c834e4f503
enable network and mgmt during Rust panic, make RTIO PLL lock failure a panic
...
Closes #198 #200
Making it a soft panic makes it more involved with a bit of code duplication - setting up mgmt requires setting up the interface and sockets. Maybe can be done a bit cleaner.
```
[spaqin@hera:~/m-labs/artiq-zynq]$ artiq_sinara_tester
****** Sinara system tester ******
[...]
ConnectionRefusedError: [Errno 111] Connection refused
[spaqin@hera:~/m-labs/artiq-zynq]$ artiq_coremgmt -D 192.168.1.56 log
[ 0.000067s] INFO(runtime): NAR3/Zynq7000 starting...
[ 0.005238s] INFO(runtime): detected gateware: GenericMaster
[ 0.016152s] INFO(libboard_zynq::i2c): PCA9548 detected
[ 0.023004s] WARN(runtime): config initialization failed: SD error: Card initialization error: No card inserted, check if the card is inserted properly.
[ 0.036730s] WARN(runtime::rtio_clocking): error reading configuration. Falling back to default.
[ 0.213000s] ERROR(runtime::rtio_clocking): RTIO PLL failed to lock
[ 0.224443s] INFO(libboard_zynq::i2c): PCA9548 detected
[ 0.256197s] INFO(runtime::comms): network addresses: MAC=e8-eb-1b-13-49-8b IPv4=192.168.1.56 IPv6-LL=fe80::eaeb:1bff:fe13:498b IPv6: no configured address
[ 0.270183s] ERROR(runtime::comms): There has been an error configuring the device: RTIO PLL failed to lock. Only mgmt interface will be available.
[ 4.000095s] INFO(libboard_zynq::eth): eth: got Link { speed: S1000, duplex: Full }
[ 33.148521s] INFO(runtime::mgmt): received connection
```
Reviewed-on: M-Labs/artiq-zynq#199
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2022-10-21 17:56:34 +08:00
dc862a9051
match ident message with mainline
2022-10-21 12:08:11 +08:00
19e60073de
kasli_soc: ident = variant name
2022-10-21 11:55:24 +08:00
a546d0f95b
Implement reboot for artiq_coremgmt
2022-10-07 18:31:11 +08:00
d6ae646790
update dependencies
2022-10-07 18:30:39 +08:00
f3310324d7
update dependencies
2022-08-26 17:37:27 +08:00
0812f22423
update dependencies
2022-07-20 17:34:26 +08:00
b638fce069
update SEEN_ASYNC_ERRORS in destination_survey ( #195 )
...
Co-authored-by: kk105 <kkl@m-kabs.hk>
Reviewed-on: M-Labs/artiq-zynq#195
Co-authored-by: kk105 <kkl@m-labs.hk>
Co-committed-by: kk105 <kkl@m-labs.hk>
2022-06-20 17:41:08 +08:00
9ec6a1feab
dyld/rebind: support rela generation with nac3ld
2022-06-01 21:27:38 +08:00
8e144e41de
reloc: impl ARM_PREL31 handling
2022-06-01 21:27:38 +08:00
512b6bac12
reloc: add PC-relative relocation support
2022-06-01 21:27:38 +08:00
e3ed41ff32
fix index table reference type
2022-06-01 18:35:50 +08:00
97a63ca8d0
dyld: add EXIDX entry type
...
The type is just for aesthetic. The interpretation of an index table entry is not our concern.
2022-06-01 18:33:19 +08:00
f0febe0ee4
change catch type to single reference
2022-05-31 18:26:30 +08:00
7a8f96dbd9
rtio_mgt: use mutex's async_lock
2022-05-25 10:39:06 +08:00
596edb480c
cargo: update zynq-rs
2022-05-25 10:37:38 +08:00
4f457d9c24
moninj: log link down at debug level
2022-05-25 10:37:38 +08:00
24df52268e
moninj: restructure timeout
...
stop logging errors if satellite is unavailable
drtio: don't even send message if link is down
2022-05-25 10:37:38 +08:00
48c9b43171
moninj: make it use async drtioaux
2022-05-25 10:37:38 +08:00
57d7f01b04
drtio: port 64-bit padding from mainline
2022-05-24 15:43:01 +08:00
efc432352e
zc706: no syncrtio for master, fixes hangs ( #188 )
2022-05-03 14:36:10 +08:00
def4d989cd
kasli_soc: fix si5324 pins routed to GTX
2022-04-25 12:33:21 +08:00
1d731a3589
zc706 master: route sma clock to si5324
2022-04-13 16:35:52 +08:00
3cf86a6335
satellites: add rtio_crg cfg
2022-04-12 13:44:53 +08:00
78bc162749
rtio_clocking: remove loop
2022-04-12 13:33:52 +08:00
14f7778732
update libconfig features
2022-04-08 10:30:21 +08:00
dcfb28ce61
fix drtioaux packet corruption
...
