Skip to content

llama: fix whole source code rebuilt on every new commit - #28278

Merged
ngxson merged 1 commit into
masterfrom
xsn/fix_rebuild_new_commit
Sep 3, 2026
Merged

llama: fix whole source code rebuilt on every new commit#28278
ngxson merged 1 commit into
masterfrom
xsn/fix_rebuild_new_commit

Conversation

@ngxson

@ngxson ngxson commented Sep 2, 2026

Copy link
Copy Markdown
Collaborator

Overview

Recently I noticed that the whole src/models dir is rebuilt on every new commit (observed locally on my machine), seems to be related to recent changes in semver

With this patch, versions are written to llama-version.h and only llama.cpp will be recompiled

Requirements

@ngxson
ngxson requested a review from danbev September 2, 2026 23:09
@ngxson
ngxson requested a review from ggerganov as a code owner September 2, 2026 23:09
@ngxson ngxson changed the title llama: fix whole source code rebuilt on each new commit llama: fix whole source code rebuilt on every new commit Sep 2, 2026
nikwen
nikwen approved these changes Sep 3, 2026
View reviewed changes

@nikwen nikwen left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks, noticed these rebuilds, too! 🙏

Even though I'll approve, I think it would make sense to wait for a review of someone else with more CMake knowledge.

In particular, it would be great to check if CMAKE_CURRENT_BINARY_DIR is the best location for this file (especially, given that we add it as an include directory now).

@ngxson
ngxson merged commit d230ddd into master Sep 3, 2026
31 of 40 checks passed
This was referenced Sep 4, 2026
Merged
Open
danbev added a commit that referenced this pull request Sep 4, 2026
@danbev
ggml : replace compile definitions with version.h.in (#28364)
86b351f 
This commit adds a cmake version configuration file to replace the
current compile definition solution for the version.

The motivation for this change is that I made a mistake and did not take
into consideration that the compile definition means that this will
become a compiler flag for all sources in the target. This means that
when a version update happens that will recompile all sources in the
target even if they have not changed.

Refs: #28278
@bernardladenthin bernardladenthin mentioned this pull request Sep 4, 2026
Merged
10 tasks
ggerganov pushed a commit to ggml-org/ggml that referenced this pull request Sep 4, 2026
@danbev @ggerganov
ggml : replace compile definitions with version.h.in (llama/28364)
72d22bd 
This commit adds a cmake version configuration file to replace the
current compile definition solution for the version.

The motivation for this change is that I made a mistake and did not take
into consideration that the compile definition means that this will
become a compiler flag for all sources in the target. This means that
when a version update happens that will recompile all sources in the
target even if they have not changed.

Refs: ggml-org/llama.cpp#28278
pull Bot pushed a commit to joyshmitz/whisper.cpp that referenced this pull request Sep 4, 2026
@danbev @ggerganov
ggml : replace compile definitions with version.h.in (llama/28364)
140e57a 
This commit adds a cmake version configuration file to replace the
current compile definition solution for the version.

The motivation for this change is that I made a mistake and did not take
into consideration that the compile definition means that this will
become a compiler flag for all sources in the target. This means that
when a version update happens that will recompile all sources in the
target even if they have not changed.

Refs: ggml-org/llama.cpp#28278
fewtarius pushed a commit to fewtarius/CachyLLama that referenced this pull request Sep 5, 2026
@ngxson @fewtarius
llama: fix whole source code rebuilt on each new commit (ggml-org#28278)
c4e5c6c
fewtarius pushed a commit to fewtarius/CachyLLama that referenced this pull request Sep 5, 2026
@danbev @fewtarius
ggml : replace compile definitions with version.h.in (ggml-org#28364)
a44f905 
This commit adds a cmake version configuration file to replace the
current compile definition solution for the version.

The motivation for this change is that I made a mistake and did not take
into consideration that the compile definition means that this will
become a compiler flag for all sources in the target. This means that
when a version update happens that will recompile all sources in the
target even if they have not changed.

Refs: ggml-org#28278
alin-o pushed a commit to alin-o/llama.cpp-qwen38 that referenced this pull request Sep 6, 2026
@danbev @carapa-ai
ggml : replace compile definitions with version.h.in (#28364)
301e21e 
This commit adds a cmake version configuration file to replace the
current compile definition solution for the version.

The motivation for this change is that I made a mistake and did not take
into consideration that the compile definition means that this will
become a compiler flag for all sources in the target. This means that
when a version update happens that will recompile all sources in the
target even if they have not changed.

Refs: ggml-org/llama.cpp#28278
alin-o pushed a commit to alin-o/llama.cpp-qwen38 that referenced this pull request Sep 6, 2026
@danbev @carapa-ai
ggml : replace compile definitions with version.h.in (#28364)
d167bc0 
This commit adds a cmake version configuration file to replace the
current compile definition solution for the version.

The motivation for this change is that I made a mistake and did not take
into consideration that the compile definition means that this will
become a compiler flag for all sources in the target. This means that
when a version update happens that will recompile all sources in the
target even if they have not changed.

Refs: ggml-org/llama.cpp#28278
alin-o pushed a commit to alin-o/llama.cpp-qwen38 that referenced this pull request Sep 6, 2026
@danbev @carapa-ai
ggml : replace compile definitions with version.h.in (#28364)
ed0c45a 
This commit adds a cmake version configuration file to replace the
current compile definition solution for the version.

The motivation for this change is that I made a mistake and did not take
into consideration that the compile definition means that this will
become a compiler flag for all sources in the target. This means that
when a version update happens that will recompile all sources in the
target even if they have not changed.

Refs: ggml-org/llama.cpp#28278
SteelPh0enix pushed a commit to SteelPh0enix/llama.cpp-qwen4exp that referenced this pull request Sep 8, 2026
@ngxson @SteelPh0enix
llama: fix whole source code rebuilt on each new commit (ggml-org#28278)
3f90421 
(cherry picked from commit d230ddd)
SteelPh0enix pushed a commit to SteelPh0enix/llama.cpp-qwen4exp that referenced this pull request Sep 8, 2026
@ngxson @SteelPh0enix
llama: fix whole source code rebuilt on each new commit (ggml-org#28278)
b2023aa 
(cherry picked from commit d230ddd)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@danbev danbev danbev approved these changes
@nikwen nikwen nikwen approved these changes
@ggerganov ggerganov Awaiting requested review from ggerganov ggerganov is a code owner

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@ngxson @danbev @nikwen