[GitHub] [tvm] Yuan-Chuan-YUE commented on pull request #8756: [CODEGEN][OpenCL]: fix tir.erf codegen to opencl directly

Yuan-Chuan-YUE commented on pull request #8756: URL: https://github.com/apache/tvm/pull/8756#issuecomment-903217680 @junrushao1994 Thank you! @csullivan Thanks a lot for helping reviewing the code! -- This is an automated message from the Apache Git Service. To respond to the

[tvm] branch main updated (fc9f582 -> d6d6367)

This is an automated email from the ASF dual-hosted git repository. junrushao pushed a change to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git. from fc9f582 [TIR] Support fold constants in specialize process (#8803) add d6d6367 Fix typos (#8787) No new

[GitHub] [tvm] junrushao1994 commented on pull request #8787: Fix typos

junrushao1994 commented on pull request #8787: URL: https://github.com/apache/tvm/pull/8787#issuecomment-903214961 Thanks @gromero @jroesch @tkonolige! It is merged -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the

[GitHub] [tvm] junrushao1994 merged pull request #8787: Fix typos

junrushao1994 merged pull request #8787: URL: https://github.com/apache/tvm/pull/8787 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail:

[GitHub] [tvm] sunjiweiswift commented on pull request #8497: [Runtime] add set_output_zero_copy

sunjiweiswift commented on pull request #8497: URL: https://github.com/apache/tvm/pull/8497#issuecomment-903200835 @jcf94 Please merge this pull request -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above

[GitHub] [tvm-vta] aasorokiin commented on pull request #32: VTA Chisel Wide memory interface.

aasorokiin commented on pull request #32: URL: https://github.com/apache/tvm-vta/pull/32#issuecomment-903198916 Updated this request to comply to Chisel changes from #33 Cannot quite get why the latest main with #33 doesn't compile lib for me (make in chisel directory). It complains:

[GitHub] [tvm-rfcs] junrushao1994 commented on pull request #23: [RFC][TIR] TIR Pinned Memory Representation

junrushao1994 commented on pull request #23: URL: https://github.com/apache/tvm-rfcs/pull/23#issuecomment-903193506 Thanks @manupa-arm for the nice proposal! Currently the memory scope of a buffer not only indicates where the buffer is allocated, but also affects schedule

[GitHub] [tvm] junrushao1994 edited a comment on pull request #8767: [TensorIR][M2a] Reorder

junrushao1994 edited a comment on pull request #8767: URL: https://github.com/apache/tvm/pull/8767#issuecomment-903177719 Thanks @jinhongyii! The PR mostly looks good to me now in terms of the algorithm side. I prefer to consolidate the logic into several semantically meaningful

[GitHub] [tvm] junrushao1994 commented on a change in pull request #8767: [TensorIR][M2a] Reorder

junrushao1994 commented on a change in pull request #8767: URL: https://github.com/apache/tvm/pull/8767#discussion_r693412063 ## File path: src/tir/schedule/primitive/loop_transformation.cc ## @@ -385,6 +511,113 @@ StmtSRef Fuse(ScheduleState self, const Array& loop_srefs) {

[GitHub] [tvm] junrushao1994 commented on pull request #8803: [TIR] Support fold constants in specialize process

junrushao1994 commented on pull request #8803: URL: https://github.com/apache/tvm/pull/8803#issuecomment-903182254 Thanks @wrongtest for the fix! @Hzfengsy would you mind discussing about the fix for Simplify some time? -- This is an automated message from the Apache Git Service.

[tvm] branch main updated: [TIR] Support fold constants in specialize process (#8803)

This is an automated email from the ASF dual-hosted git repository. junrushao pushed a commit to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git The following commit(s) were added to refs/heads/main by this push: new fc9f582 [TIR] Support fold constants in specialize

[GitHub] [tvm] junrushao1994 merged pull request #8803: [TIR] Support fold constants in specialize process

junrushao1994 merged pull request #8803: URL: https://github.com/apache/tvm/pull/8803 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail:

[GitHub] [tvm] junrushao1994 commented on pull request #8803: [TIR] Support fold constants in specialize process

junrushao1994 commented on pull request #8803: URL: https://github.com/apache/tvm/pull/8803#issuecomment-903182160 @tqchen Right. It is something we need to fix! _taking notes_ -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub

[GitHub] [tvm] junrushao1994 edited a comment on pull request #8802: Add LowerTEPass, and convert calls to LowerTE to application of LowerTEPass

junrushao1994 edited a comment on pull request #8802: URL: https://github.com/apache/tvm/pull/8802#issuecomment-903181904 @Mousius thanks for diving into the problem! Yeah, ideally we should avoid converting `Target` object to strings (i.e. avoid `target->str()`). BTW, if we want to use

