summaryrefslogtreecommitdiff
path: root/arm-unknown-riscos
Commit message (Collapse)AuthorAgeFilesLines
* RISC OS: update makerun to fe55b6bfJohn-Mark Bell2023-03-141-1/+1
|
* RISC OS: update makerun to dc839346John-Mark Bell2023-03-111-1/+1
|
* RISC OS: update makerun to a6a833a5John-Mark Bell2023-03-101-1/+1
|
* Extend POSIX locale patch to cover most APIsJohn-Mark Bell2022-05-301-27/+595
| | | | The only missing one that's defined by POSIX is strftime_l.
* UnixLib: add POSIX extended locale supportJohn-Mark Bell2022-05-301-0/+1352
| | | | | This is not a complete implementation, but is sufficient for OpenSSL3's purposes.
* arm-unknown-riscos: fetch GCC sources as a tarballJohn-Mark Bell2022-05-281-0/+11
|
* arm-unknown-riscos: update to 4.7.4r6John-Mark Bell2022-05-287-488/+9
|
* Fetchsrc: use httpsJohn-Mark Bell2022-05-281-1/+1
|
* fix RISCOS cross compiler target to correctly retrieve infozipVincent Sanders2019-02-131-18/+24
|
* cache more source components on teh ci infrastructureVincent Sanders2018-01-266-6/+88
|
* Update RISC OS toolchain a bit, hopefully it can build more nowDaniel Silverstone2017-09-195-0/+65
|
* Patch PPL version to build with gcc 6Jeffrey Lee2017-09-191-0/+175
|
* Revert "update ppl library version for risc os builds"Vincent Sanders2017-09-161-11/+0
| | | | This reverts commit a04a4d532b4a48abbc47afcc01488171e7a411d8.
* update ppl library version for risc os buildsVincent Sanders2017-09-161-0/+11
|
* Backport change to avoid SWP from upstreamJohn-Mark Bell2016-03-251-0/+304
|
* RISC OS: fix up tooling for GCCSDK 4.7.4r2John-Mark Bell2015-11-135-31/+10
| | | | | | | | | | | | * Makefile: Update to reflect upstream tree structure * recipes/patches/gcc4/binutils-plugins.p, recipes/patches/gcc4/gmp-force-build.p: Refresh context * recipes/patches/gcc4/python2.5.p: Drop: Python 2.5 is ancient * recipes/patches/gcc4/riscos.md.p: Drop: merged upstream
* Update RISC OS GCCSDK to 4.7.4v2 releaseVincent Sanders2015-11-121-2/+2
| | | | | This is required to address an issue with poll returning bad fd errors which cause netsurf fetches to fail.
* RISC OS: add patch to fix stack corruptionJohn-Mark Bell2015-08-141-0/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Given a function such as this: void foo(void) { register unsigned int sp __asm__("sp"); _exit(sp); } GCCSDK 4.7.4 release 1 will generate the following output when the optimiser is enabled (it doesn't matter which optimisation level is chosen, so long as it's >0): mov ip, sp stmfd sp!, {fp, ip, lr, pc} cmp sp, sl bllt __rt_stkovf_split_small mov r0, sp bl _exit If this function is called from a parent that has caused the current stack chunk to be fully utilised (i.e. SP on entry to foo is less than SL), then the stack chunk extender will be called. __rt_stkovf_split_small will replace the return address of the current stack frame with the address of the stack chunk cleanup function (that foo is effectively noreturn doesn't matter here). The real return address is squirreled away in a field at the base of the new stack chunk, and will be retrieved by the cleanup code. In the function prologue emitted above, however, the frame pointer is not updated before the stack check is performed. The result is that the *parent* function's stack frame will be modified instead. This causes much badness as the parent function is using a completely different stack chunk and so, when it returns to its parent, we will very likely branch through zero (if the parent's stack chunk is the initial chunk) or return to some unexpected place further up the call stack, most likely with the wrong result values (if the parent's stack chunk is not the initial chunk) To fix this, we mark rt_stkovf_v5_clobbered as using r11 (fp), in much the same way as rt_stkovf already does. This prevents the peephole optimiser optimising out the frame pointer update. This results in this much more sensible output: mov ip, sp stmfd sp!, {fp, ip, lr, pc} sub fp, ip, #4 cmp sp, sl bllt __rt_stkovf_split_small mov r0, sp bl _exit This ensures that the correct stack frame is modified by __rt_stkovf_split_small. Note that, in this particular case, foo does not return, so the stack chunk cleaning won't happen. This isn't really a problem, as the only real ways out of functions which do not return are process exit, or longjmp which, in the UnixLib implementation, explicitly cleans up stack chunks before returning control to the location specified in the jmp_buf.
