Post by @hipsterelectron@circumstances.run

@hipsterelectron@circumstances.run 3 mo. ago

reading the amd64 architecture programmer's manual:

One of the more common types of strings used by applications are byte data-type strings known as ASCII strings, which can be used to represent character data.

i agree that a varint encoding like utf-8 sucks too, but come on

15 Replies

25 Boosts

This comment is unavailable — it may have been deleted or is not visible to you.

d@nny disc@

@hipsterelectron@circumstances.run 3 mo. ago

omg and it even has a structured capability mechanism for granting per-task access to i/o address space instructions

d@nny disc@

@hipsterelectron@circumstances.run 3 mo. ago

oh FUCK yes

3.9 Memory Optimization
Generally, application software is unaware of the memory hierarchy implemented within a particular system design.

unless the application is a zero-day

The application simply sees a homogenous address space within a single level of memory.

except for POSIX i/o, which is separate

In reality, both system and processor implementations can use any number of techniques to speed up accesses into memory, doing so in a manner that is transparent to applications.

the techniques will only speed up accesses. it is transparent and will never leak side channels

Application software can be written to maximize this speed-up even though the methods used by the hardware are not visible to the application.

ABSOLUTELY FALSE!

d@nny disc@

@hipsterelectron@circumstances.run 3 mo. ago

the reason i'm pissed off about all this is because all my beautiful careful buffering and orchestration gets fucking TRASHED by the round-robin scheduler with globally-atomic IPC

d@nny disc@

@hipsterelectron@circumstances.run 3 mo. ago

Implementations of the AMD64 architecture commit the results of each instruction—that is, store the result of the executed instruction in software-visible resources, such as a register (including flags), the data cache, an internal write buffer, or memory—in program order, which is the order specified by the instruction sequence in a program.

transactional semantics with transparent memory management and speculative execution? great idea! i will now declare war on POSIX

d@nny disc@

@hipsterelectron@circumstances.run 3 mo. ago

Some system devices might be sensitive to reads.

oh i'm sorry system device pat pat

d@nny disc@

@hipsterelectron@circumstances.run 3 mo. ago

Normally, applications do not have direct access to system devices, but instead call an operating-system service routine to perform the access on the application’s behalf. In this case, it is system software’s responsibility to enforce strong read-ordering.

literally clubbing me over the head telling me not to use memory-mapped i/o

d@nny disc@

@hipsterelectron@circumstances.run 3 mo. ago

i get that there are going to be certain i/o devices like SSDs which might be able to optimize memory orderings over the memory bus. maybe i should care about that