ChangeLog revision 1.1.1.1.4.2
1 1.1.1.1.4.2 rmind 2013-06-12 Nigel Croxon <nigel.croxon (a] hp.com> 2 1.1.1.1.4.2 rmind Disable MMX and SSE 3 1.1.1.1.4.2 rmind 4 1.1.1.1.4.2 rmind GCC 4.8.0 adds some optimizations that will use movups/movaps (and use 5 1.1.1.1.4.2 rmind %xmm* registers) when they're faster, and of course that won't work at 6 1.1.1.1.4.2 rmind all since UEFI firmwares aren't guaranteed to initialize the mmx/sse 7 1.1.1.1.4.2 rmind instructions. 8 1.1.1.1.4.2 rmind 9 1.1.1.1.4.2 rmind This will be even more annoying, since most UEFI firmwares don't 10 1.1.1.1.4.2 rmind initialize the #DE or #UD trap handlers, and your backtrace will be a 11 1.1.1.1.4.2 rmind random path through uninitialized memory, occasionally including 12 1.1.1.1.4.2 rmind whatever address the IDT has for #UD, but also addresses like "0x4" and 13 1.1.1.1.4.2 rmind "0x507" that you don't normally expect to see in your call path. 14 1.1.1.1.4.2 rmind 15 1.1.1.1.4.2 rmind Signed-off-by: Peter Jones <pjones (a] redhat.com> 16 1.1.1.1.4.2 rmind 17 1.1.1.1.4.2 rmind Author: Nigel Croxon <nigel.croxon (a] hp.com> 18 1.1.1.1.4.2 rmind Date: Wed Jun 12 10:29:40 2013 -0400 19 1.1.1.1.4.2 rmind 20 1.1.1.1.4.2 rmind bug in make 3.82 expand to odd values 21 1.1.1.1.4.2 rmind 22 1.1.1.1.4.2 rmind Some Makefiles tickle a bug in make 3.82 that cause libefi.a 23 1.1.1.1.4.2 rmind and libgnuefi.a dependencies to expand to the odd values: 24 1.1.1.1.4.2 rmind 25 1.1.1.1.4.2 rmind libefi.a: boxdraw.o) smbios.o) ... 26 1.1.1.1.4.2 rmind libgnuefi.a(reloc_x86_64.o: 27 1.1.1.1.4.2 rmind 28 1.1.1.1.4.2 rmind The patch replaces libgnuefi.a($(OBJS)) & libefi.a($(OBJS)) 29 1.1.1.1.4.2 rmind with an equivalent expansion that should work with any make 30 1.1.1.1.4.2 rmind that supports $(patsubst). 31 1.1.1.1.4.2 rmind 32 1.1.1.1.4.2 rmind Author: Nigel Croxon <nigel.croxon (a] hp.com> 33 1.1.1.1.4.2 rmind Date: Wed Jun 12 09:53:01 2013 -0400 34 1.1.1.1.4.2 rmind 35 1.1.1.1.4.2 rmind support .text.* sections on x86_64 36 1.1.1.1.4.2 rmind 37 1.1.1.1.4.2 rmind Group them in .text. Also add vague linkage sections in .text. 38 1.1.1.1.4.2 rmind 39 1.1.1.1.4.2 rmind Signed-off-by: David Decotigny <decot (a] googlers.com> 40 1.1.1.1.4.2 rmind 41 1.1.1.1.4.2 rmind Author: Nigel Croxon <nigel.croxon (a] hp.com> 42 1.1.1.1.4.2 rmind Date: Wed Jun 12 09:51:36 2013 -0400 43 1.1.1.1.4.2 rmind 44 1.1.1.1.4.2 rmind cleanup and fix Make.defaults 45 1.1.1.1.4.2 rmind 46 1.1.1.1.4.2 rmind Reorder variables in Make.defaults so that they are grouped by 47 1.1.1.1.4.2 rmind functions. Also fixed ifeq (x,y) to have required syntax and make it 48 1.1.1.1.4.2 rmind work for ARCH amd64->x86_64 renaming on BSD. Also provides top-level 49 1.1.1.1.4.2 rmind Makefile with a "mkvars" target that displays effective variables. 50 1.1.1.1.4.2 rmind 51 1.1.1.1.4.2 rmind Signed-off-by: David Decotigny <decot (a] googlers.com> 52 1.1.1.1.4.2 rmind 53 1.1.1.1.4.2 rmind Author: Nigel Croxon <nigel.croxon (a] hp.com> 54 1.1.1.1.4.2 rmind Date: Wed Jun 12 09:47:16 2013 -0400 55 1.1.1.1.4.2 rmind 56 1.1.1.1.4.2 rmind automatically determine number of uefi_call_wrapper() args on x86_64 57 1.1.1.1.4.2 rmind 58 1.1.1.1.4.2 rmind Instead of asking developers to explicitly pass the number of 59 1.1.1.1.4.2 rmind parameters to the functions that get called, we determine them 60 1.1.1.1.4.2 rmind automatically at preprocessing time. This should result in more 61 1.1.1.1.4.2 rmind robust code. 62 1.1.1.1.4.2 rmind 63 1.1.1.1.4.2 rmind Argument va_num is now ignored in x86_64 code, both with and 64 1.1.1.1.4.2 rmind without HAVE_USE_MS_ABI. 65 1.1.1.1.4.2 rmind 66 1.1.1.1.4.2 rmind Credits to the macro magic given in the comments. 