coreboot March 2012

coreboot@coreboot.org

58 participants
718 discussions

SMP stop_this_cpu and AP_IN_SIPI_WAIT
by Kyösti Mälkki 12 Mar '12

12 Mar '12

There are currently two different versions of stop_this_cpu() used to halt an AP CPU after init and microcode update. Alternative #1, the default stop_this_cpu() { while (1) { hlt() } } Alternative #2, used on selected Intel Cores stop_this_cpu() { send INIT IPI message to self /* should not reach here */ while (1) { hlt() } } I would like to know the reasoning for implementing #2. Was this a solution to an observed problem? The commentary speaks about hyper-threading, while the processor models with the AP_IN_SIPI_WAIT enabled are really multi-core. I believe alternative #1 one is not sufficient for any SMP platform. If an exception raises and the AP CPU has NULL IDT, a hyper-threading CPU may shutdown its sibling CPUs in the same physical package. If there is a valid IDT and the exception handler returns, memory at CS:EIP may no longer contain the hlt() loop. Thanks, KM

2 1

Trac reminder: list of new ticket(s)
by coreboot tracker 12 Mar '12

12 Mar '12

1 0

Patch merged into coreboot/master: 30e68c8 No need to setup include paths with .s files
by gerrit＠coreboot.org 11 Mar '12

11 Mar '12

the following patch was just integrated into master: commit 30e68c8ea15af29bc3eb7a9e6886237b37d634ac Author: Patrick Georgi <patrick(a)georgi-clan.de> Date: Sun Mar 11 19:34:12 2012 +0100 No need to setup include paths with .s files They're already preprocessed, and clang whines. Change-Id: I57fe936f84a2fe1aa50ee8510fef606f2ed2ea23 Signed-off-by: Patrick Georgi <patrick(a)georgi-clan.de> See http://review.coreboot.org/782 for details. -gerrit

1 0

Patch merged into coreboot/master: b845313 malloc: size is unsigned, don't test for size < 0
by gerrit＠coreboot.org 11 Mar '12

11 Mar '12

the following patch was just integrated into master: commit b84531374751d8679d28657935ed3860a4b3851f Author: Patrick Georgi <patrick(a)georgi-clan.de> Date: Sun Mar 11 19:30:36 2012 +0100 malloc: size is unsigned, don't test for size < 0 clang complains Change-Id: Ifadf73cf377c0d1808e20731803e01101bad7e1d Signed-off-by: Patrick Georgi <patrick(a)georgi-clan.de> Build-Tested: build bot (Jenkins) at Sun Mar 11 21:22:26 2012, giving +1 Reviewed-By: Mathias Krause <minipli(a)googlemail.com> at Sun Mar 11 22:18:48 2012, giving +2 See http://review.coreboot.org/780 for details. -gerrit

1 0

New patch to review for coreboot: f499003 clang: Don't use mmx nor sse
by Patrick Georgi 11 Mar '12

11 Mar '12

Patrick Georgi (patrick(a)georgi-clan.de) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/786 -gerrit commit f499003cb68b64c427b0263db86678fcfe26f957 Author: Patrick Georgi <patrick(a)georgi-clan.de> Date: Sun Mar 11 20:44:22 2012 +0100 clang: Don't use mmx nor sse clang is much more trigger happy than gcc on those. Change-Id: Ie7c219de3cc26675692eab7361a4ad551f1c65a7 Signed-off-by: Patrick Georgi <patrick(a)georgi-clan.de> --- Makefile | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/Makefile b/Makefile index f2277d1..c5ee1f2 100644 --- a/Makefile +++ b/Makefile @@ -113,7 +113,7 @@ include $(HAVE_DOTCONFIG) ifneq ($(INNER_SCANBUILD),y) ifeq ($(CONFIG_COMPILER_LLVM_CLANG),y) -CC:=clang -m32 +CC:=clang -m32 -mno-mmx -mno-sse HOSTCC:=clang endif endif

1 0

New patch to review for coreboot: 7f1e868 Don't wrap linker script wielding ld through $(CC)
by Patrick Georgi 11 Mar '12

