fixup for 5c87b6864

Duplicate test_relocation.toml into a variant with removeall.
2025-10-31 16:14:16 +01:00 · 2021-06-13 21:40:36 -05:00 · 2021-06-13 21:39:56 -05:00 · 2021-06-13 21:38:37 -05:00 · 2021-06-13 21:11:10 -05:00 · 2021-06-13 21:10:25 -05:00
14 changed files with 2040 additions and 638 deletions
--- a/.github/workflows/post-release.yml
+++ b/.github/workflows/post-release.yml
@@ -0,0 +1,26 @@
+name: post-release
+on:
+  release:
+    branches: [master]
+    types: [released]
+
+jobs:
+  post-release:
+    runs-on: ubuntu-18.04
+    steps:
+      # trigger post-release in dependency repo, this indirection allows the
+      # dependency repo to be updated often without affecting this repo. At
+      # the time of this comment, the dependency repo is responsible for
+      # creating PRs for other dependent repos post-release.
+      - name: trigger-post-release
+        continue-on-error: true
+        run: |
+          curl -sS -X POST -H "authorization: token ${{secrets.BOT_TOKEN}}" \
+            "$GITHUB_API_URL/repos/${{secrets.POST_RELEASE_REPO}}/dispatches" \
+            -d "$(jq -n '{
+              event_type: "post-release",
+              client_payload: {
+                repo: env.GITHUB_REPOSITORY,
+                version: "${{github.event.release.tag_name}}"}}' \
+              | tee /dev/stderr)"
+
--- a/.github/workflows/release.yml
+++ b/.github/workflows/release.yml
@@ -0,0 +1,215 @@
+name: release
+on:
+  workflow_run:
+    workflows: [test]
+    branches: [master]
+    types: [completed]
+
+jobs:
+  release:
+    runs-on: ubuntu-18.04
+
+    # need to manually check for a couple things
+    # - tests passed?
+    # - we are the most recent commit on master?
+    if: ${{github.event.workflow_run.conclusion == 'success' &&
+      github.event.workflow_run.head_sha == github.sha}}
+
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          ref: ${{github.event.workflow_run.head_sha}}
+          # need workflow access since we push branches
+          # containing workflows
+          token: ${{secrets.BOT_TOKEN}}
+          # need all tags
+          fetch-depth: 0
+
+      # try to get results from tests
+      - uses: dawidd6/action-download-artifact@v2
+        continue-on-error: true
+        with:
+          workflow: ${{github.event.workflow_run.name}}
+          run_id: ${{github.event.workflow_run.id}}
+          name: results
+          path: results
+
+      - name: find-version
+        run: |
+          # rip version from lfs.h
+          LFS_VERSION="$(grep -o '^#define LFS_VERSION .*$' lfs.h \
+            | awk '{print $3}')"
+          LFS_VERSION_MAJOR="$((0xffff & ($LFS_VERSION >> 16)))"
+          LFS_VERSION_MINOR="$((0xffff & ($LFS_VERSION >>  0)))"
+
+          # find a new patch version based on what we find in our tags
+          LFS_VERSION_PATCH="$( \
+            ( git describe --tags --abbrev=0 \
+                --match="v$LFS_VERSION_MAJOR.$LFS_VERSION_MINOR.*" \
+              || echo 'v0.0.-1' ) \
+            | awk -F '.' '{print $3+1}')"
+
+          # found new version
+          LFS_VERSION="v$LFS_VERSION_MAJOR`
+            `.$LFS_VERSION_MINOR`
+            `.$LFS_VERSION_PATCH"
+          echo "LFS_VERSION=$LFS_VERSION"
+          echo "LFS_VERSION=$LFS_VERSION" >> $GITHUB_ENV
+          echo "LFS_VERSION_MAJOR=$LFS_VERSION_MAJOR" >> $GITHUB_ENV
+          echo "LFS_VERSION_MINOR=$LFS_VERSION_MINOR" >> $GITHUB_ENV
+          echo "LFS_VERSION_PATCH=$LFS_VERSION_PATCH" >> $GITHUB_ENV
+
+      # try to find previous version?
+      - name: find-prev-version
+        continue-on-error: true
+        run: |
+          LFS_PREV_VERSION="$(git describe --tags --abbrev=0 --match 'v*')"
+          echo "LFS_PREV_VERSION=$LFS_PREV_VERSION"
+          echo "LFS_PREV_VERSION=$LFS_PREV_VERSION" >> $GITHUB_ENV
+
+      # try to find results from tests
+      - name: collect-results
+        run: |
+          # previous results to compare against?
+          [ -n "$LFS_PREV_VERSION" ] && curl -sS \
+            "$GITHUB_API_URL/repos/$GITHUB_REPOSITORY/`
+              `status/$LFS_PREV_VERSION" \
+            | jq -re 'select(.sha != env.GITHUB_SHA) | .statuses[]' \
+            >> prev-results.json \
+            || true
+
+          # unfortunately these each have their own format
+          [ -e results/code-thumb.csv ] && ( \
+            export PREV="$(jq -re '
+                  select(.context == "results / code").description
+                  | capture("Code size is (?<result>[0-9]+)").result' \
+                prev-results.json || echo 0)"
+            ./scripts/code.py -u results/code-thumb.csv -s | awk '
+              NR==2 {printf "Code size,%d B",$2}
+              NR==2 && ENVIRON["PREV"]+0 != 0 {
+                printf " (%+.1f%%)",100*($2-ENVIRON["PREV"])/ENVIRON["PREV"]}
+              NR==2 {printf "\n"}' \
+            >> results.csv)
+          [ -e results/code-thumb-readonly.csv ] && ( \
+            export PREV="$(jq -re '
+                  select(.context == "results / code (readonly)").description
+                  | capture("Code size is (?<result>[0-9]+)").result' \
+                prev-results.json || echo 0)"
+            ./scripts/code.py -u results/code-thumb-readonly.csv -s | awk '
+              NR==2 {printf "Code size<br/>(readonly),%d B",$2}
+              NR==2 && ENVIRON["PREV"]+0 != 0 {
+                printf " (%+.1f%%)",100*($2-ENVIRON["PREV"])/ENVIRON["PREV"]}
+              NR==2 {printf "\n"}' \
+            >> results.csv)
+          [ -e results/code-thumb-threadsafe.csv ] && ( \
+            export PREV="$(jq -re '
+                  select(.context == "results / code (threadsafe)").description
+                  | capture("Code size is (?<result>[0-9]+)").result' \
+                prev-results.json || echo 0)"
+            ./scripts/code.py -u results/code-thumb-threadsafe.csv -s | awk '
+              NR==2 {printf "Code size<br/>(threadsafe),%d B",$2}
+              NR==2 && ENVIRON["PREV"]+0 != 0 {
+                printf " (%+.1f%%)",100*($2-ENVIRON["PREV"])/ENVIRON["PREV"]}
+              NR==2 {printf "\n"}' \
+            >> results.csv)
+          [ -e results/code-thumb-migrate.csv ] && ( \
+            export PREV="$(jq -re '
+                  select(.context == "results / code (migrate)").description
+                  | capture("Code size is (?<result>[0-9]+)").result' \
+                prev-results.json || echo 0)"
+            ./scripts/code.py -u results/code-thumb-migrate.csv -s | awk '
+              NR==2 {printf "Code size<br/>(migrate),%d B",$2}
+              NR==2 && ENVIRON["PREV"]+0 != 0 {
+                printf " (%+.1f%%)",100*($2-ENVIRON["PREV"])/ENVIRON["PREV"]}
+              NR==2 {printf "\n"}' \
+            >> results.csv)
+          [ -e results/code-thumb-error-asserts.csv ] && ( \
+            export PREV="$(jq -re '
+                  select(.context == "results / code (error-asserts)").description
+                  | capture("Code size is (?<result>[0-9]+)").result' \
+                prev-results.json || echo 0)"
+            ./scripts/code.py -u results/code-thumb-error-asserts.csv -s | awk '
+              NR==2 {printf "Code size<br/>(error-asserts),%d B",$2}
+              NR==2 && ENVIRON["PREV"]+0 != 0 {
+                printf " (%+.1f%%)",100*($2-ENVIRON["PREV"])/ENVIRON["PREV"]}
+              NR==2 {printf "\n"}' \
+            >> results.csv)
+          [ -e results/coverage.csv ] && ( \
+            export PREV="$(jq -re '
+                  select(.context == "results / coverage").description
+                  | capture("Coverage is (?<result>[0-9\\.]+)").result' \
+                prev-results.json || echo 0)"
+            ./scripts/coverage.py -u results/coverage.csv -s | awk -F '[ /%]+' '
+              NR==2 {printf "Coverage,%.1f%% of %d lines",$4,$3}
+              NR==2 && ENVIRON["PREV"]+0 != 0 {
+                printf " (%+.1f%%)",$4-ENVIRON["PREV"]}
+              NR==2 {printf "\n"}' \
+            >> results.csv)
+
+          # transpose to GitHub table
+          [ -e results.csv ] || exit 0
+          awk -F ',' '
+            {label[NR]=$1; value[NR]=$2}
+            END {
+              for (r=1; r<=NR; r++) {printf "| %s ",label[r]}; printf "|\n";
+              for (r=1; r<=NR; r++) {printf "|:--"}; printf "|\n";
+              for (r=1; r<=NR; r++) {printf "| %s ",value[r]}; printf "|\n"}' \
+            results.csv > results.txt
+          echo "RESULTS:"
+          cat results.txt
+
+      # find changes from history
+      - name: collect-changes
+        run: |
+          [ -n "$LFS_PREV_VERSION" ] || exit 0
+          # use explicit link to github commit so that release notes can
+          # be copied elsewhere
+          git log "$LFS_PREV_VERSION.." \
+            --grep='^Merge' --invert-grep \
+            --format="format:[\`%h\`](`
+              `https://github.com/$GITHUB_REPOSITORY/commit/%h) %s" \
+            > changes.txt
+          echo "CHANGES:"
+          cat changes.txt
+
+      # create and update major branches (vN and vN-prefix)
+      - name: create-major-branches
+        run: |
+          # create major branch
+          git branch "v$LFS_VERSION_MAJOR" HEAD
+
+          # create major prefix branch
+          git config user.name ${{secrets.BOT_USER}}
+          git config user.email ${{secrets.BOT_EMAIL}}
+          git fetch "https://github.com/$GITHUB_REPOSITORY.git" \
+            "v$LFS_VERSION_MAJOR-prefix" || true
+          ./scripts/prefix.py "lfs$LFS_VERSION_MAJOR"
+          git branch "v$LFS_VERSION_MAJOR-prefix" $( \
+            git commit-tree $(git write-tree) \
+              $(git rev-parse --verify -q FETCH_HEAD | sed -e 's/^/-p /') \
+              -p HEAD \
+              -m "Generated v$LFS_VERSION_MAJOR prefixes")
+          git reset --hard
+
+          # push!
+          git push --atomic origin \
+            "v$LFS_VERSION_MAJOR" \
+            "v$LFS_VERSION_MAJOR-prefix"
+
+      # build release notes
+      - name: create-release
+        run: |
+          # create release and patch version tag (vN.N.N)
+          # only draft if not a patch release
+          [ -e results.txt ] && export RESULTS="$(cat results.txt)"
+          [ -e changes.txt ] && export CHANGES="$(cat changes.txt)"
+          curl -sS -X POST -H "authorization: token ${{secrets.BOT_TOKEN}}" \
+            "$GITHUB_API_URL/repos/$GITHUB_REPOSITORY/releases" \
+            -d "$(jq -n '{
+              tag_name: env.LFS_VERSION,
+              name: env.LFS_VERSION | rtrimstr(".0"),
+              target_commitish: "${{github.event.workflow_run.head_sha}}",
+              draft: env.LFS_VERSION | endswith(".0"),
+              body: [env.RESULTS, env.CHANGES | select(.)] | join("\n\n")}' \
+              | tee /dev/stderr)"
+
--- a/.github/workflows/status.yml
+++ b/.github/workflows/status.yml
@@ -0,0 +1,55 @@
+name: status
+on:
+  workflow_run:
+    workflows: [test]
+    types: [completed]
+
+jobs:
+  status:
+    runs-on: ubuntu-18.04
+    steps:
+      # custom statuses?
+      - uses: dawidd6/action-download-artifact@v2
+        continue-on-error: true
+        with:
+          workflow: ${{github.event.workflow_run.name}}
+          run_id: ${{github.event.workflow_run.id}}
+          name: status
+          path: status
+      - name: update-status
+        continue-on-error: true
+        run: |
+          ls status
+          for s in $(shopt -s nullglob ; echo status/*.json)
+          do
+            # parse requested status
+            export STATE="$(jq -er '.state' $s)"
+            export CONTEXT="$(jq -er '.context' $s)"
+            export DESCRIPTION="$(jq -er '.description' $s)"
+            # help lookup URL for job/steps because GitHub makes
+            # it VERY HARD to link to specific jobs
+            export TARGET_URL="$(
+              jq -er '.target_url // empty' $s || (
+                export TARGET_JOB="$(jq -er '.target_job' $s)"
+                export TARGET_STEP="$(jq -er '.target_step // ""' $s)"
+                curl -sS -H "authorization: token ${{secrets.BOT_TOKEN}}" \
+                  "$GITHUB_API_URL/repos/$GITHUB_REPOSITORY/actions/runs/`
+                    `${{github.event.workflow_run.id}}/jobs" \
+                  | jq -er '.jobs[]
+                    | select(.name == env.TARGET_JOB)
+                    | .html_url
+                      + "?check_suite_focus=true"
+                      + ((.steps[]
+                        | select(.name == env.TARGET_STEP)
+                        | "#step:\(.number):0") // "")'))"
+            # update status
+            curl -sS -X POST -H "authorization: token ${{secrets.BOT_TOKEN}}" \
+              "$GITHUB_API_URL/repos/$GITHUB_REPOSITORY/statuses/`
+                `${{github.event.workflow_run.head_sha}}" \
+              -d "$(jq -n '{
+                state: env.STATE,
+                context: env.CONTEXT,
+                description: env.DESCRIPTION,
+                target_url: env.TARGET_URL}' \
+                | tee /dev/stderr)"
+          done
--- a/.github/workflows/test.yml
+++ b/.github/workflows/test.yml
@@ -0,0 +1,446 @@
+name: test
+on: [push, pull_request]
+
+env:
+  CFLAGS: -Werror
+  MAKEFLAGS: -j
+
+jobs:
+  # run tests
+  test:
+    runs-on: ubuntu-18.04
+    strategy:
+      fail-fast: false
+      matrix:
+        arch: [x86_64, thumb, mips, powerpc]
+
+    steps:
+      - uses: actions/checkout@v2
+      - name: install
+        run: |
+          # need toml, also pip3 isn't installed by default?
+          sudo apt-get update -qq
+          sudo apt-get install -qq python3 python3-pip lcov
+          sudo pip3 install toml
+          gcc --version
+
+          # setup a ram-backed disk to speed up reentrant tests
+          mkdir disks
+          sudo mount -t tmpfs -o size=100m tmpfs disks
+          TESTFLAGS="$TESTFLAGS --disk=disks/disk"
+
+          # collect coverage
+          mkdir -p coverage
+          TESTFLAGS="$TESTFLAGS --coverage=`
+            `coverage/${{github.job}}-${{matrix.arch}}.info"
+
+          echo "TESTFLAGS=$TESTFLAGS" >> $GITHUB_ENV
+
+      # cross-compile with ARM Thumb (32-bit, little-endian)
+      - name: install-thumb
+        if: ${{matrix.arch == 'thumb'}}
+        run: |
+          sudo apt-get install -qq \
+            gcc-arm-linux-gnueabi \
+            libc6-dev-armel-cross \
+            qemu-user
+          echo "CC=arm-linux-gnueabi-gcc -mthumb --static" >> $GITHUB_ENV
+          echo "EXEC=qemu-arm" >> $GITHUB_ENV
+          arm-linux-gnueabi-gcc --version
+          qemu-arm -version
+      # cross-compile with MIPS (32-bit, big-endian)
+      - name: install-mips
+        if: ${{matrix.arch == 'mips'}}
+        run: |
+          sudo apt-get install -qq \
+            gcc-mips-linux-gnu \
+            libc6-dev-mips-cross \
+            qemu-user
+          echo "CC=mips-linux-gnu-gcc --static" >> $GITHUB_ENV
+          echo "EXEC=qemu-mips" >> $GITHUB_ENV
+          mips-linux-gnu-gcc --version
+          qemu-mips -version
+      # cross-compile with PowerPC (32-bit, big-endian)
+      - name: install-powerpc
+        if: ${{matrix.arch == 'powerpc'}}
+        run: |
+          sudo apt-get install -qq \
+            gcc-powerpc-linux-gnu \
+            libc6-dev-powerpc-cross \
+            qemu-user
+          echo "CC=powerpc-linux-gnu-gcc --static" >> $GITHUB_ENV
+          echo "EXEC=qemu-ppc" >> $GITHUB_ENV
+          powerpc-linux-gnu-gcc --version
+          qemu-ppc -version
+
+      # make sure example can at least compile
+      - name: test-example
+        run: |
+          sed -n '/``` c/,/```/{/```/d; p}' README.md > test.c
+          make all CFLAGS+=" \
+            -Duser_provided_block_device_read=NULL \
+            -Duser_provided_block_device_prog=NULL \
+            -Duser_provided_block_device_erase=NULL \
+            -Duser_provided_block_device_sync=NULL \
+            -include stdio.h"
+          rm test.c
+
+      # test configurations
+      # normal+reentrant tests
+      - name: test-default
+        run: |
+          make clean
+          make test TESTFLAGS+="-nrk"
+      # NOR flash: read/prog = 1 block = 4KiB
+      - name: test-nor
+        run: |
+          make clean
+          make test TESTFLAGS+="-nrk \
+            -DLFS_READ_SIZE=1 -DLFS_BLOCK_SIZE=4096"
+      # SD/eMMC: read/prog = 512 block = 512
+      - name: test-emmc
+        run: |
+          make clean
+          make test TESTFLAGS+="-nrk \
+            -DLFS_READ_SIZE=512 -DLFS_BLOCK_SIZE=512"
+      # NAND flash: read/prog = 4KiB block = 32KiB
+      - name: test-nand
+        run: |
+          make clean
+          make test TESTFLAGS+="-nrk \
+            -DLFS_READ_SIZE=4096 -DLFS_BLOCK_SIZE=\(32*1024\)"
+      # other extreme geometries that are useful for various corner cases
+      - name: test-no-intrinsics
+        run: |
+          make clean
+          make test TESTFLAGS+="-nrk \
+            -DLFS_NO_INTRINSICS"
+      - name: test-byte-writes
+        # it just takes too long to test byte-level writes when in qemu,
+        # should be plenty covered by the other configurations
+        if: ${{matrix.arch == 'x86_64'}}
+        run: |
+          make clean
+          make test TESTFLAGS+="-nrk \
+            -DLFS_READ_SIZE=1 -DLFS_CACHE_SIZE=1"
+      - name: test-block-cycles
+        run: |
+          make clean
+          make test TESTFLAGS+="-nrk \
+            -DLFS_BLOCK_CYCLES=1"
+      - name: test-odd-block-count
+        run: |
+          make clean
+          make test TESTFLAGS+="-nrk \
+            -DLFS_BLOCK_COUNT=1023 -DLFS_LOOKAHEAD_SIZE=256"
+      - name: test-odd-block-size
+        run: |
+          make clean
+          make test TESTFLAGS+="-nrk \
+            -DLFS_READ_SIZE=11 -DLFS_BLOCK_SIZE=704"
+
+      # upload coverage for later coverage
+      - name: upload-coverage
+        uses: actions/upload-artifact@v2
+        with:
+          name: coverage
+          path: coverage
+          retention-days: 1
+
+      # update results
+      - name: results-code
+        run: |
+          mkdir -p results
+          make clean
+          make code \
+            CFLAGS+=" \
+              -DLFS_NO_ASSERT \
+              -DLFS_NO_DEBUG \
+              -DLFS_NO_WARN \
+              -DLFS_NO_ERROR" \
+            CODEFLAGS+="-o results/code-${{matrix.arch}}.csv"
+      - name: results-code-readonly
+        run: |
+          mkdir -p results
+          make clean
+          make code \
+            CFLAGS+=" \
+              -DLFS_NO_ASSERT \
+              -DLFS_NO_DEBUG \
+              -DLFS_NO_WARN \
+              -DLFS_NO_ERROR \
+              -DLFS_READONLY" \
+            CODEFLAGS+="-o results/code-${{matrix.arch}}-readonly.csv"
+      - name: results-code-threadsafe
+        run: |
+          mkdir -p results
+          make clean
+          make code \
+            CFLAGS+=" \
+              -DLFS_NO_ASSERT \
+              -DLFS_NO_DEBUG \
+              -DLFS_NO_WARN \
+              -DLFS_NO_ERROR \
+              -DLFS_THREADSAFE" \
+            CODEFLAGS+="-o results/code-${{matrix.arch}}-threadsafe.csv"
+      - name: results-code-migrate
+        run: |
+          mkdir -p results
+          make clean
+          make code \
+            CFLAGS+=" \
+              -DLFS_NO_ASSERT \
+              -DLFS_NO_DEBUG \
+              -DLFS_NO_WARN \
+              -DLFS_NO_ERROR \
+              -DLFS_MIGRATE" \
+            CODEFLAGS+="-o results/code-${{matrix.arch}}-migrate.csv"
+      - name: results-code-error-asserts
+        run: |
+          mkdir -p results
+          make clean
+          make code \
+            CFLAGS+=" \
+              -DLFS_NO_DEBUG \
+              -DLFS_NO_WARN \
+              -DLFS_NO_ERROR \
+              -D'LFS_ASSERT(test)=do {if(!(test)) {return -1;}} while(0)'" \
+            CODEFLAGS+="-o results/code-${{matrix.arch}}-error-asserts.csv"
+      - name: upload-results
+        uses: actions/upload-artifact@v2
+        with:
+          name: results
+          path: results
+      # limit reporting to Thumb, otherwise there would be too many numbers
+      # flying around for the results to be easily readable
+      - name: collect-status
+        if: ${{matrix.arch == 'thumb'}}
+        run: |
+          mkdir -p status
+          for f in $(shopt -s nullglob ; echo results/code*.csv)
+          do
+            export STEP="results-code$(
+              echo $f | sed -n 's/.*code-.*-\(.*\).csv/-\1/p')"
+            export CONTEXT="results / code$(
+              echo $f | sed -n 's/.*code-.*-\(.*\).csv/ (\1)/p')"
+            export PREV="$(curl -sS \
+              "$GITHUB_API_URL/repos/$GITHUB_REPOSITORY/status/master" \
+              | jq -re 'select(.sha != env.GITHUB_SHA) | .statuses[]
+                | select(.context == env.CONTEXT).description
+                | capture("Code size is (?<result>[0-9]+)").result' \
+              || echo 0)"
+            export DESCRIPTION="$(./scripts/code.py -u $f -s | awk '
+              NR==2 {printf "Code size is %d B",$2}
+              NR==2 && ENVIRON["PREV"]+0 != 0 {
+                printf " (%+.1f%%)",100*($2-ENVIRON["PREV"])/ENVIRON["PREV"]}')"
+            jq -n '{
+              state: "success",
+              context: env.CONTEXT,
+              description: env.DESCRIPTION,
+              target_job: "${{github.job}} (${{matrix.arch}})",
+              target_step: env.STEP}' \
+              | tee status/code$(
+                echo $f | sed -n 's/.*code-.*-\(.*\).csv/-\1/p').json
+          done
+      - name: upload-status
+        if: ${{matrix.arch == 'thumb'}}
+        uses: actions/upload-artifact@v2
+        with:
+          name: status
+          path: status
+          retention-days: 1
+
+  # run under Valgrind to check for memory errors
+  valgrind:
+    runs-on: ubuntu-18.04
+    steps:
+      - uses: actions/checkout@v2
+      - name: install
+        run: |
+          # need toml, also pip3 isn't installed by default?
+          sudo apt-get update -qq
+          sudo apt-get install -qq python3 python3-pip
+          sudo pip3 install toml
+      - name: install-valgrind
+        run: |
+          sudo apt-get update -qq
+          sudo apt-get install -qq valgrind
+          valgrind --version
+      # normal tests, we don't need to test all geometries
+      - name: test-valgrind
+        run: make test TESTFLAGS+="-k --valgrind"
+
+  # self-host with littlefs-fuse for a fuzz-like test
+  fuse:
+    runs-on: ubuntu-18.04
+    if: ${{!endsWith(github.ref, '-prefix')}}
+    steps:
+      - uses: actions/checkout@v2
+      - name: install
+        run: |
+          # need toml, also pip3 isn't installed by default?
+          sudo apt-get update -qq
+          sudo apt-get install -qq python3 python3-pip libfuse-dev
+          sudo pip3 install toml
+          fusermount -V
+          gcc --version
