SeaBIOS

seabios@seabios.org

3806 discussions

[PATCH] timer: Configure timer1 to a standard settings
by Petr Cvek 12 Aug '22

12 Aug '22

ASPI2DOS.SYS and KEYB.COM from Win 98 SE installation CD (and most likely other DOS versions too) depend on I/O port 0x61 bit 4 to be toggled. This requires timer 1 (I/O 0x41, legacy DRAM refresh) to be correctly set. Also Intel ICH7 Family Datasheet, chapter 5.8 states: Programming the counter to anything other than Mode 2 will result in undefined behavior for the REF_TOGGLE bit. Failing to have the timer 1 configured indeed causes affected OSes to freeze during the boot. Signed-off-by: Petr Cvek <petrcvekcz(a)gmail.com> --- src/hw/timer.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/src/hw/timer.c b/src/hw/timer.c index b6f102e..b272f44 100644 --- a/src/hw/timer.c +++ b/src/hw/timer.c @@ -280,4 +280,10 @@ pit_setup(void) // maximum count of 0000H = 18.2Hz outb(0x0, PORT_PIT_COUNTER0); outb(0x0, PORT_PIT_COUNTER0); + + // timer1: binary count, 16bit count, mode 2 + outb(PM_SEL_TIMER1|PM_ACCESS_WORD|PM_MODE2|PM_CNT_BINARY, PORT_PIT_MODE); + // maximum count of 0012H = 66.3kHz + outb(0x12, PORT_PIT_COUNTER1); + outb(0x0, PORT_PIT_COUNTER1); } -- 2.37.0

3 3

[PATCH 0/2] Add support for all standard resolutions
by Elliot Killick 01 Aug '22

01 Aug '22

This patch series adds support for all the standard resolutions as listed on Wikipedia. I've fixed the issue Gerd pointed out and rebased onto master. I also see that Gerd himself has recently put out a patch for 4K resolutions, however, my own ultrawide resolution of 3440x1440 remains unsupported by SeaBIOS. I've heard from some others that they have unique resolutions that are still not in SeaBIOS but they should be with this patch so I hope we can get it in. Regarding 24 bpp, probably yes, most software doesn't care about it but for maximal compatability (and because Windows has been finicky in my experience), I think we should include it. Also, there's still the chance that some people may want to use it in some unlikely scenarios which they should be able to do without recompilation. One last thing to consider is that we would then have to decide at what point do the resolutions being added with this patch become modern/new enough that it no longer makes sense to keep the 24 bpp variants. It seems like we would basically just have to draw an arbitrary line in the sand. However, if you would still like to remove 24 bpp variants then please provide me with a cut off point and I'll put it in the next patch version. Again, if we could get this in that would be fantastic because I've personally had people raise this issue with me in regards to an issue they're having in relation to one of my projects. Thanks, Elliot Elliot Killick (2): svgamodes: Add support for all standard resolutions svgamodes: Improve formatting vgasrc/svgamodes.c | 225 ++++++++++++++++++++++++++++----------------- 1 file changed, 141 insertions(+), 84 deletions(-) -- 2.34.3

1 2

[PATCH 0/2] Add support for all standard resolutions
by Elliot Killick 01 Aug '22

01 Aug '22

1 2

[PATCH] virtio-blk: use larger default request size
by Gerd Hoffmann 19 Jul '22

19 Jul '22

Bump default from 8 to 64 blocks. Using 8 by default leads to requests being splitted on qemu, which slows down boot. Some (temporary) debug logging added showed that almost all requests on a standard fedora install are less than 64 blocks, so that should bring us back to 1.15 performance levels. Signed-off-by: Gerd Hoffmann <kraxel(a)redhat.com> --- src/hw/virtio-blk.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/hw/virtio-blk.c b/src/hw/virtio-blk.c index 929ba8873dc0..9b4a05a4c424 100644 --- a/src/hw/virtio-blk.c +++ b/src/hw/virtio-blk.c @@ -95,7 +95,7 @@ virtio_blk_op(struct disk_op_s *op, int write) blk_num_max = (u16)max_io_size / vdrive->drive.blksize; else /* default blk_num_max if hardware doesnot advise a proper value */ - blk_num_max = 8; + blk_num_max = 64; if (op->count <= blk_num_max) { virtio_blk_op_one_segment(vdrive, write, sg); -- 2.36.1

2 2

[PATCH v3] Replace $(OBJDUMP) -thr with $(READELF) -WSrs
by Fangrui Song 14 Jul '22

