linux-api.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Eugene Syromiatnikov <esyr-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
To: Cyrill Gorcunov <gorcunov-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Cc: linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org,
	akpm-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org,
	avagin-5HdwGun5lf+gSpxsJD1C4w@public.gmane.org,
	xemul-5HdwGun5lf+gSpxsJD1C4w@public.gmane.org,
	mtk.manpages-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org,
	gorcunov-GEFAQzZX7r8dnm+yROfE0A@public.gmane.org,
	avagin-GEFAQzZX7r8dnm+yROfE0A@public.gmane.org,
	jbaron-JqFfY2XvxFXQT0dZR+AlfA@public.gmane.org,
	luto-kltTT9wpgjJwATOyAt5JVQ@public.gmane.org
Subject: Re: [v4,resend,2/2] kcmp: Add KCMP_EPOLL_TFD mode to compare epoll target files
Date: Sun, 17 Sep 2017 18:01:41 +0200	[thread overview]
Message-ID: <20170917160141.GA3493@asgard.redhat.com> (raw)
In-Reply-To: <20170424154423.511592110-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>

On Mon, Apr 24, 2017 at 06:39:28PM +0300, Cyrill Gorcunov wrote:
> With current epoll architecture target files are addressed
> with file_struct and file descriptor number, where the last
> is not unique. Moreover files can be transferred from another
> process via unix socket, added into queue and closed then
> so we won't find this descriptor in the task fdinfo list.
> 
> Thus to checkpoint and restore such processes CRIU needs to
> find out where exactly the target file is present to add it into
> epoll queue. For this sake one can use kcmp call where
> some particular target file from the queue is compared with
> arbitrary file passed as an argument.
> 
> Because epoll target files can have same file descriptor
> number but different file_struct a caller should explicitly
> specify the offset within.
> 
> To test if some particular file is matching entry inside
> epoll one have to
> 
>  - fill kcmp_epoll_slot structure with epoll file descriptor,
>    target file number and target file offset (in case if only
>    one target is present then it should be 0)
> 
>  - call kcmp as kcmp(pid1, pid2, KCMP_EPOLL_TFD, fd, &kcmp_epoll_slot)
>     - the kernel fetch file pointer matching file descriptor @fd of pid1
>     - lookups for file struct in epoll queue of pid2 and returns traditional
>       0,1,2 result for sorting purpose
> 
> v2:
>  - Use KCMP_FILES salt for files comparision (for convenience sake,
>    since the pointers are file structs so user can lookup over previously
>    collected files tree)
>  - Make kcmp_epoll_target as a separate helper instead of opencoding
>    it with #ifdef
> 
> v3:
>  - Use less if()s in kcmp_epoll_target for readability sake (by avagin@)
>  - Use u32 for kcmp_epoll_slot::toff instead of u64, which makes the less
>    memory pressue
> 
> Signed-off-by: Cyrill Gorcunov <gorcunov-GEFAQzZX7r8dnm+yROfE0A@public.gmane.org>
> Acked-by: Andrey Vagin <avagin-GEFAQzZX7r8dnm+yROfE0A@public.gmane.org>
> CC: Al Viro <viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org>
> CC: Andrew Morton <akpm-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org>
> CC: Pavel Emelyanov <xemul-5HdwGun5lf+gSpxsJD1C4w@public.gmane.org>
> CC: Michael Kerrisk <mtk.manpages-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
> CC: Jason Baron <jbaron-JqFfY2XvxFXQT0dZR+AlfA@public.gmane.org>
> CC: Andy Lutomirski <luto-kltTT9wpgjJwATOyAt5JVQ@public.gmane.org>
> ---
>  fs/eventpoll.c            |   42 +++++++++++++++++++++++++++++++++
>  include/linux/eventpoll.h |    3 ++
>  include/uapi/linux/kcmp.h |   10 ++++++++
>  kernel/kcmp.c             |   57 ++++++++++++++++++++++++++++++++++++++++++++++
>  4 files changed, 112 insertions(+)
> 
> Index: linux-ml.git/fs/eventpoll.c
> ===================================================================
> --- linux-ml.git.orig/fs/eventpoll.c
> +++ linux-ml.git/fs/eventpoll.c
> @@ -1000,6 +1000,48 @@ static struct epitem *ep_find(struct eve
>  	return epir;
>  }
>  
