mirror of
https://github.com/linuxkit/linuxkit.git
synced 2025-07-21 18:11:35 +00:00
Merge pull request #1625 from jcvenegas/topic/clear-containers-project
projects: Add clear containers project
This commit is contained in:
Justin Cormack
GitHub
commit
a752a3263f
@ -18,8 +18,8 @@ If you want to create a project, please submit a pull request to create a new di
|
||||
- [AWS](aws/) AWS build support
|
||||
- [Swarmd](swarmd) Standalone swarmkit based orchestrator
|
||||
- [Landlock LSM](landlock/) programmatic access control
|
||||
- [Clear Containers](clear-containers/) Clear Containers image
|
||||
|
||||
## Current projects not yet documented
|
||||
- Clear Linux integration (Intel)
|
||||
- VMWare support (VMWare)
|
||||
- ARM port and secure boot integration (ARM)
|
||||
|
||||
104
projects/clear-containers/README.md
Normal file
104
projects/clear-containers/README.md
Normal file
@ -0,0 +1,104 @@
|
||||
# Clear Containers image #
|
||||
|
||||
This project's goal is to use LinuxKit\* to generate two components for the
|
||||
Clear Containers project:
|
||||
|
||||
- Kernel: The Linux\* kernel for Intel® Clear Containers following LinuxKit
|
||||
philosophy.
|
||||
- Image or mini-os: the minimal guest user-space image (The guest OS based in
|
||||
LinuxKit is not functional for clear-containers yet).
|
||||
|
||||
## What are Clear Containers? ##
|
||||
|
||||
Intel® Clear Containers are a collection of tools, configurations, and
|
||||
techniques anchored on an implementation leveraging Intel® Architecture to
|
||||
optimize hypervisor-based containers. Intel® Clear Containers enable a small
|
||||
virtual machine with the isolation of virtual-machine technology along with
|
||||
the deployment benefits of containers.
|
||||
|
||||
Today, the Intel® Clear Containers project uses a Clear Linux\* OS for Intel®
|
||||
Architecture based guest OS image. The Clear Linux based image provides:
|
||||
|
||||
- [systemd]
|
||||
- [hyperstart]
|
||||
|
||||
## Why use LinuxKit to build a Clear Containers guest OS?##
|
||||
|
||||
Using LinuxKit to build the guest OS fits the Clear Containers guest needs. LinuxKit
|
||||
provides:
|
||||
|
||||
- An immutable Linux distribution: Write operations from the guest to the
|
||||
file system are not persistent.
|
||||
- Easy tooling with easy iteration.
|
||||
- A general purpose toolkit.
|
||||
|
||||
## Current development status ##
|
||||
|
||||
A Clear Containers compatible kernel can be generated using LinuxKit
|
||||
which has the security features of LinuxKit and the virtualization
|
||||
features required by Clear Containers.
|
||||
|
||||
## Getting started ##
|
||||
To build the kernel, enter the following commands:
|
||||
|
||||
|
||||
1. Build kernel
|
||||
|
||||
This step is required to build a kernel from source
|
||||
It will use the kernel configuration file:
|
||||
`projects/clear-containers/kernel/kernel_config` and
|
||||
the patches from `projects/clear-containers/kernel/patches-*`
|
||||
|
||||
NOTE: This step also is needed meanwhile an image
|
||||
`linuxkit/kernel-clear-containers:X.y.x `is pushed to docker-hub.
|
||||
|
||||
The following commands will build a kernel using
|
||||
`projects/clear-containers/kernel/Dockerfile` :
|
||||
|
||||
```
|
||||
pushd projects/clear-containers/kernel
|
||||
make image
|
||||
make tag
|
||||
popd
|
||||
```
|
||||
2. Get a Clear Containers kernel using moby tool
|
||||
|
||||
To generate the kernel image called `clear-containers-bzImage`, enter the
|
||||
following command:
|
||||
|
||||
```
|
||||
./bin/moby build ./projects/clear-containers/clear-containers.yml
|
||||
```
|
||||
|
||||
3. Test kernel
|
||||
|
||||
The Makefile target `qemu-lite` from `projects/clear-containers/tools`
|
||||
will launch the Clear Containers base OS and the
|
||||
with kernel built with moby. Use the `root` user to access to the test system,
|
||||
it will ask to setup a password.
