2.6.0-test1 + nvidia 4363 driver

2.6.0-test1 + nvidia 4363 driver

[Martin Zwickel]

[-- Attachment #1.1: Type: text/plain, Size: 332 bytes --]


Hi there!

Anybody got a working patch for nvidia 4363 to let it work with the 2.6.0-test1
kernel?
The 2.5 nvidia patch doesn't work for the 2.6 kernel.

Regards,
Martin

-- 
MyExcuse:
Mail server hit by UniSpammer.

Martin Zwickel <martin.zwickel@technotrend.de>
Research & Development

TechnoTrend AG <http://www.technotrend.de>

[-- Attachment #1.2: 00000008.mimetmp --]
[-- Type: application/pgp-signature, Size: 190 bytes --]

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: 2.6.0-test1 + nvidia 4363 driver

[Ian Hastie]

[-- Attachment #1: Type: text/plain, Size: 374 bytes --]

On Monday 14 Jul 2003 15:20, Martin Zwickel wrote:
> Hi there!
>
> Anybody got a working patch for nvidia 4363 to let it work with the
> 2.6.0-test1 kernel?
> The 2.5 nvidia patch doesn't work for the 2.6 kernel.

Just a quick hack to the official unofficial patch!  It's working OK for me 
right now.  You'll need to apply the 2.5 patch then this one afterwards.

-- 
Ian.

[-- Attachment #2: NVIDIA_kernel-1.0-4363-2.5-2.6.diff --]
[-- Type: text/x-diff, Size: 3815 bytes --]

--- NVIDIA_kernel-1.0-4363/nv-linux.h.old	2003-07-14 15:39:35.000000000 +0100
+++ NVIDIA_kernel-1.0-4363/nv-linux.h	2003-07-14 15:47:54.000000000 +0100
@@ -32,6 +32,8 @@
 #  define KERNEL_2_4
 #elif LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 0)
 #  define KERNEL_2_5
+#elif LINUX_VERSION_CODE < KERNEL_VERSION(2, 7, 0)
+#  define KERNEL_2_6
 #else
 #  error This driver does not support development kernels!
 #endif
@@ -46,7 +48,7 @@
 #define __SMP__
 #endif
 
-#if defined (MODVERSIONS) && !defined (KERNEL_2_5)
+#if defined (MODVERSIONS) && !(defined (KERNEL_2_5) || defined(KERNEL_2_6))
 #include <linux/modversions.h>
 #endif
 
@@ -62,7 +64,7 @@
 #include <linux/poll.h>             /* poll_wait                        */
 #include <linux/delay.h>            /* mdelay, udelay                   */
 
-#ifdef KERNEL_2_5
+#if defined(KERNEL_2_5) || defined(KERNEL_2_6)
 #include <linux/sched.h>            /* suser(), capable() replacement   */
 #include <linux/moduleparam.h>      /* module_param()                   */
 #include <linux/smp_lock.h>         /* kernel_locked                    */
@@ -126,7 +128,7 @@
 #define PUT_MODULE_SYMBOL(sym)        inter_module_put((char *) sym)
 #define NV_VMA_PRIVATE(vma)           ((vma)->vm_private_data)
 
-#ifdef KERNEL_2_5
+#if defined(KERNEL_2_5) || defined(KERNEL_2_6)
 #  define NV_DEVICE_NUMBER(_minor)      ((kdev_val(_minor)) & 0x0f)
 #  define NV_IS_CONTROL_DEVICE(_minor)  (((kdev_val(_minor)) & 0xff) == 0xff)
 #  define NV_IS_SUSER()                 capable(CAP_SYS_ADMIN)
@@ -150,7 +152,7 @@
 #  define NV_MODULE_PARAMETER(x)        MODULE_PARM(x, "i")
 #endif
 
-#ifndef KERNEL_2_5
+#if !defined(KERNEL_2_5) && !defined(KERNEL_2_6)
   typedef void irqreturn_t;
 # define IRQ_NONE
 # define IRQ_RETVAL(x)
@@ -163,7 +165,7 @@
   typedef void* devfs_handle_t;
 #endif
 
