Re: [edk2] Shell2 cannot be re-used unless started/restarted on a stable volume

On 06/24/13 23:53, Duane Voth wrote: > Does your shim.efi execute commands via the shell? If not maybe it > isn't related to this issue? It's probably not related; I only used shim (and grub) as example EFI apps for the PXE-loaded shell2 to start, from the command line. Laszlo ---

Re: [edk2] [PATCH] OvmfPkg/NvVarsFileLib: handle the inital file not found

(sorry for butting in) On 06/24/13 23:13, Jordan Justen wrote: > On Fri, Jun 21, 2013 at 1:27 AM, Michael Chang wrote: >> My fresh vm installation with OVMF firmware has problem that the OS boot >> entry are not displayed by boot manager. It happens only when the EFI >> System Partition are creat

Re: [edk2] Shell2 cannot be re-used unless started/restarted on a stable volume

Ok, oops, no the command line doesn't use exactly the same interface as gEfiShellProtocol->Execute() . There is a wrapper between gEfiShellProtocol->Execute() and InternalShellExecuteDevicePath() called EfiShellExecute() which uses ShellInfoObject.FileDevPath and ShellInfoObject.ImageDevPath to lo

Re: [edk2] [PATCH 0/3] OvmfPkg: basic SMBIOS support on QEMU

On Mon, Jun 24, 2013 at 2:03 AM, Laszlo Ersek wrote: > On 06/18/13 19:45, Laszlo Ersek wrote: >> On 06/18/13 18:18, Jordan Justen wrote: >> >>> Andrew: SmbiosLib exists in EmulatorPkg and not somewhere more common, >>> because?? I'll assume because it is difficult to add new features to >>> MdePkg

Re: [edk2] [PATCH] OvmfPkg/NvVarsFileLib: handle the inital file not found

On Mon, Jun 24, 2013 at 2:18 PM, Laszlo Ersek wrote: > On 06/24/13 23:06, Jordan Justen wrote: >> We could consider yanking all these hacks out, and just forcing people >> to use QEMU with flash support when that eventually gets finished. >> >> But, some people like the idea that the 'disk image'

Re: [edk2] [PATCH] OvmfPkg/NvVarsFileLib: handle the inital file not found

On 06/24/13 23:06, Jordan Justen wrote: > On Mon, Jun 24, 2013 at 6:12 AM, Laszlo Ersek wrote: >> ... I think I can see interaction between ReserveEmuVariableNvStore(), >> FvbInitialize(), and LoadNvVarsFromFs() now. When >> ReserveEmuVariableNvStore() runs after VM reboot, AllocateRuntimePool() >

Re: [edk2] [PATCH] OvmfPkg/NvVarsFileLib: handle the inital file not found

On Fri, Jun 21, 2013 at 1:27 AM, Michael Chang wrote: > My fresh vm installation with OVMF firmware has problem that the OS boot > entry are not displayed by boot manager. It happens only when the EFI > System Partition are created from scratch, that is there's no NvVars in > it initially. > > Aft

Re: [edk2] [PATCH] OvmfPkg/NvVarsFileLib: handle the inital file not found

On Mon, Jun 24, 2013 at 6:12 AM, Laszlo Ersek wrote: > ... I think I can see interaction between ReserveEmuVariableNvStore(), > FvbInitialize(), and LoadNvVarsFromFs() now. When > ReserveEmuVariableNvStore() runs after VM reboot, AllocateRuntimePool() > will return the same address (overlaying the

Re: [edk2] Shell2 cannot be re-used unless started/restarted on a stable volume

On 06/24/13 22:06, Duane Voth wrote: > > On Mon, Jun 24, 2013 at 1:28 PM, Laszlo Ersek > wrote: > > On 06/24/13 16:53, Duane Voth wrote: > > ... if the Shell is, for example, the payload for a PXE > > boot, the call to GetDevicePathsForImageAndFile() in Shel

Re: [edk2] Shell2 cannot be re-used unless started/restarted on a stable volume

On Mon, Jun 24, 2013 at 1:28 PM, Laszlo Ersek wrote: > On 06/24/13 16:53, Duane Voth wrote: > > ... if the Shell is, for example, the payload for a PXE > > boot, the call to GetDevicePathsForImageAndFile() in Shell.c fails > > (because the shell.efi binary cannot be accessed locally) and the Shel

Re: [edk2] Shell2 cannot be re-used unless started/restarted on a stable volume

On 06/24/13 20:45, Andrew Fish wrote: > > On Jun 24, 2013, at 11:28 AM, Laszlo Ersek > wrote: > >> IIRC Andrew may have stated at one point that *some* shell is >> required at all times for fundamental UEFI functionality. > > The shell is a UEFI application, and DOES NOT produce anything > fund

Re: [edk2] Shell2 cannot be re-used unless started/restarted on a stable volume

On Jun 24, 2013, at 11:28 AM, Laszlo Ersek wrote: > IIRC Andrew may have stated at one point that *some* shell is required > at all times for fundamental UEFI functionality. The shell is a UEFI application, and DOES NOT produce anything fundamental to UEFI functionality. The shell just consume