|
||||
|
||||
```
|
||||
cd projects/clear-containers/tools
|
||||
make qemu-lite
|
||||
```
|
||||
|
||||
The file `clear-containers-bzImage` is not a bzImage\*. The file
|
||||
is a vmlinux image, see TODO. The [cc-oci-runtime] project can be
|
||||
configured to use kernel generated by LinuxKit.
|
||||
|
||||
## TODO ##
|
||||
|
||||
* Create a Clear Container Mini-OS image.
|
||||
* Start `hyperstart` as a privileged container.
|
||||
* Add a script to launch a clear container guest.
|
||||
* Sync with official kernel LinuxKit patches.
|
||||
* Add an output format similar to `kernel+initrd` which extracts a vmlinux
|
||||
image. The Intel® Clear Containers project requires an uncompressed kernel
|
||||
to boot. Currently, the LinuxKit output format `kernel+initrd` extracts a
|
||||
bzImage from a LinuxKit packaged kernel. A way to extract the vmlinux image is
|
||||
still required.
|
||||
|
||||
[systemd]: <https://www.freedesktop.org/wiki/Software/systemd/>
|
||||
[hyperstart]: <https://github.com/clearcontainers/hyperstart>
|
||||
[cc-oci-runtime]: <https://github.com/01org/cc-oci-runtime>
|
||||
[hyperstart container]: <https://github.com/clearcontainers/hyperstart/tree/0.7.0-clearcontainers>
|
||||
26
projects/clear-containers/clear-containers.yml
Normal file
26
projects/clear-containers/clear-containers.yml
Normal file
@ -0,0 +1,26 @@
|
||||
kernel:
|
||||
image: "linuxkit/kernel-clear-containers:4.9.x"
|
||||
cmdline: "root=/dev/pmem0p1 rootflags=dax,data=ordered,errors=remount-ro rw rootfstype=ext4 tsc=reliable no_timer_check rcupdate.rcu_expedited=1 i8042.direct=1 i8042.dumbkbd=1 i8042.nopnp=1 i8042.noaux=1 noreplace-smp reboot=k panic=1 console=hvc0 console=hvc1 initcall_debug iommu=off quiet cryptomgr.notests page_poison=on"
|
||||
init:
|
||||
- mobylinux/init:8375addb923b8b88b2209740309c92aa5f2a4f9d
|
||||
onboot:
|
||||
- name: sysctl
|
||||
image: "mobylinux/sysctl:2cf2f9d5b4d314ba1bfc22b2fe931924af666d8c"
|
||||
net: host
|
||||
pid: host
|
||||
ipc: host
|
||||
capabilities:
|
||||
- CAP_SYS_ADMIN
|
||||
readonly: true
|
||||
services:
|
||||
- name: rngd
|
||||
image: "mobylinux/rngd:3dad6dd43270fa632ac031e99d1947f20b22eec9"
|
||||
capabilities:
|
||||
- CAP_SYS_ADMIN
|
||||
oomScoreAdj: -800
|
||||
readonly: true
|
||||
files:
|
||||
- path: etc/docker/daemon.json
|
||||
contents: '{"debug": true}'
|
||||
outputs:
|
||||
- format: kernel+initrd
|
||||
64
projects/clear-containers/kernel/Dockerfile
Normal file
64
projects/clear-containers/kernel/Dockerfile
Normal file
@ -0,0 +1,64 @@
|
||||
FROM linuxkit/alpine-build-kernel:cfdd576c36a52ed2dd62f237f79eeedc2dd3697b@sha256:3fe08db373a9373ba1616a485858f01ebd2d7a3cb364a099d0ed8b45fa419da2
|
||||
|
||||
|
||||
ARG KERNEL_VERSION
|
||||
ARG DEBUG=0
|
||||
|
||||
ENV KERNEL_SOURCE=https://www.kernel.org/pub/linux/kernel/v4.x/linux-${KERNEL_VERSION}.tar.xz
|
||||
|
||||
# Download kernel source code
|
||||
RUN curl -fsSL -o linux-${KERNEL_VERSION}.tar.xz ${KERNEL_SOURCE}
|
||||
RUN tar xf linux-${KERNEL_VERSION}.tar.xz && mv /linux-${KERNEL_VERSION} /linux
|
||||
WORKDIR /linux
|
||||
|
||||
ENV DEF_CONFIG_FILE=/linux/arch/x86/configs/x86_64_defconfig
|
||||
COPY kernel_config ${DEF_CONFIG_FILE}
|
||||
COPY kernel_config.debug /linux/debug_config
|
||||
|
||||
|
||||
# Enable debug
|
||||
RUN if [ $DEBUG -ne "0" ]; then \
|
||||
sed -i 's/CONFIG_PANIC_ON_OOPS=y/# CONFIG_PANIC_ON_OOPS is not set/' \
|
||||
${DEF_CONFIG_FILE}; \
|
||||
cat /linux/debug_config >> ${DEF_CONFIG_FILE}; \
|
||||
fi
|
||||
|
||||
|
||||
RUN cat ${DEF_CONFIG_FILE}
|
||||
|
||||
# Apply local patches
|
||||
COPY patches-4.9 /patches
|
||||
RUN cd /linux && \
|
||||
set -e && for patch in /patches/*.patch; do \
|
||||
echo "Applying $patch"; \
|
||||
patch -p1 < "$patch"; \
|
||||
done
|
||||
|
||||
# Build kernel
|
||||
RUN make defconfig && \
|
||||
make oldconfig && \
|
||||
perl -p -i -e "s/^EXTRAVERSION.*/EXTRAVERSION = -linuxkit/" Makefile && \
|
||||
make -j "$(getconf _NPROCESSORS_ONLN)" KCFLAGS="-fno-pie"
