aboutsummaryrefslogtreecommitdiff
path: root/libgo
AgeCommit message (Collapse)AuthorFilesLines
2014-05-07mksysinfo: Define some more non-trivial TIOC constants.Ian Lance Taylor1-0/+32
From-SVN: r210192
2014-05-07mksysinfo: Define CLONE flags.Ian Lance Taylor4-2/+12
From-SVN: r210189
2014-05-01runtime: ask $GOC rather than $CC for the version and multi-os-directoryIan Lance Taylor4-4/+4
The Go compiler may have different values for these than the C compiler. From-SVN: r209967
2014-04-25re PR go/60931 (libgo has issues when page size is not 4k)Ian Lance Taylor1-2/+12
PR go/60931 runtime: Fix garbage collector issue with non 4kB system page size The go garbage collector tracks memory in terms of 4kB pages. Most of the code checks getpagesize() at runtime and does the right thing. On a 64kB ppc64 box I see SEGVs in long running processes which has been diagnosed as a bug in scavengelist. scavengelist does a madvise(MADV_DONTNEED) without rounding the arguments to the system page size. A strace of one of the failures shows the problem: madvise(0xc211030000, 4096, MADV_DONTNEED) = 0 The kernel rounds the length up to 64kB and we mark 60kB of valid data as no longer needed. Round start up to a system page and end down before calling madvise. From-SVN: r209777
2014-04-17gofrontend: deduplicate C syscall function declarationsIan Lance Taylor1-2/+5
A gccgo language extension allows a function to be declared multiple times. Avoid the use of this extension by dedeplicating declarations in mksyscall.awk. From-SVN: r209508
2014-04-17runtime: remove use of obsolete map deletion syntaxIan Lance Taylor1-1/+1
The use of this syntax was eliminated upstream in Go 992248b2adc2, but this particular use slipped through somehow. From-SVN: r209506
2014-04-17gofrontend: avoid use of unsafe.Sizeof extensionIan Lance Taylor1-1/+1
Avoid the use of a gccgo language extension which allows unsafe.Sizeof to accept a type by passing an expression of the relevant type. From-SVN: r209503
2014-04-16libgo: Remove Solaris 8 & 9 support.Ian Lance Taylor5-55/+0
From Rainer Orth. From-SVN: r209448
2014-04-14Sync to current external repository.Chris Manghane1-24/+3
user: Ian Lance Taylor <iant@golang.org> date: Thu Apr 10 09:25:24 2014 -0700 files: go/expressions.cc description: compiler: add checks for constant overflow Prevent extremely large constants from eating all of memory. user: Chris Manghane <cmang@golang.org> date: Mon Apr 07 16:57:09 2014 -0700 files: go/gogo-tree.cc go/gogo.cc go/gogo.h go/statements.cc description: compiler: Use backend interface for variable initialization. user: Chris Manghane <cmang@golang.org> date: Thu Apr 03 19:56:05 2014 -0700 files: go/backend.h go/gogo-tree.cc go/gogo.cc go/gogo.h description: compiler: Use backend interface to build function code. changeset: 1269:6e30875d539e user: Chris Manghane <cmang@golang.org> date: Wed Apr 02 13:16:00 2014 -0700 files: go/backend.h go/gogo-tree.cc go/gogo.cc go/gogo.h description: compiler: Use backend interface for building function defer wrappers. user: Chris Manghane <cmang@golang.org> date: Mon Mar 31 12:42:49 2014 -0700 files: go/expressions.cc go/gogo-tree.cc go/gogo.cc go/gogo.h description: compiler: Use backend interface for memory allocation. user: Chris Manghane <cmang@golang.org> date: Thu Mar 27 14:22:49 2014 -0700 files: go/backend.h go/expressions.cc go/expressions.h description: compiler: Use backend interface for fixed array construction. user: Chris Manghane <cmang@golang.org> date: Mon Mar 17 21:25:04 2014 -0700 files: go/expressions.cc description: compiler: Check for loops in self-referential array types. Fixes issue 7525. user: Chris Manghane <cmang@golang.org> date: Mon Mar 17 14:31:59 2014 -0700 files: go/gogo.cc go/parse.cc description: compiler: Don't declare blank labels. Fixes issue 7539. user: Chris Manghane <cmang@golang.org> date: Mon Mar 17 13:12:32 2014 -0700 files: go/backend.h go/expressions.cc go/expressions.h