11 Mar '12

Patrick Georgi (patrick(a)georgi-clan.de) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/785 -gerrit commit 7f1e86895af6c3e092f4c45103970c0ff3677f80 Author: Patrick Georgi <patrick(a)georgi-clan.de> Date: Sun Mar 11 20:15:21 2012 +0100 Don't wrap linker script wielding ld through $(CC) clang does its own thing when linking, somewhat incompatible to our linker magic. Change-Id: I243597adcb6bc3f7343c3431d7473610c327353d Signed-off-by: Patrick Georgi <patrick(a)georgi-clan.de> --- src/arch/x86/Makefile.inc | 10 +++++----- 1 files changed, 5 insertions(+), 5 deletions(-) diff --git a/src/arch/x86/Makefile.inc b/src/arch/x86/Makefile.inc index 250c077..4494e59 100755 --- a/src/arch/x86/Makefile.inc +++ b/src/arch/x86/Makefile.inc @@ -144,7 +144,7 @@ $(objutil)/options/build_opt_tbl: $(top)/util/options/build_opt_tbl.c $(top)/src $(obj)/coreboot_ram: $(obj)/coreboot_ram.o $(src)/arch/x86/coreboot_ram.ld #ldoptions @printf " CC $(subst $(obj)/,,$(@))\n" - $(CC) -nostdlib -nostartfiles -static -o $@ -L$(obj) -T $(src)/arch/x86/coreboot_ram.ld $(obj)/coreboot_ram.o + $(LD) -m elf_i386 -o $@ -L$(obj) $(obj)/coreboot_ram.o -T $(src)/arch/x86/coreboot_ram.ld $(NM) -n $(obj)/coreboot_ram | sort > $(obj)/coreboot_ram.map $(OBJCOPY) --only-keep-debug $@ $(obj)/coreboot_ram.debug $(OBJCOPY) --strip-debug $@ @@ -152,7 +152,7 @@ $(obj)/coreboot_ram: $(obj)/coreboot_ram.o $(src)/arch/x86/coreboot_ram.ld #ldop $(obj)/coreboot_ram.o: $(obj)/arch/x86/lib/c_start.ramstage.o $$(driver-objs) $(obj)/coreboot.a $(LIBGCC_FILE_NAME) @printf " CC $(subst $(obj)/,,$(@))\n" - $(CC) -nostdlib -r -o $@ $(obj)/arch/x86/lib/c_start.ramstage.o $(driver-objs) -Wl,--wrap,__divdi3 -Wl,--wrap,__udivdi3 -Wl,--wrap,__moddi3 -Wl,--wrap,__umoddi3 -Wl,--start-group $(obj)/coreboot.a $(LIBGCC_FILE_NAME) -Wl,--end-group + $(LD) -m elf_i386 -r -o $@ $(obj)/arch/x86/lib/c_start.ramstage.o $(driver-objs) --wrap __divdi3 --wrap __udivdi3 --wrap __moddi3 --wrap __umoddi3 --start-group $(obj)/coreboot.a $(LIBGCC_FILE_NAME) --end-group $(obj)/coreboot.a: $$(ramstage-objs) @printf " AR $(subst $(obj)/,,$(@))\n" @@ -332,7 +332,7 @@ $(obj)/mainboard/$(MAINBOARDDIR)/bootblock.inc: $(src)/arch/x86/init/$(subst ",, $(obj)/bootblock.elf: $(obj)/mainboard/$(MAINBOARDDIR)/bootblock.o $(obj)/bootblock/ldscript.ld @printf " LINK $(subst $(obj)/,,$(@))\n" - $(CC) -nostdlib -nostartfiles -static -o $@ -L$(obj) -T $(obj)/bootblock/ldscript.ld $< + $(LD) -m elf_i386 -static -o $@ -L$(obj) $< -T $(obj)/bootblock/ldscript.ld $(NM) -n $(obj)/bootblock.elf | sort > $(obj)/bootblock.map $(OBJCOPY) --only-keep-debug $@ $(obj)/bootblock.debug $(OBJCOPY) --strip-debug $@ @@ -343,13 +343,13 @@ $(obj)/bootblock.elf: $(obj)/mainboard/$(MAINBOARDDIR)/bootblock.o $(obj)/bootbl $(obj)/coreboot.romstage: $(obj)/coreboot.pre1 $$(romstage-objs) $(obj)/romstage/ldscript.ld @printf " LINK $(subst $(obj)/,,$(@))\n" printf "CONFIG_ROMBASE = 0x0;\n" > $(obj)/location.ld - $(CC) -nostdlib -nostartfiles -static -o $(obj)/romstage.elf -L$(obj) -T $(obj)/romstage/ldscript.ld $(romstage-objs) + $(LD) -nostdlib -nostartfiles -static -o $(obj)/romstage.elf -L$(obj) $(romstage-objs) -T $(obj)/romstage/ldscript.ld $(OBJCOPY) -O binary $(obj)/romstage.elf $(obj)/romstage.bin printf "CONFIG_ROMBASE = 0x" > $(obj)/location.ld $(CBFSTOOL) $(obj)/coreboot.pre1 locate $(obj)/romstage.bin $(CONFIG_CBFS_PREFIX)/romstage $(CONFIG_XIP_ROM_SIZE) > $(obj)/location.txt || { echo "The romstage is larger than XIP size. Please expand the CONFIG_XIP_ROM_SIZE" ; exit 1; } cat $(obj)/location.txt >> $(obj)/location.ld printf ';\n' >> $(obj)/location.ld - $(CC) -nostdlib -nostartfiles -static -o $(obj)/romstage.elf -L$(obj) -T $(obj)/romstage/ldscript.ld $(romstage-objs) + $(LD) -nostdlib -nostartfiles -static -o $(obj)/romstage.elf -L$(obj) $(romstage-objs) -T $(obj)/romstage/ldscript.ld $(NM) -n $(obj)/romstage.elf | sort > $(obj)/romstage.map $(OBJCOPY) --only-keep-debug $(obj)/romstage.elf $(obj)/romstage.debug $(OBJCOPY) --strip-debug $(obj)/romstage.elf