+      - uses: actions/checkout@v2
+        with:
+          repository: littlefs-project/littlefs-fuse
+          ref: v2
+          path: littlefs-fuse
+      - name: setup
+        run: |
+          # copy our new version into littlefs-fuse
+          rm -rf littlefs-fuse/littlefs/*
+          cp -r $(git ls-tree --name-only HEAD) littlefs-fuse/littlefs
+
+          # setup disk for littlefs-fuse
+          mkdir mount
+          sudo chmod a+rw /dev/loop0
+          dd if=/dev/zero bs=512 count=128K of=disk
+          losetup /dev/loop0 disk
+      - name: test
+        run: |
+          # self-host test
+          make -C littlefs-fuse
+
+          littlefs-fuse/lfs --format /dev/loop0
+          littlefs-fuse/lfs /dev/loop0 mount
+
+          ls mount
+          mkdir mount/littlefs
+          cp -r $(git ls-tree --name-only HEAD) mount/littlefs
+          cd mount/littlefs
+          stat .
+          ls -flh
+          make -B test
+
+  # test migration using littlefs-fuse
+  migrate:
+    runs-on: ubuntu-18.04
+    if: ${{!endsWith(github.ref, '-prefix')}}
+    steps:
+      - uses: actions/checkout@v2
+      - name: install
+        run: |
+          # need toml, also pip3 isn't installed by default?
+          sudo apt-get update -qq
+          sudo apt-get install -qq python3 python3-pip libfuse-dev
+          sudo pip3 install toml
+          fusermount -V
+          gcc --version
+      - uses: actions/checkout@v2
+        with:
+          repository: littlefs-project/littlefs-fuse
+          ref: v2
+          path: v2
+      - uses: actions/checkout@v2
+        with:
+          repository: littlefs-project/littlefs-fuse
+          ref: v1
+          path: v1
+      - name: setup
+        run: |
+          # copy our new version into littlefs-fuse
+          rm -rf v2/littlefs/*
+          cp -r $(git ls-tree --name-only HEAD) v2/littlefs
+
+          # setup disk for littlefs-fuse
+          mkdir mount
+          sudo chmod a+rw /dev/loop0
+          dd if=/dev/zero bs=512 count=128K of=disk
+          losetup /dev/loop0 disk
+      - name: test
+        run: |
+          # compile v1 and v2
+          make -C v1
+          make -C v2
+
+          # run self-host test with v1
+          v1/lfs --format /dev/loop0
+          v1/lfs /dev/loop0 mount
+
+          ls mount
+          mkdir mount/littlefs
+          cp -r $(git ls-tree --name-only HEAD) mount/littlefs
+          cd mount/littlefs
+          stat .
+          ls -flh
+          make -B test
+
+          # attempt to migrate
+          cd ../..
+          fusermount -u mount
+
+          v2/lfs --migrate /dev/loop0
+          v2/lfs /dev/loop0 mount
+
+          # run self-host test with v2 right where we left off
+          ls mount
+          cd mount/littlefs
+          stat .
+          ls -flh
+          make -B test
+
+  # collect coverage info
+  coverage:
+    runs-on: ubuntu-18.04
+    needs: [test]
+    steps:
+      - uses: actions/checkout@v2
+      - name: install
+        run: |
+          sudo apt-get update -qq
+          sudo apt-get install -qq python3 python3-pip lcov
+          sudo pip3 install toml
+      # yes we continue-on-error nearly every step, continue-on-error
+      # at job level apparently still marks a job as failed, which isn't
+      # what we want
+      - uses: actions/download-artifact@v2
+        continue-on-error: true
+        with:
+          name: coverage
+          path: coverage
+      - name: results-coverage
+        continue-on-error: true
+        run: |
+          mkdir -p results
+          lcov $(for f in coverage/*.info ; do echo "-a $f" ; done) \
+            -o results/coverage.info
+          ./scripts/coverage.py results/coverage.info -o results/coverage.csv
+      - name: upload-results
+        uses: actions/upload-artifact@v2
+        with:
+          name: results
+          path: results
+      - name: collect-status
+        run: |
+          mkdir -p status
+          [ -e results/coverage.csv ] || exit 0
+          export STEP="results-coverage"
+          export CONTEXT="results / coverage"
+          export PREV="$(curl -sS \
+            "$GITHUB_API_URL/repos/$GITHUB_REPOSITORY/status/master" \
+            | jq -re 'select(.sha != env.GITHUB_SHA) | .statuses[]
+              | select(.context == env.CONTEXT).description
+              | capture("Coverage is (?<result>[0-9\\.]+)").result' \
+            || echo 0)"
+          export DESCRIPTION="$(
+            ./scripts/coverage.py -u results/coverage.csv -s | awk -F '[ /%]+' '
+              NR==2 {printf "Coverage is %.1f%% of %d lines",$4,$3}
+              NR==2 && ENVIRON["PREV"]+0 != 0 {
+                printf " (%+.1f%%)",$4-ENVIRON["PREV"]}')"
+          jq -n '{
+            state: "success",
+            context: env.CONTEXT,
+            description: env.DESCRIPTION,
+            target_job: "${{github.job}}",
+            target_step: env.STEP}' \
+            | tee status/coverage.json
+      - name: upload-status
+        uses: actions/upload-artifact@v2
+        with:
+          name: status
+          path: status
+          retention-days: 1
--- a/.travis.yml
+++ b/.travis.yml
@@ -1,461 +0,0 @@
-# environment variables
-env:
-  global:
-    - CFLAGS=-Werror
-    - MAKEFLAGS=-j
-
-# cache installation dirs
-cache:
-  pip: true
-  directories:
-    - $HOME/.cache/apt
-
-# common installation
-_: &install-common
-  # need toml, also pip3 isn't installed by default?
-  - sudo apt-get install python3 python3-pip
-  - sudo pip3 install toml
-  # setup a ram-backed disk to speed up reentrant tests
-  - mkdir disks
-  - sudo mount -t tmpfs -o size=100m tmpfs disks
-  - export TFLAGS="$TFLAGS --disk=disks/disk"
-
-# test cases
-_: &test-example
-  # make sure example can at least compile
-  - sed -n '/``` c/,/```/{/```/d; p}' README.md > test.c &&
-    make all CFLAGS+="
-        -Duser_provided_block_device_read=NULL
-        -Duser_provided_block_device_prog=NULL
-        -Duser_provided_block_device_erase=NULL
-        -Duser_provided_block_device_sync=NULL
-        -include stdio.h"
-# default tests
-_: &test-default
-  # normal+reentrant tests
-  - make test TFLAGS+="-nrk"
-# common real-life geometries
-_: &test-nor
-  # NOR flash: read/prog = 1 block = 4KiB
-  - make test TFLAGS+="-nrk -DLFS_READ_SIZE=1 -DLFS_BLOCK_SIZE=4096"
-_: &test-emmc
-  # eMMC: read/prog = 512 block = 512
-  - make test TFLAGS+="-nrk -DLFS_READ_SIZE=512 -DLFS_BLOCK_SIZE=512"
-_: &test-nand
-  # NAND flash: read/prog = 4KiB block = 32KiB
-  - make test TFLAGS+="-nrk -DLFS_READ_SIZE=4096 -DLFS_BLOCK_SIZE=\(32*1024\)"
-# other extreme geometries that are useful for testing various corner cases
-_: &test-no-intrinsics
-  - make test TFLAGS+="-nrk -DLFS_NO_INTRINSICS"
-_: &test-no-inline
-  - make test TFLAGS+="-nrk -DLFS_INLINE_MAX=0"
-_: &test-byte-writes
-  - make test TFLAGS+="-nrk -DLFS_READ_SIZE=1 -DLFS_CACHE_SIZE=1"
-_: &test-block-cycles
-  - make test TFLAGS+="-nrk -DLFS_BLOCK_CYCLES=1"
-_: &test-odd-block-count
-  - make test TFLAGS+="-nrk -DLFS_BLOCK_COUNT=1023 -DLFS_LOOKAHEAD_SIZE=256"
-_: &test-odd-block-size
-  - make test TFLAGS+="-nrk -DLFS_READ_SIZE=11 -DLFS_BLOCK_SIZE=704"
-
-# report size 
-_: &report-size
-  # compile and find the code size with the smallest configuration
-  - make -j1 clean size
-        OBJ="$(ls lfs*.c | sed 's/\.c/\.o/' | tr '\n' ' ')"
-        CFLAGS+="-DLFS_NO_ASSERT -DLFS_NO_DEBUG -DLFS_NO_WARN -DLFS_NO_ERROR"
-        | tee sizes
-  # update status if we succeeded, compare with master if possible
-  - |
-    if [ "$TRAVIS_TEST_RESULT" -eq 0 ]
-    then
-        CURR=$(tail -n1 sizes | awk '{print $1}')
-        PREV=$(curl -u "$GEKY_BOT_STATUSES" https://api.github.com/repos/$TRAVIS_REPO_SLUG/status/master \
-            | jq -re "select(.sha != \"$TRAVIS_COMMIT\")
-                | .statuses[] | select(.context == \"${TRAVIS_BUILD_STAGE_NAME,,}/$NAME\").description
-                | capture(\"code size is (?<size>[0-9]+)\").size" \
-            || echo 0)
-  
-        STATUS="Passed, code size is ${CURR}B"
-        if [ "$PREV" -ne 0 ]
-        then
-            STATUS="$STATUS ($(python -c "print '%+.2f' % (100*($CURR-$PREV)/$PREV.0)")%)"
-        fi
-    fi
-
-# stage control
-stages:
-  - name: test
-  - name: deploy
-    if: branch = master AND type = push
-
-# job control
-jobs:
-  # native testing
-  - &x86
-    stage: test
-    env:
-      - NAME=littlefs-x86
-    install: *install-common
-    script: [*test-example, *report-size]
-  - {<<: *x86, script: [*test-default,          *report-size]}
-  - {<<: *x86, script: [*test-nor,              *report-size]}
-  - {<<: *x86, script: [*test-emmc,             *report-size]}
-  - {<<: *x86, script: [*test-nand,             *report-size]}
-  - {<<: *x86, script: [*test-no-intrinsics,    *report-size]}
-  - {<<: *x86, script: [*test-no-inline,        *report-size]}
-  - {<<: *x86, script: [*test-byte-writes,      *report-size]}
-  - {<<: *x86, script: [*test-block-cycles,     *report-size]}
-  - {<<: *x86, script: [*test-odd-block-count,  *report-size]}
-  - {<<: *x86, script: [*test-odd-block-size,   *report-size]}
-
-  # cross-compile with ARM (thumb mode)
-  - &arm
-    stage: test
-    env:
-      - NAME=littlefs-arm
-      - CC="arm-linux-gnueabi-gcc --static -mthumb"
-      - TFLAGS="$TFLAGS --exec=qemu-arm"
-    install:
-      - *install-common
-      - sudo apt-get install
-            gcc-arm-linux-gnueabi
-            libc6-dev-armel-cross
-            qemu-user
-      - arm-linux-gnueabi-gcc --version
-      - qemu-arm -version
-    script: [*test-example, *report-size]
-  - {<<: *arm, script: [*test-default,          *report-size]}
-  - {<<: *arm, script: [*test-nor,              *report-size]}
-  - {<<: *arm, script: [*test-emmc,             *report-size]}
-  - {<<: *arm, script: [*test-nand,             *report-size]}
-  - {<<: *arm, script: [*test-no-intrinsics,    *report-size]}
-  - {<<: *arm, script: [*test-no-inline,        *report-size]}
-  # it just takes way to long to run byte-level writes in qemu,
-  # note this is still tested in the native tests
-  #- {<<: *arm, script: [*test-byte-writes,      *report-size]}
-  - {<<: *arm, script: [*test-block-cycles,     *report-size]}
-  - {<<: *arm, script: [*test-odd-block-count,  *report-size]}
-  - {<<: *arm, script: [*test-odd-block-size,   *report-size]}
-
-  # cross-compile with MIPS
-  - &mips
-    stage: test
-    env:
-      - NAME=littlefs-mips
-      - CC="mips-linux-gnu-gcc --static"
-      - TFLAGS="$TFLAGS --exec=qemu-mips"
-    install:
-      - *install-common
-      - sudo apt-get install
-            gcc-mips-linux-gnu
-            libc6-dev-mips-cross
-            qemu-user
-      - mips-linux-gnu-gcc --version
-      - qemu-mips -version
-    script: [*test-example, *report-size]
-  - {<<: *mips, script: [*test-default,          *report-size]}
-  - {<<: *mips, script: [*test-nor,              *report-size]}
-  - {<<: *mips, script: [*test-emmc,             *report-size]}
-  - {<<: *mips, script: [*test-nand,             *report-size]}
-  - {<<: *mips, script: [*test-no-intrinsics,    *report-size]}
-  - {<<: *mips, script: [*test-no-inline,        *report-size]}
-  # it just takes way to long to run byte-level writes in qemu,
-  # note this is still tested in the native tests
-  #- {<<: *mips, script: [*test-byte-writes,      *report-size]}
-  - {<<: *mips, script: [*test-block-cycles,     *report-size]}
-  - {<<: *mips, script: [*test-odd-block-count,  *report-size]}
-  - {<<: *mips, script: [*test-odd-block-size,   *report-size]}
-
-  # cross-compile with PowerPC
-  - &powerpc
-    stage: test
-    env:
-      - NAME=littlefs-powerpc
-      - CC="powerpc-linux-gnu-gcc --static"
-      - TFLAGS="$TFLAGS --exec=qemu-ppc"
-    install:
-      - *install-common
-      - sudo apt-get install
-            gcc-powerpc-linux-gnu
-            libc6-dev-powerpc-cross
-            qemu-user
-      - powerpc-linux-gnu-gcc --version
-      - qemu-ppc -version
-    script: [*test-example, *report-size]
-  - {<<: *powerpc, script: [*test-default,          *report-size]}
-  - {<<: *powerpc, script: [*test-nor,              *report-size]}
-  - {<<: *powerpc, script: [*test-emmc,             *report-size]}
-  - {<<: *powerpc, script: [*test-nand,             *report-size]}
-  - {<<: *powerpc, script: [*test-no-intrinsics,    *report-size]}
-  - {<<: *powerpc, script: [*test-no-inline,        *report-size]}
-  # it just takes way to long to run byte-level writes in qemu,
-  # note this is still tested in the native tests
-  #- {<<: *powerpc, script: [*test-byte-writes,      *report-size]}
-  - {<<: *powerpc, script: [*test-block-cycles,     *report-size]}
-  - {<<: *powerpc, script: [*test-odd-block-count,  *report-size]}
-  - {<<: *powerpc, script: [*test-odd-block-size,   *report-size]}
-
-  # test under valgrind, checking for memory errors
-  - &valgrind
-    stage: test
-    env:
-      - NAME=littlefs-valgrind
-    install:
-      - *install-common
-      - sudo apt-get install valgrind
-      - valgrind --version
-    script:
-      - make test TFLAGS+="-k --valgrind"
-
-  # test compilation in read-only mode
-  - stage: test
-    env:
-      - NAME=littlefs-readonly
-      - CC="arm-linux-gnueabi-gcc --static -mthumb"
-      - CFLAGS="-Werror -DLFS_READONLY"
-    if: branch !~ -prefix$
-    install:
-      - *install-common
-      - sudo apt-get install
-            gcc-arm-linux-gnueabi
-            libc6-dev-armel-cross
-      - arm-linux-gnueabi-gcc --version
-    # report-size will compile littlefs and report the size
-    script: [*report-size]
-
-  # test compilation in thread-safe mode
-  - stage: test
-    env:
-      - NAME=littlefs-threadsafe
-      - CC="arm-linux-gnueabi-gcc --static -mthumb"
-      - CFLAGS="-Werror -DLFS_THREADSAFE"
-    if: branch !~ -prefix$
-    install:
-      - *install-common
-      - sudo apt-get install
-            gcc-arm-linux-gnueabi
-            libc6-dev-armel-cross
-      - arm-linux-gnueabi-gcc --version
-    # report-size will compile littlefs and report the size
-    script: [*report-size]
-
-  # self-host with littlefs-fuse for fuzz test
-  - stage: test
-    env:
-      - NAME=littlefs-fuse
-    if: branch !~ -prefix$
-    install:
-      - *install-common
-      - sudo apt-get install libfuse-dev
-      - git clone --depth 1 https://github.com/geky/littlefs-fuse -b v2
-      - fusermount -V
-      - gcc --version
-
-      # setup disk for littlefs-fuse
-      - rm -rf littlefs-fuse/littlefs/*
-      - cp -r $(git ls-tree --name-only HEAD) littlefs-fuse/littlefs
-
-      - mkdir mount
-      - sudo chmod a+rw /dev/loop0
-      - dd if=/dev/zero bs=512 count=128K of=disk
-      - losetup /dev/loop0 disk
-    script:
-      # self-host test
-      - make -C littlefs-fuse
-
-      - littlefs-fuse/lfs --format /dev/loop0
-      - littlefs-fuse/lfs /dev/loop0 mount
-
-      - ls mount
-      - mkdir mount/littlefs
-      - cp -r $(git ls-tree --name-only HEAD) mount/littlefs
-      - cd mount/littlefs
-      - stat .
-      - ls -flh
-      - make -B test
-
-  # test migration using littlefs-fuse
-  - stage: test
-    env:
-      - NAME=littlefs-migration
-    if: branch !~ -prefix$
-    install:
-      - *install-common
-      - sudo apt-get install libfuse-dev
-      - git clone --depth 1 https://github.com/geky/littlefs-fuse -b v2 v2
-      - git clone --depth 1 https://github.com/geky/littlefs-fuse -b v1 v1
-      - fusermount -V
-      - gcc --version
-
-      # setup disk for littlefs-fuse
-      - rm -rf v2/littlefs/*
-      - cp -r $(git ls-tree --name-only HEAD) v2/littlefs
-
-      - mkdir mount
-      - sudo chmod a+rw /dev/loop0
-      - dd if=/dev/zero bs=512 count=128K of=disk
-      - losetup /dev/loop0 disk
-    script:
-      # compile v1 and v2
-      - make -C v1
-      - make -C v2
-
-      # run self-host test with v1
-      - v1/lfs --format /dev/loop0
-      - v1/lfs /dev/loop0 mount
-
-      - ls mount
-      - mkdir mount/littlefs
-      - cp -r $(git ls-tree --name-only HEAD) mount/littlefs
-      - cd mount/littlefs
-      - stat .
-      - ls -flh
-      - make -B test
-
-      # attempt to migrate
-      - cd ../..
-      - fusermount -u mount
-
-      - v2/lfs --migrate /dev/loop0
-      - v2/lfs /dev/loop0 mount
-
-      # run self-host test with v2 right where we left off
-      - ls mount
-      - cd mount/littlefs
-      - stat .
-      - ls -flh
-      - make -B test
-
-  # automatically create releases
-  - stage: deploy
-    env:
-      - NAME=deploy
-    script:
-      - |
-        bash << 'SCRIPT'
-        set -ev
-        # Find version defined in lfs.h
-        LFS_VERSION=$(grep -ox '#define LFS_VERSION .*' lfs.h | cut -d ' ' -f3)
-        LFS_VERSION_MAJOR=$((0xffff & ($LFS_VERSION >> 16)))
-        LFS_VERSION_MINOR=$((0xffff & ($LFS_VERSION >>  0)))
-        # Grab latests patch from repo tags, default to 0, needs finagling
-        # to get past github's pagination api
-        PREV_URL=https://api.github.com/repos/$TRAVIS_REPO_SLUG/git/refs/tags/v$LFS_VERSION_MAJOR.$LFS_VERSION_MINOR.
-        PREV_URL=$(curl -u "$GEKY_BOT_RELEASES" "$PREV_URL" -I \
-            | sed -n '/^Link/{s/.*<\(.*\)>; rel="last"/\1/;p;q0};$q1' \
-            || echo $PREV_URL)
-        LFS_VERSION_PATCH=$(curl -u "$GEKY_BOT_RELEASES" "$PREV_URL" \
-            | jq 'map(.ref | match("\\bv.*\\..*\\.(.*)$";"g")
-                .captures[].string | tonumber) | max + 1' \
-            || echo 0)
-        # We have our new version
-        LFS_VERSION="v$LFS_VERSION_MAJOR.$LFS_VERSION_MINOR.$LFS_VERSION_PATCH"
-        echo "VERSION $LFS_VERSION"
-        # Check that we're the most recent commit
-        CURRENT_COMMIT=$(curl -f -u "$GEKY_BOT_RELEASES" \
-            https://api.github.com/repos/$TRAVIS_REPO_SLUG/commits/master \
-            | jq -re '.sha')
-        [ "$TRAVIS_COMMIT" == "$CURRENT_COMMIT" ] || exit 0
-        # Create major branch
-        git branch v$LFS_VERSION_MAJOR HEAD
-        # Create major prefix branch
-        git config user.name "geky bot"
-        git config user.email "bot@geky.net"
-        git fetch https://github.com/$TRAVIS_REPO_SLUG.git \
-            --depth=50 v$LFS_VERSION_MAJOR-prefix || true
-        ./scripts/prefix.py lfs$LFS_VERSION_MAJOR
-        git branch v$LFS_VERSION_MAJOR-prefix $( \
-            git commit-tree $(git write-tree) \
-                $(git rev-parse --verify -q FETCH_HEAD | sed -e 's/^/-p /') \
-                -p HEAD \
-                -m "Generated v$LFS_VERSION_MAJOR prefixes")
-        git reset --hard
-        # Update major version branches (vN and vN-prefix)
-        git push --atomic https://$GEKY_BOT_RELEASES@github.com/$TRAVIS_REPO_SLUG.git \
-            v$LFS_VERSION_MAJOR \
-            v$LFS_VERSION_MAJOR-prefix
-        # Build release notes
-        PREV=$(git tag --sort=-v:refname -l "v*" | head -1)
-        if [ ! -z "$PREV" ]
-        then
-            echo "PREV $PREV"
-            CHANGES=$(git log --oneline $PREV.. --grep='^Merge' --invert-grep)
-            printf "CHANGES\n%s\n\n" "$CHANGES"
-        fi
-        case ${GEKY_BOT_DRAFT:-minor} in
-            true)  DRAFT=true ;;
-            minor) DRAFT=$(jq -R 'endswith(".0")' <<< "$LFS_VERSION") ;;
-            false) DRAFT=false ;;
-        esac
-        # Create the release and patch version tag (vN.N.N)
-        curl -f -u "$GEKY_BOT_RELEASES" -X POST \
-            https://api.github.com/repos/$TRAVIS_REPO_SLUG/releases \
-            -d "{
-                \"tag_name\": \"$LFS_VERSION\",
-                \"name\": \"${LFS_VERSION%.0}\",
-                \"target_commitish\": \"$TRAVIS_COMMIT\",
-                \"draft\": $DRAFT,
-                \"body\": $(jq -sR '.' <<< "$CHANGES")
-            }" #"
-        SCRIPT
-
-# manage statuses
-before_install:
-  - |
-    # don't clobber other (not us) failures
-    if ! curl https://api.github.com/repos/$TRAVIS_REPO_SLUG/status/${TRAVIS_PULL_REQUEST_SHA:-$TRAVIS_COMMIT} \
-        | jq -e ".statuses[] | select(
-            .context == \"${TRAVIS_BUILD_STAGE_NAME,,}/$NAME\" and
-            .state == \"failure\" and
-            (.target_url | endswith(\"$TRAVIS_JOB_NUMBER\") | not))"
-    then
-        curl -u "$GEKY_BOT_STATUSES" -X POST \
-            https://api.github.com/repos/$TRAVIS_REPO_SLUG/statuses/${TRAVIS_PULL_REQUEST_SHA:-$TRAVIS_COMMIT} \
-            -d "{
-                \"context\": \"${TRAVIS_BUILD_STAGE_NAME,,}/$NAME\",
-                \"state\": \"pending\",
-                \"description\": \"${STATUS:-In progress}\",
-                \"target_url\": \"$TRAVIS_JOB_WEB_URL#$TRAVIS_JOB_NUMBER\"
-            }"
-    fi
-
-after_failure:
-  - |
-    # don't clobber other (not us) failures
-    if ! curl https://api.github.com/repos/$TRAVIS_REPO_SLUG/status/${TRAVIS_PULL_REQUEST_SHA:-$TRAVIS_COMMIT} \
-        | jq -e ".statuses[] | select(
-            .context == \"${TRAVIS_BUILD_STAGE_NAME,,}/$NAME\" and
-            .state == \"failure\" and
-            (.target_url | endswith(\"$TRAVIS_JOB_NUMBER\") | not))"
-    then
-        curl -u "$GEKY_BOT_STATUSES" -X POST \
-            https://api.github.com/repos/$TRAVIS_REPO_SLUG/statuses/${TRAVIS_PULL_REQUEST_SHA:-$TRAVIS_COMMIT} \
-            -d "{
-                \"context\": \"${TRAVIS_BUILD_STAGE_NAME,,}/$NAME\",
-                \"state\": \"failure\",
-                \"description\": \"${STATUS:-Failed}\",
-                \"target_url\": \"$TRAVIS_JOB_WEB_URL#$TRAVIS_JOB_NUMBER\"
-            }"
-    fi
-
-after_success:
-  - |
-    # don't clobber other (not us) failures
-    # only update if we were last job to mark in progress,
-    # this isn't perfect but is probably good enough
-    if ! curl https://api.github.com/repos/$TRAVIS_REPO_SLUG/status/${TRAVIS_PULL_REQUEST_SHA:-$TRAVIS_COMMIT} \
-        | jq -e ".statuses[] | select(
-            .context == \"${TRAVIS_BUILD_STAGE_NAME,,}/$NAME\" and
-            (.state == \"failure\" or .state == \"pending\") and
-            (.target_url | endswith(\"$TRAVIS_JOB_NUMBER\") | not))"
-    then
-        curl -u "$GEKY_BOT_STATUSES" -X POST \
-            https://api.github.com/repos/$TRAVIS_REPO_SLUG/statuses/${TRAVIS_PULL_REQUEST_SHA:-$TRAVIS_COMMIT} \
-            -d "{
-                \"context\": \"${TRAVIS_BUILD_STAGE_NAME,,}/$NAME\",
-                \"state\": \"success\",
-                \"description\": \"${STATUS:-Passed}\",
-                \"target_url\": \"$TRAVIS_JOB_WEB_URL#$TRAVIS_JOB_NUMBER\"
-            }"
-    fi
--- a/85
+++ b/85
@@ -1,25 +1,39 @@
-TARGET = lfs.a
-ifneq ($(wildcard test.c main.c),)
-override TARGET = lfs
+ifdef BUILDDIR
+# make sure BUILDDIR ends with a slash
+override BUILDDIR := $(BUILDDIR)/
+# bit of a hack, but we want to make sure BUILDDIR directory structure
+# is correct before any commands
+$(if $(findstring n,$(MAKEFLAGS)),, $(shell mkdir -p \
+	$(BUILDDIR) \
+	$(BUILDDIR)bd \
+	$(BUILDDIR)tests))
 endif

