x86_64 UEFI set up process refactor and scripts fixes
Subsystems (2)
cover x86_64 UEFI set up process refactor and scripts fixes
Hello,
This patch series refactors the x86_64 UEFI set up process and fixes the run-tests.sh script to run under UEFI. The patches are organized as three parts.
The first part (patches 1-2) refactors the x86_64 UEFI set up process. The previous UEFI setup calls arch-specific setup functions twice and generates arch-specific data structure. As Andrew suggested [1], we refactor this process to make only one call to the arch-specific function and generate arch-neutral data structures. This simplifies the set up process and makes it easier to develop UEFI support for other architectures.
The second part (patch 3) converts several x86 test cases to Position-Independent Code (PIC) to run under UEFI. This patch is ported from the initial UEFI support patchset [2] with fixes to the 32-bit compilation.
The third part (patches 4-7) fixes the UEFI runner scripts. Patch 4 sets UEFI OVMF image as readonly. Patch 5 fixes test cases' return code under UEFI, enabling Patch 6-7 to fix the run-tests.sh script under UEFI.
This patch set is based on the uefi branch.
Best regards,
Zixuan Wang and Marc Orr
[1] https://lore.kernel.org/kvm/[email protected]/
[2] https://lore.kernel.org/kvm/[email protected]/
Marc Orr (2):
scripts: Generalize EFI check
x86 UEFI: Make run_tests.sh (mostly) work under UEFI
Zixuan Wang (5):
x86 UEFI: Remove mixed_mode
x86 UEFI: Refactor set up process
x86 UEFI: Convert x86 test cases to PIC
x86 UEFI: Set UEFI OVMF as readonly
x86 UEFI: Exit QEMU with return code
lib/efi.c | 54 ++++++-- lib/efi.h | 19 ++- lib/linux/efi.h | 317 ++++++++++++++----------------------------- lib/x86/acpi.c | 36 +++-- lib/x86/acpi.h | 5 +- lib/x86/asm/setup.h | 16 +-- lib/x86/setup.c | 153 ++++++++++----------- lib/x86/usermode.c | 3 +- scripts/common.bash | 9 +- scripts/runtime.bash | 15 +- x86/Makefile.common | 10 +- x86/Makefile.x86_64 | 7 +- x86/access.c | 9 +- x86/cet.c | 8 +- x86/efi/run | 27 +++- x86/emulator.c | 5 +- x86/eventinj.c | 8 ++ x86/run | 6 +- x86/smap.c | 13 +- x86/umip.c | 26 +++- 20 files changed, 360 insertions(+), 386 deletions(-)
Signature · 1 line
2.33.0
PATCH 1/7 x86 UEFI: Remove mixed_mode
From: Zixuan Wang <[email protected]>
Remove the mixed_mode code from efi.h as we are not supporting i386 UEFI for now.
--- lib/linux/efi.h | 317 +++++++++++++++--------------------------------- 1 file changed, 100 insertions(+), 217 deletions(-)
lib/linux/efi.h +100 -217
@@ -62,15 +62,7 @@ typedef guid_t efi_guid_t;typedef struct {efi_guid_t guid; -u32 table; -}efi_config_table_32_t ; - -typedef union { -struct { -efi_guid_t guid; -void *table; - }; -efi_config_table_32_t mixed_mode; +void *table; }efi_config_table_t ;/*
@@ -251,127 +243,76 @@ typedef struct efi_generic_dev_path efi_device_path_protocol_t;/* * EFI Boot Services table */ -union efi_boot_services { -struct { -efi_table_hdr_t hdr; -void *raise_tpl; -void *restore_tpl; - efi_status_t (__efiapi *allocate_pages)(int, int, unsigned long, - efi_physical_addr_t *); - efi_status_t (__efiapi *free_pages)(efi_physical_addr_t, - unsigned long); - efi_status_t (__efiapi *get_memory_map)(unsigned long *, void *, - unsigned long *, - unsigned long *, u32 *); - efi_status_t (__efiapi *allocate_pool)(int, unsigned long, - void **); - efi_status_t (__efiapi *free_pool)(void *); - efi_status_t (__efiapi *create_event)(u32, unsigned long, - efi_event_notify_t, void *, - efi_event_t *); - efi_status_t (__efiapi *set_timer)(efi_event_t, - EFI_TIMER_DELAY, u64); - efi_status_t (__efiapi *wait_for_event)(unsigned long, - efi_event_t *, - unsigned long *); -void *signal_event; - efi_status_t (__efiapi *close_event)(efi_event_t); -void *check_event; -void *install_protocol_interface; -void *reinstall_protocol_interface; -void *uninstall_protocol_interface; - efi_status_t (__efiapi *handle_protocol)(efi_handle_t, - efi_guid_t *, void **); -void *__reserved; -void *register_protocol_notify; - efi_status_t (__efiapi *locate_handle)(int, efi_guid_t *, - void *, unsigned long *, - efi_handle_t *); - efi_status_t (__efiapi *locate_device_path)(efi_guid_t *, - efi_device_path_protocol_t **, - efi_handle_t *); - efi_status_t (__efiapi *install_configuration_table)(efi_guid_t *, - void *); -void *load_image; -void *start_image; - efi_status_t (__efiapi *exit)(efi_handle_t, - efi_status_t, - unsigned long, - efi_char16_t *); -void *unload_image; - efi_status_t (__efiapi *exit_boot_services)(efi_handle_t, - unsigned long); -void *get_next_monotonic_count; - efi_status_t (__efiapi *stall)(unsigned long); -void *set_watchdog_timer; -void *connect_controller; - efi_status_t (__efiapi *disconnect_controller)(efi_handle_t, - efi_handle_t, - efi_handle_t); -void *open_protocol; -void *close_protocol; -void *open_protocol_information; -void *protocols_per_handle; -void *locate_handle_buffer; - efi_status_t (__efiapi *locate_protocol)(efi_guid_t *, void *, - void **); -void *install_multiple_protocol_interfaces; -void *uninstall_multiple_protocol_interfaces; -void *calculate_crc32; -void *copy_mem; -void *set_mem; -void *create_event_ex; - }; -struct { -efi_table_hdr_t hdr; -u32 raise_tpl; -u32 restore_tpl; -u32 allocate_pages; -u32 free_pages; -u32 get_memory_map; -u32 allocate_pool; -u32 free_pool; -u32 create_event; -u32 set_timer; -u32 wait_for_event; -u32 signal_event; -u32 close_event; -u32 check_event; -u32 install_protocol_interface; -u32 reinstall_protocol_interface; -u32 uninstall_protocol_interface; -u32 handle_protocol; -u32 __reserved; -u32 register_protocol_notify; -u32 locate_handle; -u32 locate_device_path; -u32 install_configuration_table; -u32 load_image; -u32 start_image; -u32 exit; -u32 unload_image; -u32 exit_boot_services; -u32 get_next_monotonic_count; -u32 stall; -u32 set_watchdog_timer; -u32 connect_controller; -u32 disconnect_controller; -u32 open_protocol; -u32 close_protocol; -u32 open_protocol_information; -u32 protocols_per_handle; -u32 locate_handle_buffer; -u32 locate_protocol; -u32 install_multiple_protocol_interfaces; -u32 uninstall_multiple_protocol_interfaces; -u32 calculate_crc32; -u32 copy_mem; -u32 set_mem; -u32 create_event_ex; - } mixed_mode; -}; - -typedef union efi_boot_services efi_boot_services_t ; +typedef struct { +efi_table_hdr_t hdr; +void *raise_tpl; +void *restore_tpl; + efi_status_t(__efiapi *allocate_pages)(int, int, unsigned long, + efi_physical_addr_t *); + efi_status_t(__efiapi *free_pages)(efi_physical_addr_t, + unsigned long); + efi_status_t(__efiapi *get_memory_map)(unsigned long *, void *, + unsigned long *, + unsigned long *, u32 *); + efi_status_t(__efiapi *allocate_pool)(int, unsigned long, + void **); + efi_status_t(__efiapi *free_pool)(void *); + efi_status_t(__efiapi *create_event)(u32, unsigned long, + efi_event_notify_t, void *, + efi_event_t *); + efi_status_t(__efiapi *set_timer)(efi_event_t, + EFI_TIMER_DELAY, u64); + efi_status_t(__efiapi *wait_for_event)(unsigned long, + efi_event_t *, + unsigned long *); +void *signal_event; + efi_status_t(__efiapi *close_event)(efi_event_t); +void *check_event; +void *install_protocol_interface; +void *reinstall_protocol_interface; +void *uninstall_protocol_interface; + efi_status_t(__efiapi *handle_protocol)(efi_handle_t, + efi_guid_t *, void **); +void *__reserved; +void *register_protocol_notify; + efi_status_t(__efiapi *locate_handle)(int, efi_guid_t *, + void *, unsigned long *, + efi_handle_t *); + efi_status_t(__efiapi *locate_device_path)(efi_guid_t *, + efi_device_path_protocol_t **, + efi_handle_t *); + efi_status_t(__efiapi *install_configuration_table)(efi_guid_t *, + void *); +void *load_image; +void *start_image; + efi_status_t(__efiapi *exit)(efi_handle_t, + efi_status_t, + unsigned long, + efi_char16_t *); +void *unload_image; + efi_status_t(__efiapi *exit_boot_services)(efi_handle_t, + unsigned long); +void *get_next_monotonic_count; + efi_status_t(__efiapi *stall)(unsigned long); +void *set_watchdog_timer; +void *connect_controller; + efi_status_t(__efiapi *disconnect_controller)(efi_handle_t, + efi_handle_t, + efi_handle_t); +void *open_protocol; +void *close_protocol; +void *open_protocol_information; +void *protocols_per_handle; +void *locate_handle_buffer; + efi_status_t(__efiapi *locate_protocol)(efi_guid_t *, void *, + void **); +void *install_multiple_protocol_interfaces; +void *uninstall_multiple_protocol_interfaces; +void *calculate_crc32; +void *copy_mem; +void *set_mem; +void *create_event_ex; +}efi_boot_services_t ;/* * Types and defines for EFI ResetSystem
