On Mon, Dec 09, 2013 at 05:42:22PM +0800, Dave Young wrote:
> Add a new setup_data type SETUP_EFI for kexec use.
> Passing the saved fw_vendor, runtime, config tables and efi runtime mappings.
>
> When entering virtual mode, directly mapping the efi runtime ragions which
> we passed in previously. And skip the step to call SetVirtualAddressMap.
>
> Specially for HP z420 workstation we need save the smbios physical address.
> The kernel boot sequence proceeds in the following order. Step 2
> requires efi.smbios to be the physical address. However, I found that on
> HP z420 EFI system table has a virtual address of SMBIOS in step 1. Hence,
> we need set it back to the physical address with the smbios in
> efi_setup_data. (When it is still the physical address, it simply sets
> the same value.)
>
> 1. efi_init() - Set efi.smbios from EFI system table
> 2. dmi_scan_machine() - Temporary map efi.smbios to access SMBIOS table
> 3. efi_enter_virtual_mode() - Map EFI ranges
>
> Tested on ovmf+qemu, lenovo thinkpad, a dell laptop and an
> HP z420 workstation.
>
> v2: refresh based on previous patch changes, code cleanup.
> v3: use ioremap instead of phys_to_virt for efi_setup
> v5: improve some code structure per comments from Matt
> Boris: improve code structure, spell fix, etc.
> Improve changelog from Toshi.
> change the variable efi_setup to the physical address of efi setup_data
> instead of the ioremapped virt address
>
> Signed-off-by: Dave Young <dyoung at redhat.com>
> ---
> arch/x86/include/asm/efi.h | 11 ++
> arch/x86/include/uapi/asm/bootparam.h | 1 +
> arch/x86/kernel/setup.c | 3 +
> arch/x86/platform/efi/efi.c | 195 ++++++++++++++++++++++++++++++----
> 4 files changed, 187 insertions(+), 23 deletions(-)
...
> @@ -115,6 +116,25 @@ static int __init setup_storage_paranoia(char *arg)
> }
> early_param("efi_no_storage_paranoia", setup_storage_paranoia);
>
> +void __init parse_efi_setup(u64 phys_addr)
> +{
> + struct setup_data *sd;
> +
> + if (!efi_enabled(EFI_64BIT)) {
> + pr_warn("SETUP_EFI not supported on 32-bit\n");
> + return;
> + }
Shouldn't this function be in two versions in efi_64.c and efi_32.c?
This way you don't need this check with cryptic printk message.
> +
> + sd = early_memremap(phys_addr, sizeof(struct setup_data));
> + if (!sd) {
> + pr_warn("efi: early_memremap setup_data failed\n");
> + return;
> + }
> + efi_setup = phys_addr + sizeof(struct setup_data);
> + nr_efi_runtime_map = (sd->len - sizeof(struct efi_setup_data)) /
> + sizeof(efi_memory_desc_t);
> + early_memunmap(sd, sizeof(struct setup_data));
> +}
>
> static efi_status_t virt_efi_get_time(efi_time_t *tm, efi_time_cap_t *tc)
> {
> @@ -494,18 +514,28 @@ static int __init efi_systab_init(void *phys)
> {
> if (efi_enabled(EFI_64BIT)) {
> efi_system_table_64_t *systab64;
> + struct efi_setup_data *data = NULL;
> u64 tmp = 0;
>
> + if (efi_setup) {
> + data = early_memremap(efi_setup, sizeof(*data));
> + if (!data)
> + return -ENOMEM;
> + }
> systab64 = early_memremap((unsigned long)phys,
> sizeof(*systab64));
> if (systab64 == NULL) {
> pr_err("Couldn't map the system table!\n");
> + if (data)
> + early_memunmap(data, sizeof(*data));
> return -ENOMEM;
> }
>
> efi_systab.hdr = systab64->hdr;
> - efi_systab.fw_vendor = systab64->fw_vendor;
> - tmp |= systab64->fw_vendor;
> +
> + efi_systab.fw_vendor = data ? (unsigned long)data->fw_vendor :
> + systab64->fw_vendor;
> + tmp |= efi_systab.fw_vendor;
> efi_systab.fw_revision = systab64->fw_revision;
> efi_systab.con_in_handle = systab64->con_in_handle;
> tmp |= systab64->con_in_handle;
> @@ -519,15 +549,20 @@ static int __init efi_systab_init(void *phys)
> tmp |= systab64->stderr_handle;
> efi_systab.stderr = systab64->stderr;
> tmp |= systab64->stderr;
> - efi_systab.runtime = (void *)(unsigned long)systab64->runtime;
> - tmp |= systab64->runtime;
> + efi_systab.runtime = data ?
> + (void *)(unsigned long)data->runtime :
> + (void *)(unsigned long)systab64->runtime;
> + tmp |= (unsigned long)efi_systab.runtime;
> efi_systab.boottime = (void *)(unsigned long)systab64->boottime;
> tmp |= systab64->boottime;
> efi_systab.nr_tables = systab64->nr_tables;
> - efi_systab.tables = systab64->tables;
> - tmp |= systab64->tables;
> + efi_systab.tables = data ? (unsigned long)data->tables :
> + systab64->tables;
> + tmp |= efi_systab.tables;
>
> early_memunmap(systab64, sizeof(*systab64));
> + if (data)
> + early_memunmap(data, sizeof(*data));
> #ifdef CONFIG_X86_32
> if (tmp >> 32) {
> pr_err("EFI data located above 4GB, disabling EFI.\n");
> @@ -631,6 +666,61 @@ static int __init efi_memmap_init(void)
> return 0;
> }
>
> +/*
> + * For kexec kernel there's some special config table entries which could be
> + * converted to virtual addresses after entering virtual mode. In kexec kernel
> + * we need the physical addresses instead, thus passing them via setup_data
> + * and update the entries to physical addresses in this function.