> +static struct epitem *ep_find_tfd(struct eventpoll *ep, int tfd, unsigned long toff)
> +{
> +	struct rb_node *rbp;
> +	struct epitem *epi;
> +
> +	for (rbp = rb_first(&ep->rbr); rbp; rbp = rb_next(rbp)) {
> +		epi = rb_entry(rbp, struct epitem, rbn);
> +		if (epi->ffd.fd == tfd) {
> +			if (toff == 0)
> +				return epi;
> +			else
> +				toff--;
> +		}
> +		cond_resched();
> +	}
> +
> +	return NULL;
> +}
> +
> +struct file *get_epoll_tfile_raw_ptr(struct file *file, int tfd,
> +				     unsigned long toff)
> +{
> +	struct file *file_raw;
> +	struct eventpoll *ep;
> +	struct epitem *epi;
> +
> +	if (!is_file_epoll(file))
> +		return ERR_PTR(-EINVAL);
> +
> +	ep = file->private_data;
> +
> +	mutex_lock(&ep->mtx);
> +	epi = ep_find_tfd(ep, tfd, toff);
> +	if (epi)
> +		file_raw = epi->ffd.file;
> +	else
> +		file_raw = ERR_PTR(-ENOENT);
> +	mutex_unlock(&ep->mtx);
> +
> +	return file_raw;
> +}
> +
>  /*
>   * This is the callback that is passed to the wait queue wakeup
>   * mechanism. It is called by the stored file descriptors when they
> Index: linux-ml.git/include/linux/eventpoll.h
> ===================================================================
> --- linux-ml.git.orig/include/linux/eventpoll.h
> +++ linux-ml.git/include/linux/eventpoll.h
> @@ -14,6 +14,7 @@
>  #define _LINUX_EVENTPOLL_H
>  
>  #include <uapi/linux/eventpoll.h>
> +#include <uapi/linux/kcmp.h>
>  
>  
>  /* Forward declarations to avoid compiler errors */
> @@ -22,6 +23,8 @@ struct file;
>  
>  #ifdef CONFIG_EPOLL
>  
> +struct file *get_epoll_tfile_raw_ptr(struct file *file, int tfd, unsigned long toff);
> +
>  /* Used to initialize the epoll bits inside the "struct file" */
>  static inline void eventpoll_init_file(struct file *file)
>  {
> Index: linux-ml.git/include/uapi/linux/kcmp.h
> ===================================================================
> --- linux-ml.git.orig/include/uapi/linux/kcmp.h
> +++ linux-ml.git/include/uapi/linux/kcmp.h
> @@ -1,6 +1,8 @@
>  #ifndef _UAPI_LINUX_KCMP_H
>  #define _UAPI_LINUX_KCMP_H
>  
> +#include <linux/types.h>
> +
>  /* Comparison type */
>  enum kcmp_type {
>  	KCMP_FILE,
> @@ -10,8 +12,16 @@ enum kcmp_type {
>  	KCMP_SIGHAND,
>  	KCMP_IO,
>  	KCMP_SYSVSEM,
> +	KCMP_EPOLL_TFD,
>  
>  	KCMP_TYPES,
>  };
>  
> +/* Slot for KCMP_EPOLL_TFD */
> +struct kcmp_epoll_slot {
> +	__u32 efd;		/* epoll file descriptor */
> +	__u32 tfd;		/* target file number */
> +	__u32 toff;		/* target offset within same numbered sequence */
> +};
> +
>  #endif /* _UAPI_LINUX_KCMP_H */
> Index: linux-ml.git/kernel/kcmp.c
> ===================================================================
> --- linux-ml.git.orig/kernel/kcmp.c
> +++ linux-ml.git/kernel/kcmp.c
> @@ -11,6 +11,10 @@
>  #include <linux/bug.h>
>  #include <linux/err.h>
>  #include <linux/kcmp.h>
> +#include <linux/capability.h>
> +#include <linux/list.h>
> +#include <linux/eventpoll.h>
> +#include <linux/file.h>
>  
>  #include <asm/unistd.h>
>  
> @@ -94,6 +98,56 @@ static int kcmp_lock(struct mutex *m1, s
>  	return err;
>  }
>  
> +#ifdef CONFIG_EPOLL
> +static int kcmp_epoll_target(struct task_struct *task1,
> +			     struct task_struct *task2,
> +			     unsigned long idx1,
> +			     struct kcmp_epoll_slot __user *uslot)
> +{
> +	struct file *filp, *filp_epoll, *filp_tgt;
> +	struct kcmp_epoll_slot slot;
> +	struct files_struct *files;
> +
> +	if (copy_from_user(&slot, uslot, sizeof(slot)))
> +		return -EFAULT;
> +
> +	filp = get_file_raw_ptr(task1, idx1);
> +	if (!filp)
> +		return -EBADF;
> +
> +	files = get_files_struct(task2);
> +	if (!files)
> +		return -EBADF;
> +
> +	spin_lock(&files->file_lock);
> +	filp_epoll = fcheck_files(files, slot.efd);
> +	if (filp_epoll)
> +		get_file(filp_epoll);
> +	else
> +		filp_tgt = ERR_PTR(-EBADF);
> +	spin_unlock(&files->file_lock);
> +	put_files_struct(files);
> +
> +	if (filp_epoll) {
> +		filp_tgt = get_epoll_tfile_raw_ptr(filp_epoll, slot.tfd, slot.toff);
> +		fput(filp_epoll);
> +	} else
I think this "else" is unnecessary here.