1 0

New patch to review for coreboot: f29db5c clang's integrated assembler isn't complete enough yet
by Patrick Georgi 11 Mar '12

11 Mar '12

Patrick Georgi (patrick(a)georgi-clan.de) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/784 -gerrit commit f29db5c1dda55cb0c520b27a90d3c134ff396999 Author: Patrick Georgi <patrick(a)georgi-clan.de> Date: Sun Mar 11 19:43:30 2012 +0100 clang's integrated assembler isn't complete enough yet So we tell clang not to use it. (.code16 and other issues) Change-Id: I881326354d7286219beadfc31f54e5a6923adb7a Signed-off-by: Patrick Georgi <patrick(a)georgi-clan.de> --- Makefile.inc | 6 ++++++ src/arch/x86/Makefile.inc | 4 ++-- 2 files changed, 8 insertions(+), 2 deletions(-) diff --git a/Makefile.inc b/Makefile.inc index 86a072a..1375684 100644 --- a/Makefile.inc +++ b/Makefile.inc @@ -58,6 +58,12 @@ ifeq ($(CONFIG_TRACE),y) ramstage-c-ccopts:= -finstrument-functions endif +ifeq ($(CONFIG_COMPILER_LLVM_CLANG),y) +romstage-s-ccopts:=-no-integrated-as +ramstage-S-ccopts:=-no-integrated-as +ramstage-s-ccopts:=-no-integrated-as +endif + smm-c-ccopts:=-D__SMM__ smm-S-ccopts:=-D__SMM__ diff --git a/src/arch/x86/Makefile.inc b/src/arch/x86/Makefile.inc index d3dac11..250c077 100755 --- a/src/arch/x86/Makefile.inc +++ b/src/arch/x86/Makefile.inc @@ -318,7 +318,7 @@ $(obj)/bootblock/bootblock.S: $$(bootblock_inc) $(obj)/mainboard/$(MAINBOARDDIR)/bootblock.o: $(obj)/mainboard/$(MAINBOARDDIR)/bootblock.s @printf " CC $(subst $(obj)/,,$(@))\n" - $(CC) -Wa,-acdlns -c -o $@ $< > $(dir $(a))/crt0.disasm + $(CC) $(romstage-s-ccopts) -Wa,-acdlns -c -o $@ $< > $(dir $(a))/crt0.disasm $(obj)/mainboard/$(MAINBOARDDIR)/bootblock.s: $(obj)/bootblock/bootblock.S $(obj)/config.h $(obj)/build.h @printf " CC $(subst $(obj)/,,$(@))\n" @@ -368,7 +368,7 @@ $(obj)/romstage/crt0.S: $$(crt0s) $(obj)/mainboard/$(MAINBOARDDIR)/crt0.romstage.o: $(obj)/mainboard/$(MAINBOARDDIR)/crt0.s @printf " CC $(subst $(obj)/,,$(@))\n" - $(CC) -Wa,-acdlns -c -o $@ $< > $(dir $(a))/crt0.disasm + $(CC) $(romstage-s-ccopts) -Wa,-acdlns -c -o $@ $< > $(dir $(a))/crt0.disasm $(obj)/mainboard/$(MAINBOARDDIR)/crt0.s: $(obj)/romstage/crt0.S $(obj)/config.h $(obj)/build.h @printf " CC $(subst $(obj)/,,$(@))\n"

