threads.impure/custom-stack-size can fail on win32 too
View Changes

Delete unused code (dead on arrival, in fact)
View Changes

Ignore errors in maybe-reparse-specifier
View Changes

Fix busted signed-sap-ref-indexed vop
View Changes

Fix file-position after unread-char with an empty buffer
View Changes

Mark threads.impure/custom-stack-size as failing
View Changes

Fix a test for -64-bit
View Changes

Print subnormal floats with their shortest digits
View Changes

Allow setting thread_control_stack_size at runtime
View Changes

Delete slow and non-critical test that came from rev 94ea2b2082
View Changes

Add local macro like current-thread-offset-sap for any thread
View Changes

arm64: fix utf8-crlf-to-character-string
View Changes

Simplify some math around altstack
View Changes

Combine redundant parts of arch_os_thread_init
View Changes

Change thread_control_stack_size to uint
View Changes

Remove THREAD_ALIGNMENT_BYTES
View Changes

ceiling/floor: enable word fast path on arm64
View Changes

Fix comment that went stale 6 years ago (rev 8ce0e3fb37)
View Changes

Fix a test
View Changes

Fix complex-tanh on large numbers
View Changes

sb-manual: document customize-target-features.lisp
View Changes

sb-manual: add @shortcontents, capitalize SBCL in the title
View Changes

Warn about bad types for make-sequence, map, concatenate
View Changes

Don't fail on empty types in a MAP transform
View Changes

combination-info might clash for different infos
View Changes

Ordering issues caused by set-difference
View Changes

Continued combination-case to cmatch transition
View Changes

Fix +/- constant association for ratios when there are integer casts
View Changes

sb-manual: more logical structure, less front-loading
View Changes

Finish removing the old combination-match macro
View Changes