|
||||
|
||||
#bzImage
|
||||
#vmlinux
|
||||
RUN cp vmlinux arch/x86_64/boot/bzImage /
|
||||
|
||||
# CC does not provide modules, not needed to distribute headers.
|
||||
#kernel-headers.tar: provides kernel headers
|
||||
RUN mkdir -p /tmp/kernel-headers/usr && \
|
||||
cd /tmp/kernel-headers && tar cf /kernel-headers.tar usr
|
||||
|
||||
# CC does no use modules do not ship it
|
||||
#kernel-modules.tar: provides kernel modules
|
||||
RUN mkdir -p /tmp/kernel-modules/lib/modules && \
|
||||
cd /tmp/kernel-modules && tar cf /kernel-modules.tar lib
|
||||
|
||||
|
||||
WORKDIR /
|
||||
|
||||
#kernel-dev.tar: provides headers .config linux/include arch/x86/include
|
||||
RUN mkdir -p /tmp/usr/src/linux-headers && \
|
||||
cd /tmp/ && tar cf /kernel-dev.tar usr/src
|
||||
|
||||
RUN printf "KERNEL_SOURCE=${KERNEL_SOURCE}\n" > /kernel-source-info
|
||||
87
projects/clear-containers/kernel/Makefile
Normal file
87
projects/clear-containers/kernel/Makefile
Normal file
@ -0,0 +1,87 @@
|
||||
DEBUG ?= 0
|
||||
|
||||
all: vmlinux push
|
||||
|
||||
# We push the image to hub twice, once with the full kernel version of
|
||||
# "linuxkit/kernel:<kernel version>.<major version>.<minor version>-<n>",
|
||||
# where "<n>" is a monotonically increasing config number, and as
|
||||
# "linuxkit/kernel:<kernel version>.<major version>.x". This version
|
||||
# number is stored in IMAGE_VERSION.
|
||||
#
|
||||
# We expect most users to us the "<kernel version>.<major version>.x"
|
||||
# variant as this simply is the latest version of a given major kernel
|
||||
# version. This version number is stored in IMAGE_MAJOR_VERSION.
|
||||
#
|
||||
# For IMAGE_VERSION, the "<n>" must be increased whenever
|
||||
# the kernel config or the patches change. We don't expect this to
|
||||
# happen very often as the minor version number gets update quite
|
||||
# frequently.
|
||||
#
|
||||
# IMAGE_VERSION is used to determine if a new image should be pushed to hub.
|
||||
KERNEL_VERSION=4.9.22
|
||||
IMAGE_VERSION=$(KERNEL_VERSION)-0
|
||||
IMAGE_MAJOR_VERSION=4.9.x
|
||||
DEPS=Dockerfile Makefile kernel_config kernel_config.debug patches-4.9
|
||||
PKG_DEPS=bzImage kernel-dev.tar kernel-headers.tar vmlinux kernel-modules.tar
|
||||
|
||||
ifdef http_proxy
|
||||
BUILD_PROXY = --build-arg http_proxy=$(http_proxy)
|
||||
RUN_PROXY = --env http_proxy=$(http_proxy)
|
||||
endif
|
||||
|
||||
ifdef https_proxy
|
||||
BUILD_PROXY += --build-arg https_proxy=$(https_proxy)
|
||||
RUN_PROXY += --env https_proxy=$(https_proxy)
|
||||
endif
|
||||
|
||||
#build a kernel using dockerfile and save image hash in kernel.tag
|
||||
kernel.tag: $(DEPS)
|
||||
