You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@arrow.apache.org by "thisisnic (via GitHub)" <gi...@apache.org> on 2026/01/24 16:21:44 UTC

[PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

thisisnic opened a new pull request, #48974:
URL: https://github.com/apache/arrow/pull/48974

   ### Rationale for this change
   
   Benchmark failing since C++20 upgrade due to lack of C++20 configuration
   
   ### What changes are included in this PR?
   
   Patch in config for C++20
   
   ### Are these changes tested?
   
   I ran it locally in a container but I'm not convinced we'll know for sure til we try it out properly.
   
   ### Are there any user-facing changes?
   
   Nope


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "pitrou (via GitHub)" <gi...@apache.org>.
pitrou commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3804022828

   > Looks like the benchmarks are building - I think the one which failed is a timeout though unsure why.
   
   We'll see if that reproduces in later benchmark runs, but this PR can certainly be merged anyway.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "pitrou (via GitHub)" <gi...@apache.org>.
pitrou commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3805774002

   That doesn't sound like an explanation to me, because there were no timeouts before. I'm having trouble connecting to https://conbench.arrow-dev.org/ so it's difficult to see the actual benchmark runtimes...


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "rok (via GitHub)" <gi...@apache.org>.
rok commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3806431096

   Conbench stable right now, but needs more work.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "pitrou (via GitHub)" <gi...@apache.org>.
pitrou commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3805802707

   If I look for "language=R" in https://conbench.arrow-dev.org/runs/3263aa975d604202b797ab81eead2317/, the benchmarks only last a couple seconds each, so this probably can't be the explanation even if they are repeated a few times each.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "rok (via GitHub)" <gi...@apache.org>.
rok commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3805858577

   > That doesn't sound like an explanation to me, because there were no timeouts before. I'm having trouble connecting to https://conbench.arrow-dev.org/ so it's difficult to see the actual benchmark runtimes...
   
   Yes, that is not an explanation.
   Let me see what's happening with conbench.
   
   > If I look for "language=R" in https://conbench.arrow-dev.org/runs/3263aa975d604202b797ab81eead2317/, the benchmarks only last a couple seconds each, so this probably can't be the explanation even if they are repeated a few times each.
   
   True. We don't need a big regression to pass over the timeout threshold. And the regression could be in environment setup or such.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "thisisnic (via GitHub)" <gi...@apache.org>.
thisisnic commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3807314987

   I compared with the log from run 305, and got the below.
   
   ___________________________________
   
   | Step                       | Old    | New    | Diff   |                                                                                                 
     |----------------------------|--------|--------|--------|                                                                                                 
     | build_arrow_cpp            | 6m     | 5m     | -0m    |                                                                                                 
     | build_arrow_python         | 4m     | 4m     | -0m    |                                                                                                 
     | cpp-micro                  | 70m    | 105m   | +35m   |                                                                                                 
     | csv-read                   | 16m    | 36m    | +19m   |                                                                                                 
     | dataframe-to-table (Py)    | 3m     | 4m     | +2m    |                                                                                                 
     | dataset-filter             | 0m     | 1m     | +1m    |                                                                                                 
     | dataset-read               | 2m     | 7m     | +4m    |                                                                                                 
     | dataset-select             | 0m     | 0m     | +0m    |                                                                                                 
     | dataset-selectivity        | 11m    | 11m    | +1m    |                                                                                                 
     | dataset-serialize          | 34m    | 45m    | +11m   |                                                                                                 
     | file-read (Py)             | 10m    | 12m    | +2m    |                                                                                                 
     | file-write (Py)            | 15m    | 16m    | +1m    |                                                                                                 
     | recursive-get-file-info    | 0m     | 5m     | +5m    |                                                                                                 
     | wide-dataframe             | 0m     | 0m     | +0m    |                                                                                                 
     | build_arrow_r              | 45m    | 45m    | +0m    |                                                                                                 
     | dataframe-to-table (R)     | 4m     | 5m     | +2m    |                                                                                                 
     | file-read (R)              | 6m     | 12m    | +6m    |                                                                                                 
     | file-write (R)             | 26m    | 36m    | +10m   |                                                                                                 
     | partitioned-dataset-filter | 7m     | 11m    | +4m    |                                                                                                 
     | tpch                       | 135m   | 48m*   | -87m   |                                                                                                 
     | **TOTAL**                  | **394m** | **420m** | **+26m** |                                                                                           
                                                                                                                                                               
     *killed before completion                                                                                                                                 
                                                                                                    
   Old job: 394m (~6.5h), completed                                                                                                                          
   New job: 420m (7h), timed out                                                                                                                             
   
   Almost everything got slower. The cumulative slowdown (~1h extra before tpch starts) means tpch doesn't have enough time to finish within the timeout.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "pitrou (via GitHub)" <gi...@apache.org>.
pitrou commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3809752593

   > Almost everything got slower. The cumulative slowdown (~1h extra before tpch starts) means tpch doesn't have enough time to finish within the timeout.
   