+# overridable target/src/tools/flags/etc
+ifneq ($(wildcard test.c main.c),)
+TARGET ?= $(BUILDDIR)lfs
+else
+TARGET ?= $(BUILDDIR)lfs.a
+endif
+
+
 CC ?= gcc
 AR ?= ar
 SIZE ?= size
+CTAGS ?= ctags
+NM ?= nm
+LCOV ?= lcov

-SRC += $(wildcard *.c bd/*.c)
-OBJ := $(SRC:.c=.o)
-DEP := $(SRC:.c=.d)
-ASM := $(SRC:.c=.s)
+SRC ?= $(wildcard *.c)
+OBJ := $(SRC:%.c=$(BUILDDIR)%.o)
+DEP := $(SRC:%.c=$(BUILDDIR)%.d)
+ASM := $(SRC:%.c=$(BUILDDIR)%.s)

 ifdef DEBUG
 override CFLAGS += -O0 -g3
 else
 override CFLAGS += -Os
 endif
-ifdef WORD
-override CFLAGS += -m$(WORD)
-endif
 ifdef TRACE
 override CFLAGS += -DLFS_YES_TRACE
 endif
@@ -28,40 +42,73 @@ override CFLAGS += -std=c99 -Wall -pedantic
 override CFLAGS += -Wextra -Wshadow -Wjump-misses-init -Wundef

 ifdef VERBOSE
-override TFLAGS += -v
+override TESTFLAGS += -v
+override CODEFLAGS += -v
+override COVERAGEFLAGS += -v
+endif
+ifdef EXEC
+override TESTFLAGS += --exec="$(EXEC)"
+endif
+ifdef BUILDDIR
+override TESTFLAGS += --build-dir="$(BUILDDIR:/=)"
+override CODEFLAGS += --build-dir="$(BUILDDIR:/=)"
+endif
+ifneq ($(NM),nm)
+override CODEFLAGS += --nm-tool="$(NM)"
 endif


-all: $(TARGET)
+# commands
+.PHONY: all build
+all build: $(TARGET)

+.PHONY: asm
 asm: $(ASM)

+.PHONY: size
 size: $(OBJ)
 	$(SIZE) -t $^

+.PHONY: tags
+tags:
+	$(CTAGS) --totals --c-types=+p $(shell find -H -name '*.h') $(SRC)
+
+.PHONY: code
+code: $(OBJ)
+	./scripts/code.py $^ $(CODEFLAGS)
+
+.PHONY: test
 test:
-	./scripts/test.py $(TFLAGS)
+	./scripts/test.py $(TESTFLAGS)
 .SECONDEXPANSION:
 test%: tests/test$$(firstword $$(subst \#, ,%)).toml
-	./scripts/test.py $@ $(TFLAGS)
+	./scripts/test.py $@ $(TESTFLAGS)

