February 2011 - OpenBIOS - mail.coreboot.org

[PATCH] RFC: Change ofmem_common.c to set memory translation properties by reference
by Mark Cave-Ayland May 4, 2011

May 4, 2011

Hi folks, The attached patch changes OFMEM so that instead of allocating new space within the Forth dictionary every time the /memory and /virtual-memory available/translations nodes are updated, we simply change the property to point directly to a static buffer. This has the effect of saving substantial amounts of memory during OpenSolaris 10 boot (in fact the final dictionary size after boot is now < 256K once again). Blue/Andreas: please could you take a look at this patch and make sure it doesn't break anything in your SPARC64/PPC tests? On the plus side, with this patch applied Milax gets to the end of its natural boot without crashing giving the following output: OpenBIOS for Sparc64 Configuration device id QEMU version 1 machine id 0 kernel cmdline CPUs: 1 x SUNW,UltraSPARC-IIi UUID: 00000000-0000-0000-0000-000000000000 Welcome to OpenBIOS v1.0 built on Oct 14 2010 20:18 Type 'help' for detailed information Trying cdrom:f... Not a bootable ELF image Not a bootable a.out image Loading FCode image... Loaded 7084 bytes entry point is 0x4000 Ignoring failed claim for va 1000000 memsz bf34e! Ignoring failed claim for va 1402000 memsz 303b3! Ignoring failed claim for va 1800000 memsz 60a30! Jumping to entry point 00000000010071d8 for type 0000000000000001... switching to new context: entry point 0x10071d8 stack 0x00000000ffe06b49 warning:interpret: exception -13 caught SunOS Release 5.11 Version MilaX_0.3.2 64-bit Copyright 1983-2008 Sun Microsystems, Inc. All rights reserved. Use is subject to license terms. spacex@:interpret: exception -13 caught kdbg-words:interpret: exception -13 caught cb-r/w:interpret: exception -13 caught (Can't load tod module) EXIT -1 > Does anyone know what the tod modules does? Is it Time Of Day (i.e. we are missing some kind of hardware clock emulation?) ATB, Mark. -- Mark Cave-Ayland - Senior Technical Architect PostgreSQL - PostGIS Sirius Corporation plc - control through freedom http://www.siriusit.co.uk t: +44 870 608 0063 Sirius Labs: http://www.siriusit.co.uk/labs

4 14

Solaris 8: ESP device enumeration issue
by Mark Cave-Ayland April 4, 2011

April 4, 2011

Hi all, So with the previous patch to change the initial timer values applied to SVN trunk, we now get much further on a Solaris 8 boot: Configuration device id QEMU version 1 machine id 32 CPUs: 1 x FMI,MB86904 UUID: 00000000-0000-0000-0000-000000000000 Welcome to OpenBIOS v1.0 built on Feb 17 2011 14:56 Type 'help' for detailed information 0 > boot cdrom:d -v Not a bootable ELF image Loading a.out image... Loaded 7680 bytes entry point is 0x4000 bootpath: /iommu/sbus/espdma/esp/sd@2,0:d Jumping to entry point 00004000 for type 00000005... switching to new context: Size: 259040+54154+47486 Bytes SunOS Release 5.8 Version Generic_108528-09 32-bit Copyright 1983-2001 Sun Microsystems, Inc. All rights reserved. Ethernet address = 52:54:0:12:34:56 Using default device instance data vac: enabled in write through mode mem = 131072K (0x8000000) avail mem = 110419968 root nexus = SUNW,SPARCstation-5 iommu0 at root: obio 0x10000000 sbus0 at iommu0: obio 0x10001000 dma0 at sbus0: SBus slot 5 0x8400000 dma0 is /iommu@0,10000000/sbus@0,10001000/espdma@5,8400000 /iommu@0,10000000/sbus@0,10001000/espdma@5,8400000/esp@5,8800000 (esp0): esp-options=0x46 esp0 at dma0: SBus slot 5 0x8800000 sparc ipl 4 esp0 is /iommu@0,10000000/sbus@0,10001000/espdma@5,8400000/esp@5,8800000 qemu: fatal: Trap 0x29 while interrupts disabled, Error state pc: f004127c npc: f0041280 General Registers: %g0-7: 00000000 f02441a0 04400fc1 00007000 f5af4e40 f0243b88 00000000 f0244020 Current Register Window: %o0-7: ffff8000 00008000 00000f00 044000c0 f5948688 ffebc000 fbe3a4b8 f0041be4 %l0-7: 04400fc0 f0041c78 f0041c7c 00000001 0000010f 00000001 0000002a fbe39f78 %i0-7: ffff8000 00008000 00000f00 044000c1 00000002 ffebc000 fbe3a020 f0041be4 Floating Point Registers: %f00: 000000000.000000 000000000.000000 000000000.000000 000000000.000000 %f04: 000000000.000000 000000000.000000 000000000.000000 000000000.000000 %f08: 000000000.000000 000000000.000000 000000000.000000 000000000.000000 %f12: 000000000.000000 000000000.000000 000000000.000000 000000000.000000 %f16: 000000000.000000 000000000.000000 000000000.000000 000000000.000000 %f20: 000000000.000000 000000000.000000 000000000.000000 000000000.000000 %f24: 000000000.000000 000000000.000000 000000000.000000 000000000.000000 %f28: 000000000.000000 000000000.000000 000000000.000000 000000000.000000 psr: 04000fc0 (icc: ---- SPE: SP-) wim: 00000001 fsr: 00080000 y: 00000000 Aborted From Artyom's OBP output, we can see that the next few lines that should appear on the console look like this: sd2 at esp0: target 2 lun 0 sd2 is /iommu at 0,10000000/sbus at 0,10001000/espdma at 5,8400000/esp at 5,8800000/sd at 2,0 root on /iommu at 0,10000000/sbus at 0,10001000/espdma at 5,8400000/esp at 5,8800000/sd at 2,0:b Therefore the crash is being caused by Solaris either trying to access the esp device or doing some kind of enumeration on the ESP bus. ATB, Mark. -- Mark Cave-Ayland - Senior Technical Architect PostgreSQL - PostGIS Sirius Corporation plc - control through freedom http://www.siriusit.co.uk t: +44 870 608 0063 Sirius Labs: http://www.siriusit.co.uk/labs

