@@ -86,30 +86,62 @@ <h2>Timetable</h2>
8686 < tr >
8787 < td > Mon 1/5</ td >
8888 < td > Both</ td >
89- < td > Course organization</ td >
90- < td > Introductory syllabus and opening slides</ td >
89+ < td >
90+ < ul class ="cell-list ">
91+ < li > Course organization</ li >
92+ < li > Semester goals and project framing</ li >
93+ </ ul >
94+ </ td >
95+ < td >
96+ < ul class ="cell-list ">
97+ < li > < a href ="https://docs.google.com/presentation/d/1-KQhZabAgIDfrjPUgkTy4F6k2BqGpFiPdeS4oC_BKsw/edit?usp=sharing "> Ganesh intro slides</ a > </ li >
98+ < li > < a href ="https://cs.rochester.edu/~sree/courses/cs6969-spring-2026/sree-intro.pdf "> Sree intro slides</ a > </ li >
99+ </ ul >
100+ </ td >
91101 < td > Semester launch</ td >
92102 </ tr >
93103 < tr >
94104 < td > Wed 1/7</ td >
95105 < td > Both</ td >
96- < td > Number systems and tools; intro to performance fundamentals</ td >
97- < td > Ganesh slides; Sree performance slides</ td >
98- < td > Asg1 released, due 1/14</ td >
106+ < td >
107+ < ul class ="cell-list ">
108+ < li > Number systems and tools</ li >
109+ < li > Intro to performance fundamentals</ li >
110+ </ ul >
111+ </ td >
112+ < td >
113+ < ul class ="cell-list ">
114+ < li > < a href ="https://docs.google.com/presentation/d/1wLwiYiBSj3b4UrlhirfCdA2O1dPMqRIJh1LMdygTYuo/edit?usp=sharing "> Ganesh slides</ a > </ li >
115+ < li > < a href ="https://cs.rochester.edu/~sree/courses/cs6969-spring-2026/sree-perf-model.pdf "> Sree performance slides</ a > </ li >
116+ </ ul >
117+ </ td >
118+ < td > Asg1 released, due 1/14. See the < a href ="https://www.overleaf.com/read/pcfjvhpzghpt#9ec326 "> assignment Overleaf</ a > .</ td >
99119 </ tr >
100120 < tr >
101121 < td > Mon 1/12</ td >
102122 < td > Sree</ td >
103123 < td > Intro to GPU performance</ td >
104- < td > GPU performance lecture material</ td >
124+ < td > < a href =" https://cs.rochester.edu/~sree/courses/cs6969-spring-2026/sree-gpu-performance.pdf " > GPU performance lecture material</ a > </ td >
105125 < td > Continue Asg1</ td >
106126 </ tr >
107127 < tr >
108128 < td > Wed 1/14</ td >
109129 < td > Both + student presenters</ td >
110- < td > Formal model of GPU execution; throughput models</ td >
111- < td > Facile, uiCA, race-effect discussion, GKLEE demo</ td >
112- < td > Asg2 assigned, due 1/21</ td >
130+ < td >
131+ < ul class ="cell-list ">
132+ < li > Formal model of GPU execution</ li >
133+ < li > Throughput models</ li >
134+ < li > Race effects and GKLEE demo</ li >
135+ </ ul >
136+ </ td >
137+ < td >
138+ < ul class ="cell-list ">
139+ < li > < a href ="https://docs.google.com/presentation/d/1mijmFYeSxJ5pfZKFdObCJYOC4lZWRzosEBBJF5XHgtM/edit?usp=sharing "> Ganesh slides</ a > </ li >
140+ < li > < a href ="https://ieeexplore.ieee.org/document/10289219 "> Facile</ a > </ li >
141+ < li > < a href ="https://dl.acm.org/doi/10.1145/3524059.3532396 "> uiCA</ a > </ li >
142+ </ ul >
143+ </ td >
144+ < td > Asg2 assigned, due 1/21. Detect races using < a href ="https://cogumbreiro.github.io/assets/faial-popl26.pdf "> Faial</ a > and optionally < a href ="https://dl.acm.org/doi/pdf/10.1145/2145816.2145844 "> GKLEE</ a > .</ td >
113145 </ tr >
114146 < tr >
115147 < td > Mon 1/19</ td >
@@ -129,8 +161,8 @@ <h2>Timetable</h2>
129161 < td > Wed 1/21</ td >
130162 < td > Guest talk</ td >
131163 < td > Modular static cost analysis and related verification ideas</ td >
132- < td > Tiago Cogumbreiro / Faial material</ td >
133- < td > Asg3 assigned, due 1/28</ td >
164+ < td > < a href =" https://cogumbreiro.github.io/assets/faial-popl26.pdf " > Tiago Cogumbreiro / Faial material</ a > </ td >
