summaryrefslogtreecommitdiffstats
path: root/contrib/gcc
Commit message (Collapse)AuthorAgeFilesLines
* Tweaks to make these files live beside our hacked config.obrien2002-06-212-9/+15
|
* Tweaks for the 64-bit compiler.obrien2002-06-211-3/+3
|
* cc -O0 and -O1 didn't do the easy optimization of alignment for space.obrien2002-06-211-1/+1
| | | | | | | Instead it attempted to do the easy optimization of alignment for time, which should be to 1-byte alignment on i386's. Submitted by: bde
* BDE prefers this organization.obrien2002-06-211-9/+11
| | | | Submitted by: bde
* cvs -j -j gave me two copies of TARGET_VERSION and I didn't noticeobrien2002-06-101-3/+0
| | | | it before committing.
* Allow one to profile FORTRAN77 programs.obrien2002-06-091-6/+33
|
* Match the DBX_OUTPUT_MAIN_SOURCE_FILE_END definition in dbxelf.h.obrien2002-06-061-2/+3
| | | | Bug found by: bde
* Allow the specification of LONG_TYPE_SIZE on the command line.obrien2002-06-061-0/+2
| | | | Submitted by: bde
* Implement "-mno-align-long-strings" which prevents pessimization of stringsobrien2002-06-062-2/+9
| | | | | | | | for space. -Os could do this, but it was easy to hack an MD version. This saves a whole 32 bytes in boot2, so I think it is worth using it. (keep how much worse gcc 3.2 will compile boot2...) Submitted by: bde (minus gcc 3.2 commentary)
* Didn't get them all in revision 1.50.obrien2002-06-061-8/+8
|
* Fix excessive alignment for the aout case, and unbreak genassym for that case.obrien2002-06-061-4/+3
| | | | | | Our aout linker has alignment hacks and doesn't need gcc to do anything special. Tested for a year by: bde :-)
* Tidy do-while.obrien2002-06-061-7/+7
| | | | Requested by: bde
* Read specs from /usr/libdata/gcc/specs if it exists.obrien2002-06-061-2/+2
|
* The brokenness in ix86_expand_clrstr is quite visible when youobrien2002-06-041-3/+3
| | | | | | compare the function with ix86_expand_movstr. Submitted by: Tor Egge
* Gcc 3.1 (-O) now generates broken inline code for memset in some cases.obrien2002-06-041-0/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This broke newfs (newfs left some garbage in a bitmap). The ASM for: #include <string.h> int x, foo[100]; main() { memset(&foo[0], 0, x); } is (at least if you have fixed function alignment): .file "z.c" .text .p2align 2,,3 .globl main .type main,@function main: pushl %ebp movl %esp, %ebp pushl %edi pushl %eax movl x, %ecx xorl %eax, %eax shrl $2, %ecx movl $foo, %edi cld rep stosl andl $-16, %esp <-- the lower bits of `len' should be loaded near here testl $2, %edi <-- this seems to be meant to test the 2^1 bit in `len' (not alignment of the pointer like it actually does). %edi is the wrong register for holding the bits, since it is still needed for the pointer. je .L2 stosw .L2: testl $1, %edi <-- similarly for the 2^0 bit. je .L3 stosb .L3: movl -4(%ebp), %edi leave ret .Lfe1: .size main,.Lfe1-main .comm foo,400,32 .comm x,4,4 .ident "GCC: (GNU) 3.1 [FreeBSD] 20020509 (prerelease)" This seems to only result in (len % 3) bytes not being cleared, since gcc doesn't seem to use the builtin memset unless it knows that the pointer is aligned. If %edi could be misaligned, then too many bytes would be set. Submitted by: BDE
* Seems we are not ready for revision 1.8.obrien2002-05-301-1/+2
|
* Make our default predefines ANSI clean.obrien2002-05-301-2/+1
|
* The special checks for %b and %D in rev 1.1 were done before skipping overobrien2002-05-261-51/+54
| | | | | | | any characters between the % and the [bD], so only plain %b and %B worked. This may un-1/2ass our -fformat-extensions support. Submitted by: bde
* 1/2assed reimplementation of c-common.c revs 1.2 (-fformat-extensions)obrien2002-05-221-6/+83
| | | | and 1.3 (printf0) for GCC 3.1.
