[PATCH v2 7/8] ipconfig: Create /proc/net/ipconfig directory
To allow ipconfig to report IP configuration details to user space
processes without cluttering /proc/net, create a new subdirectory
/proc/net/ipconfig. All files containing IP configuration details should
be written to this directory.
Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk>
---
net/ipv4/ipconfig.c | 15 +++
1 file changed, 15 insertions(+)
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index e11dfd29a929..9abf833f3a99 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -158,6 +158,9 @@ static u8 ic_domain[64];/* DNS (not NIS) domain
name */
* Private state.
*/
+/* proc_dir_entry for /proc/net/ipconfig */
+static struct proc_dir_entry *ipconfig_dir;
+
/* Name of user-selected boot device */
static char user_dev_name[IFNAMSIZ] __initdata = { 0, };
@@ -1301,6 +1304,16 @@ static const struct file_operations pnp_seq_fops = {
.llseek = seq_lseek,
.release= single_release,
};
+
+/* Create the /proc/net/ipconfig directory */
+static int ipconfig_proc_net_init(void)
+{
+ ipconfig_dir = proc_net_mkdir(_net, "ipconfig", init_net.proc_net);
+ if (!ipconfig_dir)
+ return -ENOMEM;
+
+ return 0;
+}
#endif /* CONFIG_PROC_FS */
/*
@@ -1384,6 +1397,8 @@ static int __init ip_auto_config(void)
#ifdef CONFIG_PROC_FS
proc_create("pnp", 0444, init_net.proc_net, _seq_fops);
+
+ ipconfig_proc_net_init();
#endif /* CONFIG_PROC_FS */
if (!ic_enable)
--
2.14.1
[PATCH v2 6/8] ipconfig: Correctly initialise ic_nameservers
ic_nameservers, which stores the list of name servers discovered by
ipconfig, is initialised (i.e. has all of its elements set to NONE, or
0x) by ic_nameservers_predef() in the following scenarios:
- before the "ip=" and "nfsaddrs=" kernel command line parameters are
parsed (in ip_auto_config_setup());
- before autoconfiguring via DHCP or BOOTP (in ic_bootp_init()), in
order to clear any values that may have been set after parsing "ip="
or "nfsaddrs=" and are no longer needed.
This means that ic_nameservers_predef() is not called when neither "ip="
nor "nfsaddrs=" is specified on the kernel command line. In this
scenario, every element in ic_nameservers remains set to 0x,
which is indistinguishable from ANY and causes pnp_seq_show() to write
the following (bogus) information to /proc/net/pnp:
#MANUAL
nameserver 0.0.0.0
nameserver 0.0.0.0
nameserver 0.0.0.0
This is potentially problematic for systems that blindly link
/etc/resolv.conf to /proc/net/pnp.
Ensure that ic_nameservers is also initialised when neither "ip=" nor
"nfsaddrs=" are specified by calling ic_nameservers_predef() in
ip_auto_config(), but only when ip_auto_config_setup() was not called
earlier. This causes the following to be written to /proc/net/pnp, and
is consistent with what gets written when ipconfig is configured
manually but no name servers are specified on the kernel command line:
#MANUAL
Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk>
---
net/ipv4/ipconfig.c | 13 +
1 file changed, 13 insertions(+)
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index 0f460d6d3cce..e11dfd29a929 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -750,6 +750,11 @@ static void __init ic_bootp_init_ext(u8 *e)
*/
static inline void __init ic_bootp_init(void)
{
+ /* Re-initialise all name servers to NONE, in case any were set via the
+* "ip=" or "nfsaddrs=" kernel command line parameters: any IP addresses
+* specified there will already have been decoded but are no longer
+* needed
+*/
ic_nameservers_predef();
dev_add_pack(_packet_type);
@@ -1370,6 +1375,13 @@ static int __init ip_auto_config(void)
int err;
unsigned int i;
+ /* Initialise all name servers to NONE (but only if the "ip=" or
+* "nfsaddrs=" kernel command line parameters weren't decoded, otherwise
+* we'll overwrite the IP addresses specified there)
+*/
+ if (ic_set_manually == 0)
+ ic_nameservers_predef();
+
#ifdef CONFIG_PROC_FS
proc_create("pnp", 0444, init_net.proc_net, _seq_fops);
#endif /* CONFIG_PROC_FS */
@@ -1593,6 +1605,7 @@ static int __init ip_auto_config_setup(char *addrs)
return 1;
}
+ /* Initialise all name servers to NONE */
ic_nameservers_predef();
/* Parse string for static IP assignment. */
--
2.14.1
[PATCH v2 4/8] ipconfig: BOOTP: Request CONF_NAMESERVERS_MAX name servers
When ipconfig is autoconfigured via BOOTP, the request packet initialised by ic_bootp_init_ext() always allocates 8 bytes for the name server option, limiting the BOOTP server to responding with at most 2 name servers even though ipconfig in fact supports an arbitrary number of name servers (as defined by CONF_NAMESERVERS_MAX, which is currently 3). Only request name servers in the request packet if CONF_NAMESERVERS_MAX is positive (to comply with [1, §3.8]), and allocate enough space in the packet for CONF_NAMESERVERS_MAX name servers to indicate the maximum number we can accept in response. [1] RFC 2132, "DHCP Options and BOOTP Vendor Extensions": https://tools.ietf.org/rfc/rfc2132.txt Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk> --- net/ipv4/ipconfig.c | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index bcf3c4f9882d..0f460d6d3cce 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -721,9 +721,11 @@ static void __init ic_bootp_init_ext(u8 *e) *e++ = 3; /* Default gateway request */ *e++ = 4; e += 4; +#if CONF_NAMESERVERS_MAX > 0 *e++ = 6; /* (DNS) name server request */ - *e++ = 8; - e += 8; + *e++ = 4 * CONF_NAMESERVERS_MAX; + e += 4 * CONF_NAMESERVERS_MAX; +#endif *e++ = 12; /* Host name request */ *e++ = 32; e += 32; -- 2.14.1
[PATCH v2 8/8] ipconfig: Write NTP server IPs to /proc/net/ipconfig/ntp_servers
Distributed filesystems are most effective when the server and client clocks are synchronised. Embedded devices often use NFS for their root filesystem but typically do not contain an RTC, so the clocks of the NFS server and the embedded device will be out-of-sync when the root filesystem is mounted (and may not be synchronised until late in the boot process). Extend ipconfig with the ability to export IP addresses of NTP servers it discovers to /proc/net/ipconfig/ntp_servers. They can be supplied as follows: - If ipconfig is configured manually via the "ip=" or "nfsaddrs=" kernel command line parameters, one NTP server can be specified in the new "" parameter. - If ipconfig is autoconfigured via DHCP, request DHCP option 42 in the DHCPDISCOVER message, and record the IP addresses of up to three NTP servers sent by the responding DHCP server in the subsequent DHCPOFFER message. ipconfig will only write the NTP server IP addresses it discovers to /proc/net/ipconfig/ntp_servers, one per line (in the order received from the DHCP server, if DHCP autoconfiguration is used); making use of these NTP servers is the responsibility of a user space process (e.g. an initrd/initram script that invokes an NTP client before mounting an NFS root filesystem). Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk> --- Documentation/filesystems/nfs/nfsroot.txt | 35 +++-- net/ipv4/ipconfig.c | 118 +++--- 2 files changed, 136 insertions(+), 17 deletions(-) diff --git a/Documentation/filesystems/nfs/nfsroot.txt b/Documentation/filesystems/nfs/nfsroot.txt index a1030bea60d3..d2963123eb1c 100644 --- a/Documentation/filesystems/nfs/nfsroot.txt +++ b/Documentation/filesystems/nfs/nfsroot.txt @@ -5,6 +5,7 @@ Written 1996 by Gero Kuhlmann <g...@gkminix.han.de> Updated 1997 by Martin Mares <m...@atrey.karlin.mff.cuni.cz> Updated 2006 by Nico Schottelius <nico-kernel-nfsr...@schottelius.org> Updated 2006 by Horms <ho...@verge.net.au> +Updated 2018 by Chris Novakovic <ch...