Skip to content

Commit

Permalink
Merge pull request #5505 from BOINC/dpa_apple_gpu
Browse files Browse the repository at this point in the history 
client/server/web: add 'Apple GPU' as an official GPU type.
  • Loading branch information
@AenBleidd
AenBleidd authored Feb 7, 2024
2 parents ff9954e + bb5a113 commit 3c7e18e
Show file tree
Hide file tree
Showing 32 changed files with 621 additions and 148 deletions.
7 changes: 7 additions & 0 deletions client/client_state.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -584,6 +584,13 @@ int CLIENT_STATE::init() {
coprocs.add(coprocs.intel_gpu);
}
}
if (coprocs.have_apple_gpu()) {
if (rsc_index(GPU_TYPE_APPLE)>0) {
msg_printf(NULL, MSG_INFO, "APPLE GPU info taken from cc_config.xml");
} else {
coprocs.add(coprocs.apple_gpu);
}
}
coprocs.add_other_coproc_types();

host_info.coprocs = coprocs;
Expand Down
196 changes: 165 additions & 31 deletions client/gpu_detect.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -15,38 +15,71 @@
// You should have received a copy of the GNU Lesser General Public License
// along with BOINC. If not, see <http://www.gnu.org/licenses/>.


// client-specific GPU code. Mostly GPU detection
// Detect the host's GPUs, and populate the global 'coprocs' object
//
// theory of operation:
// there are two ways of detecting GPUs:
// Theory of operation:
// There are two ways of detecting GPUs:
// - vendor-specific libraries like CUDA and CAL,
// which detect only that vendor's GPUs
// - OpenCL, which can detect multiple types of GPUs,
// including nvidia/amd/intel as well was new types
// such as ARM integrated GPUs
// - OpenCL, which can detect multiple types of GPUs:
// nvidia, amd, intel, apple and new types
//
// These libraries sometimes crash,
// and we've been unable to trap these via signal and exception handlers.
// So we do GPU detection in a separate process (boinc --detect_gpus)
// We call these libraries in a separate process (boinc --detect_gpus)
// for two reasons:
// 1) These libraries sometimes crash,
// and we've been unable to trap these via signal and exception handlers.
// 2) Some dual-GPU laptops (e.g., Macbook Pro) don't power down
// the more powerful GPU until all applications which used them exit.
// Doing GPU detection in a second, short-lived process
// saves battery life on these laptops.

// This process writes an XML file "coproc_info.xml" containing
// - lists of GPU detected via CUDA and CAL
// - lists of nvidia/amd/intel GPUs detected via OpenCL
// - lists of GPUs detected via vendor-specific APIs
// e.g. <coproc_cuda>
// - lists of nvidia/amd/intel/apple GPUs detected via OpenCL
// e.g. <nvidia_opencl>
// - a list of other GPUs detected via OpenCL
// e.g. <other_opencl>
//
// Data structures (lib/coproc.h):
// OPENCL_DEVICE_PROP: OpenCL info for a GPU
// COPROC: base class for a GPU, including OpenCL info
// COPROC_ATI etc.: derived classes including vendor-specific info
// COPROCS: a vector of COPROC objects,
// together with an instance of each vendor-specific type
//
// Also, some dual-GPU laptops (e.g., Macbook Pro) don't power
// down the more powerful GPU until all applications which used them exit.
// Doing GPU detection in a second, short-lived process
// saves battery life on these laptops.
// BOINC assumes that all GPUs of a given vendor are equivalent.
// We make this true by identifying the 'most capable' GPU of each type,
// and ignoring GPUs of that type that are not equivalent to it
// (in terms of memory size and capabilities).
// In the final COPROCS object, each COPROC has a 'count' field
// and a list of IDs (native, OpenCL, PCI) for the qualifying GPU instances.
//
// When the process finishes, the client parses the info file.
// Then for each vendor it "correlates" the GPUs, which includes:
// - matching up the OpenCL and vendor-specific descriptions, if both exist
// - finding the most capable GPU, and seeing which other GPUs
// are similar to it in hardware and RAM.
// Other GPUs are not used.
// - copy these to the COPROCS structure
// The client parses coproc_info.xml into a bunch of vectors:
// ati_gpus, etc: COPROC_* objects
// api_opencls, etc: OPENCL_DEVICE_PROPs for ATI GPUs
// other_opencls: OPENCL_DEVICE_PROPs for other GPUs
//
// Then (COPROCS::correlate_gpus):
// for each vendor (e.g. COPROC_ATI::correlate()):
// scan ati_gpus and find the most capable GPU.
// Copy its object to COPROCS.ati_gpu.
// Scan the list again, identifying equivalent instances,
// incrementing count, getting IDs
// Then (COPROCS::correlate_opencl())
// For each vendor (e.g. ATI)
// if we detected a GPU with CAL
// merge_opencl(): copy OpenCL info to COPROCS.ati_gpu
// else
// find_best_opencls(): find best OpenCL instance,
// populate COPROCS.ati_gpu
//
//
// Finally (CLIENT_STATE::init()):
// For each vendor (e.g. ATI)
// if COPROC_ATI is present (count>0)
// append it to the COPROCS.coprocs vector
// Append COPROCs for other OpenCL devices found