1 0

New patch to review for coreboot: 044489f xchg is atomic with side-effects
by Patrick Georgi 11 Mar '12

11 Mar '12

Patrick Georgi (patrick(a)georgi-clan.de) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/783 -gerrit commit 044489fc6f8e15eaafa4172186ec6a6fb79b4f90 Author: Patrick Georgi <patrick(a)georgi-clan.de> Date: Sun Mar 11 19:42:33 2012 +0100 xchg is atomic with side-effects clang doesn't know about the side effect, so we have to tell it that it's okay not to care about the result. Change-Id: Ib11890bff6779e36cf09c178d224695ea16a8ae8 Signed-off-by: Patrick Georgi <patrick(a)georgi-clan.de> --- src/include/cpu/x86/lapic.h | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/include/cpu/x86/lapic.h b/src/include/cpu/x86/lapic.h index 8b44a6c..68608ed 100644 --- a/src/include/cpu/x86/lapic.h +++ b/src/include/cpu/x86/lapic.h @@ -108,7 +108,7 @@ static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int siz static inline void lapic_write_atomic(unsigned long reg, unsigned long v) { - xchg((volatile unsigned long *)(LAPIC_DEFAULT_BASE+reg), v); + (void)xchg((volatile unsigned long *)(LAPIC_DEFAULT_BASE+reg), v); }

1 0

New patch to review for coreboot: 30e68c8 No need to setup include paths with .s files
by Patrick Georgi 11 Mar '12

11 Mar '12

Patrick Georgi (patrick(a)georgi-clan.de) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/782 -gerrit commit 30e68c8ea15af29bc3eb7a9e6886237b37d634ac Author: Patrick Georgi <patrick(a)georgi-clan.de> Date: Sun Mar 11 19:34:12 2012 +0100 No need to setup include paths with .s files They're already preprocessed, and clang whines. Change-Id: I57fe936f84a2fe1aa50ee8510fef606f2ed2ea23 Signed-off-by: Patrick Georgi <patrick(a)georgi-clan.de> --- src/arch/x86/Makefile.inc | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/arch/x86/Makefile.inc b/src/arch/x86/Makefile.inc index 4f76bd0..d3dac11 100755 --- a/src/arch/x86/Makefile.inc +++ b/src/arch/x86/Makefile.inc @@ -318,7 +318,7 @@ $(obj)/bootblock/bootblock.S: $$(bootblock_inc) $(obj)/mainboard/$(MAINBOARDDIR)/bootblock.o: $(obj)/mainboard/$(MAINBOARDDIR)/bootblock.s @printf " CC $(subst $(obj)/,,$(@))\n" - $(CC) -I$(obj) -Wa,-acdlns -c -o $@ $< > $(dir $(a))/crt0.disasm + $(CC) -Wa,-acdlns -c -o $@ $< > $(dir $(a))/crt0.disasm $(obj)/mainboard/$(MAINBOARDDIR)/bootblock.s: $(obj)/bootblock/bootblock.S $(obj)/config.h $(obj)/build.h @printf " CC $(subst $(obj)/,,$(@))\n" @@ -368,7 +368,7 @@ $(obj)/romstage/crt0.S: $$(crt0s) $(obj)/mainboard/$(MAINBOARDDIR)/crt0.romstage.o: $(obj)/mainboard/$(MAINBOARDDIR)/crt0.s @printf " CC $(subst $(obj)/,,$(@))\n" - $(CC) -I$(obj) -Wa,-acdlns -c -o $@ $< > $(dir $(a))/crt0.disasm + $(CC) -Wa,-acdlns -c -o $@ $< > $(dir $(a))/crt0.disasm $(obj)/mainboard/$(MAINBOARDDIR)/crt0.s: $(obj)/romstage/crt0.S $(obj)/config.h $(obj)/build.h @printf " CC $(subst $(obj)/,,$(@))\n"

1 0

New patch to review for coreboot: 0f446f7 printf: Remove some L modifier uses
by Patrick Georgi 11 Mar '12

