Docs: Naming temporary compilation results

Author: ashvardanian

.gitignore

@@ -7,6 +7,8 @@ build_release/
 
 # Temporary binaries
 /tmp/
-less_slow_from_ptx.cubin
 less_slow_from_cu.cubin
 less_slow_from_cu.ptx
+less_slow_sm70_from_ptx.cubin
+less_slow_sm80_from_ptx.cubin
+less_slow_sm90a_from_ptx.cubin

less_slow.cu

@@ -23,6 +23,14 @@
  *  $ nvcc -arch=sm_90a -Xptxas -v -lineinfo -cubin -o less_slow_from_cu.cubin less_slow.cu
  *  $ cuobjdump -sass less_slow_from_cu.cubin | grep -i mma
  *
+ *  Assuming how aggressively NVCC unrolls loops and the number of kernels in
+ *  this file, you may want to deduplicate them:
+ *
+ *  $ cuobjdump -sass less_slow_from_cu.cubin | grep -i mma | \
+ *  $   sed -r 's/\/\*[^*]+\*\///g' | \
+ *  $   sed -r 's/^[[:space:]]+//; s/[[:space:]]+$//' | \
+ *  $   sort -u
+ *
  *  Keep in mind the following TC generations:
  *
  *  - Volta SM70: 1st generation of TCs, server V100 cards.

less_slow_sm70.ptx

@@ -18,8 +18,16 @@
  *  You can validate this file by asking the Nvidia PTX Assembler to compile it
  *  to `.cubin` for some target architecture:
  * 
- *  $ ptxas -o less_slow_from_ptx.cubin -arch=sm_70 less_slow_sm70.ptx
- *  $ cuobjdump -sass less_slow_from_ptx.cubin | grep -i mma
+ *  $ ptxas -o less_slow_sm70_from_ptx.cubin -arch=sm_70 less_slow_sm70.ptx
+ *  $ cuobjdump -sass less_slow_sm70_from_ptx.cubin | grep -i mma
+ *
+ *  Assuming how aggressively NVCC unrolls loops and the number of kernels in
+ *  this file, you may want to deduplicate them:
+ *
+ *  $ cuobjdump -sass less_slow_sm70_from_ptx.cubin | grep -i mma | \
+ *  $   sed -r 's/\/\*[^*]+\*\///g' | \
+ *  $   sed -r 's/^[[:space:]]+//; s/[[:space:]]+$//' | \
+ *  $   sort -u
  * 
  *  @section Register File
  * 

less_slow_sm80.ptx

@@ -13,13 +13,13 @@
  *  You can validate this file by asking the Nvidia PTX Assembler to compile it
  *  to `.cubin` for some target architecture:
  * 
- *  $ ptxas -o less_slow_from_ptx.cubin -arch=sm_80 less_slow_sm80.ptx
- *  $ cuobjdump -sass less_slow_from_ptx.cubin | grep -i mma
+ *  $ ptxas -o less_slow_sm80_from_ptx.cubin -arch=sm_80 less_slow_sm80.ptx
+ *  $ cuobjdump -sass less_slow_sm80_from_ptx.cubin | grep -i mma
  *
  *  Assuming how aggressively NVCC unrolls loops and the number of kernels in
  *  this file, you may want to deduplicate them:
  *
- *  $ cuobjdump -sass less_slow_from_ptx.cubin | grep -i mma | \
+ *  $ cuobjdump -sass less_slow_sm80_from_ptx.cubin | grep -i mma | \
  *  $   sed -r 's/\/\*[^*]+\*\///g' | \
  *  $   sed -r 's/^[[:space:]]+//; s/[[:space:]]+$//' | \
  *  $   sort -u