Message-ID: <20130610185435.GB14124@sgi.com>
Date:	Mon, 10 Jun 2013 13:54:36 -0500
From:	Russ Anderson <rja@....com>
To:	Matt Fleming <matt@...sole-pimps.org>
Cc:	joeyli <jlee@...e.com>,
	Matthew Garrett <matthew.garrett@...ula.com>,
	"Fleming, Matt" <matt.fleming@...el.com>,
	"mingo@...nel.org" <mingo@...nel.org>,
	"torvalds@...ux-foundation.org" <torvalds@...ux-foundation.org>,
	"bp@...en8.de" <bp@...en8.de>, "jkosina@...e.cz" <jkosina@...e.cz>,
	"linux-efi@...r.kernel.org" <linux-efi@...r.kernel.org>,
	"x86@...nel.org" <x86@...nel.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"tglx@...utronix.de" <tglx@...utronix.de>,
	"hpa@...ux.intel.com" <hpa@...ux.intel.com>,
	"akpm@...ux-foundation.org" <akpm@...ux-foundation.org>,
	"oneukum@...e.de" <oneukum@...e.de>
Subject: Re: [PATCH] Modify UEFI anti-bricking code

On Thu, Jun 06, 2013 at 10:25:42AM +0100, Matt Fleming wrote:
> On Thu, 06 Jun, at 03:40:26PM, joeyli wrote:
> > OK, I moved volatile checking to the top of the function.
> > New version, version 3 diff result like the following.
> 
> Thanks. This is what I've now got queued up.

Is this (or similar) going to make it into 3.10?


> 
> ---
> 
> >From 118428bf3b207d9b390a27f32dfef6dc2979078d Mon Sep 17 00:00:00 2001
> From: Matthew Garrett <matthew.garrett@...ula.com>
> Date: Sat, 1 Jun 2013 16:06:20 -0400
> Subject: [PATCH] Modify UEFI anti-bricking code
> 
> This patch reworks the UEFI anti-bricking code, including an effective
> reversion of cc5a080c and 31ff2f20. It turns out that calling
> QueryVariableInfo() from boot services results in some firmware
> implementations jumping to physical addresses even after entering virtual
> mode, so until we have 1:1 mappings for UEFI runtime space this isn't
> going to work so well.
> 
> Reverting these gets us back to the situation where we'd refuse to create
> variables on some systems because they classify deleted variables as "used"
> until the firmware triggers a garbage collection run, which they won't do
> until they reach a lower threshold. This results in it being impossible to
> install a bootloader, which is unhelpful.
> 
> Feedback from Samsung indicates that the firmware doesn't need more than
> 5KB of storage space for its own purposes, so that seems like a reasonable
> threshold. However, there's still no guarantee that a platform will attempt
> garbage collection merely because it drops below this threshold. It seems
> that this is often only triggered if an attempt to write generates a
> genuine EFI_OUT_OF_RESOURCES error. We can force that by attempting to
> create a variable larger than the remaining space. This should fail, but if
> it somehow succeeds we can then immediately delete it.
> 
> I've tested this on the UEFI machines I have available, but I don't have
> a Samsung and so can't verify that it avoids the bricking problem.
> 
> Signed-off-by: Matthew Garrett <matthew.garrett@...ula.com>
> Signed-off-by: Lee, Chun-Y <jlee@...e.com> [ dummy variable cleanup ]
> Signed-off-by: Matt Fleming <matt.fleming@...el.com>
> ---
>  arch/x86/boot/compressed/eboot.c      |  47 ---------
>  arch/x86/include/asm/efi.h            |   7 --
>  arch/x86/include/uapi/asm/bootparam.h |   1 -
>  arch/x86/platform/efi/efi.c           | 188 ++++++++++++----------------------
>  4 files changed, 65 insertions(+), 178 deletions(-)
> 
> diff --git a/arch/x86/boot/compressed/eboot.c b/arch/x86/boot/compressed/eboot.c
> index 35ee62f..c205035 100644
> --- a/arch/x86/boot/compressed/eboot.c
> +++ b/arch/x86/boot/compressed/eboot.c
> @@ -251,51 +251,6 @@ static void find_bits(unsigned long mask, u8 *pos, u8 *size)
>  	*size = len;
>  }
>  
> -static efi_status_t setup_efi_vars(struct boot_params *params)
