Merge branch 'release/2.6' into lxz/corpc_err_fix_26

liuxuezhao · liuxuezhao · commit abac378bd6cf · 2026-01-16T09:32:50.000Z
diff --git a/.github/workflows/bullseye-coverage.yml b/.github/workflows/bullseye-coverage.yml
@@ -366,7 +366,7 @@ jobs:
         if: (!cancelled()) && (success() || failure()) &&
             steps.run-test.outcome != 'skipped'
         # yamllint disable-line rule:line-length
-        uses: EnricoMi/publish-unit-test-result-action@34d7c956a59aed1bfebf31df77b8de55db9bbaaf  # v2.21.0
+        uses: EnricoMi/publish-unit-test-result-action@27d65e188ec43221b20d26de30f4892fad91df2f  # v2.22.0
         with:
           check_name: ${{ env.STAGE_NAME }} Test Results
           github_token: ${{ secrets.GITHUB_TOKEN }}
@@ -634,7 +634,7 @@ jobs:
         if: (!cancelled()) && (success() || failure()) &&
             steps.run-test.outcome != 'skipped'
         # yamllint disable-line rule:line-length
-        uses: EnricoMi/publish-unit-test-result-action@34d7c956a59aed1bfebf31df77b8de55db9bbaaf  # v2.21.0
+        uses: EnricoMi/publish-unit-test-result-action@27d65e188ec43221b20d26de30f4892fad91df2f  # v2.22.0
         with:
           check_name: ${{ env.STAGE_NAME }} Test Results
           github_token: ${{ secrets.GITHUB_TOKEN }}
diff --git a/.github/workflows/ci2.yml b/.github/workflows/ci2.yml
@@ -68,7 +68,7 @@ jobs:
       - name: Publish NLT test results
         if: always()
         # yamllint disable-line rule:line-length
-        uses: EnricoMi/publish-unit-test-result-action@34d7c956a59aed1bfebf31df77b8de55db9bbaaf  # v2.21.0
+        uses: EnricoMi/publish-unit-test-result-action@27d65e188ec43221b20d26de30f4892fad91df2f  # v2.22.0
         with:
           github_token: ${{ secrets.GITHUB_TOKEN }}
           files: nlt-junit.xml
diff --git a/.github/workflows/landing-builds.yml b/.github/workflows/landing-builds.yml
@@ -143,7 +143,7 @@ jobs:
       - name: Publish NLT test results
         if: always()
         # yamllint disable-line rule:line-length
-        uses: EnricoMi/publish-unit-test-result-action@34d7c956a59aed1bfebf31df77b8de55db9bbaaf  # v2.21.0
+        uses: EnricoMi/publish-unit-test-result-action@27d65e188ec43221b20d26de30f4892fad91df2f  # v2.22.0
         with:
           github_token: ${{ secrets.GITHUB_TOKEN }}
           files: nlt-junit.xml
diff --git a/.github/workflows/ossf-scorecard.yml b/.github/workflows/ossf-scorecard.yml
@@ -71,6 +71,6 @@ jobs:
       # Upload the results to GitHub's code scanning dashboard (optional).
       # Commenting out will disable upload of results to your repo's Code Scanning dashboard
       - name: "Upload to code-scanning"
-        uses: github/codeql-action/upload-sarif@1b168cd39490f61582a9beae412bb7057a6b2c4e  # v4.31.8
+        uses: github/codeql-action/upload-sarif@5d4e8d1aca955e8d8589aabd499c5cae939e33c7  # v4.31.9
         with:
           sarif_file: results.sarif
diff --git a/.github/workflows/rpm-build-and-test-report.yml b/.github/workflows/rpm-build-and-test-report.yml
@@ -93,7 +93,7 @@ jobs:
             esac
             echo "STAGE_NAME=Build RPM on $DISTRO_NAME $DISTRO_VERSION" >> $GITHUB_ENV
       - name: Test Report
-        uses: dorny/test-reporter@fe45e9537387dac839af0d33ba56eed8e24189e8  # v2.3.0
+        uses: dorny/test-reporter@b082adf0eced0765477756c2a610396589b8c637  # v2.5.0
         with:
           artifact: ${{ env.STAGE_NAME }} test-results
           name: ${{ env.STAGE_NAME }} Test Results (dorny)
@@ -112,7 +112,7 @@ jobs:
       - name: Set variables
         run: echo "STAGE_NAME=Functional Hardware ${{ matrix.stage }}" >> $GITHUB_ENV
       - name: Test Report
-        uses: dorny/test-reporter@fe45e9537387dac839af0d33ba56eed8e24189e8  # v2.3.0
+        uses: dorny/test-reporter@b082adf0eced0765477756c2a610396589b8c637  # v2.5.0
         with:
           artifact: ${{ env.STAGE_NAME }} test-results
           name: ${{ env.STAGE_NAME }} Test Results (dorny)
diff --git a/.github/workflows/rpm-build-and-test.yml b/.github/workflows/rpm-build-and-test.yml
@@ -374,7 +374,7 @@ jobs:
         if: (!cancelled()) && (success() || failure()) &&
             steps.run-test.outcome != 'skipped'
         # yamllint disable-line rule:line-length
-        uses: EnricoMi/publish-unit-test-result-action@34d7c956a59aed1bfebf31df77b8de55db9bbaaf  # v2.21.0
+        uses: EnricoMi/publish-unit-test-result-action@27d65e188ec43221b20d26de30f4892fad91df2f  # v2.22.0
         with:
           check_name: ${{ env.STAGE_NAME }} Test Results
           github_token: ${{ secrets.GITHUB_TOKEN }}
@@ -642,7 +642,7 @@ jobs:
         if: (!cancelled()) && (success() || failure()) &&
             steps.run-test.outcome != 'skipped'
         # yamllint disable-line rule:line-length
-        uses: EnricoMi/publish-unit-test-result-action@34d7c956a59aed1bfebf31df77b8de55db9bbaaf  # v2.21.0
+        uses: EnricoMi/publish-unit-test-result-action@27d65e188ec43221b20d26de30f4892fad91df2f  # v2.22.0
         with:
           check_name: ${{ env.STAGE_NAME }} Test Results
           github_token: ${{ secrets.GITHUB_TOKEN }}
diff --git a/.github/workflows/trivy.yml b/.github/workflows/trivy.yml
@@ -68,7 +68,7 @@ jobs:
           trivy-config: 'utils/trivy/trivy.yaml'
 