   Thanks for the numbers. Yes, it seems that it's just that. It's a bit surprising, though, for example cpp-micro shouldn't have "regressed" that much (and the benchmark numbers don't seem worse generally). csv-read is quite surprising too.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "thisisnic (via GitHub)" <gi...@apache.org>.
thisisnic commented on code in PR #48974:
URL: https://github.com/apache/arrow/pull/48974#discussion_r2731986960


##########
dev/conbench_envs/hooks.sh:
##########
@@ -59,6 +59,25 @@ build_arrow_python() {
 
 build_arrow_r() {
   cat ci/etc/rprofile >> $(R RHOME)/etc/Rprofile.site
+
+  # Ensure CXX20 is configured in R's Makeconf.
+  # conda-forge's R may have empty CXX20 entries even though the compiler supports it.

Review Comment:
   Wasn't sure it was a bug but reported it at https://github.com/conda-forge/r-base-feedstock/issues/401



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "rok (via GitHub)" <gi...@apache.org>.
rok commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3799755841

   Benchmark runs are scheduled for commit f164e4dfe6241749f5a080ccb673cea7fe9d0ece. Watch https://buildkite.com/apache-arrow and https://conbench.arrow-dev.org for updates. A comment will be posted here when the runs are complete.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "thisisnic (via GitHub)" <gi...@apache.org>.
thisisnic commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3802672438

   Looks like the benchmarks are building - I think the one which failed is a timeout though unsure why. 


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "pitrou (via GitHub)" <gi...@apache.org>.
pitrou commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3805561844

   > We probably need to extend buildkite timeouts - currently they're at 7h for m5 machines, extending to 8h. Also extending t4g machines from 4h to 7h.
   
   Yeah, perhaps, but please take a look at the cause for slowness too.
   
   A C++ benchmarks run seems to take [around one hour](https://buildkite.com/apache-arrow/arrow-bci-benchmark-on-amd64-c6a-4xlarge-linux/builds/358/steps/canvas), adding the other benchmarks (Python, R) should perhaps jump the runtime to 4+ hours.
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "thisisnic (via GitHub)" <gi...@apache.org>.
thisisnic commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3795071973

   I did some more digging with :robot: and ended up here @jonkeane - I spent some time asking questions and getting verification from code, so it's a bit over my head, so let me know if this sounds feasible to you!
   
   >   Problem: conda-forge builds R with gcc 10, which has incomplete C++20 support. R's configure tests for C++20, fails, and leaves CXX20 empty in Makeconf. Even though benchmark machines install gcc 14.3.0 via the compilers package, R doesn't know to use it for C++20.                                                                                                           
   >                                                                                                                                                                                              
   >  Our fix: Check if R CMD config CXX20 returns empty, and if so, append CXX20 settings to Makeconf pointing to g++ (which resolves to gcc 14.3.0 in the conda environment).                   
     >                                                                                                                                                                                            
    > Why it works: The compiler is there and capable - R's Makeconf just wasn't configured to use it. Our patch fills in that missing config.                                                    


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "github-actions[bot] (via GitHub)" <gi...@apache.org>.
github-actions[bot] commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3794976491

   :warning: GitHub issue #48912 **has been automatically assigned in GitHub** to PR creator.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "rok (via GitHub)" <gi...@apache.org>.
rok commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3807375824

   Oh, so it's so it's a regression so big we couldn't catch it :D


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "pitrou (via GitHub)" <gi...@apache.org>.
pitrou commented on code in PR #48974:
URL: https://github.com/apache/arrow/pull/48974#discussion_r2731013823


##########
dev/conbench_envs/hooks.sh:
##########
@@ -59,6 +59,25 @@ build_arrow_python() {
 
 build_arrow_r() {
   cat ci/etc/rprofile >> $(R RHOME)/etc/Rprofile.site
+
+  # Ensure CXX20 is configured in R's Makeconf.
+  # conda-forge's R may have empty CXX20 entries even though the compiler supports it.

Review Comment:
   Did we report an issue to the corresponding conda-forge feedstock, so that we can get rid of this workaround later?



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "conbench-apache-arrow[bot] (via GitHub)" <gi...@apache.org>.
conbench-apache-arrow[bot] commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3801841070

   Thanks for your patience. Conbench analyzed the 3 benchmarking runs that have been run so far on PR commit f164e4dfe6241749f5a080ccb673cea7fe9d0ece.
   