@@ -386,24 +327,6 @@ typedef union efi_boot_services efi_boot_services_t;#define EFI_RUNTIME_SERVICES_SIGNATURE ((u64)0x5652453544e5552ULL)#define EFI_RUNTIME_SERVICES_REVISION 0x00010000 -typedef struct { -efi_table_hdr_t hdr; -u32 get_time; -u32 set_time; -u32 get_wakeup_time; -u32 set_wakeup_time; -u32 set_virtual_address_map; -u32 convert_pointer; -u32 get_variable; -u32 get_next_variable; -u32 set_variable; -u32 get_next_high_mono_count; -u32 reset_system; -u32 update_capsule; -u32 query_capsule_caps; -u32 query_variable_info; -}efi_runtime_services_32_t ; -typedef efi_status_t efi_get_time_t (efi_time_t *tm,efi_time_cap_t *tc);typedef efi_status_t efi_set_time_t (efi_time_t *tm); typedef efi_status_t efi_get_wakeup_time_t (efi_bool_t *enabled, efi_bool_t *pending,
@@ -438,25 +361,22 @@ typedef efi_status_t efi_query_variable_store_t(u32 attributes,unsigned long size, bool nonblocking); -typedef union { -struct { -efi_table_hdr_t hdr; - efi_get_time_t __efiapi *get_time; - efi_set_time_t __efiapi *set_time; - efi_get_wakeup_time_t __efiapi *get_wakeup_time; - efi_set_wakeup_time_t __efiapi *set_wakeup_time; - efi_set_virtual_address_map_t __efiapi *set_virtual_address_map; -void *convert_pointer; - efi_get_variable_t __efiapi *get_variable; - efi_get_next_variable_t __efiapi *get_next_variable; - efi_set_variable_t __efiapi *set_variable; - efi_get_next_high_mono_count_t __efiapi *get_next_high_mono_count; - efi_reset_system_t __efiapi *reset_system; - efi_update_capsule_t __efiapi *update_capsule; - efi_query_capsule_caps_t __efiapi *query_capsule_caps; - efi_query_variable_info_t __efiapi *query_variable_info; - }; -efi_runtime_services_32_t mixed_mode; +typedef struct { +efi_table_hdr_t hdr; + efi_get_time_t __efiapi *get_time; + efi_set_time_t __efiapi *set_time; + efi_get_wakeup_time_t __efiapi *get_wakeup_time; + efi_set_wakeup_time_t __efiapi *set_wakeup_time; + efi_set_virtual_address_map_t __efiapi *set_virtual_address_map; +void *convert_pointer; + efi_get_variable_t __efiapi *get_variable; + efi_get_next_variable_t __efiapi *get_next_variable; + efi_set_variable_t __efiapi *set_variable; + efi_get_next_high_mono_count_t __efiapi *get_next_high_mono_count; + efi_reset_system_t __efiapi *reset_system; + efi_update_capsule_t __efiapi *update_capsule; + efi_query_capsule_caps_t __efiapi *query_capsule_caps; + efi_query_variable_info_t __efiapi *query_variable_info; }efi_runtime_services_t ;#define EFI_SYSTEM_TABLE_SIGNATURE ((u64)0x5453595320494249ULL)
@@ -468,60 +388,23 @@ typedef union {#define EFI_1_10_SYSTEM_TABLE_REVISION ((1 << 16) | (10))#define EFI_1_02_SYSTEM_TABLE_REVISION ((1 << 16) | (02)) -typedef struct { -efi_table_hdr_t hdr; -u64 fw_vendor;/* physical addr of CHAR16 vendor string */ -u32 fw_revision; -u32 __pad1; -u64 con_in_handle; -u64 con_in; -u64 con_out_handle; -u64 con_out; -u64 stderr_handle; -u64 stderr; -u64 runtime; -u64 boottime; -u32 nr_tables; -u32 __pad2; -u64 tables; -}efi_system_table_64_t ; +typedef union efi_simple_text_input_protocol efi_simple_text_input_protocol_t ; +typedef union efi_simple_text_output_protocol efi_simple_text_output_protocol_t ;typedef struct {efi_table_hdr_t hdr; -u32 fw_vendor;/* physical addr of CHAR16 vendor string */ +unsigned long fw_vendor;/* physical addr of CHAR16 vendor string */ u32 fw_revision; -u32 con_in_handle; -u32 con_in; -u32 con_out_handle; -u32 con_out; -u32 stderr_handle; -u32 stderr; -u32 runtime; -u32 boottime; -u32 nr_tables; -u32 tables; -}efi_system_table_32_t ; - -typedef union efi_simple_text_input_protocol efi_simple_text_input_protocol_t ; -typedef union efi_simple_text_output_protocol efi_simple_text_output_protocol_t ; - -typedef union { -struct { -efi_table_hdr_t hdr; -unsigned long fw_vendor;/* physical addr of CHAR16 vendor string */ -u32 fw_revision; -unsigned long con_in_handle; -efi_simple_text_input_protocol_t *con_in; -unsigned long con_out_handle; -efi_simple_text_output_protocol_t *con_out; -unsigned long stderr_handle; -unsigned long stderr; -efi_runtime_services_t *runtime; -efi_boot_services_t *boottime; -unsigned long nr_tables; -unsigned long tables; - }; -efi_system_table_32_t mixed_mode; +unsigned long con_in_handle; +efi_simple_text_input_protocol_t *con_in; +unsigned long con_out_handle; +efi_simple_text_output_protocol_t *con_out; +unsigned long stderr_handle; +unsigned long stderr; +efi_runtime_services_t *runtime; +efi_boot_services_t *boottime; +unsigned long nr_tables; +unsigned long tables; }efi_system_table_t ;struct efi_boot_memmap {
Signature · 1 line
2.33.0
PATCH 2/7 x86 UEFI: Refactor set up process
In reply to Zixuan WangFrom: Zixuan Wang <[email protected]>
Refactor the EFI set up process. The previous set up process calls multiple arch-specific functions, now it's simplified to call only one arch-specific function:
1. (Arch neutral ) Extract EFI data structures, e.g., memory maps
2. (Arch neutral ) Exit EFI boot services
3. (Arch specific) Parse EFI data structures and set up arch-specific resources
4. (Arch neutral ) Run test cases' main functions
--- lib/efi.c | 50 ++++++++++++--- lib/efi.h | 19 ++++-- lib/x86/acpi.c | 36 ++++++----- lib/x86/acpi.h | 5 +- lib/x86/asm/setup.h | 16 +---- lib/x86/setup.c | 153 +++++++++++++++++++++----------------------- 6 files changed, 149 insertions(+), 130 deletions(-)
lib/efi.c +40 -10
@@ -65,9 +65,9 @@ out:return status; } -efi_status_t efi_exit_boot_services (void *handle,unsigned long mapkey) +efi_status_t efi_exit_boot_services (void *handle,struct efi_boot_memmap *map) { -return efi_bs_call (exit_boot_services, handle, mapkey); +return efi_bs_call (exit_boot_services, handle, *map->key_ptr); }efi_status_t efi_get_system_config_table (efi_guid_t table_guid,void **table)
@@ -88,31 +88,61 @@ efi_status_t efi_get_system_config_table(efi_guid_t table_guid, void **table)efi_status_t efi_main (efi_handle_t handle,efi_system_table_t *sys_tab) {int ret; -unsigned long mapkey =0 ;efi_status_t status;efi_bootinfo_t efi_bootinfo; efi_system_table = sys_tab; -setup_efi_bootinfo (&efi_bootinfo); - status =setup_efi_pre_boot (&mapkey, &efi_bootinfo); +/* Memory map struct values */ +efi_memory_desc_t *map =NULL ; +unsigned long map_size =0 , desc_size =0 , key =0 , buff_size =0 ; +u32 desc_ver; + +/* Set up efi_bootinfo */ + efi_bootinfo.mem_map.map = ↦ + efi_bootinfo.mem_map.map_size = &map_size; + efi_bootinfo.mem_map.desc_size = &desc_size; + efi_bootinfo.mem_map.desc_ver = &desc_ver; + efi_bootinfo.mem_map.key_ptr = &key; + efi_bootinfo.mem_map.buff_size = &buff_size; + +/* Get EFI memory map */ + status =efi_get_memory_map (&efi_bootinfo.mem_map);if (status !=EFI_SUCCESS ) { -printf ("Failed to set up before ExitBootServices, exiting.\n" ); -return status; +printf ("Failed to get memory map\n" ); + goto efi_main_error; } - status =efi_exit_boot_services (handle, mapkey); +/* +* Exit EFI boot services, let kvm-unit-tests take full control of the +* guest +*/ + status =efi_exit_boot_services (handle, &efi_bootinfo.mem_map);if (status !=EFI_SUCCESS ) {printf ("Failed to exit boot services\n" ); -return status; + goto efi_main_error; } -setup_efi (&efi_bootinfo); +/* Set up arch-specific resources */ + status =setup_efi (&efi_bootinfo); +if (status !=EFI_SUCCESS ) { +printf ("Failed to set up arch-specific resources\n" ); + goto efi_main_error; + } + +/* Run the test case */ ret =main (__argc, __argv, __environ);/* Shutdown the guest VM */ efi_rs_call (reset_system,EFI_RESET_SHUTDOWN , ret,0 ,NULL ); +/* Unreachable */ +return EFI_UNSUPPORTED ; + +efi_main_error: +/* Shutdown the guest with error EFI status */ +efi_rs_call (reset_system,EFI_RESET_SHUTDOWN , status,0 ,NULL ); +/* Unreachable */ return EFI_UNSUPPORTED ; }
lib/efi.h +14 -5
@@ -2,9 +2,7 @@#define _EFI_H_/* -* EFI-related functions in . This file's name "efi.h" is in -* conflict with GNU-EFI library's "efi.h", but does not include -* GNU-EFI headers or links against GNU-EFI. +* EFI-related functions. * * Copyright (c) 2021, Google Inc, Zixuan Wang <[email protected]> *