67 1.1.1.1.4.2 rmind 68 1.1.1.1.4.2 rmind Signed-off-by: David Decotigny <decot (a] googlers.com> 69 1.1.1.1.4.2 rmind 70 1.1.1.1.4.2 rmind Author: Nigel Croxon <nigel.croxon (a] hp.com> 71 1.1.1.1.4.2 rmind Date: Wed Jun 12 09:38:10 2013 -0400 72 1.1.1.1.4.2 rmind 73 1.1.1.1.4.2 rmind fix parameter-passing corruption on x86_64 for >= 5 args 74 1.1.1.1.4.2 rmind 75 1.1.1.1.4.2 rmind On x86_64 without HAVE_USE_MS_ABI support, uefi_call_wrapper() is a 76 1.1.1.1.4.2 rmind variadic function. Parameters >=5 are copied to the stack and, when 77 1.1.1.1.4.2 rmind passed small immediate values (and possibly other parameters), gcc 78 1.1.1.1.4.2 rmind would emit a movl instruction before calling uefi_call_wrapper(). As a 79 1.1.1.1.4.2 rmind result, only the lower 32b of these stack values are significant, the 80 1.1.1.1.4.2 rmind upper 32b potentially contain garbage. Considering that 81 1.1.1.1.4.2 rmind uefi_call_wrapper() assumes these arguments are clean 64b values 82 1.1.1.1.4.2 rmind before calling the efi_callX() trampolines, the latter may be passed 83 1.1.1.1.4.2 rmind garbage. This makes calling functions like 84 1.1.1.1.4.2 rmind EFI_PCI_IO_PROTOCOL.Mem.Read()/Write() or BS->OpenProtocol() quite 85 1.1.1.1.4.2 rmind unreliable. 86 1.1.1.1.4.2 rmind 87 1.1.1.1.4.2 rmind This patch fixes this by turning uefi_call_wrapper() into a macro that 88 1.1.1.1.4.2 rmind allows to expose the efi_callX() trampoline signatures to the callers, 89 1.1.1.1.4.2 rmind so that gcc can know upfront that it has to pass all arguments to 90 1.1.1.1.4.2 rmind efi_callX() as clean 64b values (eg. movq for immediates). The 91 1.1.1.1.4.2 rmind _cast64_efi_callX macros are just here to avoid a gcc warning, they do 92 1.1.1.1.4.2 rmind nothing otherwise. 93 1.1.1.1.4.2 rmind 94 1.1.1.1.4.2 rmind Signed-off-by: David Decotigny <decot (a] googlers.com> 95 1.1.1.1.4.2 rmind 96 1.1.1.1.4.2 rmind Author: noxorc <nigel.croxon (a] hp.com> 97 1.1.1.1.4.2 rmind Date: Wed May 15 15:26:16 2013 -0400 98 1.1.1.1.4.2 rmind 99 1.1.1.1.4.2 rmind - Removes the ElfW() macro usage from reloc_ia32.c and reloc_x86_64.c. These 100 1.1.1.1.4.2 rmind macros only exist in link.h on Linux. On FreeBSD, the equivalent macro is 101 1.1.1.1.4.2 rmind __ElfN(). But the macro usage is redundant. You're only going to compile the 102 1.1.1.1.4.2 rmind ia32 file for IA32 binaries and the x86_64 file for X64 binaries. If you had 103 1.1.1.1.4.2 rmind just one file built for both cases, then using the macro might make more 104 1.1.1.1.4.2 rmind sense. 105 1.1.1.1.4.2 rmind 106 1.1.1.1.4.2 rmind - Removes the "#define foo_t efi_foo_t" macros from reloc_ia32.c and 107 1.1.1.1.4.2 rmind reloc_x86_64.c. 108 1.1.1.1.4.2 rmind 109 1.1.1.1.4.2 rmind - Modifies inc/x86_64/efibind.h and inc/ia32/efibind.h to use the new 110 1.1.1.1.4.2 rmind definitions for uint64_t, int64_t and int8_t. The 64-bit types are now defined 111 1.1.1.1.4.2 rmind as: 112 1.1.1.1.4.2 rmind 113 1.1.1.1.4.2 rmind typedef int __attribute__((__mode__(__DI__))) int64_t; 114 1.1.1.1.4.2 rmind typedef unsigned int __attribute__((__mode__(__DI__))) uint64_t; 115 1.1.1.1.4.2 rmind 116 1.1.1.1.4.2 rmind This removes the conflict between the host types dragged in by elf.h and the 117 1.1.1.1.4.2 rmind type definitions in efibind.h that made the #define foo_t efi_foo_t" hack 118 1.1.1.1.4.2 rmind necessary. Also, int8_t is now defined as signed char instead of just char 119 1.1.1.1.4.2 rmind (assuming char == signed char is apparently not good enough). 120 1.1.1.1.4.2 rmind 121 1.1.1.1.4.2 rmind - Also modifies these files to use stdint.h instead of stdint-gcc.h. It's 122 1.1.1.1.4.2 rmind unclear if this is completely correct, but stdint-gcc.h is not present with 123 1.1.1.1.4.2 rmind all GCC installs, and if you use -std=c99 or later you will force this case to 124 1.1.1.1.4.2 rmind be hit. This also can break clang, which doesn't have a stdint-gcc.h at all. 125 1.1.1.1.4.2 rmind 126 1.1.1.1.4.2 rmind - Removes the #include of <link.h> from reloc_ia32.c and reloc_x86_64.c (since 127 1.1.1.1.4.2 rmind with the previous changes it's not needed anymore). 