165+ < td > Asg3 assigned, due 1/28. See the < a href =" https://www.overleaf.com/read/kfzrpddjsdpn#6b6082 " > Asg-3 writeup workspace </ a > . </ td >
134166 </ tr >
135167 < tr >
136168 < td > Thu 1/22</ td >
@@ -143,42 +175,68 @@ <h2>Timetable</h2>
143175 < td > Mon 1/26</ td >
144176 < td > Ganesh + David</ td >
145177 < td > AWS training; Tilus; modular scheduling</ td >
146- < td > Tilus paper and course repo experiments</ td >
178+ < td >
179+ < ul class ="cell-list ">
180+ < li > < a href ="https://docs.google.com/presentation/d/1C6vt0McbS2if2_AJoVdhWAzqrTK6owV8I71zlvUIOcE/edit?usp=sharing "> AWS training slides</ a > </ li >
181+ < li > < a href ="https://arxiv.org/pdf/2504.12984 "> Tilus paper</ a > </ li >
182+ < li > < a href ="https://github.com/parfloat/parfloat-class/tree/main/TILUS "> Tilus repo experiments</ a > </ li >
183+ </ ul >
184+ </ td >
147185 < td > AWS and low-precision kernel focus</ td >
148186 </ tr >
149187 < tr >
150188 < td > Wed 1/28</ td >
151189 < td > Ganesh</ td >
152190 < td > AWS training; Neuron architecture; Mojo</ td >
153- < td > Neuron slides and AWS instructions</ td >
154- < td > Asg4 due 2/6; Asg5 due 2/13</ td >
191+ < td >
192+ < ul class ="cell-list ">
193+ < li > < a href ="https://docs.google.com/presentation/d/1OVkwX4hO3V6tqrvNE9VwxiHOSabSzhxauVZp6ombeqU/edit?usp=sharing "> Neuron architecture slides</ a > </ li >
194+ < li > < a href ="https://www.overleaf.com/read/wxrsxmdttcgw#da21ab "> AWS writeup workspace</ a > </ li >
195+ </ ul >
196+ </ td >
197+ < td > Asg4 due 2/6; Asg5 due 2/13.</ td >
155198 </ tr >
156199 < tr >
157200 < td > Mon 2/2</ td >
158201 < td > Ganesh + Sree + students</ td >
159202 < td > AWS tensor-addition walkthrough; profiling; student talks</ td >
160- < td > Mojo in HPC; RenderMan XPU</ td >
203+ < td >
204+ < ul class ="cell-list ">
205+ < li > < a href ="https://arxiv.org/abs/2509.21039 "> Mojo in HPC</ a > </ li >
206+ < li > < a href ="https://graphics.pixar.com/library/RenderManXPU/ "> RenderMan XPU</ a > </ li >
207+ </ ul >
208+ </ td >
161209 < td > Interactive experimental session</ td >
162210 </ tr >
163211 < tr >
164212 < td > Wed 2/4</ td >
165213 < td > Both + student speakers</ td >
166214 < td > Follow-on AWS material; student presentations</ td >
167- < td > tritonBLAS; ParallelKittens</ td >
215+ < td >
216+ < ul class ="cell-list ">
217+ < li > < a href ="https://arxiv.org/abs/2512.04226 "> tritonBLAS</ a > </ li >
218+ < li > < a href ="https://arxiv.org/pdf/2511.13940 "> ParallelKittens</ a > </ li >
219+ </ ul >
220+ </ td >
168221 < td > Project-selection writeup due 2/13</ td >
169222 </ tr >
170223 < tr >
171224 < td > Mon 2/9</ td >
172225 < td > Both + student speaker</ td >
173226 < td > Discussion of Asg1-Asg4; Hoare logic for GPU programs</ td >
174- < td > Hoare logic reading; HiRace material</ td >
227+ < td >
228+ < ul class ="cell-list ">
229+ < li > < a href ="https://dl.acm.org/doi/10.1145/3001834 "> Hoare Logic of GPU Programs</ a > </ li >
230+ < li > < a href ="https://dl.acm.org/doi/10.1109/SC41406.2024.00042 "> HiRace</ a > </ li >
231+ </ ul >
232+ </ td >
175233 < td > Homework review</ td >
176234 </ tr >
177235 < tr >
178236 < td > Wed 2/11</ td >
179237 < td > Both + student speaker</ td >
180238 < td > Memory hierarchy paper discussion</ td >
181- < td > Dissecting the NVIDIA Turing T4 GPU via Microbenchmarking</ td >
239+ < td > < a href =" https://arxiv.org/pdf/1903.07486 " > Dissecting the NVIDIA Turing T4 GPU via Microbenchmarking</ a > </ td >
182240 < td > Read before class</ td >
183241 </ tr >
184242 < tr >
@@ -192,14 +250,27 @@ <h2>Timetable</h2>
192250 < td > Wed 2/18</ td >
193251 < td > Both + student speakers</ td >
194252 < td > ThunderKittens, HipKittens, TVM-FFI discussion</ td >
