M-Labs
/
artiq
mirror of https://github.com/m-labs/artiq
2
0
Fork 1
Code Issues Projects Releases Wiki Activity
8,234 Commits
18 Branches
34 Tags
36 MiB
Commit Graph

83 Commits (1852491102752c83f6689a316e61b8000454d4fa)

Author SHA1 Message Date
Egor Savkin 1852491102
add channel names to RTIO errors 2022-12-02 16:27:03 +08:00
mwojcik 4bdb4c8e11 config: error instead of empty value if key not found 2022-05-19 16:48:59 +08:00
Michael Birtwell 671453938b Require explicitly closing TcpStreams 
Instead of automatically closing and draining the TcpStream in the Drop
implementation instead expect the user to call TcpStream::close.
Add close called to all users of TcpStream.
Document the requirement to call close on TcpListener::accept, this seems
to be the only way to get a new TcpStream at the moment.
 2022-04-24 10:10:43 +08:00
pca006132 ebfeb1869f firmware: use &CSlice for lists 2022-03-10 16:30:22 +08:00
pca006132 4132c450a5 firmware: runtime changes for exception 
Ported from:
M-Labs/artiq-zynq#162

This includes new API for exception handling, some refactoring to avoid
code duplication for exception structures, and modified protocols to
send nested exceptions and avoid string allocation.
 2022-01-26 07:16:54 +08:00
Steve Fan 4a6bea479a
Host report for async error upon kernel termination (#1791) 
Closes #1644
 2021-12-04 13:33:24 +08:00
occheung de0f2d4a28 firmware: adopt endianness protocol in artiq-zynq 
Related:
artiq-zynq: M-Labs/artiq-zynq#126
artiq: #1588
 2021-09-10 13:25:12 +08:00
occheung c800b6c8d3 runtime: update rust alloc, managed 2021-09-10 13:25:09 +08:00
pca006132 8148fdb8a7
use device endian for core device protocols (#1591) 2021-01-22 16:33:21 +08:00
Sebastien Bourdeauducq 90017da484 firmware: remove obsolete watchdog code (#1458) 2020-10-15 18:38:00 +08:00
Sebastien Bourdeauducq dfad27125e runtime: relax/fix TCP keepalive settings (#1125) 2019-12-23 19:58:10 +08:00
Sebastien Bourdeauducq 87ce24e867 runtime: refactor startup and RTIO clocking initialization 2019-06-14 15:26:30 +08:00
David Nadlinger cd7a5a3683 firmware: Fix kernel RPC handling of zero-size values (e.g. empty arrays) 2019-03-31 18:33:44 +01:00
David Nadlinger b4ddf4c86b firmware: Make "unexpected reply from kernel CPU" log messages unique 
This makes it easier to localize issues based on the log output.
 2019-03-31 18:31:56 +01:00
whitequark dd829afebd firmware: fix another TOCTTOU race in sync/async RPC code. 2018-11-12 15:42:07 +00:00
whitequark 583bba8777 Revert "firmware: workaround for RPC failures" 
This reverts commit 59033d2588.
 2018-11-12 15:36:36 +00:00
whitequark 0edae64afb firmware: fix TOCTTOU race in sync/async RPC code. 
Before this commit, the main loop in session code was laid like:

  1. process_kern_queued_rpc
  2. process_host_message
  3. process_kern_message

If a host message (such as an RPC reply) caused the kernel to exit,
then any async RPCs would not complete, since RunFinished immediately
shuts down the kernel.

Fix this by reordering 1 and 2.
 2018-11-12 15:30:59 +00:00
Sebastien Bourdeauducq 59033d2588 firmware: workaround for RPC failures 2018-11-12 19:51:54 +08:00
Sebastien Bourdeauducq 8caea0e6d3 gateware,runtime: optimize RTIO kernel interface further 
* now pinning (TODO: atomicity)
* for inputs, merge request and timeout registers
 2018-11-08 18:29:24 +08:00
Sebastien Bourdeauducq 62642957cd runtime: fix DRTIO aux channel race condition 2018-09-19 11:16:21 +08:00
Sebastien Bourdeauducq c8cd830118 drtio: implement get_rtio_destination_status for kernels 2018-09-15 19:11:22 +08:00
Sebastien Bourdeauducq ae72e3a51e firmware: add support for moninj and kern_hwreq over DRTIO switching 2018-09-14 20:26:39 +08:00
Sebastien Bourdeauducq 2fff96802b runtime: remove support for building without RTIO 2018-09-10 23:09:02 +08:00
Sebastien Bourdeauducq b10d3ee4b4 make RTIO clock switch optional and simplify 
Kasli no longer has an internal RTIO clock.
Switching clocks dynamically is no longer supported.
 2018-05-18 17:41:34 +08:00
whitequark ca93b94aea firmware: move config requests to management protocol. 
They were only in session protocol because of historical reasons.
 2018-05-16 14:32:49 +00:00
Florent Kermarrec 7fe49a78d2 firmware/runtime/session: fix compilation when no rtio core 2018-05-15 23:47:47 +02:00
whitequark 9347f6e00c firmware: use a dedicated error type in the scheduler. 
After this commit, error handling does not normally allocate
(session::Error::{Load,Unexpected} still allocate, but those two
are very rare).