go/runtime.def description: compiler: Use backend interface for call expressions. user: Chris Manghane <cmang@golang.org> date: Wed Mar 12 13:34:27 2014 -0700 files: go/expressions.cc go/expressions.h go/gogo-tree.cc go/statements.cc description: compiler: Use backend interface map construction. user: Chris Manghane <cmang@golang.org> date: Tue Mar 11 12:53:06 2014 -0700 files: go/backend.h go/expressions.cc go/gogo-tree.cc go/gogo.h description: compiler: Use backend interface for string expressions. user: Chris Manghane <cmang@golang.org> date: Sat Mar 08 15:56:59 2014 -0800 files: go/backend.h go/expressions.cc go/expressions.h description: compiler: Use backend interface for array and string indexing. user: Chris Manghane <cmang@golang.org> date: Fri Mar 07 16:02:18 2014 -0800 files: go/expressions.cc description: compiler: Use backend interface for constant expressions. user: Chris Manghane <cmang@golang.org> date: Thu Mar 06 16:00:18 2014 -0800 files: go/expressions.cc description: compiler: Use backend interface for struct construction. user: Chris Manghane <cmang@golang.org> date: Wed Mar 05 13:09:37 2014 -0800 files: go/expressions.cc description: compiler: Use backend interface for type conversions. user: Chris Manghane <cmang@golang.org> date: Tue Mar 04 07:03:47 2014 -0800 files: go/expressions.cc go/expressions.h go/gogo-tree.cc go/gogo.h go/runtime.def libgo/runtime/chan.c description: compiler: Use backend interface for channel receive. user: Chris Manghane <cmang@golang.org> date: Mon Mar 03 15:18:57 2014 -0800 files: go/backend.h go/expressions.cc go/runtime.def description: compiler: Use backend interface for builtin calls. user: Chris Manghane <cmang@golang.org> date: Mon Mar 03 07:44:35 2014 -0800 files: go/expressions.cc go/expressions.h go/types.cc go/types.h description: compiler: Use backend interface for string info. user: Chris Manghane <cmang@golang.org> date: Fri Feb 28 10:45:55 2014 -0800 files: go/expressions.cc go/expressions.h go/gogo-tree.cc go/statements.cc description: compiler: Use backend interface for map indexing. user: Chris Manghane <cmang@golang.org> date: Wed Feb 26 14:13:10 2014 -0800 files: go/expressions.cc go/expressions.h description: compiler: Use backend interface for slice value expressions. user: Chris Manghane <cmang@golang.org> date: Wed Feb 26 13:12:19 2014 -0800 files: go/backend.h go/expressions.cc go/expressions.h go/gogo-tree.cc go/runtime.def go/statements.cc description: compiler: Use backend interface for interface values. user: Chris Manghane <cmang@golang.org> date: Mon Feb 24 12:30:13 2014 -0800 files: go/expressions.cc go/expressions.h go/parse.cc go/statements.cc description: compiler: Change Heap_composite_expression to Heap_expression. user: Chris Manghane <cmang@golang.org> date: Thu Feb 20 19:47:06 2014 -0800 files: go/expressions.cc go/expressions.h go/gogo-tree.cc go/gogo.cc go/gogo.h go/types.cc go/types.h description: compiler: Use backend interface for interface method table expressions. user: Chris Manghane <cmang@golang.org> date: Mon Feb 03 14:36:20 2014 -0800 files: go/expressions.cc go/expressions.h description: compiler: Add compound expressions to the frontend. * go-gcc.cc: Include "convert.h". (Gcc_backend::string_constant_expression): New function. (Gcc_backend::real_part_expression): Likewise. (Gcc_backend::imag_part_expression): Likewise. (Gcc_backend::complex_expression): Likewise. (Gcc_backend::constructor_expression): Likewise. (Gcc_backend::array_constructor_expression): Likewise. (Gcc_backend::pointer_offset_expression): Likewise. (Gcc_backend::array_index_expression): Likewise. (Gcc_backend::call_expression): Likewise. (Gcc_backend::exception_handler_statement): Likewise. (Gcc_backend::function_defer_statement): Likewise. (Gcc_backend::function_set_parameters): Likewise. (Gcc_backend::function_set_body): Likewise. (Gcc_backend::convert_expression): Handle various type conversions. From-SVN: r209393
