sleepable uprobe support

.github/actions/vmtest/action.yml

@@ -0,0 +1,35 @@
+name: 'vmtest'
+description: 'Build + run vmtest'
+inputs:
+  arch:
+    description: 'what arch to test'
+    required: true
+    default: 'x86_64'
+runs:
+  using: "composite"
+  steps:
+    # 1. Setup environment
+    - name: Setup build environment
+      uses: libbpf/ci/setup-build-env@master
+    # 2. Build
+    - name: Build kernel image
+      shell: bash
+      run: ${GITHUB_ACTION_PATH}/build.sh ${{ inputs.arch }}
+    - name: Build selftests
+      shell: bash
+      run: ${GITHUB_ACTION_PATH}/build_selftests.sh
+      env:
+        VMLINUX_BTF: ${{ github.workspace }}/vmlinux
+    # 3. Test
+    - name: Prepare rootfs
+      uses: libbpf/ci/prepare-rootfs@master
+      with:
+        project-name: 'libbpf'
+        arch: ${{ inputs.arch }}
+        kernel-root: '.'
+    - name: Run selftests
+      uses: libbpf/ci/run-qemu@master
+      with:
+        arch: ${{ inputs.arch}}
+        img: '/tmp/root.img'
+        vmlinuz: '${{ github.workspace }}/vmlinuz'

.github/actions/vmtest/build.sh

@@ -0,0 +1,17 @@
+#!/bin/bash
+
+set -euo pipefail
+
+ARCH="$1"
+
+THISDIR="$(cd $(dirname $0) && pwd)"
+
+source "${THISDIR}"/helpers.sh
+
+travis_fold start build_kernel "Building kernel"
+
+cp ${GITHUB_WORKSPACE}/travis-ci/vmtest/configs/config-latest.${ARCH} .config
+
+make -j $((4*$(nproc))) olddefconfig all > /dev/null
+
+travis_fold end build_kernel

.github/actions/vmtest/build_selftests.sh

@@ -0,0 +1,42 @@
+#!/bin/bash
+
+set -euo pipefail
+
+THISDIR="$(cd $(dirname $0) && pwd)"
+
+source "${THISDIR}"/helpers.sh
+
+travis_fold start prepare_selftests "Building selftests"
+
+LLVM_VER=15
+LIBBPF_PATH="${REPO_ROOT}"
+
+PREPARE_SELFTESTS_SCRIPT=${THISDIR}/prepare_selftests-${KERNEL}.sh
+if [ -f "${PREPARE_SELFTESTS_SCRIPT}" ]; then
+	(cd "${REPO_ROOT}/${REPO_PATH}/tools/testing/selftests/bpf" && ${PREPARE_SELFTESTS_SCRIPT})
+fi
+
+if [[ "${KERNEL}" = 'LATEST' ]]; then
+	VMLINUX_H=
+else
+	VMLINUX_H=${THISDIR}/vmlinux.h
+fi
+
+cd ${REPO_ROOT}/${REPO_PATH}
+make \
+	CLANG=clang-${LLVM_VER} \
+	LLC=llc-${LLVM_VER} \
+	LLVM_STRIP=llvm-strip-${LLVM_VER} \
+	VMLINUX_BTF="${VMLINUX_BTF}" \
+	VMLINUX_H="${VMLINUX_H}" \
+	-C "${REPO_ROOT}/${REPO_PATH}/tools/testing/selftests/bpf" \
+	-j $((4*$(nproc))) > /dev/null
+cd -
+mkdir "${LIBBPF_PATH}"/selftests
+cp -R "${REPO_ROOT}/${REPO_PATH}/tools/testing/selftests/bpf" \
+	"${LIBBPF_PATH}"/selftests
+cd "${LIBBPF_PATH}"
+rm selftests/bpf/.gitignore
+git add selftests
+
+travis_fold end prepare_selftests