@@ -13,9 +11,20 @@#include "linux/efi.h" #include <elf.h> -efi_status_t _relocate (long ldbase,Elf64_Dyn *dyn,efi_handle_t handle,efi_system_table_t *sys_tab); +/* +* efi_bootinfo_t: stores EFI-related machine info retrieved before exiting EFI +* boot services, and is then used by setup_efi(). setup_efi() cannot retrieve +* this info as it is called after ExitBootServices and thus some EFI resources +* and functions are not available. +*/ +typedef struct { +struct efi_boot_memmap mem_map; +}efi_bootinfo_t ; + +efi_status_t _relocate (long ldbase,Elf64_Dyn *dyn,efi_handle_t handle, +efi_system_table_t *sys_tab);efi_status_t efi_get_memory_map (struct efi_boot_memmap *map); -efi_status_t efi_exit_boot_services (void *handle,unsigned long mapkey); +efi_status_t efi_exit_boot_services (void *handle,struct efi_boot_memmap *map);efi_status_t efi_get_system_config_table (efi_guid_t table_guid,void **table);efi_status_t efi_main (efi_handle_t handle,efi_system_table_t *sys_tab);
lib/x86/acpi.c +21 -15
@@ -4,29 +4,35 @@#ifdef TARGET_EFI struct rsdp_descriptor *efi_rsdp =NULL ; -void setup_efi_rsdp (struct rsdp_descriptor *rsdp) { +void set_efi_rsdp (struct rsdp_descriptor *rsdp) +{ efi_rsdp = rsdp; } -static struct rsdp_descriptor *get_rsdp (void ) { +static struct rsdp_descriptor *get_rsdp (void ) +{if (efi_rsdp ==NULL ) { -printf ("Can't find RSDP from UEFI, maybe setup_efi_rsdp() was not called\n" ); +printf ("Can't find RSDP from UEFI, maybe set_efi_rsdp() was not called\n" ); }return efi_rsdp; }#else -static struct rsdp_descriptor *get_rsdp (void ) { -struct rsdp_descriptor *rsdp; -unsigned long addr; -for (addr =0xf0000 ; addr <0x100000 ; addr +=16 ) { - rsdp = (void *)addr; -if (rsdp->signature ==RSDP_SIGNATURE_8BYTE ) -break ; - } -if (addr ==0x100000 ) { -return NULL ; - } -return rsdp; +static struct rsdp_descriptor *get_rsdp (void ) +{ +struct rsdp_descriptor *rsdp; +unsigned long addr; + +for (addr =0xf0000 ; addr <0x100000 ; addr +=16 ) { + rsdp = (void *)addr; +if (rsdp->signature ==RSDP_SIGNATURE_8BYTE ) +break ; + } + +if (addr ==0x100000 ) { +return NULL ; + } + +return rsdp; }#endif /* TARGET_EFI */
lib/x86/acpi.h +1 -4
@@ -106,10 +106,7 @@ struct facs_descriptor_rev1 u8 reserved3 [40]; /* Reserved - must be zero */ }; +void set_efi_rsdp (struct rsdp_descriptor *rsdp);void *find_acpi_table_addr (u32 sig); -#ifdef TARGET_EFI -void setup_efi_rsdp (struct rsdp_descriptor *rsdp); -#endif /* TARGET_EFI */ -#endif
lib/x86/asm/setup.h +1 -15
@@ -12,21 +12,7 @@ unsigned long setup_tss(u8 *stacktop);#include "efi.h" #include "x86/amd_sev.h" -/* -* efi_bootinfo_t: stores EFI-related machine info retrieved by -* setup_efi_pre_boot(), and is then used by setup_efi(). setup_efi() cannot -* retrieve this info as it is called after ExitBootServices and thus some EFI -* resources are not available. -*/ -typedef struct { -phys_addr_t free_mem_start; -phys_addr_t free_mem_size; -struct rsdp_descriptor *rsdp; -}efi_bootinfo_t ; - -void setup_efi_bootinfo (efi_bootinfo_t *efi_bootinfo); -void setup_efi (efi_bootinfo_t *efi_bootinfo); -efi_status_t setup_efi_pre_boot (unsigned long *mapkey,efi_bootinfo_t *efi_bootinfo); +efi_status_t setup_efi (efi_bootinfo_t *efi_bootinfo);void setup_5level_page_table (void );#endif /* TARGET_EFI */
lib/x86/setup.c +72 -81
200 lines not on this page · raw message
Signature · 1 line
2.33.0
Sean Christopherson
reviewerRe: [kvm-unit-tests PATCH v1 2/7] x86 UEFI: Refactor set up process
Quoted 25 lines · Zixuan Wang
lib/x86/setup.c +24
-0+efi_status_t setup_efi (efi_bootinfo_t *efi_bootinfo) { +efi_status_t status; + + status =setup_memory_allocator (efi_bootinfo); +if (status !=EFI_SUCCESS ) { +printf ("Failed to set up memory allocator: " ); +switch (status) { +case EFI_OUT_OF_RESOURCES : +printf ("No free memory region\n" ); +break ; +default : +printf ("Unknown error\n" ); +break ; + } +return status; + } + + status =setup_rsdp (efi_bootinfo); +if (status !=EFI_SUCCESS ) { +printf ("Cannot find RSDP in EFI system table\n" ); +return status; + } + + status =setup_amd_sev ();
lib/x86/setup.c static void setup_gdt_tss(void) · line 315
Quoted 10 lines
lib/x86/setup.c +10
-0+if (status !=EFI_SUCCESS ) { +switch (status) { +case EFI_UNSUPPORTED : +/* Continue if AMD SEV is not supported */ +break ; +default : +printf ("Set up AMD SEV failed\n" ); +return status; + } + }
Looks like this is pre-existing behavior, but the switch is quite gratuituous, and arguably does the wrong thing for EFI_UNSUPPORTED here as attempting to setup SEV-ES without SEV is guaranteed to fail. And it'd be really nice if the printf() actually provided the error (below might be wrong, I don't know the type of efi_status-t).
status = setup_amd_sev();
/* Continue on if AMD SEV isn't supported, but skip SEV-ES setup. */
if (status == EFI_UNSUPPORTED)
goto continue_setup;
if (status != EFI_SUCCESS) {
printf("AMD SEV setup failed, error = %d\n", status);
return status;
}
/* Same as above, lack of SEV-ES is not a fatal error. */
status = setup_amd_sev_es();
if (status != EFI_SUCCESS && status != EFI_UNSUPPORTED) {
printf("AMD SEV-ES setup failed, error = %d\n", status);
return status;
}continue_setup:
Quoted 31 lines
lib/x86/setup.c +15
-2+ + status =setup_amd_sev_es (); +if (status !=EFI_SUCCESS ) { +switch (status) { +case EFI_UNSUPPORTED : +/* Continue if AMD SEV-ES is not supported */ +break ; +default : +printf ("Set up AMD SEV-ES failed\n" ); +return status; + } + } +reset_apic ();setup_gdt_tss ();setup_idt ();@@ -343,9 +334,9 @@ void setup_efi(efi_bootinfo_t *efi_bootinfo)enable_apic ();enable_x2apic ();smp_init (); -phys_alloc_init (efi_bootinfo->free_mem_start, efi_bootinfo->free_mem_size); -setup_efi_rsdp (efi_bootinfo->rsdp);setup_page_table (); + + return EFI_SUCCESS; }#endif /* TARGET_EFI */ Signature · 1 line
2.33.0
Re: [kvm-unit-tests PATCH v1 2/7] x86 UEFI: Refactor set up process
Quoted 62 lines · Sean Christopherson
Quoted 25 lines · Zixuan Wang
lib/x86/setup.c +24
-0+efi_status_t setup_efi (efi_bootinfo_t *efi_bootinfo) { +efi_status_t status; + + status =setup_memory_allocator (efi_bootinfo); +if (status !=EFI_SUCCESS ) { +printf ("Failed to set up memory allocator: " ); +switch (status) { +case EFI_OUT_OF_RESOURCES : +printf ("No free memory region\n" ); +break ; +default : +printf ("Unknown error\n" ); +break ; + } +return status; + } + + status =setup_rsdp (efi_bootinfo); +if (status !=EFI_SUCCESS ) { +printf ("Cannot find RSDP in EFI system table\n" ); +return status; + } + + status =setup_amd_sev ();lib/x86/setup.c static void setup_gdt_tss(void) · line 315
Quoted 10 lines
lib/x86/setup.c +10
-0+if (status !=EFI_SUCCESS ) { +switch (status) { +case EFI_UNSUPPORTED : +/* Continue if AMD SEV is not supported */ +break ; +default : +printf ("Set up AMD SEV failed\n" ); +return status; + } + }Looks like this is pre-existing behavior, but the switch is quite gratuituous, and arguably does the wrong thing for
EFI_UNSUPPORTEDhere as attempting to setup SEV-ES without SEV is guaranteed to fail. And it'd be really nice if theprintf()actually provided the error (below might be wrong, I don't know the type of efi_status-t).status = setup_amd_sev(); /* Continue on if AMD SEV isn't supported, but skip SEV-ES setup. */ if (status == EFI_UNSUPPORTED) goto continue_setup; if (status != EFI_SUCCESS) { printf("AMD SEV setup failed, error = %d\n", status); return status; } /* Same as above, lack of SEV-ES is not a fatal error. */ status = setup_amd_sev_es(); if (status != EFI_SUCCESS && status != EFI_UNSUPPORTED) { printf("AMD SEV-ES setup failed, error = %d\n", status); return status; }
Quoted 2 lines
continue_setup:
I agree. The current setup_amd_sev_es() checks if SEV is available and returns EFI_UNSUPPORTED if not, so it does no harm if called after setup_amd_sev() fails. But I think we should not rely on these underlying implementation details. I will include this part in the next version.
Best regards,
Zixuan
PATCH 3/7 x86 UEFI: Convert x86 test cases to PIC
In reply to Zixuan WangFrom: Zixuan Wang <[email protected]>
UEFI loads EFI applications to dynamic runtime addresses, so it requires all applications to be compiled as PIC (position independent code). PIC does not allow the usage of compile time absolute address.
This commit converts multiple x86 test cases to PIC so they can compile and run in UEFI:
- x86/cet.efi
- x86/emulator.c: x86/emulator.c depends on lib/x86/usermode.c. But usermode.c contains non-PIC inline assembly code. This commit converts lib/x86/usermode.c and x86/emulator.c to PIC, so x86/emulator.c can compile and run in UEFI.
