Commit Graph

14325 Commits

Author SHA1 Message Date
Jiaying Wang
3f9849a63b
Merge pull request #7892 from filecoin-project/jen/rmapp100
build: release: v1.13.2-rc7 fix: remove app image from the bundle script
2022-01-05 19:42:50 -05:00
Jennifer Wang
09cd258780 lotus v1.13.2-rc7 2022-01-05 19:27:00 -05:00
Jennifer Wang
7d65a2664d remove from the bundlde 2022-01-05 19:27:00 -05:00
Jiaying Wang
d29e3afc28
Merge pull request #7882 from filecoin-project/jen/rmapp
build: release: v1.13.2-rc6 fix: publish ci job
2022-01-05 18:35:20 -05:00
Jennifer Wang
ff949d84ad v1.13.2-rc6 prep 2022-01-04 17:58:18 -05:00
Jennifer Wang
6b0868089b stop publishing appimagine since its disabled by #7707 2022-01-04 17:57:57 -05:00
Jiaying Wang
2474c0b633
Merge pull request #7817 from filecoin-project/jen/things
build: release: v1.13.2-rc5
2022-01-04 16:33:13 -05:00
Jennifer Wang
eabb9639a3 v1.13.2-rc5 prep 2021-12-17 15:02:42 -05:00
Cory Schwartz
cfb6aca9a4 restore .circleci 2021-12-17 00:09:12 -05:00
Cory Schwartz
c37271422b try stable 2021-12-17 00:09:07 -05:00
Cory Schwartz
975090256d tmp: publish to snapcraft edge 2021-12-17 00:09:00 -05:00
Cory Schwartz
cbf308a01d remove devel grade 2021-12-17 00:08:53 -05:00
Jiaying Wang
e05fdf8923
Merge pull request #7761 from filecoin-project/jen/rcgain
build: release: v1.13.2-rc4
2021-12-10 22:06:43 -05:00
Jiaying Wang
2343778559
Update CHANGELOG.md
Co-authored-by: Aayush Rajasekaran <arajasek94@gmail.com>
2021-12-10 19:43:25 -05:00
Jennifer Wang
1ffc3ed3a6 v1.13.2-rc4 2021-12-10 16:58:58 -05:00
Łukasz Magiera
ba3c96f8c6 stores: Reduce log spam during retrievals 2021-12-10 16:47:32 -05:00
Łukasz Magiera
b7912cbe0b fix lint 2021-12-10 16:47:22 -05:00
Łukasz Magiera
dafdb7689c Fix mock ReadPiece 2021-12-10 16:47:12 -05:00
Łukasz Magiera
46ba2b6b4f fr32: Reduce MTTresh from 32M to 512k per core
This results in 64x less bytes allocated when spawning new readers
for larger pieces.