@chrisn.me.uk> @@ -79,7 +80,7 @@ nfsroot=[:][,] ip=::: - : + :: This parameter tells the kernel how to configure IP addresses of devices and also how to set up the IP routing table. It was originally called @@ -178,9 +179,18 @@ ip=::: IP address of secondary nameserver. See . - After configuration (whether manual or automatic) is complete, a file is - created at /proc/net/pnp in the following format; lines are omitted if - their respective value is empty following configuration. + IP address of a Network Time Protocol (NTP) server. + Value is exported to /proc/net/ipconfig/ntp_servers, but is + otherwise unused (see below). + + Default: None if not using autoconfiguration; determined + automatically if using autoconfiguration. + + After configuration (whether manual or automatic) is complete, two files + are created in the following format; lines are omitted if their respective + value is empty following configuration: + + - /proc/net/pnp: #PROTO: <DHCP|BOOTP|RARP|MANUAL>(depending on configuration method) domain (if autoconfigured, the DNS domain) @@ -189,13 +199,26 @@ ip=::: nameserver (tertiary name server IP) bootserver (NFS server IP) - and are requested during autoconfiguration; they - cannot be specified as part of the "ip=" kernel command line parameter. + - /proc/net/ipconfig/ntp_servers: + + (NTP server IP) + (NTP server IP) + (NTP server IP) + + and (in /proc/net/pnp) and and + (in /proc/net/ipconfig/ntp_servers) are requested during autoconfiguration; + they cannot be specified as part of the "ip=" kernel command line parameter. Because the "domain" and "nameserver" options are recognised by DNS resolvers, /etc/resolv.conf is often linked to /proc/net/pnp on systems that use an NFS root filesystem. + Note that the kernel will not synchronise the system time with any NTP + servers it discovers; this is the responsibility of a user space process + (e.g. an initrd/initramfs script that passes the IP addresses listed in + /proc/net/ipconfig/ntp_servers to an NTP client before mounting the real + root filesystem if it is on NFS). + nfsrootdebug diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index 9abf833f3a99..d839d74853fc 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -28,6 +28,9 @@ * * Multiple Nameservers in /proc/net/pnp * -- Josef Siemes <jsie...@web.de>, Aug 2002 + * + * NTP servers in /proc/net/ipconfig/ntp_servers + * -- Chris
[PATCH v2 1/8] ipconfig: Document setting of NIS domain name
ic_do_bootp_ext() is responsible for parsing the "ip=" and "nfsaddrs=" kernel parameters. If a "." character is found in parameter 4 (the client's hostname), everything before the first "." is used as the hostname, and everything after it is used as the NIS domain name (but not necessarily the DNS domain name). Document this behaviour in Documentation/filesystems/nfs/nfsroot.txt, as it is not made explicit. Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk> --- Documentation/filesystems/nfs/nfsroot.txt | 11 +++ 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/Documentation/filesystems/nfs/nfsroot.txt b/Documentation/filesystems/nfs/nfsroot.txt index 5efae00f6c7f..1513e5d663fd 100644 --- a/Documentation/filesystems/nfs/nfsroot.txt +++ b/Documentation/filesystems/nfs/nfsroot.txt @@ -123,10 +123,13 @@ ip=::: Default: Determined using autoconfiguration. - Name of the client. May be supplied by autoconfiguration, - but its absence will not trigger autoconfiguration. - If specified and DHCP is used, the user provided hostname will - be carried in the DHCP request to hopefully update DNS record. + Name of the client. If a '.' character is present, anything + before the first '.' is used as the client's hostname, and anything + after it is used as its NIS domain name. May be supplied by + autoconfiguration, but its absence will not trigger autoconfiguration. + If specified and DHCP is used, the user-provided hostname (and NIS + domain name, if present) will be carried in the DHCP request; this + may cause a DNS record to be created or updated for the client. Default: Client IP address is used in ASCII notation. -- 2.14.1
[PATCH v2 0/8] ipconfig: NTP server support, bug fixes, documentation improvements
This series (against net-next) makes various improvements to ipconfig: - Patch #1 correctly documents the behaviour of parameter 4 in the "ip=" and "nfsaddrs=" command line parameter. - Patch #2 tidies up the printk()s for reporting configured name servers. - Patch #3 fixes a bug in autoconfiguration via BOOTP whereby the IP addresses of IEN-116 name servers are requested from the BOOTP server, rather than those of DNS name servers. - Patch #4 requests the number of DNS servers specified by CONF_NAMESERVERS_MAX when autoconfiguring via BOOTP, rather than hardcoding it to 2. - Patch #5 fully documents the contents and format of /proc/net/pnp in Documentation/filesystems/nfs/nfsroot.txt. - Patch #6 fixes a bug whereby bogus information is written to /proc/net/pnp when ipconfig is not used. - Patch #7 creates a new procfs directory for ipconfig-related configuration reports at /proc/net/ipconfig. - Patch #8 allows for NTP servers to be configured (manually on the kernel command line or automatically via DHCP), enabling systems with an NFS root filesystem to synchronise their clock before mounting their root filesystem. NTP server IP addresses are written to /proc/net/ipconfig/ntp_servers. Changes from v1: - David requested that a new directory /proc/net/ipconfig be created to contain ipconfig-related configuration reports, which is implemented in the new patch #7. NTP server IPs are now written to this directory instead of /proc/net/ntp in the new patch #8. - Cong and David both requested that the modification to CREDITS be dropped. This patch has been removed from the series. Chris Novakovic (8): ipconfig: Document setting of NIS domain name ipconfig: Tidy up reporting of name servers ipconfig: BOOTP: Don't request IEN-116 name servers ipconfig: BOOTP: Request CONF_NAMESERVERS_MAX name servers ipconfig: Document /proc/net/pnp ipconfig: Correctly initialise ic_nameservers ipconfig: Create /proc/net/ipconfig directory ipconfig: Write NTP server IPs to /proc/net/ipconfig/ntp_servers Documentation/filesystems/nfs/nfsroot.txt | 70 -- net/ipv4/ipconfig.c | 151 +++--- 2 files changed, 200 insertions(+), 21 deletions(-) -- 2.14.1
[PATCH v2 2/8] ipconfig: Tidy up reporting of name servers
Commit 5e953778a2aab04929a5e7b69f53dc26e39b079e ("ipconfig: add
nameserver IPs to kernel-parameter ip=") adds the IP addresses of
discovered name servers to the summary printed by ipconfig when
configuration is complete. It appears the intention in ip_auto_config()
was to print the name servers on a new line (especially given the
spacing and lack of comma before "nameserver0="), but they're actually
printed on the same line as the NFS root filesystem configuration
summary:
[0.686186] IP-Config: Complete:
[0.686226] device=eth0, hwaddr=xx:xx:xx:xx:xx:xx, ipaddr=10.0.0.2,
mask=255.255.255.0, gw=10.0.0.1
[0.686328] host=test, domain=example.com, nis-domain=(none)
[0.686386] bootserver=10.0.0.1, rootserver=10.0.0.1, rootpath=
nameserver0=10.0.0.1
This makes it harder to read and parse ipconfig's output. Instead, print
the name servers on a separate line:
[0.791250] IP-Config: Complete:
[0.791289] device=eth0, hwaddr=xx:xx:xx:xx:xx:xx, ipaddr=10.0.0.2,
mask=255.255.255.0, gw=10.0.0.1
[0.791407] host=test, domain=example.com, nis-domain=(none)
[0.791475] bootserver=10.0.0.1, rootserver=10.0.0.1, rootpath=
[0.791476] nameserver0=10.0.0.1
Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk>
---
net/ipv4/ipconfig.c | 19 +++
1 file changed, 11 insertions(+), 8 deletions(-)
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index 43f620feb1c4..d0ea0ecc9008 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -1481,16 +1481,19 @@ static int __init ip_auto_config(void)
_servaddr, _server_addr, root_server_path);
if (ic_dev_mtu)
pr_cont(", mtu=%d", ic_dev_mtu);
- for (i = 0; i < CONF_NAMESERVERS_MAX; i++)
+ /* Name servers (if any): */
+ for (i = 0; i < CONF_NAMESERVERS_MAX; i++) {
if (ic_nameservers[i] != NONE) {
- pr_cont(" nameserver%u=%pI4",
- i, _nameservers[i]);
- break;
+ if (i == 0)
+ pr_info(" nameserver%u=%pI4",
+ i, _nameservers[i]);
+ else
+ pr_cont(", nameserver%u=%pI4",
+ i, _nameservers[i]);
}
- for (i++; i < CONF_NAMESERVERS_MAX; i++)
- if (ic_nameservers[i] != NONE)
- pr_cont(", nameserver%u=%pI4", i, _nameservers[i]);
- pr_cont("\n");
+ if (i + 1 == CONF_NAMESERVERS_MAX)
+ pr_cont("\n");
+ }
#endif /* !SILENT */
/*
--
2.14.1
[PATCH v2 5/8] ipconfig: Document /proc/net/pnp
Fully document the format used by the /proc/net/pnp file written by ipconfig, explain where its values originate from, and clarify that the tertiary name server IP and DNS domain name are only written to the file when autoconfiguration is used. Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk> --- Documentation/filesystems/nfs/nfsroot.txt | 34 ++- 1 file changed, 29 insertions(+), 5 deletions(-) diff --git a/Documentation/filesystems/nfs/nfsroot.txt b/Documentation/filesystems/nfs/nfsroot.txt index 1513e5d663fd..a1030bea60d3 100644 --- a/Documentation/filesystems/nfs/nfsroot.txt +++ b/Documentation/filesystems/nfs/nfsroot.txt @@ -110,6 +110,9 @@ ip=::: will not be triggered if it is missing and NFS root is not in operation. + Value is exported to /proc/net/pnp with the prefix "bootserver " + (see below). + Default: Determined using autoconfiguration. The address of the autoconfiguration server is used. @@ -165,12 +168,33 @@ ip=::: Default: any - IP address of first nameserver. - Value gets exported by /proc/net/pnp which is often linked - on embedded systems by /etc/resolv.conf. + IP address of primary nameserver. + Value is exported to /proc/net/pnp with the prefix "nameserver " + (see below). + + Default: None if not using autoconfiguration; determined + automatically if using autoconfiguration. + + IP address of secondary nameserver. + See . + + After configuration (whether manual or automatic) is complete, a file is + created at /proc/net/pnp in the following format; lines are omitted if + their respective value is empty following configuration. + + #PROTO: <DHCP|BOOTP|RARP|MANUAL>(depending on configuration method) + domain (if autoconfigured, the DNS domain) + nameserver (primary name server IP) + nameserver (secondary name server IP) + nameserver (tertiary name server IP) + bootserver (NFS server IP) + + and are requested during autoconfiguration; they + cannot be specified as part of the "ip=" kernel command line parameter. - IP address of second nameserver. - Same as above. + Because the "domain" and "nameserver" options are recognised by DNS + resolvers, /etc/resolv.conf is often linked to /proc/net/pnp on systems + that use an NFS root filesystem. nfsrootdebug -- 2.14.1
[PATCH v2 3/8] ipconfig: BOOTP: Don't request IEN-116 name servers
When ipconfig is autoconfigured via BOOTP, the request packet
initialised by ic_bootp_init_ext() allocates 8 bytes for tag 5 ("Name
Server" [1, §3.7]), but tag 5 in the response isn't processed by
ic_do_bootp_ext(). Instead, allocate the 8 bytes to tag 6 ("Domain Name
Server" [1, §3.8]), which is processed by ic_do_bootp_ext(), and appears
to have been the intended tag to request.
This won't cause any breakage for existing users, as tag 5 responses
provided by BOOTP servers weren't being processed anyway.
[1] RFC 2132, "DHCP Options and BOOTP Vendor Extensions":
https://tools.ietf.org/rfc/rfc2132.txt
Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk>
---
net/ipv4/ipconfig.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index d0ea0ecc9008..bcf3c4f9882d 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -721,7 +721,7 @@ static void __init ic_bootp_init_ext(u8 *e)
*e++ = 3; /* Default gateway request */
*e++ = 4;
e += 4;
- *e++ = 5; /* Name server request */
+ *e++ = 6; /* (DNS) name server request */
*e++ = 8;
e += 8;
*e++ = 12; /* Host name request */
--
2.14.1
Re: [PATCH 0/8] ipconfig: NTP server support, bug fixes, documentation improvements
On 18/04/2018 19:06, Chris Novakovic wrote: > On 18/04/2018 18:59, David Miller wrote: >> I think a plain file named /proc/net/ntp is quite confusing. It doesn't >> give any indication that it's a special file populated only by ipconfig >> and not some general NTP thing the kernel is doing. >> >> I would suggest creating a subdirectory like /proc/net/ipconfig or similar >> to put such files. Then the use and meaning is clear. > > That sounds more sensible --- I'll rework patch 7 and resend. (Would you > prefer me to post the entire series to the list again, or just that one > patch?) Sorry, I just realised how stupid that sounded: I can't remove patch 8 unless I repost the series... :)
Re: [PATCH 0/8] ipconfig: NTP server support, bug fixes, documentation improvements
On 18/04/2018 18:59, David Miller wrote: > I think a plain file named /proc/net/ntp is quite confusing. It doesn't > give any indication that it's a special file populated only by ipconfig > and not some general NTP thing the kernel is doing. > > I would suggest creating a subdirectory like /proc/net/ipconfig or similar > to put such files. Then the use and meaning is clear. That sounds more sensible --- I'll rework patch 7 and resend. (Would you prefer me to post the entire series to the list again, or just that one patch?) > Also, as noted please remove patch 8 updating CREDITS, this really isn't > done any more. Will do.
[PATCH 5/8] ipconfig: Document /proc/net/pnp
Fully document the format used by the /proc/net/pnp file written by ipconfig, explain where its values originate from, and clarify that the tertiary name server IP and DNS domain name are only written to the file when autoconfiguration is used. Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk> --- Documentation/filesystems/nfs/nfsroot.txt | 34 ++- 1 file changed, 29 insertions(+), 5 deletions(-) diff --git a/Documentation/filesystems/nfs/nfsroot.txt b/Documentation/filesystems/nfs/nfsroot.txt index 1513e5d663fd..a1030bea60d3 100644 --- a/Documentation/filesystems/nfs/nfsroot.txt +++ b/Documentation/filesystems/nfs/nfsroot.txt @@ -110,6 +110,9 @@ ip=::: will not be triggered if it is missing and NFS root is not in operation. + Value is exported to /proc/net/pnp with the prefix "bootserver " + (see below). + Default: Determined using autoconfiguration. The address of the autoconfiguration server is used. @@ -165,12 +168,33 @@ ip=::: Default: any - IP address of first nameserver. - Value gets exported by /proc/net/pnp which is often linked - on embedded systems by /etc/resolv.conf. + IP address of primary nameserver. + Value is exported to /proc/net/pnp with the prefix "nameserver " + (see below). + + Default: None if not using autoconfiguration; determined + automatically if using autoconfiguration. + + IP address of secondary nameserver. + See . + + After configuration (whether manual or automatic) is complete, a file is + created at /proc/net/pnp in the following format; lines are omitted if + their respective value is empty following configuration. + + #PROTO: <DHCP|BOOTP|RARP|MANUAL>(depending on configuration method) + domain (if autoconfigured, the DNS domain) + nameserver (primary name server IP) + nameserver (secondary name server IP) + nameserver (tertiary name server IP) + bootserver (NFS server IP) + + and are requested during autoconfiguration; they + cannot be specified as part of the "ip=" kernel command line parameter. - IP address of second nameserver. - Same as above. + Because the "domain" and "nameserver" options are recognised by DNS + resolvers, /etc/resolv.conf is often linked to /proc/net/pnp on systems + that use an NFS root filesystem. nfsrootdebug -- 2.14.1
[PATCH 2/8] ipconfig: Tidy up reporting of name servers
Commit 5e953778a2aab04929a5e7b69f53dc26e39b079e ("ipconfig: add
nameserver IPs to kernel-parameter ip=") adds the IP addresses of
discovered name servers to the summary printed by ipconfig when
configuration is complete. It appears the intention in ip_auto_config()
was to print the name servers on a new line (especially given the
spacing and lack of comma before "nameserver0="), but they're actually