128 1.1.1.1.4.2 rmind 129 1.1.1.1.4.2 rmind - Places the #include of <elf.h> after #include <efi>/#include <efilib.h> so 130 1.1.1.1.4.2 rmind that we know the types will always be defined properly, in case you build on a 131 1.1.1.1.4.2 rmind system where <elf.h> doesn't automatically pull in the right header files to 132 1.1.1.1.4.2 rmind define all the needed types. (This actually happens on VxWorks. It's harmless 133 1.1.1.1.4.2 rmind elsewhere. If you don't care about VxWorks, you can leave this out.) 134 1.1.1.1.4.2 rmind 135 1.1.1.1.4.2 rmind - Modifies setjmp_ia32.S and setjmp_x86_64.S so to change "function" to 136 1.1.1.1.4.2 rmind @function. The clang compiler doesn't like the former. Clang and GCC both like 137 1.1.1.1.4.2 rmind the latter. 138 1.1.1.1.4.2 rmind 139 1.1.1.1.4.2 rmind - Modifles Make.defaults so that if ARCH is detected as "amd64," it's changed 140 1.1.1.1.4.2 rmind to "x86_64." It happens that uname -m on 64-bit FreeBSD reports the former 141 1.1.1.1.4.2 rmind rather than the latter, which breaks the build. This may also be the case on 142 1.1.1.1.4.2 rmind some other OSes. There's a way to force uname(1) to return x86_64 as the 143 1.1.1.1.4.2 rmind machine type, but this way is a little friendlier. 144 1.1.1.1.4.2 rmind 145 1.1.1.1.4.2 rmind - Creates gnuefi/elf_ia32_fbsd_efi.lds which specifies the object file type as 146 1.1.1.1.4.2 rmind elf-ia32-freebsd. This is required for building on FreeBSD/i386, not just 147 1.1.1.1.4.2 rmind FreeBSD/amd64. 148 1.1.1.1.4.2 rmind 149 1.1.1.1.4.2 rmind - Modifies apps/Makefile to always use 150 1.1.1.1.4.2 rmind $(TOPDIR)/gnuefi/elf_$(ARCH)_fbsd_efi.lds when building on either 32-bit or 151 1.1.1.1.4.2 rmind 64-bit FreeBSD instead of just for the x86_64 case. 152 1.1.1.1.4.2 rmind 153 1.1.1.1.4.2 rmind - Changed LDFLAGS in Make.defaults to include --no-undefined. This will cause 154 1.1.1.1.4.2 rmind linking to fail if there are any unsatisfied symbols when creating foo.so 155 1.1.1.1.4.2 rmind during any of the app builds, as opposed to just silently succeeding and 156 1.1.1.1.4.2 rmind producing an unusable binary. 157 1.1.1.1.4.2 rmind 158 1.1.1.1.4.2 rmind - Changed CFLAGS to include -ffreestanding -fno-stack-protector -fno-stack- 159 1.1.1.1.4.2 rmind check. This prevents clang from inserting a call to memset() when compiling 160 1.1.1.1.4.2 rmind the RtZeroMem() and RtSetMem() routines in lib/runtime/efirtlib.c and guards 161 1.1.1.1.4.2 rmind against the native compiler in some Linux distros from adding in stack 162 1.1.1.1.4.2 rmind checking code which relies on libc help that isn't present in the EFI runtime 163 1.1.1.1.4.2 rmind environment. 164 1.1.1.1.4.2 rmind 165 1.1.1.1.4.2 rmind This does the following: 166 1.1.1.1.4.2 rmind 167 1.1.1.1.4.2 rmind - Cleans up the ia32 and x86-64 relocation code a bit (tries to break the 168 1.1.1.1.4.2 rmind dependency between the host ELF headers and the EFI runtime environment) 169 1.1.1.1.4.2 rmind - Avoids the dependency on stdint-gcc.h which may not always be available 170 1.1.1.1.4.2 rmind - Allows GNU EFI to build out of the box on both FreeBSD/i386 and 171 1.1.1.1.4.2 rmind FreeBSD/amd64 172 1.1.1.1.4.2 rmind - Allows GNU EFI to build out of the box with either GCC or clang on 173 1.1.1.1.4.2 rmind FreeBSD/i386 and FreeBSD/amd64 9.0 and later. 174 1.1.1.1.4.2 rmind - Makes things a little easier to port to VxWorks 175 1.1.1.1.4.2 rmind - Avoids creating un-runable binaries with unresolved symbol definitions 176 1.1.1.1.4.2 rmind (which can be very confusing to debug) 177 1.1.1.1.4.2 rmind 178 1.1.1.1.4.2 rmind Author: noxorc <nigel.croxon (a] hp.com> 179 1.1.1.1.4.2 rmind Date: Wed May 8 16:29:45 2013 -0400 180 1.1.1.1.4.2 rmind 181 1.1.1.1.4.2 rmind Add the definitions for TCP, UDP and IP, for both IPv4 and IPv6. 182 1.1.1.1.4.2 rmind 183 1.1.1.1.4.2 rmind 184 1.1.1.1.4.2 rmind 2013-05-02 Nigel Croxon <nigel.croxon (a] hp.com> 185 1.1.1.1.4.2 rmind * Chnage from Matt Fleming <matt.fleming (a] intel.com> 186 1.1.1.1.4.2 rmind - Preparation for adding the networking protocol definitions. 187 1.1.1.1.4.2 rmind Add the service binding protocol. 