SeaBIOS September 2016

seabios@seabios.org

8 participants
12 discussions

[PATCH 0/6] Misc tpm updates
by Kevin O'Connor 02 Dec '16

02 Dec '16

Hi Stefan, I dusted off some TPM patches I had worked on earlier in the year. Mostly just code movement, but the last patch I think helps simplify tpm_build_and_send_cmd(). Do they look okay to you? -Kevin Kevin O'Connor (6): tpm: Add comment banners to tcg.c separating major parts of spec tpm: Don't call tpm_set_failure() from tpm12_get_capability() tpm: Move code around in tcgbios.c to keep like code together acpi: Generalize find_fadt() and find_tcpa_by_rsdp() into find_acpi_table() tpm: Don't call tpm_build_and_send_cmd() from tpm20_stirrandom() tpm: Rework tpm_build_and_send_cmd() into tpm_simple_cmd() src/fw/biostables.c | 18 +- src/std/acpi.h | 7 - src/std/tcg.h | 199 +++++++++-------- src/tcgbios.c | 611 ++++++++++++++++++++++------------------------------ src/util.h | 1 + 5 files changed, 369 insertions(+), 467 deletions(-) -- 2.5.5

2 8

[PATCH 0/7] serial console support
by Gerd Hoffmann 19 Oct '16

19 Oct '16

Hi, After a looong break finally the next round of the seabios serial console patches. cheers, Gerd Gerd Hoffmann (7): std: add cp437 to unicode map kbd: make enqueue_key public, add ascii_to_keycode paravirt: read QEMU_CFG_NOGRAPHIC, store in etc/sercon-enable romfile add serial console support sercon: split-output implementation sercon: add compile time option [hack] ignore sgabios, enable sercon instead Makefile | 3 +- src/Kconfig | 5 + src/clock.c | 1 + src/cp437.c | 275 +++++++++++++++++++++++ src/cp437.h | 1 + src/fw/paravirt.c | 2 + src/kbd.c | 18 +- src/misc.c | 2 + src/optionroms.c | 17 +- src/romlayout.S | 14 ++ src/sercon.c | 660 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ src/util.h | 5 + 12 files changed, 999 insertions(+), 4 deletions(-) create mode 100644 src/cp437.c create mode 100644 src/cp437.h create mode 100644 src/sercon.c -- 1.8.3.1

2 8

[PATCH 0/5] serial console support
by Gerd Hoffmann 13 Oct '16

13 Oct '16

Hi, Next round of serial console patches, Still not ready to merge. Headline feature of this version is split mode (i.e. output on both vga and serial). Have a initial working implementation now, logic is simliar to the vgabios timer hook (as suggested by Kevin). Also changed the output logic, down to a one-cell buffer (char+attr) for output. Cursor updates are simply written to BDA now, with the actual move being done lazily when printing the next character or checking for keyboard input. That is especially useful for splitmode as we can simply skip the cursor position updates and let the vgabios do them instead. TODO list: * compile time (CONFIG_*) option. * input handling cleanups. * implement missing int10h functions. * more testing. cheers, Gerd Gerd Hoffmann (5): std: add cp437 to unicode map kbd: make enqueue_key public, add ascii_to_keycode paravirt: read QEMU_CFG_NOGRAPHIC, store in etc/sercon-enable romfile add serial console support [wip] sercon: initial split-output implementation Makefile | 7 +- src/clock.c | 1 + src/fw/paravirt.c | 2 + src/kbd.c | 17 +- src/misc.c | 2 + src/optionroms.c | 9 +- src/romlayout.S | 39 ++++ src/sercon.c | 621 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ src/std/cp437.c | 275 ++++++++++++++++++++++++ src/std/cp437.h | 1 + src/util.h | 5 + 11 files changed, 974 insertions(+), 5 deletions(-) create mode 100644 src/sercon.c create mode 100644 src/std/cp437.c create mode 100644 src/std/cp437.h -- 1.8.3.1

3 16

[PATCH v4 0/5] support booting more than 255 CPUs with QEMU
by Igor Mammedov 16 Sep '16

