similar to: [PATCH] hvmloader: fix SeaBIOS interface

Hi, These set of patches are affected by replacing bios_relocate hook with bios_load hook in hvmloader. The patches for code files config.h and hvmloader.c also contains part of the contents of Enabling UEFI BIOS(OVMF) support in Xen-unstable HVM. Is there any problem with these patches? Thank you very much. # HG changeset patch # User gbtju85@gmail.com # Replace bios_relocate hook with

This patch set mainly allows the user to build a seabios or rombios only version of hvmloader. In addition, when building a seabios only hvmloader, Option ROMs like vgabios and etherboot are no longer required, and therefore can be disabled from the build. Dependency on the bcc compiler can also be avoided the same way. v2: Separate patches for separate issues Introduced config option to

A "main" function is defined in Xen-3.1.0-source/tools/firmware/hvmloader/hvmloader.c. In this main function, rombios, vgabios and vmxassist are copied to specified address, with following commands memcpy((void *)ROMBIOS_PHYSICAL_ADDRESS, rombios, rombios_sz); ... memcpy((void *)ETHERBOOT_PHYSICAL_ADDRESS, etherboot, sizeof(etherboot)); .... But after searching the whole

Some firmware, such as OVMF relies on this value to get the size of extra memory above 4GB. Seabios in Xen doesn''t need this as it gets e820 directly from Xen. Rombios doesn''t read this value. Signed-off-by: Wei Liu <wei.liu2@citrix.com> --- tools/firmware/hvmloader/hvmloader.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git

Thank you. I notice the system then jumps to F000:FFF0 to execute. But because VMX is turned on, switching to real-mode would incur problems? I don’t find any clue to turn on the vm86 mode as Readme in the tools/firmware directory puts. Best regards, Hu Jia Yi Ext: 20430 Tel: 65-67510430 -----Original Message----- From: Cui, Dexuan [mailto:dexuan.cui@intel.com] Sent: Thursday,

Currently it is assumed PCI device BAR access < 4G memory. If there is such a device whose BAR size is larger than 4G, it must access > 4G memory address. This patch enable the 64bits big BAR support on hvmloader. v3 changes from v2: - Remain original print information v2 changes from v1 as comments by Jan. 1) Set Dynamic MMIO high memory address instead of a fixed number 640G 2) Mask

Package: xen-utils-unstable Version: 3.3-unstable+hg17192-1 Severity: important Tags: patch In tools/firmware/Makefile, INST_DIR is $(DESTDIR)/usr/$(PREFIX)/boot but $(PREFIX) is empty. hvmloader doesn't get installed. diff -ru xen-unstable-3.3-unstable+hg17192.orig/debian/patches/tools-prefix.diff xen-unstable-3.3-unstable+hg17192/debian/patches/tools-prefix.diff ---

When booting HVMs with SeaBIOS, the BIOS itself takes care of extracting option ROMs from the PCI devices. These ROMs are usually provided with by the device model (qemu). Thus, hvmloader should not require any longer the rombios, stdvga, cirrusvga or etherboot ROMs to be present. Also, the 32bitbios_support.c file is specific to rombios and should not be built when building hvmloader with

This patch set mainly allows the user to build a seabios or rombios only version of hvmloader. In addition, when building a seabios only hvmloader, Option ROMs like vgabios and etherboot are no longer required, and therefore can be disabled from the build. Dependency on the bcc compiler can also be avoided the same way. v2: Separate patches for separate issues Introduced config option to

Attached simple patch modifies the rombios and hvmloader Makefiles to default to an 8-way bochs bios for the hvmloader. Is there a reason to not define the mp config table by default for >1 processors (8), in the repository? Seems like using the hvm guest config file can the methodolody to describe a UP guest regardless of the mptables. Defaulting the rombios to MP would allow users to