> -{
> -	struct setup_data *data;
> -	struct efi_var_bootdata *efidata;
> -	u64 store_size, remaining_size, var_size;
> -	efi_status_t status;
> -
> -	if (sys_table->runtime->hdr.revision < EFI_2_00_SYSTEM_TABLE_REVISION)
> -		return EFI_UNSUPPORTED;
> -
> -	data = (struct setup_data *)(unsigned long)params->hdr.setup_data;
> -
> -	while (data && data->next)
> -		data = (struct setup_data *)(unsigned long)data->next;
> -
> -	status = efi_call_phys4((void *)sys_table->runtime->query_variable_info,
> -				EFI_VARIABLE_NON_VOLATILE |
> -				EFI_VARIABLE_BOOTSERVICE_ACCESS |
> -				EFI_VARIABLE_RUNTIME_ACCESS, &store_size,
> -				&remaining_size, &var_size);
> -
> -	if (status != EFI_SUCCESS)
> -		return status;
> -
> -	status = efi_call_phys3(sys_table->boottime->allocate_pool,
> -				EFI_LOADER_DATA, sizeof(*efidata), &efidata);
> -
> -	if (status != EFI_SUCCESS)
> -		return status;
> -
> -	efidata->data.type = SETUP_EFI_VARS;
> -	efidata->data.len = sizeof(struct efi_var_bootdata) -
> -		sizeof(struct setup_data);
> -	efidata->data.next = 0;
> -	efidata->store_size = store_size;
> -	efidata->remaining_size = remaining_size;
> -	efidata->max_var_size = var_size;
> -
> -	if (data)
> -		data->next = (unsigned long)efidata;
> -	else
> -		params->hdr.setup_data = (unsigned long)efidata;
> -
> -}
> -
>  static efi_status_t setup_efi_pci(struct boot_params *params)
>  {
>  	efi_pci_io_protocol *pci;
> @@ -1202,8 +1157,6 @@ struct boot_params *efi_main(void *handle, efi_system_table_t *_table,
>  
>  	setup_graphics(boot_params);
>  
> -	setup_efi_vars(boot_params);
> -
>  	setup_efi_pci(boot_params);
>  
>  	status = efi_call_phys3(sys_table->boottime->allocate_pool,
> diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h
> index 2fb5d58..60c89f3 100644
> --- a/arch/x86/include/asm/efi.h
> +++ b/arch/x86/include/asm/efi.h
> @@ -102,13 +102,6 @@ extern void efi_call_phys_epilog(void);
>  extern void efi_unmap_memmap(void);
>  extern void efi_memory_uc(u64 addr, unsigned long size);
>  
> -struct efi_var_bootdata {
> -	struct setup_data data;
> -	u64 store_size;
> -	u64 remaining_size;
> -	u64 max_var_size;
> -};
> -
>  #ifdef CONFIG_EFI
>  
>  static inline bool efi_is_native(void)
> diff --git a/arch/x86/include/uapi/asm/bootparam.h b/arch/x86/include/uapi/asm/bootparam.h
> index 0874424..c15ddaf 100644
> --- a/arch/x86/include/uapi/asm/bootparam.h
> +++ b/arch/x86/include/uapi/asm/bootparam.h
> @@ -6,7 +6,6 @@
>  #define SETUP_E820_EXT			1
>  #define SETUP_DTB			2
>  #define SETUP_PCI			3
> -#define SETUP_EFI_VARS			4
>  
>  /* ram_size flags */
>  #define RAMDISK_IMAGE_START_MASK	0x07FF
> diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c
> index 82089d8..5ae2eb0 100644
> --- a/arch/x86/platform/efi/efi.c
> +++ b/arch/x86/platform/efi/efi.c
> @@ -42,7 +42,6 @@
>  #include <linux/io.h>
>  #include <linux/reboot.h>
>  #include <linux/bcd.h>
> -#include <linux/ucs2_string.h>
>  
>  #include <asm/setup.h>
>  #include <asm/efi.h>
> @@ -54,12 +53,12 @@
>  
>  #define EFI_DEBUG	1
>  
> -/*
> - * There's some additional metadata associated with each
> - * variable. Intel's reference implementation is 60 bytes - bump that
> - * to account for potential alignment constraints
> - */
> -#define VAR_METADATA_SIZE 64
> +#define EFI_MIN_RESERVE 5120
> +
> +#define EFI_DUMMY_GUID \
> +	EFI_GUID(0x4424ac57, 0xbe4b, 0x47dd, 0x9e, 0x97, 0xed, 0x50, 0xf0, 0x9f, 0x92, 0xa9)
> +
> +static efi_char16_t efi_dummy_name[6] = { 'D', 'U', 'M', 'M', 'Y', 0 };
>  
>  struct efi __read_mostly efi = {
>  	.mps        = EFI_INVALID_TABLE_ADDR,
> @@ -79,13 +78,6 @@ struct efi_memory_map memmap;
>  static struct efi efi_phys __initdata;
