nosuchtim / keykit

KeyKit - an algorithmic MIDI scripting language and GUI system
Other
98 stars 7 forks source link

Realloc #3072

Closed pbarada closed 1 year ago

pbarada commented 1 year ago

Replace kmalloc/copy/kfree sequences with added krealloc. Cache task stacks for reuse (saves stack allocation/free for every note played in group tool), reallocating them if too large when cached.

pbarada commented 1 year ago

Grrr, while looking at some other bits, this branch crashes in execerror with "Uknown data type (8) in getnumval!" (at ../lib/maps.k:295). Trying to figure out why...

pbarada commented 1 year ago

Closing pull request; there's something really weird going on I need to get to the bottom of. Even backing out last commit (that uses realloc in expandstackatleast) that worked yesterday - now fails. 'make regress_stdio" worked yesterday and now fails miserably. Back to the drawing board...

nosuchtim commented 1 year ago

Probably worthwhile running some tests automatically when compiling from the top-level makefile.

On Wed, Apr 19, 2023 at 7:46 AM pbarada @.***> wrote:

Closing pull request; there's something really weird going on I need to get to the bottom of. Even backing out last commit (that uses realloc in expandstackatleast) that worked yesterday - now fails. 'make regress_stdio" worked yesterday and now fails miserably. Back to the drawing board...

— Reply to this email directly, view it on GitHub https://github.com/nosuchtim/keykit/pull/3072#issuecomment-1514870954, or unsubscribe https://github.com/notifications/unsubscribe-auth/AABESNSAHKHQZOBEBGDXJ7TXB73EFANCNFSM6AAAAAAXC33BC4 . You are receiving this because your review was requested.Message ID: @.***>

pbarada commented 1 year ago

besides "regress_stdio", what other tests are available to run?

On 4/19/23 11:25, Tim Thompson wrote:

Probably worthwhile running some tests automatically when compiling from the top-level makefile.

On Wed, Apr 19, 2023 at 7:46 AM pbarada @.***> wrote:

Closing pull request; there's something really weird going on I need to get to the bottom of. Even backing out last commit (that uses realloc in expandstackatleast) that worked yesterday - now fails. 'make regress_stdio" worked yesterday and now fails miserably. Back to the drawing board...

— Reply to this email directly, view it on GitHub https://github.com/nosuchtim/keykit/pull/3072#issuecomment-1514870954, or unsubscribe

https://github.com/notifications/unsubscribe-auth/AABESNSAHKHQZOBEBGDXJ7TXB73EFANCNFSM6AAAAAAXC33BC4 . You are receiving this because your review was requested.Message ID: @.***>

— Reply to this email directly, view it on GitHub https://github.com/nosuchtim/keykit/pull/3072#issuecomment-1514934068, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABMXGS7S7F3UVGR5FRILZVLXB77YJANCNFSM6AAAAAAXC33BC4. You are receiving this because you modified the open/close state.Message ID: @.***>

-- Peter Barada @.***

nosuchtim commented 1 year ago

There are no other automated tests.

On Wed, Apr 19, 2023 at 9:04 AM pbarada @.***> wrote:

besides "regress_stdio", what other tests are available to run?

On 4/19/23 11:25, Tim Thompson wrote:

Probably worthwhile running some tests automatically when compiling from the top-level makefile.

On Wed, Apr 19, 2023 at 7:46 AM pbarada @.***> wrote:

Closing pull request; there's something really weird going on I need to get to the bottom of. Even backing out last commit (that uses realloc in expandstackatleast) that worked yesterday - now fails. 'make regress_stdio" worked yesterday and now fails miserably. Back to the drawing board...

— Reply to this email directly, view it on GitHub <https://github.com/nosuchtim/keykit/pull/3072#issuecomment-1514870954 , or unsubscribe

< https://github.com/notifications/unsubscribe-auth/AABESNSAHKHQZOBEBGDXJ7TXB73EFANCNFSM6AAAAAAXC33BC4

. You are receiving this because your review was requested.Message ID: @.***>

— Reply to this email directly, view it on GitHub https://github.com/nosuchtim/keykit/pull/3072#issuecomment-1514934068,

or unsubscribe < https://github.com/notifications/unsubscribe-auth/ABMXGS7S7F3UVGR5FRILZVLXB77YJANCNFSM6AAAAAAXC33BC4 . You are receiving this because you modified the open/close state.Message ID: @.***>

-- Peter Barada @.***

— Reply to this email directly, view it on GitHub https://github.com/nosuchtim/keykit/pull/3072#issuecomment-1514991671, or unsubscribe https://github.com/notifications/unsubscribe-auth/AABESNQEXOIUF4CIQFPW753XCAEJJANCNFSM6AAAAAAXC33BC4 . You are receiving this because your review was requested.Message ID: @.***>