hi ralf, Thanks for the valuable input, I checked the TLB Entry and found Value in EntryLo0/1 seems absurd. Can you suggest me some way about how to approach this problem? ________________________________________________________________________ ___ xdr_partial_copy_from_skb: kaddr: ffffd000, ppage: 83f751a0, *ppage: 8117ff90 Data bus error, epc == 80161a90, ra == 80161924 ENTRYHI: 0xFFFFC000, ENTRYLO0:0x00000001, ENTRYLO1:0x001FFF5F INDEX: 0x00000009 Oops in traps.c::do_be, line 465: $0 : 00000000 10008400 ffffd000 00000000 ffffd000 83f730a0 00000978 00000000 $8 : 622f2123 732f6e69 0a230a68 74532023 20747261 ffffffff 00000010 756e694c $16: 83f730a0 ffffd000 00000998 000020a4 0000006c 00000000 ffffd000 811e3cc0 $24: 00000018 801e95ff 811e2000 811e3bc0 00000000 80161924 Hi : 00000000 Lo : 00000001 epc : 80161a90 Not tainted Status: 10008403 Cause : 0000001c Process ksoftirqd_CPU0 (pid: 3, stackpage=811e2000) Stack: 0000003c 0000003e 80036a40 80036ae4 00000998 00000998 811e3cc0 00000a04 8010813c 80036ae4 0000004c 80155f58 10008401 ffffc000 0000000a 0000003c 0000004e 80155f58 80037048 10008401 00000000 83f751a0 0000006c 00000998 811e3cc0 83f751a0 00001000 80212ff0 83fcc0b4 811e3cc0 80155f58 80155fa0 83fcd080 80212ff0 ffffd000 00001000 00000000 83f751a0 ffffd000 00001000 ... Call Trace: [<80036a40>] [<80036ae4>] [<8010813c>] [<80036ae4>] [<80155f58>] [<80155f58>] [<80037048>] [<80155f58>] [<80155fa0>] [<80160794>] [<8011d68c>] [<801560a8>] [<80185c7f>] [<80156288>] [<8015627c>] [<80140748>] [<80140738>] [<80156164>] [<80140d0c>] [<80140d00>] [<80036e50>] [<80181ff8>] [<80106420>] [<8011d898>] [<80140b04>] [<80114414>] [<8011d68c>] [<8011ec50>] [<8011d8bc>] [<8011d3a0>] [<8011d37c>] [<8011d68c>] [<8011db0c>] [<8011d318>] [<80114414>] [<8011d8bc>] [<8011d628>] [<8011d604>] [<8011d8bc>] [<8010b9c4>] ... Code: 8cac0010 8caf0014 ac880000 <ac890004> 8ca80018 8ca9001c 24a50020 2484 0020 ac8affe8 ________________________________________________________________________ ___ -----Original Message----- From: Ralf Baechle [mailto:ralf@xxxxxxxxxxxxxx] Sent: Friday, February 18, 2005 2:07 AM To: Rishabh Kumar Goel Cc: linux-mips@xxxxxxxxxxxxxx; mel@xxxxxxxxx; jbglaw@xxxxxxxxxx Subject: Re: HIGHMEM Query On Thu, Feb 17, 2005 at 10:40:22AM +0530, Rishabh@xxxxxxxxxxxx wrote: First, fix your mailer setup, it's littering emails with extra carriage returns. > I have been working on HIGHMEM fixation for 2.4.20 linux MVL kernel for > MIPS architecture. > I am getting DATA BUS ERROR(ADDR: 0xffffd000) exception when loading > (copying the data received from ethernet to the kernel space in FIXADDR > SPACE) "/sbin/init" from TARGET ROOT DIR (through NFS). I figured out > that this is because there is no corresponding page table entry for the > same. That does not make sense. If you don't have a pagetable entry the CPU will take a TLB reload exception. If the entry is invalid, the kernel will die with an oops. A bus error has different causes. Ralf The information contained in this e-mail message and in any annexure is confidential to the recipient and may contain privileged information. If you are not the intended recipient, please notify the sender and delete the message along with any annexure. You should not disclose, copy or otherwise use the information contained in the message or any annexure. Any views expressed in this e-mail are those of the individual sender except where the sender specifically states them to be the views of SoCrates Software India Pvt Ltd., Bangalore.