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2022-04-01 14:15:14 +08:00
433a9cdaf1
runtime: fix warnings on nondrtio systems
2022-03-29 10:05:11 +08:00
a79bef2243
runtime: provide/fix more libc mem functions
2022-03-28 13:24:01 +08:00
c6ef9b117c
fix previous commit
2022-03-26 20:08:11 +08:00
dcfaf587ec
firmware: add UnwrapNoneError exception
2022-03-26 15:29:40 +08:00
a92561b9d3
implement rtio_get_destination_status ( #177 )
...
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2022-03-25 18:20:05 +08:00
161044e78f
drop support for big-endian moninj
2022-03-19 23:01:36 +08:00
426500d2f9
firmware: support 64-bit moninj probes
2022-03-17 20:26:44 +08:00
ebdb08180d
drtio: demote default routing table message to info
2022-03-16 21:04:12 +08:00
0530e596ba
mgmt: remove spurious config write warning
2022-03-16 08:24:52 +08:00
7502f3a765
update dependencies
2022-03-10 17:25:40 +08:00
ae0d724bf8
runtime: use &CSlice for lists
2022-03-10 16:30:34 +08:00
6c834899e9
si5324: fix clock source
2022-03-09 13:55:36 +08:00
a22b13cc46
kasli_soc: forward SMA clkin
2022-03-09 12:43:47 +08:00
85e5c08d7f
kasli_soc: use si5324 in master
2022-03-04 13:17:53 +08:00
3c17362fad
satman: fix i2cswitch
2022-03-03 17:18:22 +08:00
4f2a0986da
rtio_clocking: fix wrong descriptions
2022-03-03 10:24:13 +08:00
4a2218641f
fix BorrowMutError in moninj
2022-03-02 15:45:17 +08:00
9a06cd9d27
expose pca954x_select api ( #167 )
...
PR accompanying to ARTIQ's PCA954X support (#1860 ).
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2022-03-02 10:52:27 +08:00
b56b50b147
add comment about EXCEPTION_ID_LOOKUP sync
2022-03-01 09:50:28 +08:00
f38117774f
runtime/eh_artiq: updated exception IDs
...
Fixes #166
2022-02-28 21:15:07 +08:00
880ba6b206
runtime: add nac3 exception symbols
2022-02-23 11:05:08 +08:00
accac99f48
updated zynq-rs with pca9547 support ( #165 )
...
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2022-02-11 13:53:58 +08:00
6f5ba46e89
runtime/eh_artiq: support exception allocation
...
The backtrace is now nested, and should be used together with the stack
pointer array to construct the full backtrace for each exception.
We now allocate exception objects in a stack, but their names are still
not allocated. This is fine for exceptions raised in the driver or artiq
code, but we will have to implement allocation for names of exceptions
raised in RPC calls. The compiler should also emit code to store the
exception names once they catch it, to prepare for later reraising.
2022-01-23 21:31:22 +08:00
8923feceac
runtime/eh_artiq: use forced unwind
...
This patches ports the LLVM libunwind newly added forced unwinding
function. This enables us to run forced unwinding to obtain correct
backtrace when uncaught exceptions occur.
This patch also changes the exception handling scheme from the standard
two-phase unwinding to single phase using forced unwinding. This brings
some performance improvement and prepared for later nested exception
support. For nested exceptions, we will have to record the backtrace
regardless if the exception is an uncaught exception, as there can be
another exception being thrown while executing the finally block for
caught exceptions, and we will lose the backtrace if we don't store it
earlier before running the cleanup pads.
2022-01-14 13:35:24 +08:00
97ca72f7f1
libunwind: enable lto
2022-01-06 14:04:04 +08:00
acaf388dbb
eh_artiq: handle catch clauses appropriately
2022-01-06 13:41:47 +08:00
8788d6458e
runtime/rpc: fixes alignment and size problem
2022-01-04 18:25:53 +08:00
efe315c21d
libdyld: accepts R_ARM_ABS32
...
Somehow this relocation type is emitted by nac3.
According to table 4-9 of ARM ELF ABI and discussion in ld bugzilla
(https://sourceware.org/bugzilla/show_bug.cgi?id=16163 ), this behaves
the same as R_ARM_GLOB_DAT and R_ARM_JUMP_SLOT.
2021-12-30 00:05:47 +08:00
84becfe2c0
report async errors upon kernel termination
...
Port of 4a6bea479aM-Labs/artiq-zynq#156
Co-authored-by: stevefan1999 <sf@m-labs.hk>
Co-committed-by: stevefan1999 <sf@m-labs.hk>
2021-12-06 17:38:55 +08:00
31fb2b388a
Support for DRTIO 100MHz ( #155 )
...
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2021-12-03 17:19:42 +08:00
e045837b67
zc706: not actually ultrascale
2021-11-29 12:48:45 +08:00
ada3f2e704
drtio: reading still needs work buffer after all
2021-11-29 12:46:08 +08:00
8be5048cd3
upgrade to new clock configuration system ( #152 )
...
As mentioned in https://github.com/m-labs/artiq/issues/1735 - this is the Zynq version.
Reviewed-on: M-Labs/artiq-zynq#152
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2021-11-29 11:17:59 +08:00
e8db2a4b49
drtio: crc from mainline, removed byte swap
2021-11-24 12:12:40 +08:00
0b27349ec4
dummy_spi -> pmod_spi
2021-10-14 16:37:13 +08:00
21eb1cab1a
zc706: added dummy spi in place of sdio
2021-10-14 15:43:51 +08:00
3096daaaee
zc706: removed nist_clock sdcard, put pmod instead
2021-10-14 15:01:38 +08:00
4fbfccf575
zc706: fix nist_qc2 extension, ams101 iostandard
2021-10-14 12:39:09 +08:00
5c40115945
make ZC706 RTIO channels consistent with KC705
...
Reviewed-on: M-Labs/artiq-zynq#147
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2021-10-13 17:20:25 +08:00
a5e3580d18
Revert "runtime: expose rint from libm"
...
This reverts commit 3582af564d
2021-10-11 08:13:26 +08:00
3582af564d
runtime: expose rint from libm
2021-10-10 20:40:29 +08:00
219c075931
added explicit runtime/satman targets for makefile
...
Reviewed-on: M-Labs/artiq-zynq#144
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2021-10-08 21:06:23 +08:00
d04a7decfe
removed simple variants from zc706
2021-10-08 11:07:12 +02:00
0efa83e956
update build scripts for DRTIO
...
Reviewed-on: M-Labs/artiq-zynq#135
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2021-10-08 16:25:13 +08:00
4fa824f42b
kasli-soc: remove irrelevant comment
2021-10-08 16:13:17 +08:00
ab0c205dd2
gateware: add DRTIO
...
Reviewed-on: M-Labs/artiq-zynq#140
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2021-10-08 16:12:30 +08:00
8d2bb09149
add satman firmware ( #136 )
...
Reviewed-on: M-Labs/artiq-zynq#136
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2021-10-08 16:04:50 +08:00
aaec0abdf6
fix build/warnings before drtio is fully merged
2021-10-06 16:17:19 +08:00
e241957419
add libbuild_zynq
...
Reviewed-on: M-Labs/artiq-zynq#141
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2021-10-06 16:16:49 +08:00
50262b3f0c
runtime: link_thread -> link_task
2021-10-06 07:59:55 +02:00
827c6c1306
runtime: switch to libio/libboard_artiq, add DRTIO mastering support
...
Reviewed-on: M-Labs/artiq-zynq#137
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2021-10-06 13:05:45 +08:00
e6863263b4
add libboard_artiq (to be shared between runtime and satman)
...
Reviewed-on: M-Labs/artiq-zynq#139
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2021-10-06 13:02:28 +08:00
d7f45d473e
add libio (to be shared between runtime and satman)
...
Reviewed-on: M-Labs/artiq-zynq#138
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2021-10-06 13:01:52 +08:00
35250b3f56
libdyld: fixed symbol relocation
...
Note that in libdyld/src/lib.rs #117-118, image pointer is already added
to the symbol offset, so we do not need to add the pointer again
2021-09-25 11:30:45 +08:00
2ed2ffe417
update dependencies
2021-08-09 15:16:54 +08:00
18e05c91e1
zc706: si5324 is not needed for standalone target
2021-08-04 09:14:19 +08:00
e3d3cb2311
si5324: bring on par with mainline ARTIQ ( #132 )
...
si5324 driver in runtime should be now equal in function to the one in artiq.
kasli-soc has no way of doing a hard reset on the peripheral, but zc706 does.
Reviewed-on: M-Labs/artiq-zynq#132
Co-authored-by: mwojcik <mw@m-labs.hk>
Co-committed-by: mwojcik <mw@m-labs.hk>
2021-08-04 09:12:38 +08:00
f543501012
si5324: remove debug print
2021-08-02 14:14:59 +08:00
111ac0c716
runtime: clock Si5324 from its crystal
2021-07-30 17:07:58 +08:00
8128dc0b56
Revert "kasli-soc: work around I2C breakage ( #130 )"
...
This reverts commit f1fd55dee5
2021-07-30 16:55:06 +08:00
dcb6129b0e
update dependencies
2021-07-05 13:56:40 +08:00
f25e261bdd
update dependencies
2021-06-25 17:12:47 +08:00
2ba4d8935d
fix compilation with nixpkgs 21.05
...
The environment variable is optional to keep compatibility with other build environments.
Closes #131
2021-06-25 15:57:39 +08:00
8c8a5d53b9
update dependencies
2021-06-19 22:51:25 +08:00
852123b42a
kasli-soc: add RTIO LEDs
2021-05-30 20:40:53 +08:00
f1fd55dee5
kasli-soc: work around I2C breakage ( #130 )
2021-05-29 17:13:41 +08:00
21d98711c1
use new smoltcp error code
2021-05-29 17:13:22 +08:00
0ae2138034
kasli-soc: preliminary si5324 support
2021-05-29 16:15:27 +08:00
1b474d2dd4
update dependencies
2021-05-29 14:20:23 +08:00
506c741238
support absence of gateware RTIO clock selection mux
2021-02-15 21:41:30 +08:00
8815f76114
kasli_soc: fix has_grabber
2021-02-15 21:41:02 +08:00
ef18fa4c6d
kasli_soc: add RTIO log channel
2021-02-15 19:56:59 +08:00
faf9714e10
add demo build for Kasli-SoC
2021-02-15 19:52:13 +08:00
8d4e42be32
remove redpitaya and coraz7 support
2021-02-15 19:30:13 +08:00
fcb38fae6c
runtime: disable TCP delayed ack
2021-02-08 03:24:18 +01:00
bfd8343876
update zynq-rs and dependencies (smoltcp 0.7.0)
2021-02-08 03:24:18 +01:00
4039431533
kasli_soc: fix eem iostandards
2021-02-07 22:34:29 +08:00
3f9bd06468
add Kasli-SoC generic gateware builder (WIP)
2021-02-07 14:44:32 +08:00
bb65074254
updated zynq-rs and IRQ handling
2021-01-28 12:56:54 +08:00
c2a6fb72f7
updated zynq-rs dependency
2021-01-26 12:38:09 +08:00
faa335461d
runtime: modified protocols to use device endian
2021-01-22 13:36:38 +08:00
be01fbd943
runtime/irq: use assembly for naked IRQ handler
...
and solves the bug introduced due to stack unwinding...
2021-01-18 17:08:04 +08:00
9e8b554c6d
runtime/kernel/core1: use correct ABI
2021-01-18 16:43:30 +08:00
b4ff6dda24
runtime: use naked function for IRQ
...
non-naked IRQ would somehow trigger interrupts after several kernel
restarts, investigating
2021-01-18 10:38:50 +08:00
35204d4716
fixed new compiler warnings
2021-01-15 17:55:58 +08:00
93493397ae
updated zynq-rs dependency
2021-01-15 17:55:47 +08:00
e5207b86db
update rust dependencies
2020-12-24 01:17:24 +01:00
28fe61b061
runtime: add feature target_kasli_soc
2020-12-23 20:11:43 +01:00
ce55e2ed23
update rust dependencies
2020-12-23 17:02:19 +01:00
cb9dae1951
update rust dependencies
2020-11-20 17:54:09 +01:00
07b425a67a
fix other compilation warnings
2020-11-16 14:57:20 +08:00
57ae8619f8
remove unnecessary no_mangle
...
no_mangle does nothing on extern items as per https://github.com/rust-lang/rust/issues/78989#issuecomment-726163973
Closes #115
2020-11-16 14:51:30 +08:00
32048ead20
gateware/coraz7: remove unused VARIANTS
2020-11-14 02:24:29 +01:00
113c8eb0b8
add coraz7 + redpitaya targets
2020-11-13 20:17:18 +01:00
9259cffeb2
i2c: add stubs for targets without i2c
2020-11-12 15:26:06 +01:00
David Nadlinger
7c336f7770
kernel/api: Add additional binary libm functions
...
