From 66d83d1297450d5f53fbaee82f055f3aca45e382 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Mon, 11 Sep 2023 17:10:59 -0400
Subject: [PATCH 01/43] enable ASAN/UBSAN in pandas CI

---
 .github/actions/build_pandas/action.yml | 4 ++--
 .github/workflows/unit-tests.yml        | 6 ++++++
 2 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/.github/actions/build_pandas/action.yml b/.github/actions/build_pandas/action.yml
index 73d7723e2fb49..ba2740ed4a631 100644
--- a/.github/actions/build_pandas/action.yml
+++ b/.github/actions/build_pandas/action.yml
@@ -25,8 +25,8 @@ runs:
     - name: Build Pandas
       run: |
         if [[ ${{ inputs.editable }} == "true" ]]; then
-          pip install -e . --no-build-isolation -v
+          pip install -e . --no-build-isolation -v  --config-settings=setup-args="-Db_sanitize=address,undefined"
         else
-          pip install . --no-build-isolation -v
+          pip install . --no-build-isolation -v --config-settings=setup-args="-Db_sanitize=address,undefined"
         fi
       shell: bash -el {0}
diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index f2b426269098b..4c3925b8e8063 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -157,6 +157,8 @@ jobs:
     - name: Build Pandas
       id: build
       uses: ./.github/actions/build_pandas
+      env:
+        CFLAGS: "$CFLAGS -fno-sanitize-recover=all"
 
     - name: Test (not single_cpu)
       uses: ./.github/actions/run-tests
@@ -164,12 +166,16 @@ jobs:
       env:
         # Set pattern to not single_cpu if not already set
         PATTERN: ${{ env.PATTERN == '' && 'not single_cpu' || matrix.pattern }}
+        ASAN_OPTIONS: detect_leaks=0
+        LD_PRELOAD: $(gcc -print-file-name=libasan.so)
 
     - name: Test (single_cpu)
       uses: ./.github/actions/run-tests
       env:
         PATTERN: 'single_cpu'
         PYTEST_WORKERS: 0
+        ASAN_OPTIONS: detect_leaks=0
+        LD_PRELOAD: $(gcc -print-file-name=libasan.so)
       if: ${{ matrix.pattern == '' && (always() && steps.build.outcome == 'success')}}
 
   macos-windows:

From 7aa2e7ad38b0e3a8a29d9b405d6dc923db3be402 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Mon, 11 Sep 2023 19:44:10 -0400
Subject: [PATCH 02/43] try input

---
 .github/actions/build_pandas/action.yml | 15 +++++++++++++--
 .github/workflows/unit-tests.yml        |  2 ++
 2 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/.github/actions/build_pandas/action.yml b/.github/actions/build_pandas/action.yml
index ba2740ed4a631..063cb4c7d3254 100644
--- a/.github/actions/build_pandas/action.yml
+++ b/.github/actions/build_pandas/action.yml
@@ -4,6 +4,9 @@ inputs:
   editable:
     description: Whether to build pandas in editable mode (default true)
     default: true
+  sanitize:
+    description: Whether sanitizers should be used or not
+    default: false
 runs:
   using: composite
   steps:
@@ -25,8 +28,16 @@ runs:
     - name: Build Pandas
       run: |
         if [[ ${{ inputs.editable }} == "true" ]]; then
-          pip install -e . --no-build-isolation -v  --config-settings=setup-args="-Db_sanitize=address,undefined"
+          if [[ ${{ inputs.sanitize }} == "true" ]]; then
+            pip install -e . --no-build-isolation -v  --config-settings=setup-args="-Db_sanitize=address,undefined"
+          else
+            pip install -e . --no-build-isolation -v
+          fi
         else
-          pip install . --no-build-isolation -v --config-settings=setup-args="-Db_sanitize=address,undefined"
+          if [[ ${{ inputs.sanitize }} == "true" ]]; then
+            pip install . --no-build-isolation -v --config-settings=setup-args="-Db_sanitize=address,undefined"
+          else
+            pip install . --no-build-isolation -v
+          fi
         fi
       shell: bash -el {0}
diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 4c3925b8e8063..d42da5ec3f42a 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -159,6 +159,8 @@ jobs:
       uses: ./.github/actions/build_pandas
       env:
         CFLAGS: "$CFLAGS -fno-sanitize-recover=all"
+      with:
+        sanitize: true
 
     - name: Test (not single_cpu)
       uses: ./.github/actions/run-tests

From a5b380813ff448cec5380d2779f06f633c671169 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Mon, 11 Sep 2023 20:00:30 -0400
Subject: [PATCH 03/43] try removing sanitize

---
 .github/workflows/unit-tests.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index d42da5ec3f42a..68b65fcae4854 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -160,7 +160,7 @@ jobs:
       env:
         CFLAGS: "$CFLAGS -fno-sanitize-recover=all"
       with:
-        sanitize: true
+        sanitize: false
 
     - name: Test (not single_cpu)
       uses: ./.github/actions/run-tests

From 7b58c6d245083d04f5bbfc7e3c6a0feb4df102f7 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Mon, 11 Sep 2023 20:03:22 -0400
Subject: [PATCH 04/43] try no CFLAGS

---
 .github/workflows/unit-tests.yml | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 68b65fcae4854..0b2752a9c1e9e 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -157,10 +157,8 @@ jobs:
     - name: Build Pandas
       id: build
       uses: ./.github/actions/build_pandas
-      env:
-        CFLAGS: "$CFLAGS -fno-sanitize-recover=all"
       with:
-        sanitize: false
+        sanitize: true
 
     - name: Test (not single_cpu)
       uses: ./.github/actions/run-tests

From 18111b01d09f576965bda7e521bfb9ac59c60da0 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Mon, 11 Sep 2023 20:19:39 -0400
Subject: [PATCH 05/43] try GH string substituion

---
 .github/workflows/unit-tests.yml | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 0b2752a9c1e9e..9cdefaa2c6df2 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -154,9 +154,17 @@ jobs:
       with:
         environment-file: ci/deps/${{ matrix.env_file }}
 
+    - name: Set sanitizer flags
+      run: |
+        echo "CFLAGS=$CFLAGS -fno-sanitize-recover=all" >> "$GITHUB_ENV"
+    - name: echo flags
+      run: |
+        printf '%s\n' "CFLAGS"
     - name: Build Pandas
       id: build
       uses: ./.github/actions/build_pandas
+      env:
+        CFLAGS: "$CFLAGS"
       with:
         sanitize: true
 

From 438cdfa2d5e82e636374606e312a85cf7f85d9d4 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Mon, 11 Sep 2023 21:32:41 -0400
Subject: [PATCH 06/43] change flags in build script

---
 .github/actions/build_pandas/action.yml | 8 ++++++--
 .github/workflows/unit-tests.yml        | 8 --------
 2 files changed, 6 insertions(+), 10 deletions(-)

diff --git a/.github/actions/build_pandas/action.yml b/.github/actions/build_pandas/action.yml
index 063cb4c7d3254..1050369ccece9 100644
--- a/.github/actions/build_pandas/action.yml
+++ b/.github/actions/build_pandas/action.yml
@@ -29,13 +29,17 @@ runs:
       run: |
         if [[ ${{ inputs.editable }} == "true" ]]; then
           if [[ ${{ inputs.sanitize }} == "true" ]]; then
-            pip install -e . --no-build-isolation -v  --config-settings=setup-args="-Db_sanitize=address,undefined"
+            CFLAGS=$CFLAGS -fno-sanitize-recover=all \
+              pip install -e . --no-build-isolation -v  \
+              --config-settings=setup-args="-Db_sanitize=address,undefined"
           else
             pip install -e . --no-build-isolation -v
           fi
         else
           if [[ ${{ inputs.sanitize }} == "true" ]]; then
-            pip install . --no-build-isolation -v --config-settings=setup-args="-Db_sanitize=address,undefined"
+            CFLAGS=$CFLAGS -fno-sanitize-recover=all \
+              pip install . --no-build-isolation -v \
+              --config-settings=setup-args="-Db_sanitize=address,undefined"
           else
             pip install . --no-build-isolation -v
           fi
diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 9cdefaa2c6df2..0b2752a9c1e9e 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -154,17 +154,9 @@ jobs:
       with:
         environment-file: ci/deps/${{ matrix.env_file }}
 
-    - name: Set sanitizer flags
-      run: |
-        echo "CFLAGS=$CFLAGS -fno-sanitize-recover=all" >> "$GITHUB_ENV"
-    - name: echo flags
-      run: |
-        printf '%s\n' "CFLAGS"
     - name: Build Pandas
       id: build
       uses: ./.github/actions/build_pandas
-      env:
-        CFLAGS: "$CFLAGS"
       with:
         sanitize: true
 

From b18cf9dd61daeb0c4e25e768941a760bb7342d46 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Mon, 11 Sep 2023 22:15:26 -0400
Subject: [PATCH 07/43] quotes

---
 .github/actions/build_pandas/action.yml | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/.github/actions/build_pandas/action.yml b/.github/actions/build_pandas/action.yml
index 1050369ccece9..6a38ebc1c5279 100644
--- a/.github/actions/build_pandas/action.yml
+++ b/.github/actions/build_pandas/action.yml
@@ -29,7 +29,7 @@ runs:
       run: |
         if [[ ${{ inputs.editable }} == "true" ]]; then
           if [[ ${{ inputs.sanitize }} == "true" ]]; then
-            CFLAGS=$CFLAGS -fno-sanitize-recover=all \
+            CFLAGS="$CFLAGS -fno-sanitize-recover=all" \
               pip install -e . --no-build-isolation -v  \
               --config-settings=setup-args="-Db_sanitize=address,undefined"
           else
@@ -37,7 +37,7 @@ runs:
           fi
         else
           if [[ ${{ inputs.sanitize }} == "true" ]]; then
-            CFLAGS=$CFLAGS -fno-sanitize-recover=all \
+            CFLAGS="$CFLAGS -fno-sanitize-recover=all" \
               pip install . --no-build-isolation -v \
               --config-settings=setup-args="-Db_sanitize=address,undefined"
           else

From 69cb6f6cdd944566f8816d873ebce471a15d02fb Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Mon, 11 Sep 2023 22:36:49 -0400
Subject: [PATCH 08/43] update script run

---
 .github/actions/run-tests/action.yml | 13 ++++++++++++-
 .github/workflows/unit-tests.yml     |  4 ++--
 2 files changed, 14 insertions(+), 3 deletions(-)

diff --git a/.github/actions/run-tests/action.yml b/.github/actions/run-tests/action.yml
index fd7c3587f2254..b66adf18a4847 100644
--- a/.github/actions/run-tests/action.yml
+++ b/.github/actions/run-tests/action.yml
@@ -1,9 +1,20 @@
 name: Run tests and report results
+inputs:
+  sanitize:
+    description: Whether sanitizers should be used or not
+    default: false
 runs:
   using: composite
   steps:
     - name: Test
-      run: ci/run_tests.sh
+      run: |
+        if [[ ${{ inputs.sanitize }} == "true" ]]; then
+          ASAN_OPTIONS=detect_leaks=0 \
+            LD_PRELOAD=$(gcc -print-file-name=libasan.so) \
+            ci/run_tests.sh
+        else
+          ci/run_tests.sh
+        fi
       shell: bash -el {0}
 
     - name: Publish test results
diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 0b2752a9c1e9e..a219cf0aacc9d 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -163,11 +163,11 @@ jobs:
     - name: Test (not single_cpu)
       uses: ./.github/actions/run-tests
       if: ${{ matrix.name != 'Pypy' }}
+      with:
+        sanitize: true
       env:
         # Set pattern to not single_cpu if not already set
         PATTERN: ${{ env.PATTERN == '' && 'not single_cpu' || matrix.pattern }}
-        ASAN_OPTIONS: detect_leaks=0
-        LD_PRELOAD: $(gcc -print-file-name=libasan.so)
 
     - name: Test (single_cpu)
       uses: ./.github/actions/run-tests

From 6f5fb1180b92e7ecd0c96e7fb4291ccfc4ad5748 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Mon, 11 Sep 2023 22:38:22 -0400
Subject: [PATCH 09/43] single_cpu updates

---
 .github/workflows/unit-tests.yml | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index a219cf0aacc9d..907a549e7b191 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -171,11 +171,11 @@ jobs:
 
     - name: Test (single_cpu)
       uses: ./.github/actions/run-tests
+      with:
+        sanitize: true
       env:
         PATTERN: 'single_cpu'
         PYTEST_WORKERS: 0
-        ASAN_OPTIONS: detect_leaks=0
-        LD_PRELOAD: $(gcc -print-file-name=libasan.so)
       if: ${{ matrix.pattern == '' && (always() && steps.build.outcome == 'success')}}
 
   macos-windows:

From 663d6d44c1aeb08e55fd4595217837f35dd3fe95 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Wed, 13 Sep 2023 20:44:28 -0400
Subject: [PATCH 10/43] asan checks for datetime funcs

---
 pandas/_libs/tslib.pyx              |  2 +-
 pandas/_libs/tslibs/conversion.pyx  |  3 ++-
 pandas/_libs/tslibs/np_datetime.pyx |  2 ++
 pandas/_libs/tslibs/offsets.pyx     |  9 +++++++++
 pandas/_libs/tslibs/period.pyx      | 10 ++++++++++
 pandas/_libs/tslibs/strptime.pyx    |  2 ++
 pandas/_libs/tslibs/timestamps.pyx  |  1 +
 7 files changed, 27 insertions(+), 2 deletions(-)

diff --git a/pandas/_libs/tslib.pyx b/pandas/_libs/tslib.pyx
index 20a18cf56779f..14b5515f6c635 100644
--- a/pandas/_libs/tslib.pyx
+++ b/pandas/_libs/tslib.pyx
@@ -94,8 +94,8 @@ def _test_parse_iso8601(ts: str):
     obj = _TSObject()
 
     string_to_dts(ts, &obj.dts, &out_bestunit, &out_local, &out_tzoffset, True)
+    check_dts_bounds(&obj.dts, NPY_FR_ns)
     obj.value = npy_datetimestruct_to_datetime(NPY_FR_ns, &obj.dts)
-    check_dts_bounds(&obj.dts)
     if out_local == 1:
         obj.tzinfo = timezone(timedelta(minutes=out_tzoffset))
         obj.value = tz_localize_to_utc_single(obj.value, obj.tzinfo)
diff --git a/pandas/_libs/tslibs/conversion.pyx b/pandas/_libs/tslibs/conversion.pyx
index 45c4d7809fe7a..025b67c195e2c 100644
--- a/pandas/_libs/tslibs/conversion.pyx
+++ b/pandas/_libs/tslibs/conversion.pyx
@@ -397,6 +397,7 @@ cdef _TSObject convert_datetime_to_tsobject(
     if nanos:
         obj.dts.ps = nanos * 1000
 
+    check_dts_bounds(&obj.dts, reso)
     obj.value = npy_datetimestruct_to_datetime(reso, &obj.dts)
 
     if obj.tzinfo is not None and not is_utc(obj.tzinfo):
@@ -404,7 +405,6 @@ cdef _TSObject convert_datetime_to_tsobject(
         pps = periods_per_second(reso)
         obj.value -= int(offset.total_seconds() * pps)
 
-    check_dts_bounds(&obj.dts, reso)
     check_overflows(obj, reso)
     return obj
 
@@ -434,6 +434,7 @@ cdef _TSObject _create_tsobject_tz_using_offset(npy_datetimestruct dts,
         datetime dt
         Py_ssize_t pos
 
+    check_dts_bounds(&dts, reso)
     value = npy_datetimestruct_to_datetime(reso, &dts)
     obj.dts = dts
     obj.tzinfo = timezone(timedelta(minutes=tzoffset))
diff --git a/pandas/_libs/tslibs/np_datetime.pyx b/pandas/_libs/tslibs/np_datetime.pyx
index c3ee68e14a8d4..84f606767ddb9 100644
--- a/pandas/_libs/tslibs/np_datetime.pyx
+++ b/pandas/_libs/tslibs/np_datetime.pyx
@@ -267,6 +267,7 @@ cdef int64_t pydatetime_to_dt64(datetime val,
     Note we are assuming that the datetime object is timezone-naive.
     """
     pydatetime_to_dtstruct(val, dts)
+    check_dts_bounds(dts, reso)
     return npy_datetimestruct_to_datetime(reso, dts)
 
 
@@ -282,6 +283,7 @@ cdef int64_t pydate_to_dt64(
     date val, npy_datetimestruct *dts, NPY_DATETIMEUNIT reso=NPY_FR_ns
 ):
     pydate_to_dtstruct(val, dts)
+    check_dts_bounds(dts, reso)
     return npy_datetimestruct_to_datetime(reso, dts)
 
 
diff --git a/pandas/_libs/tslibs/offsets.pyx b/pandas/_libs/tslibs/offsets.pyx
index 2b6f84844ff49..8e84b441d2cf7 100644
--- a/pandas/_libs/tslibs/offsets.pyx
+++ b/pandas/_libs/tslibs/offsets.pyx
@@ -63,6 +63,7 @@ from pandas._libs.tslibs.nattype cimport (
 )
 from pandas._libs.tslibs.np_datetime cimport (
     NPY_DATETIMEUNIT,
+    check_dts_bounds,
     get_unit_from_dtype,
     import_pandas_datetime,
     npy_datetimestruct,
@@ -3093,6 +3094,8 @@ cdef class SemiMonthOffset(SingleConstructorOffset):
                     days_in_month = get_days_in_month(dts.year, dts.month)
                     dts.day = min(to_day, days_in_month)
 
+                    with gil:
+                        check_dts_bounds(&dts, reso)
                     res_val = npy_datetimestruct_to_datetime(reso, &dts)
 
                 # Analogous to: out[i] = res_val
@@ -4762,6 +4765,8 @@ cdef ndarray shift_quarters(
                 dts.month = month_add_months(dts, modby * n - months_since)
                 dts.day = get_day_of_month(&dts, day_opt)
 
+                with gil:
+                    check_dts_bounds(&dts, reso)
                 res_val = npy_datetimestruct_to_datetime(reso, &dts)
 
             # Analogous to: out[i] = res_val
@@ -4820,6 +4825,8 @@ def shift_months(
                     dts.month = month_add_months(dts, months)
 
                     dts.day = min(dts.day, get_days_in_month(dts.year, dts.month))
+                    with gil:
+                        check_dts_bounds(&dts, reso)
                     res_val = npy_datetimestruct_to_datetime(reso, &dts)
 
                 # Analogous to: out[i] = res_val
@@ -4846,6 +4853,8 @@ def shift_months(
                     dts.month = month_add_months(dts, months_to_roll)
                     dts.day = get_day_of_month(&dts, day_opt)
 
+                    with gil:
+                        check_dts_bounds(&dts, reso)
                     res_val = npy_datetimestruct_to_datetime(reso, &dts)
 
                 # Analogous to: out[i] = res_val
diff --git a/pandas/_libs/tslibs/period.pyx b/pandas/_libs/tslibs/period.pyx
index c37e9cd7ef1f3..3081c54ed49c2 100644
--- a/pandas/_libs/tslibs/period.pyx
+++ b/pandas/_libs/tslibs/period.pyx
@@ -721,6 +721,8 @@ cdef int64_t unix_date_from_ymd(int year, int month, int day) noexcept nogil:
         npy_datetimestruct dts
         int64_t unix_date
 
+    with gil:
+        check_dts_bounds(&dts, NPY_FR_D)
     memset(&dts, 0, sizeof(npy_datetimestruct))
     dts.year = year
     dts.month = month
@@ -738,6 +740,8 @@ cdef int64_t dts_to_year_ordinal(npy_datetimestruct *dts, int to_end) noexcept n
     cdef:
         int64_t result
 
+    with gil:
+        check_dts_bounds(dts, NPY_DATETIMEUNIT.NPY_FR_Y)
     result = npy_datetimestruct_to_datetime(NPY_DATETIMEUNIT.NPY_FR_Y, dts)
     if dts.month > to_end:
         return result + 1
@@ -795,14 +799,20 @@ cdef int64_t get_period_ordinal(npy_datetimestruct *dts, int freq) noexcept nogi
         return dts_to_qtr_ordinal(dts, fmonth)
 
     elif freq_group == FR_WK:
+        with gil:
+            check_dts_bounds(dts, NPY_FR_D)
         unix_date = npy_datetimestruct_to_datetime(NPY_FR_D, dts)
         return unix_date_to_week(unix_date, freq - FR_WK)
 
     elif freq == FR_BUS:
+        with gil:
+            check_dts_bounds(dts, NPY_FR_D)
         unix_date = npy_datetimestruct_to_datetime(NPY_FR_D, dts)
         return DtoB(dts, 0, unix_date)
 
     unit = freq_group_code_to_npy_unit(freq)
+    with gil:
+        check_dts_bounds(dts, unit)
     return npy_datetimestruct_to_datetime(unit, dts)
 
 
diff --git a/pandas/_libs/tslibs/strptime.pyx b/pandas/_libs/tslibs/strptime.pyx
index e6924a4e24dff..4dc1894f759d5 100644
--- a/pandas/_libs/tslibs/strptime.pyx
+++ b/pandas/_libs/tslibs/strptime.pyx
@@ -325,6 +325,7 @@ def array_strptime(
             if string_to_dts_succeeded:
                 # No error reported by string_to_dts, pick back up
                 # where we left off
+                check_dts_bounds(&dts, NPY_FR_ns)
                 value = npy_datetimestruct_to_datetime(NPY_FR_ns, &dts)
                 if out_local == 1:
                     # Store the out_tzoffset in seconds
@@ -513,6 +514,7 @@ def array_strptime(
             dts.us = us
             dts.ps = ns * 1000
 
+            check_dts_bounds(&dts, NPY_FR_ns)
             iresult[i] = npy_datetimestruct_to_datetime(NPY_FR_ns, &dts)
             check_dts_bounds(&dts)
 
diff --git a/pandas/_libs/tslibs/timestamps.pyx b/pandas/_libs/tslibs/timestamps.pyx
index 944a2b0e97382..ec1e845b6f4ec 100644
--- a/pandas/_libs/tslibs/timestamps.pyx
+++ b/pandas/_libs/tslibs/timestamps.pyx
@@ -581,6 +581,7 @@ cdef class _Timestamp(ABCTimestamp):
 
         if own_tz is not None and not is_utc(own_tz):
             pydatetime_to_dtstruct(self, &dts)
+            check_dts_bounds(&dts, self._creso)
             val = npy_datetimestruct_to_datetime(self._creso, &dts) + self.nanosecond
         else:
             val = self._value

From 466056d1e2c348d45b7e3edd30a5640309bfe1ce Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 15 Sep 2023 08:01:05 -0400
Subject: [PATCH 11/43] try smaller config

---
 .github/workflows/unit-tests.yml | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 907a549e7b191..f74aaef8c0729 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -77,6 +77,11 @@ jobs:
             env_file: actions-311-numpydev.yaml
             pattern: "not slow and not network and not single_cpu"
             test_args: "-W error::DeprecationWarning -W error::FutureWarning"
+          - name: "ASAN/UBSAN"
+            env_file: actions-311-numpydev.yaml
+            pattern: "not slow and not network and not single_cpu"
+            test_args: "-W error::DeprecationWarning -W error::FutureWarning"
+            sanitize: true
           - name: "Pyarrow Nightly"
             env_file: actions-311-pyarrownightly.yaml
             pattern: "not slow and not network and not single_cpu"
@@ -90,6 +95,7 @@ jobs:
       PANDAS_COPY_ON_WRITE: ${{ matrix.pandas_copy_on_write || '0' }}
       PANDAS_CI: ${{ matrix.pandas_ci || '1' }}
       TEST_ARGS: ${{ matrix.test_args || '' }}
+      SANITIZE: ${{ matrix.sanitize || false }}
       PYTEST_WORKERS: 'auto'
       PYTEST_TARGET: ${{ matrix.pytest_target || 'pandas' }}
     concurrency:
@@ -158,13 +164,13 @@ jobs:
       id: build
       uses: ./.github/actions/build_pandas
       with:
-        sanitize: true
+        sanitize: ${{ env.SANITIZE }}
 
     - name: Test (not single_cpu)
       uses: ./.github/actions/run-tests
       if: ${{ matrix.name != 'Pypy' }}
       with:
-        sanitize: true
+        sanitize: ${{ env.SANITIZE }}
       env:
         # Set pattern to not single_cpu if not already set
         PATTERN: ${{ env.PATTERN == '' && 'not single_cpu' || matrix.pattern }}

From d4074ca90ce1f0ea1b316dd9f6e2e243edae269e Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 15 Sep 2023 15:06:54 -0400
Subject: [PATCH 12/43] checkpoint

---
 .../vendored/numpy/datetime/np_datetime.h     |  3 +-
 .../src/vendored/numpy/datetime/np_datetime.c | 59 ++++++++++++++++---
 2 files changed, 54 insertions(+), 8 deletions(-)

diff --git a/pandas/_libs/include/pandas/vendored/numpy/datetime/np_datetime.h b/pandas/_libs/include/pandas/vendored/numpy/datetime/np_datetime.h
index 6b5135f559482..83138e0dc679e 100644
--- a/pandas/_libs/include/pandas/vendored/numpy/datetime/np_datetime.h
+++ b/pandas/_libs/include/pandas/vendored/numpy/datetime/np_datetime.h
@@ -67,7 +67,8 @@ static const npy_datetimestruct _M_MAX_DTS = {
 PyObject *extract_utc_offset(PyObject *obj);
 
 npy_datetime npy_datetimestruct_to_datetime(NPY_DATETIMEUNIT base,
-                                            const npy_datetimestruct *dts);
+                                            const npy_datetimestruct *dts,
+                                            bool *overflow);
 
 void pandas_datetime_to_datetimestruct(npy_datetime val, NPY_DATETIMEUNIT fr,
                                        npy_datetimestruct *result);
diff --git a/pandas/_libs/src/vendored/numpy/datetime/np_datetime.c b/pandas/_libs/src/vendored/numpy/datetime/np_datetime.c
index 49016f79de5b9..e2101557faa93 100644
--- a/pandas/_libs/src/vendored/numpy/datetime/np_datetime.c
+++ b/pandas/_libs/src/vendored/numpy/datetime/np_datetime.c
@@ -22,6 +22,7 @@ This file is derived from NumPy 1.7. See NUMPY_LICENSE.txt
 #define NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION
 #endif  // NPY_NO_DEPRECATED_API
 
+#include <stdbool.h>
 #include <Python.h>
 
 #include <numpy/arrayobject.h>
@@ -304,17 +305,42 @@ PyObject *extract_utc_offset(PyObject *obj) {
 /*
  * Converts a datetime from a datetimestruct to a datetime based
  * on a metadata unit. The date is assumed to be valid.
+ *
+ * In case of an overflow, the library returns -1 and sets the
+ * overflow argument to true
  */
 npy_datetime npy_datetimestruct_to_datetime(NPY_DATETIMEUNIT base,
-                                            const npy_datetimestruct *dts) {
+                                            const npy_datetimestruct *dts,
+                                            bool *overflow) {
     npy_datetime ret;
 
     if (base == NPY_FR_Y) {
         /* Truncate to the year */
+      if (dts->year < INT64_MIN + 1970) {
+        *overflow = true;
+        return -1;
+      }
         ret = dts->year - 1970;
     } else if (base == NPY_FR_M) {
-        /* Truncate to the month */
-        ret = 12 * (dts->year - 1970) + (dts->month - 1);
+      /* Truncate to the month */
+      if (dts->year < INT64_MIN + 1970) {
+        *overflow = true;
+        return -1;
+      }
+      const npy_int64 years = dts->year - 1970;
+
+      if ((years > INT64_MAX / 12) || (years < INT64_MIN / 12)) {
+        *overflow = true;
+        return -1;
+      }
+      const npy_int64 months = years * 12;
+
+      if (dts->month == INT32_MIN) {
+        *overflow = true;
+        return -1;
+      }
+
+      ret = months + dts->month - 1;
     } else {
         /* Otherwise calculate the number of days to start */
         npy_int64 days = get_datetimestruct_days(dts);
@@ -325,15 +351,34 @@ npy_datetime npy_datetimestruct_to_datetime(NPY_DATETIMEUNIT base,
                 if (days >= 0) {
                     ret = days / 7;
                 } else {
-                    ret = (days - 6) / 7;
+                  if (days < INT64_MIN + 6) {
+                    *overflow = true;
+                    return -1;
+                  }
+                  ret = (days - 6) / 7;
                 }
                 break;
             case NPY_FR_D:
                 ret = days;
                 break;
-            case NPY_FR_h:
-                ret = days * 24 + dts->hour;
-                break;
+        case NPY_FR_h: {
+          if ((days > INT64_MAX / 24) || (days < INT64_MIN / 24)) {
+            *overflow = true;
+            return -1;
+          }
+          npy_int64 hours = days * 24;
+
+          if (
+              ((dts->hour > 0) && (hours > INT64_MAX - dts->hour))
+              || ((dts->hour < 0) && (hours < INT64_MIN - dts->hour))
+              ) {
+            *overflow = true;
+            return -1;
+          }
+
+          ret = hours + dts->hour;
+          break;
+        }
             case NPY_FR_m:
                 ret = (days * 24 + dts->hour) * 60 + dts->min;
                 break;

From e303ba11fde1642f01077d6bd6efed342ac40db2 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 27 Oct 2023 08:42:44 -0400
Subject: [PATCH 13/43] bool fixup

---
 .../_libs/include/pandas/vendored/numpy/datetime/np_datetime.h | 3 +--
 pandas/_libs/src/vendored/numpy/datetime/np_datetime.c         | 1 -
 2 files changed, 1 insertion(+), 3 deletions(-)

diff --git a/pandas/_libs/include/pandas/vendored/numpy/datetime/np_datetime.h b/pandas/_libs/include/pandas/vendored/numpy/datetime/np_datetime.h
index ca05ff5c11b11..e4e90a7ea24cf 100644
--- a/pandas/_libs/include/pandas/vendored/numpy/datetime/np_datetime.h
+++ b/pandas/_libs/include/pandas/vendored/numpy/datetime/np_datetime.h
@@ -67,8 +67,7 @@ static const npy_datetimestruct _M_MAX_DTS = {
 PyObject *extract_utc_offset(PyObject *obj);
 
 npy_datetime npy_datetimestruct_to_datetime(NPY_DATETIMEUNIT base,
-                                            const npy_datetimestruct *dts,
-                                            bool *overflow);
+                                            const npy_datetimestruct *dts);
 
 void pandas_datetime_to_datetimestruct(npy_datetime val, NPY_DATETIMEUNIT fr,
                                        npy_datetimestruct *result);
diff --git a/pandas/_libs/src/vendored/numpy/datetime/np_datetime.c b/pandas/_libs/src/vendored/numpy/datetime/np_datetime.c
index a0da03268604c..e3ad6fab5bdd8 100644
--- a/pandas/_libs/src/vendored/numpy/datetime/np_datetime.c
+++ b/pandas/_libs/src/vendored/numpy/datetime/np_datetime.c
@@ -23,7 +23,6 @@ This file is derived from NumPy 1.7. See NUMPY_LICENSE.txt
 #endif // NPY_NO_DEPRECATED_API
 
 #include <Python.h>
-#include <stdbool.h>
 
 #include "pandas/vendored/numpy/datetime/np_datetime.h"
 #include <numpy/arrayobject.h>

From 46d1034c46ef66ab79b38740f768870eecddc310 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Wed, 15 Nov 2023 21:54:35 -0500
Subject: [PATCH 14/43] reverts

---
 pandas/_libs/tslibs/offsets.pyx    |  9 ---------
 pandas/_libs/tslibs/period.pyx     | 10 ----------
 pandas/_libs/tslibs/timestamps.pyx |  1 -
 3 files changed, 20 deletions(-)

diff --git a/pandas/_libs/tslibs/offsets.pyx b/pandas/_libs/tslibs/offsets.pyx
index ba63947e01306..7f3a72178a359 100644
--- a/pandas/_libs/tslibs/offsets.pyx
+++ b/pandas/_libs/tslibs/offsets.pyx
@@ -68,7 +68,6 @@ from pandas._libs.tslibs.nattype cimport (
 )
 from pandas._libs.tslibs.np_datetime cimport (
     NPY_DATETIMEUNIT,
-    check_dts_bounds,
     get_unit_from_dtype,
     import_pandas_datetime,
     npy_datetimestruct,
@@ -3119,8 +3118,6 @@ cdef class SemiMonthOffset(SingleConstructorOffset):
                     days_in_month = get_days_in_month(dts.year, dts.month)
                     dts.day = min(to_day, days_in_month)
 
-                    with gil:
-                        check_dts_bounds(&dts, reso)
                     res_val = npy_datetimestruct_to_datetime(reso, &dts)
 
                 # Analogous to: out[i] = res_val
@@ -4939,8 +4936,6 @@ cdef ndarray shift_quarters(
                 dts.month = month_add_months(dts, modby * n - months_since)
                 dts.day = get_day_of_month(&dts, day_opt)
 
-                with gil:
-                    check_dts_bounds(&dts, reso)
                 res_val = npy_datetimestruct_to_datetime(reso, &dts)
 
             # Analogous to: out[i] = res_val
@@ -4999,8 +4994,6 @@ def shift_months(
                     dts.month = month_add_months(dts, months)
 
                     dts.day = min(dts.day, get_days_in_month(dts.year, dts.month))
-                    with gil:
-                        check_dts_bounds(&dts, reso)
                     res_val = npy_datetimestruct_to_datetime(reso, &dts)
 
                 # Analogous to: out[i] = res_val
@@ -5027,8 +5020,6 @@ def shift_months(
                     dts.month = month_add_months(dts, months_to_roll)
                     dts.day = get_day_of_month(&dts, day_opt)
 
-                    with gil:
-                        check_dts_bounds(&dts, reso)
                     res_val = npy_datetimestruct_to_datetime(reso, &dts)
 
                 # Analogous to: out[i] = res_val
diff --git a/pandas/_libs/tslibs/period.pyx b/pandas/_libs/tslibs/period.pyx
index 6e10d5c49c882..318e018689a78 100644
--- a/pandas/_libs/tslibs/period.pyx
+++ b/pandas/_libs/tslibs/period.pyx
@@ -730,8 +730,6 @@ cdef int64_t unix_date_from_ymd(int year, int month, int day) noexcept nogil:
         npy_datetimestruct dts
         int64_t unix_date
 
-    with gil:
-        check_dts_bounds(&dts, NPY_FR_D)
     memset(&dts, 0, sizeof(npy_datetimestruct))
     dts.year = year
     dts.month = month
@@ -749,8 +747,6 @@ cdef int64_t dts_to_year_ordinal(npy_datetimestruct *dts, int to_end) noexcept n
     cdef:
         int64_t result
 
-    with gil:
-        check_dts_bounds(dts, NPY_DATETIMEUNIT.NPY_FR_Y)
     result = npy_datetimestruct_to_datetime(NPY_DATETIMEUNIT.NPY_FR_Y, dts)
     if dts.month > to_end:
         return result + 1
@@ -808,20 +804,14 @@ cdef int64_t get_period_ordinal(npy_datetimestruct *dts, int freq) noexcept nogi
         return dts_to_qtr_ordinal(dts, fmonth)
 
     elif freq_group == FR_WK:
-        with gil:
-            check_dts_bounds(dts, NPY_FR_D)
         unix_date = npy_datetimestruct_to_datetime(NPY_FR_D, dts)
         return unix_date_to_week(unix_date, freq - FR_WK)
 
     elif freq == FR_BUS:
-        with gil:
-            check_dts_bounds(dts, NPY_FR_D)
         unix_date = npy_datetimestruct_to_datetime(NPY_FR_D, dts)
         return DtoB(dts, 0, unix_date)
 
     unit = freq_group_code_to_npy_unit(freq)
-    with gil:
-        check_dts_bounds(dts, unit)
     return npy_datetimestruct_to_datetime(unit, dts)
 
 
diff --git a/pandas/_libs/tslibs/timestamps.pyx b/pandas/_libs/tslibs/timestamps.pyx
index c67141fcaf2ab..56a6885d4a9e0 100644
--- a/pandas/_libs/tslibs/timestamps.pyx
+++ b/pandas/_libs/tslibs/timestamps.pyx
@@ -580,7 +580,6 @@ cdef class _Timestamp(ABCTimestamp):
 
         if own_tz is not None and not is_utc(own_tz):
             pydatetime_to_dtstruct(self, &dts)
-            check_dts_bounds(&dts, self._creso)
             val = npy_datetimestruct_to_datetime(self._creso, &dts) + self.nanosecond
         else:
             val = self._value

From 929c731d9698adce401e6554b20052a247c9255f Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 17 Nov 2023 15:30:35 -0800
Subject: [PATCH 15/43] known UB marker

---
 .github/workflows/unit-tests.yml            |  2 +-
 pandas/tests/io/parser/common/test_float.py | 10 +++++++++-
 pyproject.toml                              |  1 +
 scripts/tests/data/deps_minimum.toml        |  1 +
 4 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 0ade3b3fd6d34..76b72f74d7090 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -83,7 +83,7 @@ jobs:
             test_args: "-W error::DeprecationWarning -W error::FutureWarning"
           - name: "ASAN/UBSAN"
             env_file: actions-311-numpydev.yaml
-            pattern: "not slow and not network and not single_cpu"
+            pattern: "not slow and not network and not single_cpu and not known_ub"
             test_args: "-W error::DeprecationWarning -W error::FutureWarning"
             sanitize: true
           - name: "Pyarrow Nightly"
diff --git a/pandas/tests/io/parser/common/test_float.py b/pandas/tests/io/parser/common/test_float.py
index 4b23774ee2d5b..6ed32427bebf9 100644
--- a/pandas/tests/io/parser/common/test_float.py
+++ b/pandas/tests/io/parser/common/test_float.py
@@ -40,7 +40,14 @@ def test_scientific_no_exponent(all_parsers_all_precisions):
     tm.assert_frame_equal(df_roundtrip, df)
 
 
-@pytest.mark.parametrize("neg_exp", [-617, -100000, -99999999999999999])
+@pytest.mark.parametrize(
+    "neg_exp",
+    [
+        -617,
+        -100000,
+        pytest.param(-99999999999999999, marks=pytest.mark.known_ub),
+    ],
+)
 def test_very_negative_exponent(all_parsers_all_precisions, neg_exp):
     # GH#38753
     parser, precision = all_parsers_all_precisions
@@ -51,6 +58,7 @@ def test_very_negative_exponent(all_parsers_all_precisions, neg_exp):
     tm.assert_frame_equal(result, expected)
 
 
+@pytest.mark.known_ub
 @xfail_pyarrow  # AssertionError: Attributes of DataFrame.iloc[:, 0] are different
 @pytest.mark.parametrize("exp", [999999999999999999, -999999999999999999])
 def test_too_many_exponent_digits(all_parsers_all_precisions, exp, request):
diff --git a/pyproject.toml b/pyproject.toml
index 8ebd70762b2a5..d191af219ed8f 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -518,6 +518,7 @@ markers = [
   "clipboard: mark a pd.read_clipboard test",
   "arm_slow: mark a test as slow for arm64 architecture",
   "arraymanager: mark a test to run with ArrayManager enabled",
+  "known_ub: tests known to invoke undefined behavior",
 ]
 
 [tool.mypy]
diff --git a/scripts/tests/data/deps_minimum.toml b/scripts/tests/data/deps_minimum.toml
index c74ad3d17a4a9..546dc99fc10d4 100644
--- a/scripts/tests/data/deps_minimum.toml
+++ b/scripts/tests/data/deps_minimum.toml
@@ -383,6 +383,7 @@ markers = [
   "clipboard: mark a pd.read_clipboard test",
   "arm_slow: mark a test as slow for arm64 architecture",
   "arraymanager: mark a test to run with ArrayManager enabled",
+  "known_ub: tests that trigger known undefined behavior",
 ]
 
 [tool.mypy]

From 6483e07a1aa2545dd189cc59ac6bd438b833f629 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 1 Dec 2023 16:16:26 -0800
Subject: [PATCH 16/43] Finished marking tests with known UB

---
 pandas/tests/groupby/test_apply.py                        | 1 +
 pandas/tests/groupby/test_cumulative.py                   | 1 +
 pandas/tests/scalar/timedelta/methods/test_round.py       | 2 ++
 pandas/tests/scalar/timedelta/test_arithmetic.py          | 1 +
 pandas/tests/scalar/timedelta/test_timedelta.py           | 1 +
 pandas/tests/scalar/timestamp/methods/test_tz_localize.py | 1 +
 pandas/tests/scalar/timestamp/test_constructors.py        | 1 +
 pandas/tests/tools/test_to_datetime.py                    | 1 +
 8 files changed, 9 insertions(+)

diff --git a/pandas/tests/groupby/test_apply.py b/pandas/tests/groupby/test_apply.py
index 60b386adb664a..827d8a52dd977 100644
--- a/pandas/tests/groupby/test_apply.py
+++ b/pandas/tests/groupby/test_apply.py
@@ -1391,6 +1391,7 @@ def test_groupby_apply_to_series_name():
     tm.assert_series_equal(result, expected)
 
 
+@pytest.mark.known_ub
 @pytest.mark.parametrize("dropna", [True, False])
 def test_apply_na(dropna):
     # GH#28984
diff --git a/pandas/tests/groupby/test_cumulative.py b/pandas/tests/groupby/test_cumulative.py
index bf572609f3d37..ab5aeee18ba7b 100644
--- a/pandas/tests/groupby/test_cumulative.py
+++ b/pandas/tests/groupby/test_cumulative.py
@@ -60,6 +60,7 @@ def test_groupby_cumprod():
     tm.assert_series_equal(actual, expected)
 
 
+@pytest.mark.known_ub
 def test_groupby_cumprod_overflow():
     # GH#37493 if we overflow we return garbage consistent with numpy
     df = DataFrame({"key": ["b"] * 4, "value": 100_000})
diff --git a/pandas/tests/scalar/timedelta/methods/test_round.py b/pandas/tests/scalar/timedelta/methods/test_round.py
index 4beb39510413c..524a0b5736cee 100644
--- a/pandas/tests/scalar/timedelta/methods/test_round.py
+++ b/pandas/tests/scalar/timedelta/methods/test_round.py
@@ -61,6 +61,7 @@ def test_round_invalid(self):
             with pytest.raises(ValueError, match=msg):
                 t1.round(freq)
 
+    @pytest.mark.known_ub
     def test_round_implementation_bounds(self):
         # See also: analogous test for Timestamp
         # GH#38964
@@ -86,6 +87,7 @@ def test_round_implementation_bounds(self):
         with pytest.raises(OutOfBoundsTimedelta, match=msg):
             Timedelta.max.round("s")
 
+    @pytest.mark.known_ub
     @given(val=st.integers(min_value=iNaT + 1, max_value=lib.i8max))
     @pytest.mark.parametrize(
         "method", [Timedelta.round, Timedelta.floor, Timedelta.ceil]
diff --git a/pandas/tests/scalar/timedelta/test_arithmetic.py b/pandas/tests/scalar/timedelta/test_arithmetic.py
index 363ae1fa9c644..3e578db6723a7 100644
--- a/pandas/tests/scalar/timedelta/test_arithmetic.py
+++ b/pandas/tests/scalar/timedelta/test_arithmetic.py
@@ -966,6 +966,7 @@ def test_td_op_timedelta_timedeltalike_array(self, op, arr):
 
 
 class TestTimedeltaComparison:
+    @pytest.mark.known_ub
     def test_compare_pytimedelta_bounds(self):
         # GH#49021 don't overflow on comparison with very large pytimedeltas
 
diff --git a/pandas/tests/scalar/timedelta/test_timedelta.py b/pandas/tests/scalar/timedelta/test_timedelta.py
index ac605df935226..94b5a7821c20c 100644
--- a/pandas/tests/scalar/timedelta/test_timedelta.py
+++ b/pandas/tests/scalar/timedelta/test_timedelta.py
@@ -551,6 +551,7 @@ def test_timedelta_hash_equality(self):
         ns_td = Timedelta(1, "ns")
         assert hash(ns_td) != hash(ns_td.to_pytimedelta())
 
+    @pytest.mark.known_ub
     @pytest.mark.xfail(
         reason="pd.Timedelta violates the Python hash invariant (GH#44504).",
     )
diff --git a/pandas/tests/scalar/timestamp/methods/test_tz_localize.py b/pandas/tests/scalar/timestamp/methods/test_tz_localize.py
index 9df0a023730de..8169aa1035cf4 100644
--- a/pandas/tests/scalar/timestamp/methods/test_tz_localize.py
+++ b/pandas/tests/scalar/timestamp/methods/test_tz_localize.py
@@ -25,6 +25,7 @@
 
 
 class TestTimestampTZLocalize:
+    @pytest.mark.known_ub
     def test_tz_localize_pushes_out_of_bounds(self):
         # GH#12677
         # tz_localize that pushes away from the boundary is OK
diff --git a/pandas/tests/scalar/timestamp/test_constructors.py b/pandas/tests/scalar/timestamp/test_constructors.py
index 91314a497b1fb..c5f1694802930 100644
--- a/pandas/tests/scalar/timestamp/test_constructors.py
+++ b/pandas/tests/scalar/timestamp/test_constructors.py
@@ -815,6 +815,7 @@ def test_barely_out_of_bounds(self):
         with pytest.raises(OutOfBoundsDatetime, match=msg):
             Timestamp("2262-04-11 23:47:16.854775808")
 
+    @pytest.mark.known_ub
     def test_bounds_with_different_units(self):
         out_of_bounds_dates = ("1677-09-21", "2262-04-12")
 
diff --git a/pandas/tests/tools/test_to_datetime.py b/pandas/tests/tools/test_to_datetime.py
index 8139fe52c7037..627e148610348 100644
--- a/pandas/tests/tools/test_to_datetime.py
+++ b/pandas/tests/tools/test_to_datetime.py
@@ -1140,6 +1140,7 @@ def test_to_datetime_dt64s_out_of_ns_bounds(self, cache, dt, errors):
         assert ts.unit == "s"
         assert ts.asm8 == dt
 
+    @pytest.mark.known_ub
     def test_to_datetime_dt64d_out_of_bounds(self, cache):
         dt64 = np.datetime64(np.iinfo(np.int64).max, "D")
 

From b87a2102397ba3763703009382e63c111995d756 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 1 Dec 2023 21:14:30 -0800
Subject: [PATCH 17/43] dedicated CI job

---
 .github/workflows/unit-tests.yml | 47 ++++++++++++++++++++++++--------
 1 file changed, 35 insertions(+), 12 deletions(-)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 9a871ffdb5d12..9d2a97df7114c 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -93,11 +93,6 @@ jobs:
             env_file: actions-311-numpydev.yaml
             pattern: "not slow and not network and not single_cpu"
             test_args: "-W error::DeprecationWarning -W error::FutureWarning"
-          - name: "ASAN/UBSAN"
-            env_file: actions-311-numpydev.yaml
-            pattern: "not slow and not network and not single_cpu and not known_ub"
-            test_args: "-W error::DeprecationWarning -W error::FutureWarning"
-            sanitize: true
           - name: "Pyarrow Nightly"
             env_file: actions-311-pyarrownightly.yaml
             pattern: "not slow and not network and not single_cpu"
@@ -110,7 +105,6 @@ jobs:
       PANDAS_COPY_ON_WRITE: ${{ matrix.pandas_copy_on_write || '0' }}
       PANDAS_CI: ${{ matrix.pandas_ci || '1' }}
       TEST_ARGS: ${{ matrix.test_args || '' }}
-      SANITIZE: ${{ matrix.sanitize || false }}
       PYTEST_WORKERS: 'auto'
       PYTEST_TARGET: ${{ matrix.pytest_target || 'pandas' }}
       # Clipboard tests
@@ -180,22 +174,16 @@ jobs:
     - name: Build Pandas
       id: build
       uses: ./.github/actions/build_pandas
-      with:
-        sanitize: ${{ env.SANITIZE }}
 
     - name: Test (not single_cpu)
       uses: ./.github/actions/run-tests
       if: ${{ matrix.name != 'Pypy' }}
-      with:
-        sanitize: ${{ env.SANITIZE }}
       env:
         # Set pattern to not single_cpu if not already set
         PATTERN: ${{ env.PATTERN == '' && 'not single_cpu' || matrix.pattern }}
 
     - name: Test (single_cpu)
       uses: ./.github/actions/run-tests
-      with:
-        sanitize: true
       env:
         PATTERN: 'single_cpu'
         PYTEST_WORKERS: 0
@@ -317,6 +305,41 @@ jobs:
       group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-musl
       cancel-in-progress: true
 
+  ASAN/UBSAN:
+    runs-on: ubuntu-22.04
+    timeout-minutes: 90
+    steps:
+    - name: Checkout
+      uses: actions/checkout@v4
+      with:
+        fetch-depth: 0
+
+    - name: Set up Conda
+      uses: ./.github/actions/setup-conda
+      with:
+        environment-file: ci/deps/actions-311-numpydev.yaml
+
+    - name: Build Pandas
+      id: build
+      uses: ./.github/actions/build_pandas
+      with:
+        sanitize: true
+
+    - name: Test (not single_cpu)
+      uses: ./.github/actions/run-tests
+      with:
+        sanitize: true
+      env:
+        PATTERN: "not slow and not network and not single_cpu and not known_ub"
+
+    - name: Test (single_cpu)
+      uses: ./.github/actions/run-tests
+      with:
+        sanitize: true
+      env:
+        PATTERN: 'single_cpu'
+        PYTEST_WORKERS: 0
+
   python-dev:
     # This job may or may not run depending on the state of the next
     # unreleased Python version. DO NOT DELETE IT.

From 46ec0237896fb1a074ed438985b216ce5c5d95b6 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 1 Dec 2023 21:26:21 -0800
Subject: [PATCH 18/43] identifier fix

---
 .github/workflows/unit-tests.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 9d2a97df7114c..91780c80e5997 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -305,7 +305,7 @@ jobs:
       group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-musl
       cancel-in-progress: true
 
-  ASAN/UBSAN:
+  ASAN_UBSAN:
     runs-on: ubuntu-22.04
     timeout-minutes: 90
     steps:

From 8695dca20815a21b9db00d255bac7d07655e3e9f Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 1 Dec 2023 21:47:57 -0800
Subject: [PATCH 19/43] fixes

---
 .github/workflows/unit-tests.yml | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 91780c80e5997..65cb0f1389a67 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -331,14 +331,17 @@ jobs:
         sanitize: true
       env:
         PATTERN: "not slow and not network and not single_cpu and not known_ub"
+        PYTEST_WORKERS: 'auto'
+        PYTEST_TARGET: 'pandas'
 
     - name: Test (single_cpu)
       uses: ./.github/actions/run-tests
       with:
         sanitize: true
       env:
-        PATTERN: 'single_cpu'
+        PATTERN: 'single_cpu and not known_ub'
         PYTEST_WORKERS: 0
+        PYTEST_TARGET: 'pandas'
 
   python-dev:
     # This job may or may not run depending on the state of the next

From 05319ae53d5822fca6c1f0edc9b316c58c0f38f2 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 1 Dec 2023 22:28:28 -0800
Subject: [PATCH 20/43] more test skip

---
 pandas/tests/frame/test_constructors.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/pandas/tests/frame/test_constructors.py b/pandas/tests/frame/test_constructors.py
index 84189f8149d81..f13a492305659 100644
--- a/pandas/tests/frame/test_constructors.py
+++ b/pandas/tests/frame/test_constructors.py
@@ -3197,6 +3197,7 @@ def test_from_out_of_bounds_ns_datetime(
         assert item.asm8.dtype == exp_dtype
         assert dtype == exp_dtype
 
+    @pytest.mark.known_ub
     def test_out_of_s_bounds_datetime64(self, constructor):
         scalar = np.datetime64(np.iinfo(np.int64).max, "D")
         result = constructor(scalar)
@@ -3232,6 +3233,7 @@ def test_from_out_of_bounds_ns_timedelta(
         assert item.asm8.dtype == exp_dtype
         assert dtype == exp_dtype
 
+    @pytest.mark.known_ub
     @pytest.mark.parametrize("cls", [np.datetime64, np.timedelta64])
     def test_out_of_s_bounds_timedelta64(self, constructor, cls):
         scalar = cls(np.iinfo(np.int64).max, "D")

From 6d76a5733897a071aec5b36c7d2901f98d4a1c51 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 1 Dec 2023 23:24:35 -0800
Subject: [PATCH 21/43] try quotes

---
 .github/workflows/unit-tests.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 65cb0f1389a67..5711096b55034 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -339,7 +339,7 @@ jobs:
       with:
         sanitize: true
       env:
-        PATTERN: 'single_cpu and not known_ub'
+        PATTERN: "single_cpu and not known_ub"
         PYTEST_WORKERS: 0
         PYTEST_TARGET: 'pandas'
 

From f5dd440058883c39f289836c314a7cce00d5a16a Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 1 Dec 2023 23:33:43 -0800
Subject: [PATCH 22/43] simplify ci

---
 .github/actions/build_pandas/action.yml | 22 +++++-----------------
 .github/actions/run-tests/action.yml    | 13 +------------
 .github/workflows/unit-tests.yml        | 12 +++++++-----
 3 files changed, 13 insertions(+), 34 deletions(-)

diff --git a/.github/actions/build_pandas/action.yml b/.github/actions/build_pandas/action.yml
index 273a31bf36e03..db3494db4d35b 100644
--- a/.github/actions/build_pandas/action.yml
+++ b/.github/actions/build_pandas/action.yml
@@ -4,9 +4,9 @@ inputs:
   editable:
     description: Whether to build pandas in editable mode (default true)
     default: true
-  sanitize:
-    description: Whether sanitizers should be used or not
-    default: false
+  meson_args:
+    description: Extra flags to pass to meson
+    required: false
 runs:
   using: composite
   steps:
@@ -28,20 +28,8 @@ runs:
     - name: Build Pandas
       run: |
         if [[ ${{ inputs.editable }} == "true" ]]; then
-          if [[ ${{ inputs.sanitize }} == "true" ]]; then
-            CFLAGS="$CFLAGS -fno-sanitize-recover=all" \
-              pip install -e . --no-build-isolation -v  \
-              --config-settings=setup-args="-Db_sanitize=address,undefined" --no-deps
-          else
-            pip install -e . --no-build-isolation -v  --no-deps
-          fi
+          pip install -e . --no-build-isolation -v --no-deps ${{ inputs.meson_args }}
         else
-          if [[ ${{ inputs.sanitize }} == "true" ]]; then
-            CFLAGS="$CFLAGS -fno-sanitize-recover=all" \
-              pip install . --no-build-isolation -v \
-              --config-settings=setup-args="-Db_sanitize=address,undefined" --no-deps
-          else
-            pip install . --no-build-isolation -v  --no-deps
-          fi
+          pip install . --no-build-isolation -v --no-deps ${{ inputs.meson_args }}
         fi
       shell: bash -el {0}
diff --git a/.github/actions/run-tests/action.yml b/.github/actions/run-tests/action.yml
index b66adf18a4847..fd7c3587f2254 100644
--- a/.github/actions/run-tests/action.yml
+++ b/.github/actions/run-tests/action.yml
@@ -1,20 +1,9 @@
 name: Run tests and report results
-inputs:
-  sanitize:
-    description: Whether sanitizers should be used or not
-    default: false
 runs:
   using: composite
   steps:
     - name: Test
-      run: |
-        if [[ ${{ inputs.sanitize }} == "true" ]]; then
-          ASAN_OPTIONS=detect_leaks=0 \
-            LD_PRELOAD=$(gcc -print-file-name=libasan.so) \
-            ci/run_tests.sh
-        else
-          ci/run_tests.sh
-        fi
+      run: ci/run_tests.sh
       shell: bash -el {0}
 
     - name: Publish test results
diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 5711096b55034..cf558fed7f2b4 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -323,25 +323,27 @@ jobs:
       id: build
       uses: ./.github/actions/build_pandas
       with:
-        sanitize: true
+        meson_args: --config-settings=setup-args="-Db_sanitize=address,undefined"
+      env:
+        CFLAGS="$CFLAGS -fno-sanitize-recover=all"
 
     - name: Test (not single_cpu)
       uses: ./.github/actions/run-tests
-      with:
-        sanitize: true
       env:
         PATTERN: "not slow and not network and not single_cpu and not known_ub"
         PYTEST_WORKERS: 'auto'
         PYTEST_TARGET: 'pandas'
+        ASAN_OPTIONS: detect_leaks=0
+        LD_PRELOAD: $(gcc -print-file-name=libasan.so)
 
     - name: Test (single_cpu)
       uses: ./.github/actions/run-tests
-      with:
-        sanitize: true
       env:
         PATTERN: "single_cpu and not known_ub"
         PYTEST_WORKERS: 0
         PYTEST_TARGET: 'pandas'
+        ASAN_OPTIONS: detect_leaks=0
+        LD_PRELOAD: $(gcc -print-file-name=libasan.so)
 
   python-dev:
     # This job may or may not run depending on the state of the next

From 12aa1d11346fdc7916b158c09681675f4f708d50 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 1 Dec 2023 23:43:41 -0800
Subject: [PATCH 23/43] try CFLAGS

---
 .github/actions/build_pandas/action.yml | 4 ++++
 .github/workflows/unit-tests.yml        | 3 +--
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/.github/actions/build_pandas/action.yml b/.github/actions/build_pandas/action.yml
index db3494db4d35b..7b3ddd79263d3 100644
--- a/.github/actions/build_pandas/action.yml
+++ b/.github/actions/build_pandas/action.yml
@@ -7,6 +7,9 @@ inputs:
   meson_args:
     description: Extra flags to pass to meson
     required: false
+  cflags_adds:
+    description: Items to append to the CFLAGS variable
+    required: false
 runs:
   using: composite
   steps:
@@ -27,6 +30,7 @@ runs:
 
     - name: Build Pandas
       run: |
+        export CFLAGS="$CFLAGS ${{ inputs.cflags_adds }}"
         if [[ ${{ inputs.editable }} == "true" ]]; then
           pip install -e . --no-build-isolation -v --no-deps ${{ inputs.meson_args }}
         else
diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index cf558fed7f2b4..2d292faefeac1 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -324,8 +324,7 @@ jobs:
       uses: ./.github/actions/build_pandas
       with:
         meson_args: --config-settings=setup-args="-Db_sanitize=address,undefined"
-      env:
-        CFLAGS="$CFLAGS -fno-sanitize-recover=all"
+        cflags_adds: -fno-sanitize-recover=all
 
     - name: Test (not single_cpu)
       uses: ./.github/actions/run-tests

From 628d1c2754773a92531f0df0268742f6a9136d16 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Sat, 2 Dec 2023 00:34:50 -0800
Subject: [PATCH 24/43] preload args

---
 .github/actions/run-tests/action.yml |  6 +++++-
 .github/workflows/unit-tests.yml     | 10 ++++++----
 2 files changed, 11 insertions(+), 5 deletions(-)

diff --git a/.github/actions/run-tests/action.yml b/.github/actions/run-tests/action.yml
index fd7c3587f2254..02a400ab23263 100644
--- a/.github/actions/run-tests/action.yml
+++ b/.github/actions/run-tests/action.yml
@@ -1,9 +1,13 @@
 name: Run tests and report results
+inputs:
+  preload:
+    description: Preload arguments for sanitizer
+    required: false
 runs:
   using: composite
   steps:
     - name: Test
-      run: ci/run_tests.sh
+      run: ${{ inputs.preload }} ci/run_tests.sh
       shell: bash -el {0}
 
     - name: Publish test results
diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 2d292faefeac1..c5197ec745b94 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -328,21 +328,23 @@ jobs:
 
     - name: Test (not single_cpu)
       uses: ./.github/actions/run-tests
+      with:
+        preload: LD_PRELOAD=$(gcc -print-file-name=libasan.so)
       env:
         PATTERN: "not slow and not network and not single_cpu and not known_ub"
         PYTEST_WORKERS: 'auto'
         PYTEST_TARGET: 'pandas'
-        ASAN_OPTIONS: detect_leaks=0
-        LD_PRELOAD: $(gcc -print-file-name=libasan.so)
+        ASAN_OPTIONS: detect_leaks=0  # leak detection in Python not yet working
 
     - name: Test (single_cpu)
       uses: ./.github/actions/run-tests
+      with:
+        preload: LD_PRELOAD=$(gcc -print-file-name=libasan.so)
       env:
         PATTERN: "single_cpu and not known_ub"
         PYTEST_WORKERS: 0
         PYTEST_TARGET: 'pandas'
-        ASAN_OPTIONS: detect_leaks=0
-        LD_PRELOAD: $(gcc -print-file-name=libasan.so)
+        ASAN_OPTIONS: detect_leaks=0 # leak detection in Python not yet working
 
   python-dev:
     # This job may or may not run depending on the state of the next

From 1de633ee28ab2e419574c99e449957d7a7aa00ba Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Sat, 2 Dec 2023 10:02:56 -0800
Subject: [PATCH 25/43] skip single_cpu tests

---
 .github/workflows/unit-tests.yml | 19 ++++++++++---------
 1 file changed, 10 insertions(+), 9 deletions(-)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index c5197ec745b94..43f3877ae0940 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -336,15 +336,16 @@ jobs:
         PYTEST_TARGET: 'pandas'
         ASAN_OPTIONS: detect_leaks=0  # leak detection in Python not yet working
 
-    - name: Test (single_cpu)
-      uses: ./.github/actions/run-tests
-      with:
-        preload: LD_PRELOAD=$(gcc -print-file-name=libasan.so)
-      env:
-        PATTERN: "single_cpu and not known_ub"
-        PYTEST_WORKERS: 0
-        PYTEST_TARGET: 'pandas'
-        ASAN_OPTIONS: detect_leaks=0 # leak detection in Python not yet working
+    # Skipping single CPU tests for now - see pytest GH issue 641 for error
+    #- name: Test (single_cpu)
+    #  uses: ./.github/actions/run-tests
+    #  with:
+    #    preload: LD_PRELOAD=$(gcc -print-file-name=libasan.so)
+    #  env:
+    #    PATTERN: "single_cpu and not known_ub"
+    #    PYTEST_WORKERS: 0
+    #    PYTEST_TARGET: 'pandas'
+    #    ASAN_OPTIONS: detect_leaks=0 # leak detection in Python not yet working
 
   python-dev:
     # This job may or may not run depending on the state of the next

