Conversation
8e6a4d2 to
14f4ee4
Compare
The ldscripts(1.194) from official TI website[0] don't define `eheap` and `sheap`. These variables are required to get heap pointer, but allocation from heap seems not to be used currently. Therefore, DISABLE_HEAP_ALLOC is added to enable or disable `_sbrk_r` function by selection until we are sure that newlib for msp430 is verified. [0] http://www.ti.com/tool/MSP430-GCC-OPENSOURCE Signed-off-by: Justin Kim <justin.kim@collabora.com>
14f4ee4 to
d86d332
Compare
|
Rebase onto master, and fix |
|
Slight update about size problem. I investigated the differences between newlib which TI provides and locally compiled one. |
|
ping, needs rebase btw. |
|
How's this going along? |
|
@OlegHahm like to finish review? |
| # upstream GCC msp430 port (sometimes referred to as the Red Hat MSP430 | ||
| # compiler) is usually accompanied by newlib. | ||
| # The new msp430 port started with version 4.9.x, the latest release of the old | ||
| # mspgcc is 4.6.3 from the beginning of 2012 and doesn't look like it will receive |
There was a problem hiding this comment.
Is there any reason why we would want to support a compiler which has not been updated for 6 years and most likely won't receive any fix ever again?
|
|
||
| # use the nano-specs of Newlib when available | ||
| ifeq ($(shell $(LINK) -specs=nano.specs -E - 2>/dev/null >/dev/null </dev/null ; echo $$?),0) | ||
| export LINKFLAGS += -specs=nano.specs -lc -lnosys |
There was a problem hiding this comment.
don't think here is the right place to put these options. See #9281 and the PRs it mentions.
| ifeq (1,$(MSP430_USE_NEWLIB)) | ||
|
|
||
| # Temporarily disabled memory allocation from heap | ||
| export CFLAGS += -DDISABLE_HEAP_ALLOC |
There was a problem hiding this comment.
What does this do and why is it needed?
| while(1); | ||
| } | ||
|
|
||
| #ifndef DISABLE_HEAP_ALLOC |
There was a problem hiding this comment.
This part of this file is a bit different in thew current RIOT. Could you rebase? There is already a test for sbrk.
|
@justinjoy ping |
I took a quick look the merge conflict, but I found my PR is too old to fix it in a simple way. |
|
@haukepetersen @miri64 @kYc0o is there interest to support this board? Or shall we just close? |
|
Closed due to lack of momentum. |
|
I just want to add the memo label since I'd really like to work on it. |
This PR is made based on #4766. I rebased and took some commits from the original PR and they are little squashed, and some of them aren't used.
Unlike the original PR, I tend to use ldscripts from TI without changes.
Although I can build successfully, there are two issues. Please refer to the log below.
Command not foundis shown when starting build process, but I failed to find exact point of that.hexis too large forhello-world. (29Kb)