- Jul 29, 2019
-
-
Guillem Jover authored
This way we can use the exit code to check whether the tests passed or failed, and fail the package build. Signed-off-by:
Guillem Jover <guillem@hadrons.org> Signed-off-by:
Jeff Moyer <jmoyer@redhat.com>
-
Guillem Jover authored
This is a recent architecture and as such does not provide legacy support for SYS_eventfd. Declare that we need to use the new syscall. Signed-off-by:
-
Guillem Jover authored
The getconf(1) command is inherently not cross-compilation friendly. In addition PAGESIZE depends on the specific system, even within a specific arch, so using a hard-coded value is never safe. Signed-off-by:
-
Guillem Jover authored
Even though this is the same size, as the sizeof() is derived from the source expression, recent gcc versions will emit a warning, which is turned into an error by -Werror: error: argument to ‘sizeof’ in ‘strncpy’ call is the same expression as the source; did you mean to use the size of the destination? [-Werror=sizeof-pointer-memaccess] Signed-off-by: -
Guillem Jover authored
Signed-off-by:
-
Stephan Springl authored
Signed-off-by:
-
Guillem Jover authored
Signed-off-by:
-
Guillem Jover authored
- Remove unnecessary macro argument quoting. - Variables, pathnames in italics. - Keywords in bold. - Man page references in bold, followed by the man page number. - Fix TP/TQ macro usage. Signed-off-by:
-
Guillem Jover authored
- Update year. - Balance double quotes. - Remove version from source argument as recommended in man-pages(7). - Fix all sections numbers to 3. Signed-off-by:
-
Guillem Jover authored
Signed-off-by:
-
Guillem Jover authored
Signed-off-by:
-
Guillem Jover authored
Signed-off-by:
-
Guillem Jover authored
Signed-off-by:
-
Guillem Jover authored
Signed-off-by:
-
Guillem Jover authored
Signed-off-by:
-
Guillem Jover authored
Signed-off-by:
-
Guillem Jover authored
Signed-off-by:
-
Guillem Jover authored
Signed-off-by:
-
Guillem Jover authored
Signed-off-by:
-
Jeff Moyer authored
A user can now specify "LIBAIO=/path/to/libaio.so" in order to run the test harness against the installed library. Signed-off-by:
-
- Oct 22, 2018
-
-
-
Jeff Moyer authored
- Add async poll support (Christoph Hellwig) - Use canonical DESTDIR= environment variable (Thomas Petazzoni) - Add ability to disable building the shared library (Thomas Petazzoni) Signed-off-by: -
Thomas Petazzoni authored
Patch borrowed from OpenEmbedded, available at recipes/libaio/libaio-0.3.106/destdir.patch in their source tree. It just adds support for the traditional DESTDIR variable to install the library in a different sysroot than the normal /. Signed-off-by:
Thomas Petazzoni <thomas.petazzoni@bootlin.com> [JEM: I couldn't verify the initial author of this change, but it's trivial. For information on the DESTDIR convention, see: https://www.gnu.org/prep/standards/html_node/DESTDIR.html ] Signed-off-by:
-
Thomas Petazzoni authored
Currently, the libaio build sytem builds a shared library unconditionally. In some environments, this is not possible or desirable. This commit adds a new Makefile variable, ENABLE_SHARED, which defaults to "1" (i.e shared library enabled) and that allows to override this behavior, and therefore to disable the build and installation of the shared library. Signed-off-by:
Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by:
-
Jeff Moyer authored
The main reason is to document that submitting patches to the mailing list is the preferred method for contributing code. Signed-off-by: -
Jeffrey E. Moyer authored
-
Jeff Moyer authored
The result is a bitmask, so just check for the bit we're interested in. Signed-off-by: -
Jeff Moyer authored
-
- Sep 28, 2018
-
-
Guillem Jover authored
We need to link agaisnt -lgcc, on at least hppa, PPC and ARC. That's because in some corner-cases like compilation with -Os on ARC and PPC so-called millicode (basically function prologue and epilogue) implemented in libgcc.a is used. So we end up with GLOBAL UNDEFINED symbol in libaio.so and then on linkage of the final applicaiton LD fails to proceed saying: --------------------------->8---------------------- hidden symbol '__ld_r13_to_r15_ret' in .../libgcc.a(_millicodethunk_ret.o) is referenced by DSO --------------------------->8---------------------- Also it looks like in general it is not the best idea to use either "-nostartfiles" or "-nostdlib" when linking shared libs because default construtor/destructor functions won't be executed, see "5.2. Library constructor and destructor functions" in [1] So let's stop passing "-nostdlib" and "-nostartfiles" and get required stuff built-in libaio. Initial patch taken from Debian [2]. Fixes build failures in Buildroot like blktrace [3], lvm2 [4]. [1] http://tldp.org/HOWTO/Program-Library-HOWTO/miscellaneous.html [2] https://sources.debian.org/patches/libaio/0.3.111-1/01_link_libs.patch/ [3] http://autobuild.buildroot.net/results/17461209755038a30118d76acb4f43469a22a139/ [4] http://autobuild.buildroot.net/results/a5dfc87f94b97135e5cc84f6a876114891ed9dd9/ Signed-off-by:
Guillem Jover <guillem@debian.org> Signed-off-by:
Alexey Brodkin <abrodkin@synopsys.com>
-
- Mar 06, 2018
-
-
Jeff Moyer authored
- Add two new tests to the test harness (Jeff Moyer) - Generic arch dectection for padding defines (Nathan Rossi) - harness: don't hardcode page size (Jeff Moyer) - harness: add a test case for mremap (Jeff Moyer) - libaio: harness: fix build errors due to attribute warn_unused_result (Mauricio Faria de Oliveira) - libaio: harness: fix build error due to linker search order (Mauricio Faria de Oliveira) - harness: add test for allocating aio-max-nr ioctxs (Jeff Moyer) - Add support for preadv2/pwritev2 (Jeff Moyer) - syscall-generic: don't overwrite errno (Jeff Moyer) - syscall: get rid of custom syscall implementation (Jeff Moyer) - Change syscall-arm64.h to syscall-generic.h (Icenowy Zheng) - Use generic syscall number schema for RISC-V (Icenowy Zheng) - Add endian detection (LE) and bit width detection (32/64) for RISC-V (Icenowy Zheng) - Makefile: convert tag and archive targets to git (Jeff Moyer) Signed-off-by: -
Jeff Moyer authored
In the process, change the tag format to not include the release. The release portion is meant for distributers to bump when packaging changes are made; it shouldn't be used for the upstream sources. There's also no need to convert ',' to '-' in the tag. Signed-off-by:
-
- Mar 05, 2018
-
-
Icenowy Zheng authored
On the RISC-V architecture, the toolchain should define __riscv macro and a __riscv_xlen macro, which indicates the length of registers (also the length of pointers and the length of the C language long type). Use them to detect the RISC-V architecture, and set them in the LE32/LE64 codepath. Signed-off-by:
Icenowy Zheng <icenowy@aosc.io> Signed-off-by:
-
Icenowy Zheng authored
The RISC-V port of Linux kernel just adapted the generic syscall number schema. Add its support in syscall.h file. Signed-off-by:
-
Icenowy Zheng authored
The arm64 architecture support in Linux kernel just adapted the new generic syscall number schema, which is shared between most new architectures. Change syscall-arm64.h to syscall-generic.h, in order to ease the addition of support for new architectures. Signed-off-by:
-
- Jan 17, 2018
-
-
Christoph Hellwig authored
Signed-off-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
-
Christoph Hellwig authored
This is ppoll/pselect equivalent for io_getevents. It atomically executes the following sequence: sigset_t origmask; pthread_sigmask(SIG_SETMASK, &sigmask, &origmask); ret = io_getevents(ctx, min_nr, nr, events, timeout); pthread_sigmask(SIG_SETMASK, &origmask, NULL); And thus allows to safely mix aio and signals, especially together with IO_CMD_POLL. See the pselect(2) man page for a more detailed explanation. Signed-off-by:
-
Christoph Hellwig authored
Signed-off-by: -
Christoph Hellwig authored
This allows to call a 6-argument syscall using the generic syscall() function from libc. Signed-off-by:
-
- Jan 16, 2018
-
-
Christoph Hellwig authored
Now that we have poll support in mainline, remove comments about the do not use status. Signed-off-by:
-
- Jan 12, 2018
-
-
Jeff Moyer authored
Maintaining the custom system call code for each architecture is a burden. I'm not convinced that we have clobber registers correct, and at best, we are just duplicating work that has already been done by the glibc maintainers. I asked Ben what the reasoning was behind not just using syscall. This was his answer: The main issue was that glibc's pthreads implementation really sucked back during initial development and there was a use-case for having the io_XXX functions usable directly from clone()ed threads that didn't have all the glibc pthread state setup for per-cpu areas to handle per-thread errno. That made sense back then, but is rather silly today. This patch gets rid of the architecture specific system call wrappers in favor of using syscall(). We leave the per-arch header files with the syscall numbers because we'll be adding a new system call soon, and this will allow us to add the new number without requiring newer kernels on build servers. Signed-off-by:
-