2014-03-12libgo: Build math package with -ffp-contract=off on non-x86.Ian Lance Taylor2-0/+4
http://golang.org/issue/7074 shows that not using -ffp-contract=off produces the wrong result for math.Log2(1) on arm64. From-SVN: r208505
2014-03-07runtime: Fix GC bug caused by Entersyscall modifying reg.Ian Lance Taylor1-4/+20
This patch fixes a rare but serious bug. The Go garbage collector only examines Go stacks. When Go code calls a function that is not written in Go, it first calls syscall.Entersyscall. Entersyscall records the position of the Go stack pointer and saves a copy of all the registers. If the garbage collector runs while the thread is executing the non-Go code, the garbage collector fetches the stack pointer and registers from the saved location. Entersyscall saves the registers using the getcontext function. Unfortunately I didn't consider the possibility that Entersyscall might itself change a register before calling getcontext. This only matters for callee-saved registers, as caller-saved registers would be visible on the saved stack. And it only matters if Entersyscall is compiled to save and modify a callee-saved register before it calls getcontext. And it only matters if a garbage collection occurs while the non-Go code is executing. And it only matters if the only copy of a valid Go pointer happens to be in the callee-saved register when Entersyscall is called. When all those conditions are true, the Go pointer might get collected incorrectly, leading to memory corruption. This patch tries to avoid the problem by splitting Entersyscall into two functions. The first is a simple function that just calls getcontext and then calls the rest of Entersyscall. This should fix the problem, provided the simple Entersyscall function does not itself modify any callee-saved registers before calling getcontext. That seems to be true on the systems I checked. But since the argument to getcontext is an offset from a TLS variable, it won't be true on a system which needs to save callee-saved registers in order to get the address of a TLS variable. I don't know why any system would work that way, but I don't know how to rule it out. I think that on any such system this will have to be implemented in assembler. I can't put the ucontext_t structure on the stack, because this function can not split stacks, and the ucontext_t structure is large enough that it could cause a stack overflow. From-SVN: r208390
2014-03-03libgo: Update to Go 1.2.1 release.Ian Lance Taylor5-12/+70
From-SVN: r208286
2014-02-21runtime: Use a better heap location on arm64 systems.Ian Lance Taylor1-3/+27
Before this, the heap location used on a 64-bit system was not available to user-space on arm64, so the "32-bit" strategy ended up being used. So use somewhere that is available, and for bonus points is far away from where the kernel allocates address space by default. From-SVN: r207977
2014-02-03ltmain.sh: Patch for Solaris.Ian Lance Taylor1-1/+1
From Rainer Orth. From-SVN: r207432
2014-01-22libgo/configure: Test for gold with gccgo -Wl,--help, not ld --help.Ian Lance Taylor2-4/+4
From-SVN: r206937
2014-01-17re PR go/59866 (gccgo gc work buffer is misaligned)Ian Lance Taylor1-1/+1
PR go/59866 runtime: Force work variable in mgc0 to be aligned on 8-byte boundary. From-SVN: r206738
2014-01-09runtime: fix 32-bit malloc for pointers >= 0x80000000Ian Lance Taylor3-24/+12