+.PHONY: coverage
+coverage:
+	./scripts/coverage.py $(BUILDDIR)tests/*.toml.info $(COVERAGEFLAGS)
+
+# rules
 -include $(DEP)
+.SUFFIXES:

-lfs: $(OBJ)
+$(BUILDDIR)lfs: $(OBJ)
 	$(CC) $(CFLAGS) $^ $(LFLAGS) -o $@

-%.a: $(OBJ)
+$(BUILDDIR)%.a: $(OBJ)
 	$(AR) rcs $@ $^

-%.o: %.c
+$(BUILDDIR)%.o: %.c
 	$(CC) -c -MMD $(CFLAGS) $< -o $@

-%.s: %.c
+$(BUILDDIR)%.s: %.c
 	$(CC) -S $(CFLAGS) $< -o $@

+# clean everything
+.PHONY: clean
 clean:
 	rm -f $(TARGET)
 	rm -f $(OBJ)
 	rm -f $(DEP)
 	rm -f $(ASM)
-	rm -f tests/*.toml.*
+	rm -f $(BUILDDIR)tests/*.toml.*
--- a/lfs.c
+++ b/lfs.c
@@ -425,7 +425,8 @@ static inline void lfs_superblock_tole32(lfs_superblock_t *superblock) {
    superblock->attr_max    = lfs_tole32(superblock->attr_max);
 }

-static inline bool lfs_mlist_isopen(struct lfs_mlist *head,
+#ifndef LFS_NO_ASSERT
+static bool lfs_mlist_isopen(struct lfs_mlist *head,
        struct lfs_mlist *node) {
    for (struct lfs_mlist **p = &head; *p; p = &(*p)->next) {
        if (*p == (struct lfs_mlist*)node) {
@@ -435,8 +436,9 @@ static inline bool lfs_mlist_isopen(struct lfs_mlist *head,

    return false;
 }
+#endif

-static inline void lfs_mlist_remove(lfs_t *lfs, struct lfs_mlist *mlist) {
+static void lfs_mlist_remove(lfs_t *lfs, struct lfs_mlist *mlist) {
    for (struct lfs_mlist **p = &lfs->mlist; *p; p = &(*p)->next) {
        if (*p == mlist) {
            *p = (*p)->next;
@@ -445,7 +447,7 @@ static inline void lfs_mlist_remove(lfs_t *lfs, struct lfs_mlist *mlist) {
    }
 }

-static inline void lfs_mlist_append(lfs_t *lfs, struct lfs_mlist *mlist) {
+static void lfs_mlist_append(lfs_t *lfs, struct lfs_mlist *mlist) {
    mlist->next = lfs->mlist;
    lfs->mlist = mlist;
 }
@@ -465,7 +467,7 @@ static int lfs_file_rawsync(lfs_t *lfs, lfs_file_t *file);
 static int lfs_file_outline(lfs_t *lfs, lfs_file_t *file);
 static int lfs_file_flush(lfs_t *lfs, lfs_file_t *file);

-static void lfs_fs_preporphans(lfs_t *lfs, int8_t orphans);
+static int lfs_fs_preporphans(lfs_t *lfs, int8_t orphans);
 static void lfs_fs_prepmove(lfs_t *lfs,
        uint16_t id, const lfs_block_t pair[2]);
 static int lfs_fs_pred(lfs_t *lfs, const lfs_block_t dir[2],
@@ -1589,7 +1591,8 @@ static int lfs_dir_compact(lfs_t *lfs,
        // for metadata updates.
        if (end - begin < 0xff &&
                size <= lfs_min(lfs->cfg->block_size - 36,
-                    lfs_alignup(lfs->cfg->block_size/2,
+                    lfs_alignup((lfs->cfg->metadata_max ?
+                            lfs->cfg->metadata_max : lfs->cfg->block_size)/2,
                        lfs->cfg->prog_size))) {
            break;
        }
@@ -1674,7 +1677,8 @@ static int lfs_dir_compact(lfs_t *lfs,
                .crc = 0xffffffff,

                .begin = 0,
-                .end = lfs->cfg->block_size - 8,
+                .end = (lfs->cfg->metadata_max ?
+                    lfs->cfg->metadata_max : lfs->cfg->block_size) - 8,
            };

            // erase block to write to
@@ -1884,7 +1888,8 @@ static int lfs_dir_commit(lfs_t *lfs, lfs_mdir_t *dir,
            .crc = 0xffffffff,

            .begin = dir->off,
-            .end = lfs->cfg->block_size - 8,
+            .end = (lfs->cfg->metadata_max ?
+                lfs->cfg->metadata_max : lfs->cfg->block_size) - 8,
        };

        // traverse attrs that need to be written out
@@ -2061,7 +2066,10 @@ static int lfs_rawmkdir(lfs_t *lfs, const char *path) {
    // current block end of list?
    if (cwd.m.split) {
        // update tails, this creates a desync
-        lfs_fs_preporphans(lfs, +1);
+        err = lfs_fs_preporphans(lfs, +1);
+        if (err) {
+            return err;
+        }

        // it's possible our predecessor has to be relocated, and if
        // our parent is our predecessor's predecessor, this could have
@@ -2081,7 +2089,10 @@ static int lfs_rawmkdir(lfs_t *lfs, const char *path) {
        }

        lfs->mlist = cwd.next;
-        lfs_fs_preporphans(lfs, -1);
+        err = lfs_fs_preporphans(lfs, -1);
+        if (err) {
+            return err;
+        }
    }

    // now insert into our parent block
@@ -2966,7 +2977,9 @@ static lfs_ssize_t lfs_file_rawwrite(lfs_t *lfs, lfs_file_t *file,
    if ((file->flags & LFS_F_INLINE) &&
            lfs_max(file->pos+nsize, file->ctz.size) >
            lfs_min(0x3fe, lfs_min(
-                lfs->cfg->cache_size, lfs->cfg->block_size/8))) {
+                lfs->cfg->cache_size,
+                (lfs->cfg->metadata_max ?
+                    lfs->cfg->metadata_max : lfs->cfg->block_size) / 8))) {
        // inline file doesn't fit anymore
        int err = lfs_file_outline(lfs, file);
        if (err) {
@@ -3048,14 +3061,6 @@ relocate:

 static lfs_soff_t lfs_file_rawseek(lfs_t *lfs, lfs_file_t *file,
        lfs_soff_t off, int whence) {
-#ifndef LFS_READONLY
-    // write out everything beforehand, may be noop if rdonly
-    int err = lfs_file_flush(lfs, file);
-    if (err) {
-        return err;
-    }
-#endif
-
    // find new pos
    lfs_off_t npos = file->pos;
    if (whence == LFS_SEEK_SET) {
@@ -3063,7 +3068,7 @@ static lfs_soff_t lfs_file_rawseek(lfs_t *lfs, lfs_file_t *file,
    } else if (whence == LFS_SEEK_CUR) {
        npos = file->pos + off;
    } else if (whence == LFS_SEEK_END) {
-        npos = file->ctz.size + off;
+        npos = lfs_file_rawsize(lfs, file) + off;
    }

    if (npos > lfs->file_max) {
@@ -3071,6 +3076,19 @@ static lfs_soff_t lfs_file_rawseek(lfs_t *lfs, lfs_file_t *file,
        return LFS_ERR_INVAL;
    }

+    if (file->pos == npos) {
+        // noop - position has not changed
+        return npos;
+    }
+
+#ifndef LFS_READONLY
+    // write out everything beforehand, may be noop if rdonly
+    int err = lfs_file_flush(lfs, file);
+    if (err) {
+        return err;
+    }
+#endif
+
    // update pos
    file->pos = npos;
    return npos;
@@ -3101,21 +3119,22 @@ static int lfs_file_rawtruncate(lfs_t *lfs, lfs_file_t *file, lfs_off_t size) {
            return err;
        }

+        // need to set pos/block/off consistently so seeking back to
+        // the old position does not get confused
+        file->pos = size;
        file->ctz.head = file->block;
        file->ctz.size = size;
        file->flags |= LFS_F_DIRTY | LFS_F_READING;
    } else if (size > oldsize) {
        // flush+seek if not already at end
-        if (file->pos != oldsize) {
-            lfs_soff_t res = lfs_file_rawseek(lfs, file, 0, LFS_SEEK_END);
-            if (res < 0) {
-                return (int)res;
-            }
+        lfs_soff_t res = lfs_file_rawseek(lfs, file, 0, LFS_SEEK_END);
+        if (res < 0) {
+            return (int)res;
        }

        // fill with zeros
        while (file->pos < size) {
-            lfs_ssize_t res = lfs_file_rawwrite(lfs, file, &(uint8_t){0}, 1);
+            res = lfs_file_rawwrite(lfs, file, &(uint8_t){0}, 1);
            if (res < 0) {
                return (int)res;
            }
@@ -3171,7 +3190,106 @@ static int lfs_rawstat(lfs_t *lfs, const char *path, struct lfs_info *info) {
 }

 #ifndef LFS_READONLY
-static int lfs_rawremove(lfs_t *lfs, const char *path) {
+typedef int (*lfs_dir_prep_helper_t)(lfs_t *, struct lfs_mlist *, lfs_gstate_t *);
+
+static int lfs_dir_prep_remove_nonempty_folders(lfs_t *lfs, struct lfs_mlist *dir,
+        lfs_gstate_t *tmp_gstate)
+{
+    lfs_gstate_t split_gstate;
+    uint16_t id = 0;
+
+    // Walk tags stored in this directory and check for any directory
+    // tags.  Removal of directories with a directory in them can lead
+    // to additional orphans in the filesystem, so we return
+    // LFS_ERR_NOTEMPTY in this case.  Otherwise, leave the loaded
+    // directory for the tail end of the directory split to leave a proper
+    // view of the filesystem after removal.
+    while (true) {
+        if (dir->m.count == id) {
+            if (!dir->m.split) {
+                // We have iterated through the folder to the last
+                // tag.
+                break;
+            }
+
+            // Before we fetch the next block, update our fetched gstate xor
+            lfs_dir_getgstate(lfs, &dir->m, &split_gstate);
+            lfs_gstate_xor(tmp_gstate, &split_gstate);
+
+            int err = lfs_dir_fetch(lfs, &dir->m, dir->m.tail);
+            if (err) {
+                return err;
+            }
+
+            id = 0;
+        }
+
+        lfs_stag_t tag = lfs_dir_get(lfs, &dir->m, LFS_MKTAG(0x780, 0x3ff, 0),
+                LFS_MKTAG(LFS_TYPE_NAME, id, 0), NULL);
+
+        if (tag < 0) {
+            return tag;
+        }
+
+        if (lfs_tag_type3(tag) == LFS_TYPE_DIR) {
+            return LFS_ERR_NOTEMPTY;
+        }
+
+        id += 1;
+    }
+
+    return 0;
+}
+
+static int lfs_dir_prep_removal(lfs_t *lfs, struct lfs_mlist *dir,
+        lfs_mdir_t *newcwd, uint16_t newid, lfs_block_t *pair,
+        lfs_gstate_t *tmp_gstate, lfs_dir_prep_helper_t helper)
+{
+    lfs_stag_t res = lfs_dir_get(lfs, newcwd, LFS_MKTAG(0x700, 0x3ff, 0),
+            LFS_MKTAG(LFS_TYPE_STRUCT, newid, 8), pair);
+    if (res < 0) {
+        return (int)res;
+    }
+    lfs_pair_fromle32(pair);
+
+    memset(tmp_gstate, 0, sizeof(*tmp_gstate));
+
+    int err = lfs_dir_fetch(lfs, &dir->m, pair);
+    if (err) {
+        return err;
+    }
+
+    if (dir->m.count > 0 || dir->m.split) {
+        // Normal POSIX behavior wouldn't allow a non-empty
+        // folder to be removed/renamed into in this manner
+        if (NULL == helper) {
+            return LFS_ERR_NOTEMPTY;
+        }
+
+        err = helper(lfs, dir, tmp_gstate);
+        if (err) {
+            return err;
+        }
+    }
+
+    // mark fs as orphaned
+    err = lfs_fs_preporphans(lfs, +1);
+    if (err) {
+        return err;
+    }
+
+    // I know it's crazy but yes, dir can be changed by our parent's
+    // commit (if predecessor is child)
+    dir->type = 0;
+    dir->id = 0;
+    lfs->mlist = dir;
+
+    return 0;
+}
+#endif
+
+#ifndef LFS_READONLY
+static int lfs_rawremove(lfs_t *lfs, const char *path, lfs_dir_prep_helper_t helper) {
    // deorphan if we haven't yet, needed at most once after poweron
    int err = lfs_fs_forceconsistency(lfs);
    if (err) {
@@ -3185,34 +3303,16 @@ static int lfs_rawremove(lfs_t *lfs, const char *path) {
    }

    struct lfs_mlist dir;
+    lfs_block_t pair[2];
+    lfs_gstate_t tmp_gstate;
    dir.next = lfs->mlist;
    if (lfs_tag_type3(tag) == LFS_TYPE_DIR) {
-        // must be empty before removal
-        lfs_block_t pair[2];
-        lfs_stag_t res = lfs_dir_get(lfs, &cwd, LFS_MKTAG(0x700, 0x3ff, 0),
-                LFS_MKTAG(LFS_TYPE_STRUCT, lfs_tag_id(tag), 8), pair);
-        if (res < 0) {
-            return (int)res;
-        }
-        lfs_pair_fromle32(pair);
-
-        err = lfs_dir_fetch(lfs, &dir.m, pair);
-        if (err) {
+        // must be empty before removal to prevent orphans
+        err = lfs_dir_prep_removal(lfs, &dir, &cwd, lfs_tag_id(tag),
+                pair, &tmp_gstate, helper);
+        if (err < 0) {
            return err;
        }
-
-        if (dir.m.count > 0 || dir.m.split) {
-            return LFS_ERR_NOTEMPTY;
-        }
-
-        // mark fs as orphaned
-        lfs_fs_preporphans(lfs, +1);
-
-        // I know it's crazy but yes, dir can be changed by our parent's
-        // commit (if predecessor is child)
-        dir.type = 0;
-        dir.id = 0;
-        lfs->mlist = &dir;
    }

    // delete the entry
@@ -3226,13 +3326,20 @@ static int lfs_rawremove(lfs_t *lfs, const char *path) {
    lfs->mlist = dir.next;
    if (lfs_tag_type3(tag) == LFS_TYPE_DIR) {
        // fix orphan
-        lfs_fs_preporphans(lfs, -1);
-
-        err = lfs_fs_pred(lfs, dir.m.pair, &cwd);
+        err = lfs_fs_preporphans(lfs, -1);
        if (err) {
            return err;
        }

+        err = lfs_fs_pred(lfs, pair, &cwd);
+        if (err) {
+            return err;
+        }
+
+        // Merge in gstate from first block splits within the directory;
+        // lfs_dir_drop will pick up the last gstate entry.
+        lfs_gstate_xor(&lfs->gdelta, &tmp_gstate);
+
        err = lfs_dir_drop(lfs, &cwd, &dir.m);
        if (err) {
            return err;
@@ -3244,7 +3351,8 @@ static int lfs_rawremove(lfs_t *lfs, const char *path) {
 #endif

 #ifndef LFS_READONLY
-static int lfs_rawrename(lfs_t *lfs, const char *oldpath, const char *newpath) {
+static int lfs_rawrename(lfs_t *lfs, const char *oldpath,
+        const char *newpath, lfs_dir_prep_helper_t helper) {
    // deorphan if we haven't yet, needed at most once after poweron
    int err = lfs_fs_forceconsistency(lfs);
    if (err) {
@@ -3272,6 +3380,8 @@ static int lfs_rawrename(lfs_t *lfs, const char *oldpath, const char *newpath) {
    uint16_t newoldid = lfs_tag_id(oldtag);

    struct lfs_mlist prevdir;
+    lfs_block_t dir_pair[2];
+    lfs_gstate_t tmp_gstate;
    prevdir.next = lfs->mlist;
    if (prevtag == LFS_ERR_NOENT) {
        // check that name fits
@@ -3292,33 +3402,12 @@ static int lfs_rawrename(lfs_t *lfs, const char *oldpath, const char *newpath) {
        // we're renaming to ourselves??
        return 0;
    } else if (lfs_tag_type3(prevtag) == LFS_TYPE_DIR) {
-        // must be empty before removal
-        lfs_block_t prevpair[2];
-        lfs_stag_t res = lfs_dir_get(lfs, &newcwd, LFS_MKTAG(0x700, 0x3ff, 0),
-                LFS_MKTAG(LFS_TYPE_STRUCT, newid, 8), prevpair);
-        if (res < 0) {
-            return (int)res;
-        }
-        lfs_pair_fromle32(prevpair);
-
-        // must be empty before removal
-        err = lfs_dir_fetch(lfs, &prevdir.m, prevpair);
+        // must be empty before removal to prevent orphans
+        err = lfs_dir_prep_removal(lfs, &prevdir, &newcwd, newid,
+                dir_pair, &tmp_gstate, helper);
        if (err) {
            return err;
        }
-
-        if (prevdir.m.count > 0 || prevdir.m.split) {
-            return LFS_ERR_NOTEMPTY;
-        }
-
-        // mark fs as orphaned
-        lfs_fs_preporphans(lfs, +1);
-
-        // I know it's crazy but yes, dir can be changed by our parent's
-        // commit (if predecessor is child)
-        prevdir.type = 0;
-        prevdir.id = 0;
-        lfs->mlist = &prevdir;
    }

    if (!samepair) {
@@ -3355,13 +3444,20 @@ static int lfs_rawrename(lfs_t *lfs, const char *oldpath, const char *newpath) {
    lfs->mlist = prevdir.next;
    if (prevtag != LFS_ERR_NOENT && lfs_tag_type3(prevtag) == LFS_TYPE_DIR) {
        // fix orphan
-        lfs_fs_preporphans(lfs, -1);
-
-        err = lfs_fs_pred(lfs, prevdir.m.pair, &newcwd);
+        err = lfs_fs_preporphans(lfs, -1);
        if (err) {
            return err;
        }

