Skip to contentSkip to conversation

Small fix in i386 grub-setup

PATCH In review Javier Martín 2 participants 6 messages Jun 19, 2008 – Jun 21, 2008

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 Small fix in i386 grub-setup

Hi there - I have found that a grub-setup in i386 recently got borked, and now inserts the core.img filename twice in the path while looking for it the second time: the program tried to find the file "/boot/grubcore.img/core.img" which is obviously wrong. This is a fix, or so I think - it seems the right thing given the function prototype and it works for me.

Index: util/i386/pc/grub-setup.c
===================================================================
RCS file: /sources/grub/grub2/util/i386/pc/grub-setup.c,v retrieving revision 1.37

i386/pc/grub-setup.c +1 -1
diff -u -r1.37 grub-setup.c
@@ -368,7 +368,7 @@
   
   /* Make sure that GRUB reads the identical image as the OS.  */
   tmp_img = xmalloc (core_size);
-  core_path = grub_util_get_path (DEFAULT_DIRECTORY "core.img", core_file);
+  core_path = grub_util_get_path (DEFAULT_DIRECTORY, core_file);
   
   /* It is a Good Thing to sync two times.  */
   sync ();
#1 lore raw
Quoted 6 lines · Javier Martín

Hi there - I have found that a grub-setup in i386 recently got borked, and now inserts the core.img filename twice in the path while looking for it the second time: the program tried to find the file "/boot/grubcore.img/core.img" which is obviously wrong. This is a fix, or so I think - it seems the right thing given the function prototype and it works for me.

Thanks. Please could you include a ChangeLog entry?

Signature · 5 lines
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What good is a phone call… if you are unable to speak?
(as seen on /.)
#2 lore raw
Quoted 10 lines · Robert Millan
Quoted 6 lines · Javier Martín

Hi there - I have found that a grub-setup in i386 recently got borked, and now inserts the core.img filename twice in the path while looking for it the second time: the program tried to find the file "/boot/grubcore.img/core.img" which is obviously wrong. This is a fix, or so I think - it seems the right thing given the function prototype and it works for me.

Thanks. Please could you include a ChangeLog entry?

Erm, ok... I didn't know if I just had to write it here or include it in the patch, so I did the latter

Index: ChangeLog
===================================================================
RCS file: /sources/grub/grub2/ChangeLog,v
retrieving revision 1.727

ChangeLog +5 -0
diff -u -r1.727 ChangeLog
@@ -1,3 +1,8 @@
+2008-06-19  Javier Martín  <[email protected]>
+
+  * util/i386/pc-grub-setup.c (setup): Remove literal "core.img" in a call to
+  resolve the core image location that effectively appended the name twice.
+
 2008-06-19  Robert Millan  <[email protected]>
 
 	* commands/search.c (search_label, search_fs_uuid, search_file): Print

Index: util/i386/pc/grub-setup.c
===================================================================
RCS file: /sources/grub/grub2/util/i386/pc/grub-setup.c,v retrieving revision 1.37

i386/pc/grub-setup.c +1 -1
diff -u -r1.37 grub-setup.c
@@ -368,7 +368,7 @@
   
   /* Make sure that GRUB reads the identical image as the OS.  */
   tmp_img = xmalloc (core_size);
-  core_path = grub_util_get_path (DEFAULT_DIRECTORY "core.img", core_file);
+  core_path = grub_util_get_path (DEFAULT_DIRECTORY, core_file);
   
   /* It is a Good Thing to sync two times.  */
   sync ();
#4 lore raw
In reply to Robert Millan
Quoted 10 lines · Robert Millan
Quoted 6 lines · Javier Martín

Hi there - I have found that a grub-setup in i386 recently got borked, and now inserts the core.img filename twice in the path while looking for it the second time: the program tried to find the file "/boot/grubcore.img/core.img" which is obviously wrong. This is a fix, or so I think - it seems the right thing given the function prototype and it works for me.

Thanks. Please could you include a ChangeLog entry?

Oops. There's a small typo in my ChangeLog entry (pc-grub-setup.c instead of pc/grub-setup.c). It has been corrected in this version

Index: ChangeLog
===================================================================
RCS file: /sources/grub/grub2/ChangeLog,v
retrieving revision 1.727

ChangeLog +5 -0
diff -u -r1.727 ChangeLog
@@ -1,3 +1,8 @@
+2008-06-19  Javier Martín  <[email protected]>
+
+  * util/i386/pc/grub-setup.c (setup): Remove literal "core.img" in a call to
+  resolve the core image location that effectively appended the name twice.
+
 2008-06-19  Robert Millan  <[email protected]>
 
 	* commands/search.c (search_label, search_fs_uuid, search_file): Print

Index: util/i386/pc/grub-setup.c
===================================================================
RCS file: /sources/grub/grub2/util/i386/pc/grub-setup.c,v retrieving revision 1.37

i386/pc/grub-setup.c +1 -1
diff -u -r1.37 grub-setup.c
@@ -368,7 +368,7 @@
   
   /* Make sure that GRUB reads the identical image as the OS.  */
   tmp_img = xmalloc (core_size);
-  core_path = grub_util_get_path (DEFAULT_DIRECTORY "core.img", core_file);
+  core_path = grub_util_get_path (DEFAULT_DIRECTORY, core_file);
   
   /* It is a Good Thing to sync two times.  */
   sync ();
#5 lore raw
Quoted 9 lines · Javier Martín
Quoted 8 lines · Robert Millan
Quoted 6 lines · Javier Martín

Hi there - I have found that a grub-setup in i386 recently got borked, and now inserts the core.img filename twice in the path while looking for it the second time: the program tried to find the file "/boot/grubcore.img/core.img" which is obviously wrong. This is a fix, or so I think - it seems the right thing given the function prototype and it works for me.

Quoted 5 lines
Quoted 2 lines

Thanks. Please could you include a ChangeLog entry?

Oops. There's a small typo in my ChangeLog entry (pc-grub-setup.c instead of pc/grub-setup.c). It has been corrected in this version

Committed, thank you.

Signature · 5 lines
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What good is a phone call… if you are unable to speak?
(as seen on /.)
#3 lore raw
In reply to Javier Martín
Quoted 5 lines · Javier Martín
Quoted 2 lines

Thanks. Please could you include a ChangeLog entry?

Erm, ok... I didn't know if I just had to write it here or include it in the patch, so I did the latter

I think it's better to include it here (or in the patch header) instead of making it part of the patch, because otherwise it tends to break when ChangeLog changes.

But then again, it's no big deal ;-)

Signature · 5 lines
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What good is a phone call… if you are unable to speak?
(as seen on /.)