Hello Enrico, On 28.08.22 16:02, Enrico Scholz wrote: > The "tftp: allocate buffers and fifo dynamically" patch in the last > patchset broke interaction with non rfc 2347 servers (e.g. when data > transfer starts immediately after RRQ/WRQ without the OACK negotiation > phase). > > This has been fixed for both RRQ and WRQ requests. > > For WRQ requests (push), the "tsize" option will not be sent anymore > because it is always '0' and can confuse servers. > > New patches add some sanity checks which prevent modification of > internal information (blocksize or port numbers) when OACK packets > arrive in the middle of a transfer. The patch introducing the regression is still in next, which can be rebased unlike master. Perhaps it would be best you provide a fixup for the commit in question (or a revised series, if it would cause conflicts for later commits)? That way, we don't end up with breakage when doing a bisect in future. The remaining sanity checking in this series could then be stacked on top. Cheers, Ahmad > > > Enrico Scholz (8): > tftp: make debug_assert() critical when selftest is enabled. > tftp: remove sanity check of first block > tftp: split out allocation and cache initialization > tftp: accept OACK + DATA datagrams only in certain states > tftp: support non rfc 2347 servers > tftp: do not set 'tsize' in wrq requests > tftp: fix WRQ support > tftp: add some documentation about windowsize support > > Documentation/filesystems/tftp.rst | 38 ++++++ > fs/tftp.c | 189 +++++++++++++++++++---------- > 2 files changed, 166 insertions(+), 61 deletions(-) > -- Pengutronix e.K. | | Steuerwalder Str. 21 | http://www.pengutronix.de/ | 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