.github/actions/vmtest/helpers.sh

@@ -0,0 +1,44 @@
+# $1 - start or end
+# $2 - fold identifier, no spaces
+# $3 - fold section description
+travis_fold() {
+  local YELLOW='\033[1;33m'
+  local NOCOLOR='\033[0m'
+  if [ -z ${GITHUB_WORKFLOW+x} ]; then
+    echo travis_fold:$1:$2
+    if [ ! -z "${3:-}" ]; then
+      echo -e "${YELLOW}$3${NOCOLOR}"
+    fi
+    echo
+  else
+    if [ $1 = "start" ]; then
+      line="::group::$2"
+      if [ ! -z "${3:-}" ]; then
+        line="$line - ${YELLOW}$3${NOCOLOR}"
+      fi
+    else
+      line="::endgroup::"
+    fi
+    echo -e "$line"
+  fi
+}
+
+__print() {
+  local TITLE=""
+  if [[ -n $2 ]]; then
+      TITLE=" title=$2"
+  fi
+  echo "::$1${TITLE}::$3"
+}
+
+# $1 - title
+# $2 - message
+print_error() {
+  __print error $1 $2
+}
+
+# $1 - title
+# $2 - message
+print_notice() {
+  __print notice $1 $2
+}

.github/workflows/test.yml

@@ -0,0 +1,50 @@
+name: bpf-ci
+
+on:
+  pull_request:
+
+concurrency: 
+  group: ci-test-${{ github.head_ref }}
+  cancel-in-progress: true
+
+jobs:
+  VM_Test:
+    runs-on: ${{ matrix.runs_on }}
+    name: Kernel ${{ matrix.kernel }} on ${{ matrix.runs_on }} + selftests
+    timeout-minutes: 100
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          - kernel: 'LATEST'
+            runs_on: ubuntu-latest
+            arch: 'x86_64'
+          - kernel: 'LATEST'
+            runs_on: z15
+            arch: 's390x'
+    env:
+      AUTHOR_EMAIL: "$(git log -1 --pretty=\"%aE\")"
+      KERNEL: LATEST
+      REPO_ROOT: ${{ github.workspace }}
+      REPO_PATH: ""
+    steps:
+      - uses: actions/checkout@v2
+      - if: ${{ github.repository != 'kernel-patches/bpf' && github.repository != 'kernel-patches/bpf-rc' }}
+        name: Download bpf-next tree
+        uses: libbpf/ci/get-linux-source@master
+        with:
+          dest: '.kernel'
+      - if: ${{ github.repository != 'kernel-patches/bpf' && github.repository != 'kernel-patches/bpf-rc' }}
+        name: Move linux source in place
+        shell: bash
+        run: |
+          rm -rf .kernel/.git
+          cp -rf .kernel/. .
+          rm -rf .kernel
+      - uses: libbpf/ci/patch-kernel@master
+        with:
+          patches-root: '${{ github.workspace }}/travis-ci/diffs'
+          repo-root: '${{ github.workspace }}'
+      - uses: ./.github/actions/vmtest
+        with:
+          arch: ${{ matrix.arch }}

README

@@ -1,18 +0,0 @@
-Linux kernel
-============
-
-There are several guides for kernel developers and users. These guides can
-be rendered in a number of formats, like HTML and PDF. Please read
-Documentation/admin-guide/README.rst first.
-
-In order to build the documentation, use ``make htmldocs`` or
-``make pdfdocs``.  The formatted documentation can also be read online at:
-
-    https://www.kernel.org/doc/html/latest/
-
-There are various text files in the Documentation/ subdirectory,
-several of them using the Restructured Text markup notation.
-
-Please read the Documentation/process/changes.rst file, as it contains the
-requirements for building and running the kernel, and information about
-the problems which may result by upgrading your kernel.

include/linux/bpf.h

@@ -5,6 +5,7 @@
 #define _LINUX_BPF_H 1
 
 #include <uapi/linux/bpf.h>
+#include <uapi/linux/filter.h>
 
 #include <linux/workqueue.h>
 #include <linux/file.h>
@@ -22,8 +23,10 @@
 #include <linux/sched/mm.h>
 #include <linux/slab.h>
 #include <linux/percpu-refcount.h>
+#include <linux/stddef.h>
 #include <linux/bpfptr.h>
 #include <linux/btf.h>
+#include <linux/rcupdate_trace.h>
 
 struct bpf_verifier_env;
 struct bpf_verifier_log;
@@ -1068,6 +1071,40 @@ struct bpf_prog_aux {
 	};
 };
 