188 1.1.1.1.4.2 rmind 189 1.1.1.1.4.2 rmind 2013-02-21 Nigel Croxon <nigel.croxon (a] hp.com> 190 1.1.1.1.4.2 rmind * Change from Peter Jones <pjones (a] redhat.com> 191 1.1.1.1.4.2 rmind - Previously we were incorrectly passing 3 functions with 192 1.1.1.1.4.2 rmind the System V ABI to UEFI functions as EFI ABI functions. 193 1.1.1.1.4.2 rmind Mark them as EFIAPI so the compiler will (in our new 194 1.1.1.1.4.2 rmind GNU_EFI_USE_MS_ABI world) use the correct ABI. 195 1.1.1.1.4.2 rmind - These need to be EFIAPI functions because in some cases 196 1.1.1.1.4.2 rmind they call ST->ConOut->OutputString(), which is an EFIAPI 197 1.1.1.1.4.2 rmind function. (Which means that previously in cases that 198 1.1.1.1.4.2 rmind needed "cdecl", these didn't work right.) 199 1.1.1.1.4.2 rmind - If the compiler version is new enough, and GNU_EFI_USE_MS_ABI 200 1.1.1.1.4.2 rmind is defined, use the function attribute ms_abi on everything 201 1.1.1.1.4.2 rmind defined with "EFIAPI". Such calls will no longer go through 202 1.1.1.1.4.2 rmind efi_call*, and as such will be properly type-checked. 203 1.1.1.1.4.2 rmind - Honor PREFIX and LIBDIR correctly when passed in during the build. 204 1.1.1.1.4.2 rmind - Add machine type defines for i386, arm/thumb, ia64, ebc, x86_64. 205 1.1.1.1.4.2 rmind - __STDC_VERSION__ never actually gets defined unless there's a 206 1.1.1.1.4.2 rmind --std=... line. So we were accidentally defining lots of c99 207 1.1.1.1.4.2 rmind types ourself. Since it's 2012, use --std=c11 where appropriate, 208 1.1.1.1.4.2 rmind and if it's defined and we're using gcc, actually include gcc's 209 1.1.1.1.4.2 rmind stdint definitions. 210 1.1.1.1.4.2 rmind - New test application added: route80h. This is a test program 211 1.1.1.1.4.2 rmind for PciIo. It routes ioport 80h on ICH10 to PCI. This is also 212 1.1.1.1.4.2 rmind useful on a very limited set of hardware to enable use of 213 1.1.1.1.4.2 rmind a port 80h debug card. 214 1.1.1.1.4.2 rmind - New test applcation added: modelist. This lists video modes 215 1.1.1.1.4.2 rmind the GOP driver is showing us. 216 1.1.1.1.4.2 rmind * Change from Finnbarr Murphy 217 1.1.1.1.4.2 rmind - https://sourceforge.net/p/gnu-efi/feature-requests/2/ 218 1.1.1.1.4.2 rmind Please add the following status codes to <efierr.h> 219 1.1.1.1.4.2 rmind EFI_INCOMPATIBLE_VERSION 25 220 1.1.1.1.4.2 rmind EFI_SECURITY_VIOLATION 26 221 1.1.1.1.4.2 rmind EFI_CRC_ERROR 27 222 1.1.1.1.4.2 rmind EFI_END_OF_MEDIA 28 223 1.1.1.1.4.2 rmind EFI_END_OF_FILE 31 224 1.1.1.1.4.2 rmind EFI_INVALID_LANGUAGE 32 225 1.1.1.1.4.2 rmind EFI_COMPROMISED_DATA 33 226 1.1.1.1.4.2 rmind * Change from SourceForge.net Bug report 227 1.1.1.1.4.2 rmind - https://sourceforge.net/p/gnu-efi/bugs/5/ 228 1.1.1.1.4.2 rmind BufferSize is a UINT64 *. The file shipped with GNU EFI is from 229 1.1.1.1.4.2 rmind 1998 whereas the latest one is from 2004. I suspect Intel changed 230 1.1.1.1.4.2 rmind the API in order handle 64-bit systems. 231 1.1.1.1.4.2 rmind * Change from Felipe Contreras <felipe.contreras (a] gmail.com> 232 1.1.1.1.4.2 rmind - The current code seems to screw the stack at certain points. 233 1.1.1.1.4.2 rmind Multiple people have complained that gummiboot hangs right away, 234 1.1.1.1.4.2 rmind which is in part the fault of gummiboot, but happens only 235 1.1.1.1.4.2 rmind because the stack gets screwed. x86_64 EFI already aligns the 236 1.1.1.1.4.2 rmind stack, so there's no need for so much code to find a proper 237 1.1.1.1.4.2 rmind alignment, we always need to shift by 8 anyway. 238 1.1.1.1.4.2 rmind * Change from A. Steinmetz 239 1.1.1.1.4.2 rmind - https://sourceforge.net/p/gnu-efi/patches/1/ 240 1.1.1.1.4.2 rmind The patch prepares for elilo to support uefi pxe over ipv6 241 1.1.1.1.4.2 rmind See uefi spec 2.3.1 errata c page 963 as reference. 242 1.1.1.1.4.2 rmind Verfied on an ASUS Sabertooth X79 BIOS Rev. 2104 system which 243 1.1.1.1.4.2 rmind is able to do an IPv6 UEFI PXE boot. 244 1.1.1.1.4.2 rmind * Release 3.0t 245 1.1.1.1.4.2 rmind 246 1.1.1.1.4.2 rmind 2012-09-21 Nigel Croxon <nigel.croxon (a] hp.com> 247 1.1.1.1.4.2 rmind * Change from Peter Jones <pjones (a] redhat.com> 248 1.1.1.1.4.2 rmind - EFI Block I/O protocol versions 2 and 3 provide more information 249 1.1.1.1.4.2 rmind regarding physical disk layout, including alingment offset at the 250 1.1.1.1.4.2 rmind beginning of the disk ("LowestAlignedLba"), logical block size 251 1.1.1.1.4.2 rmind ("LogicalBlocksPerPhysicalBlock"), and optimal block transfer size 252 1.1.1.1.4.2 rmind ("OptimalTransferLengthGranularity"). 