Recent changes to feature-requests

deal with disappearing random source

Anne Marcel Roorda — Fri, 27 Jan 2012 09:52:14 -0000

On some router platforms hw_random is provided by the wifi driver. It isn't available before it is loaded, and disappears when the drivers are unloaded. Current behaviour prevents startup of rngd if /dev/hw_random isn't available yet, and grabs 100% cpu if it dissapears while running.

(wishlist) ethtool: man page:add "diagnostics" keyword?

Andreas Mohr — Mon, 24 Jan 2011 21:24:16 -0000

Hi,

a "diagnostic" keyword at the "-t --test" section would be very usability-useful (since the entire man page doesn't contain any "diag" keyword):
-t --test
Executes adapter selftest (diagnostics)...

Thanks!

[ethtool] Add a test for the Ethernet cable ?

Frederic MASSOT — Mon, 18 May 2009 14:17:32 -0000

Hi,

Some network adapter vendors provide a utility for Windows that allows you to have information on the status of the Ethernet cable pairs and the distance between the card and switch. Some BIOS have also this tool.

Is it possible (realistically) to have this test with ethtool?

Regards.

Add TPM RNG support to rng-tools

Sandeep K Shandilya — Tue, 11 Nov 2008 10:24:46 -0000

The TPM has a hardware random number generator that could be used as a source of entropy in rng-tools.

ethtool is linux-only

James K. Lowden — Wed, 15 Oct 2008 15:25:22 -0000

$ gmake
[...]
../ethtool.c:41:27: linux/sockios.h: No such file or directory

The description for the gkernel project indicates it's for BSDs and other non-Linux systems. Please change the description to indicate it won't build except on Linux.

ethtool -E > 82558 STEP B card not working

e_cod — Thu, 24 Jan 2008 23:40:00 -0000

History:
1. 82558 B onboard 10/100 card WOL worked under Windows XP SP2 (system IBM 6862 300PL)
2. on debian it did not kernel 2.6.23.14 e100
3. ethtool -e
Offset Values
------ ------
0x0000 00 06 29 d0 77 d9 00 01 03 00 01 02 01 47 00 00
0x0010 00 00 00 01 e5 5d d7 00 14 10 00 00 00 00 00 00
0x0020 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x0030 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x0040 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x0050 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x0060 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x0070 00 00 00 00 00 00 00 00 00 00 00 00 00 00 45 51
4.
00:03.0 Ethernet controller: Intel Corporation 82557/8/9 [Ethernet Pro 100] (rev 05)
Subsystem: IBM Unknown device 00d7
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR+ FastB2B-
Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR-
Latency: 64 (2000ns min, 14000ns max), Cache Line Size: 32 bytes
Interrupt: pin A routed to IRQ 15
Region 0: Memory at f3dff000 (32-bit, prefetchable) [size=4K]
Region 1: I/O ports at 7c60 [size=32]
Region 2: Memory at f3f00000 (32-bit, non-prefetchable) [size=1M]
Capabilities: [dc] Power Management version 1
Flags: PMEClk- DSI+ D1+ D2+ AuxCurrent=0mA PME(D0+,D1+,D2+,D3hot+,D3cold+)
Status: D0 PME-Enable- DSel=0 DScale=0 PME-
5.
changing WOL bit according to the manual: http://sourceforge.net/project/showfiles.php?group_id=42302&package_id=68544
ethtool -E eth0 magic 0x1234 offset 0x15 value 0x7d
Unfortunatelly I was not taking care about checksums.
6.
reboot
7.
e100 module does not recognize PCI NIC anymore:
lspci says:
00:03.0 0200: 1014:00d7 (rev 05)
Subsystem: 1014:00d7
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR+ FastB2B-
Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR-
Latency: 64 (2000ns min, 14000ns max), Cache Line Size: 32 bytes
Interrupt: pin A routed to IRQ 15
Region 0: Memory at f3dff000 (32-bit, prefetchable) [size=4K]
Region 1: I/O ports at 7c60 [size=32]
Region 2: Memory at f3f00000 (32-bit, non-prefetchable) [size=1M]
Capabilities: [dc] Power Management version 1
Flags: PMEClk- DSI+ D1+ D2+ AuxCurrent=0mA PME(D0+,D1+,D2+,D3hot+,D3cold+)
Status: D0 PME-Enable+ DSel=0 DScale=0 PME-

Changes:
VENDOR_ID 8086 (IBM) -> 1014
DEVICE_ID 1229 -> 00d7

the new values are coming probably from
SUBSYSTEM_VENDOR_ID
SUBSYSTEM_ID

Status:
from: Status: D0 PME-Enable- DSel=0 DScale=0 PME-
to: D0 PME-Enable+ DSel=0 DScale=0 PME-

