From: Jeff Garzik <jeff@garzik•org>
To: Ben Hutchings <bhutchings@solarflare•com>
Cc: David Miller <davem@davemloft•net>,
netdev@vger•kernel.org, linux-net-drivers@solarflare•com
Subject: Re: [PATCH net-next-2.6] sfc: Implement ethtool register dump operation
Date: Mon, 21 Jun 2010 14:08:51 -0400 [thread overview]
Message-ID: <4C1FAAB3.8000403@garzik.org> (raw)
In-Reply-To: <1277142725.2100.36.camel@achroite.uk.solarflarecom.com>
On 06/21/2010 01:52 PM, Ben Hutchings wrote:
> On Mon, 2010-06-21 at 13:39 -0400, Jeff Garzik wrote:
>> On 06/21/2010 09:06 AM, Ben Hutchings wrote:
>>> Signed-off-by: Ben Hutchings<bhutchings@solarflare•com>
>>> ---
>>> drivers/net/sfc/ethtool.c | 16 +++
>>> drivers/net/sfc/io.h | 7 +
>>> drivers/net/sfc/nic.c | 266 +++++++++++++++++++++++++++++++++++++++++++++
>>> drivers/net/sfc/nic.h | 3 +
>>> 4 files changed, 292 insertions(+), 0 deletions(-)
>>>
>>> diff --git a/drivers/net/sfc/ethtool.c b/drivers/net/sfc/ethtool.c
>>> index 22026bf..81b7f39 100644
>>> --- a/drivers/net/sfc/ethtool.c
>>> +++ b/drivers/net/sfc/ethtool.c
>>> @@ -242,6 +242,20 @@ static void efx_ethtool_get_drvinfo(struct net_device *net_dev,
>>> strlcpy(info->bus_info, pci_name(efx->pci_dev), sizeof(info->bus_info));
>>> }
>>>
>>> +static int efx_ethtool_get_regs_len(struct net_device *net_dev)
>>> +{
>>> + return efx_nic_get_regs_len(netdev_priv(net_dev));
>>> +}
>>> +
>>> +static void efx_ethtool_get_regs(struct net_device *net_dev,
>>> + struct ethtool_regs *regs, void *buf)
>>> +{
>>> + struct efx_nic *efx = netdev_priv(net_dev);
>>> +
>>> + regs->version = efx->type->revision;
>>> + efx_nic_get_regs(efx, buf);
>>
>> regs->version is really the version of hardware register dump exported
>> to userspace. You might want to hardcode this in the driver, to be
>> changed when efx_nic_regs[] array changes, rather than tying
>> regs->version directly to the SFC hardware architecture revision.
>>
>> However, if that limitation is OK with you, ie. ethtool register dump
>> code will change when h/w arch rev changes, then
>
> This is exactly what we want, as the set of defined registers (and
> fields within each register) will change in each architecture revision.
> (It may not change in each chip revision, but efx_nic_type::revision is
> supposed to reflect the architecture revision.)
>
> Although this code doesn't include an explicit format version, we can
> consider the current format to be version 0 and increment it if we ever
> want to make changes to the format for existing NIC revisions.
Yep -- that's the intented purpose. If you ever want to dump additional
registers or information, it might be useful to avoid a hard coupling
with the hardware arch revision.
Either way is fine, of course.
Jeff
prev parent reply other threads:[~2010-06-21 18:08 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-06-21 13:06 [PATCH net-next-2.6] sfc: Implement ethtool register dump operation Ben Hutchings
2010-06-21 17:39 ` Jeff Garzik
2010-06-21 17:52 ` Ben Hutchings
2010-06-21 18:08 ` Jeff Garzik [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=4C1FAAB3.8000403@garzik.org \
--to=jeff@garzik$(echo .)org \
--cc=bhutchings@solarflare$(echo .)com \
--cc=davem@davemloft$(echo .)net \
--cc=linux-net-drivers@solarflare$(echo .)com \
--cc=netdev@vger$(echo .)kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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