printed on the same line as the NFS root filesystem configuration
summary:
[0.686186] IP-Config: Complete:
[0.686226] device=eth0, hwaddr=xx:xx:xx:xx:xx:xx, ipaddr=10.0.0.2,
mask=255.255.255.0, gw=10.0.0.1
[0.686328] host=test, domain=example.com, nis-domain=(none)
[0.686386] bootserver=10.0.0.1, rootserver=10.0.0.1, rootpath=
nameserver0=10.0.0.1
This makes it harder to read and parse ipconfig's output. Instead, print
the name servers on a separate line:
[0.791250] IP-Config: Complete:
[0.791289] device=eth0, hwaddr=xx:xx:xx:xx:xx:xx, ipaddr=10.0.0.2,
mask=255.255.255.0, gw=10.0.0.1
[0.791407] host=test, domain=example.com, nis-domain=(none)
[0.791475] bootserver=10.0.0.1, rootserver=10.0.0.1, rootpath=
[0.791476] nameserver0=10.0.0.1
Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk>
---
net/ipv4/ipconfig.c | 19 +++
1 file changed, 11 insertions(+), 8 deletions(-)
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index 43f620feb1c4..d0ea0ecc9008 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -1481,16 +1481,19 @@ static int __init ip_auto_config(void)
_servaddr, _server_addr, root_server_path);
if (ic_dev_mtu)
pr_cont(", mtu=%d", ic_dev_mtu);
- for (i = 0; i < CONF_NAMESERVERS_MAX; i++)
+ /* Name servers (if any): */
+ for (i = 0; i < CONF_NAMESERVERS_MAX; i++) {
if (ic_nameservers[i] != NONE) {
- pr_cont(" nameserver%u=%pI4",
- i, _nameservers[i]);
- break;
+ if (i == 0)
+ pr_info(" nameserver%u=%pI4",
+ i, _nameservers[i]);
+ else
+ pr_cont(", nameserver%u=%pI4",
+ i, _nameservers[i]);
}
- for (i++; i < CONF_NAMESERVERS_MAX; i++)
- if (ic_nameservers[i] != NONE)
- pr_cont(", nameserver%u=%pI4", i, _nameservers[i]);
- pr_cont("\n");
+ if (i + 1 == CONF_NAMESERVERS_MAX)
+ pr_cont("\n");
+ }
#endif /* !SILENT */
/*
--
2.14.1
[PATCH 4/8] ipconfig: BOOTP: Request CONF_NAMESERVERS_MAX name servers
When ipconfig is autoconfigured via BOOTP, the request packet initialised by ic_bootp_init_ext() always allocates 8 bytes for the name server option, limiting the BOOTP server to responding with at most 2 name servers even though ipconfig in fact supports an arbitrary number of name servers (as defined by CONF_NAMESERVERS_MAX, which is currently 3). Only request name servers in the request packet if CONF_NAMESERVERS_MAX is positive (to comply with [1, §3.8]), and allocate enough space in the packet for CONF_NAMESERVERS_MAX name servers to indicate the maximum number we can accept in response. [1] RFC 2132, "DHCP Options and BOOTP Vendor Extensions": https://tools.ietf.org/rfc/rfc2132.txt Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk> --- net/ipv4/ipconfig.c | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index bcf3c4f9882d..0f460d6d3cce 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -721,9 +721,11 @@ static void __init ic_bootp_init_ext(u8 *e) *e++ = 3; /* Default gateway request */ *e++ = 4; e += 4; +#if CONF_NAMESERVERS_MAX > 0 *e++ = 6; /* (DNS) name server request */ - *e++ = 8; - e += 8; + *e++ = 4 * CONF_NAMESERVERS_MAX; + e += 4 * CONF_NAMESERVERS_MAX; +#endif *e++ = 12; /* Host name request */ *e++ = 32; e += 32; -- 2.14.1
[PATCH 8/8] CREDITS: Add Chris Novakovic
Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk> --- CREDITS | 4 1 file changed, 4 insertions(+) diff --git a/CREDITS b/CREDITS index 989cda91c427..5a13bf62c569 100644 --- a/CREDITS +++ b/CREDITS @@ -2765,6 +2765,10 @@ E: nor...@nocrew.org W: http://www.lysator.liu.se/~noring/ D: dsp56k device driver +N: Chris Novakovic +E: ch...@chrisn.me.uk +D: ipconfig: NTP server support, bug fixes, documentation + N: Michael O'Reilly E: mich...@iinet.com.au E: oreil...@tartarus.uwa.edu.au -- 2.14.1
[PATCH 6/8] ipconfig: Correctly initialise ic_nameservers
ic_nameservers, which stores the list of name servers discovered by
ipconfig, is initialised (i.e. has all of its elements set to NONE, or
0x) by ic_nameservers_predef() in the following scenarios:
- before the "ip=" and "nfsaddrs=" kernel command line parameters are
parsed (in ip_auto_config_setup());
- before autoconfiguring via DHCP or BOOTP (in ic_bootp_init()), in
order to clear any values that may have been set after parsing "ip="
or "nfsaddrs=" and are no longer needed.
This means that ic_nameservers_predef() is not called when neither "ip="
nor "nfsaddrs=" is specified on the kernel command line. In this
scenario, every element in ic_nameservers remains set to 0x,
which is indistinguishable from ANY and causes pnp_seq_show() to write
the following (bogus) information to /proc/net/pnp:
#MANUAL
nameserver 0.0.0.0
nameserver 0.0.0.0
nameserver 0.0.0.0
This is potentially problematic for systems that blindly link
/etc/resolv.conf to /proc/net/pnp.
Ensure that ic_nameservers is also initialised when neither "ip=" nor
"nfsaddrs=" are specified by calling ic_nameservers_predef() in
ip_auto_config(), but only when ip_auto_config_setup() was not called
earlier. This causes the following to be written to /proc/net/pnp, and
is consistent with what gets written when ipconfig is configured
manually but no name servers are specified on the kernel command line:
#MANUAL
Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk>
---
net/ipv4/ipconfig.c | 13 +
1 file changed, 13 insertions(+)
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index 0f460d6d3cce..e11dfd29a929 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -750,6 +750,11 @@ static void __init ic_bootp_init_ext(u8 *e)
*/
static inline void __init ic_bootp_init(void)
{
+ /* Re-initialise all name servers to NONE, in case any were set via the
+* "ip=" or "nfsaddrs=" kernel command line parameters: any IP addresses
+* specified there will already have been decoded but are no longer
+* needed
+*/
ic_nameservers_predef();
dev_add_pack(_packet_type);
@@ -1370,6 +1375,13 @@ static int __init ip_auto_config(void)
int err;
unsigned int i;
+ /* Initialise all name servers to NONE (but only if the "ip=" or
+* "nfsaddrs=" kernel command line parameters weren't decoded, otherwise
+* we'll overwrite the IP addresses specified there)
+*/
+ if (ic_set_manually == 0)
+ ic_nameservers_predef();
+
#ifdef CONFIG_PROC_FS
proc_create("pnp", 0444, init_net.proc_net, _seq_fops);
#endif /* CONFIG_PROC_FS */
@@ -1593,6 +1605,7 @@ static int __init ip_auto_config_setup(char *addrs)
return 1;
}
+ /* Initialise all name servers to NONE */
ic_nameservers_predef();
/* Parse string for static IP assignment. */
--
2.14.1
[PATCH 0/8] ipconfig: NTP server support, bug fixes, documentation improvements
This series (against net-next) makes various improvements to ipconfig. It was mistakenly posted on 2018-04-07, while net-next was closed; it still applies cleanly. - Patch #1 correctly documents the behaviour of parameter 4 in the "ip=" and "nfsaddrs=" command line parameter. - Patch #2 tidies up the printk()s for reporting configured name servers. - Patch #3 fixes a bug in autoconfiguration via BOOTP whereby the IP addresses of IEN-116 name servers are requested from the BOOTP server, rather than those of DNS name servers. - Patch #4 requests the number of DNS servers specified by CONF_NAMESERVERS_MAX when autoconfiguring via BOOTP, rather than hardcoding it to 2. - Patch #5 fully documents the contents and format of /proc/net/pnp in Documentation/filesystems/nfs/nfsroot.txt. - Patch #6 fixes a bug whereby bogus information is written to /proc/net/pnp when ipconfig is not used. - Patch #7 allows for NTP servers to be configured (manually on the kernel command line or automatically via DHCP), enabling systems with an NFS root filesystem to synchronise their clock before mounting their root filesystem. Chris Novakovic (8): ipconfig: Document setting of NIS domain name ipconfig: Tidy up reporting of name servers ipconfig: BOOTP: Don't request IEN-116 name servers ipconfig: BOOTP: Request CONF_NAMESERVERS_MAX name servers ipconfig: Document /proc/net/pnp ipconfig: Correctly initialise ic_nameservers ipconfig: Write NTP server IPs to /proc/net/ntp CREDITS: Add Chris Novakovic CREDITS | 4 + Documentation/filesystems/nfs/nfsroot.txt | 70 ++--- net/ipv4/ipconfig.c | 121 +++--- 3 files changed, 174 insertions(+), 21 deletions(-) -- 2.14.1