- x86/vmware_backdoors.c: it depends on lib/x86/usermode.c and now works without modifications
- x86/eventinj.c
- x86/smap.c
- x86/access.c
- x86/umip.c
--- lib/x86/usermode.c | 3 ++- x86/Makefile.common | 10 +++++----- x86/Makefile.x86_64 | 7 ++++--- x86/access.c | 9 +++++---- x86/cet.c | 8 +++++--- x86/emulator.c | 5 +++-- x86/eventinj.c | 8 ++++++++ x86/smap.c | 13 +++++++++---- x86/umip.c | 26 +++++++++++++++++++++++--- 9 files changed, 64 insertions(+), 25 deletions(-)
lib/x86/usermode.c +2 -1
@@ -58,7 +58,8 @@ uint64_t run_in_user(usermode_func func, unsigned int fault_vector,"pushq %[user_stack_top]\n\t" "pushfq\n\t" "pushq %[user_cs]\n\t" -"pushq $user_mode\n\t" +"lea user_mode(%%rip), %%rdx\n\t" +"pushq %%rdx\n\t" "iretq\n" "user_mode:\n\t"
x86/Makefile.common +5 -5
@@ -81,16 +81,16 @@ tests-common = $(TEST_DIR)/vmexit.$(exe) $(TEST_DIR)/tsc.$(exe) \ $(TEST_DIR)/init.$(exe) \ $(TEST_DIR)/hyperv_synic.$(exe) $(TEST_DIR)/hyperv_stimer.$(exe) \ $(TEST_DIR)/hyperv_connections.$(exe) \ - $(TEST_DIR)/tsx-ctrl.$(exe) + $(TEST_DIR)/tsx-ctrl.$(exe) \ + $(TEST_DIR)/eventinj.$(exe) \ + $(TEST_DIR)/umip.$(exe) # The following test cases are disabled when building EFI tests because they # use absolute addresses in their inline assembly code, which cannot compile # with the '-fPIC' flag ifneq ($(TARGET_EFI),y) -tests-common += $(TEST_DIR)/eventinj.$(exe) \ - $(TEST_DIR)/smap.$(exe) \ - $(TEST_DIR)/realmode.$(exe) \ - $(TEST_DIR)/umip.$(exe) +tests-common += $(TEST_DIR)/smap.$(exe) \ + $(TEST_DIR)/realmode.$(exe) endif test_cases: $(tests-common) $(tests)
x86/Makefile.x86_64 +4 -3
@@ -30,6 +30,8 @@ tests += $(TEST_DIR)/intel-iommu.$(exe) tests += $(TEST_DIR)/rdpru.$(exe) tests += $(TEST_DIR)/pks.$(exe) tests += $(TEST_DIR)/pmu_lbr.$(exe) +tests += $(TEST_DIR)/emulator.$(exe) +tests += $(TEST_DIR)/vmware_backdoors.$(exe) ifeq ($(TARGET_EFI),y) tests += $(TEST_DIR)/amd_sev.$(exe)
@@ -40,14 +42,13 @@ endif # with the '-fPIC' flag ifneq ($(TARGET_EFI),y) tests += $(TEST_DIR)/access.$(exe) -tests += $(TEST_DIR)/emulator.$(exe) tests += $(TEST_DIR)/svm.$(exe) tests += $(TEST_DIR)/vmx.$(exe) -tests += $(TEST_DIR)/vmware_backdoors.$(exe) +endif + ifneq ($(fcf_protection_full),) tests += $(TEST_DIR)/cet.$(exe) endif -endif include $(SRCDIR)/$(TEST_DIR)/Makefile.common
x86/access.c +5 -4
@@ -700,7 +700,7 @@ static int ac_test_do_access(ac_test_t *at)if (F (AC_ACCESS_TWICE )) { asmvolatile ( -"mov $fixed2, %%rsi \n\t" +"lea fixed2(%%rip), %%rsi \n\t" "mov (%[addr]), %[reg] \n\t" "fixed2:" : [reg]"=r"(r), [fault]"=a"(fault), "=b"(e)
@@ -710,7 +710,7 @@ static int ac_test_do_access(ac_test_t *at) fault =0 ; } - asmvolatile ("mov $fixed1, %%rsi \n\t" + asmvolatile ("lea fixed1(%%rip), %%rsi \n\t" "mov %%rsp, %[rsp0] \n\t" "cmp $0, %[user] \n\t" "jz do_access \n\t"
@@ -719,7 +719,8 @@ static int ac_test_do_access(ac_test_t *at)"pushq %[user_stack_top] \n\t" "pushfq \n\t" "pushq %[user_cs] \n\t" -"pushq $do_access \n\t" +"lea do_access(%%rip), %%r8\n\t" +"pushq %%r8\n\t" "iretq \n" "do_access: \n\t" "cmp $0, %[fetch] \n\t"
@@ -750,7 +751,7 @@ static int ac_test_do_access(ac_test_t *at) [user_cs]"i"(USER_CS), [user_stack_top]"r"(user_stack + sizeof user_stack), [kernel_entry_vector]"i"(0x20) - : "rsi"); + : "rsi", "r8"); asm volatile (".section .text.pf \n\t" "page_fault: \n\t"
x86/cet.c +5 -3
@@ -52,7 +52,7 @@ static u64 cet_ibt_func(void)printf ("No endbr64 instruction at jmp target, this triggers #CP...\n" ); asmvolatile ("movq $2, %rcx\n" "dec %rcx\n" -"leaq 2f, %rax\n" +"leaq 2f(%rip), %rax\n" "jmp *%rax \n" "2:\n" "dec %rcx\n" );
@@ -67,7 +67,8 @@ void test_func(void) {"pushq %[user_stack_top]\n\t" "pushfq\n\t" "pushq %[user_cs]\n\t" -"pushq $user_mode\n\t" +"lea user_mode(%%rip), %%rax\n\t" +"pushq %%rax\n\t" "iretq\n" "user_mode:\n\t"
@@ -77,7 +78,8 @@ void test_func(void) { [user_ds]"i"(USER_DS), [user_cs]"i"(USER_CS), [user_stack_top]"r"(user_stack + - sizeof(user_stack))); + sizeof(user_stack)) + : "rax"); }#define SAVE_REGS () \
x86/emulator.c +3 -2
@@ -262,12 +262,13 @@ static void test_pop(void *mem) asmvolatile ("mov %%rsp, %[tmp] \n\t" "mov %[stack_top], %%rsp \n\t" -"push $1f \n\t" +"lea 1f(%%rip), %%rax \n\t" +"push %%rax \n\t" "ret \n\t" "2: jmp 2b \n\t" "1: mov %[tmp], %%rsp" : [tmp]"=&r" (tmp) : [stack_top]"r" (stack_top) - :"memory" ); + :"memory" ,"rax" );report_pass ("ret" ); stack_top[-1 ] =0x778899 ;
x86/eventinj.c +8 -0
@@ -155,9 +155,17 @@ asm("do_iret:" "pushf"W" \n\t" "mov %cs, %ecx \n\t" "push"W" %"R "cx \n\t" +#ifndef __x86_64__ "push"W" $2f \n\t" "cmpb $0, no_test_device\n\t" // see if need to flush +#else + "leaq 2f(%rip), %rbx \n\t" + "pushq %rbx \n\t" + + "mov no_test_device(%rip), %bl \n\t" + "cmpb $0, %bl\n\t" // see if need to flush +#endif"jnz 1f\n\t" "outl %eax, $0xe4 \n\t" // flush page "1: \n\t"
x86/smap.c +9 -4
@@ -159,12 +159,17 @@ int main(int ac, char **av)init_test (i);stac (); test =-1 ; +#ifndef __x86_64__ +#define TEST "test" +#else +#define TEST "test(%rip)" +#endif asm("or $(" xstr(USER_BASE) "), %"R "sp \n""push $44 \n " -"decl test\n" +"decl " TEST "\n" "and $~(" xstr(USER_BASE) "), %"R "sp \n""pop %" R "ax\n" - "movl %eax, test"); + "movl %eax, " TEST "\n");report (pf_count ==0 && test ==44 ,"write to user stack with AC=1" );
@@ -173,10 +178,10 @@ int main(int ac, char **av) test =-1 ; asm("or $(" xstr(USER_BASE) "), %"R "sp \n""push $45 \n " -"decl test\n" +"decl " TEST "\n" "and $~(" xstr(USER_BASE) "), %"R "sp \n""pop %" R "ax\n" - "movl %eax, test"); + "movl %eax, " TEST "\n");report (pf_count ==1 && test ==45 && save ==-1 ,"write to user stack with AC=0" );
x86/umip.c +23 -3
@@ -20,10 +20,20 @@ static void gp_handler(struct ex_regs *regs) } } +#ifndef __x86_64__ +#define GP_ASM_MOVE_TO_RIP \ + "mov" W " $1f, %[expected_rip]\n\t" +#else +#define GP_ASM_MOVE_TO_RIP \ + "pushq %%rax\n\t" \ + "lea 1f(%%rip), %%rax\n\t" \ + "mov %%rax, %[expected_rip]\n\t" \ + "popq %%rax\n\t" +#endif #define GP_ASM (stmt, in, clobber) \ asm volatile ( \ - "mov" W " $1f, %[expected_rip]\n\t" \ + GP_ASM_MOVE_TO_RIP \ "movl $2f-1f, %[skip_count]\n\t" \ "1: " stmt "\n\t" \ "2: " \
@@ -125,12 +135,18 @@ static noinline int do_ring3(void (*fn)(const char *), const char *arg) "mov %%dx, %%fs\n\t" "mov %%dx, %%gs\n\t" "mov %%" R "sp, %[sp0]\n\t" /* kernel sp for exception handlers */ + "mov %[sp0], %%" R "bx\n\t" /* ebx/rbx is preserved before and after 'call' instruction */ "push" W " %%" R "dx \n\t" "lea %[user_stack_top], %%" R "dx \n\t" "push" W " %%" R "dx \n\t" "pushf" W "\n\t" "push" W " %[user_cs] \n\t" +#ifndef __x86_64__ "push" W " $1f \n\t" +#else + "lea 1f(%%rip), %%rdx \n\t" + "pushq %%rdx \n\t" +#endif "iret" W "\n" "1: \n\t"#ifndef __x86_64__
@@ -140,12 +156,16 @@ static noinline int do_ring3(void (*fn)(const char *), const char *arg)#ifndef __x86_64__ "pop %%ecx\n\t"#endif +#ifndef __x86_64__ "mov $1f, %%" R "dx\n\t" +#else + "lea 1f(%%" R "ip), %%" R "dx\n\t" +#endif "int %[kernel_entry_vector]\n\t" ".section .text.entry \n\t" "kernel_entry: \n\t" #ifdef __x86_64__ - "mov %[sp0], %%" R "sp\n\t" + "mov %%rbx, %%rsp\n\t" #else"add $(5 * " S "), %%esp\n\t" #endif
@@ -171,7 +191,7 @@ static noinline int do_ring3(void (*fn)(const char *), const char *arg) [arg]"D"(arg), [kernel_ds]"i"(KERNEL_DS), [kernel_entry_vector]"i"(0x20) - : "rcx", "rdx"); + : "rcx", "rdx", "rbx");return ret; }
Signature · 1 line
2.33.0
PATCH 4/7 x86 UEFI: Set UEFI OVMF as readonly
In reply to Zixuan WangFrom: Zixuan Wang <[email protected]>
Set readonly for UEFI OVMF image.
--- x86/efi/run | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
x86/efi/run +1 -1
@@ -54,7 +54,7 @@ popd || exit 2 # UEFI's largest allocatable memory region is large enough. EFI_RUN=y \ "$TEST_DIR/run" \ - -drive file="$EFI_UEFI",format=raw,if=pflash \ + -drive file="$EFI_UEFI",format=raw,if=pflash,readonly=on \ -drive file.dir="$EFI_TEST/$EFI_CASE/",file.driver=vvfat,file.rw=on,format=raw,if=virtio \ -net none \ -nographic \
Signature · 1 line
2.33.0
PATCH 5/7 x86 UEFI: Exit QEMU with return code
In reply to Zixuan WangFrom: Zixuan Wang <[email protected]>
kvm-unit-tests runner scripts parse QEMU exit code to determine if a test case runs successfully. But the UEFI 'reset_system' function always exits QEMU with code 0, even if the test case returns a non-zero code.
This commit fixes this issue by replacing the 'reset_system' call with an 'exit' call, which ensures QEMU exit with the correct code.
--- lib/efi.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
lib/efi.c +3 -3
@@ -87,7 +87,7 @@ efi_status_t efi_get_system_config_table(efi_guid_t table_guid, void **table)efi_status_t efi_main (efi_handle_t handle,efi_system_table_t *sys_tab) { -int ret; +unsigned long ret;efi_status_t status;efi_bootinfo_t efi_bootinfo;
@@ -134,14 +134,14 @@ efi_status_t efi_main(efi_handle_t handle, efi_system_table_t *sys_tab) ret =main (__argc, __argv, __environ);/* Shutdown the guest VM */ -efi_rs_call (reset_system,EFI_RESET_SHUTDOWN , ret,0 ,NULL ); +exit (ret);/* Unreachable */ return EFI_UNSUPPORTED ; efi_main_error:/* Shutdown the guest with error EFI status */ -efi_rs_call (reset_system,EFI_RESET_SHUTDOWN , status,0 ,NULL ); +exit (status);/* Unreachable */ return EFI_UNSUPPORTED ;
Signature · 1 line
2.33.0
Paolo Bonzini
maintainerRe: [kvm-unit-tests PATCH v1 5/7] x86 UEFI: Exit QEMU with return code
Quoted 19 lines · Zixuan Wang
From: Zixuan Wang <[email protected]>
kvm-unit-tests runner scripts parse QEMU exit code to determine if a test case runs successfully. But the UEFI '
reset_system' function always exits QEMU with code 0, even if the test case returns a non-zero code.This commit fixes this issue by replacing the '
reset_system' call with an 'exit' call, which ensures QEMU exit with the correct code.Signed-off-by Zixuan Wang <[email protected]>--- lib/efi.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)lib/efi.c +0
-0
lib/efi.c efi_status_t efi_get_system_config_table(efi_guid_t table_guid, void **table) · line 90
Quoted 5 lines
lib/efi.c +1
-1@@ -87,7 +87,7 @@ efi_status_t efi_get_system_config_table(efi_guid_t table_guid, void **table)efi_status_t efi_main (efi_handle_t handle,efi_system_table_t *sys_tab) { -int ret; +unsigned long ret;
Why this change?
Quoted 10 lines
lib/efi.c +1
-1efi_status_t status;efi_bootinfo_t efi_bootinfo;@@ -134,14 +134,14 @@ efi_status_t efi_main(efi_handle_t handle, efi_system_table_t *sys_tab) ret =main (__argc, __argv, __environ);/* Shutdown the guest VM */ -efi_rs_call (reset_system,EFI_RESET_SHUTDOWN , ret,0 ,NULL ); +exit (ret);
lib/efi.c efi_status_t efi_main(efi_handle_t handle, efi_system_table_t *sys_tab) · line 147
Quoted 10 lines
lib/efi.c +1
-1/* Unreachable */ return EFI_UNSUPPORTED ; efi_main_error:/* Shutdown the guest with error EFI status */ -efi_rs_call (reset_system,EFI_RESET_SHUTDOWN , status,0 ,NULL ); +exit (status);/* Unreachable */ return EFI_UNSUPPORTED ;
It's better to keep the exit() *and* the efi_rs_call(), I think, in case the testdev is missing and therefore the exit() does not work.
Paolo
Re: [kvm-unit-tests PATCH v1 5/7] x86 UEFI: Exit QEMU with return code
Quoted 9 lines · Paolo Bonzini
lib/efi.c efi_status_t efi_get_system_config_table(efi_guid_t table_guid, void **table) · line 90
Quoted 5 lines · Zixuan Wang
From: Zixuan Wang <[email protected]>
efi_status_tefi_main(efi_handle_t handle, efi_system_table_t *sys_tab){lib/efi.c +1
-1-int ret; +unsigned long ret;Why this change?
Didn't notice this, it should be int, thanks for pointing it out!
Quoted 18 lines
Quoted 18 lines
efi_status_t status; efi_bootinfo_t efi_bootinfo;lib/efi.c +2
-2@@ -134,14 +134,14 @@ efi_status_t efi_main(efi_handle_t handle, efi_system_table_t *sys_tab) ret =main (__argc, __argv, __environ);/* Shutdown the guest VM */ -efi_rs_call (reset_system,EFI_RESET_SHUTDOWN , ret,0 ,NULL ); +exit (ret);/* Unreachable */ return EFI_UNSUPPORTED ; efi_main_error:/* Shutdown the guest with error EFI status */ -efi_rs_call (reset_system,EFI_RESET_SHUTDOWN , status,0 ,NULL ); +exit (status);
Quoted 8 lines
Review efi_status_t efi_main(efi_handle_t handle, efi_system_table_t *sys_tab) · line 147
Quoted 2 lines
(quoted diff) +0
-0/* Unreachable */ return EFI_UNSUPPORTED;It's better to keep the
exit()*and* theefi_rs_call(), I think, in case the testdev is missing and therefore theexit()does not work.Paolo
I agree, I think there are three possible solutions:
1. keep both exit() and efi_rs_call() here, or
2. define a new function efi_exit() that calls both exit() and efi_rs_call(), or
3. add efi_rs_call() to the end of exit() function (defined in lib/x86/io.c), so many other calls to exit() can utilize EFI exit as a backup
Best regards,
Zixuan
PATCH 6/7 scripts: Generalize EFI check
In reply to Zixuan WangFrom: Marc Orr <[email protected]>
Previously, the scripts distinguish between seabios and UEFI via a hard-coded env var in the EFI run script, arch/x86/efi/run. Furthermore, this var is passed to the x86 run script, arch/x86/run, and then not available in other scripts (or to other architectures).
Replace the previous approach with a common helper function to check whether the repo has been configured to run under EFI. The helper does this by probing the config.mak file generated by configure.
--- scripts/common.bash | 5 +++++ x86/efi/run | 1 - x86/run | 6 ++++-- 3 files changed, 9 insertions(+), 3 deletions(-)
scripts/common.bash +5 -0
@@ -1,5 +1,10 @@ source config.mak +function running_under_efi() +{ + [[ ${TARGET_EFI} == "y" ]] && echo "y" || echo "" +} + function for_each_unittest() { local unittests="$1"
x86/efi/run +0 -1
@@ -52,7 +52,6 @@ popd || exit 2 # run in UEFI, some test cases, e.g. `x86/pmu.c`, require more free memory. A # simple fix is to increase the QEMU default memory size to 256MiB so that # UEFI's largest allocatable memory region is large enough. -EFI_RUN=y \ "$TEST_DIR/run" \ -drive file="$EFI_UEFI",format=raw,if=pflash,readonly=on \ -drive file.dir="$EFI_TEST/$EFI_CASE/",file.driver=vvfat,file.rw=on,format=raw,if=virtio \
x86/run +4 -2
@@ -1,5 +1,7 @@ #!/usr/bin/env bash +source scripts/common.bash + if [ -z "$STANDALONE" ]; then if [ ! -f config.mak ]; then echo "run ./configure && make first. See ./configure -h"
@@ -39,12 +41,12 @@ fi command="${qemu} --no-reboot -nodefaults $pc_testdev -vnc none -serial stdio $pci_testdev" command+=" -machine accel=$ACCEL" -if ! [ "$EFI_RUN" ]; then +if ! [ "$(running_under_efi)" ]; then command+=" -kernel" fi command="$(timeout_cmd) $command" -if [ "$EFI_RUN" ]; then +if [ "$(running_under_efi)" ]; then # Set ENVIRON_DEFAULT=n to remove '-initrd' flag for QEMU (see # 'scripts/arch-run.bash' for more details). This is because when using # UEFI, the test case binaries are passed to QEMU through the disk
Signature · 1 line
2.33.0
Paolo Bonzini
maintainerRe: [kvm-unit-tests PATCH v1 6/7] scripts: Generalize EFI check
Quoted 10 lines · Zixuan Wang
From: Marc Orr<[email protected]>
Previously, the scripts distinguish between seabios and UEFI via a hard-coded env var in the EFI run script,
arch/x86/efi/run. Furthermore, this var is passed to the x86 run script,arch/x86/run, and then not available in other scripts (or to other architectures).Replace the previous approach with a common helper function to check whether the repo has been configured to run under EFI. The helper does this by probing the
config.makfile generated byconfigure.