       - name: Upload Trivy scan results to GitHub Security tab
-        uses: github/codeql-action/upload-sarif@1b168cd39490f61582a9beae412bb7057a6b2c4e  # v4.31.8
+        uses: github/codeql-action/upload-sarif@5d4e8d1aca955e8d8589aabd499c5cae939e33c7  # v4.31.9
         with:
           sarif_file: 'trivy-results.sarif'
 
diff --git a/debian/changelog b/debian/changelog
@@ -1,3 +1,9 @@
+daos (2.6.4-13) unstable; urgency=medium
+  [ Phillip Henderson ]
+  * Bump version
+
+ -- Phillip Henderson <phillip.henderson@hpe.com>  Thu, 8 Jan 2026 12:57:00 -0500
+
 daos (2.6.4-11) unstable; urgency=medium
   [ Phillip Henderson ]
   * Bump version
diff --git a/src/client/java/pom.xml b/src/client/java/pom.xml
@@ -16,7 +16,7 @@
     <junit.platform.version>1.4.0</junit.platform.version>
     <junit.jupiter.version>5.4.0</junit.jupiter.version>
     <slf4j.version>1.7.25</slf4j.version>
-    <log4j.version>2.17.1</log4j.version>
+    <log4j.version>2.25.3</log4j.version>
     <checkstyle.failOnViolation>true</checkstyle.failOnViolation>
     <maven.compiler.source>1.8</maven.compiler.source>
     <maven.compiler.target>1.8</maven.compiler.target>
diff --git a/src/common/dav/dav_internal.h b/src/common/dav/dav_internal.h
@@ -69,6 +69,7 @@ typedef struct dav_obj {
 static inline
 struct dav_tx *utx2wtx(struct umem_wal_tx *utx)
 {
+	D_ASSERT(utx != NULL);
 	return (struct dav_tx *)&utx->utx_private;
 }
 
diff --git a/src/common/dav/tx.c b/src/common/dav/tx.c
@@ -1512,6 +1512,8 @@ dav_reserve(dav_obj_t *pop, struct dav_action *act, size_t size, uint64_t type_n
 