+struct bpf_prog {
+	u16			pages;		/* Number of allocated pages */
+	u16			jited:1,	/* Is our filter JIT'ed? */
+				jit_requested:1,/* archs need to JIT the prog */
+				gpl_compatible:1, /* Is filter GPL compatible? */
+				cb_access:1,	/* Is control block accessed? */
+				dst_needed:1,	/* Do we need dst entry? */
+				blinding_requested:1, /* needs constant blinding */
+				blinded:1,	/* Was blinded */
+				is_func:1,	/* program is a bpf function */
+				kprobe_override:1, /* Do we override a kprobe? */
+				has_callchain_buf:1, /* callchain buffer allocated? */
+				enforce_expected_attach_type:1, /* Enforce expected_attach_type checking at attach time */
+				call_get_stack:1, /* Do we call bpf_get_stack() or bpf_get_stackid() */
+				call_get_func_ip:1, /* Do we call get_func_ip() */
+				tstamp_type_access:1; /* Accessed __sk_buff->tstamp_type */
+	enum bpf_prog_type	type;		/* Type of BPF program */
+	enum bpf_attach_type	expected_attach_type; /* For some prog types */
+	u32			len;		/* Number of filter blocks */
+	u32			jited_len;	/* Size of jited insns in bytes */
+	u8			tag[BPF_TAG_SIZE];
+	struct bpf_prog_stats __percpu *stats;
+	int __percpu		*active;
+	unsigned int		(*bpf_func)(const void *ctx,
+					    const struct bpf_insn *insn);
+	struct bpf_prog_aux	*aux;		/* Auxiliary fields */
+	struct sock_fprog_kern	*orig_prog;	/* Original BPF program */
+	/* Instructions for interpreter */
+	union {
+		DECLARE_FLEX_ARRAY(struct sock_filter, insns);
+		DECLARE_FLEX_ARRAY(struct bpf_insn, insnsi);
+	};
+};
+
 struct bpf_array_aux {
 	/* Programs with direct jumps into programs part of this array. */
 	struct list_head poke_progs;
@@ -1307,6 +1344,8 @@ extern struct bpf_empty_prog_array bpf_empty_prog_array;
 
 struct bpf_prog_array *bpf_prog_array_alloc(u32 prog_cnt, gfp_t flags);
 void bpf_prog_array_free(struct bpf_prog_array *progs);
+/* Use when traversal over the bpf_prog_array uses tasks_trace rcu */
+void bpf_prog_array_free_sleepable(struct bpf_prog_array *progs);
 int bpf_prog_array_length(struct bpf_prog_array *progs);
 bool bpf_prog_array_is_empty(struct bpf_prog_array *array);
 int bpf_prog_array_copy_to_user(struct bpf_prog_array *progs,
@@ -1392,6 +1431,60 @@ bpf_prog_run_array(const struct bpf_prog_array *array,
 	return ret;
 }
 
+/**
+ * Notes on RCU design for bpf_prog_arrays containing sleepable programs:
+ *
+ * We use the tasks_trace rcu flavor read section to protect the bpf_prog_array
+ * overall. As a result, we must use the bpf_prog_array_free_sleepable
+ * in order to use the tasks_trace rcu grace period.
+ *
+ * When a non-sleepable program is inside the array, we take the rcu read
+ * section and disable preemption for that program alone, so it can access
+ * rcu-protected dynamically sized maps.
+ */
+static __always_inline u32
+bpf_prog_run_array_sleepable(const struct bpf_prog_array __rcu *array_rcu,
+			     const void *ctx, bpf_prog_run_fn run_prog)
+{
+	const struct bpf_prog_array_item *item;
+	const struct bpf_prog *prog;
+	const struct bpf_prog_array *array;
+	struct bpf_run_ctx *old_run_ctx;
+	struct bpf_trace_run_ctx run_ctx;
+	u32 ret = 1;
+
+	might_fault();
+
+	migrate_disable();
+	rcu_read_lock_trace();
+
+	array = rcu_dereference_check(array_rcu, rcu_read_lock_trace_held());
+	if (unlikely(!array))
+		goto out;
+	old_run_ctx = bpf_set_run_ctx(&run_ctx.run_ctx);
+	item = &array->items[0];
+	while ((prog = READ_ONCE(item->prog))) {
+		if (!prog->aux->sleepable) {
+			preempt_disable();
+			rcu_read_lock();
+		}
+
+		run_ctx.bpf_cookie = item->bpf_cookie;
+		ret &= run_prog(prog, ctx);
+		item++;
+
+		if (!prog->aux->sleepable) {
+			rcu_read_unlock();
+			preempt_enable();
+		}
+	}
+	bpf_reset_run_ctx(old_run_ctx);
+out:
+	rcu_read_unlock_trace();
+	migrate_enable();
+	return ret;
+}
+
 #ifdef CONFIG_BPF_SYSCALL
 DECLARE_PER_CPU(int, bpf_prog_active);
 extern struct mutex bpf_stats_enabled_mutex;

include/linux/filter.h

@@ -559,40 +559,6 @@ struct bpf_prog_stats {
 	struct u64_stats_sync syncp;
 } __aligned(2 * sizeof(u64));
 
-struct bpf_prog {
-	u16			pages;		/* Number of allocated pages */
-	u16			jited:1,	/* Is our filter JIT'ed? */
-				jit_requested:1,/* archs need to JIT the prog */
-				gpl_compatible:1, /* Is filter GPL compatible? */
-				cb_access:1,	/* Is control block accessed? */
-				dst_needed:1,	/* Do we need dst entry? */
-				blinding_requested:1, /* needs constant blinding */
-				blinded:1,	/* Was blinded */
-				is_func:1,	/* program is a bpf function */
-				kprobe_override:1, /* Do we override a kprobe? */
-				has_callchain_buf:1, /* callchain buffer allocated? */
-				enforce_expected_attach_type:1, /* Enforce expected_attach_type checking at attach time */
-				call_get_stack:1, /* Do we call bpf_get_stack() or bpf_get_stackid() */
-				call_get_func_ip:1, /* Do we call get_func_ip() */
-				tstamp_type_access:1; /* Accessed __sk_buff->tstamp_type */
-	enum bpf_prog_type	type;		/* Type of BPF program */
-	enum bpf_attach_type	expected_attach_type; /* For some prog types */
-	u32			len;		/* Number of filter blocks */
-	u32			jited_len;	/* Size of jited insns in bytes */
-	u8			tag[BPF_TAG_SIZE];
-	struct bpf_prog_stats __percpu *stats;
-	int __percpu		*active;
-	unsigned int		(*bpf_func)(const void *ctx,
-					    const struct bpf_insn *insn);
-	struct bpf_prog_aux	*aux;		/* Auxiliary fields */
-	struct sock_fprog_kern	*orig_prog;	/* Original BPF program */
-	/* Instructions for interpreter */
-	union {
-		DECLARE_FLEX_ARRAY(struct sock_filter, insns);
-		DECLARE_FLEX_ARRAY(struct bpf_insn, insnsi);
-	};
-};
-
 struct sk_filter {
 	refcount_t	refcnt;
 	struct rcu_head	rcu;

include/linux/trace_events.h

@@ -736,6 +736,7 @@ trace_trigger_soft_disabled(struct trace_event_file *file)
 
 #ifdef CONFIG_BPF_EVENTS
 unsigned int trace_call_bpf(struct trace_event_call *call, void *ctx);
+unsigned int uprobe_call_bpf(struct trace_event_call *call, void *ctx);
 int perf_event_attach_bpf_prog(struct perf_event *event, struct bpf_prog *prog, u64 bpf_cookie);
 void perf_event_detach_bpf_prog(struct perf_event *event);
 int perf_event_query_prog_array(struct perf_event *event, void __user *info);