March 2018 - SeaBIOS - mail.coreboot.org

K-TEK-M275TP-FN-BL-ML USB keyboard support
by Аладышев Константин April 28, 2018

April 28, 2018

Hello! I'm having problems with K-TEK-M275TP-FN-BL-ML keyboard (http://www.key-tek.cn/Productview.asp?id=777) It is USB device with keyboard and touchpad combined. It works in OS correctly, but doesn't work in SeaBIOS. I've tried to debug this issue and found that "usb_kbd_setup" function fails on "(epdesc->wMaxPacketSize != 8)" check, cause "epdesc->wMaxPacketSize" is actually 32. I've modified this "if" statement to pass 32 value also and now function completes successfully. But it didn't help to solve my issue. There are no interrupts from keyboard, "ehci_poll_intr" always returns -1 on check: if (token & QTD_STS_ACTIVE) { // No intrs found. return -1; } I don't know what to do now. What should I check next? Does SeaBIOS support USB keyboard+touchpad devices at all?

2 5

[PATCH] tpm: Request access to locality 0
by Stefan Berger April 12, 2018

April 12, 2018

The CRB driver has to request access to locality 0. Our QEMU CRB model will enforce that as well. Signed-off-by: Stefan Berger <stefanb(a)linux.vnet.ibm.com> --- src/hw/tpm_drivers.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/hw/tpm_drivers.c b/src/hw/tpm_drivers.c index bd971f7..e4770b3 100644 --- a/src/hw/tpm_drivers.c +++ b/src/hw/tpm_drivers.c @@ -442,6 +442,8 @@ static u32 crb_activate(u8 locty) if (!CONFIG_TCGBIOS) return 0; + writeb(CRB_REG(locty, CRB_REG_LOC_CTRL), 1); + return 0; } -- 2.5.5

2 2

[PATCH] tpm: Disable TCGBIOS by default
by Jean Lucas April 1, 2018

April 1, 2018

When TCGBIOS is enabled on a machine without a physical TPM, SeaBIOS takes over half a minute to print to the display, and over another half minute before handing control over to a boot device. Disable this option by default to have it be an explicit choice. Tested on a Dell Inspiron 660s (no TPM) configured with libgfxinit, and, separately, configured with a PCIe video card with onboard VGA BIOS. Signed-off-by: Jean Lucas <jean(a)4ray.co> --- src/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Kconfig b/src/Kconfig index 55a87cb..b5d370a 100644 --- a/src/Kconfig +++ b/src/Kconfig @@ -467,7 +467,7 @@ menu "BIOS interfaces" config TCGBIOS depends on S3_RESUME bool "TPM support and TCG BIOS extensions" - default y + default n help Provide TPM support along with TCG BIOS extensions -- 2.16.2

2 2

seabios regression in 1.11.1
by Olaf Hering March 29, 2018

March 29, 2018

Recently both qemu.git and xen.git updated their copy of seabios. Qemu is broken quite some time already, now also xen fails after updating from rel-1.10.2 to rel-1.11.1. This happens in SLE11 which uses gcc-4.3: [ 86s] Compile checking out/src/hw/blockcmd.o [ 86s] src/hw/blockcmd.c: In function 'scsi_rep_luns_scan': [ 86s] src/hw/blockcmd.c:229: error: unknown field 'cdbcmd' specified in initializer [ 86s] src/hw/blockcmd.c:229: warning: missing braces around initializer [ 86s] src/hw/blockcmd.c:229: warning: (near initialization for 'op.<anonymous>') [ 86s] src/hw/blockcmd.c:229: warning: initialization makes integer from pointer without a cast [ 86s] make[7]: *** [out/src/hw/blockcmd.o] Error 1 How can this be fixed? Olaf

2 1

Re: [SeaBIOS] seabios hangs when injecting an MSI-X interrupt in early booting.
by Kevin O'Connor March 27, 2018

