July 2013 - SeaBIOS - mail.coreboot.org

[PATCH] tools/kconfig/{m,n}conf.c: Fix typo »This interface let*s*«
by Paul Menzel Oct. 3, 2013

Oct. 3, 2013

Date: Mon, 4 Mar 2013 12:56:56 +0100 When Kconfig was introduced in SeaBIOS with commit a4c5daf0e2545361fadcad3015e10f8b23e53926 Author: Kevin O'Connor <kevin(a)koconnor.net> Date: Mon Jan 24 22:13:58 2011 -0500 Initial commit of Kconfig build tool. this typo was copied from Kconfig in Linux v2.6.38-rc2. In the meantime Linux commit »kconfig: nconf: rewrite help texts« (0b616500) [1] rewrote the texts for `nconf.c` – which might be ported too to SeaBIOS – and the typo was fixed there. But it is still present in `mconf.c` as of Linux 3.8. [1] http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=0… --- tools/kconfig/mconf.c | 2 +- tools/kconfig/nconf.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/kconfig/mconf.c b/tools/kconfig/mconf.c index d433c7a..2a24c42 100644 --- a/tools/kconfig/mconf.c +++ b/tools/kconfig/mconf.c @@ -25,7 +25,7 @@ static const char mconf_readme[] = N_( "Overview\n" "--------\n" -"This interface let you select features and parameters for the build.\n" +"This interface lets you select features and parameters for the build.\n" "Features can either be built-in, modularized, or ignored. Parameters\n" "must be entered in as decimal or hexadecimal numbers or text.\n" "\n" diff --git a/tools/kconfig/nconf.c b/tools/kconfig/nconf.c index db56377..bb797c9 100644 --- a/tools/kconfig/nconf.c +++ b/tools/kconfig/nconf.c @@ -15,7 +15,7 @@ static const char nconf_readme[] = N_( "Overview\n" "--------\n" -"This interface let you select features and parameters for the build.\n" +"This interface lets you select features and parameters for the build.\n" "Features can either be built-in, modularized, or ignored. Parameters\n" "must be entered in as decimal or hexadecimal numbers or text.\n" "\n" -- 1.7.10.4

2 2

[PATCH v2 0/5] seabios: load acpi tables from qemu
by Michael S. Tsirkin Sept. 22, 2013

Sept. 22, 2013

This is the seabios code that adds support for loading acpi tables from QEMU: it's working fine here and in fact, I am able to get exactly same tables with builtin tables and with tables coming from qemu. Changes from v1: - simplified linker interfaces along the lines suggested by Kevin - fixed lots of bugs Michael S. Tsirkin (5): linker: utility to patch in-memory ROM files pmm: add a way to test whether memory is in FSEG acpi: pack rsdp acpi: load and link tables from /etc/acpi/ acpi: add an option to disable builtin tables Makefile | 2 +- src/Kconfig | 12 +++++- src/acpi.c | 39 +++++++++++++++++ src/acpi.h | 2 +- src/linker.c | 132 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++ src/linker.h | 72 +++++++++++++++++++++++++++++++ src/paravirt.c | 2 + src/pmm.c | 24 ++++++++--- src/util.h | 2 + 9 files changed, 278 insertions(+), 9 deletions(-) create mode 100644 src/linker.c create mode 100644 src/linker.h -- MST

2 15

[PATCH] acpi: hide 64-bit PCI hole for Windows XP
by Igor Mammedov Aug. 14, 2013

Aug. 14, 2013

Windows XP (32 and 64 bit) crashes if 64-bit PCI hole is present. Use _OSI ACPI method to blacklist it and hide 64-bit PCI hole. _OSI strings reported by XP are taken from here: http://msdn.microsoft.com/library/windows/hardware/gg463275 Signed-off-by: Igor Mammedov <imammedo(a)redhat.com> --- src/acpi-dsdt.dsl | 25 +++++++++++++++++++++++++ 1 files changed, 25 insertions(+), 0 deletions(-) diff --git a/src/acpi-dsdt.dsl b/src/acpi-dsdt.dsl index 158f6b4..7b55636 100644 --- a/src/acpi-dsdt.dsl +++ b/src/acpi-dsdt.dsl @@ -33,6 +33,31 @@ DefinitionBlock ( /**************************************************************** + * OS detection + ****************************************************************/ + + Scope(\_SB) { + External(\P1V, IntObj) + + Method(_INI) { + If (CondRefOf (\_OSI, Local0)) { + /* disable 64-bit PCI window for Windows XP and its variants */ + If (\_OSI ("Windows 2001")) { + Store (Zero, P1V) + } + + If (\_OSI ("Windows 2001 SP1")) { + Store (Zero, P1V) + } + + If (\_OSI ("Windows 2001 SP2")) { + Store (Zero, P1V) + } + } + } + } + +/**************************************************************** * PCI Bus definition ****************************************************************/ -- 1.7.1

7 51

[PATCH v4] apm: fix shutdown
by Gerd Hoffmann Aug. 12, 2013

Aug. 12, 2013

Qemu commit 9ee59f3 removed the bochs bios apm interface emulation at port 0x8900. That broke poweroff via APM. Fix it by powering off the machine using the acpi pm control register. Old code is left in, so seabios wil try both poweroff methods. Cleaning that eventually up is left for another patch, after checking it isn't needed. Qemu never implemented "Standby" and "Suspend", only "Shutdown", so it looks like there might be non-qemu use cases (bochs probably). Easiest way to test this is the syslinux poweroff module; modern linux distros usually have CONFIG_APM turned off. Reported-by: Sebastian Herbszt <herbszt(a)gmx.de> Signed-off-by: Gerd Hoffmann <kraxel(a)redhat.com> --- src/acpi.c | 5 +++++ src/acpi.h | 1 + src/apm.c | 5 +++++ src/pciinit.c | 3 +++ 4 files changed, 14 insertions(+) diff --git a/src/acpi.c b/src/acpi.c index 3699898..1994838 100644 --- a/src/acpi.c +++ b/src/acpi.c @@ -18,6 +18,8 @@ #include "acpi-dsdt.hex" +u32 acpi_pm1a_cnt VARFSEG; + static void build_header(struct acpi_table_header *h, u32 sig, int len, u8 rev) { @@ -730,9 +732,12 @@ find_acpi_features(void) if (!fadt) return; u32 pm_tmr = le32_to_cpu(fadt->pm_tmr_blk); + u32 pm1a_cnt = le32_to_cpu(fadt->pm1a_cnt_blk); dprintf(4, "pm_tmr_blk=%x\n", pm_tmr); if (pm_tmr) pmtimer_setup(pm_tmr, 3579); + if (pm1a_cnt) + acpi_pm1a_cnt = pm1a_cnt; // Theoretically we should check the 'reset_reg_sup' flag, but Windows // doesn't and thus nobody seems to *set* it. If the table is large enough diff --git a/src/acpi.h b/src/acpi.h index 5d1e104..f0d24d4 100644 --- a/src/acpi.h +++ b/src/acpi.h @@ -36,6 +36,7 @@ struct rsdp_descriptor { /* Root System Descriptor Pointer */ }; extern struct rsdp_descriptor *RsdpAddr; +extern u32 acpi_pm1a_cnt; /* Table structure from Linux kernel (the ACPI tables are under the BSD license) */ diff --git a/src/apm.c b/src/apm.c index b2eac6d..de6bd99 100644 --- a/src/apm.c +++ b/src/apm.c @@ -12,6 +12,7 @@ #include "config.h" // CONFIG_* #include "biosvar.h" // GET_GLOBAL #include "paravirt.h" // runningOnQEMU +#include "acpi.h" // acpi_pm_ctl static void out_str(const char *str_cs) @@ -108,7 +109,11 @@ handle_155306(struct bregs *regs) void apm_shutdown(void) { + u16 pm1a_cnt = GET_GLOBAL(acpi_pm1a_cnt); + irq_disable(); + if (pm1a_cnt) + outw(0x2000, pm1a_cnt); out_str("Shutdown"); for (;;) hlt(); diff --git a/src/pciinit.c b/src/pciinit.c index 8370b96..0c6b0cb 100644 --- a/src/pciinit.c +++ b/src/pciinit.c @@ -15,6 +15,7 @@ #include "paravirt.h" // RamSize #include "dev-q35.h" // Q35_HOST_BRIDGE_PCIEXBAR_ADDR #include "list.h" // struct hlist_node +#include "acpi.h" // acpi_pm1a_cnt /* PM Timer ticks per second (HZ) */ #define PM_TIMER_FREQUENCY 3579545 @@ -194,6 +195,7 @@ void mch_isa_bridge_setup(struct pci_device *dev, void *arg) /* acpi enable, SCI: IRQ9 000b = irq9*/ pci_config_writeb(bdf, ICH9_LPC_ACPI_CTRL, ICH9_LPC_ACPI_CTRL_ACPI_EN); + acpi_pm1a_cnt = PORT_ACPI_PM_BASE + 0x04; pmtimer_setup(PORT_ACPI_PM_BASE + 0x08, PM_TIMER_FREQUENCY / 1000); } @@ -238,6 +240,7 @@ static void piix4_pm_setup(struct pci_device *pci, void *arg) pci_config_writel(bdf, 0x90, PORT_SMB_BASE | 1); pci_config_writeb(bdf, 0xd2, 0x09); /* enable SMBus io space */ + acpi_pm1a_cnt = PORT_ACPI_PM_BASE + 0x04; pmtimer_setup(PORT_ACPI_PM_BASE + 0x08, PM_TIMER_FREQUENCY / 1000); } -- 1.7.9.7