Good riddance to libstd_artiq.
 2018-05-15 14:19:44 +00:00
whitequark 479cb9a857 firmware: use dedicated error types for every protocol and thread. 
Good riddance to io::Error::Unrecognized.
 2018-05-15 13:27:23 +00:00
whitequark 26faaad2b7 firmware: remove io::Result<T, E> in favor of Result<T, impl Fail>. 
This doesn't yet use impl Fail since std_artiq::io::Error doesn't
impl it.
 2018-05-15 08:29:03 +00:00
whitequark 140ca729aa firmware: rename io::Error::{UnexpectedEof→UnexpectedEnd}. 
We don't even have files.
 2018-05-15 08:27:45 +00:00
whitequark 5ead27f2e1 firmware: eliminate uses of std_artiq::io::{Read,Write} from runtime. 2018-05-14 19:11:57 +00:00
whitequark cf33ba748d firmware: eliminate non-io uses of std_artiq from runtime. 
Just replace these with references to alloc.
 2018-05-14 19:11:57 +00:00
whitequark 02b8426f60 firmware: rename libboard to libboard_misoc. 
By analogy with libbuild_misoc. Hopefully one day it actually gets
moved to misoc...
 2018-05-14 18:09:44 +00:00
whitequark 6a10d54432 firmware: migrate rpc_proto to new libio. 
This closes an old and horrible issue in that some code in ksupport
may implicitly try to allocate and we'll never know until it crashes
at runtime inside liballoc_stub.

This also removes liballoc_stub.
This also removes ReadExt and WriteExt from libproto.
Hooray!
 2018-05-14 17:26:59 +00:00
whitequark 09617c9f5e firmware: migrate session_proto to new libio. 2018-05-14 15:17:32 +00:00
whitequark 873324d52b firmware: don't truncate queued RPCs (fixes #985). 2018-04-21 19:39:46 +00:00
whitequark ca419aa3c2 firmware: split out libboard_artiq from libboard. 2017-12-28 07:07:00 +00:00
whitequark d7cb4963e1 firmware: prepare config block for access from BIOS/bootloader. 
* remove liballoc dependency from mod config,
  * move mod config to libboard,
  * move config sector immediately after BIOS sector.
 2017-12-27 21:28:40 +00:00
whitequark 5a2cbe7088 runtime: remove borrow_mut!() in favor of backtraces. 2017-12-27 19:15:25 +00:00
whitequark ad8fcb8b86 runtime: has_rtio -> has_rtio_core. 
has_rtio is the cfg for the kernel CPU, has_rtio_core is the one
for the comms CPU.

Also remove a few useless #[cfg]s.
 2017-11-03 15:59:59 +00:00
whitequark 2404a0d8c8 runtime: allow #[cfg(not(has_rtio))] builds. 2017-11-03 15:49:30 +00:00
whitequark 73c76ebb9a firmware: eliminate most transmutes and document the rest. NFC. 2017-10-02 03:09:46 +00:00
whitequark c00b3fe8cd firmware: update smoltcp. 2017-09-24 22:57:27 +00:00
Sebastien Bourdeauducq 8399f8893d add kernel access to non-realtime SPI buses (#740) 2017-06-18 12:45:07 +08:00
whitequark b913d1d6f2 runtime: make a copy of startup/idle kernel firmware before loading. 
Fixes #716.
 2017-04-21 16:46:40 +00:00
whitequark 0e68eaa879 runtime: print a heap dump on out-of-memory condition. 2017-04-21 14:48:10 +00:00
Sebastien Bourdeauducq fe05aede78 firmware: DmaPlayback → DMARetrieve 2017-04-19 11:11:42 +08:00
whitequark c6e8d5c901 runtime: allow setting UART log level explicitly. 
This is way more convenient than commenting out parts
of session.rs when debugging.
 2017-04-15 08:27:18 +00:00
whitequark 0531dc45c3 DMA: erase trace before re-recording it. 
Or we could needlessly OOM replacing a large trace.
 2017-04-15 07:48:02 +00:00
whitequark 9dfe9c1248 DMA: improve recording performance. 
This commit moves DMA serialization code to the kernel CPU
(to cope with the existence of rtio_output_wide) and batches
the resulting sequences. This results in less data being transferred
between kernel and comms CPUs (24 octets with one pointer before,
18 octets with no pointers now, for the common case of rtio_output),
but most importantly reduces cache flushes, which now happen
once per 64k octets.

On average, it now takes about 15us to record a single RTIO event
in a DMA trace.

Fixes #712.
 2017-04-15 07:29:52 +00:00