I found the following answer elsewhere. Someone else asked the same question I did. I'm entering the answer below for those who might be wondering the same thing.
For those unfamiliar with ARM here is a brief intro (my interpretation). ARM is a cpu design created by a British company (Acorn?) that doesn't even make cpu's or microcontrollers. The design is apparently so good that major microcontroller manufacturers decided not to design their own cpu's but license the design from ARM instead. Consequently the cpu in microcontrollers from various vendors is the same design but the peripherals, which are on the same chip as the cpu, are designed by each vendor. This means that programs written for one manufacturer's ARM microcontroller typically won't work on others. There have been attempts to create higher level API's to abstract away the differences and make programs portable.
=== third party answer starts here ===
The GPIO hardware in your chip is
designed by TI instead of ARM and thus its headers have a slightly different coding style.
When using Cortex-M (ARM) microcontrollers, it's helpful to keep track of which part of the hardware were designed by ARM and which were designed by the silicon vendor that licensed the processor from ARM (TI in this case). ARM designed the CPU and the debug and tracing related hardware blocks (that you could even use to do some I/O, e.g. SWD) but everything else, GPIO, Timers, PWM, ADC, UART, SPI, I2S are all done by silicon vendor. Thus, even a simple low-level blink example would not port from TI to ST, NXP or Atmel chips even if they're all Cortex-M microcontrollers.
ARM had a CMSIS-Driver initiative that was aimed at creating a unified API for the common peripherals, but it's essentially dead. For portable high-level APIs, look to the
mBed or
Arduino projects.