// GPUs can also be explicitly described in cc_config.xml

Expand Down Expand Up @@ -87,14 +120,21 @@ void segv_handler(int) {
}
#endif

// the following store GPU instances during initialization
// the following vectors store the low-level info from coproc_info.xml
// - For integrated GPUs (Intel, Apple)
// there is at most one instance on current computers.
// But who knows, this might change.
// - Intel GPUs can currently only be used via OpenCL.
// But this too could change.
//
vector<COPROC_ATI> ati_gpus;
vector<COPROC_NVIDIA> nvidia_gpus;
vector<COPROC_INTEL> intel_gpus;
vector<COPROC_APPLE> apple_gpus;
vector<OPENCL_DEVICE_PROP> ati_opencls;
vector<OPENCL_DEVICE_PROP> nvidia_opencls;
vector<OPENCL_DEVICE_PROP> intel_gpu_opencls;
vector<OPENCL_DEVICE_PROP> apple_gpu_opencls;
vector<OPENCL_DEVICE_PROP> other_opencls;
vector<OPENCL_CPU_PROP> cpu_opencls;

Expand Down Expand Up @@ -140,8 +180,7 @@ void COPROCS::get(
}

// populate the global vectors
// ati_gpus, nvidia_gpus, intel_gpus,
// nvidia_opencls, etc.
// ati_gpus, nvidia_gpus, intel_gpus, nvidia_opencls, etc.
//
void COPROCS::detect_gpus(vector<string> &warnings) {
#ifdef _WIN32
Expand Down Expand Up @@ -172,7 +211,8 @@ void COPROCS::detect_gpus(vector<string> &warnings) {
catch (...) {
gpu_warning(warnings, "Caught SIGSEGV in OpenCL detection");
}
#else
#else // non-Windows

void (*old_sig)(int) = signal(SIGSEGV, segv_handler);
if (setjmp(resume)) {
gpu_warning(warnings, "Caught SIGSEGV in NVIDIA GPU detection");
Expand All @@ -187,6 +227,8 @@ void COPROCS::detect_gpus(vector<string> &warnings) {
} else {
ati.get(warnings);
}
#else
apple_gpu.get(warnings);
#endif
if (setjmp(resume)) {
gpu_warning(warnings, "Caught SIGSEGV in INTEL GPU detection");
Expand All @@ -208,6 +250,7 @@ void COPROCS::detect_gpus(vector<string> &warnings) {
// Find the most capable one, and the ones equivalent to it.
// Also correlate OpenCL GPUs with CUDA/CAL GPUs.
// Then create a single COPROC (with appropriate count)
// Also return a list of strings describing the GPUs.
//
void COPROCS::correlate_gpus(
bool use_all,
Expand All @@ -222,14 +265,17 @@ void COPROCS::correlate_gpus(
#endif
ati.correlate(use_all, ignore_gpu_instance[PROC_TYPE_AMD_GPU]);
intel_gpu.correlate(use_all, ignore_gpu_instance[PROC_TYPE_INTEL_GPU]);
#ifdef __APPLE__
apple_gpu.correlate(use_all, ignore_gpu_instance[PROC_TYPE_APPLE_GPU]);
#endif
correlate_opencl(use_all, ignore_gpu_instance);

