Commit Briefs

Stefan Sperling

add support for protecting references against 'got send -f' to gotd

ok op@



Omar Polo

gotd: Fix more double process names

Patch by Josiah Frentsos, thanks!


Stefan Sperling

make gotd repo_read store want/have commit IDs in ID sets rather than arrays

Currently only used to detect and avoid storing duplicate IDs sent in want and have lines by the client. If in the future we ever wanted to check which IDs the client has already sent us we could now do O(1) hash table lookups rather than iterating arrays. ok op@


Omar Polo

rename lib/sha1.c to lib/hash.c

It will soon grow functions to deal with sha256 too. stsp@ agrees.


Omar Polo

include sha2.h too where sha1.h is included

In preparation for wide sha256 support; stsp@ agrees. Change done mechanically with find . -iname \*.[cy] -exec sam {} + X ,x/<sha1\.h>/i/\n#include <sha2.h>



Stefan Sperling

do not expect to see a DISCONNECT message in repo processes

The parent no longer sends this message. Perform related cleanup in the shutdown path instead. ok op@



Stefan Sperling

add a gotd session process, split off from the parent process

The new session process is able to manipulate files in the repository and keeps track of the read/write client session state. The parent process now restricts its view of the filesystem to the absolute path stored in argv[0], and combines this with unveil "x" on this path. As a result the parent process can only re-exec itself. small tweaks + ok op@


Stefan Sperling

convert gotd repo_read.c and repo_write.c to single-client

Because these processes are now started on demand per client connection there is no need to keep track of multiple clients anymore. Also, these processes can now exit when a disconnect event is received. ok op, jamsek


Stefan Sperling

fork gotd repo_read/repo_write children on demand

ok op, jamsek


Stefan Sperling

switch gotd from chroot(2) to unveil(2)

In the future, gotd will fork+exec new processes for each client connection. Using unveil instead of chroot avoids having to start such processes as root. The -portable version could use chroot(2) where no equivalent to unveil(2) exists. A future component which starts new processes will be isolated as a separate process, which could run as root in the -portable version. ok op@