[PATCH 1/8] ipconfig: Document setting of NIS domain name
ic_do_bootp_ext() is responsible for parsing the "ip=" and "nfsaddrs=" kernel parameters. If a "." character is found in parameter 4 (the client's hostname), everything before the first "." is used as the hostname, and everything after it is used as the NIS domain name (but not necessarily the DNS domain name). Document this behaviour in Documentation/filesystems/nfs/nfsroot.txt, as it is not made explicit. Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk> --- Documentation/filesystems/nfs/nfsroot.txt | 11 +++ 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/Documentation/filesystems/nfs/nfsroot.txt b/Documentation/filesystems/nfs/nfsroot.txt index 5efae00f6c7f..1513e5d663fd 100644 --- a/Documentation/filesystems/nfs/nfsroot.txt +++ b/Documentation/filesystems/nfs/nfsroot.txt @@ -123,10 +123,13 @@ ip=::: Default: Determined using autoconfiguration. - Name of the client. May be supplied by autoconfiguration, - but its absence will not trigger autoconfiguration. - If specified and DHCP is used, the user provided hostname will - be carried in the DHCP request to hopefully update DNS record. + Name of the client. If a '.' character is present, anything + before the first '.' is used as the client's hostname, and anything + after it is used as its NIS domain name. May be supplied by + autoconfiguration, but its absence will not trigger autoconfiguration. + If specified and DHCP is used, the user-provided hostname (and NIS + domain name, if present) will be carried in the DHCP request; this + may cause a DNS record to be created or updated for the client. Default: Client IP address is used in ASCII notation. -- 2.14.1
[PATCH 3/8] ipconfig: BOOTP: Don't request IEN-116 name servers
When ipconfig is autoconfigured via BOOTP, the request packet
initialised by ic_bootp_init_ext() allocates 8 bytes for tag 5 ("Name
Server" [1, §3.7]), but tag 5 in the response isn't processed by
ic_do_bootp_ext(). Instead, allocate the 8 bytes to tag 6 ("Domain Name
Server" [1, §3.8]), which is processed by ic_do_bootp_ext(), and appears
to have been the intended tag to request.
This won't cause any breakage for existing users, as tag 5 responses
provided by BOOTP servers weren't being processed anyway.
[1] RFC 2132, "DHCP Options and BOOTP Vendor Extensions":
https://tools.ietf.org/rfc/rfc2132.txt
Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk>
---
net/ipv4/ipconfig.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index d0ea0ecc9008..bcf3c4f9882d 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -721,7 +721,7 @@ static void __init ic_bootp_init_ext(u8 *e)
*e++ = 3; /* Default gateway request */
*e++ = 4;
e += 4;
- *e++ = 5; /* Name server request */
+ *e++ = 6; /* (DNS) name server request */
*e++ = 8;
e += 8;
*e++ = 12; /* Host name request */
--
2.14.1
[PATCH 7/8] ipconfig: Write NTP server IPs to /proc/net/ntp
Distributed filesystems are most effective when the server and client clocks are synchronised. Embedded devices often use NFS for their root filesystem but typically do not contain an RTC, so the clocks of the NFS server and the embedded device will be out-of-sync when the root filesystem is mounted (and may not be synchronised until late in the boot process). Extend ipconfig with the ability to export IP addresses of NTP servers it discovers to /proc/net/ntp. They can be supplied as follows: - If ipconfig is configured manually via the "ip=" or "nfsaddrs=" kernel command line parameters, one NTP server can be specified in the new "" parameter. - If ipconfig is autoconfigured via DHCP, request DHCP option 42 in the DHCPDISCOVER message, and record the IP addresses of up to three NTP servers sent by the responding DHCP server in the subsequent DHCPOFFER message. ipconfig will only write the NTP server IP addresses it discovers to /proc/net/ntp, one per line (in the order received from the DHCP server, if DHCP autoconfiguration is used); making use of these NTP servers is the responsibility of a user space process (e.g. an initrd/initram script that invokes an NTP client before mounting an NFS root filesystem). Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk> --- Documentation/filesystems/nfs/nfsroot.txt | 35 +-- net/ipv4/ipconfig.c | 99 --- 2 files changed, 119 insertions(+), 15 deletions(-) diff --git a/Documentation/filesystems/nfs/nfsroot.txt b/Documentation/filesystems/nfs/nfsroot.txt index a1030bea60d3..8e51c1a39564 100644 --- a/Documentation/filesystems/nfs/nfsroot.txt +++ b/Documentation/filesystems/nfs/nfsroot.txt @@ -5,6 +5,7 @@ Written 1996 by Gero Kuhlmann <g...@gkminix.han.de> Updated 1997 by Martin Mares <m...@atrey.karlin.mff.cuni.cz> Updated 2006 by Nico Schottelius <nico-kernel-nfsr...@schottelius.org> Updated 2006 by Horms <ho...@verge.net.au> +Updated 2018 by Chris Novakovic <ch...@chrisn.me.uk> @@ -79,7 +80,7 @@ nfsroot=[:][,] ip=::: - : + :: This parameter tells the kernel how to configure IP addresses of devices and also how to set up the IP routing table. It was originally called @@ -178,9 +179,18 @@ ip=::: IP address of secondary nameserver. See . - After configuration (whether manual or automatic) is complete, a file is - created at /proc/net/pnp in the following format; lines are omitted if - their respective value is empty following configuration. + IP address of a Network Time Protocol (NTP) server. + Value is exported to /proc/net/ntp, but is otherwise unused + (see below). + + Default: None if not using autoconfiguration; determined + automatically if using autoconfiguration. + + After configuration (whether manual or automatic) is complete, two files + are created in the following format; lines are omitted if their respective + value is empty following configuration: + + - /proc/net/pnp: #PROTO: <DHCP|BOOTP|RARP|MANUAL>(depending on configuration method) domain (if autoconfigured, the DNS domain) @@ -189,13 +199,26 @@ ip=::: nameserver (tertiary name server IP) bootserver (NFS server IP) - and are requested during autoconfiguration; they - cannot be specified as part of the "ip=" kernel command line parameter. + - /proc/net/ntp: + + (NTP server IP) + (NTP server IP) + (NTP server IP) + + and (in /proc/net/pnp) and and + (in /proc/net/ntp) are requested during autoconfiguration; they cannot be + specified as part of the "ip=" kernel command line parameter. Because the "domain" and "nameserver" options are recognised by DNS resolvers, /etc/resolv.conf is often linked to /proc/net/pnp on systems that use an NFS root filesystem. + Note that the kernel will not synchronise the system time with any NTP + servers it discovers; this is the responsibility of a user space process + (e.g. an initrd/initramfs script that passes the IP addresses listed in + /proc/net/ntp to an NTP client before mounting the real root filesystem + if it is on NFS). + nfsrootdebug diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index e11dfd29a929..a5d68e506494 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -28,6 +28,9 @@ * * Multiple Nameservers in /proc/net/pnp * -- Josef Siemes <jsie...@web.de>, Aug 2002 + * + * NTP servers in /proc/net/ntp + * -- Chris Novakovic <ch...@chrisn.me.uk>, April 2018 */ #include @@ -93,6 +96,7 @@ #define CONF_TIME
Re: [PATCH 0/8] ipconfig: NTP server support, bug fixes, documentation improvements
On 07/04/2018 17:29, David Miller wrote: > From: Chris Novakovic <ch...@chrisn.me.uk> > Date: Sat, 7 Apr 2018 05:08:55 +0100 > >> This series (against net-next) makes various improvements to ipconfig: > > The net-next tree is closed at this time, please resubmit this series > when that tree opens up again. > > Thank you. Will do --- apologies for the noise.