253 1.1.1.1.4.2 rmind * Release 3.0r 254 1.1.1.1.4.2 rmind 255 1.1.1.1.4.2 rmind 2012-04-30 Nigel Croxon <nigel.croxon (a] hp.com> 256 1.1.1.1.4.2 rmind * Change from Matt Fleming <matt.fleming (a] intel.com> 257 1.1.1.1.4.2 rmind - The .reloc section is now 4096-byte boundary for x86_64. 258 1.1.1.1.4.2 rmind Without this patch the .reloc section will not adhere to 259 1.1.1.1.4.2 rmind the alignment value in the FileAlignment field (512 bytes by 260 1.1.1.1.4.2 rmind default) of the PE/COFF header. This results in a signed 261 1.1.1.1.4.2 rmind executable failing to boot in a secure boot environment. 262 1.1.1.1.4.2 rmind * Release 3.0q 263 1.1.1.1.4.2 rmind 264 1.1.1.1.4.2 rmind 2011-12-12 Nigel Croxon <nigel.croxon (a] hp.com> 265 1.1.1.1.4.2 rmind * Changes from Fenghua Yu <fenghua.yu (a] intel.com> 266 1.1.1.1.4.2 rmind - This fixes redefined types compilation failure for tcc.c on x86_64 machines. 267 1.1.1.1.4.2 rmind * Release 3.0p 268 1.1.1.1.4.2 rmind 269 1.1.1.1.4.2 rmind 2011-11-15 Nigel Croxon <nigel.croxon (a] hp.com> 270 1.1.1.1.4.2 rmind * Changes from Darren Hart <dvhart (a] linux.intel.com> 271 1.1.1.1.4.2 rmind - Conditionally assign toolchain binaries to allow overriding them. 272 1.1.1.1.4.2 rmind - Force a dependency on lib for gnuefi. 273 1.1.1.1.4.2 rmind * Release 3.0n 274 1.1.1.1.4.2 rmind 275 1.1.1.1.4.2 rmind 2011-08-23 Nigel Croxon <nigel.croxon (a] hp.com> 276 1.1.1.1.4.2 rmind * Changes from Peter Jones <pjones (a] redhat.com> 277 1.1.1.1.4.2 rmind - Add guarantee 16-byte stack alignment on x86_64. 278 1.1.1.1.4.2 rmind - Add routine to make callbacks work. 279 1.1.1.1.4.2 rmind - Add apps/tcc.efi to test calling convention. 280 1.1.1.1.4.2 rmind * Release 3.0m 281 1.1.1.1.4.2 rmind 282 1.1.1.1.4.2 rmind 2011-07-22 Nigel Croxon <nigel.croxon (a] hp.com> 283 1.1.1.1.4.2 rmind * Changed Makefiles from GPL to BSD. 284 1.1.1.1.4.2 rmind * Changes from Peter Jones <pjones (a] redhat.com> 285 1.1.1.1.4.2 rmind - Add ifdefs for ia64 to mirror ia32 and x86-64 so that 286 1.1.1.1.4.2 rmind one can build with GCC. 287 1.1.1.1.4.2 rmind - Add headers for PciIo. 288 1.1.1.1.4.2 rmind - Add the UEFI 2.x bits for EFI_BOOT_SERVICES 289 1.1.1.1.4.2 rmind - Add an ignore for .note.GNU-stack section in X86-64 linker maps. 290 1.1.1.1.4.2 rmind * Release 3.0l 291 1.1.1.1.4.2 rmind 292 1.1.1.1.4.2 rmind 2011-04-07 Nigel Croxon <nigel.croxon (a] hp.com> 293 1.1.1.1.4.2 rmind * Change license from GPL to BSD. 294 1.1.1.1.4.2 rmind * Release 3.0j 295 1.1.1.1.4.2 rmind 296 1.1.1.1.4.2 rmind 2009-09-12 Julien BLACHE <jb (a] jblache.org> 297 1.1.1.1.4.2 rmind * Add support for FreeBSD. 298 1.1.1.1.4.2 rmind * Release 3.0i 299 1.1.1.1.4.2 rmind 300 1.1.1.1.4.2 rmind 2009-09-11 Julien BLACHE <jb (a] jblache.org> 301 1.1.1.1.4.2 rmind * Fix elf_ia32_efi.lds linker script to be compatible with the new 302 1.1.1.1.4.2 rmind linker behaviour. Patch from the RedHat bugzilla 492183. 303 1.1.1.1.4.2 rmind 304 1.1.1.1.4.2 rmind 2009-06-18 Nigel Croxon <nigel.croxon (a] hp.com> 305 1.1.1.1.4.2 rmind * Release 3.0h 306 1.1.1.1.4.2 rmind 307 1.1.1.1.4.2 rmind 2008-11-06 Nigel Croxon <nigel.croxon (a] hp.com> 308 1.1.1.1.4.2 rmind * Fix to not having any relocations at all. 309 1.1.1.1.4.2 rmind 310 1.1.1.1.4.2 rmind 2008-09-18 Nigel Croxon <nigel.croxon (a] hp.com> 311 1.1.1.1.4.2 rmind * Use LIBDIR in makefiles 312 1.1.1.1.4.2 rmind * Add setjmp/longjmp 313 1.1.1.1.4.2 rmind * Fixes incorrect section attribute in crt0-efi-ia32.S 314 1.1.1.1.4.2 rmind * Adds value EfiResetShutdown to enum EFI_RESET_TYPE 315 1.1.1.1.4.2 rmind * Fixes a RAW warning in reloc_ia64.S 316 1.1.1.1.4.2 rmind * Adds the USB HCI device path structure in the headers 317 1.1.1.1.4.2 rmind patches were supplied by Peter Jones @ RedHat 318 1.1.1.1.4.2 rmind 319 1.1.1.1.4.2 rmind 2008-02-22 Nigel Croxon <nigel.croxon (a] hp.com> 320 1.1.1.1.4.2 rmind * Added '-mno-red-zone' to x68_64 compiles. 