Re: Sign-off for the IP1000A driver before inclusion
jesse\(??\) <[EMAIL PROTECTED]> :
[...]
> --Remove Threshold comfig and RxDMAInt from ipg_io_config(). Remove relative
> define form ipg.h
Na??ve question: why ?
Hardware assisted Rx irq mitigation is cool (TM).
> --Remove and Rewrite ipg_config_autoneg() function.
- Is there any hardware reason why the bits in MAC_CTRL are set one (*1*) at
a time through repeated use of iowrite32(ioread32(ioaddr+MAC_CTRL), ...) ?
I already changed it in a7302ea588bccc3df88bc8162af8b4d4acd0e380 so I
wonder if it was reverted by accident or not.
- The rewrite messed up the codingstyle (switch/case + some spaces/tabs +
see below)
[...]
struct ipg_nic_private* sp=netdev_priv(dev);
void __iomem *ioaddr = sp->ioaddr;
u8 phyctrl;
u32 asicctrl;
int fiber;
int gig;
int fullduplex;
int txflowcontrol;
int rxflowcontrol;
[...]
/* Configure full duplex, and flow control. */
if (fullduplex == 1)
{
/* Configure IPG for full duplex operation. */
printk(KERN_INFO "setting full duplex, ");
iowrite32(ioread32(ioaddr+MAC_CTRL)
|IPG_MC_DUPLEX_SELECT_FD,ioaddr+MAC_CTRL);
if (txflowcontrol == 1)
{
printk("TX flow control");
iowrite32(ioread32(ioaddr+MAC_CTRL)
|IPG_MC_TX_FLOW_CONTROL_ENABLE,ioaddr+MAC_CTRL);
} else
{
- At the top of ipg_config_autoneg(), I noticed that:
phyctrl = ioread8(ioaddr + PHY_CTRL);
turned into:
phyctrl = ioread32(ioaddr +PHY_CTRL);
I changed it back to ioread8() since that's what the driver uses everywhere
else when reading PHY_CTRL. Is it correct ?
Updated patch against fe0dd7dcecf6fbcc3b34099e064f7c0065caef71 below.
The patch has not been committed in the git tree so far.
diff --git a/drivers/net/ipg.c b/drivers/net/ipg.c
index 754ddb5..fe6ea02 100644
--- a/drivers/net/ipg.c
+++ b/drivers/net/ipg.c
@@ -5,6 +5,7 @@
* IC Plus IP1000 Gigabit Ethernet Adapter Linux Driver v2.01
* by IC Plus Corp. 2003
*
+ * Initiator:
* Craig Rich
* Sundance Technology, Inc.
* 1485 Saratoga Avenue
@@ -13,6 +14,14 @@
* 408 873 4117
* www.sundanceti.com
* [EMAIL PROTECTED]
+ *
+ * Current Maintainer:
+ * Sorbica Shieh.
+ * 10F, No.47, Lane 2, Kwang-Fu RD.
+ * Sec. 2, Hsin-Chu, Taiwan, R.O.C.
+ * http://www.icplus.com.tw
+ * [EMAIL PROTECTED]
+ * Signed-off-by: Sorbica Shieh <[EMAIL PROTECTED]>
*/
#include
#include
@@ -478,13 +487,6 @@ static int ipg_config_autoneg(struct net
void __iomem *ioaddr = sp->ioaddr;
u32 asicctrl;
u32 mac_ctl;
- int phyaddr = 0;
- u16 status = 0;
- u16 advertisement;
- u16 linkpartner_ability;
- u16 gigadvertisement;
- u16 giglinkpartner_ability;
- u16 techabilities;
int fiber;
int gig;
int fullduplex;
@@ -511,14 +513,12 @@ static int ipg_config_autoneg(struct net
*/
sp->tenmbpsmode = 0;
- printk("Link speed = ");
+ printk(KERN_INFO "Link speed = ");
/* Determine actual speed of operation. */
switch (phyctrl & IPG_PC_LINK_SPEED) {
case IPG_PC_LINK_SPEED_10MBPS:
printk("10Mbps.\n");
- printk(KERN_INFO "%s: 10Mbps operational mode enabled.\n",
- dev->name);
sp->tenmbpsmode = 1;
break;
case IPG_PC_LINK_SPEED_100MBPS:
@@ -530,282 +530,49 @@ static int ipg_config_autoneg(struct net
break;
default:
printk("undefined!\n");
- }
-
- fiber = ipg_sti_fiber_detect(dev);
-
- /* Determine if auto-negotiation resolution is necessary.