Results in about 30% speedup in 1G unpad benchmark on AMD TR 2950x
2021-12-10 16:47:07 -05:00
Łukasz Magiera
b4c1e340ea piecereader: Avoid allocating 1024MB slices per read 2021-12-10 16:47:01 -05:00
Łukasz Magiera
a438e6fa73 piecereader: Avoid redundant roundtrips when seeking 2021-12-10 16:46:57 -05:00
Łukasz Magiera
b21d3ded2f piecereader: Move closer to storage 2021-12-10 16:46:52 -05:00
Jiaying Wang
09a5c2e025
Merge pull request #7752 from filecoin-project/jen/13.2rc3
build: release: v1.13.2-rc3
2021-12-09 22:26:16 -05:00
Jennifer Wang
3f9f9eb667 v1.13.2-rc3 2021-12-09 21:15:49 -05:00
Jennifer Wang
29cdee761b update to ipld-legacy to v0.1.1 2021-12-09 21:06:56 -05:00
Jiaying Wang
67bedc4ace
Merge pull request #7745 from filecoin-project/jen/v1.13.2rc2
build: release: v1.13.2-rc2
2021-12-08 02:38:06 -05:00
Jennifer Wang
55ae9e8414 update changelog 2021-12-08 02:26:11 -05:00
Jennifer Wang
4768c425d1 bump version to v1.13.2-rc2 2021-12-08 02:04:39 -05:00
Łukasz Magiera
b4bdb89f65 CARv2 v2.1.0 2021-12-08 01:52:00 -05:00
Łukasz Magiera
ea9130256c dagstore pieceReader: Fix wrong ErrUnexpectedEOF return in ReadAt 2021-12-08 01:51:56 -05:00
Łukasz Magiera
cde687bbe1 dagstore pieceReader: Always read full in ReadAt 2021-12-08 01:51:47 -05:00
Łukasz Magiera
9725d72a80 Add metrics to dagstore piecereader 2021-12-08 01:51:43 -05:00
Travis Person
fbd3c90b80 disable building of appimage on release
appimage testing is currently broken due to a requirement to have access
to /dev/snd, which is not present on the image we are using in circleci.
2021-12-08 01:02:17 -05:00
Jiaying Wang
52c7fa2427
Merge pull request #7718 from filecoin-project/jen/new-prep
v1.13.2-rc1
2021-11-30 19:23:09 -05:00
Jennifer Wang
02fa0df81c update change log 2021-11-30 19:06:11 -05:00
Jennifer Wang
027017d0d2 update the version to v1.13.2-rc1 2021-11-30 19:03:24 -05:00
Łukasz Magiera
26c9120d22
Merge pull request #7714 from filecoin-project/feat/sched-review
Address Scheduler enhancements (#7703) review
2021-11-30 23:03:03 +01:00
Łukasz Magiera
71329f6c41 Address Scheduler enhancements (#7703) review 2021-11-30 20:50:40 +01:00
Łukasz Magiera
73f16f08e3
Merge pull request #7703 from filecoin-project/feat/scheduler-enhancements
Scheduler enhancements
2021-11-30 19:14:25 +01:00
Łukasz Magiera
330cfc33ee worker: Typo in resources cmd usage
Co-authored-by: Aayush Rajasekaran <arajasek94@gmail.com>
2021-11-30 02:06:58 +01:00
Łukasz Magiera
cf20b0b2b8 worker: Command to print resource-table env vars 2021-11-30 02:06:58 +01:00
Łukasz Magiera
001ecbb561 fix lint 2021-11-30 02:06:58 +01:00
Łukasz Magiera
a597b072b8 fix sched tests 2021-11-30 02:06:58 +01:00
Łukasz Magiera
f25efecb74 worker: Test resource table overrides 2021-11-30 02:06:58 +01:00
Łukasz Magiera
6d52d8552b Fix docsgen 2021-11-30 02:06:58 +01:00
Łukasz Magiera
c9a2ff4007 cleanup worker resource overrides 2021-11-30 02:06:58 +01:00
Łukasz Magiera
b961e1aab5 sched resources: Separate Parallelism defaults depending on GPU presence 2021-11-30 02:06:58 +01:00
Łukasz Magiera
36868a8749 sched: C2 is not all-core load 2021-11-30 02:06:58 +01:00
Clint Armstrong
4ef8543128 Permit workers to override resource table
In an environment with heterogenious worker nodes, a universal resource
table for all workers does not allow effective scheduling of tasks. Some
workers may have different proof cache settings, changing the required
memory for different tasks. Some workers may have a different count of
CPUs per core-complex, changing the max parallelism of PC1.

This change allows workers to customize these parameters with
environment variables. A worker could set the environment variable
PC1_MIN_MEMORY for example to customize the minimum memory requirement
for PC1 tasks. If no environment variables are specified, the resource
table on the miner is used, except for PC1 parallelism.

If PC1_MAX_PARALLELISM is not specified, and
FIL_PROOFS_USE_MULTICORE_SDR is set, PC1_MAX_PARALLELSIM will
automatically be set to FIL_PROOFS_MULTICORE_SDR_PRODUCERS + 1.
2021-11-30 02:06:58 +01:00
Clint Armstrong
93e4656a27 Use a float to represent GPU utilization
Before this change workers can only be allocated one GPU task,
regardless of how much of the GPU resources that task uses, or how many
GPUs are in the system.

This makes GPUUtilization a float which can represent that a task needs
a portion, or multiple GPUs. GPUs are accounted for like RAM and CPUs so
that workers with more GPUs can be allocated more tasks.

A known issue is that PC2 cannot use multiple GPUs. And even if the
worker has multiple GPUs and is allocated multiple PC2 tasks, those
tasks will only run on the first GPU.

This could result in unexpected behavior when a worker with multiple
GPUs is assigned multiple PC2 tasks. But this should not suprise any
existing users who upgrade, as any existing users who run workers with
multiple GPUs should already know this and be running a worker per GPU
for PC2. But now those users have the freedom to customize the GPU
utilization of PC2 to be less than one and effectively run multiple PC2
processes in a single worker.

C2 is capable of utilizing multiple GPUs, and now workers can be
customized for C2 accordingly.
2021-11-30 02:06:58 +01:00