EFI chainloader improvement
cover EFI chainloader improvement
This series improves the EFI chainloader. I've noticed for a while now that chainloading would fail when root=memdisk. It didn't really make sense because I was specifying the image to chainload as device+path, so why would it care about what my root was. But I noticed that if I changed the root to the device the image file was located on, then chainloading worked. The second patch fixes this by removing some previous assumptions that I don't believe are valid (eg. that LoadImage needs a valid device path).
Glenn
Glenn Washburn (2):
efi/chainloader: Do not require a valid $root when chainloading
docs: Document that extra arguments to chainloader on EFI
docs/grub.texi | 7 +++++-- grub-core/loader/efi/chainloader.c | 31 +++++++++++------------------- 2 files changed, 16 insertions(+), 22 deletions(-)
Signature · 1 line
2.34.1
PATCH 1/2 efi/chainloader: Do not require a valid $root when chainloading
The EFI chainloader checks that a device path can be created for the $root device before allowing chainloading to a given file. This is probably to ensure that the given file can be accessed and loaded by the firmware. However, since GRUB is loading the image itself, the firmware need not be able to access the file location of the image. So remove this check.
Also, this fixes an issue where chainloading an image file on a location that is accessible by the firmware, eg. (hd0,1)/efi/boot.efi, would fail when root is a location inaccessible by the firmware, eg. memdisk.
Use GRUB_EFI_BYTES_TO_PAGES() instead of donig the calculation explicitly.
Add comment noting the section where the load options for the chainloaded EFI application is constructed.
--- grub-core/loader/efi/chainloader.c | 31 +++++++++++------------------- 1 file changed, 11 insertions(+), 20 deletions(-)
grub-core/loader/efi/chainloader.c +11 -20
@@ -32,6 +32,7 @@#include <grub/efi/api.h> #include <grub/efi/efi.h> #include <grub/efi/disk.h> +#include <grub/efi/memory.h> #include <grub/command.h> #include <grub/i18n.h> #include <grub/net.h>
@@ -239,11 +240,7 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)),if (! file) goto fail; -/* Get the root device's device path. */ - dev =grub_device_open (0 ); -if (! dev) - goto fail; - + dev = file->device;if (dev->disk) dev_handle =grub_efidisk_get_device_handle (dev->disk);else if (dev->net && dev->net->server)
@@ -267,18 +264,15 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)),if (dev_handle) dp =grub_efi_get_device_path (dev_handle); -if (! dp) +if (dp !=NULL ) { -grub_error (GRUB_ERR_BAD_DEVICE ,"not a valid root device" ); - goto fail; - } - - file_path =make_file_path (dp, filename); -if (! file_path) - goto fail; + file_path =make_file_path (dp, filename); +if (file_path ==NULL ) + goto fail; -grub_printf ("file path: " ); -grub_efi_print_device_path (file_path); +grub_printf ("file path: " ); +grub_efi_print_device_path (file_path); + } size =grub_file_size (file);if (!size)
@@ -287,7 +281,7 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)), filename); goto fail; } - pages = (((grub_efi_uintn_t ) size + ((1 <<12 ) -1 )) >>12 ); + pages = (grub_efi_uintn_t )GRUB_EFI_BYTES_TO_PAGES (size); status = efi_call_4 (b->allocate_pages, GRUB_EFI_ALLOCATE_ANY_PAGES, GRUB_EFI_LOADER_CODE,
@@ -370,6 +364,7 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)), } loaded_image->device_handle = dev_handle; +/* Build load options with arguments from chainloader command line. */ if (argc >1 ) {int i, len;
@@ -400,7 +395,6 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)), }grub_file_close (file); -grub_device_close (dev);/* We're finished with the source image buffer and file path now. */ efi_call_2 (b->free_pages, address, pages);
@@ -411,9 +405,6 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)), fail: -if (dev) -grub_device_close (dev); -if (file)grub_file_close (file);
Signature · 1 line
2.34.1
PATCH 2/2 docs: Document that extra arguments to chainloader on EFI
In reply to Glenn WashburnExtra arguments given to chainloader on EFI platforms will be sent to the chainloaded application. Also, minor edit in the chainloading section to note that chainloading can be a jump via the firmware and not necessarily in real mode (which does not exist on some achitectures).
