Re: Linux 2.6.36-rc3

[Yinghai Lu <yinghai@kernel.org>]

On 08/29/2010 12:24 PM, Torsten Kaiser wrote:
> On Sun, Aug 29, 2010 at 7:25 PM, Linus Torvalds
> <torvalds@linux-foundation.org> wrote:
>> You all know the drill by now: another week, another -rc.
> 
> I'm still seeing the regression from -rc1 that the start of the early
> serial output only contains garbage.
> 
> I'm blaming commit 8fee13a48e4879fba57725f6d9513df4bfa8e9f3 (x86,
> setup: enable early console output from the decompressor), because if
> I'm disabling the new serial_putchar() in
> arch/x86/boot/compressed/misc.c by putting "return;" in its first
> line, the output returns to normal.
> 
> Here the different outputs from the serial console:
> Start of output from 2.6.35:
> [    0.000000] Linux version 2.6.35 (root@treogen) (gcc version 4.4.4
> (Gentoo 4.4.4-r1 p1.0, pie-0.4.5) ) #1 SMP Mon Aug 2 19:54:57 CEST
> 2010
> [    0.000000] Command line: fastboot earlyprintk=serial,ttyS0,115200
> console=ttyS0,115200 console=tty1 crypt_root=/dev/md3 radeon.modeset=1
> video=1280x1024
> [    0.000000] BIOS-provided physical RAM map:
> [    0.000000]  BIOS-e820: 0000000000000000 - 000000000009fc00 (usable)
> [    0.000000]  BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved)
> [    0.000000]  BIOS-e820: 00000000000e4000 - 0000000000100000 (reserved)
> 
> Start of output from 2.6.36-rc1..3:
> {binary garbage}00000000 - 000000000009fc00 (usable)
> [    0.000000]  BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved)
> [    0.000000]  BIOS-e820: 00000000000e4000 - 0000000000100000 (reserved)
> 
> Start of output with disabled serial_putchar():
> [    0.000000] Linux version 2.6.36-rc3 (root@treogen) (gcc version
> 4.4.4 (Gentoo 4.4.4-r1 p1.0, pie-0.4.5) ) #1 SMP Sun Aug 29 19:03:00
> CEST 2010
> [    0.000000] Command line: fastboot earlyprintk=serial,ttyS0,115200
> console=ttyS0,115200 console=tty1 crypt_root=/dev/md3 radeon.modeset=1
> video=1280x1024
> [    0.000000] BIOS-provided physical RAM map:
> [    0.000000]  BIOS-e820: 0000000000000000 - 000000000009fc00 (usable)
> [    0.000000]  BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved)
> [    0.000000]  BIOS-e820: 00000000000e4000 - 0000000000100000 (reserved)
> 
> I can't see anything wrong with this commit, but it looks to me, as if
> the newly added decompressor output gets corrupted somewhere. Could it
> be my nondefault baud rate at 115200?

this one should fix the problem.

please check

[PATCH] x86, setup: fix earlyprint=serial,ttyS0...

Torsten reported that there is garbage output,
after commit 8fee13a48e4879fba57725f6d9513df4bfa8e9f3 (x86,
setup: enable early console output from the decompressor)

it turns out we missed the offset for that case.

Reported-by: Torsten Kaiser <just.for.lkml@googlemail.com>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>

Index: linux-2.6/arch/x86/boot/early_serial_console.c
===================================================================
--- linux-2.6.orig/arch/x86/boot/early_serial_console.c
+++ linux-2.6/arch/x86/boot/early_serial_console.c
@@ -58,7 +58,7 @@ static void parse_earlyprintk(void)
 		if (arg[pos] == ',')
 			pos++;
 
-		if (!strncmp(arg, "ttyS", 4)) {
+		if (!strncmp(arg + pos, "ttyS", 4)) {
 			static const int bases[] = { 0x3f8, 0x2f8 };
 			int idx = 0;