16 Sep '16

Changelog since: v3: * don't move CPU counting around, leave it as is and consolidate apic_id accounting only * add and use qemu_*_present_cpus_count() helpers to init/get boot cpus count at boot time and only qemu_get_present_cpus_count() at resume time * fix S3 wakeup stuck in resume vector * If cpu were hotplugged during smp_scan, report it to user instead of silently hanging if it came online while smp_scan() waits for APs v2: * rebase on top of current master /smp_scan() changes/ v1: * s/count_cpu/apic_id_init/ * merge handle_x2apic() into apic_id_init() RFC: * move out max-cpus check out of mptable_setup() * factor out CPU counting/apic ID detection in separate function * return back accidentially deleted debug message with APIC ID * drop unused code in smp_setup() According to SDM, if CPUs have APIC ID more than 254 firmware should pass control to OS in x2APIC mode. This series adds x2APIC bootstrap initialization. Works fine with not yet rebased on master QEMU side of x2APIC support: https://www.mail-archive.com/qemu-devel@nongnu.org/msg390671.html Rebased variant probably will be sent next week. Note: S3 wakeup works as expected if linux guest is running with IRQ remapping enabled. However it seems that kernel is buggy with IRQ remapping disabled as kernel disables CPUs with APIC ID > 254 and on resume from S3 hangs somewere after getting control from Seabios. Igor Mammedov (4): paravirt: disable legacy bios tables in case of more than 255 CPUs error out if present cpus count changed during SMP bringup add helpers to read etc/boot-cpus at resume time support booting with more than 255 CPUs Kevin O'Connor (1): smp: consolidate CPU APIC ID detection and accounting src/fw/paravirt.h | 3 +++ src/x86.h | 1 + src/fw/paravirt.c | 44 +++++++++++++++++++++++++++++++++++-- src/fw/smp.c | 65 ++++++++++++++++++++++++++++++++++++++----------------- 4 files changed, 91 insertions(+), 22 deletions(-) -- 2.7.4

2 7

[PATCH] boot: Show boot splash until all threads are complete
by Paul Menzel 16 Sep '16

16 Sep '16

[Please find the unmangled patch attached. I have no idea how to turn off line wrapping in Mozilla Thunderbird.] Currently, even with `etc/threads` set to 2 in CBFS, the boot splash is only shown for a short period, while SeaBIOS detects the hard disk on the ASRock E350M1, which takes 10 seconds with this particular hard drive model. So move the call to wait for the threads to finish further up, so it is before the instruction to disable the boot splash. On the ASRock E350M1 the boot splash is shown the whole time now. Note, sometimes the GRUB menu from the MBR is not shown, and boot splash can still be seen. The screen only changes, once the Linux kernel boots. This seems to be a GRUB issue though [1]. [1] http://metadata.ftp-master.debian.org/changelogs//main/g/grub2/grub2_2.02~b… Suggested-by: Kevin O'Connor <kevin(a)koconnor.net> Signed-off-by: Paul Menzel <pmenzel(a)molgen.mpg.de> --- src/boot.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/boot.c b/src/boot.c index 706b7df..af3c627 100644 --- a/src/boot.c +++ b/src/boot.c @@ -468,6 +468,7 @@ interactive_bootmenu(void) u32 menutime = romfile_loadint("etc/boot-menu-wait", DEFAULT_BOOTMENU_WAIT); enable_bootsplash(); int scan_code = get_keystroke(menutime); + wait_threads(); disable_bootsplash(); if (scan_code != menukey) return; @@ -476,7 +477,6 @@ interactive_bootmenu(void) ; printf("Select boot device:\n\n"); - wait_threads(); // Show menu items int maxmenu = 0; -- 2.4.1