--- docs/grub.texi | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
docs/grub.texi +5 -2
@@ -965,7 +965,7 @@ information. Operating systems that do not support Multiboot and do not have specific support in GRUB (specific support is available for Linux, FreeBSD, NetBSD and OpenBSD) must be chain-loaded, which involves loading another boot -loader and jumping to it in real mode. +loader and jumping to it in real mode or via the firmware. The @command{chainloader} command (@pxref{chainloader}) is used to set this up. It is normally also necessary to load some GRUB modules and set the
@@ -4011,10 +4011,13 @@ a list of commands that could use more documentation: @node chainloader @subsection chainloader -@deffn Command chainloader [@option{--force}] file +@deffn Command chainloader [@option{--force}] file [args...] Load @var{file} as a chain-loader. Like any other file loaded by the filesystem code, it can use the blocklist notation (@pxref{Block list syntax}) to grab the first sector of the current partition with @samp{+1}. +On EFI platforms, any arguments after @var{file} will be sent to the loaded +image. + If you specify the option @option{--force}, then load @var{file} forcibly, whether it has a correct signature or not. This is required when you want to load a defective boot loader, such as SCO UnixWare 7.1.
Signature · 1 line
2.34.1
Re: [PATCH 1/2] efi/chainloader: Do not require a valid $root when chainloading
In reply to Glenn WashburnHi,
This is interesting. I had to work around this same issue in loopback to allow chainloading from loopback devices see
https://github.com/rhboot/grub2/commit/0e5cb733f3cb227293ea58397ea10891519095f0
Quoted 115 lines · Glenn Washburn
The EFI chainloader checks that a device path can be created for the $root device before allowing chainloading to a given file. This is probably to ensure that the given file can be accessed and loaded by the firmware. However, since GRUB is loading the image itself, the firmware need not be able to access the file location of the image. So remove this check.
Also, this fixes an issue where chainloading an image file on a location that is accessible by the firmware, eg. (hd0,1)/efi/boot.efi, would fail when root is a location inaccessible by the firmware, eg. memdisk.
Use
GRUB_EFI_BYTES_TO_PAGES()instead of donig the calculation explicitly.Add comment noting the section where the load options for the chainloaded EFI application is constructed.
Signed-off-by Glenn Washburn <[email protected]>--- grub-core/loader/efi/chainloader.c | 31 +++++++++++------------------- 1 file changed, 11 insertions(+), 20 deletions(-)grub-core/loader/efi/chainloader.c +11
-21@@ -32,6 +32,7 @@#include <grub/efi/api.h> #include <grub/efi/efi.h> #include <grub/efi/disk.h> +#include <grub/efi/memory.h> #include <grub/command.h> #include <grub/i18n.h> #include <grub/net.h> @@ -239,11 +240,7 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)),if (! file) goto fail; -/* Get the root device's device path. */ - dev =grub_device_open (0 ); -if (! dev) - goto fail; - + dev = file->device;if (dev->disk) dev_handle =grub_efidisk_get_device_handle (dev->disk);else if (dev->net && dev->net->server)@@ -267,18 +264,15 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)),if (dev_handle) dp =grub_efi_get_device_path (dev_handle); -if (! dp) +if (dp !=NULL ) { -grub_error (GRUB_ERR_BAD_DEVICE ,"not a valid root device" ); - goto fail; - } - - file_path =make_file_path (dp, filename); -if (! file_path) - goto fail; + file_path =make_file_path (dp, filename); +if (file_path ==NULL ) + goto fail; -grub_printf ("file path: " ); -grub_efi_print_device_path (file_path); +grub_printf ("file path: " ); +grub_efi_print_device_path (file_path); + } size =grub_file_size (file);if (!size)@@ -287,7 +281,7 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)), filename); goto fail; } - pages = (((grub_efi_uintn_t ) size + ((1 <<12 ) -1 )) >>12 ); + pages = (grub_efi_uintn_t )GRUB_EFI_BYTES_TO_PAGES (size); status = efi_call_4 (b->allocate_pages, GRUB_EFI_ALLOCATE_ANY_PAGES, GRUB_EFI_LOADER_CODE,@@ -370,6 +364,7 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)), } loaded_image->device_handle = dev_handle; +/* Build load options with arguments from chainloader command line. */ if (argc >1 ) {int i, len;@@ -400,7 +395,6 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)), }grub_file_close (file); -grub_device_close (dev);/* We're finished with the source image buffer and file path now. */ efi_call_2 (b->free_pages, address, pages);@@ -411,9 +405,6 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)), fail: -if (dev) -grub_device_close (dev); -if (file)grub_file_close (file); --2.34.1
