Re: [Qemu-devel] [PATCH 11/13] 9p: darwin: Mark mknod as unsupported
On Thu, May 31, 2018 at 7:06 PM, Keno Fischer wrote: > On Thu, May 31, 2018 at 6:56 PM, Keno Fischer wrote: My concern was that allowing this would cause unexpected behavior, since the device numbers will differ between OS X and Linux. Though maybe this isn't the place to worry about that. >>> >>> The numbers may differ indeed but we don't really care since the >>> server never opens device files. This is just a directory entry. >> >> Ok, let me try to implement it. However, I don't think it is possible >> to implement mknodat (or at least I can't think of how) on Darwin >> directly. I could use regular mknod, but presumably, this is used >> to avoid a race condition between creating the device and setting >> the permissions. Can you think of a good way to resolve that? > > Would it work to fchdir in to the directory and use a cwd-relative > mknod, then fchdir back? Or do we need to maintain the cwd > while in this code? Sorry for the triple-self-post here, but I took a look at the Darwin kernel source and there's an unexposed (from the Darwin C library) syscall that only changes the current thread's cwd. That seems like it should be safe, so I'll go ahead and use that to implement mknodat. I'll also submit a feature request to apple to implement mknodat.
Re: [Qemu-devel] [PATCH 11/13] 9p: darwin: Mark mknod as unsupported
On Thu, May 31, 2018 at 6:56 PM, Keno Fischer wrote: >>> My concern was that allowing this would cause unexpected >>> behavior, since the device numbers will differ between OS X >>> and Linux. Though maybe this isn't the place to worry about >>> that. >> >> The numbers may differ indeed but we don't really care since the >> server never opens device files. This is just a directory entry. > > Ok, let me try to implement it. However, I don't think it is possible > to implement mknodat (or at least I can't think of how) on Darwin > directly. I could use regular mknod, but presumably, this is used > to avoid a race condition between creating the device and setting > the permissions. Can you think of a good way to resolve that? Would it work to fchdir in to the directory and use a cwd-relative mknod, then fchdir back? Or do we need to maintain the cwd while in this code?
Re: [Qemu-devel] [PATCH 11/13] 9p: darwin: Mark mknod as unsupported
>> My concern was that allowing this would cause unexpected >> behavior, since the device numbers will differ between OS X >> and Linux. Though maybe this isn't the place to worry about >> that. > > The numbers may differ indeed but we don't really care since the > server never opens device files. This is just a directory entry. Ok, let me try to implement it. However, I don't think it is possible to implement mknodat (or at least I can't think of how) on Darwin directly. I could use regular mknod, but presumably, this is used to avoid a race condition between creating the device and setting the permissions. Can you think of a good way to resolve that?
Re: [Qemu-devel] [PATCH 11/13] 9p: darwin: Mark mknod as unsupported
On Thu, 31 May 2018 12:37:56 -0400 Keno Fischer wrote: > >> +#ifdef CONFIG_DARWIN > >> +/* Darwin doesn't have mknodat and it's unlikely to work anyway, > > > > What's unlikely to work ? > > > > My concern was that allowing this would cause unexpected > behavior, since the device numbers will differ between OS X > and Linux. Though maybe this isn't the place to worry about > that. The numbers may differ indeed but we don't really care since the server never opens device files. This is just a directory entry.
Re: [Qemu-devel] [PATCH 11/13] 9p: darwin: Mark mknod as unsupported
>> +#ifdef CONFIG_DARWIN >> +/* Darwin doesn't have mknodat and it's unlikely to work anyway, > > What's unlikely to work ? > My concern was that allowing this would cause unexpected behavior, since the device numbers will differ between OS X and Linux. Though maybe this isn't the place to worry about that.
Re: [Qemu-devel] [PATCH 11/13] 9p: darwin: Mark mknod as unsupported
On Sat, 26 May 2018 01:23:13 -0400
k...@juliacomputing.com wrote:
> From: Keno Fischer
>
> Signed-off-by: Keno Fischer
> ---
> hw/9pfs/9p-local.c | 9 +
> 1 file changed, 9 insertions(+)
>
> diff --git a/hw/9pfs/9p-local.c b/hw/9pfs/9p-local.c
> index c55ea25..3e358b7 100644
> --- a/hw/9pfs/9p-local.c
> +++ b/hw/9pfs/9p-local.c
> @@ -669,6 +669,13 @@ static int local_mknod(FsContext *fs_ctx, V9fsPath
> *dir_path,
> return -1;
> }
>
> +#ifdef CONFIG_DARWIN
> +/* Darwin doesn't have mknodat and it's unlikely to work anyway,
What's unlikely to work ?
> + so let's just mark it as unsupported */
> +err = -1;
> +errno = EOPNOTSUPP;
> +goto out;
> +#else
Please introduce qemu_mknodat() with distinct implementations for linux
and darwin.
> if (fs_ctx->export_flags & V9FS_SM_MAPPED ||
> fs_ctx->export_flags & V9FS_SM_MAPPED_FILE) {
> err = mknodat(dirfd, name, fs_ctx->fmode | S_IFREG, 0);
> @@ -699,6 +706,8 @@ static int local_mknod(FsContext *fs_ctx, V9fsPath
> *dir_path,
>
> err_end:
> unlinkat_preserve_errno(dirfd, name, 0);
> +#endif
> +
> out:
> close_preserve_errno(dirfd);
> return err;