From 3e295c56a4c3233fe2b9d525f3e94931d5ad3e83 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Sat, 2 Dec 2023 10:04:48 -0800
Subject: [PATCH 26/43] wording

---
 scripts/tests/data/deps_minimum.toml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/tests/data/deps_minimum.toml b/scripts/tests/data/deps_minimum.toml
index 3230359f0decf..c5eed77928271 100644
--- a/scripts/tests/data/deps_minimum.toml
+++ b/scripts/tests/data/deps_minimum.toml
@@ -383,7 +383,7 @@ markers = [
   "clipboard: mark a pd.read_clipboard test",
   "arm_slow: mark a test as slow for arm64 architecture",
   "arraymanager: mark a test to run with ArrayManager enabled",
-  "known_ub: tests that trigger known undefined behavior",
+  "known_ub: tests known to invoke undefined behavior",
 ]
 
 [tool.mypy]

From d5809b82ca74537c4b2f39e260c7ccd31d4c12d3 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Mon, 4 Dec 2023 16:36:48 -0800
Subject: [PATCH 27/43] removed unneeded marker

---
 pandas/tests/groupby/test_apply.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/pandas/tests/groupby/test_apply.py b/pandas/tests/groupby/test_apply.py
index 827d8a52dd977..60b386adb664a 100644
--- a/pandas/tests/groupby/test_apply.py
+++ b/pandas/tests/groupby/test_apply.py
@@ -1391,7 +1391,6 @@ def test_groupby_apply_to_series_name():
     tm.assert_series_equal(result, expected)
 
 
-@pytest.mark.known_ub
 @pytest.mark.parametrize("dropna", [True, False])
 def test_apply_na(dropna):
     # GH#28984

From 6266422371022ba227ae01634c40090e26635784 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Mon, 4 Dec 2023 16:43:05 -0800
Subject: [PATCH 28/43] float set implementations

---
 .../_libs/include/pandas/vendored/klib/khash_python.h  | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/pandas/_libs/include/pandas/vendored/klib/khash_python.h b/pandas/_libs/include/pandas/vendored/klib/khash_python.h
index 5a933b45d9e21..78528e27296fe 100644
--- a/pandas/_libs/include/pandas/vendored/klib/khash_python.h
+++ b/pandas/_libs/include/pandas/vendored/klib/khash_python.h
@@ -130,13 +130,23 @@ static inline khuint32_t kh_float32_hash_func(float val) {
   KHASH_INIT(name, khfloat64_t, khval_t, 1, kh_float64_hash_func,              \
              kh_floats_hash_equal)
 
+#define KHASH_SET_INIT_FLOAT64(name)                                           \
+  KHASH_INIT(name, khfloat64_t, char, 0, kh_float64_hash_func,                 \
+             kh_floats_hash_equal)
+
 KHASH_MAP_INIT_FLOAT64(float64, size_t)
+KHASH_SET_INIT_FLOAT64(float64_set)
 
 #define KHASH_MAP_INIT_FLOAT32(name, khval_t)                                  \
   KHASH_INIT(name, khfloat32_t, khval_t, 1, kh_float32_hash_func,              \
              kh_floats_hash_equal)
 
+#define KHASH_SET_INIT_FLOAT32(name)                                           \
+  KHASH_INIT(name, khfloat32_t, char, 0, kh_float32_hash_func,                 \
+             kh_floats_hash_equal)
+
 KHASH_MAP_INIT_FLOAT32(float32, size_t)
+KHASH_SET_INIT_FLOAT32(float32_set)
 
 static inline khint32_t kh_complex128_hash_func(khcomplex128_t val) {
   return kh_float64_hash_func(val.real) ^ kh_float64_hash_func(val.imag);

From b68a533a7f642a9dd2de16e29cd0266a928effe8 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Mon, 4 Dec 2023 17:07:50 -0800
Subject: [PATCH 29/43] Revert "float set implementations"

This reverts commit 6266422371022ba227ae01634c40090e26635784.
---
 .../_libs/include/pandas/vendored/klib/khash_python.h  | 10 ----------
 1 file changed, 10 deletions(-)

diff --git a/pandas/_libs/include/pandas/vendored/klib/khash_python.h b/pandas/_libs/include/pandas/vendored/klib/khash_python.h
index 78528e27296fe..5a933b45d9e21 100644
--- a/pandas/_libs/include/pandas/vendored/klib/khash_python.h
+++ b/pandas/_libs/include/pandas/vendored/klib/khash_python.h
@@ -130,23 +130,13 @@ static inline khuint32_t kh_float32_hash_func(float val) {
   KHASH_INIT(name, khfloat64_t, khval_t, 1, kh_float64_hash_func,              \
              kh_floats_hash_equal)
 
-#define KHASH_SET_INIT_FLOAT64(name)                                           \
-  KHASH_INIT(name, khfloat64_t, char, 0, kh_float64_hash_func,                 \
-             kh_floats_hash_equal)
-
 KHASH_MAP_INIT_FLOAT64(float64, size_t)
-KHASH_SET_INIT_FLOAT64(float64_set)
 
 #define KHASH_MAP_INIT_FLOAT32(name, khval_t)                                  \
   KHASH_INIT(name, khfloat32_t, khval_t, 1, kh_float32_hash_func,              \
              kh_floats_hash_equal)
 
-#define KHASH_SET_INIT_FLOAT32(name)                                           \
-  KHASH_INIT(name, khfloat32_t, char, 0, kh_float32_hash_func,                 \
-             kh_floats_hash_equal)
-
 KHASH_MAP_INIT_FLOAT32(float32, size_t)
-KHASH_SET_INIT_FLOAT32(float32_set)
 
 static inline khint32_t kh_complex128_hash_func(khcomplex128_t val) {
   return kh_float64_hash_func(val.real) ^ kh_float64_hash_func(val.imag);

From a03ad1e84f52646a8f0ef4144fa591d45263b03c Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 15 Dec 2023 17:17:16 -0500
Subject: [PATCH 30/43] change marker name

---
 .github/workflows/unit-tests.yml                          | 4 ++--
 pandas/tests/frame/test_constructors.py                   | 4 ++--
 pandas/tests/groupby/test_cumulative.py                   | 2 +-
 pandas/tests/io/parser/common/test_float.py               | 4 ++--
 pandas/tests/scalar/timedelta/methods/test_round.py       | 4 ++--
 pandas/tests/scalar/timedelta/test_arithmetic.py          | 2 +-
 pandas/tests/scalar/timedelta/test_timedelta.py           | 2 +-
 pandas/tests/scalar/timestamp/methods/test_tz_localize.py | 2 +-
 pandas/tests/scalar/timestamp/test_constructors.py        | 2 +-
 pandas/tests/tools/test_to_datetime.py                    | 2 +-
 pyproject.toml                                            | 2 +-
 scripts/tests/data/deps_minimum.toml                      | 2 +-
 12 files changed, 16 insertions(+), 16 deletions(-)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 7367d57cc8066..8d49a8a3f8e50 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -331,7 +331,7 @@ jobs:
       with:
         preload: LD_PRELOAD=$(gcc -print-file-name=libasan.so)
       env:
-        PATTERN: "not slow and not network and not single_cpu and not known_ub"
+        PATTERN: "not slow and not network and not single_cpu and not skip_ubsan"
         PYTEST_WORKERS: 'auto'
         PYTEST_TARGET: 'pandas'
         ASAN_OPTIONS: detect_leaks=0  # leak detection in Python not yet working
@@ -342,7 +342,7 @@ jobs:
     #  with:
     #    preload: LD_PRELOAD=$(gcc -print-file-name=libasan.so)
     #  env:
-    #    PATTERN: "single_cpu and not known_ub"
+    #    PATTERN: "single_cpu and not skip_ubsan"
     #    PYTEST_WORKERS: 0
     #    PYTEST_TARGET: 'pandas'
     #    ASAN_OPTIONS: detect_leaks=0 # leak detection in Python not yet working
diff --git a/pandas/tests/frame/test_constructors.py b/pandas/tests/frame/test_constructors.py
index aebeff88730b9..f64d6a886fe9f 100644
--- a/pandas/tests/frame/test_constructors.py
+++ b/pandas/tests/frame/test_constructors.py
@@ -3206,7 +3206,7 @@ def test_from_out_of_bounds_ns_datetime(
         assert item.asm8.dtype == exp_dtype
         assert dtype == exp_dtype
 
-    @pytest.mark.known_ub
+    @pytest.mark.skip_ubsan
     def test_out_of_s_bounds_datetime64(self, constructor):
         scalar = np.datetime64(np.iinfo(np.int64).max, "D")
         result = constructor(scalar)
@@ -3242,7 +3242,7 @@ def test_from_out_of_bounds_ns_timedelta(
         assert item.asm8.dtype == exp_dtype
         assert dtype == exp_dtype
 
-    @pytest.mark.known_ub
+    @pytest.mark.skip_ubsan
     @pytest.mark.parametrize("cls", [np.datetime64, np.timedelta64])
     def test_out_of_s_bounds_timedelta64(self, constructor, cls):
         scalar = cls(np.iinfo(np.int64).max, "D")
diff --git a/pandas/tests/groupby/test_cumulative.py b/pandas/tests/groupby/test_cumulative.py
index ab5aeee18ba7b..1bdbef6d50c4c 100644
--- a/pandas/tests/groupby/test_cumulative.py
+++ b/pandas/tests/groupby/test_cumulative.py
@@ -60,7 +60,7 @@ def test_groupby_cumprod():
     tm.assert_series_equal(actual, expected)
 
 
-@pytest.mark.known_ub
+@pytest.mark.skip_ubsan
 def test_groupby_cumprod_overflow():
     # GH#37493 if we overflow we return garbage consistent with numpy
     df = DataFrame({"key": ["b"] * 4, "value": 100_000})
diff --git a/pandas/tests/io/parser/common/test_float.py b/pandas/tests/io/parser/common/test_float.py
index 6ed32427bebf9..6069c23936297 100644
--- a/pandas/tests/io/parser/common/test_float.py
+++ b/pandas/tests/io/parser/common/test_float.py
@@ -45,7 +45,7 @@ def test_scientific_no_exponent(all_parsers_all_precisions):
     [
         -617,
         -100000,
-        pytest.param(-99999999999999999, marks=pytest.mark.known_ub),
+        pytest.param(-99999999999999999, marks=pytest.mark.skip_ubsan),
     ],
 )
 def test_very_negative_exponent(all_parsers_all_precisions, neg_exp):
@@ -58,7 +58,7 @@ def test_very_negative_exponent(all_parsers_all_precisions, neg_exp):
     tm.assert_frame_equal(result, expected)
 
 
-@pytest.mark.known_ub
+@pytest.mark.skip_ubsan
 @xfail_pyarrow  # AssertionError: Attributes of DataFrame.iloc[:, 0] are different
 @pytest.mark.parametrize("exp", [999999999999999999, -999999999999999999])
 def test_too_many_exponent_digits(all_parsers_all_precisions, exp, request):
diff --git a/pandas/tests/scalar/timedelta/methods/test_round.py b/pandas/tests/scalar/timedelta/methods/test_round.py
index 524a0b5736cee..e54adb27d126b 100644
--- a/pandas/tests/scalar/timedelta/methods/test_round.py
+++ b/pandas/tests/scalar/timedelta/methods/test_round.py
@@ -61,7 +61,7 @@ def test_round_invalid(self):
             with pytest.raises(ValueError, match=msg):
                 t1.round(freq)
 
-    @pytest.mark.known_ub
+    @pytest.mark.skip_ubsan
     def test_round_implementation_bounds(self):
         # See also: analogous test for Timestamp
         # GH#38964
@@ -87,7 +87,7 @@ def test_round_implementation_bounds(self):
         with pytest.raises(OutOfBoundsTimedelta, match=msg):
             Timedelta.max.round("s")
 