_______________________________________________
Grub-devel mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/grub-devel
Signature · 3 lines
okurrr, Dimitri
Re: [PATCH 1/2] efi/chainloader: Do not require a valid $root when chainloading
Quoted 5 lines · On Fri, 26 Aug 2022 10:12:18 +0100 Dimitri John Ledkov…
Hi,
This is interesting. I had to work around this same issue in loopback to allow chainloading from loopback devices see
https://github.com/rhboot/grub2/commit/0e5cb733f3cb227293ea58397ea10891519095f0
Hmm, now that I think about this more, the current code is in some way more general. For instance, here's how you can get around your loopback issue. Suppose you have a loopback device, (d), backed by a file, (hd0,1)/fat.img, with an efi application at /efi/app.efi and you want to chainload that. To achieve the same thing as your patch, but with current code you would do:
root=(hd0,1)
chainloader (d)/efi/app.efi
So the current code allows settings the device_handle independently of where the chainloaded file is residing. Its annoying for me when my root is a memdisk, which forces me to change root. I'm guessing you have your root set to the loopback device, which is what caused you headaches. My question for the EFI gurus or anyone who might have experience with the expectation of other EFI apps is: How are EFI apps in the wild known to use the device_handle of their loaded image? Are there any that do bad things when it is NULL? And if so, do we really care about them? And how might apps do something undesirable if passed a device_handle not associated with where they were actually loaded from? As I'm seeing it, the device_handle is meant to be basically like a handle to CWD.
So now I'm thinking that perhaps a better way to do this to preserve functionality with existing code is to add a "-d" option to chainloader that where the device_handle can be passed implicitly. The new way to do the above would be like this:
chainloader -d (hd0,1) (d)/efi/app.efi
The default behavior should be what the end user most likely would want. So I'm thinking that not providing a -d would try to get the device handle from the chainloaded file. If that fails use root, and if that fails provide a NULL device_handle. Or should we not do fallbacks and just have NULL device_handle if getting a device path for the file fails?
From my perspective as a user (before understanding the issue more), it didn't make sense that I should have to set root to chainload a file on a device, partition, and filesystem that the firmware natively understood. So I don't think chainloader should have this behavior to avoid confusion (though thankfully there is an error message that clued me in to how to resolve the issue).
Glenn
Quoted 121 lines
Quoted 115 lines · Glenn Washburn
The EFI chainloader checks that a device path can be created for the $root device before allowing chainloading to a given file. This is probably to ensure that the given file can be accessed and loaded by the firmware. However, since GRUB is loading the image itself, the firmware need not be able to access the file location of the image. So remove this check.
Also, this fixes an issue where chainloading an image file on a location that is accessible by the firmware, eg. (hd0,1)/efi/boot.efi, would fail when root is a location inaccessible by the firmware, eg. memdisk.
Use
GRUB_EFI_BYTES_TO_PAGES()instead of donig the calculation explicitly.Add comment noting the section where the load options for the chainloaded EFI application is constructed.