+        err = lfs_fs_pred(lfs, dir_pair, &newcwd);
+        if (err) {
+            return err;
+        }
+
+        // Merge in gstate from first split blocks in the directory;
+        // lfs_dir_drop will pick up the other gstate entries.
+        lfs_gstate_xor(&lfs->gdelta, &tmp_gstate);
+
        err = lfs_dir_drop(lfs, &newcwd, &prevdir.m);
        if (err) {
            return err;
@@ -3536,6 +3632,8 @@ static int lfs_init(lfs_t *lfs, const struct lfs_config *cfg) {
        lfs->attr_max = LFS_ATTR_MAX;
    }

+    LFS_ASSERT(lfs->cfg->metadata_max <= lfs->cfg->block_size);
+
    // setup default state
    lfs->root[0] = LFS_BLOCK_NULL;
    lfs->root[1] = LFS_BLOCK_NULL;
@@ -3616,12 +3714,6 @@ static int lfs_rawformat(lfs_t *lfs, const struct lfs_config *cfg) {
            goto cleanup;
        }

-        // sanity check that fetch works
-        err = lfs_dir_fetch(lfs, &root, (const lfs_block_t[2]){0, 1});
-        if (err) {
-            goto cleanup;
-        }
-
        // force compaction to prevent accidentally mounting any
        // older version of littlefs that may live on disk
        root.erased = false;
@@ -3629,6 +3721,12 @@ static int lfs_rawformat(lfs_t *lfs, const struct lfs_config *cfg) {
        if (err) {
            goto cleanup;
        }
+
+        // sanity check that fetch works
+        err = lfs_dir_fetch(lfs, &root, (const lfs_block_t[2]){0, 1});
+        if (err) {
+            goto cleanup;
+        }
    }

 cleanup:
@@ -3829,7 +3927,7 @@ int lfs_fs_rawtraverse(lfs_t *lfs,
                if (err) {
                    return err;
                }
-            } else if (includeorphans && 
+            } else if (includeorphans &&
                    lfs_tag_type3(tag) == LFS_TYPE_DIRSTRUCT) {
                for (int i = 0; i < 2; i++) {
                    err = cb(data, (&ctz.head)[i]);
@@ -3986,7 +4084,10 @@ static int lfs_fs_relocate(lfs_t *lfs,

    if (tag != LFS_ERR_NOENT) {
        // update disk, this creates a desync
-        lfs_fs_preporphans(lfs, +1);
+        int err = lfs_fs_preporphans(lfs, +1);
+        if (err) {
+            return err;
+        }

        // fix pending move in this pair? this looks like an optimization but
        // is in fact _required_ since relocating may outdate the move.
@@ -4003,7 +4104,7 @@ static int lfs_fs_relocate(lfs_t *lfs,
        }

        lfs_pair_tole32(newpair);
-        int err = lfs_dir_commit(lfs, &parent, LFS_MKATTRS(
+        err = lfs_dir_commit(lfs, &parent, LFS_MKATTRS(
                {LFS_MKTAG_IF(moveid != 0x3ff,
                    LFS_TYPE_DELETE, moveid, 0), NULL},
                {tag, newpair}));
@@ -4013,7 +4114,10 @@ static int lfs_fs_relocate(lfs_t *lfs,
        }

        // next step, clean up orphans
-        lfs_fs_preporphans(lfs, -1);
+        err = lfs_fs_preporphans(lfs, -1);
+        if (err) {
+            return err;
+        }
    }

    // find pred
@@ -4052,11 +4156,13 @@ static int lfs_fs_relocate(lfs_t *lfs,
 #endif

 #ifndef LFS_READONLY
-static void lfs_fs_preporphans(lfs_t *lfs, int8_t orphans) {
+static int lfs_fs_preporphans(lfs_t *lfs, int8_t orphans) {
    LFS_ASSERT(lfs_tag_size(lfs->gstate.tag) > 0 || orphans >= 0);
    lfs->gstate.tag += orphans;
    lfs->gstate.tag = ((lfs->gstate.tag & ~LFS_MKTAG(0x800, 0, 0)) |
            ((uint32_t)lfs_gstate_hasorphans(&lfs->gstate) << 31));
+
+    return 0;
 }
 #endif

@@ -4173,8 +4279,7 @@ static int lfs_fs_deorphan(lfs_t *lfs) {
    }

    // mark orphans as fixed
-    lfs_fs_preporphans(lfs, -lfs_gstate_getorphans(&lfs->gstate));
-    return 0;
+    return lfs_fs_preporphans(lfs, -lfs_gstate_getorphans(&lfs->gstate));
 }
 #endif

@@ -4723,7 +4828,7 @@ static int lfs_rawmigrate(lfs_t *lfs, const struct lfs_config *cfg) {

                lfs1_entry_tole32(&entry1.d);
                err = lfs_dir_commit(lfs, &dir2, LFS_MKATTRS(
-                        {LFS_MKTAG(LFS_TYPE_CREATE, id, 0)},
+                        {LFS_MKTAG(LFS_TYPE_CREATE, id, 0), NULL},
                        {LFS_MKTAG_IF_ELSE(isdir,
                            LFS_TYPE_DIR, id, entry1.d.nlen,
                            LFS_TYPE_REG, id, entry1.d.nlen),
@@ -4828,7 +4933,7 @@ static int lfs_rawmigrate(lfs_t *lfs, const struct lfs_config *cfg) {

        lfs_superblock_tole32(&superblock);
        err = lfs_dir_commit(lfs, &dir2, LFS_MKATTRS(
-                {LFS_MKTAG(LFS_TYPE_CREATE, 0, 0)},
+                {LFS_MKTAG(LFS_TYPE_CREATE, 0, 0), NULL},
                {LFS_MKTAG(LFS_TYPE_SUPERBLOCK, 0, 8), "littlefs"},
                {LFS_MKTAG(LFS_TYPE_INLINESTRUCT, 0, sizeof(superblock)),
                    &superblock}));
@@ -4954,7 +5059,7 @@ int lfs_remove(lfs_t *lfs, const char *path) {
    }
    LFS_TRACE("lfs_remove(%p, \"%s\")", (void*)lfs, path);

-    err = lfs_rawremove(lfs, path);
+    err = lfs_rawremove(lfs, path, NULL);

    LFS_TRACE("lfs_remove -> %d", err);
    LFS_UNLOCK(lfs->cfg);
@@ -4962,6 +5067,24 @@ int lfs_remove(lfs_t *lfs, const char *path) {
 }
 #endif

+#ifndef LFS_READONLY
+int lfs_removeall(lfs_t *lfs, const char *path) {
+    int err = LFS_LOCK(lfs->cfg);
+    if (err) {
+        return err;
+    }
+    LFS_TRACE("lfs_removeall(%p, \"%s\")", (void*)lfs, path);
+
+    // Note: We pass in a helper pointer here so that this extra
+    //       logic can be dropped if it is never referenced
+    err = lfs_rawremove(lfs, path, lfs_dir_prep_remove_nonempty_folders);
+
+    LFS_TRACE("lfs_removeall -> %d", err);
+    LFS_UNLOCK(lfs->cfg);
+    return err;
+}
+#endif
+
 #ifndef LFS_READONLY
 int lfs_rename(lfs_t *lfs, const char *oldpath, const char *newpath) {
    int err = LFS_LOCK(lfs->cfg);
@@ -4970,7 +5093,25 @@ int lfs_rename(lfs_t *lfs, const char *oldpath, const char *newpath) {
    }
    LFS_TRACE("lfs_rename(%p, \"%s\", \"%s\")", (void*)lfs, oldpath, newpath);

-    err = lfs_rawrename(lfs, oldpath, newpath);
+    err = lfs_rawrename(lfs, oldpath, newpath, NULL);
+
+    LFS_TRACE("lfs_rename -> %d", err);
+    LFS_UNLOCK(lfs->cfg);
+    return err;
+}
+#endif
+
+#ifndef LFS_READONLY
+int lfs_rename_with_removeall(lfs_t *lfs, const char *oldpath, const char *newpath) {
+    int err = LFS_LOCK(lfs->cfg);
+    if (err) {
+        return err;
+    }
+    LFS_TRACE("lfs_rename(%p, \"%s\", \"%s\")", (void*)lfs, oldpath, newpath);
+
+    // Note: We pass in a helper pointer here so that this extra
+    //       logic can be dropped if it is never referenced
+    err = lfs_rawrename(lfs, oldpath, newpath, lfs_dir_prep_remove_nonempty_folders);

    LFS_TRACE("lfs_rename -> %d", err);
    LFS_UNLOCK(lfs->cfg);
--- a/lfs.h
+++ b/lfs.h
@@ -22,7 +22,7 @@ extern "C"
 // Software library version
 // Major (top-nibble), incremented on backwards incompatible changes
 // Minor (bottom-nibble), incremented on feature additions
-#define LFS_VERSION 0x00020003
+#define LFS_VERSION 0x00020004
 #define LFS_VERSION_MAJOR (0xffff & (LFS_VERSION >> 16))
 #define LFS_VERSION_MINOR (0xffff & (LFS_VERSION >>  0))

@@ -207,7 +207,7 @@ struct lfs_config {
    // Number of erasable blocks on the device.
    lfs_size_t block_count;

-    // Number of erase cycles before littlefs evicts metadata logs and moves 
+    // Number of erase cycles before littlefs evicts metadata logs and moves
    // the metadata to another block. Suggested values are in the
    // range 100-1000, with large values having better performance at the cost
    // of less consistent wear distribution.
@@ -256,6 +256,12 @@ struct lfs_config {
    // larger attributes size but must be <= LFS_ATTR_MAX. Defaults to
    // LFS_ATTR_MAX when zero.
    lfs_size_t attr_max;
+
+    // Optional upper limit on total space given to metadata pairs in bytes. On
+    // devices with large blocks (e.g. 128kB) setting this to a low size (2-8kB)
+    // can help bound the metadata compaction time. Must be <= block_size.
+    // Defaults to block_size when zero.
+    lfs_size_t metadata_max;
 };

 // File info structure
@@ -452,6 +458,17 @@ int lfs_unmount(lfs_t *lfs);
 int lfs_remove(lfs_t *lfs, const char *path);
 #endif

+#ifndef LFS_READONLY
+// Removes a file or directory
+//
+// If removing a directory, the directory must not have
+// any directories but it may contain files.  This is
+// non-POSIX behavior, and thus is a different call
+// than lfs_remove(...)
+// Returns a negative error code on failure.
+int lfs_removeall(lfs_t *lfs, const char *path);
+#endif
+
 #ifndef LFS_READONLY
 // Rename or move a file or directory
 //
@@ -462,6 +479,19 @@ int lfs_remove(lfs_t *lfs, const char *path);
 int lfs_rename(lfs_t *lfs, const char *oldpath, const char *newpath);
 #endif

+#ifndef LFS_READONLY
+// Rename or move a file or directory
+//
+// If the destination exists, it must match the source in type.
+// If the destination is a directory, it may not contain
+// any directories but it may contain files.  This is
+// non-POSIX behavior, and thus is a different call
+// than lfs_rename(...)
+//
+// Returns a negative error code on failure.
+int lfs_rename_with_removeall(lfs_t *lfs, const char *oldpath, const char *newpath);
+#endif
+
 // Find info about a file or directory
 //
 // Fills out the info structure, based on the specified file or directory.
--- a/lfs_util.h
+++ b/lfs_util.h
@@ -49,6 +49,7 @@ extern "C"
 // code footprint

 // Logging functions
+#ifndef LFS_TRACE
 #ifdef LFS_YES_TRACE
 #define LFS_TRACE_(fmt, ...) \
    printf("%s:%d:trace: " fmt "%s\n", __FILE__, __LINE__, __VA_ARGS__)
@@ -56,7 +57,9 @@ extern "C"
 #else
 #define LFS_TRACE(...)
 #endif
+#endif

+#ifndef LFS_DEBUG
 #ifndef LFS_NO_DEBUG
 #define LFS_DEBUG_(fmt, ...) \
    printf("%s:%d:debug: " fmt "%s\n", __FILE__, __LINE__, __VA_ARGS__)
@@ -64,7 +67,9 @@ extern "C"
 #else
 #define LFS_DEBUG(...)
 #endif
+#endif

+#ifndef LFS_WARN
 #ifndef LFS_NO_WARN
 #define LFS_WARN_(fmt, ...) \
    printf("%s:%d:warn: " fmt "%s\n", __FILE__, __LINE__, __VA_ARGS__)
@@ -72,7 +77,9 @@ extern "C"
 #else
 #define LFS_WARN(...)
 #endif
+#endif

+#ifndef LFS_ERROR
 #ifndef LFS_NO_ERROR
 #define LFS_ERROR_(fmt, ...) \
    printf("%s:%d:error: " fmt "%s\n", __FILE__, __LINE__, __VA_ARGS__)
@@ -80,13 +87,16 @@ extern "C"
 #else
 #define LFS_ERROR(...)
 #endif
+#endif

 // Runtime assertions
+#ifndef LFS_ASSERT
 #ifndef LFS_NO_ASSERT
 #define LFS_ASSERT(test) assert(test)
 #else
 #define LFS_ASSERT(test)
 #endif
+#endif