2 1

[PATCH] fix virtio-pci
by Gerd Hoffmann 16 Sep '16

16 Sep '16

virtio-pci calls pci_enable_{io,mem}bar with the bar number, but the functions expect the bar base register offset. Reported-by: Igor Mammedov <imammedo(a)redhat.com> Signed-off-by: Gerd Hoffmann <kraxel(a)redhat.com> --- src/hw/virtio-pci.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/hw/virtio-pci.c b/src/hw/virtio-pci.c index 8efc826..e5c2c33 100644 --- a/src/hw/virtio-pci.c +++ b/src/hw/virtio-pci.c @@ -449,7 +449,7 @@ void vp_init_simple(struct vp_device *vp, struct pci_device *pci) switch (vp_cap->mode) { case VP_ACCESS_IO: { - u32 addr = pci_enable_iobar(pci, vp_cap->bar); + u32 addr = pci_enable_iobar(pci, base); if (!addr) return; vp_cap->ioaddr = addr + offset; @@ -458,7 +458,7 @@ void vp_init_simple(struct vp_device *vp, struct pci_device *pci) } case VP_ACCESS_MMIO: { - void *addr = pci_enable_membar(pci, vp_cap->bar); + void *addr = pci_enable_membar(pci, base); if (!addr) return; vp_cap->memaddr = addr + offset; -- 1.8.3.1

3 2

[PATCH] kbd: Move extended and release events out of special key detection switch
by Kevin O'Connor 16 Sep '16

16 Sep '16

Move checking for release and extended scancodes to the top of __process_key(). Signed-off-by: Kevin O'Connor <kevin(a)koconnor.net> --- src/kbd.c | 70 +++++++++++++++++++++++++++------------------------------------ 1 file changed, 30 insertions(+), 40 deletions(-) diff --git a/src/kbd.c b/src/kbd.c index 5945e99..31e47e3 100644 --- a/src/kbd.c +++ b/src/kbd.c @@ -439,73 +439,65 @@ kbd_prtscr(int key_release) static void __process_key(u8 scancode) { - // Check for multi-key sequences + // Check for multi-scancode key sequences u8 flags1 = GET_BDA(kbd_flag1); - if (flags1 & KF1_LAST_E1) { - // Part of "pause" key (sequence is e1 1d 45 e1 9d c5) - if ((scancode & ~0x80) == 0x1d) - // Second key of sequence - ignore + int key_release = scancode & 0x80; + if (key_release) { + if (scancode == 0xe0) { + // Extended key + SET_BDA(kbd_flag1, flags1 | KF1_LAST_E0); + return; + } else if (scancode == 0xe1) { + // Start of pause key sequence + SET_BDA(kbd_flag1, flags1 | KF1_LAST_E1); + return; + } + scancode &= ~0x80; + } + if (flags1 & (KF1_LAST_E0|KF1_LAST_E1)) { + if (flags1 & KF1_LAST_E1 && scancode == 0x1d) + // Ignore second scancode of pause key (e1 1d 45 e1 9d c5) return; - // Third key of sequence - clear flag for next key - SET_BDA(kbd_flag1, flags1 & ~KF1_LAST_E1); + // Clear E0/E1 flag in memory for next key