> +
> +	if (IS_ERR(filp_tgt))
> +		return PTR_ERR(filp_tgt);
> +
> +	return kcmp_ptr(filp, filp_tgt, KCMP_FILE);
> +}
> +#else
> +static int kcmp_epoll_target(struct task_struct *task1,
> +			     struct task_struct *task2,
> +			     unsigned long idx1,
> +			     struct kcmp_epoll_slot __user *uslot)
> +{
> +	return -EOPNOTSUPP;
> +}
> +#endif
> +
>  SYSCALL_DEFINE5(kcmp, pid_t, pid1, pid_t, pid2, int, type,
>  		unsigned long, idx1, unsigned long, idx2)
>  {
> @@ -165,6 +219,9 @@ SYSCALL_DEFINE5(kcmp, pid_t, pid1, pid_t
>  		ret = -EOPNOTSUPP;
>  #endif
>  		break;
> +	case KCMP_EPOLL_TFD:
> +		ret = kcmp_epoll_target(task1, task2, idx1, (void *)idx2);
> +		break;
>  	default:
>  		ret = -EINVAL;
>  		break;

      parent reply	other threads:[~2017-09-17 16:01 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-04-24 15:39 [patch v4 resend 2/2] kcmp: Add KCMP_EPOLL_TFD mode to compare epoll target files Cyrill Gorcunov
     [not found] ` <20170424154423.511592110-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2017-05-12 22:00   ` Andrew Morton
     [not found]     ` <20170512150018.b931c7f5295dd7484845fcec-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>
2017-05-12 22:14       ` Cyrill Gorcunov
2017-05-12 22:41   ` Jann Horn
2017-05-12 22:53     ` Cyrill Gorcunov
     [not found]       ` <20170512225340.GD1881-ZmlpmtaulQd+urZeOPWqwQ@public.gmane.org>
2017-05-13  1:45         ` Andrei Vagin
     [not found]           ` <20170513014508.GA21900-1ViLX0X+lBJGNQ1M2rI3KwRV3xvJKrda@public.gmane.org>
2017-05-13  6:55             ` Cyrill Gorcunov
     [not found]               ` <20170513065514.GE1881-ZmlpmtaulQd+urZeOPWqwQ@public.gmane.org>
2017-05-13  7:15                 ` Cyrill Gorcunov
2017-09-17 16:01   ` Eugene Syromiatnikov [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20170917160141.GA3493@asgard.redhat.com \
    --to=esyr-h+wxahxf7alqt0dzr+alfa@public.gmane.org \
    --cc=akpm-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org \
    --cc=avagin-5HdwGun5lf+gSpxsJD1C4w@public.gmane.org \
    --cc=avagin-GEFAQzZX7r8dnm+yROfE0A@public.gmane.org \
    --cc=gorcunov-GEFAQzZX7r8dnm+yROfE0A@public.gmane.org \
    --cc=gorcunov-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
    --cc=jbaron-JqFfY2XvxFXQT0dZR+AlfA@public.gmane.org \
    --cc=linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=luto-kltTT9wpgjJwATOyAt5JVQ@public.gmane.org \
    --cc=mtk.manpages-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
    --cc=viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org \
    --cc=xemul-5HdwGun5lf+gSpxsJD1C4w@public.gmane.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).