Re: [edk2] Shell2 cannot be re-used unless started/restarted on a stable volume

On 06/24/13 16:53, Duane Voth wrote: > When Shell2 (ShellPkg) is started it records the ImageDevPath and > FileDevPath for itself for use in gEfiShellProtocol->Execute() (which > any other app is allowed to call to parse and execute a shell command > line). However if the Shell is, for example, th

Re: [edk2] [PATCH] OvmfPkg/NvVarsFileLib: handle the inital file not found

Michael, On 06/24/13 15:12, Laszlo Ersek wrote: > On the other hand, I think we should implement this more simply. What do > you think of the attached patch? Self-NAK for that, it's broken. PlatformBdsRestoreNvVarsFromHardDisk VisitAllInstancesOfProtocol -- for each Simple File System; pe

Re: [edk2] StdLib getenv not POSIX compliant

Ah, ok, well then the python startup code is broken. :) Both versions (Python-2.7.2/Modules/getpath.c and Efi/getpath.c) contain: static void calculate_path(void) { extern char *Py_GetProgramName(void); static char delimiter[2] = {DELIM, '\0'}; static char separator[2] = {SEP, '\0'

[edk2] Shell2 cannot be re-used unless started/restarted on a stable volume

When Shell2 (ShellPkg) is started it records the ImageDevPath and FileDevPath for itself for use in gEfiShellProtocol->Execute() (which any other app is allowed to call to parse and execute a shell command line). However if the Shell is, for example, the payload for a PXE boot, the call to GetDevi

Re: [edk2] [PATCH] OvmfPkg/NvVarsFileLib: handle the inital file not found

On 06/24/13 15:12, Laszlo Ersek wrote: > On the other hand, I think we should implement this more simply. What do > you think of the attached patch? ... In any case, your patch fixes a very annoying bug. If Jordan prefers your patch over my suggestion (or, of course, either of you points out a bu

Re: [edk2] [PATCH] OvmfPkg/NvVarsFileLib: handle the inital file not found

On 06/21/13 10:27, Michael Chang wrote: > My fresh vm installation with OVMF firmware has problem that the OS boot > entry are not displayed by boot manager. It happens only when the EFI > System Partition are created from scratch, that is there's no NvVars in > it initially. > > After some testin

Re: [edk2] [PATCH 2/2] UnixPkg, Nt32Pkg, EmulatorPkg and OvmfPkg: PEI variable does not robustly handle crashes during Reclaim().

Based on the impact of PATCH 1, it is to provide some platform changes. === EmulatorPkg: Use FaultTolerantWritePei driver. 1. The signature of working block header needs to be updated to gWorkingBlockSignatureGuid because FTW write header and record will be updated and exposed t

Re: [edk2] [PATCH 2/2] MdeModulePkg and SecurityPkg: PEI variable does not robustly handle crashes during Reclaim().

Based on the impact of PATCH 1, it is to provide some platform changes. === EmulatorPkg: Use FaultTolerantWritePei driver. 1. The signature of working block header needs to be updated to gWorkingBlockSignatureGuid because FTW write header and record will be updated and exposed t

[edk2] Recall: [PATCH 2/2] MdeModulePkg and SecurityPkg: PEI variable does not robustly handle crashes during Reclaim().

Zeng, Star would like to recall the message, "[edk2] [PATCH 2/2] MdeModulePkg and SecurityPkg: PEI variable does not robustly handle crashes during Reclaim().". -- This SF.net email is sponsored by Windows: Build for Win

Re: [edk2] [PATCH] MdeModulePkg: fix compilation errors

On 24 June 2013 10:08, Ryan Harkin wrote: > On 24 June 2013 08:44, Dong, Eric wrote: >> This change is not correct, the "Buffer" variable is not defined. > > Hmmm, isn't that curious! I've not really looked that this change > before, which is my mistake for accepting it, but you're right, Buffer

Re: [edk2] [PATCH] MdeModulePkg: fix compilation errors

On 24 June 2013 11:44, Leif Lindholm wrote: > On 24 June 2013 10:08, Ryan Harkin wrote: >> On 24 June 2013 08:44, Dong, Eric wrote: >>> This change is not correct, the "Buffer" variable is not defined. >> >> Hmmm, isn't that curious! I've not really looked that this change >> before, which is m

Re: [edk2] [PATCH] MdeModulePkg: fix compilation errors

On 24 June 2013 08:44, Dong, Eric wrote: > This change is not correct, the "Buffer" variable is not defined. Hmmm, isn't that curious! I've not really looked that this change before, which is my mistake for accepting it, but you're right, Buffer is not defined. I think look into this one proper

Re: [edk2] [PATCH 0/3] OvmfPkg: basic SMBIOS support on QEMU

On 06/18/13 19:45, Laszlo Ersek wrote: > On 06/18/13 18:18, Jordan Justen wrote: > >> Andrew: SmbiosLib exists in EmulatorPkg and not somewhere more common, >> because?? I'll assume because it is difficult to add new features to >> MdePkg. >> >> Laszlo, what do you think of using Andrew's >> Emula

Re: [edk2] [PATCH] MdeModulePkg: fix compilation errors

This change is not correct, the "Buffer" variable is not defined. Thanks, Eric -Original Message- From: Ryan Harkin [mailto:ryan.har...@linaro.org] Sent: Friday, June 21, 2013 3:48 PM To: ryan.har...@linaro.org; edk2-devel@lists.sourceforge.net; patc...@linaro.org; boot-architect...@l