 // Builtin functions, these may be replaced by more efficient
--- a/scripts/code.py
+++ b/scripts/code.py
@@ -0,0 +1,214 @@
+#!/usr/bin/env python3
+#
+# Script to find code size at the function level. Basically just a bit wrapper
+# around nm with some extra conveniences for comparing builds. Heavily inspired
+# by Linux's Bloat-O-Meter.
+#
+
+import os
+import glob
+import itertools as it
+import subprocess as sp
+import shlex
+import re
+import csv
+import collections as co
+
+
+OBJ_PATHS = ['*.o', 'bd/*.o']
+
+def collect(paths, **args):
+    results = co.defaultdict(lambda: 0)
+    pattern = re.compile(
+        '^(?P<size>[0-9a-fA-F]+)' +
+        ' (?P<type>[%s])' % re.escape(args['type']) +
+        ' (?P<func>.+?)$')
+    for path in paths:
+        # note nm-tool may contain extra args
+        cmd = args['nm_tool'] + ['--size-sort', path]
+        if args.get('verbose'):
+            print(' '.join(shlex.quote(c) for c in cmd))
+        proc = sp.Popen(cmd,
+            stdout=sp.PIPE,
+            stderr=sp.PIPE if not args.get('verbose') else None,
+            universal_newlines=True)
+        for line in proc.stdout:
+            m = pattern.match(line)
+            if m:
+                results[(path, m.group('func'))] += int(m.group('size'), 16)
+        proc.wait()
+        if proc.returncode != 0:
+            if not args.get('verbose'):
+                for line in proc.stderr:
+                    sys.stdout.write(line)
+            sys.exit(-1)
+
+    flat_results = []
+    for (file, func), size in results.items():
+        # map to source files
+        if args.get('build_dir'):
+            file = re.sub('%s/*' % re.escape(args['build_dir']), '', file)
+        # discard internal functions
+        if func.startswith('__'):
+            continue
+        # discard .8449 suffixes created by optimizer
+        func = re.sub('\.[0-9]+', '', func)
+        flat_results.append((file, func, size))
+
+    return flat_results
+
+def main(**args):
+    # find sizes
+    if not args.get('use', None):
+        # find .o files
+        paths = []
+        for path in args['obj_paths']:
+            if os.path.isdir(path):
+                path = path + '/*.o'
+
+            for path in glob.glob(path):
+                paths.append(path)
+
+        if not paths:
+            print('no .obj files found in %r?' % args['obj_paths'])
+            sys.exit(-1)
+
+        results = collect(paths, **args)
+    else:
+        with open(args['use']) as f:
+            r = csv.DictReader(f)
+            results = [
+                (   result['file'],
+                    result['function'],
+                    int(result['size']))
+                for result in r]
+
+    total = 0
+    for _, _, size in results:
+        total += size
+
+    # find previous results?
+    if args.get('diff'):
+        with open(args['diff']) as f:
+            r = csv.DictReader(f)
+            prev_results = [
+                (   result['file'],
+                    result['function'],
+                    int(result['size']))
+                for result in r]
+
+        prev_total = 0
+        for _, _, size in prev_results:
+            prev_total += size
+
+    # write results to CSV
+    if args.get('output'):
+        with open(args['output'], 'w') as f:
+            w = csv.writer(f)
+            w.writerow(['file', 'function', 'size'])
+            for file, func, size in sorted(results):
+                w.writerow((file, func, size))
+
+    # print results
+    def dedup_entries(results, by='function'):
+        entries = co.defaultdict(lambda: 0)
+        for file, func, size in results:
+            entry = (file if by == 'file' else func)
+            entries[entry] += size
+        return entries
+
+    def diff_entries(olds, news):
+        diff = co.defaultdict(lambda: (0, 0, 0, 0))
+        for name, new in news.items():
+            diff[name] = (0, new, new, 1.0)
+        for name, old in olds.items():
+            _, new, _, _ = diff[name]
+            diff[name] = (old, new, new-old, (new-old)/old if old else 1.0)
+        return diff
+
+    def print_header(by=''):
+        if not args.get('diff'):
+            print('%-36s %7s' % (by, 'size'))
+        else:
+            print('%-36s %7s %7s %7s' % (by, 'old', 'new', 'diff'))
+
+    def print_entries(by='function'):
+        entries = dedup_entries(results, by=by)
+
+        if not args.get('diff'):
+            print_header(by=by)
+            for name, size in sorted(entries.items()):
+                print("%-36s %7d" % (name, size))
+        else:
+            prev_entries = dedup_entries(prev_results, by=by)
+            diff = diff_entries(prev_entries, entries)
+            print_header(by='%s (%d added, %d removed)' % (by,
+                sum(1 for old, _, _, _ in diff.values() if not old),
+                sum(1 for _, new, _, _ in diff.values() if not new)))
+            for name, (old, new, diff, ratio) in sorted(diff.items(),
+                    key=lambda x: (-x[1][3], x)):
+                if ratio or args.get('all'):
+                    print("%-36s %7s %7s %+7d%s" % (name,
+                        old or "-",
+                        new or "-",
+                        diff,
+                        ' (%+.1f%%)' % (100*ratio) if ratio else ''))
+
+    def print_totals():
+        if not args.get('diff'):
+            print("%-36s %7d" % ('TOTAL', total))
+        else:
+            ratio = (total-prev_total)/prev_total if prev_total else 1.0
+            print("%-36s %7s %7s %+7d%s" % (
+                'TOTAL',
+                prev_total if prev_total else '-',
+                total if total else '-',
+                total-prev_total,
+                ' (%+.1f%%)' % (100*ratio) if ratio else ''))
+
+    if args.get('quiet'):
+        pass
+    elif args.get('summary'):
+        print_header()
+        print_totals()
+    elif args.get('files'):
+        print_entries(by='file')
+        print_totals()
+    else:
+        print_entries(by='function')
+        print_totals()
+
+if __name__ == "__main__":
+    import argparse
+    import sys
+    parser = argparse.ArgumentParser(
+        description="Find code size at the function level.")
+    parser.add_argument('obj_paths', nargs='*', default=OBJ_PATHS,
+        help="Description of where to find *.o files. May be a directory \
+            or a list of paths. Defaults to %r." % OBJ_PATHS)
+    parser.add_argument('-v', '--verbose', action='store_true',
+        help="Output commands that run behind the scenes.")
+    parser.add_argument('-o', '--output',
+        help="Specify CSV file to store results.")
+    parser.add_argument('-u', '--use',
+        help="Don't compile and find code sizes, instead use this CSV file.")
+    parser.add_argument('-d', '--diff',
+        help="Specify CSV file to diff code size against.")
+    parser.add_argument('-a', '--all', action='store_true',
+        help="Show all functions, not just the ones that changed.")
+    parser.add_argument('--files', action='store_true',
+        help="Show file-level code sizes. Note this does not include padding! "
+            "So sizes may differ from other tools.")
+    parser.add_argument('-s', '--summary', action='store_true',
+        help="Only show the total code size.")
+    parser.add_argument('-q', '--quiet', action='store_true',
+        help="Don't show anything, useful with -o.")
+    parser.add_argument('--type', default='tTrRdDbB',
+        help="Type of symbols to report, this uses the same single-character "
+            "type-names emitted by nm. Defaults to %(default)r.")
+    parser.add_argument('--nm-tool', default=['nm'], type=lambda x: x.split(),
+        help="Path to the nm tool to use.")
+    parser.add_argument('--build-dir',
+        help="Specify the relative build directory. Used to map object files \
+            to the correct source files.")
+    sys.exit(main(**vars(parser.parse_args())))
--- a/scripts/coverage.py
+++ b/scripts/coverage.py
@@ -0,0 +1,254 @@
+#!/usr/bin/env python3
+#
+# Parse and report coverage info from .info files generated by lcov
+#
+import os
+import glob
+import csv
+import re
+import collections as co
+import bisect as b
+
+
+INFO_PATHS = ['tests/*.toml.info']
+
+def collect(paths, **args):
+    file = None
+    funcs = []
+    lines = co.defaultdict(lambda: 0)
+    pattern = re.compile(
+        '^(?P<file>SF:/?(?P<file_name>.*))$'
+        '|^(?P<func>FN:(?P<func_lineno>[0-9]*),(?P<func_name>.*))$'
+        '|^(?P<line>DA:(?P<line_lineno>[0-9]*),(?P<line_hits>[0-9]*))$')
+    for path in paths:
+        with open(path) as f:
+            for line in f:
+                m = pattern.match(line)
+                if m and m.group('file'):
+                    file = m.group('file_name')
+                elif m and file and m.group('func'):
+                    funcs.append((file, int(m.group('func_lineno')),
+                        m.group('func_name')))
+                elif m and file and m.group('line'):
+                    lines[(file, int(m.group('line_lineno')))] += (
+                        int(m.group('line_hits')))
+
+    # map line numbers to functions
+    funcs.sort()
+    def func_from_lineno(file, lineno):
+        i = b.bisect(funcs, (file, lineno))
+        if i and funcs[i-1][0] == file:
+            return funcs[i-1][2]
+        else:
+            return None
+
+    # reduce to function info
+    reduced_funcs = co.defaultdict(lambda: (0, 0))
+    for (file, line_lineno), line_hits in lines.items():
+        func = func_from_lineno(file, line_lineno)
+        if not func:
+            continue
+        hits, count = reduced_funcs[(file, func)]
+        reduced_funcs[(file, func)] = (hits + (line_hits > 0), count + 1)
+
+    results = []
+    for (file, func), (hits, count) in reduced_funcs.items():
+        # discard internal/testing functions (test_* injected with
+        # internal testing)
+        if func.startswith('__') or func.startswith('test_'):
+            continue
+        # discard .8449 suffixes created by optimizer
+        func = re.sub('\.[0-9]+', '', func)
+        results.append((file, func, hits, count))
+
+    return results
+
+
+def main(**args):
+    # find coverage
+    if not args.get('use'):
+        # find *.info files
+        paths = []
+        for path in args['info_paths']:
+            if os.path.isdir(path):
+                path = path + '/*.gcov'
+
+            for path in glob.glob(path):
+                paths.append(path)
+
+        if not paths:
+            print('no .info files found in %r?' % args['info_paths'])
+            sys.exit(-1)
+
+        results = collect(paths, **args)
+    else:
+        with open(args['use']) as f:
+            r = csv.DictReader(f)
+            results = [
+                (   result['file'],
+                    result['function'],
+                    int(result['hits']),
+                    int(result['count']))
+                for result in r]
+
+    total_hits, total_count = 0, 0
+    for _, _, hits, count in results:
+        total_hits += hits
+        total_count += count
+
+    # find previous results?
+    if args.get('diff'):
+        with open(args['diff']) as f:
+            r = csv.DictReader(f)
+            prev_results = [
+                (   result['file'],
+                    result['function'],
+                    int(result['hits']),
+                    int(result['count']))
+                for result in r]
+
+        prev_total_hits, prev_total_count = 0, 0
+        for _, _, hits, count in prev_results:
+            prev_total_hits += hits
+            prev_total_count += count
+
+    # write results to CSV
+    if args.get('output'):
+        with open(args['output'], 'w') as f:
+            w = csv.writer(f)
+            w.writerow(['file', 'function', 'hits', 'count'])
+            for file, func, hits, count in sorted(results):
+                w.writerow((file, func, hits, count))
+
+    # print results
+    def dedup_entries(results, by='function'):
+        entries = co.defaultdict(lambda: (0, 0))
+        for file, func, hits, count in results:
+            entry = (file if by == 'file' else func)
+            entry_hits, entry_count = entries[entry]
+            entries[entry] = (entry_hits + hits, entry_count + count)
+        return entries
+
+    def diff_entries(olds, news):
+        diff = co.defaultdict(lambda: (0, 0, 0, 0, 0, 0, 0))
+        for name, (new_hits, new_count) in news.items():
+            diff[name] = (
+                0, 0,
+                new_hits, new_count,
+                new_hits, new_count,
+                (new_hits/new_count if new_count else 1.0) - 1.0)
+        for name, (old_hits, old_count) in olds.items():
+            _, _, new_hits, new_count, _, _, _ = diff[name]
+            diff[name] = (
+                old_hits, old_count,
+                new_hits, new_count,
+                new_hits-old_hits, new_count-old_count,
+                ((new_hits/new_count if new_count else 1.0)
+                    - (old_hits/old_count if old_count else 1.0)))
+        return diff
+
+    def print_header(by=''):
+        if not args.get('diff'):
+            print('%-36s %19s' % (by, 'hits/line'))
+        else:
+            print('%-36s %19s %19s %11s' % (by, 'old', 'new', 'diff'))
+
+    def print_entries(by='function'):
+        entries = dedup_entries(results, by=by)
+
+        if not args.get('diff'):
+            print_header(by=by)
+            for name, (hits, count) in sorted(entries.items()):
+                print("%-36s %11s %7s" % (name,
+                    '%d/%d' % (hits, count)
+                        if count else '-',
+                    '%.1f%%' % (100*hits/count)
+                        if count else '-'))
+        else:
+            prev_entries = dedup_entries(prev_results, by=by)
+            diff = diff_entries(prev_entries, entries)
+            print_header(by='%s (%d added, %d removed)' % (by,
+                sum(1 for _, old, _, _, _, _, _ in diff.values() if not old),
+                sum(1 for _, _, _, new, _, _, _ in diff.values() if not new)))
+            for name, (
+                    old_hits, old_count,
+                    new_hits, new_count,
+                    diff_hits, diff_count, ratio) in sorted(diff.items(),
+                        key=lambda x: (-x[1][6], x)):
+                if ratio or args.get('all'):
+                    print("%-36s %11s %7s %11s %7s %11s%s" % (name,
+                        '%d/%d' % (old_hits, old_count)
+                            if old_count else '-',
+                        '%.1f%%' % (100*old_hits/old_count)
+                            if old_count else '-',
+                        '%d/%d' % (new_hits, new_count)
+                            if new_count else '-',
+                        '%.1f%%' % (100*new_hits/new_count)
+                            if new_count else '-',
+                        '%+d/%+d' % (diff_hits, diff_count),
+                        ' (%+.1f%%)' % (100*ratio) if ratio else ''))
+
+    def print_totals():
+        if not args.get('diff'):
+            print("%-36s %11s %7s" % ('TOTAL',
+                '%d/%d' % (total_hits, total_count)
+                    if total_count else '-',
+                '%.1f%%' % (100*total_hits/total_count)
+                    if total_count else '-'))
+        else:
+            ratio = ((total_hits/total_count
+                    if total_count else 1.0)
+                - (prev_total_hits/prev_total_count
+                    if prev_total_count else 1.0))
+            print("%-36s %11s %7s %11s %7s %11s%s" % ('TOTAL',
+                '%d/%d' % (prev_total_hits, prev_total_count)
+                    if prev_total_count else '-',
+                '%.1f%%' % (100*prev_total_hits/prev_total_count)
+                    if prev_total_count else '-',
+                '%d/%d' % (total_hits, total_count)
+                    if total_count else '-',
+                '%.1f%%' % (100*total_hits/total_count)
+                    if total_count else '-',
+                '%+d/%+d' % (total_hits-prev_total_hits,
+                    total_count-prev_total_count),
+                ' (%+.1f%%)' % (100*ratio) if ratio else ''))
+
+    if args.get('quiet'):
+        pass
+    elif args.get('summary'):
+        print_header()
+        print_totals()
+    elif args.get('files'):
+        print_entries(by='file')
+        print_totals()
+    else:
+        print_entries(by='function')
+        print_totals()
+
+if __name__ == "__main__":
+    import argparse
+    import sys
+    parser = argparse.ArgumentParser(
+        description="Parse and report coverage info from .info files \
+            generated by lcov")
+    parser.add_argument('info_paths', nargs='*', default=INFO_PATHS,
+        help="Description of where to find *.info files. May be a directory \
+            or list of paths. *.info files will be merged to show the total \
+            coverage. Defaults to %r." % INFO_PATHS)
+    parser.add_argument('-v', '--verbose', action='store_true',
+        help="Output commands that run behind the scenes.")
+    parser.add_argument('-o', '--output',
+        help="Specify CSV file to store results.")
+    parser.add_argument('-u', '--use',
+        help="Don't do any work, instead use this CSV file.")
+    parser.add_argument('-d', '--diff',
+        help="Specify CSV file to diff code size against.")
+    parser.add_argument('-a', '--all', action='store_true',
+        help="Show all functions, not just the ones that changed.")
+    parser.add_argument('--files', action='store_true',
+        help="Show file-level coverage.")
+    parser.add_argument('-s', '--summary', action='store_true',
+        help="Only show the total coverage.")
+    parser.add_argument('-q', '--quiet', action='store_true',
+        help="Don't show anything, useful with -o.")
+    sys.exit(main(**vars(parser.parse_args())))
--- a/scripts/test.py
+++ b/scripts/test.py
@@ -20,19 +20,50 @@ import pty
 import errno
 import signal

-TESTDIR = 'tests'
+TEST_PATHS = 'tests'
 RULES = """
+# add block devices to sources
+TESTSRC ?= $(SRC) $(wildcard bd/*.c)
+
 define FLATTEN
-tests/%$(subst /,.,$(target)): $(target)
+%(path)s%%$(subst /,.,$(target)): $(target)
    ./scripts/explode_asserts.py $$< -o $$@
 endef
-$(foreach target,$(SRC),$(eval $(FLATTEN)))
-
-include tests/*.d
+$(foreach target,$(TESTSRC),$(eval $(FLATTEN)))

+-include %(path)s*.d
 .SECONDARY:
-%.test: %.test.o $(foreach f,$(subst /,.,$(SRC:.c=.o)),%.$f)
+
+%(path)s.test: %(path)s.test.o \\
+        $(foreach t,$(subst /,.,$(TESTSRC:.c=.o)),%(path)s.$t)
    $(CC) $(CFLAGS) $^ $(LFLAGS) -o $@
+
+# needed in case builddir is different
+%(path)s%%.o: %(path)s%%.c
+    $(CC) -c -MMD $(CFLAGS) $< -o $@
+"""
+COVERAGE_RULES = """
+%(path)s.test: override CFLAGS += -fprofile-arcs -ftest-coverage
+
+# delete lingering coverage
+%(path)s.test: | %(path)s.info.clean
+.PHONY: %(path)s.info.clean
+%(path)s.info.clean:
+    rm -f %(path)s*.gcda
+
+# accumulate coverage info
+.PHONY: %(path)s.info
+%(path)s.info:
+    $(strip $(LCOV) -c \\
+        $(addprefix -d ,$(wildcard %(path)s*.gcda)) \\
+        --rc 'geninfo_adjust_src_path=$(shell pwd)' \\
+        -o $@)
+    $(LCOV) -e $@ $(addprefix /,$(SRC)) -o $@
+ifdef COVERAGETARGET
+    $(strip $(LCOV) -a $@ \\
+        $(addprefix -a ,$(wildcard $(COVERAGETARGET))) \\
+        -o $(COVERAGETARGET))
+endif
 """
 GLOBALS = """
 //////////////// AUTOGENERATED TEST ////////////////
@@ -119,6 +150,8 @@ class TestCase:
        self.if_ = config.get('if', None)
        self.in_ = config.get('in', None)

+        self.result = None
+
    def __str__(self):
        if hasattr(self, 'permno'):
            if any(k not in self.case.defines for k in self.defines):
@@ -179,7 +212,7 @@ class TestCase:
                len(self.filter) >= 2 and
                self.filter[1] != self.permno):
            return False
-        elif args.get('no_internal', False) and self.in_ is not None:
+        elif args.get('no_internal') and self.in_ is not None:
            return False
        elif self.if_ is not None:
            if_ = self.if_
@@ -213,7 +246,7 @@ class TestCase:
                try:
                    with open(disk, 'w') as f:
                        f.truncate(0)
-                    if args.get('verbose', False):
+                    if args.get('verbose'):
                        print('truncate --size=0', disk)
                except FileNotFoundError:
                    pass
@@ -237,14 +270,14 @@ class TestCase:
                    '-ex', 'r'])
            ncmd.extend(['--args'] + cmd)

-            if args.get('verbose', False):
+            if args.get('verbose'):
                print(' '.join(shlex.quote(c) for c in ncmd))
            signal.signal(signal.SIGINT, signal.SIG_IGN)
            sys.exit(sp.call(ncmd))

        # run test case!
        mpty, spty = pty.openpty()
-        if args.get('verbose', False):
+        if args.get('verbose'):
            print(' '.join(shlex.quote(c) for c in cmd))
        proc = sp.Popen(cmd, stdout=spty, stderr=spty)
        os.close(spty)
@@ -260,7 +293,7 @@ class TestCase:
                        break
                    raise
                stdout.append(line)
-                if args.get('verbose', False):
+                if args.get('verbose'):
                    sys.stdout.write(line)
                # intercept asserts
                m = re.match(
@@ -299,7 +332,7 @@ class ValgrindTestCase(TestCase):
        return not self.leaky and super().shouldtest(**args)

    def test(self, exec=[], **args):
-        verbose = args.get('verbose', False)
+        verbose = args.get('verbose')
        uninit = (self.defines.get('LFS_ERASE_VALUE', None) == -1)
        exec = [
            'valgrind',
@@ -351,12 +384,17 @@ class TestSuite:
        self.name = os.path.basename(path)
        if self.name.endswith('.toml'):
            self.name = self.name[:-len('.toml')]
-        self.path = path
+        if args.get('build_dir'):
+            self.toml = path
+            self.path = args['build_dir'] + '/' + path
+        else:
+            self.toml = path
+            self.path = path
        self.classes = classes
        self.defines = defines.copy()
        self.filter = filter

-        with open(path) as f:
+        with open(self.toml) as f:
            # load tests
            config = toml.load(f)

@@ -467,7 +505,7 @@ class TestSuite:

    def build(self, **args):
        # build test files
-        tf = open(self.path + '.test.c.t', 'w')
+        tf = open(self.path + '.test.tc', 'w')
        tf.write(GLOBALS)
        if self.code is not None:
            tf.write('#line %d "%s"\n' % (self.code_lineno, self.path))
@@ -477,7 +515,7 @@ class TestSuite:
        for case in self.cases:
            if case.in_ not in tfs:
                tfs[case.in_] = open(self.path+'.'+
-                    case.in_.replace('/', '.')+'.t', 'w')
+                    re.sub('(\.c)?$', '.tc', case.in_.replace('/', '.')), 'w')
                tfs[case.in_].write('#line 1 "%s"\n' % case.in_)
                with open(case.in_) as f:
                    for line in f:
@@ -516,25 +554,33 @@ class TestSuite:

        # write makefiles
        with open(self.path + '.mk', 'w') as mk:
-            mk.write(RULES.replace(4*' ', '\t'))
+            mk.write(RULES.replace(4*' ', '\t') % dict(path=self.path))
            mk.write('\n')

