user/sven/linux.git/drivers/rtc/rtc-efi.c, branch v4.9.261

rtc: efi: Fail probing if RTC reads don't work

2016-06-06T15:07:33Z

While the EFI spec mandates an RTC, not every implementation actually adheres to that rule (or can adhere to it - some systems just don't have an RTC). For those, we really don't want to probe the EFI RTC driver at all, because if we do we'd get a non-functional driver that does nothing useful but only spills our kernel log with warnings. Signed-off-by: Alexander Graf Acked-by: Ard Biesheuvel Signed-off-by: Alexandre Belloni

rtc: efi: add efi_procfs in efi_rtc_ops

2016-01-11T19:19:55Z

Add efi_procfs in efi_rtc_ops to show rtc-efi info in /proc/driver/rtc. Most of the code comes from efi_rtc_proc_show() in efirtc. Signed-off-by: Geliang Tang Signed-off-by: Alexandre Belloni

rtc: efi: use correct EFI 'epoch'

2015-06-24T23:13:44Z

The rtc-efi driver declares that the EFI 'epoch' is 1/1/1998, but the UEFI spec does not define it at all. It does define a range of [1900, 9999] for the 'Year' member of the EFI_TIME struct, so let's use 1900 as the minimum year and not 1998. Also, move the validation of the year to the convert_from_efi_time() routine where all other EFI_TIME fields are validated as well. This prevents rtc_read_time() failures when the RTC that backs the EFI time services is set to a date before 1998, e.g., when it has lost power. This also optimizes the compute_wday() routine, by replacing the for loop with a simple arithmetic expression, and by reusing the yearday value that we need to compute anyway when populating the rtc_time::tm_yday field. Cc: Alessandro Zummo Cc: Alexandre Belloni Cc: rtc-linux@googlegroups.com Signed-off-by: Ard Biesheuvel Signed-off-by: Alexandre Belloni

rtc: efi: Update author email address

2015-06-19T18:03:20Z

I'm no longer employed by HP. Signed-off-by: dann frazier Signed-off-by: Alexandre Belloni

Merge tag 'efi-next' of git://git.kernel.org/pub/scm/linux/kernel/git/mfleming/efi into x86/efi

2015-01-29T18:16:40Z

Pull EFI updates from Matt Fleming: " - Move efivarfs from the misc filesystem section to pseudo filesystem, since that's a more logical and accurate place - Leif Lindholm - Update efibootmgr URL in Kconfig help - Peter Jones - Improve accuracy of EFI guid function names - Borislav Petkov - Expose firmware platform size in sysfs for the benefit of EFI boot loader installers and other utilities - Steve McIntyre - Cleanup __init annotations for arm64/efi code - Ard Biesheuvel - Mark the UIE as unsupported for rtc-efi - Ard Biesheuvel - Fix memory leak in error code path of runtime map code - Dan Carpenter - Improve robustness of get_memory_map() by removing assumptions on the size of efi_memory_desc_t (which could change in future spec versions) and querying the firmware instead of guessing about the memmap size - Ard Biesheuvel - Remove superfluous guid unparse calls - Ivan Khoronzhuk - Delete unnecessary chosen@0 DT node FDT code since was duplicated from code in drivers/of and is entirely unnecessary - Leif Lindholm There's nothing super scary, mainly cleanups, and a merge from Ricardo who kindly picked up some patches from the linux-efi mailing list while I was out on annual leave in December. Perhaps the biggest risk is the get_memory_map() change from Ard, which changes the way that both the arm64 and x86 EFI boot stub build the early memory map. It would be good to have it bake in linux-next for a while. " Signed-off-by: Ingo Molnar

efi: rtc-efi: Mark UIE as unsupported

2015-01-20T15:46:21Z

Tools like hwclock attempt to enable the RTC update interrupt (UIE) to maximize the accuracy of the reported time value. The EFI rtc does not have interrupt capability so this is a pointless exercise to begin with, but the generic RTC framework ends up issuing a SetWakeupTime() Runtime Services call before drawing that conclusion on its own. Instead, we can mark UIE as unsupported at driver probe time. The net result is the same, but without the spurious SetWakeupTime() call. Signed-off-by: Ard Biesheuvel Signed-off-by: Matt Fleming

Merge branch 'platform/remove_owner' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux into driver-core-next

2014-11-04T03:53:56Z

Remove all .owner fields from platform drivers

rtc: drop owner assignment from platform_drivers

2014-10-20T14:21:31Z

A platform_driver does not need to set an owner, it will be populated by the driver core. Signed-off-by: Wolfram Sang

Merge branch 'next' into efi-next-merge

2014-10-03T21:15:56Z

Conflicts: arch/x86/boot/compressed/eboot.c

efi: rtc-efi: Export platform:rtc-efi as module alias

2014-10-03T17:41:04Z

When the rtc-efi driver is built as a module, we already register the EFI rtc as a platform device if UEFI Runtime Services are enabled. To wire it up to udev, and let the module be loaded automatically, we need to export the 'platform:rtc-efi' alias from the module. Signed-off-by: Ard Biesheuvel Cc: Alessandro Zummo Cc: Mark Salter Cc: Dave Young Signed-off-by: Matt Fleming