3 16

[PATCH] Fix PowerPC build
by Aurelien Jarno Feb. 19, 2011

Feb. 19, 2011

Given openbios overrides <asm/types.h> with its own version, it should provides at least the same types for the included kernel headers. This fixes the following build failure: | HOSTCC host/kernel/bootstrap.o | In file included from /usr/include/asm/ptrace.h:28, | from /usr/include/asm/sigcontext.h:12, | from /usr/include/bits/sigcontext.h:29, | from /usr/include/signal.h:340, | from ../kernel/bootstrap.c:14: | /usr/include/linux/types.h:27: error: expected '=', ',', ';', 'asm' or '__attribute__' before '__le16' | /usr/include/linux/types.h:28: error: expected '=', ',', ';', 'asm' or '__attribute__' before '__be16' | /usr/include/linux/types.h:29: error: expected '=', ',', ';', 'asm' or '__attribute__' before '__le32' | /usr/include/linux/types.h:30: error: expected '=', ',', ';', 'asm' or '__attribute__' before '__be32' | /usr/include/linux/types.h:31: error: expected '=', ',', ';', 'asm' or '__attribute__' before '__le64' | /usr/include/linux/types.h:32: error: expected '=', ',', ';', 'asm' or '__attribute__' before '__be64' | /usr/include/linux/types.h:34: error: expected '=', ',', ';', 'asm' or '__attribute__' before '__sum16' | /usr/include/linux/types.h:35: error: expected '=', ',', ';', 'asm' or '__attribute__' before '__wsum' | In file included from /usr/include/asm/sigcontext.h:12, | from /usr/include/bits/sigcontext.h:29, | from /usr/include/signal.h:340, | from ../kernel/bootstrap.c:14: | /usr/include/asm/ptrace.h:201: error: expected specifier-qualifier-list before '__u32' | In file included from /usr/include/asm/sigcontext.h:12, | from /usr/include/bits/sigcontext.h:29, | from /usr/include/signal.h:340, | from ../kernel/bootstrap.c:14: | /usr/include/asm/ptrace.h:223: error: expected specifier-qualifier-list before '__u32' | make[1]: *** [host/kernel/bootstrap.o] Error 1 Signed-off-by: Aurelien Jarno <aurelien(a)aurel32.net> --- a/include/arch/ppc/types.h +++ b/include/arch/ppc/types.h @@ -75,4 +75,10 @@ typedef short s16; typedef int s32; typedef long long s64; +/* Needed by kernel includes */ +typedef u8 __u8; +typedef u16 __u16; +typedef u32 __u32; +typedef u64 __u64; + #endif -- Aurelien Jarno GPG: 1024D/F1BCDB73 aurelien(a)aurel32.net http://www.aurel32.net