March 27, 2018

On Tue, Mar 27, 2018 at 10:08:20AM +0800, Zhe Liu wrote: > Dear Kevin, > Thanks for getting back to me. Your replay helps me a lot. And there > are two things I want to make clear. > 1. > The real-mode handlers installed at 0x60-0x66 are intended for > software interrupts. > During the booting of the vm seabios, which modules will raise the software > interrupts? (Qemu? KVM? VirtualMachine?) Software interrupts are for guest software to raise and interally catch. Some old DOS-era software does this. > 2. In src/post.c: init_ivt, the handlers for 0x60-0x66 is set to none > as below. > for (i=0x60; i<=0x66; i++) > SET_IVT(i, SEGOFF(0, 0)); > If I change the handlers of 0x60-0x66 to default handler, which is > "iret" directly, is there some side effects? It might break some old DOS-era software. Notably, though, if something external is raising random interrupts, it's likely going to cause havoc with both modern and old software. -Kevin

1 0

[PATCH v2 0/3] Fix CRB probing to work with real hardware
by Stefan Berger March 26, 2018

March 26, 2018

This series of patches attempts to fix the probing of the CRB interface for real hardware. Stephen Douthit tested the changes on real hardware and it seems to fix an existing problem. Stefan v1->v2: - test LOC_STATE register for set tpmRegValidSts flag and unset locAssigned flag without writing to any other registers Stefan Berger (3): tpm: Wait for tpmRegValidSts flag on CRQ interface before probing tpm: revert return values for successful/failed CRB probing tpm: when CRB is active, select, lock it, and check addresses src/hw/tpm_drivers.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) -- 2.5.5

4 17

Re: [SeaBIOS] seabios hangs when injecting an MSI-X interrupt in early booting.
by Kevin O'Connor March 26, 2018

March 26, 2018

On Thu, Mar 22, 2018 at 11:40:46AM +0800, Zhe Liu wrote: > Hi! > Recently, I encountered a seabios hanging problem. When I try to inject an > MSI-X interrupt along with SMI in qemu like below: > *************** patch below********************* > diff --git a/target-i386/kvm.c b/target-i386/kvm.c- a/target-i386/kvm.c > > > --- a/target-i386/kvm.c > +++ b/target-i386/kvm.c > > @@ -104,6 +104,7 @@ static uint32_t num_architectural_pmu_counters; > static int has_xsave; > static int has_xcrs; > static int has_pit_state2; > +static int msi_count; > > static bool has_msr_mcg_ext_ctl; > > @@ -2870,12 +2871, 26 @@ void kvm_arch_pre_run(CPUState *cpu, struct > kvm_run *run) > DPRINTF("inject SMI\n"); > + MSIMessage msg = {.address = 0x0, .data = 0x4062}; > + if (msi_count == 1) { > + ret = kvm_irqchip_send_msi(kvm_state, msg); > + if (ret <0) { > + QEMU_LOG(LOG_ALERT, "MSI lost %s\n", strerror(-ret)); > + } > + } > + msi_count++; > > ret = kvm_vcpu_ioctl(cpu, KVM_SMI); > if (ret < 0) > **************patch above************************* > > Then, when the vm will hang when booting. > There will be only one line in the vnc displayer: > "Guest has not initialized the display(yet)" > > I went through the codes(kvm qemu seabios), and I found the check_irqs > function in seabios will allow interrupt for a while. > When the interrupt triggered, the vcpu will lookup the seabios IDT table > and jump to the interrupt handler. > > However, in ivt_init(void), the entries from 0x60 to 0x66 will be cleared, > which is configured by the patch below. > > https://github.com/coreboot/seabios/commit/b164d2c1b8ff2dd764dcf064e2624dd4… > > So the vcpu will jump to an undefined entry instead of the default entry. > > ========================================================================= > I removed the patch, and the bug seems to disappear. I wonder if this is an > approprite solution. This doesn't seem like a SeaBIOS issue. The real-mode handlers installed at 0x60-0x66 are intended for software interrupts; not for hardware raised interrupts. I don't know why the machine irq routing would deliver a hardware interrupt to that vector, but I suspect that is what needs to be looked at. -Kevin

3 2

[PATCH] ps2port: Enable keyboard write for i8042 before reset
by Jean Lucas March 22, 2018