* the processed hash fileobrien2002-05-171-0/+472
|
* MD_EXEC_PREFIX doesn't work for the cross-arch compiler.ru2002-05-161-1/+7
| | | | | | | The change also makes the `cc -print-search-dirs' output sane (the pre-3.1 way) in the non-cross case. Draft reviewed by: obrien
* Merge c-common.c rev 1.10 into GCC 3.1obrien2002-05-141-1/+3
| | | | | | | Clarify that the "yields only last 2 digits of year in some locales" warning does not apply to BSD. Submitted by: ache
* If you want to not warn on K&R main(); this is how it should be implemented.obrien2002-05-142-1/+17
|
* Merge choose-temp.c rev 1.2 (use /tmp 1st, and then /var/tmp) into GCC 3.1.obrien2002-05-141-2/+4
|
* Add framework for our kernel printf enhancements.obrien2002-05-143-0/+35
|
* do-while ASM_OUTPUT_ALIGN to enable its use in more places.obrien2002-05-121-3/+5
|
* Fixes for building a.out bits.obrien2002-05-121-5/+23
| | | | Submitted by: bde
* Localize the version number at the request of the GCC developers.obrien2002-05-111-1/+3
|
* Add tweaks needed when using as the system compiler.obrien2002-05-101-1/+21
|
* Change symbol values to match freebsd.h to avoid warnings.obrien2002-05-101-2/+2
|
* Remove files not part of GCC 3.1.obrien2002-05-10205-80711/+0
|
* Import data for 3.0.2.obrien2002-05-102-22/+20
|
* Use the stock (3.1 pre) file.obrien2002-05-092-170/+338
|
* Use the stock file for now (but leave a note behind about our OBE changes).obrien2002-05-091-42/+203
|
* Merge rev 1.2 (printf format error fixes) in its 1.8 form into Gcc 3.1 pre.obrien2002-05-091-341/+405
|
* Use the stock (3.1) file.obrien2002-05-091-11/+39
|
* We use the stock (3.1) file.obrien2002-05-091-153/+316
|
* Merge rev 1.2 (-fformat-extensions) and rev 1.7 (optimization downgrade)obrien2002-05-091-50/+77
| | | | into the Gcc 3.1 prerelease.
* Use the stock (3.1 pre) file.obrien2002-05-091-32/+57
|
* Use the stock (3.1) version.obrien2002-05-091-0/+18
|
* Merge rev 1.2 [_exit() over exit()] into the Gcc 3.1 prerelease.obrien2002-05-091-66/+124
|
* Merge revs 1.{2,15} (FREEBSD_NATIVE); 1.{5,13} (ELF, and objformat support);obrien2002-05-091-62/+47
| | | | | 1.{16,23,25} (better cross building control); 1.21 ("GCC_OPTIONS") into the GCC 3.1 prerelease.
* Merge rev 1.2 (a.out support) into the GCC 3.1 prerelease.obrien2002-05-091-158/+280
|
* Use the stock 3.1 prerelease version of this for now.obrien2002-05-091-2/+7
|
* Use the stock 3.1 prerelease file now. Rev 1.2 (add_bb) just doesn'tobrien2002-05-091-23/+23
| | | | cleanly fit in any more.
* Use the stock (3.1) file.obrien2002-05-091-2/+20
|
* We use the stock (3.1) file now.obrien2002-05-091-497/+631
|
* Use the stock (3.1) file now.obrien2002-05-091-151/+18
| | | | The choosing of TMPDIR has moved to make-temp-file.c.
* Use the stock 3.1 prerelease file.obrien2002-05-091-1/+11
|
* Use the stock 3.1-prerelease file.obrien2002-05-091-10/+142
|
OpenPOWER on IntegriCloud