14 Jul '22

objdump -h relies heavily on BFD internals and the BFD flags. The output is difficult to emulate in llvm-objdump. (llvm-objdump -h has a different output https://reviews.llvm.org/D57146). Switch to READELF, so that llvm-readelf can be used as an alternative. readelf is generally better than objdump when dumping ELF section/symbol information. Signed-off-by: Fangrui Song <maskray(a)google.com> --- Changes from v2 (https://mail.coreboot.org/hyperkitty/list/seabios@seabios.org/message/4AZCP…) * Change an objdump comment to mention readelf * For your convenience, this patch is available on https://github.com/MaskRay/seabios/tree/readelf --- Makefile | 18 +++--- scripts/checkrom.py | 2 +- scripts/layoutrom.py | 145 ++++++++++++++++++++----------------------- 3 files changed, 79 insertions(+), 86 deletions(-) diff --git a/Makefile b/Makefile index c108f87..5c6a838 100644 --- a/Makefile +++ b/Makefile @@ -20,7 +20,7 @@ endif AS=$(CROSS_PREFIX)as LD=$(CROSS_PREFIX)ld OBJCOPY=$(CROSS_PREFIX)objcopy -OBJDUMP=$(CROSS_PREFIX)objdump +READELF=$(CROSS_PREFIX)readelf STRIP=$(CROSS_PREFIX)strip PYTHON=python CPP=cpp @@ -165,14 +165,14 @@ $(OUT)romlayout.o: src/romlayout.S $(OUT)autoconf.h $(OUT)asm-offsets.h $(OUT)romlayout16.lds: $(OUT)ccode32flat.o $(OUT)code32seg.o $(OUT)ccode16.o $(OUT)romlayout.o src/version.c scripts/layoutrom.py scripts/buildversion.py @echo " Building ld scripts" - $(Q)$(PYTHON) ./scripts/buildversion.py -e "$(EXTRAVERSION)" -t "$(CC);$(AS);$(LD);$(OBJCOPY);$(OBJDUMP);$(STRIP)" $(OUT)autoversion.h + $(Q)$(PYTHON) ./scripts/buildversion.py -e "$(EXTRAVERSION)" -t "$(CC);$(AS);$(LD);$(OBJCOPY);$(READELF);$(STRIP)" $(OUT)autoversion.h $(Q)$(CC) $(CFLAGS32FLAT) -c src/version.c -o $(OUT)version.o $(Q)$(LD) $(LD32BIT_FLAG) -r $(OUT)ccode32flat.o $(OUT)version.o -o $(OUT)code32flat.o $(Q)$(LD) $(LD32BIT_FLAG) -r $(OUT)ccode16.o $(OUT)romlayout.o -o $(OUT)code16.o - $(Q)$(OBJDUMP) -thr $(OUT)code32flat.o > $(OUT)code32flat.o.objdump - $(Q)$(OBJDUMP) -thr $(OUT)code32seg.o > $(OUT)code32seg.o.objdump - $(Q)$(OBJDUMP) -thr $(OUT)code16.o > $(OUT)code16.o.objdump - $(Q)$(PYTHON) ./scripts/layoutrom.py $(OUT)code16.o.objdump $(OUT)code32seg.o.objdump $(OUT)code32flat.o.objdump $(OUT)$(KCONFIG_AUTOHEADER) $(OUT)romlayout16.lds $(OUT)romlayout32seg.lds $(OUT)romlayout32flat.lds + $(Q)$(READELF) -WSrs $(OUT)code32flat.o > $(OUT)code32flat.o.readelf + $(Q)$(READELF) -WSrs $(OUT)code32seg.o > $(OUT)code32seg.o.readelf + $(Q)$(READELF) -WSrs $(OUT)code16.o > $(OUT)code16.o.readelf + $(Q)$(PYTHON) ./scripts/layoutrom.py $(OUT)code16.o.readelf $(OUT)code32seg.o.readelf $(OUT)code32flat.o.readelf $(OUT)$(KCONFIG_AUTOHEADER) $(OUT)romlayout16.lds $(OUT)romlayout32seg.lds $(OUT)romlayout32flat.lds # These are actually built by scripts/layoutrom.py above, but by pulling them # into an extra rule we prevent make -j from spawning layoutrom.py 4 times. @@ -193,9 +193,9 @@ $(OUT)rom.o: $(OUT)rom16.noexec.o $(OUT)rom32seg.noexec.o $(OUT)code32flat.o $(O $(OUT)bios.bin.prep: $(OUT)rom.o scripts/checkrom.py @echo " Prepping $@" $(Q)rm -f $(OUT)bios.bin $(OUT)Csm16.bin $(OUT)bios.bin.elf - $(Q)$(OBJDUMP) -thr $< > $<.objdump + $(Q)$(READELF) -WSrs $< > $<.readelf $(Q)$(OBJCOPY) -O binary $< $(OUT)bios.bin.raw - $(Q)$(PYTHON) ./scripts/checkrom.py $<.objdump $(CONFIG_ROM_SIZE) $(OUT)bios.bin.raw $(OUT)bios.bin.prep + $(Q)$(PYTHON) ./scripts/checkrom.py $<.readelf $(CONFIG_ROM_SIZE) $(OUT)bios.bin.raw $(OUT)bios.bin.prep $(OUT)bios.bin: $(OUT)bios.bin.prep @echo " Creating $@" @@ -237,7 +237,7 @@ $(OUT)vgaentry.o: vgasrc/vgaentry.S $(OUT)autoconf.h $(OUT)asm-offsets.h $(OUT)vgarom.o: $(OUT)vgaccode16.o $(OUT)vgaentry.o $(OUT)vgasrc/vgalayout.lds vgasrc/vgaversion.c scripts/buildversion.py @echo " Linking $@" - $(Q)$(PYTHON) ./scripts/buildversion.py -e "$(EXTRAVERSION)" -t "$(CC);$(AS);$(LD);$(OBJCOPY);$(OBJDUMP);$(STRIP)" $(OUT)autovgaversion.h + $(Q)$(PYTHON) ./scripts/buildversion.py -e "$(EXTRAVERSION)" -t "$(CC);$(AS);$(LD);$(OBJCOPY);$(READELF);$(STRIP)" $(OUT)autovgaversion.h $(Q)$(CC) $(CFLAGS16) -c vgasrc/vgaversion.c -o $(OUT)vgaversion.o $(Q)$(LD) --gc-sections -T $(OUT)vgasrc/vgalayout.lds $(OUT)vgaccode16.o $(OUT)vgaentry.o $(OUT)vgaversion.o -o $@ diff --git a/scripts/checkrom.py b/scripts/checkrom.py index a5b15a4..4a23f0a 100755 --- a/scripts/checkrom.py +++ b/scripts/checkrom.py @@ -21,7 +21,7 @@ def main(): # Read in symbols objinfofile = open(objinfo, 'r') - symbols = layoutrom.parseObjDump(objinfofile, 'in')[1] + symbols = layoutrom.parseReadElf(objinfofile, 'in')[1] # Read in raw file f = open(rawfile, 'rb') diff --git a/scripts/layoutrom.py b/scripts/layoutrom.py index 7bc4a57..f4b94ea 100755 --- a/scripts/layoutrom.py +++ b/scripts/layoutrom.py @@ -541,87 +541,80 @@ class Reloc: class Symbol: name = offset = section = None -# Read in output from objdump -def parseObjDump(file, fileid): +def parseReadElf(file, fileid): # sections = [section, ...] sections = [] sectionmap = {} # symbols[symbolname] = symbol symbols = {} + lines = file.readlines() - state = None - for line in file.readlines(): - line = line.rstrip() - if line == 'Sections:': - state = 'section' - continue - if line == 'SYMBOL TABLE:': - state = 'symbol' - continue - if line.startswith('RELOCATION RECORDS FOR ['): - sectionname = line[24:-2] - if sectionname.startswith('.debug_'): - # Skip debugging sections (to reduce parsing time) - state = None - continue - state = 'reloc' - relocsection = sectionmap[sectionname] - continue + i = 0 + while not lines[i].startswith('Section Headers'): + i += 1 + i += 3 # Skip [Nr] and [ 0] + while ']' in lines[i]: + parts = lines[i][lines[i].find(']')+1:].split() + if len(parts) == 10: + name, _, _, _, size, _, _, _, _, align = parts + else: # not SHF_ALLOC + name, _, _, _, size, _, _, _, align = parts + section = Section() + section.name = name + section.size = int(size, 16) + section.align = int(align) + section.fileid = fileid + section.relocs = [] + sections.append(section) + sectionmap[name] = section + i += 1 - if state == 'section': - try: - idx, name, size, vma, lma, fileoff, align = line.split() - if align[:3] != '2**': - continue - section = Section() - section.name = name - section.size = int(size, 16) - section.align = 2**int(align[3:]) - section.fileid = fileid - section.relocs = [] - sections.append(section) - sectionmap[name] = section - except ValueError: - pass + while not lines[i].startswith('Symbol table '): + i += 1 + i += 3 # Skip Num: and 0: + while i < len(lines) and ':' in lines[i]: + colon = lines[i].find(':') + parts = lines[i].split() + try: + ndx = int(parts[6]) + name = sections[ndx-1].name + except ValueError: # ABS or UND + ndx = 0 + name = '' + if parts[3] != 'SECTION': + name = parts[7] + symbol = Symbol() + symbol.size = int(parts[2]) + symbol.offset = int(parts[1], 16) + symbol.name = name + symbol.section = sections[ndx-1] if ndx > 0 else None + symbols[name] = symbol + i += 1 + + i = 0 + while i < len(lines): + if not lines[i].startswith('Relocation section'): + i += 1 continue - if state == 'symbol': - try: - parts = line[17:].split() - if len(parts) == 3: - sectionname, size, name = parts - elif len(parts) == 4 and parts[2] == '.hidden': - sectionname, size, hidden, name = parts - else: - continue - symbol = Symbol() - symbol.size = int(size, 16) - symbol.offset = int(line[:8], 16) - symbol.name = name - symbol.section = sectionmap.get(sectionname) - symbols[name] = symbol - except ValueError: - pass + rel_name = lines[i][20:lines[i].index("'", 20)] + i += 2 + if rel_name.startswith('.rel.debug_'): + # Skip debugging sections (to reduce parsing time) continue - if state == 'reloc': - try: - off, type, symbolname = line.split() - reloc = Reloc() - reloc.offset = int(off, 16) - reloc.type = type - reloc.symbolname = symbolname - reloc.symbol = symbols.get(symbolname) - if reloc.symbol is None: - # Some binutils (2.20.1) give section name instead - # of a symbol - create a dummy symbol. - reloc.symbol = symbol = Symbol() - symbol.size = 0 - symbol.offset = 0 - symbol.name = symbolname - symbol.section = sectionmap.get(symbolname) - symbols[symbolname] = symbol - relocsection.relocs.append(reloc) - except ValueError: - pass + assert rel_name.startswith('.rel') + relocated = sectionmap[rel_name[4:]] + while lines[i][0].isalnum(): + line = lines[i] + off, _, typ, _, symbolname = line.split() + reloc = Reloc() + reloc.offset = int(off, 16) + reloc.type = typ + reloc.symbolname = symbolname + reloc.symbol = symbols.get(symbolname) + assert reloc.symbol is not None + relocated.relocs.append(reloc) + i += 1 + return sections, symbols # Parser for constants in simple C header files. @@ -644,15 +637,15 @@ def main(): # Get output name in16, in32seg, in32flat, cfgfile, out16, out32seg, out32flat = sys.argv[1:] - # Read in the objdump information + # Read in the readelf information infile16 = open(in16, 'r') infile32seg = open(in32seg, 'r') infile32flat = open(in32flat, 'r') # infoX = (sections, symbols) - info16 = parseObjDump(infile16, '16') - info32seg = parseObjDump(infile32seg, '32seg') - info32flat = parseObjDump(infile32flat, '32flat') + info16 = parseReadElf(infile16, '16') + info32seg = parseReadElf(infile32seg, '32seg') + info32flat = parseReadElf(infile32flat, '32flat') # Read kconfig config file config = scanconfig(cfgfile) -- 2.37.0.144.g8ac04bfd2-goog