[GitHub] [tvm] junrushao1994 commented on pull request #8802: Add LowerTEPass, and convert calls to LowerTE to application of LowerTEPass

junrushao1994 commented on pull request #8802: URL: https://github.com/apache/tvm/pull/8802#issuecomment-903181904 @Mousius thanks for diving into the problem! Yeah, ideally we should avoid converting `Target` object to strings. BTW, if we want to use its kind as a string (e.g. the kind

[GitHub] [tvm] junrushao1994 commented on pull request #8787: Fix typos

junrushao1994 commented on pull request #8787: URL: https://github.com/apache/tvm/pull/8787#issuecomment-903179757 I retriggered the CI and let's get it merged ASAP -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the

[tvm] branch main updated: [CODEGEN][OpenCL]: fix tir.erf codegen to opencl directly (#8756)

This is an automated email from the ASF dual-hosted git repository. junrushao pushed a commit to branch main in repository https://gitbox.apache.org/repos/asf/tvm.git The following commit(s) were added to refs/heads/main by this push: new 4b9881e [CODEGEN][OpenCL]: fix tir.erf codegen to

[GitHub] [tvm] junrushao1994 merged pull request #8756: [CODEGEN][OpenCL]: fix tir.erf codegen to opencl directly

junrushao1994 merged pull request #8756: URL: https://github.com/apache/tvm/pull/8756 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail:

[GitHub] [tvm] junrushao1994 commented on pull request #8767: [TensorIR][M2a] Reorder

junrushao1994 commented on pull request #8767: URL: https://github.com/apache/tvm/pull/8767#issuecomment-903177719 Thanks @jinhongyii! The PR mostly looks good to me now in terms of the algorithm side. I prefer to consolidate the logic into several semantically meaningful methods,

[GitHub] [tvm] junrushao1994 commented on a change in pull request #8767: [TensorIR][M2a] Reorder

junrushao1994 commented on a change in pull request #8767: URL: https://github.com/apache/tvm/pull/8767#discussion_r693402064 ## File path: src/tir/schedule/primitive/loop_transformation.cc ## @@ -385,6 +511,116 @@ StmtSRef Fuse(ScheduleState self, const Array& loop_srefs) {

[GitHub] [tvm] junrushao1994 commented on a change in pull request #8767: [TensorIR][M2a] Reorder

junrushao1994 commented on a change in pull request #8767: URL: https://github.com/apache/tvm/pull/8767#discussion_r693402002 ## File path: src/tir/schedule/primitive/loop_transformation.cc ## @@ -385,6 +511,116 @@ StmtSRef Fuse(ScheduleState self, const Array& loop_srefs) {

[GitHub] [tvm] junrushao1994 commented on a change in pull request #8767: [TensorIR][M2a] Reorder

junrushao1994 commented on a change in pull request #8767: URL: https://github.com/apache/tvm/pull/8767#discussion_r693401580 ## File path: src/tir/schedule/primitive/loop_transformation.cc ## @@ -385,6 +511,116 @@ StmtSRef Fuse(ScheduleState self, const Array& loop_srefs) {

[GitHub] [tvm] junrushao1994 commented on a change in pull request #8767: [TensorIR][M2a] Reorder

junrushao1994 commented on a change in pull request #8767: URL: https://github.com/apache/tvm/pull/8767#discussion_r693401332 ## File path: src/tir/schedule/primitive/loop_transformation.cc ## @@ -385,6 +511,116 @@ StmtSRef Fuse(ScheduleState self, const Array& loop_srefs) {

[GitHub] [tvm] junrushao1994 commented on a change in pull request #8767: [TensorIR][M2a] Reorder

junrushao1994 commented on a change in pull request #8767: URL: https://github.com/apache/tvm/pull/8767#discussion_r693401171 ## File path: src/tir/schedule/primitive/loop_transformation.cc ## @@ -385,6 +511,116 @@ StmtSRef Fuse(ScheduleState self, const Array& loop_srefs) {

[GitHub] [tvm] junrushao1994 commented on a change in pull request #8767: [TensorIR][M2a] Reorder

junrushao1994 commented on a change in pull request #8767: URL: https://github.com/apache/tvm/pull/8767#discussion_r693400687 ## File path: src/tir/schedule/primitive/loop_transformation.cc ## @@ -385,6 +511,116 @@ StmtSRef Fuse(ScheduleState self, const Array& loop_srefs) {

[GitHub] [tvm] junrushao1994 commented on a change in pull request #8767: [TensorIR][M2a] Reorder

junrushao1994 commented on a change in pull request #8767: URL: https://github.com/apache/tvm/pull/8767#discussion_r693400463 ## File path: src/tir/schedule/primitive/loop_transformation.cc ## @@ -385,6 +511,116 @@ StmtSRef Fuse(ScheduleState self, const Array& loop_srefs) {