11 Mar '12

Patrick Georgi (patrick(a)georgi-clan.de) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/781 -gerrit commit 0f446f71c39bd8bff9a399826e584a2863217ca7 Author: Patrick Georgi <patrick(a)georgi-clan.de> Date: Sun Mar 11 19:31:03 2012 +0100 printf: Remove some L modifier uses We use the L modifier in a non-standard way (for long long instead of long double, which we have no business with). clang complains, to reduce its use, to make emulation/qemu-x86 happier. Long term, we should consider eliminating public uses of 'L' (but internal use in vtxprintf to denote long long is fine) Change-Id: If9a17d9ae9925cdc8736445e7d5eedc59c7028c6 Signed-off-by: Patrick Georgi <patrick(a)georgi-clan.de> --- src/devices/device.c | 8 ++++---- src/devices/device_util.c | 4 ++-- src/devices/pnp_device.c | 2 +- 3 files changed, 7 insertions(+), 7 deletions(-) diff --git a/src/devices/device.c b/src/devices/device.c index a2619bf..0d2bf8f 100644 --- a/src/devices/device.c +++ b/src/devices/device.c @@ -322,7 +322,7 @@ static void compute_resources(struct bus *bus, struct resource *bridge, (resource->flags & IORESOURCE_ASSIGNED)) { printk(BIOS_ERR, "Resource limit looks wrong! (no APIC?)\n"); - printk(BIOS_ERR, "%s %02lx limit %08Lx\n", + printk(BIOS_ERR, "%s %02lx limit %08llx\n", dev_path(dev), resource->index, resource->limit); } @@ -627,7 +627,7 @@ static void avoid_fixed_resources(struct device *dev) for (res = dev->resource_list; res; res = res->next) { if ((res->flags & IORESOURCE_FIXED)) continue; - printk(BIOS_SPEW, "%s:@%s %02lx limit %08Lx\n", __func__, + printk(BIOS_SPEW, "%s:@%s %02lx limit %08llx\n", __func__, dev_path(dev), res->index, res->limit); if ((res->flags & MEM_MASK) == PREF_TYPE && (res->limit < limits.pref.limit)) @@ -660,9 +660,9 @@ static void avoid_fixed_resources(struct device *dev) else continue; - printk(BIOS_SPEW, "%s2: %s@%02lx limit %08Lx\n", __func__, + printk(BIOS_SPEW, "%s2: %s@%02lx limit %08llx\n", __func__, dev_path(dev), res->index, res->limit); - printk(BIOS_SPEW, "\tlim->base %08Lx lim->limit %08Lx\n", + printk(BIOS_SPEW, "\tlim->base %08llx lim->limit %08llx\n", lim->base, lim->limit); /* Is the resource outside the limits? */ diff --git a/src/devices/device_util.c b/src/devices/device_util.c index 5b182e1..526490a 100644 --- a/src/devices/device_util.c +++ b/src/devices/device_util.c @@ -571,7 +571,7 @@ void report_resource_stored(device_t dev, struct resource *resource, sprintf(buf, "bus %02x ", dev->link_list->secondary); #endif } - printk(BIOS_DEBUG, "%s %02lx <- [0x%010Lx - 0x%010Lx] size 0x%08Lx " + printk(BIOS_DEBUG, "%s %02lx <- [0x%010llx - 0x%010llx] size 0x%08llx " "gran 0x%02x %s%s%s\n", dev_path(dev), resource->index, base, end, resource->size, resource->gran, buf, resource_type(resource), comment); @@ -786,7 +786,7 @@ void show_one_resource(int debug_level, struct device *dev, */ do_printk(debug_level, "%s %02lx <- [0x%010llx - 0x%010llx] " - "size 0x%08Lx gran 0x%02x %s%s%s\n", dev_path(dev), + "size 0x%08llx gran 0x%02x %s%s%s\n", dev_path(dev), resource->index, base, end, resource->size, resource->gran, buf, resource_type(resource), comment); } diff --git a/src/devices/pnp_device.c b/src/devices/pnp_device.c index a6d428a..fb6c9d6 100644 --- a/src/devices/pnp_device.c +++ b/src/devices/pnp_device.c @@ -107,7 +107,7 @@ void pnp_read_resources(device_t dev) static void pnp_set_resource(device_t dev, struct resource *resource) { if (!(resource->flags & IORESOURCE_ASSIGNED)) { - printk(BIOS_ERR, "ERROR: %s %02lx %s size: 0x%010Lx " + printk(BIOS_ERR, "ERROR: %s %02lx %s size: 0x%010llx " "not assigned\n", dev_path(dev), resource->index, resource_type(resource), resource->size); return;

1 0

← Newer
1
...
29
30
31
32
33
34
35
...
72
Older →

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

coreboot March 2012