It should be possible to just use
[ "${TARGET_EFI}" == "y" ]as the test:
x86/efi/run +0 -1
@@ -52,7 +52,6 @@ popd || exit 2 # run in UEFI, some test cases, e.g. `x86/pmu.c`, require more free memory. A # simple fix is to increase the QEMU default memory size to 256MiB so that # UEFI's largest allocatable memory region is large enough. -EFI_RUN=y \ "$TEST_DIR/run" \ -drive file="$EFI_UEFI",format=raw,if=pflash,readonly=on \ -drive file.dir="$EFI_TEST/$EFI_CASE/",file.driver=vvfat,file.rw=on,format=raw,if=virtio \
x86/run +2 -2
@@ -39,12 +39,12 @@ fi command="${qemu} --no-reboot -nodefaults $pc_testdev -vnc none -serial stdio $pci_testdev" command+=" -machine accel=$ACCEL" -if ! [ "$EFI_RUN" ]; then +if [ ${TARGET_EFI} != "y" ]; then command+=" -kernel" fi command="$(timeout_cmd) $command" -if [ "$EFI_RUN" ]; then +if [ ${TARGET_EFI} = "y" ]; then # Set ENVIRON_DEFAULT=n to remove '-initrd' flag for QEMU (see # 'scripts/arch-run.bash' for more details). This is because when using # UEFI, the test case binaries are passed to QEMU through the disk
Paolo
Re: [kvm-unit-tests PATCH v1 6/7] scripts: Generalize EFI check
Quoted 51 lines · Paolo Bonzini
Quoted 10 lines · Zixuan Wang
From: Marc Orr<[email protected]>
Previously, the scripts distinguish between seabios and UEFI via a hard-coded env var in the EFI run script,
arch/x86/efi/run. Furthermore, this var is passed to the x86 run script,arch/x86/run, and then not available in other scripts (or to other architectures).Replace the previous approach with a common helper function to check whether the repo has been configured to run under EFI. The helper does this by probing the
config.makfile generated byconfigure.It should be possible to just use
[ "${TARGET_EFI}" == "y" ]as the test:
x86/efi/run +0
-1@@ -52,7 +52,6 @@ popd || exit 2 # run in UEFI, some test cases, e.g. `x86/pmu.c`, require more free memory. A # simple fix is to increase the QEMU default memory size to 256MiB so that # UEFI's largest allocatable memory region is large enough. -EFI_RUN=y \ "$TEST_DIR/run" \ -drive file="$EFI_UEFI",format=raw,if=pflash,readonly=on \ -drive file.dir="$EFI_TEST/$EFI_CASE/",file.driver=vvfat,file.rw=on,format=raw,if=virtio \x86/run +2
-2@@ -39,12 +39,12 @@ fi command="${qemu} --no-reboot -nodefaults $pc_testdev -vnc none -serial stdio $pci_testdev" command+=" -machine accel=$ACCEL" -if ! [ "$EFI_RUN" ]; then +if [ ${TARGET_EFI} != "y" ]; then command+=" -kernel" fi command="$(timeout_cmd) $command" -if [ "$EFI_RUN" ]; then +if [ ${TARGET_EFI} = "y" ]; then # Set ENVIRON_DEFAULT=n to remove '-initrd' flag for QEMU (see # 'scripts/arch-run.bash' for more details). This is because when using # UEFI, the test case binaries are passed to QEMU through the disk
Quoted 2 lines
Paolo
Agreed. That SGTM.
PATCH 7/7 x86 UEFI: Make run_tests.sh (mostly) work under UEFI
In reply to Zixuan WangFrom: Marc Orr <[email protected]>
Make several fixes so that run_tests.sh can run the test cases when the repo is configured to run under UEFI. Specifically:
1. The scripts are enlighted to run the .efi test cases rather than the .flat test cases.
2. The scripts are enlighted to run an empty file, _NO_FILE_4Uhere_, which is used to sanity check QEMU and QEMU flags.
3. The scripts are updated to always run with -smp 1, since SMP is not yet supported under UEFI.
Notably, QEMU's -append flag still does not work. This will need to be fixed in future commits. For now, test cases that use -append are marked SKIP.
--- scripts/common.bash | 4 ++-- scripts/runtime.bash | 15 +++++++++++++-- x86/efi/run | 24 ++++++++++++++++++++---- 3 files changed, 35 insertions(+), 8 deletions(-)
scripts/common.bash +2 -2
@@ -26,7 +26,7 @@ function for_each_unittest() if [[ "$line" =~ ^\[(.*)\]$ ]]; then rematch=${BASH_REMATCH[1]} if [ -n "${testname}" ]; then - $(arch_cmd) "$cmd" "$testname" "$groups" "$smp" "$kernel" "$opts" "$arch" "$check" "$accel" "$timeout" + $(arch_cmd) "$cmd" "$testname" "$groups" "$smp" "$kernel" "$opts" "$arch" "$check" "$accel" "$timeout" "$(running_under_efi)" fi testname=$rematch smp=1
@@ -56,7 +56,7 @@ function for_each_unittest() fi done if [ -n "${testname}" ]; then - $(arch_cmd) "$cmd" "$testname" "$groups" "$smp" "$kernel" "$opts" "$arch" "$check" "$accel" "$timeout" + $(arch_cmd) "$cmd" "$testname" "$groups" "$smp" "$kernel" "$opts" "$arch" "$check" "$accel" "$timeout" "$(running_under_efi)" fi exec {fd}<&- }
scripts/runtime.bash +13 -2
@@ -81,6 +81,11 @@ function run() local check="${CHECK:-$7}" local accel="$8" local timeout="${9:-$TIMEOUT}" # unittests.cfg overrides the default + local running_under_efi="${10}" + + if [ "$running_under_efi" ]; then + kernel=$(basename $kernel .flat) + fi if [ -z "$testname" ]; then return
@@ -127,8 +132,14 @@ function run() fi last_line=$(premature_failure > >(tail -1)) && { - print_result "SKIP" $testname "" "$last_line" - return 77 + skip=true + if [ "${running_under_efi}" ] && [[ "${last_line}" =~ "Reset" ]]; then + skip=false + fi + if [ ${skip} == true ]; then + print_result "SKIP" $testname "" "$last_line" + return 77 + fi } cmdline=$(get_cmdline $kernel)
x86/efi/run +20 -4
@@ -34,16 +34,27 @@ shift 1 # This host dir will be loaded by QEMU as a FAT32 image # - Make UEFI startup script that runs the .efi on boot mkdir -p "$EFI_TEST/$EFI_CASE/" -cp "$EFI_SRC/$EFI_CASE.efi" "$EFI_TEST/$EFI_CASE/" +if [ $EFI_CASE != "_NO_FILE_4Uhere_" ]; then + cp "$EFI_SRC/$EFI_CASE.efi" "$EFI_TEST/$EFI_CASE/" +else + touch "$EFI_TEST/$EFI_CASE/$EFI_CASE.efi" +fi pushd "$EFI_TEST/$EFI_CASE" || exit 2 # 'startup.nsh' is the default script executed by UEFI on boot # Use this script to run the test binary automatically -cat << EOF >startup.nsh +if [ $EFI_CASE != "_NO_FILE_4Uhere_" ]; then + cat << EOF >startup.nsh @echo -off fs0: "$EFI_CASE.efi" EOF +else + cat << EOF >startup.nsh +@echo -off +reset -s +EOF +fi popd || exit 2 # Run test case with 256MiB QEMU memory. QEMU default memory size is 128MiB.
@@ -52,11 +63,16 @@ popd || exit 2 # run in UEFI, some test cases, e.g. `x86/pmu.c`, require more free memory. A # simple fix is to increase the QEMU default memory size to 256MiB so that # UEFI's largest allocatable memory region is large enough. +# +# Also, pass in an EFI-specific smp count (i.e., `-smp 1`) as the last argument +# to x86/run. This `smp` flag overrides any previous `smp` flags (e.g., +# `-smp 4`). This is necessary because KVM-Unit-Tests do not currently support +# SMP under UEFI. This last flag should be removed when this issue is resolved. "$TEST_DIR/run" \ -drive file="$EFI_UEFI",format=raw,if=pflash,readonly=on \ -drive file.dir="$EFI_TEST/$EFI_CASE/",file.driver=vvfat,file.rw=on,format=raw,if=virtio \ -net none \ -nographic \ - -smp "$EFI_SMP" \ -m 256 \ - "$@" + "$@" \ + -smp "$EFI_SMP"
Signature · 1 line
2.33.0
Paolo Bonzini
maintainerQuoted 19 lines · Zixuan Wang
Hello,
This patch series refactors the
x86_64UEFI set up process and fixes therun-tests.shscript to run under UEFI. The patches are organized as three parts.The first part (patches 1-2) refactors the
x86_64UEFI set up process. The previous UEFI setup calls arch-specific setup functions twice and generates arch-specific data structure. As Andrew suggested [1], we refactor this process to make only one call to the arch-specific function and generate arch-neutral data structures. This simplifies the set up process and makes it easier to develop UEFI support for other architectures.The second part (patch 3) converts several x86 test cases to Position-Independent Code (PIC) to run under UEFI. This patch is ported from the initial UEFI support patchset [2] with fixes to the 32-bit compilation.
Quoted 5 lines
The third part (patches 4-7) fixes the UEFI runner scripts. Patch 4 sets UEFI OVMF image as readonly. Patch 5 fixes test cases' return code under UEFI, enabling Patch 6-7 to fix the
run-tests.shscript under UEFI.This patch set is based on the
uefibranch.
Thank you, for patches 1-6 I have squashed the patches when applicable (1, 4, 5, 6) and queued the others (2 and 3).
I did not queue patch 7 yet, it seems okay but I want to understand better the changes it needs in the harness and what is missing. I'll take a look during the week.