3 2

[PATCH v4 0/2] pci: load memory window setup from host
by Michael S. Tsirkin Aug. 7, 2013

Aug. 7, 2013

This is a refactoring of patch PATCH v2: pci: load memory window setup from host. Please note qemu merged the required interface already. Please review, and consider for release. Changes from v3: - fix bug causing windows to crash - better debug output (when enabled) Changes from v2: - address comments by Kevin: avoid dynamic memory allocation refactor code to avoid special-casing loaded windows - split out to two patch series Changes from v1: - fix bug in 64 bit range check - address Kevin's comments: move file load into pciinit.c dont reorder initialization sizeof style fix Michael S. Tsirkin (2): pciinit: refactor mem init code pci: load memory window setup from host src/pciinit.c | 145 ++++++++++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 121 insertions(+), 24 deletions(-) -- MST

3 8

Open Firmware path on boot selection screen.
by Weinz Christian Aug. 1, 2013

Aug. 1, 2013

Hello, I suggest adding the Open Firmware paths for devices SeaBIOS can boot from (example: /pci@i0cf8/usb@12,2/usb-*@4) to the boot selection screen. I needed to buy a serial adapter and a null modem just to get that string. That string is important since you cannot modify the default boot sequence without it. Best Regards, Christian Weinz

3 2

vgabios: more x86emu issues
by Gerd Hoffmann July 30, 2013

July 30, 2013

Hi, We already have tools/vgafixup.py to workaround some x86emu issues. Looks like some more is needed, reportly[1] there are more xorg segfaults which look alot like x86emu isses. They trigger with stdvga and qxl (which are handled using the vesa driver), and they only happen on x86_64 where vm86 mode can not be used to run the vgabios. Guest in question is RHEL-5 / centos-5. Xorg version is 1.1.1 Some people digged deep into the x86emu changelogs when this showed up for the first time. Anyone remembers the details? Are there known x86emu bugs where we don't have workarounds for? cheers, Gerd [1] https://bugzilla.redhat.com/show_bug.cgi?id=987097

2 1

[PATCH v3] apm: fix shutdown
by Gerd Hoffmann July 26, 2013

July 26, 2013

Qemu commit 9ee59f3 removed the bochs bios apm interface emulation at port 0x8900. That broke poweroff via APM. Fix it by powering off the machine using the acpi pm control register. Old code is left in, so seabios wil try both poweroff methods. Cleaning that eventually up is left for another patch, after checking it isn't needed. Qemu never implemented "Standby" and "Suspend", only "Shutdown", so it looks like there might be non-qemu use cases (bochs probably). Easiest way to test this is the syslinux poweroff module; modern linux distros usually have CONFIG_APM turned off. Reported-by: Sebastian Herbszt <herbszt(a)gmx.de> Signed-off-by: Gerd Hoffmann <kraxel(a)redhat.com> --- src/acpi.c | 5 +++++ src/acpi.h | 1 + src/apm.c | 2 ++ 3 files changed, 8 insertions(+) diff --git a/src/acpi.c b/src/acpi.c index 3699898..44ba507 100644 --- a/src/acpi.c +++ b/src/acpi.c @@ -18,6 +18,8 @@ #include "acpi-dsdt.hex" +u32 acpi_pm1a_cnt VARLOW = PORT_ACPI_PM_BASE + 0x04; + static void build_header(struct acpi_table_header *h, u32 sig, int len, u8 rev) { @@ -730,9 +732,12 @@ find_acpi_features(void) if (!fadt) return; u32 pm_tmr = le32_to_cpu(fadt->pm_tmr_blk); + u32 pm1a_cnt = le32_to_cpu(fadt->pm1a_cnt_blk); dprintf(4, "pm_tmr_blk=%x\n", pm_tmr); if (pm_tmr) pmtimer_setup(pm_tmr, 3579); + if (pm1a_cnt) + acpi_pm1a_cnt = pm1a_cnt; // Theoretically we should check the 'reset_reg_sup' flag, but Windows // doesn't and thus nobody seems to *set* it. If the table is large enough diff --git a/src/acpi.h b/src/acpi.h index 5d1e104..f0d24d4 100644 --- a/src/acpi.h +++ b/src/acpi.h @@ -36,6 +36,7 @@ struct rsdp_descriptor { /* Root System Descriptor Pointer */ }; extern struct rsdp_descriptor *RsdpAddr; +extern u32 acpi_pm1a_cnt; /* Table structure from Linux kernel (the ACPI tables are under the BSD license) */ diff --git a/src/apm.c b/src/apm.c index b2eac6d..b3b351c 100644 --- a/src/apm.c +++ b/src/apm.c @@ -12,6 +12,7 @@ #include "config.h" // CONFIG_* #include "biosvar.h" // GET_GLOBAL #include "paravirt.h" // runningOnQEMU +#include "acpi.h" // acpi_pm_ctl static void out_str(const char *str_cs) @@ -109,6 +110,7 @@ void apm_shutdown(void) { irq_disable(); + outw(0x2000, acpi_pm1a_cnt); out_str("Shutdown"); for (;;) hlt(); -- 1.7.9.7