[PATCH v2 7/8] ipconfig: Write NTP server IPs to /proc/net/ntp
Distributed filesystems are most effective when the server and client clocks are synchronised. Embedded devices often use NFS for their root filesystem but typically do not contain an RTC, so the clocks of the NFS server and the embedded device will be out-of-sync when the root filesystem is mounted (and may not be synchronised until late in the boot process). Extend ipconfig with the ability to export IP addresses of NTP servers it discovers to /proc/net/ntp. They can be supplied as follows: - If ipconfig is configured manually via the "ip=" or "nfsaddrs=" kernel command line parameters, one NTP server can be specified in the new "" parameter. - If ipconfig is autoconfigured via DHCP, request DHCP option 42 in the DHCPDISCOVER message, and record the IP addresses of up to three NTP servers sent by the responding DHCP server in the subsequent DHCPOFFER message. ipconfig will only write the NTP server IP addresses it discovers to /proc/net/ntp, one per line (in the order received from the DHCP server, if DHCP autoconfiguration is used); making use of these NTP servers is the responsibility of a user space process (e.g. an initrd/initram script that invokes an NTP client before mounting an NFS root filesystem). Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk> --- Documentation/filesystems/nfs/nfsroot.txt | 35 +-- net/ipv4/ipconfig.c | 99 --- 2 files changed, 119 insertions(+), 15 deletions(-) diff --git a/Documentation/filesystems/nfs/nfsroot.txt b/Documentation/filesystems/nfs/nfsroot.txt index a1030bea60d3..8e51c1a39564 100644 --- a/Documentation/filesystems/nfs/nfsroot.txt +++ b/Documentation/filesystems/nfs/nfsroot.txt @@ -5,6 +5,7 @@ Written 1996 by Gero Kuhlmann <g...@gkminix.han.de> Updated 1997 by Martin Mares <m...@atrey.karlin.mff.cuni.cz> Updated 2006 by Nico Schottelius <nico-kernel-nfsr...@schottelius.org> Updated 2006 by Horms <ho...@verge.net.au> +Updated 2018 by Chris Novakovic <ch...@chrisn.me.uk> @@ -79,7 +80,7 @@ nfsroot=[:][,] ip=::: - : + :: This parameter tells the kernel how to configure IP addresses of devices and also how to set up the IP routing table. It was originally called @@ -178,9 +179,18 @@ ip=::: IP address of secondary nameserver. See . - After configuration (whether manual or automatic) is complete, a file is - created at /proc/net/pnp in the following format; lines are omitted if - their respective value is empty following configuration. + IP address of a Network Time Protocol (NTP) server. + Value is exported to /proc/net/ntp, but is otherwise unused + (see below). + + Default: None if not using autoconfiguration; determined + automatically if using autoconfiguration. + + After configuration (whether manual or automatic) is complete, two files + are created in the following format; lines are omitted if their respective + value is empty following configuration: + + - /proc/net/pnp: #PROTO: <DHCP|BOOTP|RARP|MANUAL>(depending on configuration method) domain (if autoconfigured, the DNS domain) @@ -189,13 +199,26 @@ ip=::: nameserver (tertiary name server IP) bootserver (NFS server IP) - and are requested during autoconfiguration; they - cannot be specified as part of the "ip=" kernel command line parameter. + - /proc/net/ntp: + + (NTP server IP) + (NTP server IP) + (NTP server IP) + + and (in /proc/net/pnp) and and + (in /proc/net/ntp) are requested during autoconfiguration; they cannot be + specified as part of the "ip=" kernel command line parameter. Because the "domain" and "nameserver" options are recognised by DNS resolvers, /etc/resolv.conf is often linked to /proc/net/pnp on systems that use an NFS root filesystem. + Note that the kernel will not synchronise the system time with any NTP + servers it discovers; this is the responsibility of a user space process + (e.g. an initrd/initramfs script that passes the IP addresses listed in + /proc/net/ntp to an NTP client before mounting the real root filesystem + if it is on NFS). + nfsrootdebug diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index e11dfd29a929..a5d68e506494 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -28,6 +28,9 @@ * * Multiple Nameservers in /proc/net/pnp * -- Josef Siemes <jsie...@web.de>, Aug 2002 + * + * NTP servers in /proc/net/ntp + * -- Chris Novakovic <ch...@chrisn.me.uk>, April 2018 */ #include @@ -93,6 +96,7 @@ #define CONF_TIME
Re: [PATCH 7/8] ipconfig: Write NTP server IPs to /proc/net/ntp
On 07/04/2018 05:09, Chris Novakovic wrote: > Distributed filesystems are most effective when the server and client > clocks are synchronised. Embedded devices often use NFS for their > root filesystem but typically do not contain an RTC, so the clocks of > the NFS server and the embedded device will be out-of-sync when the root > filesystem is mounted (and may not be synchronised until late in the > boot process). > > Extend ipconfig with the ability to export IP addresses of NTP servers > it discovers to /proc/net/ntp. They can be supplied as follows: > > - If ipconfig is configured manually via the "ip=" or "nfsaddrs=" >kernel command line parameters, one NTP server can be specified in >the new "" parameter. > - If ipconfig is autoconfigured via DHCP, request DHCP option 42 in >the DHCPDISCOVER message, and record the IP addresses of up to three >NTP servers sent by the responding DHCP server in the subsequent >DHCPOFFER message. > > ipconfig will only write the NTP server IP addresses it discovers to > /proc/net/ntp, one per line (in the order received from the DHCP server, > if DHCP autoconfiguration is used); making use of these NTP servers is > the responsibility of a user space process (e.g. an initrd/initram > script that invokes an NTP client before mounting an NFS root > filesystem). > > Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk> Sorry, I just spotted a minor typo in the commit message and documentation in this patch --- I'll send a v2 shortly.
[PATCH 2/8] ipconfig: Tidy up reporting of name servers
Commit 5e953778a2aab04929a5e7b69f53dc26e39b079e ("ipconfig: add
nameserver IPs to kernel-parameter ip=") adds the IP addresses of
discovered name servers to the summary printed by ipconfig when
configuration is complete. It appears the intention in ip_auto_config()
was to print the name servers on a new line (especially given the
spacing and lack of comma before "nameserver0="), but they're actually
printed on the same line as the NFS root filesystem configuration
summary:
[0.686186] IP-Config: Complete:
[0.686226] device=eth0, hwaddr=xx:xx:xx:xx:xx:xx, ipaddr=10.0.0.2,
mask=255.255.255.0, gw=10.0.0.1
[0.686328] host=test, domain=example.com, nis-domain=(none)
[0.686386] bootserver=10.0.0.1, rootserver=10.0.0.1, rootpath=
nameserver0=10.0.0.1
This makes it harder to read and parse ipconfig's output. Instead, print
the name servers on a separate line:
[0.791250] IP-Config: Complete:
[0.791289] device=eth0, hwaddr=xx:xx:xx:xx:xx:xx, ipaddr=10.0.0.2,
mask=255.255.255.0, gw=10.0.0.1
[0.791407] host=test, domain=example.com, nis-domain=(none)
[0.791475] bootserver=10.0.0.1, rootserver=10.0.0.1, rootpath=
[0.791476] nameserver0=10.0.0.1
Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk>
---
net/ipv4/ipconfig.c | 19 +++
1 file changed, 11 insertions(+), 8 deletions(-)
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index 43f620feb1c4..d0ea0ecc9008 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -1481,16 +1481,19 @@ static int __init ip_auto_config(void)
_servaddr, _server_addr, root_server_path);
if (ic_dev_mtu)
pr_cont(", mtu=%d", ic_dev_mtu);
- for (i = 0; i < CONF_NAMESERVERS_MAX; i++)
+ /* Name servers (if any): */
+ for (i = 0; i < CONF_NAMESERVERS_MAX; i++) {
if (ic_nameservers[i] != NONE) {
- pr_cont(" nameserver%u=%pI4",
- i, _nameservers[i]);
- break;
+ if (i == 0)
+ pr_info(" nameserver%u=%pI4",
+ i, _nameservers[i]);
+ else
+ pr_cont(", nameserver%u=%pI4",
+ i, _nameservers[i]);
}
- for (i++; i < CONF_NAMESERVERS_MAX; i++)
- if (ic_nameservers[i] != NONE)
- pr_cont(", nameserver%u=%pI4", i, _nameservers[i]);
- pr_cont("\n");
+ if (i + 1 == CONF_NAMESERVERS_MAX)
+ pr_cont("\n");
+ }
#endif /* !SILENT */
/*
--
2.14.1
[PATCH 3/8] ipconfig: BOOTP: Don't request IEN-116 name servers
When ipconfig is autoconfigured via BOOTP, the request packet
initialised by ic_bootp_init_ext() allocates 8 bytes for tag 5 ("Name
Server" [1, §3.7]), but tag 5 in the response isn't processed by
ic_do_bootp_ext(). Instead, allocate the 8 bytes to tag 6 ("Domain Name
Server" [1, §3.8]), which is processed by ic_do_bootp_ext(), and appears
to have been the intended tag to request.
