July 2012 - SeaBIOS - mail.coreboot.org

Re: [SeaBIOS] [PATCH] geodevga: make use of readl and writel for mmio access
by Kevin O'Connor Aug. 27, 2012

Aug. 27, 2012

On Tue, Jul 17, 2012 at 12:08:59PM +0200, Christian Gmeiner wrote: > This commit just simplifies some code. > > Signed-off-by: Christian Gmeiner <christian.gmeiner(a)gmail.com> > --- > vgasrc/geodevga.c | 48 +++++++++--------------------------------------- > 1 file changed, 9 insertions(+), 39 deletions(-) > > diff --git a/vgasrc/geodevga.c b/vgasrc/geodevga.c > index 5c6caf0..4e3d912 100644 > --- a/vgasrc/geodevga.c > +++ b/vgasrc/geodevga.c [...] > /* Set mmio registers > * there may be some timing issues here, the reads seem > * to slow things down enough work reliably > */ > > - reg = geode_memRead(vp+VP_MISC); > + reg = readl(vp+VP_MISC); The code is run in 16bit "big real" mode. This means all of memory is available, but one still needs to be careful about segment usage. SeaBIOS always calls the option roms with a %ss == %ds == 0, but I don't think it is wise to rely upon that. So, it's better to use GET_FARVAR(0, *(vp+VP_MISC)). -Kevin

2 1

Re: [SeaBIOS] [BUG] BSOD on Win2003 Server when 64bit PCI resource is present
by Kevin O'Connor Aug. 6, 2012

Aug. 6, 2012

On Thu, Jul 26, 2012 at 03:38:47PM +0000, Alexey Korolev wrote: > HI, > > Current version of Seabios is causing blue screen on Windows2003 when 64bit PCI resource is present and occupies high memory. > > BSOD Error code is: 0x000000A5 (0x02, 0xFFFFfADF6A446880, 0x1, 0xFFFFFADFAA34690) > > The issue is localized, it is related to presence of 64bit resource in _CRS method. > > If we disable a 64bit region from _CRS the Win2003 load normally but this doesn't allow Windows to use 64bit resources. > > At the moment I have no idea how to fix this. Please help! Unfortunately, it's very difficult to debug acpi issues on Windows. Gerd's been on vacation this week - so, lets give him a chance to look at it when he gets back. If it can't be resolved, we'll need to revert the patch that broke Win2003. > P/S > Yet another issue is related to debug messages. > If I add any DBUG call in acpi-dst it will cause BSOD on win2003 (win2008 works fine) I've seen this as well (on WinXP - I don't have Win2003). Unfortunately, I haven't been able to find out why the debug macro fails on these older versions of Windows. -Kevin

4 3

[SeaBIOS PATCH 0/2] Allow non-contiguous APIC IDs (v3)
by Eduardo Habkost Aug. 1, 2012

Aug. 1, 2012

Changes v2 -> v3: - Report I/O APIC ID = 0 on MP-table, too Changes v1 -> v2: - Patch 1/2: cosmetic whitespace change - Patch 2/2: use size suffixes on asm instructions on smp.c - New patch descriptions Eduardo Habkost (2): report real I/O APIC ID (0) on MADT and MP-table (v3) allow CPUs to have non-contiguous Local APIC IDs (v2) src/acpi-dsdt.dsl | 4 +++- src/acpi.c | 11 ++++++----- src/config.h | 1 + src/mptable.c | 4 ++-- src/smp.c | 17 +++++++++++++++++ src/util.h | 1 + 6 files changed, 30 insertions(+), 8 deletions(-) -- 1.7.10.4

4 7

[PATCH 0/6] Build PCI hotplug SSDT from a single template
by Paolo Bonzini Aug. 1, 2012

Aug. 1, 2012

More than 1kb of data is taken by the 32 copies of the PCI hotplug SSDT methods. We can build them from a single template like we do for CPUs. This series does exactly this. Patches 1 and 2 prepare for the change, by moving other pieces of ssdt-pcihd.dsl out of the way. Patch 3 is also a simple rename and patch 4 fixes a bug in acpi_extract. Patches 5 and 6 finally do the movement. Paolo Bonzini (6): acpi: move PCNT to DSDT acpi: move s3/s4/s5 to build_ssdt acpi: rename Processor SSDT constants acpi_extract: fix off-by-one acpi_extract: detect DeviceOp acpi: build PCI hotplug devices from a single template Makefile | 2 +- src/acpi-dsdt.dsl | 39 +++++++++++- src/acpi.c | 161 ++++++++++++++++++++++++++----------------------- src/ssdt-pcihp.dsl | 124 ++++--------------------------------- src/ssdt-susp.dsl | 41 +++++++++++++ tools/acpi_extract.py | 30 ++++++++- 6 files changed, 206 insertions(+), 191 deletions(-) create mode 100644 src/ssdt-susp.dsl -- 1.7.10.4

3 12

[PATCH] EHCI: increase CONNECT detection delay
by Sven Schnelle Aug. 1, 2012

Aug. 1, 2012

1ms seems to short on my chromebook, increasing that to 5ms makes EHCI work. Signed-off-by: Sven Schnelle <svens(a)stackframe.org> --- src/usb-ehci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/usb-ehci.c b/src/usb-ehci.c index 3c0be13..b53caa1 100644 --- a/src/usb-ehci.c +++ b/src/usb-ehci.c @@ -82,7 +82,7 @@ ehci_hub_detect(struct usbhub_s *hub, u32 port) writel(portreg, portsc); msleep(EHCI_TIME_POSTPOWER); } else { - msleep(1); // XXX - time for connect to be detected. + msleep(5); // XXX - time for connect to be detected. } portsc = readl(portreg); -- 1.7.10.4

2 2

[PATCH v2] Add an IPMI SMBIOS entry
by minyard＠acm.org July 31, 2012

July 31, 2012

From: Corey Minyard <cminyard(a)mvista.com> An IPMI device is being added to the qemu code, and it has an SMBIOS entry to describe the interface characteristics. So add the SMBIOS entry to the BIOS so it can handle this. Signed-off-by: Corey Minyard <cminyard(a)mvista.com> --- Makefile | 2 +- src/ipmi.c | 31 +++++++++++++++++++++++++++++++ src/ipmi.h | 32 ++++++++++++++++++++++++++++++++ src/post.c | 2 ++ src/smbios.c | 55 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ src/smbios.h | 12 ++++++++++++ src/util.h | 5 +++++ 7 files changed, 138 insertions(+), 1 deletions(-) create mode 100644 src/ipmi.c create mode 100644 src/ipmi.h diff --git a/Makefile b/Makefile index dfdec5c..1b6eb96 100644 --- a/Makefile +++ b/Makefile @@ -13,7 +13,7 @@ SRCBOTH=misc.c stacks.c pmm.c output.c util.c block.c floppy.c ata.c mouse.c \ pnpbios.c pirtable.c vgahooks.c ramdisk.c pcibios.c blockcmd.c \ usb.c usb-uhci.c usb-ohci.c usb-ehci.c usb-hid.c usb-msc.c \ virtio-ring.c virtio-pci.c virtio-blk.c virtio-scsi.c apm.c ahci.c \ - usb-uas.c lsi-scsi.c + usb-uas.c lsi-scsi.c ipmi.c SRC16=$(SRCBOTH) system.c disk.c font.c SRC32FLAT=$(SRCBOTH) post.c shadow.c memmap.c coreboot.c boot.c \ acpi.c smm.c mptable.c smbios.c pciinit.c optionroms.c mtrr.c \ diff --git a/src/ipmi.c b/src/ipmi.c new file mode 100644 index 0000000..bedcdc0 --- /dev/null +++ b/src/ipmi.c @@ -0,0 +1,31 @@ +// IPMI setup information +// +// Copyright (C) 2012 Corey Minyard <cminyard(a)mvista.com> +// +// This file may be distributed under the terms of the GNU LGPLv3 license. + +#include "ipmi.h" +#include "util.h" + +struct ipmi_info ipmi_info VAR16VISIBLE; + +void +ipmi_setup(void) +{ + u8 *data; + int size; + + data = romfile_loadfile("etc/ipmi", &size); + if (!data || (size < sizeof(ipmi_info))) { + ipmi_info.interface = 0; /* Disable */ + return; + } + + if (size > sizeof(ipmi_info)) + size = sizeof(ipmi_info); + + memcpy(&ipmi_info, data, size); + free(data); + + ipmi_info.base_addr = le64_to_cpu(ipmi_info.base_addr); +} diff --git a/src/ipmi.h b/src/ipmi.h new file mode 100644 index 0000000..85fdf42 --- /dev/null +++ b/src/ipmi.h @@ -0,0 +1,32 @@ +// IPMI setup information +// +// Copyright (C) 2012 Corey Minyard <cminyard(a)mvista.com> +// +// This file may be distributed under the terms of the GNU LGPLv3 license. + +#ifndef __IPMI_H +#define __IPMI_H + +#include "config.h" // CONFIG_COREBOOT +#include "types.h" + +#define IPMI_MEM_SPACE 0 +#define IPMI_IO_SPACE 1 + +struct ipmi_info { + u8 str_version; /* Version of this structure */ + u8 interface; + u8 reg_space; + u8 reg_spacing; + u8 slave_addr; + u8 irq; + u8 version; + u8 reserved1; + u64 