was served stale for good without a word. lock_slot now logs the step that failed, as the scan does, and leaves out only another request at work.
25 hours	Open the lock files close-on-exec	Bryce Kwon	-1/+3
	Each exec filter and snapshot compressor a fill runs inherited the slot's lock descriptor beside the stdout it writes into. An flock lasts until its last copy closes and nothing waits for a process a filter leaves behind, so a fill killed before its rename could leave the slot locked, and served stale or uncached, for as long as such a process lived. The scan holds its lock across no exec and opens the same way so that the two stay alike.
26 hours	Give stdout back from a slot in one place	Bryce Kwon	-10/+11
26 hours	Return nothing from two calls nobody checks	Bryce Kwon	-3/+2
26 hours	Format the cache listing's times where it prints	Bryce Kwon	-17/+6
26 hours	Close the cache's descriptors through one helper	Bryce Kwon	-33/+15
26 hours	Drop the cache's checks for arguments never NULL	Bryce Kwon	-14/+0
	The cache root always holds a default or a copy, the key comes from a strbuf, and every string hashed is the same or a macro expansion.
26 hours	Write the cache's errors with fprintf like the rest	Bryce Kwon	-25/+10
2026-10-17	Lock a cache slot with flock	Bryce Kwon	-7/+6
	The slot lock was a POSIX record lock, which a process loses the moment any descriptor on the file closes, and the fill puts the lock file on stdout, which every exec filter and snapshot compressor closes with its dup2. A second request on the slot could then rebuild it under the first and splice the two pages. An flock holds until the last descriptor goes.
2026-10-17	Abandon a fill whose last flush fails to write	Bryce Kwon	-2/+4
	The slot being filled was forgotten before the page's final flush, so a write that failed there, on a full disk for one, could not abandon the fill: cgit died into the lock file instead and the visitor got nothing. The mid-page path already delivered the page from the lock file.
2026-10-17	Widen the cache ttl before multiplying it	Bryce Kwon	-1/+3
	A ttl in the tens of millions of minutes overflowed the int it was multiplied in.
2026-10-16	Read and write slot keys and pages in full	Bryce Kwon	-2/+2
	xread and xwrite may move fewer bytes than asked. A short write of the key left a slot that could never match, a short read of it reported a collision, and a page write that moved nothing looped for ever.
2026-10-16	Stop the sendfile loop at the end of a short slot	Bryce Kwon	-0/+4
	sendfile returns zero at the end of the file, and the loop only ended once the offset reached the size recorded when the slot was opened, so a slot cut shorter in between spun the request forever. Only Linux builds this path. A stand-in header type checked it here.
2026-10-01	End a request quietly when the client has gone	Bryce Kwon	-1/+2
	A write that failed with EPIPE on an uncached page died into an error page, which died again at exit when the page could not be sent either, and git logged a recursion warning on every dropped connection. A filter that exits early fails a write the same way, so the quiet exit is kept for writes that reach the client itself.
2026-10-01	Keep clone files and oversized responses out of the cache	Bryce Kwon	-3/+22
	The dumb transport reads files that already sit on the disk, so a pack copied into a slot cost that disk twice and the request a second write of every byte. A snapshot took a slot whatever its size, so a visitor naming distinct refs and ids could fill the cache root with archives. `cache-max-slot-size`, 64 MB unless set, now serves a larger response from the lock file and drops it, along with any expired copy it would have replaced.
2026-09-22	Align with spaces instead of tabs	Bryce Kwon	-7/+7
2026-09-21	Harden the request path, scan and error recovery	Bryce Kwon	-22/+49
2026-09-21	Log a cache slot that cannot be published	Bryce Kwon	-2/+14
2026-09-21	Fail cleanly when a filter cannot run	Bryce Kwon	-0/+3
	A filter program that could not be run answered with two responses, and a filter that exited without reading its input ended cgit with the page half written.
2026-09-06	Clean up the whole tree	Bryce Kwon	-34/+29
2026-09-06	Trim the comments and dead code across the tree	Bryce Kwon	-24/+14
2026-08-23	Log scan lock failures that are not contention	Bryce Kwon	-7/+7
2026-08-23	Give up a scan or slot whose lock file was renamed	Bryce Kwon	-0/+13
2026-08-23	Serve nothing further once a fill is abandoned	Bryce Kwon	-3/+34
2026-08-22	Log when a cache slot is claimed by a colliding key	Bryce Kwon	-0/+8
2026-08-12	Restyle the sources and fix the audit's findings	Bryce Kwon	-218/+246
2026-08-08	Bound the search and cache key a request can ask	Bryce Kwon	-0/+15
2026-08-08	Gather page output into one buffer	Bryce Kwon	-0/+1
2026-08-06	Settle the names, types and layout of the sources	Bryce Kwon	-10/+9
2026-07-25	Bound the cache listing key output	Bryce Kwon	-2/+9
2026-07-16	Point the fork's links and credits at itself	Bryce Kwon	-1/+1
	AUTHORS now records the upstream developers as the project's history rather than as maintainers of the fork.
2026-07-15	Reorganize into source, assets and libraries	Bryce Kwon	-0/+475
	The C sources move to source/, the served files to assets/, the bundled Git submodule to libraries/git and the filter scripts to extensions/. Everything the build generates now lands in build/, so a clean is a single remove and the tree stays clean.