event + SET_BDA(kbd_flag1, flags1 & ~(KF1_LAST_E0|KF1_LAST_E1)); } - if (flags1 & KF1_LAST_E0) - // Clear E0 flag in memory for next key event - SET_BDA(kbd_flag1, flags1 & ~KF1_LAST_E0); // Check for special keys - int key_release = scancode & 0x80; switch (scancode) { - case 0xe0: - // Extended key - SET_BDA(kbd_flag1, flags1 | KF1_LAST_E0); - return; - case 0xe1: - // Start of pause key sequence - SET_BDA(kbd_flag1, flags1 | KF1_LAST_E1); - return; - - case 0x3a: /* Caps Lock press */ - case 0xba: /* Caps Lock release */ + case 0x3a: /* Caps Lock */ kbd_set_flag(key_release, KF0_CAPS, 0, KF0_CAPSACTIVE); return; - case 0x2a: /* L Shift press */ - case 0xaa: /* L Shift release */ + case 0x2a: /* L Shift */ if (flags1 & KF1_LAST_E0) // Ignore fake shifts return; kbd_set_flag(key_release, KF0_LSHIFT, 0, 0); return; - case 0x36: /* R Shift press */ - case 0xb6: /* R Shift release */ + case 0x36: /* R Shift */ if (flags1 & KF1_LAST_E0) // Ignore fake shifts return; kbd_set_flag(key_release, KF0_RSHIFT, 0, 0); return; - case 0x1d: /* Ctrl press */ - case 0x9d: /* Ctrl release */ + case 0x1d: /* Ctrl */ if (flags1 & KF1_LAST_E0) kbd_set_flag(key_release, KF0_CTRLACTIVE, KF1_RCTRL, 0); else kbd_set_flag(key_release, KF0_CTRLACTIVE | KF0_LCTRL, 0, 0); return; - case 0x38: /* Alt press */ - case 0xb8: /* Alt release */ + case 0x38: /* Alt */ if (flags1 & KF1_LAST_E0) kbd_set_flag(key_release, KF0_ALTACTIVE, KF1_RALT, 0); else kbd_set_flag(key_release, KF0_ALTACTIVE | KF0_LALT, 0, 0); return; - case 0x45: /* Num Lock press */ - case 0xc5: /* Num Lock release */ + case 0x45: /* Num Lock */ if (flags1 & KF1_LAST_E1) // XXX - pause key. return; kbd_set_flag(key_release, KF0_NUM, 0, KF0_NUMACTIVE); return; - case 0x46: /* Scroll Lock press */ - case 0xc6: /* Scroll Lock release */ + case 0x46: /* Scroll Lock */ if (flags1 & KF1_LAST_E0) { kbd_ctrl_break(key_release); return; @@ -513,20 +505,18 @@ __process_key(u8 scancode) kbd_set_flag(key_release, KF0_SCROLL, 0, KF0_SCROLLACTIVE); return; - case 0x37: - case 0xb7: + case 0x37: /* * */ if (flags1 & KF1_LAST_E0) { kbd_prtscr(key_release); return; } break; - case 0x54: - case 0xd4: + case 0x54: /* SysReq */ kbd_sysreq(key_release); return; - case 0x53: + case 0x53: /* Del */ if ((GET_BDA(kbd_flag0) & (KF0_CTRLACTIVE|KF0_ALTACTIVE)) - == (KF0_CTRLACTIVE|KF0_ALTACTIVE)) { + == (KF0_CTRLACTIVE|KF0_ALTACTIVE) && !key_release) { // Ctrl+alt+del - reset machine. SET_BDA(soft_reset_flag, 0x1234); reset(); -- 2.5.5