BUILD=$$( tar cf - $^ | docker build -f $< $(BUILD_PROXY) --build-arg DEBUG=$(DEBUG) --build-arg KERNEL_VERSION=$(KERNEL_VERSION) -q - ) && [ -n "$$BUILD" ] && echo "Built $$BUILD" && echo "$$BUILD" > $@
|
||||
|
||||
#Extract $(PKG) from image with built kernel
|
||||
vmlinux: kernel.tag
|
||||
mkdir -p x86_64
|
||||
docker run --rm --net=none --log-driver=none \
|
||||
$(shell cat kernel.tag) \
|
||||
tar cf - $(PKG_DEPS) | tar xf - -C x86_64
|
||||
cp x86_64/kernel-modules.tar kernel.tar
|
||||
# FIXME: Remove when linuxkit allow get vmlinux from docker images
|
||||
# Rename vmlinux to bzImage
|
||||
cp x86_64/vmlinux bzImage
|
||||
cp x86_64/vmlinux $@
|
||||
|
||||
.PHONY: image push tag
|
||||
|
||||
MEDIA_TOYBOX=linuxkit/toybox-media:d7e82a7d19ccc84c9071fa7a88ecaa58ae958f7c@sha256:4c7d25f2be2429cd08417c36e04161cb924e46f3e419ee33a0aa9ff3a0942e02
|
||||
|
||||
BASE="$MEDIA_TOYBOX"
|
||||
IMAGE=kernel-clear-containers
|
||||
|
||||
default: push
|
||||
|
||||
Dockerfile.media:
|
||||
printf "FROM $(MEDIA_TOYBOX)\nADD . /\n" > $@
|
||||
|
||||
image: Dockerfile.media vmlinux bzImage kernel.tar $(DEPS)
|
||||
tar cf - $^ | docker build --no-cache -t $(IMAGE):build -f Dockerfile.media -
|
||||
|
||||
push: image
|
||||
docker pull linuxkit/$(IMAGE):$(IMAGE_VERSION) || \
|
||||
(docker tag $(IMAGE):build linuxkit/$(IMAGE):$(IMAGE_VERSION) && \
|
||||
docker push linuxkit/$(IMAGE):$(IMAGE_VERSION) && \
|
||||
docker tag $(IMAGE):build linuxkit/$(IMAGE):$(IMAGE_MAJOR_VERSION) && \
|
||||
docker push linuxkit/$(IMAGE):$(IMAGE_MAJOR_VERSION))
|
||||
docker rmi $(IMAGE):build
|
||||
rm -f hash
|
||||
|
||||
tag: image
|
||||
(docker tag $(IMAGE):build linuxkit/$(IMAGE):$(IMAGE_VERSION) && \
|
||||
docker tag $(IMAGE):build linuxkit/$(IMAGE):$(IMAGE_MAJOR_VERSION))
|
||||
docker rmi $(IMAGE):build
|
||||
rm -f hash
|
||||
|
||||
.PHONY: clean
|
||||
clean:
|
||||
rm -rf x86_64 lib usr sbin kernel.tag Dockerfile.media vmlinux bzImage kernel.tar etc
|
||||
|
||||
.DELETE_ON_ERROR:
|
||||
2369
projects/clear-containers/kernel/kernel_config
Normal file
2369
projects/clear-containers/kernel/kernel_config
Normal file
File diff suppressed because it is too large
Load Diff
2337
projects/clear-containers/kernel/kernel_config.debug
Normal file
2337
projects/clear-containers/kernel/kernel_config.debug
Normal file
File diff suppressed because it is too large
Load Diff
@ -0,0 +1,34 @@
|
||||
From ae8bbac0d97941a8d54451c98479dc14dddf9678 Mon Sep 17 00:00:00 2001
|
||||
From: Arjan van de Ven <arjan@linux.intel.com>
|
||||
Date: Wed, 11 Feb 2015 16:19:26 -0600
|
||||
Subject: [PATCH 1/4] cpuidle: skip synchronize_rcu() on single CPU systems
|
||||
|
||||
synchronize_rcu() is pretty expensive, and on single CPU systems we don't need
|
||||
it in this specific case, so skip it.