2 2

Force 1024x768 resolution on SeaBIOS with coreboot native graphics
by Felix Freeman 01 Jul '22

01 Jul '22

Hello, I have noticed several problems when using coreboot native graphics ("high resolution" linear framebuffer) with a resolution of 1024x768, on SeaBIOS + SeaVGABIOS: - Boot menus (GRUB, Syslinux) look at a resolution of 640x480. - I get a warning in Debian "graphical" installer "Undefined video mode number: 314", asking me to select a mode, if I select a resolution of 1024x768 it boots fine, if I don't, after 30 seconds, it crashes. - When I use iPXE with SeaBIOS I can't get video on "text based installers" (which use ncurses, for example Debian text based installers). However: - In iPXE I get video on other "graphical" installers (e.g. Debian graphical installer). - I have noticed that in "video mode" it is able to change resolution to show a bootsplash (mode 0x144 on vgasrc/svgamodes.c) and then goes back to 640x480 on "text console". - In iPXE the text display resolution is 1024x7480, however the menus are 640x480. I don't know if its implementation is incomplete, or because of technical reasons SeaVGABIOS is unable to behave correctly without a proprietary VGA BIOS. I think a first step might be to be able to force a minimum resolution in text and video modes. So, my first question is: what is needed to force SeaBIOS/SeaVGABIOS to display a resolution of 1024x768 instead of 640x480 when in "text mode" and "video mode"? And my second question is whether with some work the coreboot native graphics implementation could behave as good as (or close to) a proprietary VGA BIOS. Best regards, Felix