 	if (palloc_reserve(pop->do_heap, size, NULL, NULL, type_num,
 		0, 0, 0, act) != 0) {
+		if (!tx_inprogress)
+			lw_tx_end(pop, NULL);
 		DAV_API_END();
 		return 0;
 	}
@@ -1565,9 +1567,26 @@ dav_publish(dav_obj_t *pop, struct dav_action *actv, size_t actvcnt)
 void
 dav_cancel(dav_obj_t *pop, struct dav_action *actv, size_t actvcnt)
 {
+	int rc, tx_inprogress = 0;
+
 	DAV_DBG("actvcnt=%zu", actvcnt);
+	if (get_tx()->stage != DAV_TX_STAGE_NONE)
+		tx_inprogress = 1;
+
 	DAV_API_START();
+	if (!tx_inprogress) {
+		rc = lw_tx_begin(pop);
+		if (rc) {
+			D_ERROR("Failed to start local tx. %d\n", rc);
+			return;
+		}
+	}
+
 	palloc_cancel(pop->do_heap, actv, actvcnt);
+
+	if (!tx_inprogress)
+		lw_tx_end(pop, NULL);
+
 	DAV_API_END();
 }
 
diff --git a/src/common/tests/umem_test_bmem.c b/src/common/tests/umem_test_bmem.c
@@ -1,6 +1,6 @@
 /**
  * (C) Copyright 2019-2023 Intel Corporation.
- * (C) Copyright 2023 Hewlett Packard Enterprise Development LP.
+ * (C) Copyright 2023-2025 Hewlett Packard Enterprise Development LP.
  *
  * SPDX-License-Identifier: BSD-2-Clause-Patent
  */
@@ -1193,7 +1193,8 @@ test_tx_reserve_publish_cancel(void **state)
 	assert_int_equal(memcmp(rsrv_ptr1, local_buf, 980), 0);
 	assert_int_equal(memcmp(rsrv_ptr2, local_buf, 128), 0);
 	umem_cancel(umm, rsrvd_act);
-	validate_persist_activity(1, 0);
+	/* umem_cacnel() internally started tx, which increased one additional resrv_cnt */
+	validate_persist_activity(2, 0);
 	utest_get_scm_used_space(arg->ta_utx, &cur_mem_used);
 	assert_true(cur_mem_used >= initial_mem_used);
 	umoff = umem_atomic_alloc(umm, 980, UMEM_TYPE_ANY);
diff --git a/src/container/srv_target.c b/src/container/srv_target.c
@@ -183,8 +183,8 @@ cont_aggregate_runnable(struct ds_cont_child *cont, struct sched_request *req,
 
 	if (ds_pool_is_rebuilding(pool) && !vos_agg) {
 		D_DEBUG(DB_EPC, DF_CONT ": skip EC aggregation during rebuild %d, %d.\n",
-			DP_CONT(cont->sc_pool->spc_uuid, cont->sc_uuid), pool->sp_rebuilding,
-			pool->sp_rebuild_scan);
+			DP_CONT(cont->sc_pool->spc_uuid, cont->sc_uuid),
+			atomic_load(&pool->sp_rebuilding), pool->sp_rebuild_scan);
 		return false;
 	}
 
diff --git a/src/include/daos_srv/daos_engine.h b/src/include/daos_srv/daos_engine.h
@@ -707,10 +707,10 @@ ds_object_migrate_send(struct ds_pool *pool, uuid_t pool_hdl_uuid, uuid_t cont_u
 		       uint32_t new_gl_ver, unsigned int migrate_opc, uint64_t *enqueue_id,
 		       uint32_t *max_delay);
 int
-ds_migrate_object(struct ds_pool *pool, uuid_t po_hdl, uuid_t co_hdl, uuid_t co_uuid,
-		  uint32_t version, uint32_t generation, uint64_t max_eph, uint32_t opc,
-		  daos_unit_oid_t *oids, daos_epoch_t *epochs, daos_epoch_t *punched_epochs,
-		  unsigned int *shards, uint32_t count, unsigned int tgt_idx, uint32_t new_gl_ver);
+ds_migrate_object(uuid_t pool_uuid, uuid_t po_hdl, uuid_t co_hdl, uuid_t co_uuid, uint32_t version,
+		  uint32_t generation, uint64_t max_eph, uint32_t opc, daos_unit_oid_t *oids,
+		  daos_epoch_t *epochs, daos_epoch_t *punched_epochs, unsigned int *shards,
+		  uint32_t count, unsigned int tgt_idx, uint32_t new_gl_ver);
 void
 ds_migrate_stop(struct ds_pool *pool, uint32_t ver, unsigned int generation);
 
diff --git a/src/include/daos_srv/pool.h b/src/include/daos_srv/pool.h
@@ -25,6 +25,7 @@
 #include <daos_pool.h>
 #include <daos_security.h>
 #include <gurt/telemetry_common.h>
+#include <gurt/atomic.h>
 #include <daos_srv/rdb.h>
 