   There were 2 benchmark results indicating a performance regression:
   
   - Pull Request Run on `amd64-c6a-4xlarge-linux` at [2026-01-26 15:35:45Z](https://conbench.arrow-dev.org/compare/runs/55976d7d8d6040f7a67af93172cc3845...91bdfe02bed44a9b91e6d3d8bcf26101/)
     - [`UniqueInt64` (C++) with params=0, source=cpp-micro, suite=arrow-compute-vector-hash-benchmark](https://conbench.arrow-dev.org/compare/benchmarks/0697406e25e476d380008d4607507a9a...069778f3c2c3720d8000182b544e1fcb)
     - [`HashIntegers` (C++) with source=cpp-micro, suite=arrow-hashing-benchmark](https://conbench.arrow-dev.org/compare/benchmarks/0697406bdff07c258000f50ff4121fae...069778f2009e773d8000a4f6fc3472c1)
   
   The [full Conbench report](https://github.com/apache/arrow/runs/61527580443) has more details.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "jonkeane (via GitHub)" <gi...@apache.org>.
jonkeane commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3795776664

   That sounds ok to me if it works!


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "thisisnic (via GitHub)" <gi...@apache.org>.
thisisnic merged PR #48974:
URL: https://github.com/apache/arrow/pull/48974


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "conbench-apache-arrow[bot] (via GitHub)" <gi...@apache.org>.
conbench-apache-arrow[bot] commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3808796679

   After merging your PR, Conbench analyzed the 3 benchmarking runs that have been run so far on merge-commit de06a522ffdd32f7d9fd942fe8d1e3305139d780.
   
   There was 1 benchmark result indicating a performance regression:
   
   - Commit Run on `amd64-m5-4xlarge-linux` at [2026-01-27 21:29:34Z](https://conbench.arrow-dev.org/compare/runs/77f8db9c57ee4845a0109c95819b0246...f829ddc0be944d5aa38aa416c7c3a15b/)
     - [`file-read` (Python) with compression=uncompressed, dataset=fanniemae_2016Q4, file_type=feather, output_type=table](https://conbench.arrow-dev.org/compare/benchmarks/06978f17e99c758a80006b2bd123aa40...069791e075a1796b800000049aaffaee)
   
   The [full Conbench report](https://github.com/apache/arrow/runs/61689847532) has more details. It also includes information about 18 possible false positives for unstable benchmarks that are known to sometimes produce them.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "rok (via GitHub)" <gi...@apache.org>.
rok commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3805624739

   AFAIR R benchmarks include [tpc-h](https://github.com/arctosalliance/benchmarks/blob/main/benchmarks.json#L126) and other magical incantations.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "thisisnic (via GitHub)" <gi...@apache.org>.
thisisnic commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3806956367

   I got :robot: to parse the logs for the job which timed out, longest running components are highlighted in bold.
   
    | Step                       | Start | End   | Duration |                                                                                                 
     |----------------------------|-------|-------|----------|                                                                                                 
     | build_arrow_cpp            | 6m    | 11m   | 5m       |                                                                                                 
     | build_arrow_python         | 11m   | 15m   | 4m       |                                                                                                 
     | cpp-micro                  | 15m   | 120m  | **105m** |                                                                                                 
     | csv-read                   | 121m  | 156m  | 36m      |                                                                                                 
     | dataframe-to-table (Py)    | 156m  | 161m  | 4m       |                                                                                                 
     | dataset-filter             | 161m  | 162m  | 1m       |                                                                                                 
     | dataset-read               | 162m  | 169m  | 7m       |                                                                                                 
     | dataset-select             | 169m  | 169m  | 0m       |                                                                                                 
     | dataset-selectivity        | 169m  | 180m  | 11m      |                                                                                                 
     | dataset-serialize          | 180m  | 225m  | **45m**  |                                                                                                 
     | file-read (Py)             | 225m  | 237m  | 12m      |                                                                                                 
     | file-write (Py)            | 237m  | 253m  | 16m      |                                                                                                 
     | recursive-get-file-info    | 253m  | 258m  | 5m       |                                                                                                 
     | wide-dataframe             | 258m  | 258m  | 0m       |                                                                                                 
     | build_arrow_r              | 258m  | 304m  | **45m**  |                                                                                                 
     | dataframe-to-table (R)     | 305m  | 311m  | 5m       |                                                                                                 
     | file-read (R)              | 313m  | 325m  | 12m      |                                                                                                 
     | file-write (R)             | 325m  | 361m  | **36m**  |                                                                                                 
     | partitioned-dataset-filter | 361m  | 372m  | 11m      |                                                                                                 
     | tpch                       | 372m  | 420m  | **48m+** (killed) | 


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "rok (via GitHub)" <gi...@apache.org>.
rok commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3805382041

   > Looks like the benchmarks are building - I think the one which failed is a timeout though unsure why.
   
   We probably need to extend buildkite timeouts - currently they're at 7h for m5 machines, extending to 8h. Also extending t4g machines from 4h to 7h.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] GH-48912: [R] Configure C++20 in conda R on continuous benchmarking [arrow]

Posted by "pitrou (via GitHub)" <gi...@apache.org>.
pitrou commented on PR #48974:
URL: https://github.com/apache/arrow/pull/48974#issuecomment-3799754971

   @ursabot please benchmark


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org