Also factored out (f64, f64) -> f64 libm wrappers into
a macro, similar to the unary ones.
2020-11-11 01:24:44 +01:00
291a782db0
fix compiler warning
...
Feature is now stable.
2020-11-06 12:22:34 +08:00
479e6afd12
update Rust dependencies
2020-11-06 12:20:48 +08:00
7dbffadf08
mgmt: implemented config write
2020-11-04 21:16:47 +08:00
b7155c9ded
Makefile: cleanup
2020-10-14 13:06:15 +08:00
5c62d6a141
update dependencies, disable custom compiler_builtins ( #113 )
2020-10-13 21:51:40 +08:00
eab839aed0
follow changes in zynq-rs
2020-10-13 19:12:55 +08:00
a374d8a02f
runtime/kernel/dma: reduced replay overhead
...
We can just flush the cache once when we get the handle, instead of
everytime before replay.
2020-09-09 21:25:03 +08:00
03d9827a5a
acpki: working
2020-09-09 21:24:49 +08:00
86b9045417
use liconfig, libcoreio, szl from zynq-rs
2020-09-09 18:44:12 +08:00
7e26a87aed
fix previous commit
2020-09-09 18:12:39 +08:00
a277e89b3a
Makefile: fix runtime.bin target
2020-09-09 17:01:14 +08:00
1e742cc390
updated zynq-rs dependency
2020-09-07 16:18:50 +08:00
2fe73505c8
improve i2c error reporting
2020-09-06 00:38:28 +08:00
36d8ffec3b
expose i2c to kernels
2020-09-06 00:11:19 +08:00
91ed035bef
makefile: fix szl rebuild ( #108 )
2020-09-06 00:10:44 +08:00
d5a91a7697
updated zynq-rs for more CPU options
2020-09-04 16:43:07 +08:00
5da76f2abb
enabled cpu program flow prediction
2020-09-04 13:25:17 +08:00
5e4bf8bbf7
runtime/comms: Faster RPC alloc
...
We do busy polling for some time before doing await, for small
allocations we could avoid the context switching and reduce the latency.
2020-09-03 16:58:44 +08:00
cdc8ad8aee
runtime/kernel/control: fixed memory leak
2020-09-03 16:51:51 +08:00
805f1d4eff
runtime: increased heap size
2020-09-02 10:15:52 +08:00
ae07c05db4
runtime: optimize for speed and fix deadlock
...
The previous method of taking the channel could cause deadlock, we now
use semaphore to signal if the channel is available instead of busy
polling the mutex.
2020-09-02 10:15:52 +08:00
b0706f470d
runtime: set default log level to Info
2020-09-01 17:11:21 +08:00
ccf8ae5b5d
szl: implemented #96
...
SZL no longer do self-extraction for runtime binary, it would boot from
SD/ethernet depending on the boot mode settings.
This allows a larger runtime binary, so we can optimize for speed in the
runtime firmware for better performance, and allow more features to be
added later.
2020-09-01 15:57:20 +08:00
050b2457a4
runtime/main: removed bitstream loading code
2020-09-01 15:43:54 +08:00
eb78e4e2da
libconfig: refactored load_pl into bootgen
...
Now allows loading firmware.
2020-09-01 14:48:19 +08:00
afecc83ecf
libconfig/net_settings: made ipv6 optional feature
...
This is to prepare for szl, which cannot use ipv6 due to memory
limitation.
2020-09-01 14:48:19 +08:00
04437e876c
libconfig/load_pl: added alignment for devc buffer
...
According to the TRM, the buffer should be 64B aligned.
Without the alignment would cause failure for the DMA transaction.
It seems that the allocator would give some alignment, but to be more
correct we should specify that with the alloc interface.
2020-09-01 14:48:19 +08:00
