We're about to need an atomic uint8_t for state operations.
Unfortunately, we're at the point where things won't get inlined into the key
methods unless they're force-inlined. This is embarassing and we should do
something about it, but in the meantime we'll force-inline a little more when we
need to.
When configured with --with-lg-page, it's possible for the configured page size
to be greater than the system page size, in which case the page address may only
be aligned with the system page size.
When configuring out-of-tree (source directory is not build directory),
the generated include files from the build directory should have higher
priority than those in the source dir.
This is especially helpful when cross-compiling.
Signed-off-by: Christoph Muellner <christoph.muellner@theobroma-systems.com>
This patch allows to override the lg-vaddr values, which
are defined by the build machine's CPUID information (x86_64)
or default values (other architectures like aarch64).
Signed-off-by: Christoph Muellner <christoph.muellner@theobroma-systems.com>
Instead of setting a fix value of 48 allowed VA bits,
we distiguish between LP64 and ILP32.
Testsuite result with LP64:
Test suite summary: pass: 13/13, skip: 0/13, fail: 0/13
Testsuit result with ILP32:
Test suite summary: pass: 13/13, skip: 0/13, fail: 0/13
Signed-off-by: Christoph Muellner <christoph.muellner@theobroma-systems.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Previously, we would leak the extent and memory associated with a salvageable
portion of an extent that we were trying to split in three, in the case where
the first split attempt succeeded and the second failed.
Right now we always make our TLS use the initial-exec model if the compiler
supports it. This change allows configure-time disabling of this setting, which
can be helpful when dynamically loading jemalloc is the only option.
Looking at the thread counts in our services, jemalloc's background thread
is useful, but mostly idle. Add a config option to tune down the number of threads.
szind and slab bits are read on fast path, where compiler generated two memory
loads separately for them before this diff. Manually operate on the bits to
avoid the extra memory load.
preserve_lru feature adds lots of complication, for little value.
Removing it means merged extents are re-added to the lru list, and may
take longer to madvise away than they otherwise would.
Canaries after removal seem flat for several services (no change).