2 3

[SeaVGABIOS] No VBE modes for card on PCIe Root Port
by Kevin Locke 09 Jun '22

09 Jun '22

Hi All, With QEMU 7.0.0 and SeaBIOS 1.16.0, virtual machines with a PCIe graphics card connected to a PCIe Root Port, rather than the Root Complex, as recommended by the QEMU PCI Express Guidelines[1], are unable to use VESA BIOS Extensions display modes. For example, qemu-system-x86_64 \ -no-user-config \ -nodefaults \ -machine q35,accel=kvm \ -m 1G \ -cdrom "$iso" \ -device VGA works well and can use many VBE display modes, while qemu-system-x86_64 \ -no-user-config \ -nodefaults \ -machine q35,accel=kvm \ -m 1G \ -cdrom "$iso" \ -device pcie-root-port,id=pci.1,bus=pcie.0 \ -device VGA,bus=pci.1 only shows a few text display modes. This is the case for VGA, qxl-vga, and virtio-vga devices. `info qtree` shows the device as `type: PCIE` in both configurations. High-resolution graphics modes are available when using UEFI (OVMF) in both configurations. Am I correct in assuming this is a limitation of SeaVGABIOS? Is this expected behavior, or a bug? If it's unlikely to change soon, would it make sense for me to propose documenting this limitation in the QEMU PCI Express Guidelines? I'd appreciate any insights you might have. Thanks, Kevin [1]: https://github.com/qemu/qemu/blob/v7.0.0/docs/pcie.txt

2 2

How does call16 work from protected mode
by Himanshu Chauhan 07 Jun '22

07 Jun '22

Hi, I am trying to understand the call of 16-bit code from protected mode. I am tracing the code from "enable_vga_console" function. I see that it further calls "call16_int10". Further down somewhere (probably check_irqs), pause instruction is executed. I am trying to understand what is expected here. I assume pause is executed and SeaBIOS waits for an interrupt to arrive to finish the call16 processing. But I don't see RTC being programmed for interrupt. For which interrupt is SeaBIOS waiting? NMI? Regards Himanshu

1 0

[PATCH 1/3] handle_1591: [INT15/AH=91h] Always set AH=0 and clear CF
by Lev Kujawski 02 Jun '22

02 Jun '22

References: R. Brown, "INT 15 - OS HOOK - DEVICE POST (AT,PS)", in The x86 Interrupt List. https://www.cs.cmu.edu/~ralf/files.html [Updated: 2018-02-28]. F. v. Gilluwe, The Undocumented PC. Boston, MA: Addison-Wesley, 1997, p. 780. Signed-off-by: Lev Kujawski <lkujaw(a)member.fsf.org> --- src/system.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/system.c b/src/system.c index 438e60e..ec99347 100644 --- a/src/system.c +++ b/src/system.c @@ -232,6 +232,7 @@ handle_1590(struct bregs *regs) static void handle_1591(struct bregs *regs) { + set_code_success(regs); } // keyboard intercept -- 2.34.1

2 5

Open-Source Firmware Conference 2022
by Christian Walter 02 Jun '22

02 Jun '22

** *Dear SeaBIOS community,* * the OSFC is the biggest open-source firmware conference worldwide made by and for developers. After two years we can finally meet in person again. This year's OSFC will take place in Gothenburg, Sweden from September 19th - 21th. We will have a two day conference, followed by a one-day hackathon where we can hack together on hardware and firmware. Currently the Call for Participation is open, and I’d like to invite everyone in this community to submit talks and present new innovations, updates or problem talks. Additionally, you can already buy early bird tickets, so feel free to check out the homepage https://www.osfc.io <https://www.osfc.io>for more information. SeaBIOS and it's community has been a constant part of open-source firmware - and we'd love to see contributions from you on the OSFC 2022! If you have any questions feel free to drop me an email! Thank you! Chris * -- *Christian Walter* *Head of Firmware Development / Cyber Security * 9elements GmbH, Kortumstraße 19-21, 44787 Bochum, Germany Email: christian.walter(a)9elements.com Phone: _+49 234 68 94 188 <tel:+492346894188>_ Mobile: _+49 176 70845047 <tel:+4917670845047>_ Sitz der Gesellschaft: Bochum Handelsregister: Amtsgericht Bochum, HRB 17519 Geschäftsführung: Sebastian Deutsch, Eray Basar Datenschutzhinweise nach Art. 13 DSGVO <https://9elements.com/privacy>

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

SeaBIOS