This patch series adds the ability to selectively disable the S3 and S4 ACPI power states for HVM guests. Since there is a general move towards retiring the hvm_info_table structure, the first patch moves the acpi_enabled flag out of the hvm_info_table and into a xenstore key (platform/acpi). The second patch then introduces the acpi_s3 and acpi_s4 configuration parameters to the xl config file

Hello, I found out what the problem in my other thread was, now I need to solve it: I need to port a modified Bochs bios to Xen x86. However, a file of the original source seems missing: acpi-dsdt.hex. It''s neither in the tools/firmware/hvmloader/acpi directory nor in the tools/firmware/rombios one. Also applying the diffs or manually editing the source probably won''t work since

Hi, Please post any comments or questions if you get around to testing them. Your comments are very welcome! Best regards, -Gonglei > > >>> On 22.10.13 at 06:08, "Gonglei (Arei)" <arei.gonglei@huawei.com> > wrote: > > > Hi, guys. The new patch has been modified based on the principles you > > > suggested, thank you so much. > > >

I've uploaded seabios 1.7.4-2 to debian archive which now ships 2 versions of the bios binary -- small 128Kb one and bios-256k.bin (with size = 256Kb, obviously). For now, both are built with Xen support. So for now, it is enough to just rebuild xen with seabios >= 1.7.4-2~ to get this fixed without patching. However, please apply the mentioned patch (or update to xen-4.3.1 version)

> -----Original Message----- > From: xen-staging-bounces@lists.xensource.com > [mailto:xen-staging-bounces@lists.xensource.com] On Behalf Of > Xen staging patchbot-unstable > Sent: 24 April 2007 12:16 > To: xen-staging@lists.xensource.com > Subject: [Xen-staging] [xen-unstable] hvm rombios: Fix int15, > func 0x86 (udelay). > > # HG changeset patch > # User

I am unable to build Xen on an x86 FC3 platform where I routinely build and test Xen: make[3]: Leaving directory `/tmp/xen-unstable.hg/tools/pygrub'' make[3]: Entering directory `/tmp/xen-unstable.hg/tools/firmware'' make[4]: Entering directory `/tmp/xen- unstable.hg/tools/firmware/rombios'' gcc -o biossums biossums.c gcc -DBX_SMP_PROCESSORS=1 -E -P rombios.c >

[HVM] Pick up SMBIOS entry point in the ROMBIOS. Signed-off-by: Andrew D. Ball <aball@us.ibm.com> diff -r 1d817bfc5ed9 tools/firmware/rombios/rombios.c --- a/tools/firmware/rombios/rombios.c Fri Aug 11 14:22:54 2006 +0100 +++ b/tools/firmware/rombios/rombios.c Fri Aug 11 14:55:22 2006 -0400 @@ -9443,6 +9443,43 @@ rom_scan_increment: mov ds, ax ret +#ifdef HVMASSIST + +; Copy the

Package: xen-utils-4.3 Version: 4.3.0-3 Severity: important Tags: upstream patch When using Xen compiled with seabios-1.7.4-1 (now in unstable/testing), HVMs fail immediately upon starting with a VCPU triple fault: (XEN) HVM1: Loading SeaBIOS ... (XEN) hvm.c:1253:d1 Triple fault on VCPU0 - invoking HVM shutdown action 1. This is because SeaBIOS can now be greater than 128k, which breaks an

I ran into this when I was reviewing what was outstanding to get NetWare to run fully virtualized. The BIOS emulator code has the segment and offset backward in the returned buffer for the int 13h, fn 48h BIOS call. BTW: This is fixed in current Bochs emulator upstream code. Signed-off-by: Bruce Rogers <brogers@novell.com> diff -r 59b8768d0d0d tools/firmware/rombios/rombios.c ---

HI: I try to let guest os to DHA(directly hardware acceess) physical network card, and also want to let rombios (which is not DHA) to initialize networks card''s pnp option Rom(the pnp option rom is DHA). however , I find the init program(offset by 3 from 55aa header signature) within option rom can not run properly ---- it cause stact underflow . and when retf form option rom ,system