-#ifdef KERNEL_2_5
+#if defined(KERNEL_2_5) || defined(KERNEL_2_6)
 #define NV_DEVFS_REGISTER(_name, _minor)                            \
 ({                                                                  \
     devfs_handle_t __handle = NULL;                                 \
@@ -193,7 +195,7 @@
  * relevant releases to date use it. This version was backported to 2.4 by
  * RedHat without means to identify the change, hence this hack.
  */
-#ifdef KERNEL_2_5
+#if defined(KERNEL_2_5) || defined(KERNEL_2_6)
 #define NV_REMAP_PAGE_RANGE(a, b...)    remap_page_range(vma, a, ## b)
 #else
 #if defined(REMAP_PAGE_RANGE_5)
--- NVIDIA_kernel-1.0-4363/nv.c.old	2003-07-14 15:51:39.000000000 +0100
+++ NVIDIA_kernel-1.0-4363/nv.c	2003-07-14 15:49:48.000000000 +0100
@@ -720,7 +720,7 @@
 
     nv_printf(NV_DBG_ERRORS, "nvidia: loading %s\n", pNVRM_ID);
 
-#if defined(CONFIG_DEVFS_FS) && !defined(KERNEL_2_5)
+#if defined(CONFIG_DEVFS_FS) && !(defined(KERNEL_2_5) || defined(KERNEL_2_6))
     rc = devfs_register_chrdev(nv_major, "nvidia", &nv_fops);
 #else
     rc = register_chrdev(nv_major, "nvidia", &nv_fops);
@@ -806,7 +806,7 @@
     return 0;
 
  failed:
-#if defined(CONFIG_DEVFS_FS) && !defined(KERNEL_2_5)
+#if defined(CONFIG_DEVFS_FS) && !(defined(KERNEL_2_5) || defined(KERNEL_2_6))
     devfs_unregister_chrdev(nv_major, "nvidia");
 #else
     unregister_chrdev(nv_major, "nvidia");
@@ -856,7 +856,7 @@
         }
     }
 
-#if defined(CONFIG_DEVFS_FS) && !defined(KERNEL_2_5)
+#if defined(CONFIG_DEVFS_FS) && !(defined(KERNEL_2_5) || defined(KERNEL_2_6))
     rc = devfs_unregister_chrdev(nv_major, "nvidia");
 #else
     rc = unregister_chrdev(nv_major, "nvidia");
@@ -1461,7 +1461,7 @@
 
     switch (_IOC_NR(cmd))
     {
-#if !defined(KERNEL_2_5)
+#if !defined(KERNEL_2_5) && !defined(KERNEL_2_6)
         /* debug tool; zap the module use count so we can unload driver */
         /*             even if it is confused */
         case _IOC_NR(NV_IOCTL_MODULE_RESET):

Re: 2.6.0-test1 + nvidia 4363 driver

[Joshua Schmidlkofer]

http://www.minion.de posted updated patches.  They work pretty find for
me.


js


On Mon, 2003-07-14 at 07:20, Martin Zwickel wrote:
> Hi there!
> 
> Anybody got a working patch for nvidia 4363 to let it work with the 2.6.0-test1
> kernel?
> The 2.5 nvidia patch doesn't work for the 2.6 kernel.
> 
> Regards,
> Martin

Re: 2.6.0-test1 + nvidia 4363 driver

[Fernando Sanchez]

hello Joshua,
is there any mirror for the patch? I can't get to the site you pointed. 
Thanks,

Fernando

Joshua Schmidlkofer wrote:
> http://www.minion.de posted updated patches.  They work pretty find for
> me.
> 
> 
> js
> 
> 
> On Mon, 2003-07-14 at 07:20, Martin Zwickel wrote:
> 
>>Hi there!
>>
>>Anybody got a working patch for nvidia 4363 to let it work with the 2.6.0-test1
>>kernel?
>>The 2.5 nvidia patch doesn't work for the 2.6 kernel.
>>
>>Regards,
>>Martin
> 
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
> 
> 
> 


-- 


Fernando Sanchez
Dpto. Sistemas USFQ

Re: 2.6.0-test1 + nvidia 4363 driver

[oford]

[-- Attachment #1: Type: text/plain, Size: 485 bytes --]

On Wed, 2003-07-16 at 17:41, Fernando Sanchez wrote:
> hello Joshua,
> is there any mirror for the patch? I can't get to the site you pointed. 
> Thanks,
> 
> Fernando
> 
> Joshua Schmidlkofer wrote:
> > http://www.minion.de posted updated patches.  They work pretty find for
> > me.