// NOTE: OpenCL can report a max of only 4GB.
//
for (i=0; i<cpu_opencls.size(); i++) {
gstate.host_info.opencl_cpu_prop[gstate.host_info.num_opencl_cpu_platforms++] = cpu_opencls[i];
}

// make list of GPU descriptions

for (i=0; i<nvidia_gpus.size(); i++) {
// This is really CUDA description
nvidia_gpus[i].description(buf, sizeof(buf));
Expand Down Expand Up @@ -323,14 +369,23 @@ void COPROCS::correlate_gpus(
descs.push_back(string(buf));
}

// Create descriptions for OpenCL Apple GPUs
//
for (i=0; i<apple_gpu_opencls.size(); i++) {
apple_gpu_opencls[i].description(buf, sizeof(buf), proc_type_name(PROC_TYPE_APPLE_GPU));
descs.push_back(string(buf));
}

// Create descriptions for other OpenCL GPUs
//
int max_other_coprocs = MAX_RSC-1; // coprocs[0] is reserved for CPU
if (have_nvidia()) max_other_coprocs--;
if (have_ati()) max_other_coprocs--;
if (have_intel_gpu()) max_other_coprocs--;
if (have_apple_gpu()) max_other_coprocs--;

// TODO: Should we implement cc_config ignore vectors for other (future) OpenCL coprocessors?
// TODO: Should we implement cc_config ignore vectors
// for other (future) OpenCL coprocessors?

for (i=0; i<other_opencls.size(); i++) {
if ((int)i > max_other_coprocs) {
Expand All @@ -350,14 +405,16 @@ void COPROCS::correlate_gpus(
ati_gpus.clear();
nvidia_gpus.clear();
intel_gpus.clear();
apple_gpus.clear();
ati_opencls.clear();
nvidia_opencls.clear();
intel_gpu_opencls.clear();
apple_gpu_opencls.clear();
cpu_opencls.clear();
}

// This is called from CLIENT_STATE::init()
// after adding NVIDIA, ATI and Intel GPUs
// after adding NVIDIA/ATI/Intel/Apple GPUs
// If we don't care about the order of GPUs in COPROCS::coprocs[],
// this code could be included at the end of COPROCS::correlate_gpus().
//
Expand Down Expand Up @@ -433,6 +490,9 @@ int COPROCS::write_coproc_info_file(vector<string> &warnings) {
for (i=0; i<intel_gpus.size(); ++i) {
intel_gpus[i].write_xml(mf, false);
}
for (i=0; i<apple_gpus.size(); ++i) {
apple_gpus[i].write_xml(mf, false);
}
for (i=0; i<ati_opencls.size(); ++i) {
ati_opencls[i].write_xml(mf, "ati_opencl", true);
}
Expand All @@ -442,6 +502,9 @@ int COPROCS::write_coproc_info_file(vector<string> &warnings) {
for (i=0; i<intel_gpu_opencls.size(); ++i) {
intel_gpu_opencls[i].write_xml(mf, "intel_gpu_opencl", true);
}
for (i=0; i<apple_gpu_opencls.size(); ++i) {
apple_gpu_opencls[i].write_xml(mf, "apple_gpu_opencl", true);
}
for (i=0; i<other_opencls.size(); i++) {
other_opencls[i].write_xml(mf, "other_opencl", true);
}
Expand Down Expand Up @@ -469,18 +532,22 @@ int COPROCS::read_coproc_info_file(vector<string> &warnings) {
COPROC_ATI my_ati_gpu;
COPROC_NVIDIA my_nvidia_gpu;
COPROC_INTEL my_intel_gpu;
COPROC_APPLE my_apple_gpu;
OPENCL_DEVICE_PROP ati_opencl;
OPENCL_DEVICE_PROP nvidia_opencl;
OPENCL_DEVICE_PROP intel_gpu_opencl;
OPENCL_DEVICE_PROP apple_gpu_opencl;
OPENCL_DEVICE_PROP other_opencl;
OPENCL_CPU_PROP cpu_opencl;

ati_gpus.clear();
nvidia_gpus.clear();
intel_gpus.clear();
apple_gpus.clear();
ati_opencls.clear();
nvidia_opencls.clear();
intel_gpu_opencls.clear();
apple_gpu_opencls.clear();
other_opencls.clear();
cpu_opencls.clear();