321 1.1.1.1.4.2 rmind Patch provided by Mats Andersson. 322 1.1.1.1.4.2 rmind 323 1.1.1.1.4.2 rmind 2008-01-23 Nigel Croxon <nigel.croxon (a] hp.com> 324 1.1.1.1.4.2 rmind * release 3.0e to support x86_64 325 1.1.1.1.4.2 rmind EFI calling convention, the stack should be aligned in 16 bytes 326 1.1.1.1.4.2 rmind to make it possible to use SSE2 in EFI boot services. 327 1.1.1.1.4.2 rmind This patch fixes this issue. Patch provided by Huang Ying from Intel. 328 1.1.1.1.4.2 rmind 329 1.1.1.1.4.2 rmind 2007-05-11 Nigel Croxon <nigel.croxon (a] hp.com> 330 1.1.1.1.4.2 rmind * release 3.0d to support x86_64 from Chandramouli Narayanan 331 1.1.1.1.4.2 rmind from Intel and based on 3.0c-1 332 1.1.1.1.4.2 rmind 333 1.1.1.1.4.2 rmind 2006-03-21 Stephane Eranian <eranian (a] hpl.hp.com> 334 1.1.1.1.4.2 rmind * merged patch to support gcc-4.1 submitted by 335 1.1.1.1.4.2 rmind Raymund Will from Novell/SuSE 336 1.1.1.1.4.2 rmind 337 1.1.1.1.4.2 rmind 2006-03-20 Stephane Eranian <eranian (a] hpl.hp.com> 338 1.1.1.1.4.2 rmind * updated ia-64 and ia-32 linker scripts to 339 1.1.1.1.4.2 rmind match latest gcc. The new gcc may put functions in 340 1.1.1.1.4.2 rmind .text* sections. patch submitted by H.J. Lu from Intel. 341 1.1.1.1.4.2 rmind 342 1.1.1.1.4.2 rmind 2004-11-19 Stephane Eranian <eranian (a] hpl.hp.com> 343 1.1.1.1.4.2 rmind * added patch to ignore .eh_frame section for IA-32. Patch 344 1.1.1.1.4.2 rmind submitted by Jim Wilson 345 1.1.1.1.4.2 rmind 346 1.1.1.1.4.2 rmind 2004-09-23 Stephane Eranian <eranian (a] hpl.hp.com> 347 1.1.1.1.4.2 rmind * added patch to discard unwind sections, newer toolchains 348 1.1.1.1.4.2 rmind complained about them. Patch submitted by Jesse Barnes from SGI. 349 1.1.1.1.4.2 rmind 350 1.1.1.1.4.2 rmind 2003-09-29 Stephane Eranian <eranian (a] hpl.hp.com> 351 1.1.1.1.4.2 rmind * updated elf_ia64_efi.lds to reflect new data sections 352 1.1.1.1.4.2 rmind created by gcc-3.3. Patch provided by Andreas Schwab from Suse. 353 1.1.1.1.4.2 rmind 354 1.1.1.1.4.2 rmind 2003-06-20 Stephane Eranian <eranian (a] hpl.hp.com> 355 1.1.1.1.4.2 rmind * updated elf_ia64_efi.lds and elf_ia32_efi.lds to include 356 1.1.1.1.4.2 rmind new types data sections produced by recent version of gcc-3.x 357 1.1.1.1.4.2 rmind 358 1.1.1.1.4.2 rmind 2002-02-22 Stephane Eranian <eranian (a] hpl.hp.com> 359 1.1.1.1.4.2 rmind * release 3.0a 360 1.1.1.1.4.2 rmind * modified both IA-64 and IA-32 loader scripts to add support for the 361 1.1.1.1.4.2 rmind new .rodata sections names (such as rodata.str2.8). Required 362 1.1.1.1.4.2 rmind for new versions of gcc3.x. 363 1.1.1.1.4.2 rmind 364 1.1.1.1.4.2 rmind 2001-06-20 Stephane Eranian <eranian (a] hpl.hp.com> 365 1.1.1.1.4.2 rmind * release 3.0 366 1.1.1.1.4.2 rmind * split gnu-efi package in two different packages: the libary+include+crt and the bootloader. 367 1.1.1.1.4.2 rmind * removed W2U() hack and related files to get from wide-char to unicode. 368 1.1.1.1.4.2 rmind * Use -fshort-wchar option for unicode. 369 1.1.1.1.4.2 rmind * restructured Makefiles now install under INSTALLROOT. 370 1.1.1.1.4.2 rmind 371 1.1.1.1.4.2 rmind 2001-04-06 Stephane Eranian <eranian (a] hpl.hp.com> 372 1.1.1.1.4.2 rmind 373 1.1.1.1.4.2 rmind * incorporated patches from David and Michael Johnston at Intel 374 1.1.1.1.4.2 rmind to get the package to compile for IA-32 linux target. 375 1.1.1.1.4.2 rmind 376 1.1.1.1.4.2 rmind * Fixed ELILO to compile for Ia-32 (does not execute yet, though): 377 1.1.1.1.4.2 rmind Makefile and start_kernel() function. 378 1.1.1.1.4.2 rmind 379 1.1.1.1.4.2 rmind 2001-04-06 Andreas Schwab <schwab (a] suse.de> 380 1.1.1.1.4.2 rmind 381 1.1.1.1.4.2 rmind * Fixed config.c to 382 1.1.1.1.4.2 rmind get the timeout directive to do something. implemented the global 383 1.1.1.1.4.2 rmind root= directive. 