5 messages, 4 authors, 2007-03-21 · open the first message on its own page
From: Valentine Barshak <hidden>
Date: 2007-03-20 19:29:58
Add PPC440EPx/440GRx EMAC along with Marvell 88E1111 and ET1011C PHY support. The M88E1111 chip can be found on Sequoia board. Signed-off-by: Valentine Barshak <redacted>
From: Sergei Shtylyov <hidden>
Date: 2007-03-20 19:35:34
Hello. Valentine Barshak wrote:
Add PPC440EPx/440GRx EMAC along with Marvell 88E1111 and ET1011C PHY support. The M88E1111 chip can be found on Sequoia board.
Signed-off-by: Valentine Barshak <redacted>
To netdev@vger.kernel.org, please.
WBR, Sergei
From: Olof Johansson <hidden>
Date: 2007-03-21 15:50:31
On Tue, Mar 20, 2007 at 10:29:12PM +0300, Valentine Barshak wrote:
diff -ruN linux.orig/drivers/net/ibm_emac/ibm_emac_core.c linux/drivers/net/ibm_emac/ibm_emac_core.c--- linux.orig/drivers/net/ibm_emac/ibm_emac_core.c 2007-03-16 18:03:51.000000000 +0300 +++ linux/drivers/net/ibm_emac/ibm_emac_core.c 2007-03-18 18:53:08.000000000 +0300@@ -86,7 +86,8 @@ static u32 busy_phy_map; #if defined(CONFIG_IBM_EMAC_PHY_RX_CLK_FIX) && \ - (defined(CONFIG_405EP) || defined(CONFIG_440EP) || defined(CONFIG_440GR)) + (defined(CONFIG_405EP) || defined(CONFIG_440EP) || defined(CONFIG_440GR) || \ + defined(CONFIG_440EPX) || defined(CONFIG_440GRX))
This is getting out of control. Please add a config option for this, and add "select" statements to the Kconfig file where any of the above CONFIG_4.* are defined.
/* 405EP has "EMAC to PHY Control Register" (CPC0_EPCTL) which can help us * with PHY RX clock problem. * 440EP/440GR has more sane SDR0_MFR register implementation than 440GX, which diff -ruN linux.orig/drivers/net/ibm_emac/ibm_emac.h linux/drivers/net/ibm_emac/ibm_emac.h--- linux.orig/drivers/net/ibm_emac/ibm_emac.h 2007-03-16 18:03:51.000000000 +0300 +++ linux/drivers/net/ibm_emac/ibm_emac.h 2007-03-18 18:53:08.000000000 +0300@@ -26,7 +26,7 @@ #if !defined(CONFIG_405GP) && !defined(CONFIG_405GPR) && !defined(CONFIG_405EP) && \ !defined(CONFIG_440GP) && !defined(CONFIG_440GX) && !defined(CONFIG_440SP) && \ !defined(CONFIG_440EP) && !defined(CONFIG_NP405H) && !defined(CONFIG_440SPE) && \ - !defined(CONFIG_440GR) + !defined(CONFIG_440GR) && !defined(CONFIG_440EPX) && !defined(CONFIG_440GRX)
Same here. I know you only added a couple more, but it's reached critical mass (well, it did a while ago). -Olof
From: Eugene Surovegin <hidden>
Date: 2007-03-21 16:28:50
On Wed, Mar 21, 2007 at 11:04:38AM -0500, Olof Johansson wrote:
On Tue, Mar 20, 2007 at 10:29:12PM +0300, Valentine Barshak wrote:quoted
diff -ruN linux.orig/drivers/net/ibm_emac/ibm_emac.h linux/drivers/net/ibm_emac/ibm_emac.h--- linux.orig/drivers/net/ibm_emac/ibm_emac.h 2007-03-16 18:03:51.000000000 +0300 +++ linux/drivers/net/ibm_emac/ibm_emac.h 2007-03-18 18:53:08.000000000 +0300@@ -26,7 +26,7 @@ #if !defined(CONFIG_405GP) && !defined(CONFIG_405GPR) && !defined(CONFIG_405EP) && \ !defined(CONFIG_440GP) && !defined(CONFIG_440GX) && !defined(CONFIG_440SP) && \ !defined(CONFIG_440EP) && !defined(CONFIG_NP405H) && !defined(CONFIG_440SPE) && \ - !defined(CONFIG_440GR) + !defined(CONFIG_440GR) && !defined(CONFIG_440EPX) && !defined(CONFIG_440GRX)Same here. I know you only added a couple more, but it's reached critical mass (well, it did a while ago).
No, it's not same here. Please, take a time and look at the full code this particular snippet has nothing to do with Konfig stuff. One can argue that it can be removed altogether, but I added this ugly check for a reason - people were adding support for new 4xx SoCs blindly in the past without checking that EMAC registers weren't changed slightly. So, Valentine, keep this particular ugly ifdef as it is. Also, Valentine, next time CC me with any EMAC changes because I'm not actively reading PPC maillists these days. -- Eugene
From: Valentine Barshak <hidden>
Date: 2007-03-21 18:14:41
Eugene Surovegin wrote:
On Wed, Mar 21, 2007 at 11:04:38AM -0500, Olof Johansson wrote:quoted
On Tue, Mar 20, 2007 at 10:29:12PM +0300, Valentine Barshak wrote:quoted
diff -ruN linux.orig/drivers/net/ibm_emac/ibm_emac.h linux/drivers/net/ibm_emac/ibm_emac.h--- linux.orig/drivers/net/ibm_emac/ibm_emac.h 2007-03-16 18:03:51.000000000 +0300 +++ linux/drivers/net/ibm_emac/ibm_emac.h 2007-03-18 18:53:08.000000000 +0300@@ -26,7 +26,7 @@ #if !defined(CONFIG_405GP) && !defined(CONFIG_405GPR) && !defined(CONFIG_405EP) && \ !defined(CONFIG_440GP) && !defined(CONFIG_440GX) && !defined(CONFIG_440SP) && \ !defined(CONFIG_440EP) && !defined(CONFIG_NP405H) && !defined(CONFIG_440SPE) && \ - !defined(CONFIG_440GR) + !defined(CONFIG_440GR) && !defined(CONFIG_440EPX) && !defined(CONFIG_440GRX)Same here. I know you only added a couple more, but it's reached critical mass (well, it did a while ago).No, it's not same here. Please, take a time and look at the full code this particular snippet has nothing to do with Konfig stuff. One can argue that it can be removed altogether, but I added this ugly check for a reason - people were adding support for new 4xx SoCs blindly in the past without checking that EMAC registers weren't changed slightly. So, Valentine, keep this particular ugly ifdef as it is. Also, Valentine, next time CC me with any EMAC changes because I'm not actively reading PPC maillists these days.
Please, take a look at the updated 440EPx EMAC patch. I've added IBM_EMAC_CLK_SEL option which is selected if both IBM_EMAC_PHY_RX_CLK_FIX is set and the processor is not 440GX. The other "ugly" ifdef is left unchanged. Thanks.