base_addr; +} PACKED; + +extern struct ipmi_info ipmi_info; + +void ipmi_setup(void); + +#endif diff --git a/src/post.c b/src/post.c index 0f31b4c..69834b6 100644 --- a/src/post.c +++ b/src/post.c @@ -28,6 +28,7 @@ #include "virtio-blk.h" // virtio_blk_setup #include "virtio-scsi.h" // virtio_scsi_setup #include "lsi-scsi.h" // lsi_scsi_setup +#include "ipmi.h" // ipmi_setup /**************************************************************** @@ -255,6 +256,7 @@ maininit(void) pnp_setup(); kbd_setup(); mouse_setup(); + ipmi_setup(); init_bios_tables(); // Run vga option rom diff --git a/src/smbios.c b/src/smbios.c index fc84aad..f480a8d 100644 --- a/src/smbios.c +++ b/src/smbios.c @@ -8,6 +8,7 @@ #include "util.h" // dprintf #include "paravirt.h" // qemu_cfg_smbios_load_field #include "smbios.h" // struct smbios_entry_point +#include "ipmi.h" // struct ipmi_info struct smbios_entry_point *SMBiosAddr; @@ -422,6 +423,58 @@ smbios_init_type_32(void *start) return start+2; } +/* Type 38 -- System Boot Information */ +static void * +smbios_init_type_38(void *start, struct ipmi_info *info) +{ + struct smbios_type_38 *p = (struct smbios_type_38 *)start; + + p->header.type = 38; + p->header.length = sizeof(struct smbios_type_38); + p->header.handle = 0x2100; + + p->interface_type = info->interface; + + p->base_addr = info->base_addr & ~1ULL; + /* Bit 0 goes into a special place */ + p->base_addr_mod_and_irq_info = (info->base_addr & 1) << 4; + + switch (info->reg_spacing) { + case 4: + p->base_addr_mod_and_irq_info |= (1 << 6); + break; + case 16: + p->base_addr_mod_and_irq_info |= (2 << 6); + break; + case 1: + default: + /* zero is the right value */ + break; + } + + if (info->reg_space) + p->base_addr |= 1; /* I/O space */ + + if (info->version) + p->ipmi_version = info->version; + else + p->ipmi_version = 0x15; + + if (info->slave_addr) + p->i2c_slave_addr = info->slave_addr; + else + p->i2c_slave_addr = 0x20; + + p->nv_storage_dev_addr = 0; + + p->interrupt_number = info->irq; + + start += sizeof(struct smbios_type_38); + *((u16 *)start) = 0; + + return start+2; +} + /* Type 127 -- End of Table */ static void * smbios_init_type_127(void *start) @@ -510,6 +563,8 @@ smbios_init(void) } add_struct(32, p); + if (ipmi_info.interface) + add_struct(38, p, &ipmi_info); /* Add any remaining provided entries before the end marker */ for (i = 0; i < 256; i++) qemu_cfg_smbios_load_external(i, &p, &nr_structs, &max_struct_size, diff --git a/src/smbios.h b/src/smbios.h index 9d54e80..1935335 100644 --- a/src/smbios.h +++ b/src/smbios.h @@ -160,6 +160,18 @@ struct smbios_type_32 { u8 boot_status; } PACKED; +/* SMBIOS type 38 - IPMI Information */ +struct smbios_type_38 { + struct smbios_structure_header header; + u8 interface_type; + u8 ipmi_version; + u8 i2c_slave_addr; + u8 nv_storage_dev_addr; + u64 base_addr; + u8 base_addr_mod_and_irq_info; + u8 interrupt_number; +} PACKED; + /* SMBIOS type 127 -- End-of-table */ struct smbios_type_127 { struct smbios_structure_header header; diff --git a/src/util.h b/src/util.h index ef8ec7c..415f518 100644 --- a/src/util.h +++ b/src/util.h @@ -134,6 +134,11 @@ static inline u32 le32_to_cpu(u32 x) return x; } +static inline u64 le64_to_cpu(u64 x) +{ + return x; +} + static inline u32 getesp(void) { u32 esp; asm("movl %%esp, %0" : "=rm"(esp)); -- 1.7.4.1

3 2

Re: [SeaBIOS] Compiling SeaBIOS for coreboot has problems with its ACPI code
by Moore, Robert July 31, 2012