This won't cause any breakage for existing users, as tag 5 responses
provided by BOOTP servers weren't being processed anyway.
[1] RFC 2132, "DHCP Options and BOOTP Vendor Extensions":
https://tools.ietf.org/rfc/rfc2132.txt
Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk>
---
net/ipv4/ipconfig.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index d0ea0ecc9008..bcf3c4f9882d 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -721,7 +721,7 @@ static void __init ic_bootp_init_ext(u8 *e)
*e++ = 3; /* Default gateway request */
*e++ = 4;
e += 4;
- *e++ = 5; /* Name server request */
+ *e++ = 6; /* (DNS) name server request */
*e++ = 8;
e += 8;
*e++ = 12; /* Host name request */
--
2.14.1
[PATCH 0/8] ipconfig: NTP server support, bug fixes, documentation improvements
This series (against net-next) makes various improvements to ipconfig: - Patch #1 correctly documents the behaviour of parameter 4 in the "ip=" and "nfsaddrs=" command line parameter. - Patch #2 tidies up the printk()s for reporting configured name servers. - Patch #3 fixes a bug in autoconfiguration via BOOTP whereby the IP addresses of IEN-116 name servers are requested from the BOOTP server, rather than those of DNS name servers. - Patch #4 requests the number of DNS servers specified by CONF_NAMESERVERS_MAX when autoconfiguring via BOOTP, rather than hardcoding it to 2. - Patch #5 fully documents the contents and format of /proc/net/pnp in Documentation/filesystems/nfs/nfsroot.txt. - Patch #6 fixes a bug whereby bogus information is written to /proc/net/pnp when ipconfig is not used. - Patch #7 allows for NTP servers to be configured (manually on the kernel command line or automatically via DHCP), enabling systems with an NFS root filesystem to synchronise their clock before mounting their root filesystem. Patch #7 gets a few warnings when run through checkpatch.pl, but I felt it'd be better to use the same style as surrounding code where appropriate, rather than adhering strictly to the net/ style guide. Chris Novakovic (8): ipconfig: Document setting of NIS domain name ipconfig: Tidy up reporting of name servers ipconfig: BOOTP: Don't request IEN-116 name servers ipconfig: BOOTP: Request CONF_NAMESERVERS_MAX name servers ipconfig: Document /proc/net/pnp ipconfig: Correctly initialise ic_nameservers ipconfig: Write NTP server IPs to /proc/net/ntp CREDITS: Add Chris Novakovic CREDITS | 4 + Documentation/filesystems/nfs/nfsroot.txt | 70 ++--- net/ipv4/ipconfig.c | 121 +++--- 3 files changed, 174 insertions(+), 21 deletions(-) -- 2.14.1
[PATCH 5/8] ipconfig: Document /proc/net/pnp
Fully document the format used by the /proc/net/pnp file written by ipconfig, explain where its values originate from, and clarify that the tertiary name server IP and DNS domain name are only written to the file when autoconfiguration is used. Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk> --- Documentation/filesystems/nfs/nfsroot.txt | 34 ++- 1 file changed, 29 insertions(+), 5 deletions(-) diff --git a/Documentation/filesystems/nfs/nfsroot.txt b/Documentation/filesystems/nfs/nfsroot.txt index 1513e5d663fd..a1030bea60d3 100644 --- a/Documentation/filesystems/nfs/nfsroot.txt +++ b/Documentation/filesystems/nfs/nfsroot.txt @@ -110,6 +110,9 @@ ip=::: will not be triggered if it is missing and NFS root is not in operation. + Value is exported to /proc/net/pnp with the prefix "bootserver " + (see below). + Default: Determined using autoconfiguration. The address of the autoconfiguration server is used. @@ -165,12 +168,33 @@ ip=::: Default: any - IP address of first nameserver. - Value gets exported by /proc/net/pnp which is often linked - on embedded systems by /etc/resolv.conf. + IP address of primary nameserver. + Value is exported to /proc/net/pnp with the prefix "nameserver " + (see below). + + Default: None if not using autoconfiguration; determined + automatically if using autoconfiguration. + + IP address of secondary nameserver. + See . + + After configuration (whether manual or automatic) is complete, a file is + created at /proc/net/pnp in the following format; lines are omitted if + their respective value is empty following configuration. + + #PROTO: <DHCP|BOOTP|RARP|MANUAL>(depending on configuration method) + domain (if autoconfigured, the DNS domain) + nameserver (primary name server IP) + nameserver (secondary name server IP) + nameserver (tertiary name server IP) + bootserver (NFS server IP) + + and are requested during autoconfiguration; they + cannot be specified as part of the "ip=" kernel command line parameter. - IP address of second nameserver. - Same as above. + Because the "domain" and "nameserver" options are recognised by DNS + resolvers, /etc/resolv.conf is often linked to /proc/net/pnp on systems + that use an NFS root filesystem. nfsrootdebug -- 2.14.1
[PATCH 4/8] ipconfig: BOOTP: Request CONF_NAMESERVERS_MAX name servers
When ipconfig is autoconfigured via BOOTP, the request packet initialised by ic_bootp_init_ext() always allocates 8 bytes for the name server option, limiting the BOOTP server to responding with at most 2 name servers even though ipconfig in fact supports an arbitrary number of name servers (as defined by CONF_NAMESERVERS_MAX, which is currently 3). Only request name servers in the request packet if CONF_NAMESERVERS_MAX is positive (to comply with [1, §3.8]), and allocate enough space in the packet for CONF_NAMESERVERS_MAX name servers to indicate the maximum number we can accept in response. [1] RFC 2132, "DHCP Options and BOOTP Vendor Extensions": https://tools.ietf.org/rfc/rfc2132.txt Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk> --- net/ipv4/ipconfig.c | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index bcf3c4f9882d..0f460d6d3cce 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -721,9 +721,11 @@ static void __init ic_bootp_init_ext(u8 *e) *e++ = 3; /* Default gateway request */ *e++ = 4; e += 4; +#if CONF_NAMESERVERS_MAX > 0 *e++ = 6; /* (DNS) name server request */ - *e++ = 8; - e += 8; + *e++ = 4 * CONF_NAMESERVERS_MAX; + e += 4 * CONF_NAMESERVERS_MAX; +#endif *e++ = 12; /* Host name request */ *e++ = 32; e += 32; -- 2.14.1
[PATCH 1/8] ipconfig: Document setting of NIS domain name
ic_do_bootp_ext() is responsible for parsing the "ip=" and "nfsaddrs=" kernel parameters. If a "." character is found in parameter 4 (the client's hostname), everything before the first "." is used as the hostname, and everything after it is used as the NIS domain name (but not necessarily the DNS domain name). Document this behaviour in Documentation/filesystems/nfs/nfsroot.txt, as it is not made explicit. Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk> --- Documentation/filesystems/nfs/nfsroot.txt | 11 +++ 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/Documentation/filesystems/nfs/nfsroot.txt b/Documentation/filesystems/nfs/nfsroot.txt index 5efae00f6c7f..1513e5d663fd 100644 --- a/Documentation/filesystems/nfs/nfsroot.txt +++ b/Documentation/filesystems/nfs/nfsroot.txt @@ -123,10 +123,13 @@ ip=::: Default: Determined using autoconfiguration. - Name of the client. May be supplied by autoconfiguration, - but its absence will not trigger autoconfiguration. - If specified and DHCP is used, the user provided hostname will - be carried in the DHCP request to hopefully update DNS record. + Name of the client. If a '.' character is present, anything + before the first '.' is used as the client's hostname, and anything + after it is used as its NIS domain name. May be supplied by + autoconfiguration, but its absence will not trigger autoconfiguration. + If specified and DHCP is used, the user-provided hostname (and NIS + domain name, if present) will be carried in the DHCP request; this + may cause a DNS record to be created or updated for the client. Default: Client IP address is used in ASCII notation. -- 2.14.1
[PATCH 6/8] ipconfig: Correctly initialise ic_nameservers
ic_nameservers, which stores the list of name servers discovered by
ipconfig, is initialised (i.e. has all of its elements set to NONE, or
0x) by ic_nameservers_predef() in the following scenarios:
- before the "ip=" and "nfsaddrs=" kernel command line parameters are
parsed (in ip_auto_config_setup());
- before autoconfiguring via DHCP or BOOTP (in ic_bootp_init()), in
order to clear any values that may have been set after parsing "ip="
or "nfsaddrs=" and are no longer needed.
