Skip to content

fix(gpu): guard qq synchronization without projectors #7816

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
GenZeal-Lin wants to merge 3 commits into
base: develop
Choose a base branch
from
+32 −23
Open

fix(gpu): guard qq synchronization without projectors #7816

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
a6bf6c6
fix(gpu): guard qq synchronization without projectors
Aug 14, 2026
4ba66ae
Merge branch 'develop' into fix/7681-cuda-fft-grid
GenZeal-Lin Aug 14, 2026
d1b6840
test(gpu): re-enable cases fixed by #7681
Aug 14, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 15 additions & 6 deletions source/source_pw/module_pwdft/vnl_pw.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -768,10 +768,13 @@ void pseudopot_cell_vnl::init_vnl(UnitCell& cell, const ModulePW::PW_Basis* rho_
castmem_d2s_h2d_op()(this->s_nhtol, this->nhtol.c, this->nhtol.nr * this->nhtol.nc);
castmem_d2s_h2d_op()(this->s_nhtolm, this->nhtolm.c, this->nhtolm.nr * this->nhtolm.nc);
castmem_d2s_h2d_op()(this->s_tab, this->tab.ptr, this->tab.getSize());
castmem_d2s_h2d_op()(this->s_qq_nt, this->qq_nt.ptr, this->qq_nt.getSize());
castmem_z2c_h2d_op()(this->c_qq_so, this->qq_so.ptr, this->qq_so.getSize());
if (this->nhm > 0)
{
castmem_d2s_h2d_op()(this->s_qq_nt, this->qq_nt.ptr, this->qq_nt.getSize());
castmem_z2c_h2d_op()(this->c_qq_so, this->qq_so.ptr, this->qq_so.getSize());
}
}
if (PARAM.globalv.has_double_data)
if (PARAM.globalv.has_double_data && this->nhm > 0)
{
syncmem_z2z_h2d_op()(this->z_qq_so, this->qq_so.ptr, this->qq_so.getSize());
}
Expand All @@ -782,7 +785,10 @@ void pseudopot_cell_vnl::init_vnl(UnitCell& cell, const ModulePW::PW_Basis* rho_
syncmem_d2d_h2d_op()(this->d_nhtol, this->nhtol.c, this->nhtol.nr * this->nhtol.nc);
syncmem_d2d_h2d_op()(this->d_nhtolm, this->nhtolm.c, this->nhtolm.nr * this->nhtolm.nc);
syncmem_d2d_h2d_op()(this->d_tab, this->tab.ptr, this->tab.getSize());
syncmem_d2d_h2d_op()(this->d_qq_nt, this->qq_nt.ptr, this->qq_nt.getSize());
if (this->nhm > 0)
{
syncmem_d2d_h2d_op()(this->d_qq_nt, this->qq_nt.ptr, this->qq_nt.getSize());
}
}
else
{
Expand All @@ -792,8 +798,11 @@ void pseudopot_cell_vnl::init_vnl(UnitCell& cell, const ModulePW::PW_Basis* rho_
castmem_d2s_h2h_op()(this->s_nhtol, this->nhtol.c, this->nhtol.nr * this->nhtol.nc);
castmem_d2s_h2h_op()(this->s_nhtolm, this->nhtolm.c, this->nhtolm.nr * this->nhtolm.nc);
castmem_d2s_h2h_op()(this->s_tab, this->tab.ptr, this->tab.getSize());
castmem_d2s_h2h_op()(this->s_qq_nt, this->qq_nt.ptr, this->qq_nt.getSize());
castmem_z2c_h2h_op()(this->c_qq_so, this->qq_so.ptr, this->qq_so.getSize());
if (this->nhm > 0)
{
castmem_d2s_h2h_op()(this->s_qq_nt, this->qq_nt.ptr, this->qq_nt.getSize());
castmem_z2c_h2h_op()(this->c_qq_so, this->qq_so.ptr, this->qq_so.getSize());
}
}
// There's no need to synchronize double precision pointers while in a CPU environment.
}
Expand Down
34 changes: 17 additions & 17 deletions tests/01_PW/CASES_GPU.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,10 +17,10 @@ scf_out_elf
013_PW_ONCV_LDA
014_PW_UPF201_BLPS
015_PW_GTH
#016_PW_BLPS
#017_PW_LPS6
#018_PW_LPS8
#019_PW_Coulomb
016_PW_BLPS
017_PW_LPS6
018_PW_LPS8
019_PW_Coulomb
020_PW_kspace
021_PW_kspace3
022_PW_CG
Expand Down Expand Up @@ -113,17 +113,17 @@ scf_out_elf
#208_PW_CG_float
209_PW_DFTHALF
210_PW_kspace_shift
#801_PW_LT_sc
#802_PW_LT_fcc
801_PW_LT_sc
802_PW_LT_fcc
#803_PW_LT_bcc
#804_PW_LT_hex
#805_PW_LT_trigonal
#806_PW_LT_st
#807_PW_LT_bct
#808_PW_LT_so
#809_PW_LT_baco
#810_PW_LT_fco
#811_PW_LT_bco
#812_PW_LT_sm
#813_PW_LT_bacm
#814_PW_LT_triclinic
804_PW_LT_hex
805_PW_LT_trigonal
806_PW_LT_st
807_PW_LT_bct
808_PW_LT_so
809_PW_LT_baco
810_PW_LT_fco
811_PW_LT_bco
812_PW_LT_sm
813_PW_LT_bacm
814_PW_LT_triclinic
Loading