3 5

openfirmware boot partition layout
by Dinar Valeev Feb. 18, 2011

Feb. 18, 2011

Hi, I'm playing with kvm on powerpc (I have LPAR'ed IBM Power5 machine). I have a couple of disk images which are working perfectly. (I've tested it by deploying it using VIO server with usual dd command directly to LPAR disk) But when I'm trying to attach same image as qemu (with KVM enabled) disk. VM won't boot. It just stuck on openbios banner. I have another image with old mac partition table (which is working under qemu) So I have actually two questions: 1) Is booting from PReP partition (type 0x41) is supported? (PReP just consists raw copy of yaboot binary) 2) When VM have LVM root then it required to have FAT partition (type 0x06) with kernel, initrd, yaboot and its config. Does this scenario supported as well? Here is usual partition setups I have: non lvm setup: Number Start End Size Type File system Flags 1 32.3kB 8225kB 8193kB primary boot, , , , , , , , , type=41, prep, 2 8225kB 897MB 888MB primary ext3 , , , , , , , , , type=83, , lvm setup: Number Start End Size Type File system Flags 1 32.3kB 49.4MB 49.3MB primary fat16 boot, , , , , , , , , type=06, , 2 49.4MB 979MB 929MB primary , , , , , lvm, , , , type=8e, , I'll glad to provide additional information or participate in testing in the future. Thanks in advance, Dinar

4 7

[commit] r1033 - trunk/openbios-devel/drivers
by repository service Feb. 18, 2011

Feb. 18, 2011

Author: mcayland Date: Fri Feb 18 21:26:57 2011 New Revision: 1033 URL: http://tracker.coreboot.org/trac/openbios/changeset/1033 Log: Fix level 14 and level 10 timers under Solaris 8 on SPARC32. During boot, if the level 14 timer limit is not set then the Solaris kernel panics with a division by zero error. Fix this by setting it to the default value as documented in obio.h. Also if the level 10 timer is not set with a limit of 0, the Solaris kernel constantly outputs "spurious interrupt at processor level 10" messages during boot. Signed-off-by: Mark Cave-Ayland <mark.cave-ayland(a)siriusit.co.uk> Modified: trunk/openbios-devel/drivers/obio.c Modified: trunk/openbios-devel/drivers/obio.c ============================================================================== --- trunk/openbios-devel/drivers/obio.c Thu Feb 17 21:00:49 2011 (r1032) +++ trunk/openbios-devel/drivers/obio.c Fri Feb 18 21:26:57 2011 (r1033) @@ -296,8 +296,8 @@ counter_regs = (struct sun4m_timer_regs *)ofmem_map_io(base + (uint64_t)offset, sizeof(*counter_regs)); counter_regs->cfg = 0xffffffff; - counter_regs->l10_timer_limit = (((1000000/100) + 1) << 10); - counter_regs->cpu_timers[0].l14_timer_limit = 0; + counter_regs->l10_timer_limit = 0; + counter_regs->cpu_timers[0].l14_timer_limit = 0x9c4000; /* see comment in obio.h */ counter_regs->cpu_timers[0].cntrl = 1; for (i = 0; i < SUN4M_NCPU; i++) {

1 0

[PATCH] Fix level 14 and level 10 timers under Solaris 8 on SPARC32.
by mark.cave-ayland＠siriusit.co.uk Feb. 18, 2011

Feb. 18, 2011

From: Mark Cave-Ayland <mark.cave-ayland(a)siriusit.co.uk> During boot, if the level 14 timer limit is not set then the Solaris kernel panics with a division by zero error. Fix this by setting it to the default value as documented in obio.h. Also if the level 10 timer is not set with a limit of 0, the Solaris kernel constantly outputs "spurious interrupt at processor level 10" messages during boot. Signed-off-by: Mark Cave-Ayland <mark.cave-ayland(a)siriusit.co.uk> --- openbios-devel/drivers/obio.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/openbios-devel/drivers/obio.c b/openbios-devel/drivers/obio.c index 2153c44..3adf9ac 100644 --- a/openbios-devel/drivers/obio.c +++ b/openbios-devel/drivers/obio.c @@ -296,8 +296,8 @@ ob_counter_init(uint64_t base, unsigned long offset) counter_regs = (struct sun4m_timer_regs *)ofmem_map_io(base + (uint64_t)offset, sizeof(*counter_regs)); counter_regs->cfg = 0xffffffff; - counter_regs->l10_timer_limit = (((1000000/100) + 1) << 10); - counter_regs->cpu_timers[0].l14_timer_limit = 0; + counter_regs->l10_timer_limit = 0; + counter_regs->cpu_timers[0].l14_timer_limit = 0x9c4000; /* see comment in obio.h */ counter_regs->cpu_timers[0].cntrl = 1; for (i = 0; i < SUN4M_NCPU; i++) { -- 1.7.2.3