>  static efi_system_table_t efi_systab __initdata;
>  
> -static u64 efi_var_store_size;
> -static u64 efi_var_remaining_size;
> -static u64 efi_var_max_var_size;
> -static u64 boot_used_size;
> -static u64 boot_var_size;
> -static u64 active_size;
> -
>  unsigned long x86_efi_facility;
>  
>  /*
> @@ -188,53 +180,8 @@ static efi_status_t virt_efi_get_next_variable(unsigned long *name_size,
>  					       efi_char16_t *name,
>  					       efi_guid_t *vendor)
>  {
> -	efi_status_t status;
> -	static bool finished = false;
> -	static u64 var_size;
> -
> -	status = efi_call_virt3(get_next_variable,
> -				name_size, name, vendor);
> -
> -	if (status == EFI_NOT_FOUND) {
> -		finished = true;
> -		if (var_size < boot_used_size) {
> -			boot_var_size = boot_used_size - var_size;
> -			active_size += boot_var_size;
> -		} else {
> -			printk(KERN_WARNING FW_BUG  "efi: Inconsistent initial sizes\n");
> -		}
> -	}
> -
> -	if (boot_used_size && !finished) {
> -		unsigned long size = 0;
> -		u32 attr;
> -		efi_status_t s;
> -		void *tmp;
> -
> -		s = virt_efi_get_variable(name, vendor, &attr, &size, NULL);
> -
> -		if (s != EFI_BUFFER_TOO_SMALL || !size)
> -			return status;
> -
> -		tmp = kmalloc(size, GFP_ATOMIC);
> -
> -		if (!tmp)
> -			return status;
> -
> -		s = virt_efi_get_variable(name, vendor, &attr, &size, tmp);
> -
> -		if (s == EFI_SUCCESS && (attr & EFI_VARIABLE_NON_VOLATILE)) {
> -			var_size += size;
> -			var_size += ucs2_strsize(name, 1024);
> -			active_size += size;
> -			active_size += VAR_METADATA_SIZE;
> -			active_size += ucs2_strsize(name, 1024);
> -		}
> -
> -		kfree(tmp);
> -	}
> -
> -	return status;
> +	return efi_call_virt3(get_next_variable,
> +			      name_size, name, vendor);
>  }
>  
>  static efi_status_t virt_efi_set_variable(efi_char16_t *name,
> @@ -243,34 +190,9 @@ static efi_status_t virt_efi_set_variable(efi_char16_t *name,
>  					  unsigned long data_size,
>  					  void *data)
>  {
> -	efi_status_t status;
> -	u32 orig_attr = 0;
> -	unsigned long orig_size = 0;
> -
> -	status = virt_efi_get_variable(name, vendor, &orig_attr, &orig_size,
> -				       NULL);
> -
> -	if (status != EFI_BUFFER_TOO_SMALL)
> -		orig_size = 0;
> -
> -	status = efi_call_virt5(set_variable,
> -				name, vendor, attr,
> -				data_size, data);
> -
> -	if (status == EFI_SUCCESS) {
> -		if (orig_size) {
> -			active_size -= orig_size;
> -			active_size -= ucs2_strsize(name, 1024);
> -			active_size -= VAR_METADATA_SIZE;
> -		}
> -		if (data_size) {
> -			active_size += data_size;
> -			active_size += ucs2_strsize(name, 1024);
> -			active_size += VAR_METADATA_SIZE;
> -		}
> -	}
> -
> -	return status;
> +	return efi_call_virt5(set_variable,
> +			      name, vendor, attr,
> +			      data_size, data);
>  }
>  
>  static efi_status_t virt_efi_query_variable_info(u32 attr,
> @@ -786,9 +708,6 @@ void __init efi_init(void)
>  	char vendor[100] = "unknown";
>  	int i = 0;
>  	void *tmp;
> -	struct setup_data *data;
> -	struct efi_var_bootdata *efi_var_data;
> -	u64 pa_data;
>  
>  #ifdef CONFIG_X86_32
>  	if (boot_params.efi_info.efi_systab_hi ||
> @@ -806,22 +725,6 @@ void __init efi_init(void)
>  	if (efi_systab_init(efi_phys.systab))
>  		return;
>  
> -	pa_data = boot_params.hdr.setup_data;
> -	while (pa_data) {
> -		data = early_ioremap(pa_data, sizeof(*efi_var_data));
> -		if (data->type == SETUP_EFI_VARS) {
> -			efi_var_data = (struct efi_var_bootdata *)data;
> -
> -			efi_var_store_size = efi_var_data->store_size;
> -			efi_var_remaining_size = efi_var_data->remaining_size;
> -			efi_var_max_var_size = efi_var_data->max_var_size;
> -		}
> -		pa_data = data->next;
> -		early_iounmap(data, sizeof(*efi_var_data));
> -	}
> -