-* First check for fiber based media 10/100 media.
-*/
- if ((fiber == 1) &&
- (asicctrl & (IPG_AC_PHY_SPEED10 | IPG_AC_PHY_SPEED100))) {
- printk(KERN_INFO
- "%s: Fiber based PHY, setting full duplex, no flow
control.\n",
- dev->name);
-
- mac_ctl = ioread32(ioaddr + MAC_CTRL);
- mac_ctl |= IPG_MC_DUPLEX_SELECT_FD;
- mac_ctl &= ~IPG_MC_TX_FLOW_CONTROL_ENABLE;
- mac_ctl &= ~IPG_MC_RX_FLOW_CONTROL_ENABLE;
-
- iowrite32(IPG_MC_RSVD_MASK & mac_ctl, ioaddr + MAC_CTRL);
-
return 0;
}
- /* Determine if PHY is auto-negotiation capable. */
- phyaddr = ipg_find_phyaddr(dev);
-
- if (phyaddr == -1) {
- printk(KERN_INFO
- "%s: Error on read to GMII/MII Status register.\n",
- dev->name);
- return -EILSEQ;
- }
- sp->mii_if.phy_id = phyaddr;
-
-
Re: Sign-off for the IP1000A driver before inclusion
Hi Jesse, On Tue, 30 May 2006, jesse\(??\)~H~H\) wrote: > Sorry for that. I try to generate the patch file. But I only can use > "diff -uN" to generate it. The "diff --git" is not work. I still try to > generate it. I assume you're using Francois' git tree, right? What you need to do is: - Check out netdev-ipg branch: git checkout netdev-ipg - Create a new branch against netdev-ipg (this is where you'll do the work): git checkout -b ipg-mine - Commit patches in your ipg-mine branch. Use git update-index and git commit for this. Please commit each changeset separately. You can use my patchset as a starting point: http://www.cs.helsinki.fi/u/penberg/linux/ip1000-jesse/ - When you're done, you can generate diffs against the netdev-ipg branch: git format-patch netdev-ipg ipg-mine - If you want, you can then delete your working branch: git checkout netdev-ipg ; git branch -D ipg-mine On Tue, 30 May 2006, jesse\(??\)~H~H\) wrote: > --Changelog: > --Updata mentainer information > --Remove some default phy params > --Remove Threshold comfig and RxDMAInt from ipg_io_config(). Remove relative > define form ipg.h > --Remove and Rewrite ipg_config_autoneg() function. The changelog isn't telling me much. Why are you removing default phy params and the threshold config? Pekka - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Sign-off for the IP1000A driver before inclusion
On Mon, 29 May 2006, Andrew Morton wrote: > It takes people quite a few iterations to get patch preparation worked out. > > Pekka, if you have time, perhaps you can extract the patches for us? Jesse, I wasn't able to work out the ipg_config_autoneg() changes, but the rest of them are at: http://www.cs.helsinki.fi/u/penberg/linux/ip1000-jesse/ They're missing changelogs, though, so I'd much appreciate if Jesse submitted them properly to Francois who's maintaining the ipg git tree. Pekka - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Sign-off for the IP1000A driver before inclusion
On Tue, 30 May 2006 09:16:29 +0300 (EEST) Pekka J Enberg <[EMAIL PROTECTED]> wrote: > On Tue, 30 May 2006, jesse\(??\)~H~H\) wrote: > > I had did some modification of ipg.h and ipg.c. Please see attach file. > > 20060521-2.6.17-rc4-git-ip1000-test.patch__icplus_modify.zip > > Please send them as regular patches. Please > Documentation/SubmittingPatches and the following URL for details: > > http://www.zipworld.com.au/~akpm/linux/patches/stuff/tpp.txt > It takes people quite a few iterations to get patch preparation worked out. Pekka, if you have time, perhaps you can extract the patches for us? - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Sign-off for the IP1000A driver before inclusion
