diff options
| author | John Wickerson <j.wickerson@imperial.ac.uk> | 2020-09-14 21:28:23 +0000 |
|---|---|---|
| committer | overleaf <overleaf@localhost> | 2020-09-14 21:28:29 +0000 |
| commit | ad13398545960436523c1b739a84bc5c60ff7d5c (patch) | |
| tree | 8e5c848cc538ff3df10d40555cad8a41f2b35020 | |
| parent | 9592cda0cd4aef71be15f2aca001a3f12c9a83e1 (diff) | |
| download | fccm21_esrhls-ad13398545960436523c1b739a84bc5c60ff7d5c.tar.gz fccm21_esrhls-ad13398545960436523c1b739a84bc5c60ff7d5c.zip | |
Update on Overleaf.
| -rw-r--r-- | method-new.tex | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/method-new.tex b/method-new.tex index 934969f..c985d59 100644 --- a/method-new.tex +++ b/method-new.tex @@ -164,7 +164,7 @@ Having generated HLS-friendly programs and automatically augmented them with dir For each HLS tool in turn, we compile the C program to RTL and then simulate the RTL. Independently, we also compile the C program using GCC and execute it. Additionally, to ensure that our testing is scalable for a large number of large, random programs, we also enforce several time-outs: we set a 5-minute time-out for C execution and a 2-hour time-out for RTL simulation. -Whenever a time-out is reached +We do not count time-outs as bugs, but we record them. \paragraph{Types of bugs} There two types of bugs that we can encounter in this testing setup: programs that cause the HLS tool to crash during compilation (e.g. an unhandled assertion violation or a segmentation fault), and programs where the software execution and the RTL simulation do not return the same value. @@ -175,7 +175,7 @@ Programs that cause either type of error are given to the reduction stage, which We were only able to test one version of LegUp HLS (version 4.0). LegUp 7.5 is GUI-based and not suitable for scripting. %For all tools, we set a 5-minute and 2-hour for C simulation and RTL simulation respectively. -In cases where simulation takes longer, we do not consider them as failures or crashes and we record them. +%In cases where simulation takes longer, we do not consider them as failures or crashes and we record them. % When testing Intel HLS, we place three time-outs since its execution is generally slower than other HLS tools: a \ref{XX}-minute time-out for C compilation, a \ref{XX}-hour time out for HLS compilation and a \ref{XX}-hour time-out for co-simulation. % And the number of timeouts placed has increased to 4. The first timeout sets when compiling the C++ program to CPU and returning an executable once finished. The second timeout is placed when running the executable to get the C++ result. The third timeout, which been given the most extended period, is at synthesizing the design and generating the co-simulation executable. Finally, running the co-simulation executable requires the fourth timeout. The test case can be dumped at any timeout period if the task is not finished within the limited time. |