195- < td > OpenReview paper; HipKittens; TVM-FFI</ td >
253+ < td >
254+ < ul class ="cell-list ">
255+ < li > < a href ="https://openreview.net/pdf?id=0fJfVOSUra "> ThunderKittens</ a > </ li >
256+ < li > < a href ="https://arxiv.org/abs/2511.08083 "> HipKittens</ a > </ li >
257+ < li > < a href ="https://tvm.apache.org/ffi/ "> TVM-FFI</ a > and < a href ="https://github.com/apache/tvm-ffi?tab=readme-ov-file "> repo</ a > </ li >
258+ </ ul >
259+ </ td >
196260 < td > Paper-discussion format</ td >
197261 </ tr >
198262 < tr >
199263 < td > Mon 2/23</ td >
200264 < td > Ganesh + student speakers</ td >
201265 < td > MLIR-AIR paper and software tryout</ td >
202- < td > MLIR paper; MLIR-AIR; AIR2CUDA; Alex Zinenko talk</ td >
266+ < td >
267+ < ul class ="cell-list ">
268+ < li > < a href ="https://dl.acm.org/doi/epdf/10.1145/3318464.3380595 "> MLIR paper</ a > </ li >
269+ < li > < a href ="https://arxiv.org/abs/2510.14871 "> MLIR-AIR paper</ a > </ li >
270+ < li > < a href ="https://github.com/parfloat/parfloat-class/tree/main/AIR2CUDA "> AIR2CUDA</ a > </ li >
271+ < li > < a href ="https://youtu.be/hkgWi0oN_L8?si=Cl_H2Se2AlYvKmqN "> Alex Zinenko talk</ a > </ li >
272+ </ ul >
273+ </ td >
203274 < td > Unit-test and software exploration</ td >
204275 </ tr >
205276 < tr >
@@ -213,14 +284,26 @@ <h2>Timetable</h2>
213284 < td > Mon 3/2</ td >
214285 < td > Ganesh + students</ td >
215286 < td > Faial race-checking and GKLEE</ td >
216- < td > Faial FMSD paper; GKLEE paper; NVBit basics</ td >
287+ < td >
288+ < ul class ="cell-list ">
289+ < li > < a href ="https://cogumbreiro.github.io/assets/faial-fmsd23.pdf "> Faial FMSD paper</ a > </ li >
290+ < li > < a href ="https://dl.acm.org/doi/pdf/10.1145/2145816.2145844 "> GKLEE paper</ a > </ li >
291+ < li > < a href ="https://dl.acm.org/doi/10.1145/3352460.3358307 "> NVBit paper</ a > </ li >
292+ </ ul >
293+ </ td >
217294 < td > Project-idea discussion</ td >
218295 </ tr >
219296 < tr >
220297 < td > Wed 3/4</ td >
221298 < td > Students</ td >
222299 < td > Brief project idea presentations</ td >
223- < td > Correctness tooling slide deck; Goldberg FP reading</ td >
300+ < td >
301+ < ul class ="cell-list ">
302+ < li > < a href ="https://docs.google.com/presentation/d/13Wm5mQldICWj227BsHivin1ZhueXy4UJKfQorlbQ4KY/edit?usp=sharing "> Correctness tooling slide deck</ a > </ li >
303+ < li > < a href ="https://dl.acm.org/doi/10.1145/103162.103163 "> Goldberg FP reading</ a > </ li >
304+ < li > < a href ="https://dl.acm.org/doi/pdf/10.1145/3736112.3736144 "> Additional reading</ a > </ li >
305+ </ ul >
306+ </ td >
224307 < td > Project pitching</ td >
225308 </ tr >
226309 < tr >
@@ -255,7 +338,12 @@ <h2>Timetable</h2>
255338 < td > Mon 3/23</ td >
256339 < td > Guest / project mode</ td >
257340 < td > SLEEK paper and code discussion</ td >
258- < td > SLEEK IPDPS 2026 paper and repository</ td >
341+ < td >
342+ < ul class ="cell-list ">
343+ < li > < a href ="https://userweb.cs.txstate.edu/~burtscher/papers/ipdps26.pdf "> SLEEK paper</ a > </ li >
344+ < li > < a href ="https://github.com/burtscher/SLEEK/ "> SLEEK repository</ a > </ li >
345+ </ ul >
346+ </ td >
259347 < td > Andrew Rodriguez presentation</ td >
260348 </ tr >
261349 < tr >
@@ -276,7 +364,7 @@ <h2>Timetable</h2>
276364 < td > Wed 4/1</ td >
277365 < td > Discussion</ td >
278366 < td > cuFuzz discussion</ td >
279- < td > NVIDIA research link and slides </ td >
367+ < td > < a href =" https:// research.nvidia.com/publication/2026-03_hunting-cuda-bugs-scale-cufuzz " > NVIDIA cuFuzz research page </ a > </ td >
280368 < td > Tooling / bug-finding focus</ td >
281369 </ tr >
282370 < tr >
0 commit comments