The spans array is allocated in runtime_mallocinit. On a 32-bit system the number of entries in the spans array is MaxArena32 / PageSize, which (2U << 30) / (1 << 12) == (1 << 19). So we are allocating an array that can hold 19 bits for an index that can hold 20 bits. According to the comment in the function, this is intentional: we only allocate enough spans (and bitmaps) for a 2G arena, because allocating more would probably be wasteful. But since the span index is simply the upper 20 bits of the memory address, this scheme only works if memory addresses are limited to the low 2G of memory. That would be OK if we were careful to enforce it, but we're not. What we are careful to enforce, in functions like runtime_MHeap_SysAlloc, is that we always return addresses between the heap's arena_start and arena_start + MaxArena32. We generally get away with it because we start allocating just after the program end, so we only run into trouble with programs that allocate a lot of memory, enough to get past address 0x80000000. This changes the code that computes a span index to subtract arena_start on 32-bit systems just as we currently do on 64-bit systems. From-SVN: r206501
2014-01-08re PR go/59430 (os/user FAILs on Solaris)Ian Lance Taylor5-10/+50
PR go/59430 os/user: Use POSIX functions on Solaris. From-SVN: r206412
2014-01-08re PR go/59433 (Many 64-bit Go tests SEGV on Solaris)Ian Lance Taylor1-16/+45
PR go/59433 net: Don't use stack space for fd_sets when using select. From-SVN: r206411
2014-01-06go/build: Set GOARCH on arm64 systems.Ian Lance Taylor6-5/+32
I am reliably informed that the architecture name and letter for the plan9/inferno compilers for 64-bit ARM systems will be "arm64" and "7" respectively, so let's get that bit in nice and early. From Michael Hudson-Doyle. https://codereview.appspot.com/34830045/ From-SVN: r206374
2014-01-06runtime: Remove unused runtime_cpuid variables.Ian Lance Taylor1-5/+0
From-SVN: r206353
2013-12-28net: work around Solaris connect issue when server closes socketIan Lance Taylor1-0/+10
On Solaris, if you do a in-progress connect, and then the server accepts and closes the socket, the client's later attempt to complete the connect will fail with EINVAL. Handle this case by assuming that the connect succeeded. This code is weird enough that it is implemented as Solaris-only so that it doesn't hide a real error on a different OS. See http://golang.org/issue/6828. From-SVN: r206232
2013-12-27re PR go/59506 (net FAILs (timeout) on alpha)Ian Lance Taylor1-1/+1
PR go/59506 net: use DialTimeout in TestSelfConnect Backported from master repository. This avoids problems with systems that take a long time to find out nothing is listening, while still testing for the self-connect misfeature since a self-connect should be fast. With this we may be able to remove the test for non-Linux systems. Tested (on GNU/Linux) by editing selfConnect in tcpsock_posix.go to always return false and verifying that TestSelfConnect then fails with and without this change. Idea from Uros Bizjak. From-SVN: r206224
2013-12-25Revert unwanted commit.Uros Bizjak1-1/+1
From-SVN: r206201
2013-12-25re PR target/59422 (Support more targets for function multi versioning)Allan Sandfeld Jensen1-1/+1
gcc/ 2013-12-25 Allan Sandfeld Jensen <sandfeld@kde.org> H.J. Lu <hongjiu.lu@intel.com> PR target/59422 * config/i386/i386.c (get_builtin_code_for_version): Handle PROCESSOR_HASWELL, PROCESSOR_SILVERMONT, PROCESSOR_BTVER1, PROCESSOR_BTVER2, PROCESSOR_BDVER3 and PROCESSOR_BDVER4. Change priority of PROCESSOR_BDVER1 to P_PROC_XOP. (fold_builtin_cpu): Add "ivybridge", "haswell", "bonnell", "silvermont", "bobcat" and "jaguar" CPU names. Add "sse4a", "fma4", "xop" and "fma" ISA names. libgcc/ 2013-12-25 Allan Sandfeld Jensen <sandfeld@kde.org> H.J. Lu <hongjiu.lu@intel.com> PR target/59422 * config/i386/cpuinfo.c (enum processor_types): Add AMD_BOBCAT and AMD_JAGUAR. (enum processor_subtypes): Add AMDFAM15H_BDVER3, AMDFAM15H_BDVER4, INTEL_COREI7_IVYBRIDGE and INTEL_COREI7_HASWELL. (enum processor_features): Add FEATURE_SSE4_A, FEATURE_FMA4, FEATURE_XOP and FEATURE_FMA. (get_amd_cpu): Handle AMD_BOBCAT, AMD_JAGUAR, AMDFAM15H_BDVER2 and AMDFAM15H_BDVER3. (get_intel_cpu): Handle INTEL_COREI7 and INTEL_COREI7_HASWELL. (get_available_features): Handle FEATURE_FMA, FEATURE_SSE4_A, FEATURE_FMA4 and FEATURE_XOP. testsuite/ 2013-12-25 Allan Sandfeld Jensen <sandfeld@kde.org> PR target/59422 * gcc.target/i386/funcspec-5.c (test_fma, test_xop, test_no_fma, test_no_xop, test_arch_corei7, test_arch_corei7_avx, test_arch_core_avx2, test_arch_bdver1, test_arch_bdver2, test_arch_bdver3, test_tune_corei7, test_tune_corei7_avx, test_tune_core_avx2, test_tune_bdver1, test_tune_bdver2 and test_tune_bdver3): New function prototypes. From-SVN: r206200