384 1.1.1.1.4.2 rmind 385 1.1.1.1.4.2 rmind * Fix the efi_main() to deal with the -C option properly 386 1.1.1.1.4.2 rmind 387 1.1.1.1.4.2 rmind 2001-04-05 Stephane Eranian <eranian (a] hpl.hp.com> 388 1.1.1.1.4.2 rmind 389 1.1.1.1.4.2 rmind * update efi library to latest EFI toolkit 1.02 as distributed 390 1.1.1.1.4.2 rmind by Intel. Fixed header + library files to compile with GCC 391 1.1.1.1.4.2 rmind 392 1.1.1.1.4.2 rmind * merged ELI and LILO (as of gnu-efi-1.1) together, mostly 393 1.1.1.1.4.2 rmind taking the config file feature of ELI. 394 1.1.1.1.4.2 rmind 395 1.1.1.1.4.2 rmind * renamed LILO to ELILO to make the distinction 396 1.1.1.1.4.2 rmind 397 1.1.1.1.4.2 rmind * restructured code to make it easier to understand and maintain 398 1.1.1.1.4.2 rmind 399 1.1.1.1.4.2 rmind * fixed FPSWA driver checking and loading: we try all possible 400 1.1.1.1.4.2 rmind files and let the driver itself figure out if it is the most 401 1.1.1.1.4.2 rmind recent. 402 1.1.1.1.4.2 rmind * added support for compression (gzip) but keep support for plain 403 1.1.1.1.4.2 rmind ELF image. ELILO autodetects the format 404 1.1.1.1.4.2 rmind 405 1.1.1.1.4.2 rmind * change the way the kernel is invoked. Now we call it in 406 1.1.1.1.4.2 rmind physical memory mode. This breaks the dependency between the 407 1.1.1.1.4.2 rmind kernel code and the loader. No more lilo_start.c madness. 408 1.1.1.1.4.2 rmind 409 1.1.1.1.4.2 rmind * changed the way the boot_params are passed. We don't use the 410 1.1.1.1.4.2 rmind ZERO_PAGE_ADDR trick anymore. Instead we use EFI runtime memory. 411 1.1.1.1.4.2 rmind The address of the structure is passed to the kernel in r28 412 1.1.1.1.4.2 rmind by our convention. 413 1.1.1.1.4.2 rmind 414 1.1.1.1.4.2 rmind * released as gnu-efi-2.0 415 1.1.1.1.4.2 rmind 416 1.1.1.1.4.2 rmind 2001-04-03 David Mosberger <davidm (a] hpl.hp.com> 417 1.1.1.1.4.2 rmind 418 1.1.1.1.4.2 rmind * gnuefi/reloc_ia32.c (_relocate): Change return type from "void" 419 1.1.1.1.4.2 rmind to "int". Return error status if relocation fails for some 420 1.1.1.1.4.2 rmind reason. 421 1.1.1.1.4.2 rmind 422 1.1.1.1.4.2 rmind * gnuefi/elf_ia32_efi.lds: Drop unneeded ".rel.reloc" section. 423 1.1.1.1.4.2 rmind 424 1.1.1.1.4.2 rmind * gnuefi/crt0-efi-ia32.S (_start): Exit if _relocate() returns with 425 1.1.1.1.4.2 rmind non-zero exit status. 426 1.1.1.1.4.2 rmind 427 1.1.1.1.4.2 rmind * inc/ia32/efibind.h [__GNUC__]: Force 8-byte alignment for 64-bit 428 1.1.1.1.4.2 rmind types as that is what EFI appears to be expecting, despite the 429 1.1.1.1.4.2 rmind "#pragma pack()" at the beginning of the file! 430 1.1.1.1.4.2 rmind 431 1.1.1.1.4.2 rmind 2001-03-29 David Mosberger <davidm (a] hpl.hp.com> 432 1.1.1.1.4.2 rmind 433 1.1.1.1.4.2 rmind * gnuefi/reloc_ia32.c: Add a couple of defines to work around 434 1.1.1.1.4.2 rmind libc/efilib collision on uint64_t et al. 435 1.1.1.1.4.2 rmind (_relocate): Use ELF32_R_TYPE() instead of ELFW(R_TYPE)(). 436 1.1.1.1.4.2 rmind 437 1.1.1.1.4.2 rmind * gnuefi/crt0-efi-ia32.S (dummy): Add a dummy relocation entry. 438 1.1.1.1.4.2 rmind 439 1.1.1.1.4.2 rmind 2001-03-29 David Mosberger <davidm (a] hpl.hp.com> 440 1.1.1.1.4.2 rmind 441 1.1.1.1.4.2 rmind * gnuefi/reloc_ia32.c: Add a couple of defines to work around 442 1.1.1.1.4.2 rmind libc/efilib collision on uint64_t et al. 443 1.1.1.1.4.2 rmind (_relocate): Use ELF32_R_TYPE() instead of ELFW(R_TYPE)(). 444 1.1.1.1.4.2 rmind 445 1.1.1.1.4.2 rmind * gnuefi/crt0-efi-ia32.S (dummy): Add a dummy relocation entry. 446 1.1.1.1.4.2 rmind 447 1.1.1.1.4.2 rmind 2000-10-26 David Mosberger <davidm (a] hpl.hp.com> 448 1.1.1.1.4.2 rmind 449 1.1.1.1.4.2 rmind * gnuefi/elf_ia64_efi.lds: Mention .rela.sdata. 450 1.1.1.1.4.2 rmind 451 1.1.1.1.4.2 rmind * Make.defaults (CFLAGS): Remove -nostdinc flags so we can pick 452 1.1.1.1.4.2 rmind up the C compiler's stdarg.h. 453 1.1.1.1.4.2 rmind 454 1.1.1.1.4.2 rmind * inc/stdarg.h: Remove this file. It's not correct for gcc (nor 455 1.1.1.1.4.2 rmind most other optimizing compilers). 