8.
I tried without success:
setpci -v -G -s 3 VENDOR_ID=8086 DEVICE_ID=1229
output:
Trying method 1......using /sys/bus/pci...OK
Decided to use Linux-sysfs
00:03.0:00 8086
00:03.0:02 1229

new flash BIOS update

9.
What could be next? any idea?
thanks in advance

New wake up condition: wake-after-delay

Javier Cardona — Wed, 16 Jan 2008 19:28:00 -0000

The libertas wireless module provides a programmable wake up condition: the host configures a wake up delay, and the wireless module will raise a wake up signal after the delay expires. This is useful for testing the wake up sequence without having to generate wake up traffic.
Could you consider the attached patch for inclusion?

Thanks,

Javier

From: Javier Cardona <javier@cozybit.com>
Date: Wed, 16 Jan 2008 11:08:40 -0800
Subject: [PATCH] New wake-after-delay option.

A wake-after-delay wake up condition allows testing the wake up sequence without having to generate wireless traffic.
---
ethtool-copy.h | 1 +
ethtool.c | 7 ++++++-
2 files changed, 7 insertions(+), 1 deletions(-)

diff --git a/ethtool-copy.h b/ethtool-copy.h
index 3a63224..1efe77a 100644
--- a/ethtool-copy.h
+++ b/ethtool-copy.h
@@ -500,5 +500,6 @@ struct ethtool_ops {
#define WAKE_ARP (1 << 4)
#define WAKE_MAGIC (1 << 5)
#define WAKE_MAGICSECURE (1 << 6) /* only meaningful if WAKE_MAGIC */
+#define WAKE_DELAY (1 << 7) /* wake after delay, for testing */

#endif /* _LINUX_ETHTOOL_H */
diff --git a/ethtool.c b/ethtool.c
index 3adf843..b01d16d 100644
--- a/ethtool.c
+++ b/ethtool.c
@@ -106,7 +106,7 @@ static struct option {
" [ advertise %%x ]\n"
" [ phyad %%d ]\n"
" [ xcvr internal|external ]\n"
- " [ wol p|u|m|b|a|g|s|d... ]\n"
+ " [ wol p|u|m|b|a|g|s|y|d... ]\n"
" [ sopass %%x:%%x:%%x:%%x:%%x:%%x ]\n"
" [ msglvl %%d ] \n" },
{ "-a", "--show-pause", MODE_GPAUSE, "Show pause options" },
@@ -942,6 +942,9 @@ static int parse_wolopts(char *optstr, u32 *data)
case 's':
*data |= WAKE_MAGICSECURE;
break;
+ case 'y':
+ *data |= WAKE_DELAY;
+ break;
case 'd':
*data = 0;
break;
@@ -975,6 +978,8 @@ static char *unparse_wolopts(int wolopts)
*p++ = 'g';
if (wolopts & WAKE_MAGICSECURE)
*p++ = 's';
+ if (wolopts & WAKE_DELAY)
+ *p++ = 'y';
} else {

Status of link connection?

Anonymous — Fri, 27 Jul 2007 10:22:54 -0000

I use mii-tool to check the current link status (speed and if it is on-line). As I understand, ethtool is a replacement for mii-tool. Is it possible to get similar information from ethtool?

rngd syslog interval parameter

Anonymous — Sun, 15 Jul 2007 16:26:34 -0000

I understand the need to be able to check at regular interval whether the entropy source is still correctly functionning but rngd wakes up hard drives from ther sleep state every time it has to dump stats to the syslog.
Could we have the same parameter as rngtest:
-t n, --timedstats=n (default: 0)
Dump statistics every n secods, if n is not zero.

[quoted from man rngtest)

This way we could set a larger interval if needed.

thanks

support limits on autoneg advertised speed

Paul Raines — Fri, 29 Sep 2006 19:12:06 -0000

I have a box with a gigabit NIC using the tg3 driver on
a switch port that is gigabit. I have no access to the
switch. I want to limit the box to 100Mbps. If I just
run:

ethtool -s eth0 autoneg off speed 100 duplex full

I end up getting lots of RX errors as shown by ifconfig
as the switch port is still sending at gigabit.

ethtool should have a way to do autoneg but with limits
on the speed. So that doing

ethtool -s eth0 autoneg on speed 100,10

for example would only negotiate with the switch port
for speeds up to 100Mbps.

The tg3 driver has no way to pass parameters to it like
the old bcm203x driver did to achieve this. They
dropped support of parameter in favor of people using
ethtool but ethtool does not support autoneg limits.