|
||||
|
||||
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
|
||||
Signed-off-by: Miguel Bernal Marin <miguel.bernal.marin@linux.intel.com>
|
||||
---
|
||||
drivers/cpuidle/cpuidle.c | 5 ++++-
|
||||
1 file changed, 4 insertions(+), 1 deletion(-)
|
||||
|
||||
diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c
|
||||
index c73207ab..224cefc0 100644
|
||||
--- a/drivers/cpuidle/cpuidle.c
|
||||
+++ b/drivers/cpuidle/cpuidle.c
|
||||
@@ -307,8 +307,11 @@ void cpuidle_uninstall_idle_handler(void)
|
||||
/*
|
||||
* Make sure external observers (such as the scheduler)
|
||||
* are done looking at pointed idle states.
|
||||
+ * This is only relevant if there is more than one cpu,
|
||||
+ * if there is only one CPU, that is us... and we're
|
||||
+ * coherent to ourselves.
|
||||
*/
|
||||
- synchronize_rcu();
|
||||
+
|
||||
}
|
||||
|
||||
/**
|
||||
--
|
||||
2.12.2
|
||||
|
||||
@ -0,0 +1,38 @@
|
||||
From e1b22ee1b1a8160cb03112d89defd9bf8ae8baf1 Mon Sep 17 00:00:00 2001
|
||||
From: Arjan van de Ven <arjan@linux.intel.com>
|
||||
Date: Wed, 11 Feb 2015 16:25:16 -0600
|
||||
Subject: [PATCH 2/4] sysrq: skip synchronize_rcu() if there is no old op
|
||||
|
||||
synchronize_rcu() is expensive. Currently it is called as part of the sysrq
|
||||
registration/unregistration, which happens during boot several times.
|
||||
Now, the reason for the synchronize_rcu() is to allow an old registered
|
||||
operation to expire properly... which is pointless if the old operation
|
||||
is NULL...
|
||||
So we can save the common case of the old operation being NULL a lot of time
|
||||
by just checking for non-NULL prior to the synchronize_rcu()
|
||||
|
||||
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
|
||||
Signed-off-by: Miguel Bernal Marin <miguel.bernal.marin@linux.intel.com>
|
||||
---
|
||||
drivers/tty/sysrq.c | 4 +++-
|
||||
1 file changed, 3 insertions(+), 1 deletion(-)
|
||||
|
||||
diff --git a/drivers/tty/sysrq.c b/drivers/tty/sysrq.c
|
||||
index 701c085b..c60c7ba5 100644
|
||||
--- a/drivers/tty/sysrq.c
|
||||
+++ b/drivers/tty/sysrq.c
|
||||
@@ -1065,8 +1065,10 @@ static int __sysrq_swap_key_ops(int key, struct sysrq_key_op *insert_op_p,
|
||||
* A concurrent __handle_sysrq either got the old op or the new op.
|
||||
* Wait for it to go away before returning, so the code for an old
|
||||
* op is not freed (eg. on module unload) while it is in use.
|
||||
+ * This is only relevant if the old op is not NULL of course.
|
||||
*/
|
||||
- synchronize_rcu();
|
||||
+ if (remove_op_p)
|
||||
+ synchronize_rcu();
|
||||
|
||||
return retval;
|
||||
}
|
||||
--
|
||||
2.12.2
|
||||
|
||||
@ -0,0 +1,39 @@
|
||||
From 58736c7e70371aa602e284812d493108b25070c3 Mon Sep 17 00:00:00 2001
|
||||
From: Arjan van de Ven <arjan@linux.intel.com>
|
||||
Date: Mon, 22 Jun 2015 09:33:33 -0500
|
||||
Subject: [PATCH 3/4] init: no wait for the known devices
|
||||
|
||||
No wait for the known devices to complete their probing
|
||||
|
||||
Author: Arjan van de Ven <arjan@linux.intel.com>
|
||||
|
||||
Signed-off-by: Miguel Bernal Marin <miguel.bernal.marin@linux.intel.com>
|
||||
---
|
||||
init/do_mounts.c | 4 +++-
|
||||
1 file changed, 3 insertions(+), 1 deletion(-)
|
||||
|
||||
diff --git a/init/do_mounts.c b/init/do_mounts.c
|
||||
index dea5de95..da840946 100644
|
||||
--- a/init/do_mounts.c
|
||||
+++ b/init/do_mounts.c
|
||||
@@ -28,6 +28,7 @@
|
||||
#include <linux/slab.h>
|
||||
#include <linux/ramfs.h>
|
||||
#include <linux/shmem_fs.h>
|
||||
+#include <linux/async.h>
|
||||
|
||||
#include <linux/nfs_fs.h>
|
||||
#include <linux/nfs_fs_sb.h>
|
||||
@@ -563,7 +564,8 @@ void __init prepare_namespace(void)
|
||||
* For example, it is not atypical to wait 5 seconds here
|
||||
* for the touchpad of a laptop to initialize.