456 1.1.1.1.4.2 rmind 457 1.1.1.1.4.2 rmind 2000-10-10 Stephane Eranian <eranian (a] hpl.hp.com> 458 1.1.1.1.4.2 rmind 459 1.1.1.1.4.2 rmind * cleaned up the error message and printing of those. 460 1.1.1.1.4.2 rmind * added support to load the FPSWA from a file in case support is not 461 1.1.1.1.4.2 rmind present in the firmware already 462 1.1.1.1.4.2 rmind * fixed split_args() to do the right thing when you have leading spaces 463 1.1.1.1.4.2 rmind before kernel name 464 1.1.1.1.4.2 rmind * changed the argify() function to rely on \0 instead of LoadOptionSize 465 1.1.1.1.4.2 rmind as the field seems to be broken with current firmware 466 1.1.1.1.4.2 rmind * bumped version to 1.0 467 1.1.1.1.4.2 rmind 468 1.1.1.1.4.2 rmind 2000-10-04 David Mosberger <davidm (a] hpl.hp.com> 469 1.1.1.1.4.2 rmind 470 1.1.1.1.4.2 rmind * gnuefi/reloc_ia64.S: Reserve space for up to 750 function descriptors. 471 1.1.1.1.4.2 rmind 472 1.1.1.1.4.2 rmind * gnuefi/elf_ia64_efi.lds: Add .sdata section for small data and 473 1.1.1.1.4.2 rmind put __gp in the "middle" of it. 474 1.1.1.1.4.2 rmind 475 1.1.1.1.4.2 rmind * gnuefi/crt0-efi-ia64.S (_start): Use movl/add to load 476 1.1.1.1.4.2 rmind gp-relative addresses that could be out of the range of the addl 477 1.1.1.1.4.2 rmind offset. 478 1.1.1.1.4.2 rmind * gnuefi/reloc_ia64.S (_relocate): Ditto. 479 1.1.1.1.4.2 rmind 480 1.1.1.1.4.2 rmind * apps/Makefile: Remove standard rules and include Make.rules instead. 481 1.1.1.1.4.2 rmind * lilo/Makefile: Ditto. 482 1.1.1.1.4.2 rmind 483 1.1.1.1.4.2 rmind * Make.rules: New file. 484 1.1.1.1.4.2 rmind 485 1.1.1.1.4.2 rmind 2000-08-04 Stephane Eranian <eranian (a] hpl.hp.com> 486 1.1.1.1.4.2 rmind * released version 0.9 487 1.1.1.1.4.2 rmind * incorporated ACPI changes for Asuza by NEC < kouchi (a] hpc.bs1.fc.nec.co.jp> 488 1.1.1.1.4.2 rmind * added support for initrd (-i option) original ELI code from Bill Nottingham <notting (a] redhat.com>) 489 1.1.1.1.4.2 rmind * lots of cleanups 490 1.1.1.1.4.2 rmind * got rid of #ifdef LILO_DEBUG and uses macro instead 491 1.1.1.1.4.2 rmind * fix a few extra memory leaks in create_boot_params() 492 1.1.1.1.4.2 rmind * added exit capability just before starting the kernel 493 1.1.1.1.4.2 rmind 494 1.1.1.1.4.2 rmind 2000-06-22 David Mosberger <davidm (a] hpl.hp.com> 495 1.1.1.1.4.2 rmind 496 1.1.1.1.4.2 rmind * gnuefi/elf_ia64_efi.lds: Add .srodata, .ctors, .IA64.unwind, 497 1.1.1.1.4.2 rmind .IA64.unwind_info to .data section and .rela.ctors to .rela 498 1.1.1.1.4.2 rmind section. 499 1.1.1.1.4.2 rmind 500 1.1.1.1.4.2 rmind 2000-04-03 David Mosberger <davidm (a] hpl.hp.com> 501 1.1.1.1.4.2 rmind 502 1.1.1.1.4.2 rmind * lilo/lilo.c (LILO_VERSION): Up version number to 0.9. 503 1.1.1.1.4.2 rmind 504 1.1.1.1.4.2 rmind * gnuefi/elf_ia64_efi.lds: Include .IA_64.unwind and 505 1.1.1.1.4.2 rmind .IA_64.unwind_info in .data segment to avoid EFI load error 506 1.1.1.1.4.2 rmind "ImageAddress: pointer outside of image" error due to the .dynsym 507 1.1.1.1.4.2 rmind relocations against these sections. 508 1.1.1.1.4.2 rmind 509 1.1.1.1.4.2 rmind * ChangeLog: Moved from lilo/ChangeLogs. 510 1.1.1.1.4.2 rmind 511 1.1.1.1.4.2 rmind * gnuefi/reloc_ia64.S: fixed typo: .space directive had constant 512 1.1.1.1.4.2 rmind 100 hardcoded instead of using MAX_FUNCTION_DESCRIPTORS 513 1.1.1.1.4.2 rmind macro. Duh. 514 1.1.1.1.4.2 rmind 515 1.1.1.1.4.2 rmind Fri Mar 17 15:19:18 PST 2000 Stephane Eranian <eranian (a] hpl.hp.com> 516 1.1.1.1.4.2 rmind 517 1.1.1.1.4.2 rmind * Released 0.8 518 1.1.1.1.4.2 rmind * replace the getopt.c with new version free with better license 519 1.1.1.1.4.2 rmind * created a documentation file 520 1.1.1.1.4.2 rmind * fix a couple of memory leaks 521 1.1.1.1.4.2 rmind * code cleanups 522 1.1.1.1.4.2 rmind * created a separate directory for lilo in the gnu-efi package. 523 1.1.1.1.4.2 rmind * added support for the BOOT_IMAGE argument to kernel 524 1.1.1.1.4.2 rmind * default is to build natively now 525
Indexes created Fri Oct 31 11:10:07 GMT 2025