July 31, 2012

Yes, for whatever reason, it is definitely an issue with the newer version of bison. None of the code below is output with bison 2.4.2: #if YYDEBUG extern int AslCompilerdebug; #endif #ifdef YYPARSE_PARAM #if defined __STDC__ || defined __cplusplus int AslCompilerparse (void *YYPARSE_PARAM); #else int AslCompilerparse (); #endif #else /* ! YYPARSE_PARAM */ #if defined __STDC__ || defined __cplusplus int AslCompilerparse (void); #else int AslCompilerparse (); #endif #endif /* ! YYPARSE_PARAM */ > -----Original Message----- > From: Idwer Vollering [mailto:vidwer@gmail.com] > Sent: Tuesday, July 31, 2012 2:20 PM > To: Moore, Robert > Cc: Michael S. Tsirkin; Kevin O'Connor; seabios(a)seabios.org; Tang, Feng > Subject: Re: [SeaBIOS] Compiling SeaBIOS for coreboot has problems with > its ACPI code > > 2012/7/31 Moore, Robert <robert.moore(a)intel.com>: > > And, could you please send the bison output file "aslcompilerparse.h" > > > > Thanks, > > Bob > > Attached.

1 0

Re: [SeaBIOS] Compiling SeaBIOS for coreboot has problems with its ACPI code
by Idwer Vollering July 31, 2012

July 31, 2012

2012/7/31 Moore, Robert <robert.moore(a)intel.com>: > And, could you please send the bison output file "aslcompilerparse.h" > > Thanks, > Bob Attached.

1 0

Re: [SeaBIOS] Compiling SeaBIOS for coreboot has problems with its ACPI code
by Moore, Robert July 31, 2012

July 31, 2012

And, could you please send the bison output file "aslcompilerparse.h" Thanks, Bob > -----Original Message----- > From: Moore, Robert > Sent: Tuesday, July 31, 2012 1:37 PM > To: 'Idwer Vollering' > Cc: Michael S. Tsirkin; Kevin O'Connor; seabios(a)seabios.org; Tang, Feng > Subject: RE: [SeaBIOS] Compiling SeaBIOS for coreboot has problems with > its ACPI code > > This looks like it may be a Bison issue, since the file > "aslcompilerparse.h" is in fact an intermediate file created by Bison. > With Bison version 2.4.1, there is no declaration of "AslCompilerparse" > within the aslcompilerparse.h file. > > Could you try an earlier version of Bison? > > > > > In file included from aslanalyze.c:118:0: > aslcompilerparse.h:738:5: error: redundant redeclaration of > 'AslCompilerparse' [-Werror=redundant-decls] > > In file included from aslanalyze.c:117:0: > aslcompiler.h:163:1: note: previous declaration of 'AslCompilerparse' > was > > > > > -----Original Message----- > > From: Idwer Vollering [mailto:vidwer@gmail.com] > > Sent: Tuesday, July 31, 2012 1:27 PM > > To: Moore, Robert > > Cc: Michael S. Tsirkin; Kevin O'Connor; seabios(a)seabios.org; Tang, > > Feng > > Subject: Re: [SeaBIOS] Compiling SeaBIOS for coreboot has problems > > with its ACPI code > > > > 2012/7/31 Moore, Robert <robert.moore(a)intel.com>: > > > On Debian, we are using > > > > > > Flex 2.5.35 > > > Bison 2.4.1 > > > > My build machine runs 64-bit archlinux and has gcc 4.7.1, flex 2.5.36 > > and bison 2.6 installed. > > > > P.S. please use reply-to-all, so everyone is informed.

1 0

Re: [SeaBIOS] Compiling SeaBIOS for coreboot has problems with its ACPI code
by Moore, Robert July 31, 2012

July 31, 2012

This looks like it may be a Bison issue, since the file "aslcompilerparse.h" is in fact an intermediate file created by Bison. With Bison version 2.4.1, there is no declaration of "AslCompilerparse" within the aslcompilerparse.h file. Could you try an earlier version of Bison? In file included from aslanalyze.c:118:0: aslcompilerparse.h:738:5: error: redundant redeclaration of 'AslCompilerparse' [-Werror=redundant-decls] In file included from aslanalyze.c:117:0: aslcompiler.h:163:1: note: previous declaration of 'AslCompilerparse' was > -----Original Message----- > From: Idwer Vollering [mailto:vidwer@gmail.com] > Sent: Tuesday, July 31, 2012 1:27 PM > To: Moore, Robert > Cc: Michael S. Tsirkin; Kevin O'Connor; seabios(a)seabios.org; Tang, Feng > Subject: Re: [SeaBIOS] Compiling SeaBIOS for coreboot has problems with > its ACPI code > > 2012/7/31 Moore, Robert <robert.moore(a)intel.com>: > > On Debian, we are using > > > > Flex 2.5.35 > > Bison 2.4.1 > > My build machine runs 64-bit archlinux and has gcc 4.7.1, flex 2.5.36 > and bison 2.6 installed. > > P.S. please use reply-to-all, so everyone is informed.

1 0