-    @pytest.mark.known_ub
+    @pytest.mark.skip_ubsan
     @given(val=st.integers(min_value=iNaT + 1, max_value=lib.i8max))
     @pytest.mark.parametrize(
         "method", [Timedelta.round, Timedelta.floor, Timedelta.ceil]
diff --git a/pandas/tests/scalar/timedelta/test_arithmetic.py b/pandas/tests/scalar/timedelta/test_arithmetic.py
index 3e578db6723a7..a5d786cfba427 100644
--- a/pandas/tests/scalar/timedelta/test_arithmetic.py
+++ b/pandas/tests/scalar/timedelta/test_arithmetic.py
@@ -966,7 +966,7 @@ def test_td_op_timedelta_timedeltalike_array(self, op, arr):
 
 
 class TestTimedeltaComparison:
-    @pytest.mark.known_ub
+    @pytest.mark.skip_ubsan
     def test_compare_pytimedelta_bounds(self):
         # GH#49021 don't overflow on comparison with very large pytimedeltas
 
diff --git a/pandas/tests/scalar/timedelta/test_timedelta.py b/pandas/tests/scalar/timedelta/test_timedelta.py
index 94b5a7821c20c..d4398f66e6f89 100644
--- a/pandas/tests/scalar/timedelta/test_timedelta.py
+++ b/pandas/tests/scalar/timedelta/test_timedelta.py
@@ -551,7 +551,7 @@ def test_timedelta_hash_equality(self):
         ns_td = Timedelta(1, "ns")
         assert hash(ns_td) != hash(ns_td.to_pytimedelta())
 
-    @pytest.mark.known_ub
+    @pytest.mark.skip_ubsan
     @pytest.mark.xfail(
         reason="pd.Timedelta violates the Python hash invariant (GH#44504).",
     )
diff --git a/pandas/tests/scalar/timestamp/methods/test_tz_localize.py b/pandas/tests/scalar/timestamp/methods/test_tz_localize.py
index 8169aa1035cf4..af3dee1880d2e 100644
--- a/pandas/tests/scalar/timestamp/methods/test_tz_localize.py
+++ b/pandas/tests/scalar/timestamp/methods/test_tz_localize.py
@@ -25,7 +25,7 @@
 
 
 class TestTimestampTZLocalize:
-    @pytest.mark.known_ub
+    @pytest.mark.skip_ubsan
     def test_tz_localize_pushes_out_of_bounds(self):
         # GH#12677
         # tz_localize that pushes away from the boundary is OK
diff --git a/pandas/tests/scalar/timestamp/test_constructors.py b/pandas/tests/scalar/timestamp/test_constructors.py
index d229d59335476..3975f3c46aaa1 100644
--- a/pandas/tests/scalar/timestamp/test_constructors.py
+++ b/pandas/tests/scalar/timestamp/test_constructors.py
@@ -822,7 +822,7 @@ def test_barely_out_of_bounds(self):
         with pytest.raises(OutOfBoundsDatetime, match=msg):
             Timestamp("2262-04-11 23:47:16.854775808")
 
-    @pytest.mark.known_ub
+    @pytest.mark.skip_ubsan
     def test_bounds_with_different_units(self):
         out_of_bounds_dates = ("1677-09-21", "2262-04-12")
 
diff --git a/pandas/tests/tools/test_to_datetime.py b/pandas/tests/tools/test_to_datetime.py
index ed0af5c2144e8..368a50022b1b7 100644
--- a/pandas/tests/tools/test_to_datetime.py
+++ b/pandas/tests/tools/test_to_datetime.py
@@ -1140,7 +1140,7 @@ def test_to_datetime_dt64s_out_of_ns_bounds(self, cache, dt, errors):
         assert ts.unit == "s"
         assert ts.asm8 == dt
 
-    @pytest.mark.known_ub
+    @pytest.mark.skip_ubsan
     def test_to_datetime_dt64d_out_of_bounds(self, cache):
         dt64 = np.datetime64(np.iinfo(np.int64).max, "D")
 
diff --git a/pyproject.toml b/pyproject.toml
index 0365f21830f2e..ea6c801f241ba 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -524,7 +524,7 @@ markers = [
   "clipboard: mark a pd.read_clipboard test",
   "arm_slow: mark a test as slow for arm64 architecture",
   "arraymanager: mark a test to run with ArrayManager enabled",
-  "known_ub: tests known to invoke undefined behavior",
+  "skip_ubsan: Tests known to fail UBSAN check",
 ]
 
 [tool.mypy]
diff --git a/scripts/tests/data/deps_minimum.toml b/scripts/tests/data/deps_minimum.toml
index c5eed77928271..968b3b170f38a 100644
--- a/scripts/tests/data/deps_minimum.toml
+++ b/scripts/tests/data/deps_minimum.toml
@@ -383,7 +383,7 @@ markers = [
   "clipboard: mark a pd.read_clipboard test",
   "arm_slow: mark a test as slow for arm64 architecture",
   "arraymanager: mark a test to run with ArrayManager enabled",
-  "known_ub: tests known to invoke undefined behavior",
+  "skip_ubsan: tests known to invoke undefined behavior",
 ]
 
 [tool.mypy]

From 656edb1d2d75cc77cd4f7a6c45fef15fc9814e57 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 15 Dec 2023 17:21:58 -0500
Subject: [PATCH 31/43] dedicated actions file

---
 .github/workflows/unit-tests.yml    | 21 ++++++++++----------
 ci/deps/actions-311-sanitizers.yaml | 30 +++++++++++++++++++++++++++++
 2 files changed, 40 insertions(+), 11 deletions(-)
 create mode 100644 ci/deps/actions-311-sanitizers.yaml

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 8d49a8a3f8e50..d585d61e2f082 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -317,7 +317,7 @@ jobs:
     - name: Set up Conda
       uses: ./.github/actions/setup-conda
       with:
-        environment-file: ci/deps/actions-311-numpydev.yaml
+        environment-file: ci/deps/actions-311-sanitizers.yaml
 
     - name: Build Pandas
       id: build
@@ -336,16 +336,15 @@ jobs:
         PYTEST_TARGET: 'pandas'
         ASAN_OPTIONS: detect_leaks=0  # leak detection in Python not yet working
 
-    # Skipping single CPU tests for now - see pytest GH issue 641 for error
-    #- name: Test (single_cpu)
-    #  uses: ./.github/actions/run-tests
-    #  with:
-    #    preload: LD_PRELOAD=$(gcc -print-file-name=libasan.so)
-    #  env:
-    #    PATTERN: "single_cpu and not skip_ubsan"
-    #    PYTEST_WORKERS: 0
-    #    PYTEST_TARGET: 'pandas'
-    #    ASAN_OPTIONS: detect_leaks=0 # leak detection in Python not yet working
+    - name: Test (single_cpu)
+      uses: ./.github/actions/run-tests
+      with:
+        preload: LD_PRELOAD=$(gcc -print-file-name=libasan.so)
+      env:
+        PATTERN: "single_cpu and not skip_ubsan"
+        PYTEST_WORKERS: 0
+        PYTEST_TARGET: 'pandas'
+        ASAN_OPTIONS: detect_leaks=0 # leak detection in Python not yet working
 
   python-dev:
     # This job may or may not run depending on the state of the next
diff --git a/ci/deps/actions-311-sanitizers.yaml b/ci/deps/actions-311-sanitizers.yaml
new file mode 100644
index 0000000000000..4744d622fc687
--- /dev/null
+++ b/ci/deps/actions-311-sanitizers.yaml
@@ -0,0 +1,30 @@
+name: pandas-dev
+channels:
+  - conda-forge
+dependencies:
+  - python=3.11
+
+  # build dependencies
+  - versioneer[toml]
+  - cython>=0.29.33  
+  - meson[ninja]=1.2.1
+  - meson-python=0.13.1
+
+  # test dependencies
+  - pytest>=7.3.2
+  - pytest-cov
+  - pytest-xdist>=2.2.0
+  - pytest-localserver>=0.7.1
+  - pytest-qt>=4.2.0
+  - boto3
+
+  # required dependencies
+  - python-dateutil
+  - numpy<2
+  - pytz
+
+  # pandas dependencies
+  - pip
+
+  - pip:
+    - "tzdata>=2022.7"

From 2aabda1e4fa96e187fbb0acd3649e296d370f307 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 15 Dec 2023 17:29:22 -0500
Subject: [PATCH 32/43] consolidated into matrix

---
 .github/actions/run-tests/action.yml |  5 ++-
 .github/workflows/unit-tests.yml     | 57 ++++++++--------------------
 2 files changed, 20 insertions(+), 42 deletions(-)

diff --git a/.github/actions/run-tests/action.yml b/.github/actions/run-tests/action.yml
index 02a400ab23263..b4778b74df335 100644
--- a/.github/actions/run-tests/action.yml
+++ b/.github/actions/run-tests/action.yml
@@ -3,11 +3,14 @@ inputs:
   preload:
     description: Preload arguments for sanitizer
     required: false
+  asan_options:
+    description: Arguments for Address Sanitizer (ASAN)
+    required: false
 runs:
   using: composite
   steps:
     - name: Test
-      run: ${{ inputs.preload }} ci/run_tests.sh
+      run: ${{ inputs.asan_options }} ${{ inputs.preload }} ci/run_tests.sh
       shell: bash -el {0}
 
     - name: Publish test results
diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index d585d61e2f082..5b26bdb669744 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -96,6 +96,13 @@ jobs:
           - name: "Pyarrow Nightly"
             env_file: actions-311-pyarrownightly.yaml
             pattern: "not slow and not network and not single_cpu"
+          - name: "ASAN / UBSAN"
+            env_file: actions-311-sanitizers.yaml
+            pattern: "not slow and not network and not single_cpu and not skip_ubsan"
+            asan_options="detect_leaks=0"
+            preload: LD_PRELOAD=$(gcc -print-file-name=libasan.so)
+            meson_args: --config-settings=setup-args="-Db_sanitize=address,undefined"
+            cflags_adds: -fno-sanitize-recover=all
       fail-fast: false
     name: ${{ matrix.name || format('ubuntu-latest {0}', matrix.env_file) }}
     env:
@@ -174,16 +181,25 @@ jobs:
     - name: Build Pandas
       id: build
       uses: ./.github/actions/build_pandas
+      with:
+        meson_args: --config-settings=setup-args="-Db_sanitize=address,undefined"
+        cflags_adds: -fno-sanitize-recover=all      
 
     - name: Test (not single_cpu)
       uses: ./.github/actions/run-tests
       if: ${{ matrix.name != 'Pypy' }}
+      with:
+        preload: ${{ matrix.preload }}
+        asan_options: ${{ matrix.asan_options }}        
       env:
         # Set pattern to not single_cpu if not already set
         PATTERN: ${{ env.PATTERN == '' && 'not single_cpu' || matrix.pattern }}
 
     - name: Test (single_cpu)
       uses: ./.github/actions/run-tests
+      with:
+        preload: ${{ matrix.preload }}
+        asan_options: ${{ matrix.asan_options }}
       env:
         PATTERN: 'single_cpu'
         PYTEST_WORKERS: 0
@@ -305,47 +321,6 @@ jobs:
       group: ${{ github.event_name == 'push' && github.run_number || github.ref }}-musl
       cancel-in-progress: true
 
-  ASAN_UBSAN:
-    runs-on: ubuntu-22.04
-    timeout-minutes: 90
-    steps:
-    - name: Checkout
-      uses: actions/checkout@v4
-      with:
-        fetch-depth: 0
-
-    - name: Set up Conda
-      uses: ./.github/actions/setup-conda
-      with:
-        environment-file: ci/deps/actions-311-sanitizers.yaml
-
-    - name: Build Pandas
-      id: build
-      uses: ./.github/actions/build_pandas
-      with:
-        meson_args: --config-settings=setup-args="-Db_sanitize=address,undefined"
-        cflags_adds: -fno-sanitize-recover=all
-
-    - name: Test (not single_cpu)
-      uses: ./.github/actions/run-tests
-      with:
-        preload: LD_PRELOAD=$(gcc -print-file-name=libasan.so)
-      env:
-        PATTERN: "not slow and not network and not single_cpu and not skip_ubsan"
-        PYTEST_WORKERS: 'auto'
-        PYTEST_TARGET: 'pandas'
-        ASAN_OPTIONS: detect_leaks=0  # leak detection in Python not yet working
-
-    - name: Test (single_cpu)
-      uses: ./.github/actions/run-tests
-      with:
-        preload: LD_PRELOAD=$(gcc -print-file-name=libasan.so)
-      env:
-        PATTERN: "single_cpu and not skip_ubsan"
-        PYTEST_WORKERS: 0
-        PYTEST_TARGET: 'pandas'
-        ASAN_OPTIONS: detect_leaks=0 # leak detection in Python not yet working
-
   python-dev:
     # This job may or may not run depending on the state of the next
     # unreleased Python version. DO NOT DELETE IT.