Paolo
Quoted 39 lines
Best regards,
Zixuan Wang and Marc Orr[1] https://lore.kernel.org/kvm/[email protected]/
[2] https://lore.kernel.org/kvm/[email protected]/Marc Orr (2):
scripts: Generalize EFI check
x86 UEFI: Makerun_tests.sh(mostly) work under UEFIZixuan Wang (5):
x86 UEFI: Removemixed_mode
x86 UEFI: Refactor set up process
x86 UEFI: Convert x86 test cases to PIC
x86 UEFI: Set UEFI OVMF as readonly
x86 UEFI: Exit QEMU with return codelib/efi.c | 54 ++++++-- lib/efi.h | 19 ++- lib/linux/efi.h | 317 ++++++++++++++----------------------------- lib/x86/acpi.c | 36 +++-- lib/x86/acpi.h | 5 +- lib/x86/asm/setup.h | 16 +-- lib/x86/setup.c | 153 ++++++++++----------- lib/x86/usermode.c | 3 +- scripts/common.bash | 9 +- scripts/runtime.bash | 15 +- x86/Makefile.common | 10 +- x86/Makefile.x86_64 | 7 +- x86/access.c | 9 +- x86/cet.c | 8 +- x86/efi/run | 27 +++- x86/emulator.c | 5 +- x86/eventinj.c | 8 ++ x86/run | 6 +- x86/smap.c | 13 +- x86/umip.c | 26 +++- 20 files changed, 360 insertions(+), 386 deletions(-)
Quoted 25 lines · Paolo Bonzini
Quoted 23 lines · Zixuan Wang
Hello,
This patch series refactors the
x86_64UEFI set up process and fixes therun-tests.shscript to run under UEFI. The patches are organized as three parts.The first part (patches 1-2) refactors the
x86_64UEFI set up process. The previous UEFI setup calls arch-specific setup functions twice and generates arch-specific data structure. As Andrew suggested [1], we refactor this process to make only one call to the arch-specific function and generate arch-neutral data structures. This simplifies the set up process and makes it easier to develop UEFI support for other architectures.The second part (patch 3) converts several x86 test cases to Position-Independent Code (PIC) to run under UEFI. This patch is ported from the initial UEFI support patchset [2] with fixes to the 32-bit compilation.
The third part (patches 4-7) fixes the UEFI runner scripts. Patch 4 sets UEFI OVMF image as readonly. Patch 5 fixes test cases' return code under UEFI, enabling Patch 6-7 to fix the
run-tests.shscript under UEFI.
Quoted 8 lines
Quoted 1 line
This patch set is based on the
uefibranch.Thank you, for patches 1-6 I have squashed the patches when applicable (1, 4, 5, 6) and queued the others (2 and 3).
I did not queue patch 7 yet, it seems okay but I want to understand better the changes it needs in the harness and what is missing. I'll take a look during the week.
SGTM, thank you! Zixuan and I discussed a few things that are missing:
1. Test cases that take the -append arg are currently marked SKIP. Two issues need to be resolved here. First, we're not using QEMU's -kernel flag for EFI test cases [1]. And the -append flag does not work without the -kernel flag. I don't understand the details on why we don't use the -kernel flag myself. Maybe Zixuan can elaborate. Second, assuming we fix the first issue, then we need to enlighten the KVM-Unit-Tests under UEFI to parse kernel command line arguments and pass them down to the test cases via argv. Zixuan pointed out to me that there is some prior work from Drew [2] that we should be able to follow to make this work. So I'm hoping that Zixuan and I can work together on solving these issues to get the argument passing working next.
2. We need a way to annotate test cases in x86/unittests.cfg as known to work under SEV. I'm thinking of doing this via new (very broad) test groups in unittests.cfg. I _think_ SEV is the primary scenario we care about. However, folks may care about running the test cases under UEFI outside of SEV. For example, last time I checked, emulator runs OK under UEFI minus SEV-ES but fails under SEV-ES. And similarly, while most test cases work under UEFI minus SEV, there are a few that do mis-behave -- and it probably makes sense to document this (e.g., via annotations in unittests.cfg). Also, there are many variations of SEV (SEV, SEV-ES, SEV-SNP)... And hopefully some of this will eventually be applicable to TDX as well. So many testgroups is not a good solution. I'm not sure.
3. Multi-CPU needs to be made to work under UEFI. For now, patch #7 forces all EFI test cases to run with 1 vCPU. I chatted with Brijesh, and he mentioned that Varad would like to work on this. However, if anything here changes, please let me know, because we can work on this as well. But for now, I'm not planning to work on it so we can avoid duplicating work.
4. UEFI runs a lot slower than SEABIOS. It doesn't help that the test harness launches QEMU more than once for each test case (i.e., it runs the _NO_FILE_4Uhere_ scenario to check QEMU arguments). I'm not sure how much of an issue this is in practice. Depending on the answer, I know Zixuan had some ideas on how to speed this up in the current test harness. Or maybe we can explore an alternative to the _NO_FILE_4Uhere_ approach instead.
Zixuan: Please add/correct anything as needed!
[1] https://gitlab.com/kvm-unit-tests/kvm-unit-tests/-/blob/uefi/x86/run#L42-44
[2] https://github.com/rhdrjones/kvm-unit-tests/blob/target-efi/scripts/mkefi.sh
Thanks,
Marc
Quoted 38 lines · Marc Orr
Quoted 25 lines · Paolo Bonzini
Quoted 23 lines · Zixuan Wang
Hello,
This patch series refactors the
x86_64UEFI set up process and fixes therun-tests.shscript to run under UEFI. The patches are organized as three parts.The first part (patches 1-2) refactors the
x86_64UEFI set up process. The previous UEFI setup calls arch-specific setup functions twice and generates arch-specific data structure. As Andrew suggested [1], we refactor this process to make only one call to the arch-specific function and generate arch-neutral data structures. This simplifies the set up process and makes it easier to develop UEFI support for other architectures.The second part (patch 3) converts several x86 test cases to Position-Independent Code (PIC) to run under UEFI. This patch is ported from the initial UEFI support patchset [2] with fixes to the 32-bit compilation.
The third part (patches 4-7) fixes the UEFI runner scripts. Patch 4 sets UEFI OVMF image as readonly. Patch 5 fixes test cases' return code under UEFI, enabling Patch 6-7 to fix the
run-tests.shscript under UEFI.Quoted 8 lines
Quoted 1 line
This patch set is based on the
uefibranch.Thank you, for patches 1-6 I have squashed the patches when applicable (1, 4, 5, 6) and queued the others (2 and 3).
I did not queue patch 7 yet, it seems okay but I want to understand better the changes it needs in the harness and what is missing. I'll take a look during the week.
SGTM, thank you! Zixuan and I discussed a few things that are missing:
Quoted 12 lines
1. Test cases that take the
-appendarg are currently markedSKIP. Two issues need to be resolved here. First, we're not using QEMU's-kernelflag for EFI test cases [1]. And the-appendflag does not work without the-kernelflag. I don't understand the details on why we don't use the-kernelflag myself. Maybe Zixuan can elaborate. Second, assuming we fix the first issue, then we need to enlighten the KVM-Unit-Tests under UEFI to parse kernel command line arguments and pass them down to the test cases viaargv. Zixuan pointed out to me that there is some prior work from Drew [2] that we should be able to follow to make this work. So I'm hoping that Zixuan and I can work together on solving these issues to get the argument passing working next.
Thank you for the detailed summary!
Current kvm-unit-tests pass an EFI binary as part of a disk image, instead of using the -kernel argument.
I just tested the -kernel argument and it seems to work with EFI binaries, and more importantly, it's really fast (bypassing the default 5-second user input waiting). I will update the x86/efi/run to use -kernel argument to pass the EFI binaries.
Since -kernel is working, I can start to investigate how to use -append to pass arguments. If that doesn't work well, an alternative approach could be:
1. (host) create a file args.txt in the disk image, which contains all the arguments needed
2. (guest) call UEFI filesystem interface to read this args.txt from the disk image, parse it and pass the arguments to main()
Quoted 4 lines
2. We need a way to annotate test cases in
x86/unittests.cfgas known to work under SEV. I'm thinking of doing this via new (very broad) test groups inunittests.cfg. I _think_ SEV is the primary scenario we care about. However, folks may care about running the test
Quoted 8 lines
cases under UEFI outside of SEV. For example, last time I checked, emulator runs OK under UEFI minus SEV-ES but fails under SEV-ES. And similarly, while most test cases work under UEFI minus SEV, there are a few that do mis-behave -- and it probably makes sense to document this (e.g., via annotations in
unittests.cfg). Also, there are many variations of SEV (SEV, SEV-ES, SEV-SNP)... And hopefully some of this will eventually be applicable to TDX as well. So many testgroups is not a good solution. I'm not sure.
Adding an efi group seems helpful. E.g., the current x86/smap.c does not work under UEFI; but the run-tests.sh still tries to run this test case, even if this test case is not compiled.
Quoted 5 lines
3. Multi-CPU needs to be made to work under UEFI. For now, patch #7 forces all EFI test cases to run with 1 vCPU. I chatted with Brijesh, and he mentioned that Varad would like to work on this. However, if anything here changes, please let me know, because we can work on this as well. But for now, I'm not planning to work on it so we can avoid
Quoted 8 lines
duplicating work.
4. UEFI runs a lot slower than SEABIOS. It doesn't help that the test harness launches QEMU more than once for each test case (i.e., it runs the_NO_FILE_4Uhere_scenario to check QEMU arguments). I'm not sure how much of an issue this is in practice. Depending on the answer, I know Zixuan had some ideas on how to speed this up in the current test harness. Or maybe we can explore an alternative to the_NO_FILE_4Uhere_approach instead.
As the -kernel argument now works with the EFI binaries and is significantly faster, this should not be an issue anymore. We just need to update the runner scripts to use -kernel argument.
Quoted 7 lines
Zixuan: Please add/correct anything as needed!
[1] https://gitlab.com/kvm-unit-tests/kvm-unit-tests/-/blob/uefi/x86/run#L42-44
[2] https://github.com/rhdrjones/kvm-unit-tests/blob/target-efi/scripts/mkefi.shThanks,
Marc
Best regards,
Zixuan
Quoted 102 lines · Zixuan Wang
Quoted 38 lines · Marc Orr
Quoted 25 lines · Paolo Bonzini
Quoted 23 lines · Zixuan Wang
Hello,
This patch series refactors the
x86_64UEFI set up process and fixes therun-tests.shscript to run under UEFI. The patches are organized as three parts.The first part (patches 1-2) refactors the
x86_64UEFI set up process. The previous UEFI setup calls arch-specific setup functions twice and generates arch-specific data structure. As Andrew suggested [1], we refactor this process to make only one call to the arch-specific function and generate arch-neutral data structures. This simplifies the set up process and makes it easier to develop UEFI support for other architectures.The second part (patch 3) converts several x86 test cases to Position-Independent Code (PIC) to run under UEFI. This patch is ported from the initial UEFI support patchset [2] with fixes to the 32-bit compilation.