* Update pinned GCCSDK revision to include recent fixes, importantly gmplib ↵Rob Kendrick2015-03-041-1/+1
| | | | download URL
* force arm-unknowm-riscos gmp build to x86_64 otherwise compiling on VM hosts ↵Vincent Sanders2014-12-101-0/+11
| | | | fails
* unecessary to set version explictlyVincent Sanders2014-12-091-11/+0
|
* force gcc 4.7 svn versionVincent Sanders2014-12-091-0/+11
|
* Merge branch 'rjek/gccsdk-4.7.4' of ssh://git.netsurf-browser.org/toolchains ↵rjek/gccsdk-4.7.4Rob Kendrick2014-11-161-0/+3
|\ | | | | | | into rjek/gccsdk-4.7.4
| * Remove trampoline binaries created by newer GCCSDKRob Kendrick2014-11-161-0/+3
| |
* | Build RISC OS binutils with --enable-pluginsRob Kendrick2014-11-161-0/+11
|/
* Upgrade GCCSDK and OSLibRob Kendrick2014-11-163-387/+7
| | | | | | - GCCSDK is tied to a specific trunk revision, which appears to work and gives us 4.7.4 - OSLib upgraded from specific trunk revision to 7.00 release.
* Further texinfo fixesJohn-Mark Bell2014-11-151-48/+151
|
* Further texinfo fixes for binutils 2.21John-Mark Bell2014-11-151-1/+175
|
* Further fixups for GCCSDK buildJohn-Mark Bell2014-11-151-1/+41
|
* Speculative fix for RISC OS GCCSDK build.John-Mark Bell2014-11-151-0/+48
|
* ARM-Unknown-RISCOS: fix OSLib SVN URI.John-Mark Bell2014-01-051-1/+1
|
* make spidermonkey 1.8.5 build without JIT on RISC OSVincent Sanders2012-12-161-0/+2
| | | | add support for spidermonkey 1.7.0 source
* Dirty hack to make GCCSDK build with Python2.5John-Mark Bell2012-11-071-0/+11
|
* Switch from Infozip's FTP site to sourceforge for downloadDaniel Silverstone2012-11-031-2/+2
|
* Fix GIT URIs, add Squeeze to targets.John-Mark Bell2012-10-071-3/+3
|
* Change squeeze var. names.Michael Drake2012-10-071-3/+3
|
* Add squeeze. Get makerun from git. Untested.Michael Drake2012-10-071-4/+21
|
* Remove @.Michael Drake2012-10-011-1/+1
|
* Update comments.Michael Drake2012-10-011-2/+2
|
* Update to GCCSDK 4.1.2 v2.Michael Drake2012-10-014-126/+2
|
* Update to build from scratchVincent Sanders2012-07-171-0/+2
| | | | Add basic README containing dependancies and risc os build instructions
* Patch to fix bug in UnixLib stack extension causing the signal handler stack ↵John Mark Bell2011-10-271-0/+88
| | | | | | to be trampled. svn path=/toolchains/; revision=13081
* Fixup bitrot:John Mark Bell2011-10-274-41/+127
| | | | | | | | * Makefile now uses the same scheme as the other toolchains for dependency tracking * Binutils 2.21 is no longer available so migrate to 2.21.1 with a partial backport of GCCSDK r5195 * Introduce patch to fix bug in defmod that causes OSLib builds to fail with GAS 2.21.1 (will be upstreamed asap) svn path=/toolchains/; revision=13080
* Make my views on installing build host tooling into the target environment clearJohn Mark Bell2011-01-041-0/+1
| | | | svn path=/toolchains/; revision=11195
* Add Infozip to toolchain (patches shamelessly stolen from the GCCSDK ↵John Mark Bell2011-01-044-1/+252
| | | | | | autobuilder) svn path=/toolchains/; revision=11194
* Add makerun to toolchainJohn Mark Bell2011-01-021-1/+15
| | | | svn path=/toolchains/; revision=11187
* Move OSLib build to the arm-unknown-riscos toolchainJohn Mark Bell2011-01-021-5/+44
| | | | | | Add CCRes to the toolchain svn path=/toolchains/; revision=11183
* Build runes for the RISC OS toolchainJohn Mark Bell2010-12-302-0/+41
svn path=/toolchains/; revision=11170