Several gentoo folks were having trouble reaching that site as well.

So here is a mirror:

http://www.arghblech.com/NVIDIA_FreeBSD-1.0-3203-5.0.diff

-- 
Owen Ford <oford@ev1.net>

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: 2.6.0-test1 + nvidia 4363 driver

[oford]

[-- Attachment #1: Type: text/plain, Size: 684 bytes --]

On Wed, 2003-07-16 at 19:31, oford wrote:
> On Wed, 2003-07-16 at 17:41, Fernando Sanchez wrote:
> > hello Joshua,
> > is there any mirror for the patch? I can't get to the site you pointed. 
> > Thanks,
> > 
> > Fernando
> > 
> > Joshua Schmidlkofer wrote:
> > > http://www.minion.de posted updated patches.  They work pretty find for
> > > me.
> 
> Several gentoo folks were having trouble reaching that site as well.
> 
> So here is a mirror:
> 
> http://www.arghblech.com/NVIDIA_FreeBSD-1.0-3203-5.0.diff

And I'll include the correct link for free this time! *sigh*

http://www.arghblech.com/NVIDIA_kernel-1.0-4363-2.5.diff

-- 
Owen Ford <oford@ev1.net>

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: 2.6.0-test1 + nvidia 4363 driver

[Fernando Sanchez]

oford wrote:
> On Wed, 2003-07-16 at 19:31, oford wrote:
> 
>>On Wed, 2003-07-16 at 17:41, Fernando Sanchez wrote:
>>
>>>hello Joshua,
>>>is there any mirror for the patch? I can't get to the site you pointed. 
>>>Thanks,
>>>
>>>Fernando
>>>
>>>Joshua Schmidlkofer wrote:
>>>
>>>>http://www.minion.de posted updated patches.  They work pretty find for
>>>>me.
>>
>>Several gentoo folks were having trouble reaching that site as well.
>>
>>So here is a mirror:
>>
>>http://www.arghblech.com/NVIDIA_FreeBSD-1.0-3203-5.0.diff
> 
> 
> And I'll include the correct link for free this time! *sigh*

I'm glad the first time is for free :-)


> 
> http://www.arghblech.com/NVIDIA_kernel-1.0-4363-2.5.diff
> 


-- 


Fernando Sanchez
Dpto. Sistemas USFQ

Re: 2.6.0-test1 + nvidia 4363 driver

[Martin Schlemmer]

On Thu, 2003-07-17 at 02:31, oford wrote:
> On Wed, 2003-07-16 at 17:41, Fernando Sanchez wrote:
> > hello Joshua,
> > is there any mirror for the patch? I can't get to the site you pointed. 
> > Thanks,
> > 
> > Fernando
> > 
> > Joshua Schmidlkofer wrote:
> > > http://www.minion.de posted updated patches.  They work pretty find for
> > > me.
> 
> Several gentoo folks were having trouble reaching that site as well.
> 
> So here is a mirror:
> 
> http://www.arghblech.com/NVIDIA_FreeBSD-1.0-3203-5.0.diff

Ok, but why do the Gentoo folk hassle ?  I already updated
nvidia-kernel-1.0.4363-r3 to work fine with 2.6.0-test1 on
the 14th ...


Cheers,

-- 
Martin Schlemmer

Re: 2.6.0-test1 + nvidia 4363 driver

[Ian Hastie]

On Monday 14 Jul 2003 20:34, Anich Gregor wrote:
> >
> > Just a quick hack to the official unofficial patch!  It's working OK for
> > me right now.  You'll need to apply the 2.5 patch then this one
> > afterwards.
>
> Seems like some people have problems applying your patch.

Strange you should say that as yours is the only reply I've seen.  Is there 
somewhere else that someone is saying they're having problems?

> I have put up a small howto how to compile the NVIDIA_kernel
> with linux-2.6 at
> http://deltaanime.ath.cx/~blight/nvidia_kernel_with_2.6.html The only thing
> what breaks it is that the version is no longer below 2, 6, 0 which the 2.5
> patch uses to check wether it's 2.5 or not ;)

Yes, I know that works, but I felt like extending the definitions instead of 
making it treat 2.5 and 2.6 as if they're identical.  OK, they probably will 
be now, but I've never likes taking things for granted.

-- 
Ian.