 /* Pool service (opaque) */
@@ -91,7 +92,7 @@ struct ds_pool {
 	 * rebuild job.
 	 */
 	uint32_t                 sp_rebuild_gen;
-	int			sp_rebuilding;
+	ATOMIC int               sp_rebuilding;
 	/**
 	 * someone has already messaged this pool to for rebuild scan,
 	 * NB: all xstreams can do lockless-write on it but it's OK
@@ -213,7 +214,7 @@ struct ds_pool_svc_op_val {
 static inline bool
 ds_pool_is_rebuilding(struct ds_pool *pool)
 {
-	return (pool->sp_rebuilding > 0 || pool->sp_rebuild_scan > 0);
+	return (atomic_load(&pool->sp_rebuilding) > 0 || pool->sp_rebuild_scan > 0);
 }
 
 /* encode metadata RPC operation key: HLC time first, in network order, for keys sorted by time.
diff --git a/src/object/srv_ec_aggregate.c b/src/object/srv_ec_aggregate.c
@@ -2288,9 +2288,9 @@ ec_aggregate_yield(struct ec_agg_param *agg_param)
 	int	rc;
 
 	if (ds_pool_is_rebuilding(agg_param->ap_pool_info.api_pool)) {
-		D_INFO(DF_UUID": abort ec aggregation, sp_rebuilding %d\n",
+		D_INFO(DF_UUID ": abort ec aggregation, sp_rebuilding %d\n",
 		       DP_UUID(agg_param->ap_pool_info.api_pool->sp_uuid),
-		       agg_param->ap_pool_info.api_pool->sp_rebuilding);
+		       atomic_load(&agg_param->ap_pool_info.api_pool->sp_rebuilding));
 		return true;
 	}
 
@@ -2494,10 +2494,10 @@ agg_iterate_pre_cb(daos_handle_t ih, vos_iter_entry_t *entry,
 	 * (see obj_inflight_io_check()).
 	 */
 	if (ds_pool_is_rebuilding(agg_param->ap_pool_info.api_pool)) {
-		D_INFO(DF_CONT" abort as rebuild started, sp_rebuilding %d\n",
-			DP_CONT(agg_param->ap_pool_info.api_pool_uuid,
-				agg_param->ap_pool_info.api_cont_uuid),
-			agg_param->ap_pool_info.api_pool->sp_rebuilding);
+		D_INFO(DF_CONT " abort as rebuild started, sp_rebuilding %d\n",
+		       DP_CONT(agg_param->ap_pool_info.api_pool_uuid,
+			       agg_param->ap_pool_info.api_cont_uuid),
+		       atomic_load(&agg_param->ap_pool_info.api_pool->sp_rebuilding));
 		return -1;
 	}
 
@@ -2522,9 +2522,9 @@ agg_iterate_pre_cb(daos_handle_t ih, vos_iter_entry_t *entry,
 	}
 
 	if (rc < 0) {
-		D_ERROR(DF_UUID" EC aggregation (rebuilding %d) failed: "DF_RC"\n",
+		D_ERROR(DF_UUID " EC aggregation (rebuilding %d) failed: " DF_RC "\n",
 			DP_UUID(agg_param->ap_pool_info.api_pool->sp_uuid),
-			agg_param->ap_pool_info.api_pool->sp_rebuilding, DP_RC(rc));
+			atomic_load(&agg_param->ap_pool_info.api_pool->sp_rebuilding), DP_RC(rc));
 		return rc;
 	}
 