On Tue, 30 May 2006, jesse\(??\)~H~H\) wrote: > I had did some modification of ipg.h and ipg.c. Please see attach file. > 20060521-2.6.17-rc4-git-ip1000-test.patch__icplus_modify.zip Please send them as regular patches. Please Documentation/SubmittingPatches and the following URL for details: http://www.zipworld.com.au/~akpm/linux/patches/stuff/tpp.txt Thanks. Pekka - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Sign-off for the IP1000A driver before inclusion
On Wed, 24 May 2006, jesse\(??\)\) wrote: > We had read this document, but we can't find any blank at which we can sign. > Could you tell us where we can sign it off. You sign it off in the patch submission. It's enough that you agree on it and let us know a sign off line: Signed-off-by: John Doe <[EMAIL PROTECTED]> which the patch submitter (probably Francois) adds to the patch when it is being submitted. There is no actual documentation you need to sign off. Thanks. Pekka - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Sign-off for the IP1000A driver before inclusion
Dear Pekka: We had read this document, but we can't find any blank at which we can sign. Could you tell us where we can sign it off. Thanks! Jesse - Original Message - From: "Pekka Enberg" <[EMAIL PROTECTED]> To: "jesse(建興)" <[EMAIL PROTECTED]> Cc: "Francois Romieu" <[EMAIL PROTECTED]>; "David Vrabel" <[EMAIL PROTECTED]>; ; ; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]> Sent: Tuesday, May 23, 2006 3:01 PM Subject: Sign-off for the IP1000A driver before inclusion Hi Jesse, On 5/23/06, jesse\(建興\) <[EMAIL PROTECTED]> wrote: > If any problem, please feel free to contact me. [EMAIL PROTECTED] As per: http://marc.theaimsgroup.com/?l=linux-kernel&m=114666027122495 we still need someone from IC Plus to sign off on the IP1000A driver for it to be merged with mainline Linux. For more details on the Developer's Certificate of Origin, please refer to: http://www.osdl.org/newsroom/press_releases/2004/2004_05_24_dco.html or Documentation/SubmittingPatches. Pekka - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Sign-off for the IP1000A driver before inclusion
Hi Pekka: I will send this document to our product manager. After he agreed, we will sign it off. Thank you for this information. Jesse - Original Message - From: "Pekka Enberg" <[EMAIL PROTECTED]> To: "jesse(建興)" <[EMAIL PROTECTED]> Cc: "Francois Romieu" <[EMAIL PROTECTED]>; "David Vrabel" <[EMAIL PROTECTED]>; ; ; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]> Sent: Tuesday, May 23, 2006 3:01 PM Subject: Sign-off for the IP1000A driver before inclusion Hi Jesse, On 5/23/06, jesse\(建興\) <[EMAIL PROTECTED]> wrote: > If any problem, please feel free to contact me. [EMAIL PROTECTED] As per: http://marc.theaimsgroup.com/?l=linux-kernel&m=114666027122495 we still need someone from IC Plus to sign off on the IP1000A driver for it to be merged with mainline Linux. For more details on the Developer's Certificate of Origin, please refer to: http://www.osdl.org/newsroom/press_releases/2004/2004_05_24_dco.html or Documentation/SubmittingPatches. Pekka - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
