msys2 3.3.6 binutils fixes #203
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Binutils >= 2.41 started making .rsrc section read-only, which caused memset(_cygheap_start, ...) to segfault. Instead, put _cygheap_start in the .cygheap section, which makes more sense anyway, but may result in reducing the difference between _cygheap_start and _cygheap_end by the fraction of a page not used in the .rsrc section. Fixes msys2#200 See-also: https://cygwin.com/pipermail/cygwin/2024-February/255472.html Signed-off-by: Jeremy Drake <github@jdrake.com>
Also allow that linkage to be dynamic, as libzstd-devel doesn't currently provide a static library. (cherry picked from commit 1387ea9)
Rather than guessing, based on just the presence of libbfd, add an explicit configuration option, to build dumper or not, defaulting to building it. This might have some use when bootstrapping Cygwin for a new architecture, or when building your own Cygwin-targetted cross-compiler, rather than installing one from the copr, along with the dependencies of libbfd. (cherry picked from commit 1b5fc91) (cherry picked from commit d7a76c6)
The `bfd.h` header file that is included in `binutils` has the line
`#include "ansidecl.h"`, which is fragile because it prefers Cygwin's
`include/ansidecl.h` (as opposed to `#include <ansidecl.h>`, which would
only look in the system include paths).
This matters because as of v2.42, `bfd.h` also makes use of the
`ATTRIBUTE_WARN_UNUSED_RESULT` macro.
So let's just copy that macro (and while at it, the other `ATTRIBUTE_*`
macros) from binutils' `ansidecl.h` file, to avoid compile errors while
compiling `dumper.o` that look like this:
/usr/include/bfd.h:2770:1: error: expected initializer before ‘ATTRIBUTE_WARN_UNUSED_RESULT’
2770 | ATTRIBUTE_WARN_UNUSED_RESULT;
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
(cherry picked from commit 44b5e4e)
|
lgtm |
jeremyd2019
added a commit
to jeremyd2019/MSYS2-packages
that referenced
this pull request
Feb 23, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes #200
Wound up having to backport more fixes for binutils than I expected, to get dumper.exe included again.
Corinna suggested in https://cygwin.com/pipermail/cygwin/2024-February/255472.html that there was probably no good reason for
_cygheap_startbeing in.rsrcinstead of.cygheapsection, so moved it. This at least works well enough to run and update other packages on both i686 and x86_64, so is definitely an improvement on the old behavior of crashing immediately.As I mentioned in the commit message, this may result in reducing the difference between
_cygheap_startand_cygheap_endby the fraction of a page not used in the.rsrcsection. This works out to moving_cygheap_startforward by 3064 bytes on i686, from 0x408 bytes into the.rsrcpage to the start of the.cygheapsection, which should have reduced the difference between_cygheap_endand_cygheap_startfrom 2137080 to 2134016 bytes. But for some reason, in practice this difference is now 2138112 bytes.On x86_64, the used portion of
.rsrcis also0x408bytes, and the difference between end and start went from 3210232 to 3182592 bytes.