Daswc

[hatal175]

No description provided.

[hatal175]

Just take a look at any rel. I think you'll see most don't have an rodata section and in most you'll see mismatches in sdata and sdata2 locations.

…

On Thu, Apr 17, 2025, 01:34 TakaRikka ***@***.***> wrote: ***@***.**** commented on this pull request. ------------------------------ In configure.py <#2398 (comment)>: > ] +# I think that debug version doesn't pass these flags but the build fails on a relocation whats the reasoning for this thought? it seems unlikely based on what we know about rels for this to be true, but maybe theres a similar thing that needs to be adjusted for correct behavior — Reply to this email directly, view it on GitHub <#2398 (review)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACYQXXJIEJ2TMTRLX6N2TQ32Z3LHXAVCNFSM6AAAAAB3JFGHVSVHI2DSMVQWIX3LMV43YUDVNRWFEZLROVSXG5CSMV3GSZLXHMZDONZUGAYDKNZSG4> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[hatal175]

Of course, my sampling might be wrong, so a second look is welcome.

…

On Thu, Apr 17, 2025, 05:48 Tal Hayon ***@***.***> wrote: Just take a look at any rel. I think you'll see most don't have an rodata section and in most you'll see mismatches in sdata and sdata2 locations. On Thu, Apr 17, 2025, 01:34 TakaRikka ***@***.***> wrote: > ***@***.**** commented on this pull request. > ------------------------------ > > In configure.py > <#2398 (comment)>: > > > ] > > +# I think that debug version doesn't pass these flags but the build fails on a relocation > > whats the reasoning for this thought? it seems unlikely based on what we > know about rels for this to be true, but maybe theres a similar thing that > needs to be adjusted for correct behavior > > — > Reply to this email directly, view it on GitHub > <#2398 (review)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/ACYQXXJIEJ2TMTRLX6N2TQ32Z3LHXAVCNFSM6AAAAAB3JFGHVSVHI2DSMVQWIX3LMV43YUDVNRWFEZLROVSXG5CSMV3GSZLXHMZDONZUGAYDKNZSG4> > . > You are receiving this because you authored the thread.Message ID: > ***@***.***> >

[TakaRikka]

i think the problem here is that in the shield version, some of the rels were instead just made into dol TUs, like d_a_grass and d_a_swc, etc. meaning d_a_swc would be compiled with dol flags for shield, but rel flags for GC

[hatal175]

Hmm. I'll adjust the pool request.

…

On Thu, Apr 17, 2025, 06:27 TakaRikka ***@***.***> wrote: i think the problem here is that in the shield version, some of the rels were instead just made into dol TUs, like d_a_grass and d_a_swc, etc. meaning d_a_swc would be compiled with dol flags for shield, but rel flags for GC — Reply to this email directly, view it on GitHub <#2398 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACYQXXOHSLJIRQIO5ZVO4ID2Z4NTVAVCNFSM6AAAAAB3JFGHVSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQMJRGY2DEMRUGQ> . You are receiving this because you authored the thread.Message ID: ***@***.***> *TakaRikka* left a comment (zeldaret/tp#2398) <#2398 (comment)> i think the problem here is that in the shield version, some of the rels were instead just made into dol TUs, like d_a_grass and d_a_swc, etc. meaning d_a_swc would be compiled with dol flags for shield, but rel flags for GC — Reply to this email directly, view it on GitHub <#2398 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACYQXXOHSLJIRQIO5ZVO4ID2Z4NTVAVCNFSM6AAAAAB3JFGHVSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQMJRGY2DEMRUGQ> . You are receiving this because you authored the thread.Message ID: ***@***.***>