diff --git a/src/object/srv_internal.h b/src/object/srv_internal.h
@@ -1,5 +1,6 @@
 /**
  * (C) Copyright 2016-2024 Intel Corporation.
+ * (C) Copyright 2025 Hewlett Packard Enterprise Development LP
  *
  * SPDX-License-Identifier: BSD-2-Clause-Patent
  */
@@ -72,12 +73,8 @@ struct migrate_pool_tls {
 	/* The ULT number on each target xstream, which actually refer
 	 * back to the item within mpt_obj/dkey_ult_cnts array.
 	 */
-	ATOMIC uint32_t		*mpt_tgt_obj_ult_cnt;
-	ATOMIC uint32_t		*mpt_tgt_dkey_ult_cnt;
-
-	/* ULT count array from all targets, obj: enumeration, dkey:fetch/update */
-	ATOMIC uint32_t		*mpt_obj_ult_cnts;
-	ATOMIC uint32_t		*mpt_dkey_ult_cnts;
+	uint32_t                 mpt_tgt_obj_ult_cnt;
+	uint32_t                 mpt_tgt_dkey_ult_cnt;
 
 	/* reference count for the structure */
 	uint64_t		mpt_refcount;
@@ -92,16 +89,11 @@ struct migrate_pool_tls {
 	uint32_t		mpt_inflight_max_ult;
 	uint32_t		mpt_opc;
 
-	ABT_cond		mpt_init_cond;
-	ABT_mutex		mpt_init_mutex;
-
 	/* The new layout version for upgrade job */
 	uint32_t		mpt_new_layout_ver;
 
 	/* migrate leader ULT */
-	unsigned int		mpt_ult_running:1,
-				mpt_init_tls:1,
-				mpt_fini:1;
+	unsigned int             mpt_ult_running : 1, mpt_fini : 1;
 
 	/* migration init error */
 	int			mpt_init_err;
diff --git a/src/object/srv_obj.c b/src/object/srv_obj.c
@@ -2425,17 +2425,17 @@ obj_inflight_io_check(struct ds_cont_child *child, uint32_t opc,
 	if (opc == DAOS_OBJ_RPC_ENUMERATE && flags & ORF_FOR_MIGRATION) {
 		/* EC aggregation is still inflight, rebuild should wait until it's paused */
 		if (ds_cont_child_ec_aggregating(child)) {
-			D_ERROR(DF_CONT" ec aggregate still active, rebuilding %d\n",
+			D_ERROR(DF_CONT " ec aggregate still active, rebuilding %d\n",
 				DP_CONT(child->sc_pool->spc_uuid, child->sc_uuid),
-				child->sc_pool->spc_pool->sp_rebuilding);
+				atomic_load(&child->sc_pool->spc_pool->sp_rebuilding));
 			return -DER_UPDATE_AGAIN;
 		}
 	}
 
 	if (!obj_is_modification_opc(opc) && (opc != DAOS_OBJ_RPC_CPD || flags & ORF_CPD_RDONLY))
 		return 0;
 
-	if (child->sc_pool->spc_pool->sp_rebuilding) {
+	if (atomic_load(&child->sc_pool->spc_pool->sp_rebuilding)) {
 		uint32_t version;
 
 		ds_rebuild_running_query(child->sc_pool_uuid, RB_OP_REBUILD,
diff --git a/src/object/srv_obj_migrate.c b/src/object/srv_obj_migrate.c
diff --git a/src/pool/srv_target.c b/src/pool/srv_target.c
diff --git a/src/pool/srv_util.c b/src/pool/srv_util.c
diff --git a/src/rebuild/scan.c b/src/rebuild/scan.c
diff --git a/src/rebuild/srv.c b/src/rebuild/srv.c
diff --git a/src/tests/ftest/ior/intercept_multi_client.yaml b/src/tests/ftest/ior/intercept_multi_client.yaml
diff --git a/src/tests/ftest/util/ior_intercept_test_base.py b/src/tests/ftest/util/ior_intercept_test_base.py
diff --git a/utils/rpms/daos.spec b/utils/rpms/daos.spec

Original file line number	Diff line number	Diff line change
`@@ -69,6 +69,7 @@ typedef struct dav_obj {`
`69`	`69`	`static inline`
`70`	`70`	`struct dav_tx utx2wtx(struct umem_wal_tx utx)`
`71`	`71`	`{`
	`72`	`+ D_ASSERT(utx != NULL);`
`72`	`73`	`return (struct dav_tx *)&utx->utx_private;`
`73`	`74`	`}`
`74`	`75`