March 22, 2018

This enables the keyboard on some Chromebooks after a hard reset. Tested on an Acer C720 and a Dell Inspiron 660s. Signed-off-by: Jean Lucas <jean(a)4ray.co> --- src/hw/ps2port.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/src/hw/ps2port.c b/src/hw/ps2port.c index 88b1059..ac409de 100644 --- a/src/hw/ps2port.c +++ b/src/hw/ps2port.c @@ -475,6 +475,11 @@ ps2_keyboard_setup(void *data) return; } + // Keyboard write command + ret = i8042_command(I8042_CMD_CTL_WCTR, NULL); + if (ret) + return; + // Controller keyboard test. ret = i8042_command(I8042_CMD_KBD_TEST, param); if (ret) @@ -505,11 +510,6 @@ ps2_keyboard_setup(void *data) return; } - /* Disable keyboard */ - ret = ps2_kbd_command(ATKBD_CMD_RESET_DIS, NULL); - if (ret) - return; - // Set scancode command (mode 2) param[0] = 0x02; ret = ps2_kbd_command(ATKBD_CMD_SSCANSET, param); -- 2.16.2

2 1

seabios hangs when injecting an MSI-X interrupt in early booting.
by Zhe Liu March 21, 2018

March 21, 2018

Hi! Recently, I encountered a seabios hanging problem. When I try to inject an MSI-X interrupt along with SMI in qemu like below: *************** patch below********************* diff --git a/target-i386/kvm.c b/target-i386/kvm.c- a/target-i386/kvm.c --- a/target-i386/kvm.c +++ b/target-i386/kvm.c @@ -104,6 +104,7 @@ static uint32_t num_architectural_pmu_counters; static int has_xsave; static int has_xcrs; static int has_pit_state2; +static int msi_count; static bool has_msr_mcg_ext_ctl; @@ -2870,12 +2871, 26 @@ void kvm_arch_pre_run(CPUState *cpu, struct kvm_run *run) DPRINTF("inject SMI\n"); + MSIMessage msg = {.address = 0x0, .data = 0x4062}; + if (msi_count == 1) { + ret = kvm_irqchip_send_msi(kvm_state, msg); + if (ret <0) { + QEMU_LOG(LOG_ALERT, "MSI lost %s\n", strerror(-ret)); + } + } + msi_count++; ret = kvm_vcpu_ioctl(cpu, KVM_SMI); if (ret < 0) **************patch above************************* Then, when the vm will hang when booting. There will be only one line in the vnc displayer: "Guest has not initialized the display(yet)" I went through the codes(kvm qemu seabios), and I found the check_irqs function in seabios will allow interrupt for a while. When the interrupt triggered, the vcpu will lookup the seabios IDT table and jump to the interrupt handler. However, in ivt_init(void), the entries from 0x60 to 0x66 will be cleared, which is configured by the patch below. https://github.com/coreboot/seabios/commit/b164d2c1b8ff2dd764dcf064e2624dd4… So the vcpu will jump to an undefined entry instead of the default entry. ========================================================================= I removed the patch, and the bug seems to disappear. I wonder if this is an approprite solution. ========================================================================= There is an another strange thing: I opened kvm trace, and it showed like below: ************************ kvm_inj_exception: #UD(0x0) kvm_entry: vcpu0 kvm_run: vcpu 0 to guest mode kvm_run: vcpu 0 out guest mode kvm_exit: reason EXCEPTION_NMI rip 0x3 info 0 80000306 kvm_emulate_insn: 0:3:f0 53 (real) kvm_inj_exception: #UD(0x0) kvm_entry: vcpu0 kvm_run: vcpu 0 to guest mode kvm_run: vcpu 0 out guest mode kvm_exit: reason EXCEPTION_NMI rip 0x3 info 0 80000306 kvm_emulate_insn: 0:3:f0 53 (real) ************************ It seems the emulation got wrong at cs:eip = 0:0x3, but the entry is set to 0 in ivt_init(void). thanks!

1 0

preparing 1.11.1 stable release
by Gerd Hoffmann March 19, 2018

March 19, 2018

Hi, Time to prepare a 1.11-stable release, so we can pick up bugfixes for the upcoming qemu release. Looking at the commits I think these should go in: a3c93bd81d build: Use git describe --always 42812e062a shadow: Don't invoke a shutdown on reboot unless in a reboot loop 14d91c353e pci: fix 'io hints' capability for RedHat PCI bridges 0541f2f0f2 paravirt: Only enable sercon in NOGRAPHIC mode if no other console specified Comments? Other suggestions? thanks, Gerd

3 3