Linux 2.4.20-rc2

Linux 2.4.20-rc2

[Marcelo Tosatti]

So here goes -rc2, fixing the lcall DoS.



Summary of changes from v2.4.20-rc1 to v2.4.20-rc2
============================================

<cel@citi.umich.edu>:
  o sock_writable not appropriate for TCP sockets

<hch@sgi.com>:
  o fix file system corruption under load

<jgarzik@redhat.com>:
  o Use dev_kfree_skb_any not dev_kfree_skb in tg3 net driver function tg3_free_rings.

<marcelo@freak.distro.conectiva>:
  o Undo latest hid-input fixes: they are broken
  o Reverse order of BK config checkout entries
  o Changed EXTRAVERSION to -rc2

<mkp@mkp.net>:
  o Update credits

<rth@are.twiddle.net>:
  o Fix carry ripple in 3 and 4 word addition and subtraction macros

<tytso@think.thunk.org>:
  o HTREE backwards compatibility patch

Alan Cox <alan@lxorguk.ukuu.org.uk>:
  o Enable the merged AMD pm driver

Andries E. Brouwer <Andries.Brouwer@cwi.nl>:
  o [TCP] Do not update rcv_nxt until ts_recent is updated

Ben Collins <bcollins@debian.org>:
  o [TG3]: TG3_HW_STATUS_SIZE should be 0x50 not 0x80

c-d.hailfinger.kernel.2002-q4@gmx.net <c-d.hailfinger.kernel.2002-Q4@gmx.net>:
  o restore framebuffer console after suspend

David S. Miller <davem@nuts.ninka.net>:
  o [SPARC64]: Translate SO_{SND,RCV}TIMEO socket options
  o [SPARC64]: Handle kernel integer divide by zero properly
  o [SPARC64]: Check DRM_NEW not DRM in ioctl32.c
  o [SPARC64]: Fix accidental clobbering of register on cheetahplus

David S. Miller <davem@redhat.com>:
  o Fix tg3 net driver to properly disable interrupts during some TX operations

Edward Peng <edward_peng@dlink.com.tw>:
  o sundance net driver updates

Joshua Uziel <uzi@uzix.org>:
  o [SPARC64]: 0x22/0x10 is Ultra-I/spitfire

Pete Zaitcev <zaitcev@redhat.com>:
  o [sparc] Fix off-by-one in s/g handling

Petr Vandrovec <VANDROVE@vc.cvut.cz>:
  o Fix ncpfs file creation issue

Petr Vandrovec <vandrove@vc.cvut.cz>:
  o Fix lcall DoS

r.e.wolff@bitwizard.nl <R.E.Wolff@BitWizard.nl>:
  o Fix SX driver detection

Tom Rini <trini@kernel.crashing.org>:
  o Fix a thinko in arch/ppc/kernel/ppc_ksyms.c

Trond Myklebust <trond.myklebust@fys.uio.no>:
  o another kmap imbalance in 2.4.x/2.5.x RPC

Vojtech Pavlik <vojtech@suse.cz>:
  o Add vt8235 support

Re: Linux 2.4.20-rc2 in2000.c and t128.c build fixes (resend)

[Andreas Steinmetz]

[-- Attachment #1: Type: text/plain, Size: 954 bytes --]

Marcelo,
it seems the gcc 3.2 build fixes for in2000.c and t128.c got lost.

Please see:

http://marc.theaimsgroup.com/?l=linux-kernel&m=103641735125372&w=2
http://marc.theaimsgroup.com/?l=linux-kernel&m=103641959127368&w=2
http://marc.theaimsgroup.com/?l=linux-kernel&m=103641876226674&w=2

Example error without the patch:

gcc -D__KERNEL__ -I/usr/src/linux-2.4.20rc2/include -Wall 
-Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common 
-fomit-frame-pointer -pipe -mpreferred-stack-boundary=2 -mcpu=i686 
-march=i686 -falign-loops -falign-jumps -falign-functions   -nostdinc 
-iwithprefix include -DKBUILD_BASENAME=in2000  -c -o in2000.o in2000.c
in2000.c:1919: base_tab causes a section type conflict
in2000.c:1926: int_tab causes a section type conflict
make[3]: *** [in2000.o] Error 1