This means that ic_nameservers_predef() is not called when neither "ip="
nor "nfsaddrs=" is specified on the kernel command line. In this
scenario, every element in ic_nameservers remains set to 0x,
which is indistinguishable from ANY and causes pnp_seq_show() to write
the following (bogus) information to /proc/net/pnp:
#MANUAL
nameserver 0.0.0.0
nameserver 0.0.0.0
nameserver 0.0.0.0
This is potentially problematic for systems that blindly link
/etc/resolv.conf to /proc/net/pnp.
Ensure that ic_nameservers is also initialised when neither "ip=" nor
"nfsaddrs=" is specified by calling ic_nameservers_predef() in
ip_auto_config(), but only when ip_auto_config_setup() was not called
earlier. This causes the following to be written to /proc/net/pnp, and
is consistent with what gets written when ipconfig is configured
manually but no name servers are specified on the kernel command line:
#MANUAL
Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk>
---
net/ipv4/ipconfig.c | 13 +
1 file changed, 13 insertions(+)
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index 0f460d6d3cce..e11dfd29a929 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -750,6 +750,11 @@ static void __init ic_bootp_init_ext(u8 *e)
*/
static inline void __init ic_bootp_init(void)
{
+ /* Re-initialise all name servers to NONE, in case any were set via the
+* "ip=" or "nfsaddrs=" kernel command line parameters: any IP addresses
+* specified there will already have been decoded but are no longer
+* needed
+*/
ic_nameservers_predef();
dev_add_pack(_packet_type);
@@ -1370,6 +1375,13 @@ static int __init ip_auto_config(void)
int err;
unsigned int i;
+ /* Initialise all name servers to NONE (but only if the "ip=" or
+* "nfsaddrs=" kernel command line parameters weren't decoded, otherwise
+* we'll overwrite the IP addresses specified there)
+*/
+ if (ic_set_manually == 0)
+ ic_nameservers_predef();
+
#ifdef CONFIG_PROC_FS
proc_create("pnp", 0444, init_net.proc_net, _seq_fops);
#endif /* CONFIG_PROC_FS */
@@ -1593,6 +1605,7 @@ static int __init ip_auto_config_setup(char *addrs)
return 1;
}
+ /* Initialise all name servers to NONE */
ic_nameservers_predef();
/* Parse string for static IP assignment. */
--
2.14.1
[PATCH 7/8] ipconfig: Write NTP server IPs to /proc/net/ntp
Distributed filesystems are most effective when the server and client clocks are synchronised. Embedded devices often use NFS for their root filesystem but typically do not contain an RTC, so the clocks of the NFS server and the embedded device will be out-of-sync when the root filesystem is mounted (and may not be synchronised until late in the boot process). Extend ipconfig with the ability to export IP addresses of NTP servers it discovers to /proc/net/ntp. They can be supplied as follows: - If ipconfig is configured manually via the "ip=" or "nfsaddrs=" kernel command line parameters, one NTP server can be specified in the new "" parameter. - If ipconfig is autoconfigured via DHCP, request DHCP option 42 in the DHCPDISCOVER message, and record the IP addresses of up to three NTP servers sent by the responding DHCP server in the subsequent DHCPOFFER message. ipconfig will only write the NTP server IP addresses it discovers to /proc/net/ntp, one per line (in the order received from the DHCP server, if DHCP autoconfiguration is used); making use of these NTP servers is the responsibility of a user space process (e.g. an initrd/initram script that invokes an NTP client before mounting an NFS root filesystem). Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk> --- Documentation/filesystems/nfs/nfsroot.txt | 35 +-- net/ipv4/ipconfig.c | 99 --- 2 files changed, 119 insertions(+), 15 deletions(-) diff --git a/Documentation/filesystems/nfs/nfsroot.txt b/Documentation/filesystems/nfs/nfsroot.txt index a1030bea60d3..4d55470f7ca9 100644 --- a/Documentation/filesystems/nfs/nfsroot.txt +++ b/Documentation/filesystems/nfs/nfsroot.txt @@ -5,6 +5,7 @@ Written 1996 by Gero Kuhlmann <g...@gkminix.han.de> Updated 1997 by Martin Mares <m...@atrey.karlin.mff.cuni.cz> Updated 2006 by Nico Schottelius <nico-kernel-nfsr...@schottelius.org> Updated 2006 by Horms <ho...@verge.net.au> +Updated 2018 by Chris Novakovic <ch...@chrisn.me.uk> @@ -79,7 +80,7 @@ nfsroot=[:][,] ip=::: - : + :: This parameter tells the kernel how to configure IP addresses of devices and also how to set up the IP routing table. It was originally called @@ -178,9 +179,18 @@ ip=::: IP address of secondary nameserver. See . - After configuration (whether manual or automatic) is complete, a file is - created at /proc/net/pnp in the following format; lines are omitted if - their respective value is empty following configuration. + IP address of a Network Time Protocol (NTP) server. + Value is exported to /proc/net/ntp, but is otherwise unused + (see below). + + Default: None if not using autoconfiguration; determined + automatically if using autoconfiguration. + + After configuration (whether manual or automatic) is complete, two files + are created in the following format; lines are omitted if their respective + value is empty following configuration: + + - /proc/net/pnp: #PROTO: <DHCP|BOOTP|RARP|MANUAL>(depending on configuration method) domain (if autoconfigured, the DNS domain) @@ -189,13 +199,26 @@ ip=::: nameserver (tertiary name server IP) bootserver (NFS server IP) - and are requested during autoconfiguration; they - cannot be specified as part of the "ip=" kernel command line parameter. + - /proc/net/ntp: + + (NTP server IP) + (NTP server IP) + (NTP server IP) + + and (in /proc/net/pnp) and and + (in /proc/net/ntp) are requested during autoconfiguration; they cannot be + specified as part of the "ip=" kernel command line parameter. Because the "domain" and "nameserver" options are recognised by DNS resolvers, /etc/resolv.conf is often linked to /proc/net/pnp on systems that use an NFS root filesystem. + Note that the kernel will not synchronise the system time with any NTP + servers it discovers; this is the responsibility of a user space process + (e.g. an initrd/initramfs script that passes the IP addresses listed in + /proc/net/ntp to an NTP client before mounting the real root filesystem + if it is on NFS). + nfsrootdebug diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index e11dfd29a929..a5d68e506494 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -28,6 +28,9 @@ * * Multiple Nameservers in /proc/net/pnp * -- Josef Siemes <jsie...@web.de>, Aug 2002 + * + * NTP servers in /proc/net/ntp + * -- Chris Novakovic <ch...@chrisn.me.uk>, April 2018 */ #include @@ -93,6 +96,7 @@ #define CONF_TIME
[PATCH 8/8] CREDITS: Add Chris Novakovic
Signed-off-by: Chris Novakovic <ch...@chrisn.me.uk> --- CREDITS | 4 1 file changed, 4 insertions(+) diff --git a/CREDITS b/CREDITS index 989cda91c427..5a13bf62c569 100644 --- a/CREDITS +++ b/CREDITS @@ -2765,6 +2765,10 @@ E: nor...@nocrew.org W: http://www.lysator.liu.se/~noring/ D: dsp56k device driver +N: Chris Novakovic +E: ch...@chrisn.me.uk +D: ipconfig: NTP server support, bug fixes, documentation + N: Michael O'Reilly E: mich...@iinet.com.au E: oreil...@tartarus.uwa.edu.au -- 2.14.1