> -	boot_used_size = efi_var_store_size - efi_var_remaining_size;
> -
>  	set_bit(EFI_SYSTEM_TABLES, &x86_efi_facility);
>  
>  	/*
> @@ -1085,6 +988,13 @@ void __init efi_enter_virtual_mode(void)
>  		runtime_code_page_mkexec();
>  
>  	kfree(new_memmap);
> +
> +	/* clean DUMMY object */
> +	efi.set_variable(efi_dummy_name, &EFI_DUMMY_GUID,
> +			 EFI_VARIABLE_NON_VOLATILE |
> +			 EFI_VARIABLE_BOOTSERVICE_ACCESS |
> +			 EFI_VARIABLE_RUNTIME_ACCESS,
> +			 0, NULL);
>  }
>  
>  /*
> @@ -1136,33 +1046,65 @@ efi_status_t efi_query_variable_store(u32 attributes, unsigned long size)
>  	efi_status_t status;
>  	u64 storage_size, remaining_size, max_size;
>  
> +	if (!(attributes & EFI_VARIABLE_NON_VOLATILE))
> +		return 0;
> +
>  	status = efi.query_variable_info(attributes, &storage_size,
>  					 &remaining_size, &max_size);
>  	if (status != EFI_SUCCESS)
>  		return status;
>  
> -	if (!max_size && remaining_size > size)
> -		printk_once(KERN_ERR FW_BUG "Broken EFI implementation"
> -			    " is returning MaxVariableSize=0\n");
>  	/*
>  	 * Some firmware implementations refuse to boot if there's insufficient
>  	 * space in the variable store. We account for that by refusing the
>  	 * write if permitting it would reduce the available space to under
> -	 * 50%. However, some firmware won't reclaim variable space until
> -	 * after the used (not merely the actively used) space drops below
> -	 * a threshold. We can approximate that case with the value calculated
> -	 * above. If both the firmware and our calculations indicate that the
> -	 * available space would drop below 50%, refuse the write.
> +	 * 5KB. This figure was provided by Samsung, so should be safe.
>  	 */
> +	if ((remaining_size - size < EFI_MIN_RESERVE) &&
> +		!efi_no_storage_paranoia) {
> +
> +		/*
> +		 * Triggering garbage collection may require that the firmware
> +		 * generate a real EFI_OUT_OF_RESOURCES error. We can force
> +		 * that by attempting to use more space than is available.
> +		 */
> +		unsigned long dummy_size = remaining_size + 1024;
> +		void *dummy = kmalloc(dummy_size, GFP_ATOMIC);
> +
> +		status = efi.set_variable(efi_dummy_name, &EFI_DUMMY_GUID,
> +					  EFI_VARIABLE_NON_VOLATILE |
> +					  EFI_VARIABLE_BOOTSERVICE_ACCESS |
> +					  EFI_VARIABLE_RUNTIME_ACCESS,
> +					  dummy_size, dummy);
> +
> +		if (status == EFI_SUCCESS) {
> +			/*
> +			 * This should have failed, so if it didn't make sure
> +			 * that we delete it...
> +			 */
> +			efi.set_variable(efi_dummy_name, &EFI_DUMMY_GUID,
> +					 EFI_VARIABLE_NON_VOLATILE |
> +					 EFI_VARIABLE_BOOTSERVICE_ACCESS |
> +					 EFI_VARIABLE_RUNTIME_ACCESS,
> +					 0, dummy);
> +		}
>  
> -	if (!storage_size || size > remaining_size ||
> -	    (max_size && size > max_size))
> -		return EFI_OUT_OF_RESOURCES;
> +		/*
> +		 * The runtime code may now have triggered a garbage collection
> +		 * run, so check the variable info again
> +		 */
> +		status = efi.query_variable_info(attributes, &storage_size,
> +						 &remaining_size, &max_size);
>  
> -	if (!efi_no_storage_paranoia &&
> -	    ((active_size + size + VAR_METADATA_SIZE > storage_size / 2) &&
> -	     (remaining_size - size < storage_size / 2)))
> -		return EFI_OUT_OF_RESOURCES;
> +		if (status != EFI_SUCCESS)
> +			return status;
> +
> +		/*
> +		 * There still isn't enough room, so return an error
> +		 */
> +		if (remaining_size - size < EFI_MIN_RESERVE)
> +			return EFI_OUT_OF_RESOURCES;
> +	}
>  
>  	return EFI_SUCCESS;
>  }
> -- 
> 1.8.1.4
> 
> -- 
> Matt Fleming, Intel Open Source Technology Center

-- 
Russ Anderson, OS RAS/Partitioning Project Lead  
SGI - Silicon Graphics Inc          rja@....com
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists