> On Jun 12, 2023, at 2:52 AM, Nikos Nikoleris <nikos.nikoleris@xxxxxxx> wrote: > > On 12/06/2023 08:52, Andrew Jones wrote: >> On Sat, Jun 10, 2023 at 01:32:59AM -0700, Nadav Amit wrote: >>> >>>> On May 30, 2023, at 9:08 AM, Nikos Nikoleris <nikos.nikoleris@xxxxxxx> wrote: >>>> >>>> Hello, >>>> >>>> This series adds initial support for building arm64 tests as EFI >>>> apps and running them under QEMU. Much like x86_64, we import external >>>> dependencies from gnu-efi and adapt them to work with types and other >>>> assumptions from kvm-unit-tests. In addition, this series adds support >>>> for enumerating parts of the system using ACPI. >>> >>> Just an issue I encountered, which I am not sure is arm64 specific: >>> >>> All the printf’s in efi_main() are before current_thread_info() is >>> initialized (or even memset’d to zero, as done in setup_efi). >>> >>> But printf() calls puts() which checks if mmu_enabled(). And >>> mmu_enabled() uses is_user() and current_thread_info()->cpu, both >>> of which read uninitialized data from current_thread_info(). >>> >>> IOW: Any printf in efi_main() can cause a crash. >>> >> Nice catch, Nadav. Nikos, shouldn't we drop the memset() in setup_efi and >> put a zero_range call, similar to the one in arm/cstart64.S which zero's >> the thread-info area, in arm/efi/crt0-efi-aarch64.S? > > While I haven't run into any problems with this in this series, I had in a previous version and back then the solution was this patch: > > 993c37be - arm/arm64: Zero BSS and stack at startup > > So I agree we should drop the memset and call some macro like zero_range in arm/efi/crt0-efi-aarch64.S. > > Let me know if you want me to send a patch for this. Thanks. I am still struggling to run the tests on my environment. Why the heck frame-pointers are disabled on arm64? Perhaps I’ll send my patch to enable them (and add one on exception handling). Anyhow, I was wondering, since it was not clearly mentioned in the cover-letter: which tests were run on what environment? Did they all pass or are there still some open issues? [ I ask for my enabling efforts. Not blaming or anything. ] Thanks again for the hard work, Nikos (and Andrew).