From 3056e5f682b194feadc78f5c7f96008c6ca363af Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 15 Dec 2023 17:41:04 -0500
Subject: [PATCH 33/43] fixup

---
 .github/workflows/unit-tests.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 5b26bdb669744..5d3e407246420 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -99,7 +99,7 @@ jobs:
           - name: "ASAN / UBSAN"
             env_file: actions-311-sanitizers.yaml
             pattern: "not slow and not network and not single_cpu and not skip_ubsan"
-            asan_options="detect_leaks=0"
+            asan_options: "ASAN_OPTIONS=detect_leaks=0"
             preload: LD_PRELOAD=$(gcc -print-file-name=libasan.so)
             meson_args: --config-settings=setup-args="-Db_sanitize=address,undefined"
             cflags_adds: -fno-sanitize-recover=all

From 89b2b80086859f46dd0d7f2f4f1bb40031a4e754 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 15 Dec 2023 18:00:40 -0500
Subject: [PATCH 34/43] typos

---
 .github/workflows/unit-tests.yml | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 5d3e407246420..829c1b6b18063 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -182,8 +182,8 @@ jobs:
       id: build
       uses: ./.github/actions/build_pandas
       with:
-        meson_args: --config-settings=setup-args="-Db_sanitize=address,undefined"
-        cflags_adds: -fno-sanitize-recover=all      
+        meson_args: ${{ matrix.meson_args }}
+        cflags_adds: ${{ matrix.cflags_adds }}
 
     - name: Test (not single_cpu)
       uses: ./.github/actions/run-tests

From d591b78ddd98e2f8212e35a604e2d76bbb49caca Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 15 Dec 2023 20:42:28 -0500
Subject: [PATCH 35/43] fixups

---
 .github/workflows/unit-tests.yml    | 2 +-
 ci/deps/actions-311-sanitizers.yaml | 3 ++-
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 829c1b6b18063..6854f7a924066 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -190,7 +190,7 @@ jobs:
       if: ${{ matrix.name != 'Pypy' }}
       with:
         preload: ${{ matrix.preload }}
-        asan_options: ${{ matrix.asan_options }}        
+        asan_options: ${{ matrix.asan_options }}
       env:
         # Set pattern to not single_cpu if not already set
         PATTERN: ${{ env.PATTERN == '' && 'not single_cpu' || matrix.pattern }}
diff --git a/ci/deps/actions-311-sanitizers.yaml b/ci/deps/actions-311-sanitizers.yaml
index 4744d622fc687..18e7a360a56c9 100644
--- a/ci/deps/actions-311-sanitizers.yaml
+++ b/ci/deps/actions-311-sanitizers.yaml
@@ -6,7 +6,7 @@ dependencies:
 
   # build dependencies
   - versioneer[toml]
-  - cython>=0.29.33  
+  - cython>=0.29.33
   - meson[ninja]=1.2.1
   - meson-python=0.13.1
 
@@ -17,6 +17,7 @@ dependencies:
   - pytest-localserver>=0.7.1
   - pytest-qt>=4.2.0
   - boto3
+  - hypothesis>=6.46.1
 
   # required dependencies
   - python-dateutil

From 644206637ea54b82d468d45e79ddeae96fb5013d Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Fri, 15 Dec 2023 22:40:31 -0500
Subject: [PATCH 36/43] add qt?

---
 ci/deps/actions-311-sanitizers.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/ci/deps/actions-311-sanitizers.yaml b/ci/deps/actions-311-sanitizers.yaml
index 18e7a360a56c9..dcd381066b0ea 100644
--- a/ci/deps/actions-311-sanitizers.yaml
+++ b/ci/deps/actions-311-sanitizers.yaml
@@ -18,6 +18,7 @@ dependencies:
   - pytest-qt>=4.2.0
   - boto3
   - hypothesis>=6.46.1
+  - pyqt>=5.15.9
 
   # required dependencies
   - python-dateutil

From 02bf20daaa306ce84214e56173abb14f9923f4a4 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Tue, 19 Dec 2023 15:40:20 -0500
Subject: [PATCH 37/43] intentional UB with verbose

---
 .github/workflows/unit-tests.yml                   | 1 +
 pandas/_libs/src/vendored/ujson/python/objToJSON.c | 3 +++
 2 files changed, 4 insertions(+)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 6854f7a924066..4c5839a035261 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -103,6 +103,7 @@ jobs:
             preload: LD_PRELOAD=$(gcc -print-file-name=libasan.so)
             meson_args: --config-settings=setup-args="-Db_sanitize=address,undefined"
             cflags_adds: -fno-sanitize-recover=all
+            test_args: "-v"
       fail-fast: false
     name: ${{ matrix.name || format('ubuntu-latest {0}', matrix.env_file) }}
     env:
diff --git a/pandas/_libs/src/vendored/ujson/python/objToJSON.c b/pandas/_libs/src/vendored/ujson/python/objToJSON.c
index 8bba95dd456de..98b8a44949d60 100644
--- a/pandas/_libs/src/vendored/ujson/python/objToJSON.c
+++ b/pandas/_libs/src/vendored/ujson/python/objToJSON.c
@@ -2062,5 +2062,8 @@ PyObject *objToJSON(PyObject *Py_UNUSED(self), PyObject *args,
     encoder->free(ret);
   }
 
+  int myvar = INT_MAX;
+  printf("This value should overflow: %d\n", myvar + 1);
+
   return newobj;
 }

From 01070f35dad1bcc7f5345976fb2f58a5c3575b25 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Wed, 20 Dec 2023 18:00:32 -0500
Subject: [PATCH 38/43] disable pytest-xdist

---
 .github/workflows/unit-tests.yml                   | 4 ++--
 pandas/_libs/src/vendored/ujson/python/objToJSON.c | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index 4c5839a035261..57f9893d36044 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -103,7 +103,7 @@ jobs:
             preload: LD_PRELOAD=$(gcc -print-file-name=libasan.so)
             meson_args: --config-settings=setup-args="-Db_sanitize=address,undefined"
             cflags_adds: -fno-sanitize-recover=all
-            test_args: "-v"
+            pytest_workers: -1  # disable pytest-xdist as it swallows stderr from ASAN
       fail-fast: false
     name: ${{ matrix.name || format('ubuntu-latest {0}', matrix.env_file) }}
     env:
@@ -113,7 +113,7 @@ jobs:
       PANDAS_COPY_ON_WRITE: ${{ matrix.pandas_copy_on_write || '0' }}
       PANDAS_CI: ${{ matrix.pandas_ci || '1' }}
       TEST_ARGS: ${{ matrix.test_args || '' }}
-      PYTEST_WORKERS: 'auto'
+      PYTEST_WORKERS: ${{ matrix.pytest_workers || 'auto' }}
       PYTEST_TARGET: ${{ matrix.pytest_target || 'pandas' }}
       # Clipboard tests
       QT_QPA_PLATFORM: offscreen
diff --git a/pandas/_libs/src/vendored/ujson/python/objToJSON.c b/pandas/_libs/src/vendored/ujson/python/objToJSON.c
index 98b8a44949d60..a54c6bccab43a 100644
--- a/pandas/_libs/src/vendored/ujson/python/objToJSON.c
+++ b/pandas/_libs/src/vendored/ujson/python/objToJSON.c
@@ -2062,8 +2062,8 @@ PyObject *objToJSON(PyObject *Py_UNUSED(self), PyObject *args,
     encoder->free(ret);
   }
 
-  int myvar = INT_MAX;
-  printf("This value should overflow: %d\n", myvar + 1);
+  int someints[2] = {0, 1};
+  printf("the OOB value is: %d\n", someints[2]);
 
   return newobj;
 }

From 57ed2863ea68686abb617d29a580cf166079ac61 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Wed, 20 Dec 2023 18:12:00 -0500
Subject: [PATCH 39/43] original issue

---
 pandas/_libs/src/vendored/ujson/python/objToJSON.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/pandas/_libs/src/vendored/ujson/python/objToJSON.c b/pandas/_libs/src/vendored/ujson/python/objToJSON.c
index a54c6bccab43a..98b8a44949d60 100644
--- a/pandas/_libs/src/vendored/ujson/python/objToJSON.c
+++ b/pandas/_libs/src/vendored/ujson/python/objToJSON.c
@@ -2062,8 +2062,8 @@ PyObject *objToJSON(PyObject *Py_UNUSED(self), PyObject *args,
     encoder->free(ret);
   }
 
-  int someints[2] = {0, 1};
-  printf("the OOB value is: %d\n", someints[2]);
+  int myvar = INT_MAX;
+  printf("This value should overflow: %d\n", myvar + 1);
 
   return newobj;
 }

From 677da0e4bce131d6992c320920885eded2a5d796 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Wed, 20 Dec 2023 18:45:37 -0500
Subject: [PATCH 40/43] remove UB

---
 pandas/_libs/src/vendored/ujson/python/objToJSON.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/pandas/_libs/src/vendored/ujson/python/objToJSON.c b/pandas/_libs/src/vendored/ujson/python/objToJSON.c
index 98b8a44949d60..8bba95dd456de 100644
--- a/pandas/_libs/src/vendored/ujson/python/objToJSON.c
+++ b/pandas/_libs/src/vendored/ujson/python/objToJSON.c
@@ -2062,8 +2062,5 @@ PyObject *objToJSON(PyObject *Py_UNUSED(self), PyObject *args,
     encoder->free(ret);
   }
 
-  int myvar = INT_MAX;
-  printf("This value should overflow: %d\n", myvar + 1);
-
   return newobj;
 }

From af0150a8ec5058219a3628bbfb950bca21fceb8c Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Wed, 20 Dec 2023 20:57:31 -0500
Subject: [PATCH 41/43] Revert "remove UB"

This reverts commit 677da0e4bce131d6992c320920885eded2a5d796.
---
 pandas/_libs/src/vendored/ujson/python/objToJSON.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/pandas/_libs/src/vendored/ujson/python/objToJSON.c b/pandas/_libs/src/vendored/ujson/python/objToJSON.c
index 8bba95dd456de..98b8a44949d60 100644
--- a/pandas/_libs/src/vendored/ujson/python/objToJSON.c
+++ b/pandas/_libs/src/vendored/ujson/python/objToJSON.c
@@ -2062,5 +2062,8 @@ PyObject *objToJSON(PyObject *Py_UNUSED(self), PyObject *args,
     encoder->free(ret);
   }
 
+  int myvar = INT_MAX;
+  printf("This value should overflow: %d\n", myvar + 1);
+
   return newobj;
 }

From 4647f123a3b477bb64f5261f90b1a8874dae712e Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Wed, 20 Dec 2023 22:11:08 -0500
Subject: [PATCH 42/43] merge fixup

---
 pyproject.toml                       | 1 -
 scripts/tests/data/deps_minimum.toml | 1 -
 2 files changed, 2 deletions(-)

diff --git a/pyproject.toml b/pyproject.toml
index edee426e20b2a..ca19f463edf40 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -523,7 +523,6 @@ markers = [
   "db: tests requiring a database (mysql or postgres)",
   "clipboard: mark a pd.read_clipboard test",
   "arm_slow: mark a test as slow for arm64 architecture",
-  "arraymanager: mark a test to run with ArrayManager enabled",
   "skip_ubsan: Tests known to fail UBSAN check",
 ]
 
diff --git a/scripts/tests/data/deps_minimum.toml b/scripts/tests/data/deps_minimum.toml
index 968b3b170f38a..3be6be17d1ee2 100644
--- a/scripts/tests/data/deps_minimum.toml
+++ b/scripts/tests/data/deps_minimum.toml
@@ -382,7 +382,6 @@ markers = [
   "db: tests requiring a database (mysql or postgres)",
   "clipboard: mark a pd.read_clipboard test",
   "arm_slow: mark a test as slow for arm64 architecture",
-  "arraymanager: mark a test to run with ArrayManager enabled",
   "skip_ubsan: tests known to invoke undefined behavior",
 ]
 

From cba79f61ce80d791540b469145af306e014c8d33 Mon Sep 17 00:00:00 2001
From: Will Ayd <william.ayd@icloud.com>
Date: Wed, 20 Dec 2023 23:40:27 -0500
Subject: [PATCH 43/43] remove UB

---
 pandas/_libs/src/vendored/ujson/python/objToJSON.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/pandas/_libs/src/vendored/ujson/python/objToJSON.c b/pandas/_libs/src/vendored/ujson/python/objToJSON.c
index 98b8a44949d60..8bba95dd456de 100644
--- a/pandas/_libs/src/vendored/ujson/python/objToJSON.c
+++ b/pandas/_libs/src/vendored/ujson/python/objToJSON.c
@@ -2062,8 +2062,5 @@ PyObject *objToJSON(PyObject *Py_UNUSED(self), PyObject *args,
     encoder->free(ret);
   }
 
-  int myvar = INT_MAX;
-  printf("This value should overflow: %d\n", myvar + 1);
-
   return newobj;
 }