The third part (patches 4-7) fixes the UEFI runner scripts. Patch 4 sets UEFI OVMF image as readonly. Patch 5 fixes test cases' return code under UEFI, enabling Patch 6-7 to fix the
run-tests.shscript under UEFI.Quoted 8 lines
Quoted 1 line
This patch set is based on the
uefibranch.Thank you, for patches 1-6 I have squashed the patches when applicable (1, 4, 5, 6) and queued the others (2 and 3).
I did not queue patch 7 yet, it seems okay but I want to understand better the changes it needs in the harness and what is missing. I'll take a look during the week.
SGTM, thank you! Zixuan and I discussed a few things that are missing:
Quoted 12 lines
1. Test cases that take the
-appendarg are currently markedSKIP. Two issues need to be resolved here. First, we're not using QEMU's-kernelflag for EFI test cases [1]. And the-appendflag does not work without the-kernelflag. I don't understand the details on why we don't use the-kernelflag myself. Maybe Zixuan can elaborate. Second, assuming we fix the first issue, then we need to enlighten the KVM-Unit-Tests under UEFI to parse kernel command line arguments and pass them down to the test cases viaargv. Zixuan pointed out to me that there is some prior work from Drew [2] that we should be able to follow to make this work. So I'm hoping that Zixuan and I can work together on solving these issues to get the argument passing working next.Thank you for the detailed summary!
Current kvm-unit-tests pass an EFI binary as part of a disk image, instead of using the
-kernelargument.I just tested the
-kernelargument and it seems to work with EFI binaries, and more importantly, it's really fast (bypassing the default 5-second user input waiting). I will update thex86/efi/runto use-kernelargument to pass the EFI binaries.Since
-kernelis working, I can start to investigate how to use-appendto pass arguments. If that doesn't work well, an alternative approach could be:1. (host) create a file
args.txtin the disk image, which contains all the arguments needed
2. (guest) call UEFI filesystem interface to read thisargs.txtfrom the disk image, parse it and pass the arguments tomain()Quoted 4 lines
2. We need a way to annotate test cases in
x86/unittests.cfgas known to work under SEV. I'm thinking of doing this via new (very broad) test groups inunittests.cfg. I _think_ SEV is the primary scenario we care about. However, folks may care about running the testQuoted 8 lines
cases under UEFI outside of SEV. For example, last time I checked, emulator runs OK under UEFI minus SEV-ES but fails under SEV-ES. And similarly, while most test cases work under UEFI minus SEV, there are a few that do mis-behave -- and it probably makes sense to document this (e.g., via annotations in
unittests.cfg). Also, there are many variations of SEV (SEV, SEV-ES, SEV-SNP)... And hopefully some of this will eventually be applicable to TDX as well. So many testgroups is not a good solution. I'm not sure.Adding an
efigroup seems helpful. E.g., the currentx86/smap.cdoes not work under UEFI; but therun-tests.shstill tries to run this test case, even if this test case is not compiled.Quoted 5 lines
3. Multi-CPU needs to be made to work under UEFI. For now, patch #7 forces all EFI test cases to run with 1 vCPU. I chatted with Brijesh, and he mentioned that Varad would like to work on this. However, if anything here changes, please let me know, because we can work on this as well. But for now, I'm not planning to work on it so we can avoid
Quoted 8 lines
duplicating work.
4. UEFI runs a lot slower than SEABIOS. It doesn't help that the test harness launches QEMU more than once for each test case (i.e., it runs the_NO_FILE_4Uhere_scenario to check QEMU arguments). I'm not sure how much of an issue this is in practice. Depending on the answer, I know Zixuan had some ideas on how to speed this up in the current test harness. Or maybe we can explore an alternative to the_NO_FILE_4Uhere_approach instead.
Quoted 3 lines
As the
-kernelargument now works with the EFI binaries and is significantly faster, this should not be an issue anymore. We just need to update the runner scripts to use-kernelargument.
You can add an additional '-kernel' + EFI binary runner if you want, but the goal of being able to run kvm-unit-tests on bare-metal means we shouldn't be counting on QEMU/OVMF to do magic stuff with the kernel. We need to build disk images. Argument passing works with EFI apps, when implemented, so that's not a problem. I also created a script that uses the framework's for_each_unittest to generate an EFI script that allowed each test to be easily run with its arguments.
Thanks,
drew
Quoted 12 lines
Quoted 8 lines
Zixuan: Please add/correct anything as needed!
[1] https://gitlab.com/kvm-unit-tests/kvm-unit-tests/-/blob/uefi/x86/run#L42-44
[2] https://github.com/rhdrjones/kvm-unit-tests/blob/target-efi/scripts/mkefi.shThanks,
MarcBest regards,
Zixuan
Quoted 116 lines · Andrew Jones
Quoted 102 lines · Zixuan Wang
Quoted 38 lines · Marc Orr
Quoted 25 lines · Paolo Bonzini
Quoted 23 lines · Zixuan Wang
Hello,
This patch series refactors the
x86_64UEFI set up process and fixes therun-tests.shscript to run under UEFI. The patches are organized as three parts.The first part (patches 1-2) refactors the
x86_64UEFI set up process. The previous UEFI setup calls arch-specific setup functions twice and generates arch-specific data structure. As Andrew suggested [1], we refactor this process to make only one call to the arch-specific function and generate arch-neutral data structures. This simplifies the set up process and makes it easier to develop UEFI support for other architectures.The second part (patch 3) converts several x86 test cases to Position-Independent Code (PIC) to run under UEFI. This patch is ported from the initial UEFI support patchset [2] with fixes to the 32-bit compilation.
The third part (patches 4-7) fixes the UEFI runner scripts. Patch 4 sets UEFI OVMF image as readonly. Patch 5 fixes test cases' return code under UEFI, enabling Patch 6-7 to fix the
run-tests.shscript under UEFI.Quoted 8 lines
Quoted 1 line
This patch set is based on the
uefibranch.Thank you, for patches 1-6 I have squashed the patches when applicable (1, 4, 5, 6) and queued the others (2 and 3).
I did not queue patch 7 yet, it seems okay but I want to understand better the changes it needs in the harness and what is missing. I'll take a look during the week.
SGTM, thank you! Zixuan and I discussed a few things that are missing:
Quoted 12 lines
1. Test cases that take the
-appendarg are currently markedSKIP. Two issues need to be resolved here. First, we're not using QEMU's-kernelflag for EFI test cases [1]. And the-appendflag does not work without the-kernelflag. I don't understand the details on why we don't use the-kernelflag myself. Maybe Zixuan can elaborate. Second, assuming we fix the first issue, then we need to enlighten the KVM-Unit-Tests under UEFI to parse kernel command line arguments and pass them down to the test cases viaargv. Zixuan pointed out to me that there is some prior work from Drew [2] that we should be able to follow to make this work. So I'm hoping that Zixuan and I can work together on solving these issues to get the argument passing working next.Thank you for the detailed summary!
Current kvm-unit-tests pass an EFI binary as part of a disk image, instead of using the
-kernelargument.I just tested the
-kernelargument and it seems to work with EFI binaries, and more importantly, it's really fast (bypassing the default 5-second user input waiting). I will update thex86/efi/runto use-kernelargument to pass the EFI binaries.Since
-kernelis working, I can start to investigate how to use-appendto pass arguments. If that doesn't work well, an alternative approach could be:1. (host) create a file
args.txtin the disk image, which contains all the arguments needed
2. (guest) call UEFI filesystem interface to read thisargs.txtfrom the disk image, parse it and pass the arguments tomain()Quoted 4 lines
2. We need a way to annotate test cases in
x86/unittests.cfgas known to work under SEV. I'm thinking of doing this via new (very broad) test groups inunittests.cfg. I _think_ SEV is the primary scenario we care about. However, folks may care about running the testQuoted 8 lines
cases under UEFI outside of SEV. For example, last time I checked, emulator runs OK under UEFI minus SEV-ES but fails under SEV-ES. And similarly, while most test cases work under UEFI minus SEV, there are a few that do mis-behave -- and it probably makes sense to document this (e.g., via annotations in
unittests.cfg). Also, there are many variations of SEV (SEV, SEV-ES, SEV-SNP)... And hopefully some of this will eventually be applicable to TDX as well. So many testgroups is not a good solution. I'm not sure.Adding an
efigroup seems helpful. E.g., the currentx86/smap.cdoes not work under UEFI; but therun-tests.shstill tries to run this test case, even if this test case is not compiled.Quoted 5 lines
3. Multi-CPU needs to be made to work under UEFI. For now, patch #7 forces all EFI test cases to run with 1 vCPU. I chatted with Brijesh, and he mentioned that Varad would like to work on this. However, if anything here changes, please let me know, because we can work on this as well. But for now, I'm not planning to work on it so we can avoid
Quoted 8 lines
duplicating work.
4. UEFI runs a lot slower than SEABIOS. It doesn't help that the test harness launches QEMU more than once for each test case (i.e., it runs the_NO_FILE_4Uhere_scenario to check QEMU arguments). I'm not sure how much of an issue this is in practice. Depending on the answer, I know Zixuan had some ideas on how to speed this up in the current test harness. Or maybe we can explore an alternative to the_NO_FILE_4Uhere_approach instead.Quoted 3 lines
As the
-kernelargument now works with the EFI binaries and is significantly faster, this should not be an issue anymore. We just need to update the runner scripts to use-kernelargument.You can add an additional '-kernel' + EFI binary runner if you want, but the goal of being able to run kvm-unit-tests on bare-metal means we shouldn't be counting on QEMU/OVMF to do magic stuff with the kernel. We need to build disk images. Argument passing works with EFI apps, when implemented, so that's not a problem. I also created a script that uses the framework's
for_each_unittestto generate an EFI script that allowed each test to be easily run with its arguments.
Quoted 2 lines
Thanks,
drew
I see, I think an alternative approach is to rename test case binaries to UEFI default binary filename, which is EFI/BOOT/BOOTX64.EFI for x86_64. This should work just like the -kernel argument. I will explore this approach with the argument passing mechanisms.
Best regards,
Zixuan