Expand Down Expand Up @@ -536,6 +603,16 @@ int COPROCS::read_coproc_info_file(vector<string> &warnings) {
}
continue;
}
if (xp.match_tag("coproc_apple_gpu")) {
retval = my_apple_gpu.parse(xp);
if (retval) {
my_apple_gpu.clear();
} else {
my_apple_gpu.device_num = (int)apple_gpus.size();
apple_gpus.push_back(my_apple_gpu);
}
continue;
}

if (xp.match_tag("ati_opencl")) {
ati_opencl.clear();
Expand Down Expand Up @@ -572,6 +649,17 @@ int COPROCS::read_coproc_info_file(vector<string> &warnings) {
}
continue;
}
if (xp.match_tag("apple_gpu_opencl")) {
apple_gpu_opencl.clear();
retval = apple_gpu_opencl.parse(xp, "/apple_gpu_opencl");
if (retval) {
apple_gpu_opencl.clear();
} else {
apple_gpu_opencl.is_used = COPROC_IGNORED;
apple_gpu_opencls.push_back(apple_gpu_opencl);
}
continue;
}

if (xp.match_tag("other_opencl")) {
other_opencl.clear();
Expand Down Expand Up @@ -754,3 +842,49 @@ void gpu_warning(vector<string> &warnings, const char* msg) {
fprintf(stderr, "%s\n", msg);
warnings.push_back(msg);
}

#ifdef __APPLE__
#include "mac/mac_spawn.h"
void COPROC_APPLE::get(vector<string>&) {
int retval = callPosixSpawn(
"sh -c '/usr/sbin/system_profiler SPDisplaysDataType > temp'"
);
if (retval) return;
FILE* f = fopen("temp", "r");
if (!f) return;
char buf[256], chipset_model[256];
int n, metalv;
bool have_model=false, have_ncores=false, have_metalv=false;
while (fgets(buf, sizeof(buf), f)) {
if (sscanf(buf, "%*[ ]Chipset Model: %[^\n]", chipset_model) == 1) {
have_model = true;
} else if (sscanf(buf, "%*[ ]Total Number of Cores: %d", &n) == 1) {
have_ncores = true;
} else if (sscanf(buf, "%*[ ]Metal Support: Metal %d", &metalv) == 1) {
have_metalv = true;
}
}
fclose(f);

if (!have_model || !have_ncores || !have_metalv) return;

// system_profiler reports Intel integrated GPUs on Intel Macs.
// Ignore them.
//
if (!strstr(chipset_model, "Apple")) return;

COPROC_APPLE ca;
ca.count = 1;
safe_strcpy(ca.model, chipset_model);
ca.ncores = n;
ca.metal_support = metalv;
ca.have_metal = true;
apple_gpus.push_back(ca);
}

void COPROC_APPLE::correlate(bool, vector<int> &ignore_devs) {
if (!ignore_devs.empty() && ignore_devs[0]==0) return;
if (apple_gpus.empty()) return;
*this = apple_gpus[0];
}
#endif
2 changes: 2 additions & 0 deletions client/gpu_detect.h
View file
Original file line number Diff line number Diff line change
Expand Up @@ -26,9 +26,11 @@
extern std::vector<COPROC_ATI> ati_gpus;
extern std::vector<COPROC_NVIDIA> nvidia_gpus;
extern std::vector<COPROC_INTEL> intel_gpus;
extern std::vector<COPROC_APPLE> apple_gpus;
extern std::vector<OPENCL_DEVICE_PROP> nvidia_opencls;
extern std::vector<OPENCL_DEVICE_PROP> ati_opencls;
extern std::vector<OPENCL_DEVICE_PROP> intel_gpu_opencls;
extern std::vector<OPENCL_DEVICE_PROP> apple_gpu_opencls;
extern std::vector<OPENCL_DEVICE_PROP> other_opencls;
extern std::vector<OPENCL_CPU_PROP> cpu_opencls;

Expand Down
11 changes: 0 additions & 11 deletions client/gpu_intel.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -41,14 +41,3 @@ using std::string;

#include "client_msgs.h"
#include "gpu_detect.h"

void COPROC_INTEL::get(
vector<string>&
) {
}

void COPROC_INTEL::correlate(
bool,
vector<int>&
) {
}
Loading

0 comments on commit 3c7e18e

Please sign in to comment.