|
||||
*/
|
||||
- wait_for_device_probe();
|
||||
+ //wait_for_device_probe();
|
||||
+ async_synchronize_full();
|
||||
|
||||
md_run_setup();
|
||||
|
||||
--
|
||||
2.12.2
|
||||
|
||||
@ -0,0 +1,131 @@
|
||||
From 627247f7d57ce964617f6580d9c5e1184850ac5d Mon Sep 17 00:00:00 2001
|
||||
From: Eric Van Hensbergen <ericvh@gmail.com>
|
||||
Date: Tue, 21 Apr 2015 12:46:29 -0700
|
||||
Subject: [PATCH 4/4] fs/9p: fix create-unlink-getattr idiom
|
||||
|
||||
Fixes several outstanding bug reports of not being able to getattr from an
|
||||
open file after an unlink. This patch cleans up transient fids on an unlink
|
||||
and will search open fids on a client if it detects a dentry that appears to
|
||||
have been unlinked. This search is necessary because fstat does not pass fd
|
||||
information through the VFS API to the filesystem, only the dentry which for
|
||||
9p has an imperfect match to fids.
|
||||
|
||||
Inherent in this patch is also a fix for the qid handling on create/open
|
||||
which apparently wasn't being set correctly and was necessary for the search
|
||||
to succeed.
|
||||
|
||||
A possible optimization over this fix is to include accounting of open fids
|
||||
with the inode in the private data (in a similar fashion to the way we track
|
||||
transient fids with dentries). This would allow a much quicker search for
|
||||
a matching open fid.
|
||||
|
||||
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
|
||||
---
|
||||
fs/9p/fid.c | 30 ++++++++++++++++++++++++++++++
|
||||
fs/9p/vfs_inode.c | 4 ++++
|
||||
net/9p/client.c | 5 ++++-
|
||||
3 files changed, 38 insertions(+), 1 deletion(-)
|
||||
|
||||
diff --git a/fs/9p/fid.c b/fs/9p/fid.c
|
||||
index 60fb4746..e19c9cf7 100644
|
||||
--- a/fs/9p/fid.c
|
||||
+++ b/fs/9p/fid.c
|
||||
@@ -54,6 +54,33 @@ void v9fs_fid_add(struct dentry *dentry, struct p9_fid *fid)
|
||||
}
|
||||
|
||||
/**
|
||||
+ * v9fs_fid_find_global - search for a fid off of the client list
|
||||
+ * @inode: return a fid pointing to a specific inode
|
||||
+ * @uid: return a fid belonging to the specified user
|
||||
+ *
|
||||
+ */
|
||||
+
|
||||
+static struct p9_fid *v9fs_fid_find_inode(struct inode *inode, kuid_t uid)
|
||||
+{
|
||||
+ struct p9_client *clnt = v9fs_inode2v9ses(inode)->clnt;
|
||||
+ struct p9_fid *fid, *fidptr, *ret = NULL;
|
||||
+ unsigned long flags;
|
||||
+
|
||||
+ p9_debug(P9_DEBUG_VFS, " inode: %p\n", inode);
|
||||
+
|
||||
+ spin_lock_irqsave(&clnt->lock, flags);
|
||||
+ list_for_each_entry_safe(fid, fidptr, &clnt->fidlist, flist) {
|
||||
+ if (uid_eq(fid->uid, uid) &&
|
||||
+ (inode->i_ino == v9fs_qid2ino(&fid->qid))) {
|
||||
+ ret = fid;
|
||||
+ break;
|
||||
+ }
|
||||
+ }
|
||||
+ spin_unlock_irqrestore(&clnt->lock, flags);
|
||||
+ return ret;
|
||||
+}
|
||||
+
|
||||
+/**
|
||||
* v9fs_fid_find - retrieve a fid that belongs to the specified uid
|
||||
* @dentry: dentry to look for fid in
|
||||
* @uid: return fid that belongs to the specified user
|
||||
@@ -80,6 +107,9 @@ static struct p9_fid *v9fs_fid_find(struct dentry *dentry, kuid_t uid, int any)
|
||||
}
|
||||
}
|
||||
spin_unlock(&dentry->d_lock);
|
||||
+ } else {
|
||||
+ if (dentry->d_inode)
|
||||
+ ret = v9fs_fid_find_inode(dentry->d_inode, uid);
|
||||
}
|
||||
|
||||
return ret;
|
||||
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