1 1

guest is unable to boot from virtio drive
by Igor Mammedov 16 Sep '16

16 Sep '16

This happens only when combination of build options CONFIG_RELOCATE_INIT=n + CONFIG_DEBUG_LEVEL=3 to reproduce compile seabios defconfig + above options and start guest with a single virtio drive: -drive if=virtio,file=guest.img As result one gets: No bootable device. here is debug log: [...] init virtio-blk /3ffba000\ Start thread |3ffba000| found virtio-blk at 00:04.0 |3ffba000| pci dev 0:4 virtio cap at 0x84 type 5 [pci cfg access] |3ffba000| WARNING - internal error detected at pci_enable_membar:197! |3ffba000| ERROR: queue size is 0 |3ffba000| fail to find vq for virtio-blk 00:04.0 \3ffba000/ End thread init virtio-scsi init lsi53c895a init esp init megasas init pvscsi init MPT init lpt Found 1 lpt ports init serial Found 1 serial ports |3ffbc000| PS2 keyboard initialized \3ffbc000/ End thread |3ffbb000| DVD/CD [ata1-0: QEMU DVD-ROM ATAPI-4 DVD/CD] |3ffbb000| Searching bootorder for: /pci@i0cf8/*@1,1/drive@1/disk@0 |3ffbb000| Registering bootable: DVD/CD [ata1-0: QEMU DVD-ROM ATAPI-4 DVD/CD] (type:3 prio:103 data:f5a70) \3ffbb000/ End thread All threads complete. Scan for option roms WARNING - Unable to allocate resource at copy_rom:234! Searching bootorder for: /rom(a)genroms/kvmvapic.bin Registering bootable: Legacy option rom (type:129 prio:101 data:c9800003) Searching bootorder for: HALT Running option rom at c980:0003 Mapping floppy drive 0x000f5ae0 Mapping cd drive 0x000f5a70 finalize PMM malloc finalize Space available for UMB: cc000-d7800, f5660-f5a40 Returned 131072 bytes of ZoneHigh e820 map has 7 items: 0: 0000000000000000 - 000000000009fc00 = 1 RAM 1: 000000000009fc00 - 00000000000a0000 = 2 RESERVED 2: 00000000000f0000 - 0000000000100000 = 2 RESERVED 3: 0000000000100000 - 000000003ffe0000 = 1 RAM 4: 000000003ffe0000 - 0000000040000000 = 2 RESERVED 5: 00000000feffc000 - 00000000ff000000 = 2 RESERVED 6: 00000000fffc0000 - 0000000100000000 = 2 RESERVED locking shadow ram Jump to int19 enter handle_19: NULL Booting from Hard Disk... invalid handle_legacy_disk:729: a=00000201 b=00000000 c=00000001 d=00000080 ds=0000 es=07c0 ss=c900 si=00000000 di=00000000 bp=00000000 sp=0000fcde cs=f000 ip=ceef f=0202 Boot failed: could not read the boot disk enter handle_18: NULL Booting from Floppy... Floppy_drive_recal 0 Floppy_enable_controller Floppy_media_sense on drive 0 found rate 0 floppy error: 40 00 00 00 00 01 02 invalid basic_access:158: a=00000200 b=00000000 c=00000001 d=00000000 ds=0000 es=07c0 ss=c900 si=00000000 di=00000000 bp=00000000 sp=0000fcde cs=f000 ip=ceef f=0202 Boot failed: could not read the boot disk enter handle_18: NULL Booting from DVD/CD... Device reports MEDIUM NOT PRESENT scsi_is_ready returned -1 Boot failed: Could not read from CDROM (code 0003) enter handle_18: NULL No bootable device.

1 1

TPM implementation
by Wim Vervoorn 16 Sep '16

16 Sep '16

Hello Kevin/Stefan, I can see a lot of work has been done already on the tpm implementation in SeaBIOS. Can you update me on the status of this? Should I still consider this to be work in progress or is this close to a stage where it can actually be used in a production level project? Best Regards, Wim Vervoorn Eltan B.V. Ambachtstraat 23 5481 SM Schijndel The Netherlands T : +31-(0)73-594 46 64 E : wvervoorn(a)eltan.com W : http://www.eltan.com "THIS MESSAGE CONTAINS CONFIDENTIAL INFORMATION. UNLESS YOU ARE THE INTENDED RECIPIENT OF THIS MESSAGE, ANY USE OF THIS MESSAGE IS STRICTLY PROHIBITED. IF YOU HAVE RECEIVED THIS MESSAGE IN ERROR, PLEASE IMMEDIATELY NOTIFY THE SENDER BY TELEPHONE +31-(0)73-5944664 OR REPLY EMAIL, AND IMMEDIATELY DELETE THIS MESSAGE AND ALL COPIES."

1 0

Re: [SeaBIOS] [Qemu-devel] [RFC PATCH v2 00/12] Guest startup time optimization
by Gerd Hoffmann 12 Sep '16

12 Sep '16

Hi, > > AFAIK latest seabios already supports DMA. > > It's easy to add more config options for seabios > > if you are so inclined. > > Yes, I tried that, the fw_cfg overhead in SeaBios/linuxboot optrom is > already not a big issue for us. However, there are still some other > code in SeaBIOS needs to be investigated. What exactly? Also a few suggestions: * Try the seabios master branch. There have been some improvements recently which are not in the 1.9 release. * Use a stripped down config. When booting a kernel directly you don't need storage and usb drivers, also no keyboard. That all reduces init time. * Initializing all pci devices (placing bars in address space and programming them) can probably be skipped and left to the linux kernel to handle. But when trying take care that the effect might be that we shift just startup time from seabios to linux kernel, the placement needs to happen somewhere after all. * When discussing these things add the seabios list (seabios(a)seabios.org) to cc: Added the list now. cheers, Gerd

2 1

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

SeaBIOS September 2016