Nixl agent - #1
Draft
tshmilnvidia wants to merge 1168 commits into
Draft
Conversation
tshmilnvidia
marked this pull request as draft
June 25, 2025 14:46
tshmilnvidia
force-pushed
the
nixl_agent
branch
3 times, most recently
from
June 29, 2025 08:41
6dc47e6 to
74b88cf
Compare
tshmilnvidia
force-pushed
the
nixl_agent
branch
4 times, most recently
from
July 6, 2025 11:55
7210604 to
96a28f6
Compare
tshmilnvidia
force-pushed
the
nixl_agent
branch
5 times, most recently
from
July 16, 2025 10:11
57f2533 to
b7dbd4b
Compare
glevnv
force-pushed
the
nixl_agent
branch
8 times, most recently
from
August 19, 2025 09:34
271fd4c to
4237fc5
Compare
…#7022) Signed-off-by: Fanrong Li <23290157+lfr-0531@users.noreply.github.com>
Signed-off-by: Yiqing Yan <yiqingy@nvidia.com>
Signed-off-by: qqiao <qqiao@nvidia.com>
Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com>
Signed-off-by: Patrice Castonguay <55748270+pcastonguay@users.noreply.github.com>
…0 branch (NVIDIA#7043) Signed-off-by: Iman Tabrizian <10105175+tabrizian@users.noreply.github.com> Co-authored-by: Iman Tabrizian <10105175+Tabrizian@users.noreply.github.com>
…orch (NVIDIA#6761) Signed-off-by: Michal Guzek <mguzek@nvidia.com>
tshmilnvidia
force-pushed
the
nixl_agent
branch
from
August 19, 2025 16:27
4237fc5 to
1ac01d4
Compare
…DIA#7458) Signed-off-by: leslie-fang25 <leslief@nvidia.com>
…ility (NVIDIA#7313) Signed-off-by: Frida Hou <201670829+Fridah-nv@users.noreply.github.com>
…IDIA#7114) Signed-off-by: Yao Yao <lowsfer@users.noreply.github.com>
…VIDIA#7338) Signed-off-by: Lizhi Zhou <1432185+reasonsolo@users.noreply.github.com>
…VIDIA#7492) Signed-off-by: Stanley Sun <190317771+StanleySun639@users.noreply.github.com>
NVIDIA#7442) Signed-off-by: Jin Li <59594262+liji-nv@users.noreply.github.com>
* Why? Initial profiling showed there were multiple D2H / H2D copies being scheduled in the mistral 3.1 small model. * What? This commit removes those unnecessary copies by returning `image_sizes` as a simple list instead of a tensor. Signed-off-by: William Zhang <133824995+2ez4bz@users.noreply.github.com>
Signed-off-by: qqiao <qqiao@nvidia.com>
Signed-off-by: Yanchao Lu <yanchaol@nvidia.com>
Signed-off-by: greg-kwasniewski1 <213329731+greg-kwasniewski1@users.noreply.github.com>
Signed-off-by: sorenwu <sorenwu@tencent.com>
Signed-off-by: Yiqing Yan <yiqingy@nvidia.com>
…ger (NVIDIA#7437) Signed-off-by: linquanh <linquanh@nvidia.com>
tshmilnvidia
force-pushed
the
nixl_agent
branch
2 times, most recently
from
September 4, 2025 10:34
351b864 to
507570f
Compare
…dDockerImage (NVIDIA#6729) Signed-off-by: ZhanruiSunCh <184402041+ZhanruiSunCh@users.noreply.github.com> Signed-off-by: Zhanrui Sun <184402041+ZhanruiSunCh@users.noreply.github.com> Co-authored-by: Yanchao Lu <yanchaol@nvidia.com>
…er than ep_size (NVIDIA#7331) Signed-off-by: Fred Wei <20514172+WeiHaocheng@users.noreply.github.com>
Signed-off-by: Wanli Jiang <35160485+Wanli-Jiang@users.noreply.github.com>
…VIDIA#7489) Signed-off-by: junq <22017000+QiJune@users.noreply.github.com>
Signed-off-by: Ivy Zhang <25222398+crazydemo@users.noreply.github.com>
Passing mode & directory parameters to relevant onboard & offload functions. Signed-off-by: Tomer Shmilovich <tshmilovich@nvidia.com>
tshmilnvidia
force-pushed
the
nixl_agent
branch
2 times, most recently
from
September 8, 2025 08:33
becd2a4 to
6b92b32
Compare
Implement class LoopbackAgent. Signed-off-by: Tomer Shmilovich <tshmilovich@nvidia.com>
Signed-off-by: Tomer Shmilovich <tshmilovich@nvidia.com>
Signed-off-by: Tomer Shmilovich <tshmilovich@nvidia.com>
Signed-off-by: Guy Lev <glev@nvidia.com>
Signed-off-by: Guy Lev <glev@nvidia.com>
Signed-off-by: Tomer Shmilovich <tshmilovich@nvidia.com>
tshmilnvidia
force-pushed
the
nixl_agent
branch
from
September 8, 2025 08:44
6b92b32 to
ed5fd42
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
20 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Nixl support for GDS
Commit 1 & 2 ("BlockManager copy constructor fix" + "Pass mode & directory") are both fixes that I believe are needed in TRT-LLM first, not related to NIXL.
Commit 3 ("GDS Debug") can be ignored for review.
Commit 4 & 5 ("Nixl Loopback Agent" + "copyBlock with NixlLoopbackAgent") Are the relevant commits for GDS w/ NIXL.