The attached and combined patch is modified to reflect Alan's comment as 
referenced above.
-- 
Andreas Steinmetz
D.O.M. Datenverarbeitung GmbH

[-- Attachment #2: rc2-fixes.patch --]
[-- Type: text/plain, Size: 923 bytes --]

diff -rNu old/drivers/scsi/in2000.c linux/drivers/scsi/in2000.c
--- old/drivers/scsi/in2000.c	2001-09-30 21:26:07.000000000 +0200
+++ linux/drivers/scsi/in2000.c	2002-11-04 14:18:58.000000000 +0100
@@ -1916,14 +1916,14 @@
    0
    };
 
-static const unsigned short base_tab[] in2000__INITDATA = {
+static unsigned short base_tab[] in2000__INITDATA = {
    0x220,
    0x200,
    0x110,
    0x100,
    };
 
-static const int int_tab[] in2000__INITDATA = {
+static int int_tab[] in2000__INITDATA = {
    15,
    14,
    11,
diff -rNu old/drivers/scsi/t128.c linux/drivers/scsi/t128.c
--- old/drivers/scsi/t128.c	2001-12-21 18:41:55.000000000 +0100
+++ kubux/drivers/scsi/t128.c	2002-11-04 14:21:47.000000000 +0100
@@ -142,7 +142,7 @@
 
 #define NO_BASES (sizeof (bases) / sizeof (struct base))
 
-static const struct signature {
+static struct signature {
 	const char *string;
 	int offset;
 } signatures[] __initdata = {
 

Re: Linux 2.4.20-rc2

[Adrian Bunk]

I got the following error at the final linking of 2.4.20-rc2:

<--   snip  -->

...
        -o vmlinux
fs/fs.o(.text+0x53bc3): In function `presto_free_cache':
: undefined reference to `presto_dentry_slab'
make: *** [vmlinux] Error 1

<--  snip  -->


K S Sreeram <sreeram@tachyontech.net> proposed the following patch two
weeks ago:

--- a/fs/intermezzo/dcache.c	Mon Oct 21 10:56:57 2002
+++ b/fs/intermezzo/dcache.c	Mon Oct 21 10:56:57 2002
@@ -48,7 +48,7 @@
 
 #include <linux/intermezzo_fs.h>
 
-static kmem_cache_t * presto_dentry_slab;
+kmem_cache_t * presto_dentry_slab;
 
 /* called when a cache lookup succeeds */
 static int presto_d_revalidate(struct dentry *de, int flag)



cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: Linux 2.4.20-rc2

[Keith Owens]

On Fri, 15 Nov 2002 13:10:07 -0200 (BRST), 
Marcelo Tosatti <marcelo@conectiva.com.br> wrote:
>So here goes -rc2, fixing the lcall DoS.

No obvious changes that affect kdb, so unless somebody reports
problems, use kdb-v2.5-2.4.20-rc1 with -rc2.

Re: Linux 2.4.20-rc2

[Robert Read]

This patch is obviously correct and has been applied to our
tree.  It would be great it was applied to 2.4.20-rc2.

robert

* Adrian Bunk (bunk@fs.tum.de) [021115 15:19]:
> I got the following error at the final linking of 2.4.20-rc2:
> 
> <--   snip  -->
> 
> ...
>         -o vmlinux
> fs/fs.o(.text+0x53bc3): In function `presto_free_cache':
> : undefined reference to `presto_dentry_slab'
> make: *** [vmlinux] Error 1
> 
> <--  snip  -->
> 
> 
> K S Sreeram <sreeram@tachyontech.net> proposed the following patch two
> weeks ago:
> 
> --- a/fs/intermezzo/dcache.c	Mon Oct 21 10:56:57 2002
> +++ b/fs/intermezzo/dcache.c	Mon Oct 21 10:56:57 2002
> @@ -48,7 +48,7 @@
>  
>  #include <linux/intermezzo_fs.h>
>  
> -static kmem_cache_t * presto_dentry_slab;
> +kmem_cache_t * presto_dentry_slab;
>  
>  /* called when a cache lookup succeeds */
>  static int presto_d_revalidate(struct dentry *de, int flag)
> 
> 
> 
> cu
> Adrian
> 
> -- 
> 
>        "Is there not promise of rain?" Ling Tan asked suddenly out
>         of the darkness. There had been need of rain for many days.
>        "Only a promise," Lao Er said.
>                                        Pearl S. Buck - Dragon Seed
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by: To learn the basics of securing 
> your web site with SSL, click here to get a FREE TRIAL of a Thawte 
> Server Certificate: http://www.gothawte.com/rd524.html
> _______________________________________________
> intermezzo-discuss mailing list
> intermezzo-discuss@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/intermezzo-discuss

Re: Linux 2.4.20-rc2

[Adrian Bunk]

Hi Marcelo,

the patch below is still needed to fix a .text.exit error.

Please apply
Adrian


--- linux-2.4.19-full-nohotplug/drivers/scsi/ips.c.old	2002-10-04 18:49:10.000000000 +0200
+++ linux-2.4.19-full-nohotplug/drivers/scsi/ips.c	2002-10-04 18:50:02.000000000 +0200
@@ -305,21 +305,21 @@
        name:		ips_hot_plug_name,
        id_table:	ips_pci_table,
        probe:		ips_insert_device,
-       remove:		ips_remove_device,
+       remove:		__devexit_p(ips_remove_device),
    }; 
            
    struct pci_driver ips_pci_driver_5i = {
        name:		ips_hot_plug_name,
        id_table:	ips_pci_table_5i,
        probe:		ips_insert_device,
-       remove:		ips_remove_device,
+       remove:		__devexit_p(ips_remove_device),
    };
 
    struct pci_driver ips_pci_driver_i960 = {
        name:		ips_hot_plug_name,
        id_table:	ips_pci_table_i960,
        probe:		ips_insert_device,
-       remove:		ips_remove_device,
+       remove:		__devexit_p(ips_remove_device),
    };
 
 #endif

Linux 2.4.20-rc2 screwy ac97_codec.c:codec_id()

[Paul]

	Hi;

	Im pretty sure this is broken, but I dont know exactly
what it is trying to do.
	The first snprintf is overwritten regardless-- missing
else block? And its format string should probably be "%4X:%4X",
because whats there wont fit in the buffer.
	Then the first 3 chars in the string are filled in
with raw numbers (For my card, non-ascii) and then a single
decimal digit?? (This string is printed out during boot time--
which is how I noticed it because of the 'garbage' chars.)
	I dont know what a PnP string is supposed to look like...

Paul
set@pobox.com

--- linux-2.4.19/drivers/sound/ac97_codec.c     2002-08-03 00:39:44.000000000 +0 000
+++ linux-2.4.20/drivers/sound/ac97_codec.c     2002-11-15 14:56:52.000000000 +0 000
@@ -654,6 +654,27 @@
 }
  
/**
+ *     codec_id        -  Turn id1/id2 into a PnP string
+ *     @id1: Vendor ID1
+ *     @id2: Vendor ID2
+ *     @buf: 10 byte buffer
+ *
+ *     Fills buf with a zero terminated PnP ident string for the id1/id2
+ *     pair. For convenience the return is the passed in buffer pointer.
+ */
+ 
+static char *codec_id(u16 id1, u16 id2, char *buf)
+{
+       if(id1&0x8080)
+               snprintf(buf, 10, "%0x4X:%0x4X", id1, id2);
+       buf[0] = (id1 >> 8);
+       buf[1] = (id1 & 0xFF);
+       buf[2] = (id2 >> 8);
+       snprintf(buf+3, 7, "%d", id2&0xFF);
+       return buf;
+}

Re: Linux 2.4.20-rc2 screwy ac97_codec.c:codec_id()

[Alan Cox]

On Fri, 2002-11-22 at 01:14, Paul wrote:
> 	Im pretty sure this is broken, but I dont know exactly
> what it is trying to do.
> 	The first snprintf is overwritten regardless-- missing
> else block? And its format string should probably be "%4X:%4X",
> because whats there wont fit in the buffer.
> 	Then the first 3 chars in the string are filled in
> with raw numbers (For my card, non-ascii) and then a single
> decimal digit?? (This string is printed out during boot time--
> which is how I noticed it because of the 'garbage' chars.)
> 	I dont know what a PnP string is supposed to look like...

There is an else missing you are correct


> +       if(id1&0x8080)
> +               snprintf(buf, 10, "%0x4X:%0x4X", id1, id2);

else 
{

> +       buf[0] = (id1 >> 8);
> +       buf[1] = (id1 & 0xFF);
> +       buf[2] = (id2 >> 8);
> +       snprintf(buf+3, 7, "%d", id2&0xFF);

}

> +       return buf;
> +}
> 
> 
> -
> 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/

Re: Linux 2.4.20-rc2 screwy ac97_codec.c:codec_id()

[Paul]

[-- Attachment #1: Type: text/plain, Size: 780 bytes --]

Alan Cox <alan@lxorguk.ukuu.org.uk>, on Fri Nov 22, 2002 [01:57:06 AM] said:
> On Fri, 2002-11-22 at 01:14, Paul wrote:
> > 	Im pretty sure this is broken, but I dont know exactly
> > what it is trying to do.
> > 	The first snprintf is overwritten regardless-- missing
> > else block? And its format string should probably be "%4X:%4X",
> > because whats there wont fit in the buffer.
> 
> There is an else missing you are correct
> 
	Hi;
	
	Well, just for the heck of it, here is a patch that
puts the else block in, and fixes the format so that it matches
previous output and fits in the buffer. If the powers that be
wanted "1234:ABCD" instead of "0x1234:0xabcd", then the
format string would be "%04X:%04X" and CODEC_ID_BUFSZ could
go back to 10.
	Tested.

Paul
set@pobox.com

[-- Attachment #2: 2.4.20-rc2-ac97_codec.patch --]
[-- Type: text/plain, Size: 1799 bytes --]

--- 2.4.20-rc2/drivers/sound/ac97_codec.c.orig	2002-11-21 13:54:36.000000000 -0500
+++ 2.4.20-rc2/drivers/sound/ac97_codec.c	2002-11-21 22:49:18.000000000 -0500
@@ -52,6 +52,8 @@
 #include <linux/ac97_codec.h>
 #include <asm/uaccess.h>
 
+#define CODEC_ID_BUFSZ 14
+
 static int ac97_read_mixer(struct ac97_codec *codec, int oss_channel);
 static void ac97_write_mixer(struct ac97_codec *codec, int oss_channel, 
 			     unsigned int left, unsigned int right);
@@ -657,7 +659,7 @@
  *	codec_id	-  Turn id1/id2 into a PnP string
  *	@id1: Vendor ID1
  *	@id2: Vendor ID2
- *	@buf: 10 byte buffer
+ *	@buf: CODEC_ID_BUFSZ byte buffer
  *
  *	Fills buf with a zero terminated PnP ident string for the id1/id2
  *	pair. For convenience the return is the passed in buffer pointer.
@@ -665,12 +667,14 @@
  
 static char *codec_id(u16 id1, u16 id2, char *buf)
 {
-	if(id1&0x8080)
-		snprintf(buf, 10, "%0x4X:%0x4X", id1, id2);
-	buf[0] = (id1 >> 8);
-	buf[1] = (id1 & 0xFF);
-	buf[2] = (id2 >> 8);
-	snprintf(buf+3, 7, "%d", id2&0xFF);
+	if(id1&0x8080) {
+		snprintf(buf, CODEC_ID_BUFSZ, "0x%04x:0x%04x", id1, id2);
+	} else {
+		buf[0] = (id1 >> 8);
+		buf[1] = (id1 & 0xFF);
+		buf[2] = (id2 >> 8);
+		snprintf(buf+3, CODEC_ID_BUFSZ - 3, "%d", id2&0xFF);
+	}
 	return buf;
 }
  
@@ -702,7 +706,7 @@
 	u16 id1, id2;
 	u16 audio, modem;
 	int i;
-	char cidbuf[10];
+	char cidbuf[CODEC_ID_BUFSZ];
 
 	/* probing AC97 codec, AC97 2.0 says that bit 15 of register 0x00 (reset) should 
 	 * be read zero.
@@ -746,7 +750,7 @@
 	}
 	if (codec->name == NULL)
 		codec->name = "Unknown";
-	printk(KERN_INFO "ac97_codec: AC97 %s codec, id: %s(%s)\n", 
+	printk(KERN_INFO "ac97_codec: AC97 %s codec, id: %s (%s)\n", 
 		modem ? "Modem" : (audio ? "Audio" : ""),
 	       codec_id(id1, id2, cidbuf), codec->name);