grsecurity forums

Hello,

I have problem with IBM ServeRAID and grsecurity (same as http://forums.grsecurity.net/viewtopic.php?p=5601&)
I tried 2.6.14 and 2.6.17 with the same effect, clean sources working fine but when I apply grsecurity patch (even when none of its options aren't enabled) and recompile, new kernel does not recognize raid controller anymore (kernel panic ).

Unfortunately there is no error messages indicating problem (neither during compilation or boot)

I will be very grateful for any advices.

Best Regards
POSTER

POSTER wrote:I have problem with IBM ServeRAID and grsecurity (same as http://forums.grsecurity.net/viewtopic.php?p=5601&)
I tried 2.6.14 and 2.6.17 with the same effect, clean sources working fine but when I apply grsecurity patch (even when none of its options aren't enabled) and recompile, new kernel does not recognize raid controller anymore (kernel panic ).

i'd at least need the same information that was mentioned in that other thread (to compare boot logs).

Without further information i would suspect:

diff -urNp linux-2.6.17.11/arch/i386/Kconfig linux-2.6.17.11/arch/i386/Kconfig
--- linux-2.6.17.11/arch/i386/Kconfig 2006-08-07 00:18:54.000000000 -0400
+++ linux-2.6.17.11/arch/i386/Kconfig 2006-09-01 16:20:28.000000000 -0400
@@ -948,7 +948,7 @@ config PCI
choice
prompt "PCI access mode"
depends on PCI && !X86_VISWS
- default PCI_GOANY
+ default PCI_GODIRECT
---help---
On PCI systems, the BIOS can be used to detect the PCI devices and
determine their configuration. However, some old PCI motherboards
@@ -980,7 +980,7 @@ endchoice

config PCI_BIOS
bool
- depends on !X86_VISWS && PCI && (PCI_GOBIOS || PCI_GOANY)
+ depends on !X86_VISWS && PCI && PCI_GOBIOS
default y

config PCI_DIRECT


But post the logs

Dr_Napalm wrote:Without further information i would suspect:

diff -urNp linux-2.6.17.11/arch/i386/Kconfig linux-2.6.17.11/arch/i386/Kconfig
--- linux-2.6.17.11/arch/i386/Kconfig 2006-08-07 00:18:54.000000000 -0400
+++ linux-2.6.17.11/arch/i386/Kconfig 2006-09-01 16:20:28.000000000 -0400

this chunk is new to 2.6.17 so it can't affect 2.6.14 where the problem occurs as well.

Hi,

Sorry to refresh this post but it seems that the problem came back with the 2.6.24.5 flavor.

I didn't had the problem with 2.6.19.2.

Again, no problem when using vanilla 2.6.24.5 kernel.

Does someone knows what done in the previous version to solve the problem?

Server is an IBM x345 using ServeRAID 5i.

Thanks for your help.

secumap wrote:Does someone knows what done in the previous version to solve the problem?

Server is an IBM x345 using ServeRAID 5i.

nothing in particular was done to fix this problem since we didn't find out the cause...