Skip to content

Conversation

gbaraldi
Copy link
Member

This does not fix the underlying issue that can occur here which is a collision of build_ids.lo between modules in IR decompression. Fixing that requires a somewhat significant overhaul to the serialization of IR (probably using the module identity as a key). This does mean we use a lot more of the bits available here so it makes collisions a lot less likely( they were already extremely rare) but hrtime does tend to only use the lower bits of a 64 bit integer and this will hopefully add some more randomness and make this even less likely

@gbaraldi gbaraldi added backport 1.10 Change should be backported to the 1.10 release backport 1.11 Change should be backported to release-1.11 backport 1.12 Change should be backported to release-1.12 labels Apr 28, 2025
@KristofferC
Copy link
Member

KristofferC commented Apr 28, 2025

Would it be possible to also add a check that the root_blocks array does not have duplicated entries so in case something goes wrong we won't silently run with bad data but get a hard error?

@KristofferC KristofferC mentioned this pull request Apr 29, 2025
53 tasks
@gbaraldi gbaraldi force-pushed the gb/better-buildid branch from 794a26c to eb401f3 Compare April 30, 2025 21:33
@kpamnany kpamnany linked an issue May 1, 2025 that may be closed by this pull request
@gbaraldi gbaraldi added the merge me PR is reviewed. Merge when all tests are passing label May 2, 2025
@kpamnany kpamnany merged commit 7157407 into master May 3, 2025
6 of 8 checks passed
@kpamnany kpamnany deleted the gb/better-buildid branch May 3, 2025 15:33
kpamnany pushed a commit to RelationalAI/julia that referenced this pull request May 3, 2025
This does not fix the underlying issue that can occur here which is a
collision of build_ids.lo between modules in IR decompression. Fixing
that requires a somewhat significant overhaul to the serialization of IR
(probably using the module identity as a key). This does mean we use a
lot more of the bits available here so it makes collisions a lot less
likely( they were already extremely rare) but hrtime does tend to only
use the lower bits of a 64 bit integer and this will hopefully add some
more randomness and make this even less likely
kpamnany pushed a commit to RelationalAI/julia that referenced this pull request May 3, 2025
This does not fix the underlying issue that can occur here which is a
collision of build_ids.lo between modules in IR decompression. Fixing
that requires a somewhat significant overhaul to the serialization of IR
(probably using the module identity as a key). This does mean we use a
lot more of the bits available here so it makes collisions a lot less
likely( they were already extremely rare) but hrtime does tend to only
use the lower bits of a 64 bit integer and this will hopefully add some
more randomness and make this even less likely
kpamnany added a commit to RelationalAI/julia that referenced this pull request May 3, 2025
This does not fix the underlying issue that can occur here which is a
collision of build_ids.lo between modules in IR decompression. Fixing
that requires a somewhat significant overhaul to the serialization of IR
(probably using the module identity as a key). This does mean we use a
lot more of the bits available here so it makes collisions a lot less
likely( they were already extremely rare) but hrtime does tend to only
use the lower bits of a 64 bit integer and this will hopefully add some
more randomness and make this even less likely

Co-authored-by: Gabriel Baraldi <[email protected]>
KristofferC pushed a commit that referenced this pull request May 5, 2025
This does not fix the underlying issue that can occur here which is a
collision of build_ids.lo between modules in IR decompression. Fixing
that requires a somewhat significant overhaul to the serialization of IR
(probably using the module identity as a key). This does mean we use a
lot more of the bits available here so it makes collisions a lot less
likely( they were already extremely rare) but hrtime does tend to only
use the lower bits of a 64 bit integer and this will hopefully add some
more randomness and make this even less likely

(cherry picked from commit 7157407)
@giordano giordano removed the merge me PR is reviewed. Merge when all tests are passing label May 5, 2025
@KristofferC KristofferC removed the backport 1.12 Change should be backported to release-1.12 label May 9, 2025
charleskawczynski pushed a commit to charleskawczynski/julia that referenced this pull request May 12, 2025
This does not fix the underlying issue that can occur here which is a
collision of build_ids.lo between modules in IR decompression. Fixing
that requires a somewhat significant overhaul to the serialization of IR
(probably using the module identity as a key). This does mean we use a
lot more of the bits available here so it makes collisions a lot less
likely( they were already extremely rare) but hrtime does tend to only
use the lower bits of a 64 bit integer and this will hopefully add some
more randomness and make this even less likely
KristofferC pushed a commit that referenced this pull request Jun 4, 2025
This does not fix the underlying issue that can occur here which is a
collision of build_ids.lo between modules in IR decompression. Fixing
that requires a somewhat significant overhaul to the serialization of IR
(probably using the module identity as a key). This does mean we use a
lot more of the bits available here so it makes collisions a lot less
likely( they were already extremely rare) but hrtime does tend to only
use the lower bits of a 64 bit integer and this will hopefully add some
more randomness and make this even less likely