Signed-off-by Glenn Washburn <[email protected]>--- grub-core/loader/efi/chainloader.c | 31 +++++++++++------------------- 1 file changed, 11 insertions(+), 20 deletions(-)grub-core/loader/efi/chainloader.c +11
-21@@ -32,6 +32,7 @@#include <grub/efi/api.h> #include <grub/efi/efi.h> #include <grub/efi/disk.h> +#include <grub/efi/memory.h> #include <grub/command.h> #include <grub/i18n.h> #include <grub/net.h> @@ -239,11 +240,7 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)),if (! file) goto fail; -/* Get the root device's device path. */ - dev =grub_device_open (0 ); -if (! dev) - goto fail; - + dev = file->device;if (dev->disk) dev_handle =grub_efidisk_get_device_handle (dev->disk);else if (dev->net && dev->net->server)@@ -267,18 +264,15 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)),if (dev_handle) dp =grub_efi_get_device_path (dev_handle); -if (! dp) +if (dp !=NULL ) { -grub_error (GRUB_ERR_BAD_DEVICE ,"not a valid root device" ); - goto fail; - } - - file_path =make_file_path (dp, filename); -if (! file_path) - goto fail; + file_path =make_file_path (dp, filename); +if (file_path ==NULL ) + goto fail; -grub_printf ("file path: " ); -grub_efi_print_device_path (file_path); +grub_printf ("file path: " ); +grub_efi_print_device_path (file_path); + } size =grub_file_size (file);if (!size)@@ -287,7 +281,7 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)), filename); goto fail; } - pages = (((grub_efi_uintn_t ) size + ((1 <<12 ) -1 )) >>12 ); + pages = (grub_efi_uintn_t )GRUB_EFI_BYTES_TO_PAGES (size); status = efi_call_4 (b->allocate_pages, GRUB_EFI_ALLOCATE_ANY_PAGES, GRUB_EFI_LOADER_CODE,@@ -370,6 +364,7 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)), } loaded_image->device_handle = dev_handle; +/* Build load options with arguments from chainloader command line. */ if (argc >1 ) {int i, len;@@ -400,7 +395,6 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)), }grub_file_close (file); -grub_device_close (dev);/* We're finished with the source image buffer and file path now. */ efi_call_2 (b->free_pages, address, pages);@@ -411,9 +405,6 @@ grub_cmd_chainloader (grub_command_t cmd __attribute__ ((unused)), fail: -if (dev) -grub_device_close (dev); -if (file)grub_file_close (file); --2.34.1
_______________________________________________
Grub-devel mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/grub-devel
Quoted 8 lines · On Thu, 25 Aug 2022 23:32:01 -0500 Glenn Washburn…
This series improves the EFI chainloader. I've noticed for a while now that chainloading would fail when root=memdisk. It didn't really make sense because I was specifying the image to chainload as device+path, so why would it care about what my root was. But I noticed that if I changed the root to the device the image file was located on, then chainloading worked. The second patch fixes this by removing some previous assumptions that I don't believe are valid (eg. that LoadImage needs a valid device path).
In a previous response to Dimitri Ledkov, I got confused as to what the DevicePath parameter to BootService.LoadImage was supposed to be. I was thinking it could be a kind of equivalent to the shell's current working directory. According to the spec 2.10 7.4.1, it is "the DeviceHandle specific file path from which the image is loaded." So really $root should have nothing to do with it, unless the chainloaded file is not specified with a leading device component. My conclusion is that the current implementation violates the spec.
The right thing to do is to try to make a device path from the path of the chainloaded file. If a device path can not be made (eg. the file can not be accessed from the firmware) then the DevicePath parameter should be NULL, which is valid because the parameter is marked optional.
So, Daniel, actually I think this patch does the right thing and is ready for review.
Glenn
Quoted 9 lines
Glenn Washburn (2):
efi/chainloader: Do not require a valid $root when chainloading
docs: Document that extra arguments to chainloader on EFIdocs/grub.texi | 7 +++++-- grub-core/loader/efi/chainloader.c | 31+++++++++++------------------- 2 files changed, 16 insertions(+), 22 deletions(-)
Quoted 8 lines · Glenn Washburn
This series improves the EFI chainloader. I've noticed for a while now that chainloading would fail when root=memdisk. It didn't really make sense because I was specifying the image to chainload as device+path, so why would it care about what my root was. But I noticed that if I changed the root to the device the image file was located on, then chainloading worked. The second patch fixes this by removing some previous assumptions that I don't believe are valid (eg. that LoadImage needs a valid device path).
Quoted 5 lines
Glenn
Glenn Washburn (2):
efi/chainloader: Do not require a valid $root when chainloading
docs: Document that extra arguments to chainloader on EFI
The patch set LGTM. Sadly it does not apply due to significant changes in the EFI code. Could you rebase it on top of [1].
Daniel
[1] https://lists.gnu.org/archive/html/grub-devel/2023-05/msg00147.html