+            # add coverage hooks?
+            if args.get('coverage'):
+                mk.write(COVERAGE_RULES.replace(4*' ', '\t') % dict(
+                    path=self.path))
+                mk.write('\n')
+
            # add truely global defines globally
            for k, v in sorted(self.defines.items()):
-                mk.write('%s: override CFLAGS += -D%s=%r\n' % (
-                    self.path+'.test', k, v))
+                mk.write('%s.test: override CFLAGS += -D%s=%r\n'
+                    % (self.path, k, v))

            for path in tfs:
                if path is None:
                    mk.write('%s: %s | %s\n' % (
                        self.path+'.test.c',
-                        self.path,
-                        self.path+'.test.c.t'))
+                        self.toml,
+                        self.path+'.test.tc'))
                else:
                    mk.write('%s: %s %s | %s\n' % (
                        self.path+'.'+path.replace('/', '.'),
-                        self.path, path,
-                        self.path+'.'+path.replace('/', '.')+'.t'))
+                        self.toml,
+                        path,
+                        self.path+'.'+re.sub('(\.c)?$', '.tc',
+                            path.replace('/', '.'))))
                mk.write('\t./scripts/explode_asserts.py $| -o $@\n')

        self.makefile = self.path + '.mk'
@@ -557,7 +603,7 @@ class TestSuite:
                if not args.get('verbose', True):
                    sys.stdout.write(FAIL)
                    sys.stdout.flush()
-                if not args.get('keep_going', False):
+                if not args.get('keep_going'):
                    if not args.get('verbose', True):
                        sys.stdout.write('\n')
                    raise
@@ -579,30 +625,30 @@ def main(**args):

    # and what class of TestCase to run
    classes = []
-    if args.get('normal', False):
+    if args.get('normal'):
        classes.append(TestCase)
-    if args.get('reentrant', False):
+    if args.get('reentrant'):
        classes.append(ReentrantTestCase)
-    if args.get('valgrind', False):
+    if args.get('valgrind'):
        classes.append(ValgrindTestCase)
    if not classes:
        classes = [TestCase]

    suites = []
-    for testpath in args['testpaths']:
+    for testpath in args['test_paths']:
        # optionally specified test case/perm
        testpath, *filter = testpath.split('#')
        filter = [int(f) for f in filter]

        # figure out the suite's toml file
        if os.path.isdir(testpath):
-            testpath = testpath + '/test_*.toml'
+            testpath = testpath + '/*.toml'
        elif os.path.isfile(testpath):
            testpath = testpath
        elif testpath.endswith('.toml'):
-            testpath = TESTDIR + '/' + testpath
+            testpath = TEST_PATHS + '/' + testpath
        else:
-            testpath = TESTDIR + '/' + testpath + '.toml'
+            testpath = TEST_PATHS + '/' + testpath + '.toml'

        # find tests
        for path in glob.glob(testpath):
@@ -628,7 +674,7 @@ def main(**args):
        list(it.chain.from_iterable(['-f', m] for m in makefiles)) +
        [target for target in targets])
    mpty, spty = pty.openpty()
-    if args.get('verbose', False):
+    if args.get('verbose'):
        print(' '.join(shlex.quote(c) for c in cmd))
    proc = sp.Popen(cmd, stdout=spty, stderr=spty)
    os.close(spty)
@@ -642,14 +688,14 @@ def main(**args):
                break
            raise
        stdout.append(line)
-        if args.get('verbose', False):
+        if args.get('verbose'):
            sys.stdout.write(line)
        # intercept warnings
        m = re.match(
            '^{0}([^:]+):(\d+):(?:\d+:)?{0}{1}:{0}(.*)$'
            .format('(?:\033\[[\d;]*.| )*', 'warning'),
            line)
-        if m and not args.get('verbose', False):
+        if m and not args.get('verbose'):
            try:
                with open(m.group(1)) as f:
                    lineno = int(m.group(2))
@@ -662,27 +708,26 @@ def main(**args):
            except:
                pass
    proc.wait()
-
    if proc.returncode != 0:
-        if not args.get('verbose', False):
+        if not args.get('verbose'):
            for line in stdout:
                sys.stdout.write(line)
-        sys.exit(-3)
+        sys.exit(-1)

    print('built %d test suites, %d test cases, %d permutations' % (
        len(suites),
        sum(len(suite.cases) for suite in suites),
        sum(len(suite.perms) for suite in suites)))

-    filtered = 0
+    total = 0
    for suite in suites:
        for perm in suite.perms:
-            filtered += perm.shouldtest(**args)
-    if filtered != sum(len(suite.perms) for suite in suites):
-        print('filtered down to %d permutations' % filtered)
+            total += perm.shouldtest(**args)
+    if total != sum(len(suite.perms) for suite in suites):
+        print('filtered down to %d permutations' % total)

    # only requested to build?
-    if args.get('build', False):
+    if args.get('build'):
        return 0

    print('====== testing ======')
@@ -697,15 +742,12 @@ def main(**args):
    failed = 0
    for suite in suites:
        for perm in suite.perms:
-            if not hasattr(perm, 'result'):
-                continue
-
            if perm.result == PASS:
                passed += 1
-            else:
+            elif isinstance(perm.result, TestFailure):
                sys.stdout.write(
                    "\033[01m{path}:{lineno}:\033[01;31mfailure:\033[m "
-                    "{perm} failed with {returncode}\n".format(
+                    "{perm} failed\n".format(
                        perm=perm, path=perm.suite.path, lineno=perm.lineno,
                        returncode=perm.result.returncode or 0))
                if perm.result.stdout:
@@ -723,11 +765,33 @@ def main(**args):
                sys.stdout.write('\n')
                failed += 1

-    if args.get('gdb', False):
+    if args.get('coverage'):
+        # collect coverage info
+        # why -j1? lcov doesn't work in parallel because of gcov limitations
+        cmd = (['make', '-j1', '-f', 'Makefile'] +
+            list(it.chain.from_iterable(['-f', m] for m in makefiles)) +
+            (['COVERAGETARGET=%s' % args['coverage']]
+                if isinstance(args['coverage'], str) else []) +
+            [suite.path + '.info' for suite in suites
+                if any(perm.result == PASS for perm in suite.perms)])
+        if args.get('verbose'):
+            print(' '.join(shlex.quote(c) for c in cmd))
+        proc = sp.Popen(cmd,
+            stdout=sp.PIPE if not args.get('verbose') else None,
+            stderr=sp.STDOUT if not args.get('verbose') else None,
+            universal_newlines=True)
+        proc.wait()
+        if proc.returncode != 0:
+            if not args.get('verbose'):
+                for line in proc.stdout:
+                    sys.stdout.write(line)
+            sys.exit(-1)
+
+    if args.get('gdb'):
        failure = None
        for suite in suites:
            for perm in suite.perms:
-                if getattr(perm, 'result', PASS) != PASS:
+                if isinstance(perm.result, TestFailure):
                    failure = perm.result
        if failure is not None:
            print('======= gdb ======')
@@ -735,20 +799,22 @@ def main(**args):
            failure.case.test(failure=failure, **args)
            sys.exit(0)

-    print('tests passed: %d' % passed)
-    print('tests failed: %d' % failed)
+    print('tests passed %d/%d (%.2f%%)' % (passed, total,
+        100*(passed/total if total else 1.0)))
+    print('tests failed %d/%d (%.2f%%)' % (failed, total,
+        100*(failed/total if total else 1.0)))
    return 1 if failed > 0 else 0

 if __name__ == "__main__":
    import argparse
    parser = argparse.ArgumentParser(
        description="Run parameterized tests in various configurations.")
-    parser.add_argument('testpaths', nargs='*', default=[TESTDIR],
+    parser.add_argument('test_paths', nargs='*', default=[TEST_PATHS],
        help="Description of test(s) to run. By default, this is all tests \
            found in the \"{0}\" directory. Here, you can specify a different \
-            directory of tests, a specific file, a suite by name, and even a \
-            specific test case by adding brackets. For example \
-            \"test_dirs[0]\" or \"{0}/test_dirs.toml[0]\".".format(TESTDIR))
+            directory of tests, a specific file, a suite by name, and even \
+            specific test cases and permutations. For example \
+            \"test_dirs#1\" or \"{0}/test_dirs.toml#1#1\".".format(TEST_PATHS))
    parser.add_argument('-D', action='append', default=[],
        help="Overriding parameter definitions.")
    parser.add_argument('-v', '--verbose', action='store_true',
@@ -769,10 +835,19 @@ if __name__ == "__main__":
        help="Run tests normally.")
    parser.add_argument('-r', '--reentrant', action='store_true',
        help="Run reentrant tests with simulated power-loss.")
-    parser.add_argument('-V', '--valgrind', action='store_true',
+    parser.add_argument('--valgrind', action='store_true',
        help="Run non-leaky tests under valgrind to check for memory leaks.")
-    parser.add_argument('-e', '--exec', default=[], type=lambda e: e.split(' '),
+    parser.add_argument('--exec', default=[], type=lambda e: e.split(),
        help="Run tests with another executable prefixed on the command line.")
-    parser.add_argument('-d', '--disk',
+    parser.add_argument('--disk',
        help="Specify a file to use for persistent/reentrant tests.")
+    parser.add_argument('--coverage', type=lambda x: x if x else True,
+        nargs='?', const='',
+        help="Collect coverage information during testing. This uses lcov/gcov \
+            to accumulate coverage information into *.info files. May also \
+            a path to a *.info file to accumulate coverage info into.")
+    parser.add_argument('--build-dir',
+        help="Build relative to the specified directory instead of the \
+            current directory.")
+
    sys.exit(main(**vars(parser.parse_args())))
--- a/tests/test_relocations_with_removeall.toml
+++ b/tests/test_relocations_with_removeall.toml
@@ -0,0 +1,305 @@
+# specific corner cases worth explicitly testing for
+[[case]] # dangling split dir test
+define.ITERATIONS = 20
+define.COUNT = 10
+define.LFS_BLOCK_CYCLES = [8, 1]
+code = '''
+    lfs_format(&lfs, &cfg) => 0;
+    // fill up filesystem so only ~16 blocks are left
+    lfs_mount(&lfs, &cfg) => 0;
+    lfs_file_open(&lfs, &file, "padding", LFS_O_CREAT | LFS_O_WRONLY) => 0;
+    memset(buffer, 0, 512);
+    while (LFS_BLOCK_COUNT - lfs_fs_size(&lfs) > 16) {
+        lfs_file_write(&lfs, &file, buffer, 512) => 512;
+    }
+    lfs_file_close(&lfs, &file) => 0;
+    // make a child dir to use in bounded space
+    lfs_mkdir(&lfs, "child") => 0;
+    lfs_unmount(&lfs) => 0;
+
+    lfs_mount(&lfs, &cfg) => 0;
+    for (int j = 0; j < ITERATIONS; j++) {
+        for (int i = 0; i < COUNT; i++) {
+            sprintf(path, "child/test%03d_loooooooooooooooooong_name", i);
+            lfs_file_open(&lfs, &file, path, LFS_O_CREAT | LFS_O_WRONLY) => 0;
+            lfs_file_close(&lfs, &file) => 0;
+        }
+
+        lfs_dir_open(&lfs, &dir, "child") => 0;
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        for (int i = 0; i < COUNT; i++) {
+            sprintf(path, "test%03d_loooooooooooooooooong_name", i);
+            lfs_dir_read(&lfs, &dir, &info) => 1;
+            strcmp(info.name, path) => 0;
+        }
+        lfs_dir_read(&lfs, &dir, &info) => 0;
+        lfs_dir_close(&lfs, &dir) => 0;
+
+        if (j == ITERATIONS-1) {
+            break;
+        }
+
+        for (int i = 0; i < COUNT; i++) {
+            sprintf(path, "child/test%03d_loooooooooooooooooong_name", i);
+            lfs_removeall(&lfs, path) => 0;
+        }
+    }
+    lfs_unmount(&lfs) => 0;
+
+    lfs_mount(&lfs, &cfg) => 0;
+    lfs_dir_open(&lfs, &dir, "child") => 0;
+    lfs_dir_read(&lfs, &dir, &info) => 1;
+    lfs_dir_read(&lfs, &dir, &info) => 1;
+    for (int i = 0; i < COUNT; i++) {
+        sprintf(path, "test%03d_loooooooooooooooooong_name", i);
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        strcmp(info.name, path) => 0;
+    }
+    lfs_dir_read(&lfs, &dir, &info) => 0;
+    lfs_dir_close(&lfs, &dir) => 0;
+    for (int i = 0; i < COUNT; i++) {
+        sprintf(path, "child/test%03d_loooooooooooooooooong_name", i);
+        lfs_removeall(&lfs, path) => 0;
+    }
+    lfs_unmount(&lfs) => 0;
+'''
+
+[[case]] # outdated head test
+define.ITERATIONS = 20
+define.COUNT = 10
+define.LFS_BLOCK_CYCLES = [8, 1]
+code = '''
+    lfs_format(&lfs, &cfg) => 0;
+    // fill up filesystem so only ~16 blocks are left
+    lfs_mount(&lfs, &cfg) => 0;
+    lfs_file_open(&lfs, &file, "padding", LFS_O_CREAT | LFS_O_WRONLY) => 0;
+    memset(buffer, 0, 512);
+    while (LFS_BLOCK_COUNT - lfs_fs_size(&lfs) > 16) {
+        lfs_file_write(&lfs, &file, buffer, 512) => 512;
+    }
+    lfs_file_close(&lfs, &file) => 0;
+    // make a child dir to use in bounded space
+    lfs_mkdir(&lfs, "child") => 0;
+    lfs_unmount(&lfs) => 0;
+
+    lfs_mount(&lfs, &cfg) => 0;
+    for (int j = 0; j < ITERATIONS; j++) {
+        for (int i = 0; i < COUNT; i++) {
+            sprintf(path, "child/test%03d_loooooooooooooooooong_name", i);
+            lfs_file_open(&lfs, &file, path, LFS_O_CREAT | LFS_O_WRONLY) => 0;
+            lfs_file_close(&lfs, &file) => 0;
+        }
+
+        lfs_dir_open(&lfs, &dir, "child") => 0;
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        for (int i = 0; i < COUNT; i++) {
+            sprintf(path, "test%03d_loooooooooooooooooong_name", i);
+            lfs_dir_read(&lfs, &dir, &info) => 1;
+            strcmp(info.name, path) => 0;
+            info.size => 0;
+
+            sprintf(path, "child/test%03d_loooooooooooooooooong_name", i);
+            lfs_file_open(&lfs, &file, path, LFS_O_WRONLY) => 0;
+            lfs_file_write(&lfs, &file, "hi", 2) => 2;
+            lfs_file_close(&lfs, &file) => 0;
+        }
+        lfs_dir_read(&lfs, &dir, &info) => 0;
+
+        lfs_dir_rewind(&lfs, &dir) => 0;
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        for (int i = 0; i < COUNT; i++) {
+            sprintf(path, "test%03d_loooooooooooooooooong_name", i);
+            lfs_dir_read(&lfs, &dir, &info) => 1;
+            strcmp(info.name, path) => 0;
+            info.size => 2;
+
+            sprintf(path, "child/test%03d_loooooooooooooooooong_name", i);
+            lfs_file_open(&lfs, &file, path, LFS_O_WRONLY) => 0;
+            lfs_file_write(&lfs, &file, "hi", 2) => 2;
+            lfs_file_close(&lfs, &file) => 0;
+        }
+        lfs_dir_read(&lfs, &dir, &info) => 0;
+
+        lfs_dir_rewind(&lfs, &dir) => 0;
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        lfs_dir_read(&lfs, &dir, &info) => 1;
+        for (int i = 0; i < COUNT; i++) {
+            sprintf(path, "test%03d_loooooooooooooooooong_name", i);
+            lfs_dir_read(&lfs, &dir, &info) => 1;
+            strcmp(info.name, path) => 0;
+            info.size => 2;
+        }
+        lfs_dir_read(&lfs, &dir, &info) => 0;
+        lfs_dir_close(&lfs, &dir) => 0;
+
+        for (int i = 0; i < COUNT; i++) {
+            sprintf(path, "child/test%03d_loooooooooooooooooong_name", i);
+            lfs_removeall(&lfs, path) => 0;
+        }
+    }
+    lfs_unmount(&lfs) => 0;
+'''
+
+[[case]] # reentrant testing for relocations, this is the same as the
+         # orphan testing, except here we also set block_cycles so that
+         # almost every tree operation needs a relocation
+reentrant = true
+# TODO fix this case, caused by non-DAG trees
+if = '!(DEPTH == 3 && LFS_CACHE_SIZE != 64)'
+define = [
+    {FILES=6,  DEPTH=1, CYCLES=20, LFS_BLOCK_CYCLES=1},
+    {FILES=26, DEPTH=1, CYCLES=20, LFS_BLOCK_CYCLES=1},
+    {FILES=3,  DEPTH=3, CYCLES=20, LFS_BLOCK_CYCLES=1},
+]
+code = '''
+    err = lfs_mount(&lfs, &cfg);
+    if (err) {
+        lfs_format(&lfs, &cfg) => 0;
+        lfs_mount(&lfs, &cfg) => 0;
+    }
+
+    srand(1);
+    const char alpha[] = "abcdefghijklmnopqrstuvwxyz";
+    for (int i = 0; i < CYCLES; i++) {
+        // create random path
+        char full_path[256];
+        for (int d = 0; d < DEPTH; d++) {
+            sprintf(&full_path[2*d], "/%c", alpha[rand() % FILES]);
+        }
+
+        // if it does not exist, we create it, else we destroy
+        int res = lfs_stat(&lfs, full_path, &info);
+        if (res == LFS_ERR_NOENT) {
+            // create each directory in turn, ignore if dir already exists
+            for (int d = 0; d < DEPTH; d++) {
+                strcpy(path, full_path);
+                path[2*d+2] = '\0';
+                err = lfs_mkdir(&lfs, path);
+                assert(!err || err == LFS_ERR_EXIST);
+            }
+
+            for (int d = 0; d < DEPTH; d++) {
+                strcpy(path, full_path);
+                path[2*d+2] = '\0';
+                lfs_stat(&lfs, path, &info) => 0;
+                assert(strcmp(info.name, &path[2*d+1]) == 0);
+                assert(info.type == LFS_TYPE_DIR);
+            }
+        } else {
+            // is valid dir?
+            assert(strcmp(info.name, &full_path[2*(DEPTH-1)+1]) == 0);
+            assert(info.type == LFS_TYPE_DIR);
+
+            // try to delete path in reverse order, ignore if dir is not empty
+            for (int d = DEPTH-1; d >= 0; d--) {
+                strcpy(path, full_path);
+                path[2*d+2] = '\0';
+                err = lfs_removeall(&lfs, path);
+                assert(!err || err == LFS_ERR_NOTEMPTY);
+            }
+
+            lfs_stat(&lfs, full_path, &info) => LFS_ERR_NOENT;
+        }
+    }
+    lfs_unmount(&lfs) => 0;
+'''
+
+[[case]] # reentrant testing for relocations, but now with random renames!
+reentrant = true
+# TODO fix this case, caused by non-DAG trees
+if = '!(DEPTH == 3 && LFS_CACHE_SIZE != 64)'
+define = [
+    {FILES=6,  DEPTH=1, CYCLES=20, LFS_BLOCK_CYCLES=1},
+    {FILES=26, DEPTH=1, CYCLES=20, LFS_BLOCK_CYCLES=1},
+    {FILES=3,  DEPTH=3, CYCLES=20, LFS_BLOCK_CYCLES=1},
+]
+code = '''
+    err = lfs_mount(&lfs, &cfg);
+    if (err) {
+        lfs_format(&lfs, &cfg) => 0;
+        lfs_mount(&lfs, &cfg) => 0;
+    }
+
+    srand(1);
+    const char alpha[] = "abcdefghijklmnopqrstuvwxyz";
+    for (int i = 0; i < CYCLES; i++) {
+        // create random path
+        char full_path[256];
+        for (int d = 0; d < DEPTH; d++) {
+            sprintf(&full_path[2*d], "/%c", alpha[rand() % FILES]);
+        }
+
+        // if it does not exist, we create it, else we destroy
+        int res = lfs_stat(&lfs, full_path, &info);
+        assert(!res || res == LFS_ERR_NOENT);
+        if (res == LFS_ERR_NOENT) {
+            // create each directory in turn, ignore if dir already exists
+            for (int d = 0; d < DEPTH; d++) {
+                strcpy(path, full_path);
+                path[2*d+2] = '\0';
+                err = lfs_mkdir(&lfs, path);
+                assert(!err || err == LFS_ERR_EXIST);
+            }
+
+            for (int d = 0; d < DEPTH; d++) {
+                strcpy(path, full_path);
+                path[2*d+2] = '\0';
+                lfs_stat(&lfs, path, &info) => 0;
+                assert(strcmp(info.name, &path[2*d+1]) == 0);
+                assert(info.type == LFS_TYPE_DIR);
+            }
+        } else {
+            assert(strcmp(info.name, &full_path[2*(DEPTH-1)+1]) == 0);
+            assert(info.type == LFS_TYPE_DIR);
+
+            // create new random path
+            char new_path[256];
+            for (int d = 0; d < DEPTH; d++) {
+                sprintf(&new_path[2*d], "/%c", alpha[rand() % FILES]);
+            }
+
+            // if new path does not exist, rename, otherwise destroy
+            res = lfs_stat(&lfs, new_path, &info);
+            assert(!res || res == LFS_ERR_NOENT);
+            if (res == LFS_ERR_NOENT) {
+                // stop once some dir is renamed
+                for (int d = 0; d < DEPTH; d++) {
+                    strcpy(&path[2*d], &full_path[2*d]);
+                    path[2*d+2] = '\0';
+                    strcpy(&path[128+2*d], &new_path[2*d]);
+                    path[128+2*d+2] = '\0';
+                    err = lfs_rename(&lfs, path, path+128);
+                    assert(!err || err == LFS_ERR_NOTEMPTY);
+                    if (!err) {
+                        strcpy(path, path+128);
+                    }
+                }
+
+                for (int d = 0; d < DEPTH; d++) {
+                    strcpy(path, new_path);
+                    path[2*d+2] = '\0';
+                    lfs_stat(&lfs, path, &info) => 0;
+                    assert(strcmp(info.name, &path[2*d+1]) == 0);
+                    assert(info.type == LFS_TYPE_DIR);
+                }
+                
+                lfs_stat(&lfs, full_path, &info) => LFS_ERR_NOENT;
+            } else {
+                // try to delete path in reverse order,
+                // ignore if dir is not empty
+                for (int d = DEPTH-1; d >= 0; d--) {
+                    strcpy(path, full_path);
+                    path[2*d+2] = '\0';
+                    err = lfs_removeall(&lfs, path);
+                    assert(!err || err == LFS_ERR_NOTEMPTY);
+                }
+
+                lfs_stat(&lfs, full_path, &info) => LFS_ERR_NOENT;
+            }
+        }
+    }
+    lfs_unmount(&lfs) => 0;
+'''
--- a/tests/test_truncate.toml
+++ b/tests/test_truncate.toml
@@ -392,3 +392,48 @@ code = '''

    lfs_unmount(&lfs) => 0;
 '''
+
+[[case]] # noop truncate
+define.MEDIUMSIZE = [32, 2048]
+code = '''
+    lfs_format(&lfs, &cfg) => 0;
+    lfs_mount(&lfs, &cfg) => 0;
+    lfs_file_open(&lfs, &file, "baldynoop",
+            LFS_O_RDWR | LFS_O_CREAT) => 0;
+
+    strcpy((char*)buffer, "hair");
+    size = strlen((char*)buffer);
+    for (lfs_off_t j = 0; j < MEDIUMSIZE; j += size) {
+        lfs_file_write(&lfs, &file, buffer, size) => size;
+
+        // this truncate should do nothing
+        lfs_file_truncate(&lfs, &file, j+size) => 0;
+    }
+    lfs_file_size(&lfs, &file) => MEDIUMSIZE;
+
+    lfs_file_seek(&lfs, &file, 0, LFS_SEEK_SET) => 0;
+    // should do nothing again
+    lfs_file_truncate(&lfs, &file, MEDIUMSIZE) => 0;
+    lfs_file_size(&lfs, &file) => MEDIUMSIZE;
+
+    for (lfs_off_t j = 0; j < MEDIUMSIZE; j += size) {
+        lfs_file_read(&lfs, &file, buffer, size) => size;
+        memcmp(buffer, "hair", size) => 0;
+    }
+    lfs_file_read(&lfs, &file, buffer, size) => 0;
+
+    lfs_file_close(&lfs, &file) => 0;
+    lfs_unmount(&lfs) => 0;
+
+    // still there after reboot?
+    lfs_mount(&lfs, &cfg) => 0;
+    lfs_file_open(&lfs, &file, "baldynoop", LFS_O_RDWR) => 0;
+    lfs_file_size(&lfs, &file) => MEDIUMSIZE;
+    for (lfs_off_t j = 0; j < MEDIUMSIZE; j += size) {
+        lfs_file_read(&lfs, &file, buffer, size) => size;
+        memcmp(buffer, "hair", size) => 0;
+    }
+    lfs_file_read(&lfs, &file, buffer, size) => 0;
+    lfs_file_close(&lfs, &file) => 0;
+    lfs_unmount(&lfs) => 0;
+'''
Author	SHA1	Message	Date
Tim Nordell	e3ace9ce98	fixup for `5c87b6864` Duplicate test_relocation.toml into a variant with removeall.	2021-06-13 21:40:36 -05:00
Tim Nordell	40bebef368	fixup for `5c87b6864` Add in lfs_removeall(...) and lfs_rename_with_removeall(...).	2021-06-13 21:39:56 -05:00
Tim Nordell	2612ef130b	fixup for `5c87b6864` Move the logic for determining if we should support split directories into a helper function. This permits compiled outputs that do not utilize this to drop this code easily at the linker level. Additionally, drop the default behavior.	2021-06-13 21:38:37 -05:00
Tim Nordell	a3ea209503	fixup for `5c87b6864` Add in blockers for the normal POSIX like entry into the removal and rename code; pending naming of the actual function.	2021-06-13 21:11:10 -05:00
Tim Nordell	45d02e5750	fixup for `5c87b6864` Add in missing global state xor for each sub page within a folder.	2021-06-13 21:10:25 -05:00
Tim Nordell	5c87b6864f	Support removal of entire directories that contain only files Removing directories containing other directories is problematic from an orphan perspective, but removing a directory only containing files is fairly easy to do. Add logic that checks if a directory only contains files, and if it does, permit removal of this directory.	2021-06-10 20:12:09 -05:00
Tim Nordell	8c1931ac55	Consolidate some of the logic for removals of folders This routine checks that the folder is empty before removing and is found in two places. We can consolidate this for changes to the logic.	2021-05-12 17:14:15 -05:00
Christopher Haster	1863dc7883	Merge pull request #519 from littlefs-project/devel Minor release: v2.4	2021-01-19 18:50:34 -06:00
Christopher Haster	3d4e4f2085	Bumped minor version to v2.4	2021-01-18 20:23:54 -06:00
Christopher Haster	a2c744c8f8	Merge pull request #516 from littlefs-project/ci-revamp Adopt GitHub Actions, bring in a number of script/Makefile improvements	2021-01-18 18:38:42 -06:00
Christopher Haster	c0cc0a417e	Enabled overriding of LFS_ASSERT/TRACE/DEBUG/etc This is useful for testing the new erroring assert behavior in CI. Asserts do not error by default, so this macro needs to be overriden. It is possible to test this behavior using the existing option of overriding lfs_util.h with a custom file, by using a small sed one-line script. But this is much simpler. This does raise the question if more of the configuration options in lfs_util.h should be opened up for function-like macro overrides.	2021-01-18 14:01:53 -06:00
Christopher Haster	bca64d76cf	Merge branch 'devel' into ci-revamp Needed to bring in new "error-asserts" configuration	2021-01-18 12:23:25 -06:00
Christopher Haster	cab1d6cca6	Merge pull request #514 from mon/feature/assert_early_return lfs_fs_preporphans: return int to alllow graceful LFS_ASSERT	2021-01-18 11:53:47 -06:00
Will	c9eed1f181	Add test to ensure asserts can return	2021-01-18 11:50:39 -06:00
Will	e7e4b352bd	lfs_fs_preporphans ret int for graceful LFS_ASSERT	2021-01-18 11:50:33 -06:00
Christopher Haster	9449ef4be4	Merge pull request #511 from embeddedt/fix_lseek Skip flushing file if lfs_file_rawseek() doesn't change position	2021-01-18 11:47:56 -06:00
Christopher Haster	cfe779fc08	Merge pull request #508 from littlefs-project/fix-sanity-check Moved sanity check in lfs_format after compaction	2021-01-18 11:47:23 -06:00
Christopher Haster	0db6466984	Merge pull request #502 from mon/feature/meta_limits Add metadata_max config to help performance on devices with large blocks	2021-01-18 11:45:34 -06:00
Christopher Haster	21488d9e06	Fixed incorrect documentation in test.py The argparse documented an outdated format, and was off by 1. Found by sender6	2021-01-18 11:41:51 -06:00
Christopher Haster	10a08833c6	Moved lfs_mdir_isopen behind LFS_NO_ASSERT lfs_mdir_isopen goes unused if asserts are disabled, and this caused an "unused function" warning on Clang (curiously not on GCC since the function was static inline, commonly used for header-only functions). Also removed "inline" from the lfs_mdir_* functions as these involve linked-list operations and really shouldn't be inlined. And since they are static, inlining should occur automatically if there is a benefit. Found by dpgeorge	2021-01-18 11:41:18 -06:00
Christopher Haster	47d6b2fcf3	Removed unnecessary truncate condition thanks to new seek optimization	2021-01-11 00:14:34 -06:00
Christopher Haster	745d98cde0	Fixed lfs_file_truncate issue where internal state may not be flushed This was caused by the new lfs_file_rawseek optimization that can skip flushing when calculated file->pos is unchanged combined with an implicit expectation in lfs_file_truncate that lfs_file_rawseek unconditionally sets file->pos. Because of this assumption, lfs_file_truncate could leave file->pos in an outdated state while changing the internal file metadata. Humorously, this was always gauranteed to trigger the skip in lfs_file_rawseek when we try to restore the file->pos, leaving the file->cache used to do the CTZ skip-list lookup in a potentially bad state. The easiest fix is to just update file->pos correctly. Note we don't want to explicitly flush since we can leverage the same noop optimization if we truncate to the file position. Which I've added a test for.	2021-01-11 00:14:34 -06:00
Themba Dube	3216b07c3b	Use lfs_file_rawsize to calculate LFS_SEEK_END position	2021-01-11 00:14:30 -06:00
Christopher Haster	6592719d28	Removed .travis.yml Now that it's been replaced by GitHub workflows (in .github/workflows)	2021-01-10 13:20:14 -06:00
Christopher Haster	c9110617b3	Added post-release script, cleaned up workflows This helps an outstanding maintainer annoyance: updating dependencies to bring in new versions on each littlefs release. But instead of adding a bunch of scripts to the tail end of the release workflow, the post-release script just triggers a single "repository_dispatch" event in the newly created littlefs.post-release repo. From there any number of post-release workflows can be run. This indirection should let the post-release scripts move much quicker than littlefs itself, which helps offset how fragile these sort of scripts are. --- Also finished cleaning up the workflows now that they are mostly working.	2021-01-10 13:20:11 -06:00
Christopher Haster	104d65113d	Reduced build sources to just the core littlefs Currently this is just lfs.c and lfs_util.c. Previously this included the block devices, but this meant all of the scripts needed to explicitly deselect the block devices to avoid reporting build size/coverage info on them. Note that test.py still explicitly adds the block devices for compiling tests, which is their main purpose. Humorously this means the block devices will probably be compiled into most builds in this repo anyways.	2021-01-10 04:03:16 -06:00
Christopher Haster	6d3e4ac33e	Brought over the release workflow This is pretty much a cleaned up version of the release script that ran on Travis. This biggest change is that now the release script also collecs the build results into a table as part of the change notes, which is a nice addition.	2021-01-10 04:03:13 -06:00
Christopher Haster	9d6546071b	Fixed a recompilation issue in CI, tweaked coverage.py a bit more This was lost in the Travis -> GitHub transition, in serializing some of the jobs, I missed that we need to clean between tests with different geometry configurations. Otherwise we end up running outdated binaries, which explains some of the weird test behavior we were seeing. Also tweaked a few script things: - Better subprocess error reporting (dump stderr on failure) - Fixed a BUILDDIR rule issue in test.py - Changed test-not-run status to None instead of undefined	2021-01-10 03:21:28 -06:00
Christopher Haster	b84fb6bcc5	Added BUILDDIR, a bit of script reworking Now littlefs's Makefile can work with a custom build directory for compilation output. Just set the BUILDDIR variable and the Makefile will take care of the rest. make BUILDDIR=build size This makes it very easy to compare builds with different compile-time configurations or different cross-compilers. This meant most of code.py's build isolation is no longer needed, so revisted the scripts and cleaned/tweaked a number of things. Also bought code.py in line with coverage.py, fixing some of the inconsistencies that were created while developing these scripts. One change to note was removing the inline measuring logic, I realized this feature is unnecessary thanks to GCC's -fkeep-static-functions and -fno-inline flags.	2021-01-10 03:21:21 -06:00
Christopher Haster	887f3660ed	Switched to lcov for coverage collection, greatly simplified coverage.py Since we already have fairly complicated scriptts, I figured it wouldn't be too hard to use the gcov tools and directly parse their output. Boy was I wrong. The gcov intermediary format is a bit of a mess. In version 5.4, a text-based intermediary format is written to a single .gcov file per executable. This changed sometime before version 7.5, when it started writing separate .gcov files per .o files. And in version 9 this intermediary format has been entirely replaced with an incompatible json format! Ironically, this means the internal-only .gcda/.gcno binary format has actually been more stable than the intermediary format. Also there's no way to avoid temporary .gcov files generated in the project root, which risks messing with how test.py runs parallel tests. Fortunately this looks like it will be fixed in gcov version 9. --- Ended up switching to lcov, which was the right way to go. lcov handles all of the gcov parsing, provides an easily parsable output, and even provides a set of higher-level commands to manage coverage collection from different runs. Since this is all provided by lcov, was able to simplify coverage.py quite a bit. Now it just parses the .info files output by lcov.	2021-01-10 02:21:33 -06:00
Christopher Haster	eeeceb9e30	Added coverage.py, and optional coverage info to test.py Now coverage information can be collected if you provide the --coverage to test.py. Internally this uses GCC's gcov instrumentation along with a new script, coverage.py, to parse *.gcov files. The main use for this is finding coverage info during CI runs. There's a risk that the instrumentation may make it more difficult to debug, so I decided to not make coverage collection enabled by default.	2021-01-10 02:12:45 -06:00
Christopher Haster	b2235e956d	Added GitHub workflows to run tests Mostly taken from .travis.yml, biggest changes were around how to get the status updates to work. We can't use a token on PRs the same way we could in Travis, so instead we use a second workflow that checks every pull request for "status" artifacts, and create the actual statuses in the "workflow_run" event, where we have full access to repo secrets.	2021-01-09 23:42:49 -06:00
Themba Dube	6bb4043154	Skip flushing file if lfs_file_rawseek() doesn't change position	2020-12-24 14:05:46 -05:00
Christopher Haster	2b804537b0	Moved sanity check in lfs_format after compaction After a bit of tweaking in 9dde5c7 to write out all superblocks during lfs_format, additional writes were added after the sanity checking normally done at the end. This turned out to be a problem when porting littlefs, as it makes it easy for addressing issues to not get caught during lfs_format. Found by marekr, tristanclare94, and mjs513	2020-12-22 11:47:48 -06:00
Christopher Haster	d804c2d3b7	Added scripts/code_size.py, for more in-depth code-size reporting Inspired by Linux's Bloat-O-Meter, code_size.py wraps nm to provide function-level code size, and supports detailed comparison between different builds. One difference is that code_size.py invokes littlefs's build system similarly to test.py, creating a duplicate build in the "sizes" directory. This makes it easy to monitor a cross-compiled build size while simultaneously testing on the host machine.	2020-12-19 18:49:57 -06:00
Will	37f4de2976	Remove inline_files_max and lfs_t entry for metadata_max	2020-12-18 13:05:20 +10:00
Will	6b16dafb4d	Add metadata_max and inline_file_max to config We have seen poor read performance on NAND flashes with 128kB blocks. The root cause is inline files having to traverse many sets of metadata pairs inside the current block before being fully reconstructed. Simply disabling inline files is not enough, as the metadata will still fill up the block and eventually need to be compacted. By allowing configuration of how much size metadata takes up, along with limiting (or disabling) inline file size, we achieve read performance improvements on an order of magnitude.	2020-12-15 12:59:32 +10:00