[GitHub] [tvm] junrushao1994 commented on a change in pull request #8767: [TensorIR][M2a] Reorder

junrushao1994 commented on a change in pull request #8767: URL: https://github.com/apache/tvm/pull/8767#discussion_r693400173 ## File path: src/tir/schedule/primitive/loop_transformation.cc ## @@ -385,6 +511,113 @@ StmtSRef Fuse(ScheduleState self, const Array& loop_srefs) {

[GitHub] [tvm] junrushao1994 commented on a change in pull request #8767: [TensorIR][M2a] Reorder

junrushao1994 commented on a change in pull request #8767: URL: https://github.com/apache/tvm/pull/8767#discussion_r693399727 ## File path: src/tir/schedule/primitive/loop_transformation.cc ## @@ -385,6 +511,116 @@ StmtSRef Fuse(ScheduleState self, const Array& loop_srefs) {

[GitHub] [tvm] junrushao1994 commented on a change in pull request #8767: [TensorIR][M2a] Reorder

junrushao1994 commented on a change in pull request #8767: URL: https://github.com/apache/tvm/pull/8767#discussion_r693399446 ## File path: src/tir/schedule/primitive/loop_transformation.cc ## @@ -385,6 +511,116 @@ StmtSRef Fuse(ScheduleState self, const Array& loop_srefs) {

[GitHub] [tvm] junrushao1994 commented on a change in pull request #8767: [TensorIR][M2a] Reorder

junrushao1994 commented on a change in pull request #8767: URL: https://github.com/apache/tvm/pull/8767#discussion_r693399202 ## File path: src/tir/schedule/primitive/loop_transformation.cc ## @@ -385,6 +511,116 @@ StmtSRef Fuse(ScheduleState self, const Array& loop_srefs) {

[GitHub] [tvm] junrushao1994 commented on a change in pull request #8767: [TensorIR][M2a] Reorder

junrushao1994 commented on a change in pull request #8767: URL: https://github.com/apache/tvm/pull/8767#discussion_r693398936 ## File path: src/tir/schedule/primitive/loop_transformation.cc ## @@ -385,6 +511,113 @@ StmtSRef Fuse(ScheduleState self, const Array& loop_srefs) {

[GitHub] [tvm] chiwwang commented on pull request #8448: [Draft PR] Relay IR visualizer.

chiwwang commented on pull request #8448: URL: https://github.com/apache/tvm/pull/8448#issuecomment-903158848 A quick status update. Following functionalities are done: 1. Show node-information with the zoom-in/out level of the plot. 2. Interactive legend, clicking the legend item to

[GitHub] [tvm] chiwwang commented on pull request #8668: [Draft PR] Visualization Relay IR on terminal

chiwwang commented on pull request #8668: URL: https://github.com/apache/tvm/pull/8668#issuecomment-903155532 Hi @kueitang Maybe we can re-use this RFC, https://discuss.tvm.apache.org/t/rfc-visualizing-relay-program-as-graph/4825 There are some valuable thoughts there :) Back

[GitHub] [tvm] jinhongyii commented on a change in pull request #8767: [TensorIR][M2a] Reorder

jinhongyii commented on a change in pull request #8767: URL: https://github.com/apache/tvm/pull/8767#discussion_r693361836 ## File path: src/tir/schedule/primitive/loop_transformation.cc ## @@ -385,6 +511,113 @@ StmtSRef Fuse(ScheduleState self, const Array& loop_srefs) {

[GitHub] [tvm-rfcs] tqchen commented on pull request #23: [RFC][TIR] TIR Pinned Memory Representation

tqchen commented on pull request #23: URL: https://github.com/apache/tvm-rfcs/pull/23#issuecomment-903112642 cc @masahi @kparzysz-quic @csullivan -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to

[GitHub] [tvm-rfcs] tqchen commented on pull request #23: [RFC][TIR] TIR Pinned Memory Representation

tqchen commented on pull request #23: URL: https://github.com/apache/tvm-rfcs/pull/23#issuecomment-903112540 Thank you @manupa-arm for the proposal. What you described is closest to the "tag" field of the storage scope, which we put as part of the pointer and associate with most of

[GitHub] [tvm] Mousius commented on pull request #8802: Add LowerTEPass, and convert calls to LowerTE to application of LowerTEPass

Mousius commented on pull request #8802: URL: https://github.com/apache/tvm/pull/8802#issuecomment-903111720 Hi @electriclilies, I just took a look at your branch using `Target` and I think the runtime issues stem from `src/relay/backend/build_module.cc`: ``` Map GetIRModule()