2013-12-12runtime: Fix defer of unlock thread at program startup.Ian Lance Taylor6-5/+19
Don't free stack allocated defer block. Also ensure we have a Go context in a few more places before freeing the block. From-SVN: r205940
2013-12-12reflect: Fix MakeFunc returning float32 or float64 on 386.Ian Lance Taylor2-15/+26
From-SVN: r205932
2013-12-12compiler, reflect, runtime: Implement method values in reflect.Ian Lance Taylor7-44/+120
From-SVN: r205913
2013-12-11reflect, runtime: Let reflect.MakeFunc functions call recover.Ian Lance Taylor6-2/+109
From-SVN: r205908
2013-12-06re PR go/59408 (Many Go tests FAIL with notesleep not on g0)Ian Lance Taylor1-0/+4
PR go/59408 runtime: Don't require g != m->g0 in sema notesleep. From-SVN: r205756
2013-12-04runtime: Use pthread_sigmask instead of sigprocmask.Ian Lance Taylor3-4/+4
From-SVN: r205652
2013-12-03runtime: Fix prototype and one use of runtime_traceback.Ian Lance Taylor2-2/+2
From Richard Biener. From-SVN: r205634
2013-12-01libgo: Avoid some cases of getting callers recursively.Ian Lance Taylor3-0/+19
Avoids hanging inside older versions of glibc that do not support recurive calls to dl_iterate_phdr. From-SVN: r205561
2013-11-30reflect: Rename struct field to be consistent in assembler and Go.Ian Lance Taylor1-2/+2
From-SVN: r205555
2013-11-30reflect: Fix MakeFunc for 386 when returning a struct.Ian Lance Taylor2-1/+14
When a 386 function returns a struct, it needs to return using an rtd instruction that pops the hidden struct parameter off the stack. That wasn't happening. From-SVN: r205551
2013-11-27libgo: Update to current Go library.Ian Lance Taylor30-60/+241
From-SVN: r205426
2013-11-26runtime: Fix handling of surrogate pairs in string([]rune).Ian Lance Taylor1-0/+4
From-SVN: r205422
2013-11-25syscall: Set SizeofSockaddrAny to the value the go distribution usesIan Lance Taylor1-1/+1
In particular this means that the names Getsockname returns are not truncated to 26 characters. Fixes issue 6829 https://codereview.appspot.com/31840043/ From-SVN: r205368
2013-11-24syscall: Only call varargs libc functions from C code.Ian Lance Taylor11-11/+71
From-SVN: r205321
2013-11-22libgo: Update libtool support for powerpc64le-linux-gnu.Ian Lance Taylor2-8/+8
From Ulrich Weigand. From-SVN: r205287
2013-11-19runtime: Update for change to libbacktrace library.Ian Lance Taylor1-1/+1
From-SVN: r205031
2013-11-19reflect: Handle calls to functions that take or return empty structsIan Lance Taylor2-1/+44
Fixes issue 6761 This simple change seems to work fine, slightly to my surprise. This includes the tests I submitted to the main Go repository at https://codereview.appspot.com/26570046 From-SVN: r205001
2013-11-19gotest: Recognize PPC ELF v2 function pointers in text section.Ian Lance Taylor1-1/+1
From-SVN: r205000
2013-11-19libgo: Fix typo for is_dragonfly in configure script.Ian Lance Taylor2-2/+2
From-SVN: r204999
2013-11-15runtime: Use runtime_m to get m value after call to runtime_mcall.Ian Lance Taylor1-1/+4
From-SVN: r204853
2013-11-14runtime: Don't use filename without '/' for backtrace library.Ian Lance Taylor1-0/+7
Fixes http://golang.org/issue/6715. From-SVN: r204828
2013-11-14net: On Solaris use Darwin keepalive code.Ian Lance Taylor3-27/+2
From-SVN: r204819
2013-11-14runtime: Add netpoll code that uses select.Ian Lance Taylor8-59/+270
Required for Solaris support. From-SVN: r204817
2013-11-14runtime: Fix GC flag in when allocating memory from cgo.Ian Lance Taylor1-1/+1
From-SVN: r204815
2013-11-14go/build: Add all known gccgo architectures to list.Ian Lance Taylor1-1/+1
From-SVN: r204796
generated by cgit v1.1 at 2024-09-26 21:56:57 +0000