|
||||
index 30ca770c..c00487ea 100644
|
||||
--- a/fs/9p/vfs_inode.c
|
||||
+++ b/fs/9p/vfs_inode.c
|
||||
@@ -624,6 +624,10 @@ static int v9fs_remove(struct inode *dir, struct dentry *dentry, int flags)
|
||||
|
||||
v9fs_invalidate_inode_attr(inode);
|
||||
v9fs_invalidate_inode_attr(dir);
|
||||
+
|
||||
+ /* invalidate all fids associated with dentry */
|
||||
+ /* NOTE: This will not include open fids */
|
||||
+ dentry->d_op->d_release(dentry);
|
||||
}
|
||||
return retval;
|
||||
}
|
||||
diff --git a/net/9p/client.c b/net/9p/client.c
|
||||
index 3fc94a49..98980bac 100644
|
||||
--- a/net/9p/client.c
|
||||
+++ b/net/9p/client.c
|
||||
@@ -1208,7 +1208,7 @@ struct p9_fid *p9_client_walk(struct p9_fid *oldfid, uint16_t nwname,
|
||||
if (nwname)
|
||||
memmove(&fid->qid, &wqids[nwqids - 1], sizeof(struct p9_qid));
|
||||
else
|
||||
- fid->qid = oldfid->qid;
|
||||
+ memmove(&fid->qid, &oldfid->qid, sizeof(struct p9_qid));
|
||||
|
||||
kfree(wqids);
|
||||
return fid;
|
||||
@@ -1261,6 +1261,7 @@ int p9_client_open(struct p9_fid *fid, int mode)
|
||||
p9_is_proto_dotl(clnt) ? "RLOPEN" : "ROPEN", qid.type,
|
||||
(unsigned long long)qid.path, qid.version, iounit);
|
||||
|
||||
+ memmove(&fid->qid, &qid, sizeof(struct p9_qid));
|
||||
fid->mode = mode;
|
||||
fid->iounit = iounit;
|
||||
|
||||
@@ -1306,6 +1307,7 @@ int p9_client_create_dotl(struct p9_fid *ofid, char *name, u32 flags, u32 mode,
|
||||
(unsigned long long)qid->path,
|
||||
qid->version, iounit);
|
||||
|
||||
+ memmove(&ofid->qid, qid, sizeof(struct p9_qid));
|
||||
ofid->mode = mode;
|
||||
ofid->iounit = iounit;
|
||||
|
||||
@@ -1351,6 +1353,7 @@ int p9_client_fcreate(struct p9_fid *fid, char *name, u32 perm, int mode,
|
||||
(unsigned long long)qid.path,
|
||||
qid.version, iounit);
|
||||
|
||||
+ memmove(&fid->qid, &qid, sizeof(struct p9_qid));
|
||||
fid->mode = mode;
|
||||
fid->iounit = iounit;
|
||||
|
||||
--
|
||||
2.12.2
|
||||
|
||||
10
projects/clear-containers/tools/Dockerfile
Normal file
10
projects/clear-containers/tools/Dockerfile
Normal file
@ -0,0 +1,10 @@
|
||||
FROM fedora:25
|
||||
|
||||
RUN dnf install -y 'dnf-command(config-manager)'
|
||||
RUN dnf config-manager --add-repo \
|
||||
http://download.opensuse.org/repositories/home:clearlinux:preview:clear-containers-2.1/Fedora\_25/home:clearlinux:preview:clear-containers-2.1.repo
|
||||
|
||||
RUN dnf install -y qemu-lite clear-containers-image linux-container
|
||||
COPY qemu.sh /bin/qemu.sh
|
||||
WORKDIR /root
|
||||
ENTRYPOINT ["qemu.sh"]
|
||||
35
projects/clear-containers/tools/Makefile
Normal file
35
projects/clear-containers/tools/Makefile
Normal file
@ -0,0 +1,35 @@
|
||||
MKFILE_PATH := $(abspath $(lastword $(MAKEFILE_LIST)))
|
||||
MKFILE_DIR := $(dir $(MKFILE_PATH))
|
||||
ifdef http_proxy
|
||||
BUILD_PROXY = --build-arg http_proxy=$(http_proxy)
|
||||
RUN_PROXY = --env http_proxy=$(http_proxy)
|
||||
endif
|
||||
|
||||
ifdef https_proxy
|
||||
BUILD_PROXY += --build-arg https_proxy=$(https_proxy)
|
||||
RUN_PROXY += --env https_proxy=$(https_proxy)
|
||||
endif
|
||||
|
||||
CMDLINE_FILE=clear-containers-cmdline
|
||||
KERNEL_FILE=clear-containers-bzImage
|
||||
|
||||
IMAGE=qemu-lite
|
||||
|
||||
docker_image: Dockerfile
|
||||
docker build $(BUILD_PROXY) -t clearcontainers/$(IMAGE) .