Rewrite:
"A number of config table entries get remapped to virtual addresses
after entering EFI virtual mode. However, the kexec kernel requires
their physical addresses therefore we pass them via setup_data and
correct those entries to their respective physical addresses here."
> + *
> + * Currently only handles smbios which is necessary for HP z420.
Didn't we say that this behavior is coming from a generic UEFI fw
implementation and if so, no need to mention z420?
> + */
> +static int __init efi_reuse_config(u64 tables, int nr_tables)
> +{
> + int i, sz, ret = 0;
> + void *p, *tablep;
> + struct efi_setup_data *data;
> +
> + if (!efi_setup)
> + return 0;
> +
> + if (!efi_enabled(EFI_64BIT))
> + return 0;
> +
> + data = early_memremap(efi_setup, sizeof(*data));
> + if (!data) {
> + ret = -ENOMEM;
> + goto out;
> + }
> +
> + if (!data->smbios)
> + goto out_memremap;
> +
> + sz = sizeof(efi_config_table_64_t);
> +
> + p = tablep = early_memremap(tables, nr_tables * sz);
> + if (!p) {
> + pr_err("Could not map Configuration table!\n");
> + ret = -ENOMEM;
> + goto out_memremap;
> + }
> +
> + for (i = 0; i < efi.systab->nr_tables; i++) {
> + efi_guid_t guid;
> +
> + guid = ((efi_config_table_64_t *)p)->guid;
> +
> + if (!efi_guidcmp(guid, SMBIOS_TABLE_GUID))
> + ((efi_config_table_64_t *)p)->table = data->smbios;
> + p += sz;
> + }
> + early_memunmap(tablep, nr_tables * sz);
> +
> +out_memremap:
> + early_memunmap(data, sizeof(*data));
> +out:
> + return ret;
> +}
> +
> void __init efi_init(void)
> {
> efi_char16_t *c16;
> @@ -676,6 +766,8 @@ void __init efi_init(void)
> efi.systab->hdr.revision >> 16,
> efi.systab->hdr.revision & 0xffff, vendor);
>
> + efi_reuse_config(efi.systab->tables, efi.systab->nr_tables);
> +
> if (efi_config_init(arch_tables))
> return;
>
> @@ -886,6 +978,50 @@ out_krealloc:
> }
>
> /*
> + * Map efi regions which was passed via setup_data. The virt_addr is a fixed
were
> + * addr which was used in first kernel in case kexec boot.
^
of a
> + */
> +static int __init map_regions_fixed(void)
> +{
> + int i, s, ret = 0;
> + u64 end, systab;
> + unsigned long size;
> + efi_memory_desc_t *md;
> + struct efi_setup_data *data;
> +
> + s = sizeof(*data) + nr_efi_runtime_map * sizeof(data->map[0]);
> + data = early_memremap(efi_setup, s);
> + if (!data) {
> + ret = -ENOMEM;
> + goto out;
> + }
newline.
> + for (i = 0, md = data->map; i < nr_efi_runtime_map; i++, md++) {
> + efi_map_region_fixed(md); /* FIXME: add error handling */
> + size = md->num_pages << PAGE_SHIFT;
> + end = md->phys_addr + size;
> +
> + systab = (u64) (unsigned long) efi_phys.systab;
> + if (md->phys_addr <= systab && systab < end) {
> + systab += md->virt_addr - md->phys_addr;
> + efi.systab = (efi_system_table_t *)(unsigned long)systab;
> + }
> + ret = save_runtime_map(md, i);
Wait a minute, this is executed in the second, kexec-ed kernel, right?
Why do we need to save the map there too?
> + if (ret)
> + goto out_save_runtime;
> + }
> +
> + early_memunmap(data, s);
> + return 0;
> +
> +out_save_runtime:
> + kfree(efi_runtime_map);
> + nr_efi_runtime_map = 0;
> + early_memunmap(data, s);
> +out:
> + return ret;
> +}
> +
> +/*
> * This function will switch the EFI runtime services to virtual mode.
> * Essentially, we look through the EFI memmap and map every region that
> * has the runtime attribute bit set in its memory descriptor into the
> @@ -901,12 +1037,16 @@ out_krealloc:
> * so that we're in a different address space when calling a runtime
> * function. For function arguments passing we do copy the PGDs of the
> * kernel page table into ->trampoline_pgd prior to each call.
> + *
> + * Specially for kexec boot, efi runtime maps in previous kernel should
> + * be passed in via setup_data. In that case runtime ranges will be mapped
> + * to the same virtual addresses exactly same as the ones in previous kernel.
"... to the same ..exactly same as ... " sounds funny. What's wrong with
"... to the same virtual addresses as the first kernel."
or if you really insist on "exact":
"... to the same exact virtual addresses as the first kernel."
> */
> void __init efi_enter_virtual_mode(void)
> {
--
Regards/Gruss,
Boris.
Sent from a fat crate under my desk. Formatting is fine.
--