2 1

[PATCH v2] apm: fix shutdown
by Gerd Hoffmann July 26, 2013

July 26, 2013

Qemu commit 9ee59f3 removed the bochs bios apm interface emulation at port 0x8900. That broke poweroff via APM. Fix it by powering off the machine using the acpi pm control register. Old code is left in, so seabios wil try both poweroff methods. Cleaning that eventually up is left for another patch, after checking it isn't needed. Qemu never implemented "Standby" and "Suspend", only "Shutdown", so it looks like there might be non-qemu use cases (bochs probably). Easiest way to test this is the syslinux poweroff module; modern linux distros usually have CONFIG_APM turned off. Reported-by: Sebastian Herbszt <herbszt(a)gmx.de> Signed-off-by: Gerd Hoffmann <kraxel(a)redhat.com> --- src/acpi.c | 5 +++++ src/acpi.h | 1 + src/apm.c | 2 ++ 3 files changed, 8 insertions(+) diff --git a/src/acpi.c b/src/acpi.c index 3699898..2062033 100644 --- a/src/acpi.c +++ b/src/acpi.c @@ -18,6 +18,8 @@ #include "acpi-dsdt.hex" +u32 acpi_pm1a_cnt = PORT_ACPI_PM_BASE + 0x04; + static void build_header(struct acpi_table_header *h, u32 sig, int len, u8 rev) { @@ -730,9 +732,12 @@ find_acpi_features(void) if (!fadt) return; u32 pm_tmr = le32_to_cpu(fadt->pm_tmr_blk); + u32 pm1a_cnt = le32_to_cpu(fadt->pm1a_cnt_blk); dprintf(4, "pm_tmr_blk=%x\n", pm_tmr); if (pm_tmr) pmtimer_setup(pm_tmr, 3579); + if (pm1a_cnt) + acpi_pm1a_cnt = pm1a_cnt; // Theoretically we should check the 'reset_reg_sup' flag, but Windows // doesn't and thus nobody seems to *set* it. If the table is large enough diff --git a/src/acpi.h b/src/acpi.h index 5d1e104..f0d24d4 100644 --- a/src/acpi.h +++ b/src/acpi.h @@ -36,6 +36,7 @@ struct rsdp_descriptor { /* Root System Descriptor Pointer */ }; extern struct rsdp_descriptor *RsdpAddr; +extern u32 acpi_pm1a_cnt; /* Table structure from Linux kernel (the ACPI tables are under the BSD license) */ diff --git a/src/apm.c b/src/apm.c index b2eac6d..b3b351c 100644 --- a/src/apm.c +++ b/src/apm.c @@ -12,6 +12,7 @@ #include "config.h" // CONFIG_* #include "biosvar.h" // GET_GLOBAL #include "paravirt.h" // runningOnQEMU +#include "acpi.h" // acpi_pm_ctl static void out_str(const char *str_cs) @@ -109,6 +110,7 @@ void apm_shutdown(void) { irq_disable(); + outw(0x2000, acpi_pm1a_cnt); out_str("Shutdown"); for (;;) hlt(); -- 1.7.9.7

1 0

[PATCH] apm: fix shutdown
by Gerd Hoffmann July 26, 2013

July 26, 2013

Qemu commit 9ee59f3 removed the bochs bios apm interface emulation at port 0x8900. That broke poweroff via APM. Fix it by powering off the machine using the acpi pm control register. Old code is left in, so seabios wil try both poweroff methods. Cleaning that eventually up is left for another patch, after checking it isn't needed. Qemu never implemented "Standby" and "Suspend", only "Shutdown", so it looks like there might be non-qemu use cases (bochs probably). Easiest way to test this is the syslinux poweroff module; modern linux distros usually have CONFIG_APM turned off. Reported-by: Sebastian Herbszt <herbszt(a)gmx.de> Signed-off-by: Gerd Hoffmann <kraxel(a)redhat.com> --- src/apm.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/apm.c b/src/apm.c index b2eac6d..15f37dd 100644 --- a/src/apm.c +++ b/src/apm.c @@ -109,6 +109,7 @@ void apm_shutdown(void) { irq_disable(); + outw(0x2000, PORT_ACPI_PM_BASE + 0x04); out_str("Shutdown"); for (;;) hlt(); -- 1.7.9.7

2 1