|
||||
|
||||
qemu-lite: docker_image $(KERNEL_FILE) $(CMDLINE_FILE)
|
||||
docker run --privileged --rm -ti \
|
||||
-v $$(pwd)/$(CMDLINE_FILE):/root/clear-containers-cmdline \
|
||||
-v $$(pwd)/$(KERNEL_FILE):/root/clear-containers-vmlinux \
|
||||
clearcontainers/$(IMAGE) \
|
||||
bash
|
||||
|
||||
$(KERNEL_FILE):
|
||||
$(MKFILE_DIR)/../../../bin/moby build $(MKFILE_DIR)/../clear-containers.yml
|
||||
|
||||
push: docker_image
|
||||
docker push clearcontainers/$(IMAGE)
|
||||
|
||||
clean:
|
||||
rm -rf *-bzImage *-cmdline *.img
|
||||
49
projects/clear-containers/tools/qemu.sh
Executable file
49
projects/clear-containers/tools/qemu.sh
Executable file
@ -0,0 +1,49 @@
|
||||
#!/bin/bash
|
||||
|
||||
set -e
|
||||
|
||||
function die {
|
||||
echo >&2 "$@"
|
||||
exit 1
|
||||
}
|
||||
|
||||
img=/usr/share/clear-containers/clear-containers.img
|
||||
img=$(readlink -f "$img")
|
||||
img_size=$(du -b "${img}" | awk '{print $1}')
|
||||
|
||||
kernel="$(pwd)/clear-containers-vmlinux"
|
||||
kernel_cmdline_file="$(pwd)/clear-containers-cmdline"
|
||||
[ -f "${img}" ] || die "Image s required"
|
||||
[ -f "${kernel}" ] || die "Kernel is required"
|
||||
[ -f ${kernel_cmdline_file} ] || \
|
||||
die "Kernel cmdline file is required"
|
||||
|
||||
kernel_cmdline=$(cat "$kernel_cmdline_file")
|
||||
|
||||
cmd="/usr/bin/qemu-lite-system-x86_64"
|
||||
cmd="$cmd -machine pc-lite,accel=kvm,kernel_irqchip,nvdimm"
|
||||
cmd="$cmd -device nvdimm,memdev=mem0,id=nv0"
|
||||
#image
|
||||
cmd="$cmd -object memory-backend-file,id=mem0,mem-path=${img},size=${img_size}"
|
||||
#memory
|
||||
cmd="$cmd -m 2G,slots=2,maxmem=3G"
|
||||
#kernel
|
||||
cmd="$cmd -kernel ${kernel}"
|
||||
cmd="$cmd -append '${kernel_cmdline}'"
|
||||
#cpu
|
||||
cmd="$cmd -smp 2,sockets=1,cores=2,threads=1"
|
||||
cmd="$cmd -cpu host"
|
||||
#clock
|
||||
cmd="$cmd -rtc base=utc,driftfix=slew"
|
||||
cmd="$cmd -no-user-config"
|
||||
cmd="$cmd -nodefaults"
|
||||
cmd="$cmd -global"
|
||||
cmd="$cmd kvm-pit.lost_tick_policy=discard"
|
||||
#console
|
||||
cmd="$cmd -device virtio-serial-pci,id=virtio-serial0"
|
||||
cmd="$cmd -chardev stdio,id=charconsole0,signal=off"
|
||||
cmd="$cmd -device virtconsole,chardev=charconsole0,id=console0"
|
||||
cmd="$cmd -nographic"
|
||||
cmd="$cmd -vga none"
|
||||
|
||||
eval "$cmd"
|
||||
Loading…
Reference in New Issue
Block a user