(cherry picked from commit 7157407)
This was referenced Jun 4, 2025
KristofferC pushed a commit that referenced this pull request Jun 5, 2025
This does not fix the underlying issue that can occur here which is a
collision of build_ids.lo between modules in IR decompression. Fixing
that requires a somewhat significant overhaul to the serialization of IR
(probably using the module identity as a key). This does mean we use a
lot more of the bits available here so it makes collisions a lot less
likely( they were already extremely rare) but hrtime does tend to only
use the lower bits of a 64 bit integer and this will hopefully add some
more randomness and make this even less likely

(cherry picked from commit 7157407)
KristofferC pushed a commit that referenced this pull request Jun 5, 2025
This does not fix the underlying issue that can occur here which is a
collision of build_ids.lo between modules in IR decompression. Fixing
that requires a somewhat significant overhaul to the serialization of IR
(probably using the module identity as a key). This does mean we use a
lot more of the bits available here so it makes collisions a lot less
likely( they were already extremely rare) but hrtime does tend to only
use the lower bits of a 64 bit integer and this will hopefully add some
more randomness and make this even less likely

(cherry picked from commit 7157407)
KristofferC pushed a commit that referenced this pull request Jun 5, 2025
This does not fix the underlying issue that can occur here which is a
collision of build_ids.lo between modules in IR decompression. Fixing
that requires a somewhat significant overhaul to the serialization of IR
(probably using the module identity as a key). This does mean we use a
lot more of the bits available here so it makes collisions a lot less
likely( they were already extremely rare) but hrtime does tend to only
use the lower bits of a 64 bit integer and this will hopefully add some
more randomness and make this even less likely

(cherry picked from commit 7157407)
KristofferC pushed a commit that referenced this pull request Jun 5, 2025
This does not fix the underlying issue that can occur here which is a
collision of build_ids.lo between modules in IR decompression. Fixing
that requires a somewhat significant overhaul to the serialization of IR
(probably using the module identity as a key). This does mean we use a
lot more of the bits available here so it makes collisions a lot less
likely( they were already extremely rare) but hrtime does tend to only
use the lower bits of a 64 bit integer and this will hopefully add some
more randomness and make this even less likely

(cherry picked from commit 7157407)
KristofferC pushed a commit that referenced this pull request Jun 5, 2025
This does not fix the underlying issue that can occur here which is a
collision of build_ids.lo between modules in IR decompression. Fixing
that requires a somewhat significant overhaul to the serialization of IR
(probably using the module identity as a key). This does mean we use a
lot more of the bits available here so it makes collisions a lot less
likely( they were already extremely rare) but hrtime does tend to only
use the lower bits of a 64 bit integer and this will hopefully add some
more randomness and make this even less likely

(cherry picked from commit 7157407)
KristofferC pushed a commit that referenced this pull request Jun 5, 2025
This does not fix the underlying issue that can occur here which is a
collision of build_ids.lo between modules in IR decompression. Fixing
that requires a somewhat significant overhaul to the serialization of IR
(probably using the module identity as a key). This does mean we use a
lot more of the bits available here so it makes collisions a lot less
likely( they were already extremely rare) but hrtime does tend to only
use the lower bits of a 64 bit integer and this will hopefully add some
more randomness and make this even less likely

(cherry picked from commit 7157407)
KristofferC pushed a commit that referenced this pull request Jul 3, 2025
This does not fix the underlying issue that can occur here which is a
collision of build_ids.lo between modules in IR decompression. Fixing
that requires a somewhat significant overhaul to the serialization of IR
(probably using the module identity as a key). This does mean we use a
lot more of the bits available here so it makes collisions a lot less
likely( they were already extremely rare) but hrtime does tend to only
use the lower bits of a 64 bit integer and this will hopefully add some
more randomness and make this even less likely

(cherry picked from commit 7157407)
@KristofferC KristofferC removed the backport 1.11 Change should be backported to release-1.11 label Aug 19, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
backport 1.10 Change should be backported to the 1.10 release
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Julia 1.10: segfault in jl_rettype_inferred
4 participants