3 2

[PATCH] ESP : Fix SCSI READ command length
by Olivier DANET Feb. 17, 2011

Feb. 17, 2011

Signed-off-by: Olivier Danet <odanet(a)caramail.com> --- diff -rup a/drivers/esp.c b/drivers/esp.c --- a/drivers/esp.c 2011-01-27 15:11:52.000000000 +0100 +++ b/drivers/esp.c 2011-01-27 15:13:34.000000000 +0100 @@ -147,7 +147,7 @@ ob_sd_read_sector(esp_private_t *esp, sd sd->id, offset); // Setup command = Read(10) - memset(esp->buffer, 0, 10); + memset(esp->buffer, 0, 11); esp->buffer[0] = 0x80; esp->buffer[1] = READ_10; @@ -159,7 +159,7 @@ ob_sd_read_sector(esp_private_t *esp, sd esp->buffer[8] = 0; esp->buffer[9] = 1; - if (do_command(esp, sd, 10, sd->bs)) + if (do_command(esp, sd, 11, sd->bs)) return 0; return 0;

4 5

[commit] r1032 - trunk/openbios-devel/drivers
by repository service Feb. 17, 2011

Feb. 17, 2011

Author: blueswirl Date: Thu Feb 17 21:00:49 2011 New Revision: 1032 URL: http://tracker.coreboot.org/trac/openbios/changeset/1032 Log: ESP : Fix SCSI READ command length Signed-off-by: Olivier Danet <odanet(a)caramail.com> Signed-off-by: Blue Swirl <blauwirbel(a)gmail.com> Modified: trunk/openbios-devel/drivers/esp.c Modified: trunk/openbios-devel/drivers/esp.c ============================================================================== --- trunk/openbios-devel/drivers/esp.c Sun Feb 13 23:30:14 2011 (r1031) +++ trunk/openbios-devel/drivers/esp.c Thu Feb 17 21:00:49 2011 (r1032) @@ -147,7 +147,7 @@ sd->id, offset); // Setup command = Read(10) - memset(esp->buffer, 0, 10); + memset(esp->buffer, 0, 11); esp->buffer[0] = 0x80; esp->buffer[1] = READ_10; @@ -159,7 +159,7 @@ esp->buffer[8] = 0; esp->buffer[9] = 1; - if (do_command(esp, sd, 10, sd->bs)) + if (do_command(esp, sd, 11, sd->bs)) return 0; return 0;

1 0

State of AIX
by Grieve, Shane Feb. 14, 2011

Feb. 14, 2011

I have been following the AIX booting issue for a while now and was wondering if there is any progress. I have access to AIX hardware and I am happy to help test. Thanks Shane

5 11

More work on Solaris 8 SPARC32 crash
by Mark Cave-Ayland Feb. 14, 2011

Feb. 14, 2011

Hi all, I've spent a bit more time today trying to figure out why this crashes under OpenBIOS and so decided to start working backwards through the -d in_asm QEMU output to try and figure out why we seem to get stuck in an infinite loop. Anyhow the short version is that I've traced the series of functions that get called just before we get stuck in the infinite loop and it looks like this: 0xf0044384: cbe_set_level 0xf0063954: splr 0xf0044268: cbe_enable 0xf0062cf0: __div64 ... invokes "ta 2" trap instruction 0xf004127c: sys_trap() 0xf0064658: vpanic () 0xf0064618: panic_trigger () 0xf005bf90: trap ... infinite loop ... The interesting part here is that the fatal trap sequence is being invoked from __div64 which makes me think it is some kind of divide by zero error. I'm not exactly sure what the cbe_* functions are doing, but a quick browse of the OpenSolaris source shows that it's related to PIL/RTC things. My current feeling is that maybe OpenBIOS isn't doing something right with PIL state somewhere, or we're reading back a zero value from either the clock (or a missing OpenBIOS property) which is causing us to die in the division in cbe_enable(). And another thing, what does cbe actually stand for? ATB, Mark. -- Mark Cave-Ayland - Senior Technical Architect PostgreSQL - PostGIS Sirius Corporation plc - control through freedom http://www.siriusit.co.uk t: +44 870 608 0063 Sirius Labs: http://www.siriusit.co.uk/labs

3 8