1
0
mirror of https://github.com/hashcat/hashcat.git synced 2024-11-23 00:28:11 +00:00

OpenCL Runtime: Added a warning about using Mesa OpenCL runtime

This commit is contained in:
jsteube 2017-02-23 11:48:21 +01:00
parent 4e2adc031a
commit aa5ed1ca89
2 changed files with 17 additions and 1 deletions

View File

@ -88,6 +88,7 @@
- OpenCL Kernel: Renumbered hash-mode 7600 to 4521 - OpenCL Kernel: Renumbered hash-mode 7600 to 4521
- OpenCL Kernel: Remove "static" keyword from function declarations; Causes older Intel OpenCL runtimes to fail compiling - OpenCL Kernel: Remove "static" keyword from function declarations; Causes older Intel OpenCL runtimes to fail compiling
- OpenCL Runtime: Updated AMDGPU-Pro driver version check, do warn if version 16.60 is detected which is known to be broken - OpenCL Runtime: Updated AMDGPU-Pro driver version check, do warn if version 16.60 is detected which is known to be broken
- OpenCL Runtime: Added a warning about using Mesa OpenCL runtime
- Outfile Check: Fixed a memory leak for failed outfile reads - Outfile Check: Fixed a memory leak for failed outfile reads
- Restore: Add some checks on the rd->cwd variable in restore case - Restore: Add some checks on the rd->cwd variable in restore case
- Rule Engine: Fixed several memory leaks in case loading of rules failed - Rule Engine: Fixed several memory leaks in case loading of rules failed

View File

@ -2515,6 +2515,21 @@ int opencl_ctx_devices_init (hashcat_ctx_t *hashcat_ctx, const int comptime)
if (platform_skipped == true) continue; if (platform_skipped == true) continue;
if (user_options->force == false)
{
if (platform_vendor_id == VENDOR_ID_MESA)
{
event_log_error (hashcat_ctx, "Mesa (Gallium) OpenCL platform detected!");
event_log_warning (hashcat_ctx, "It is known to cause errors which an unexperienced user misinterpret as a bug in hashcat");
event_log_warning (hashcat_ctx, "You are STRONGLY encouraged to use the driver as listed in docs/readme.txt");
event_log_warning (hashcat_ctx, "You can use --opencl-platforms to manually deselect the platform and get rid of this error");
event_log_warning (hashcat_ctx, "You can use --force to override this but do not post error reports if you do so");
return -1;
}
}
hc_device_param_t *devices_param = opencl_ctx->devices_param; hc_device_param_t *devices_param = opencl_ctx->devices_param;
for (u32 platform_devices_id = 0; platform_devices_id < platform_devices_cnt; platform_devices_id++) for (u32 platform_devices_id = 0; platform_devices_id < platform_devices_cnt; platform_devices_id++)
@ -2827,7 +2842,7 @@ int opencl_ctx_devices_init (hashcat_ctx_t *hashcat_ctx, const int comptime)
{ {
if (device_param->device_vendor_id == VENDOR_ID_AMD_USE_INTEL) if (device_param->device_vendor_id == VENDOR_ID_AMD_USE_INTEL)
{ {
if (user_options->force == 0) if (user_options->force == false)
{ {
if (user_options->quiet == false) event_log_warning (hashcat_ctx, "* Device #%u: Not a native Intel OpenCL runtime, expect massive speed loss", device_id + 1); if (user_options->quiet == false) event_log_warning (hashcat_ctx, "* Device #%u: Not a native Intel OpenCL runtime, expect massive speed loss", device_id + 1);
if (user_options->quiet == false) event_log_warning (hashcat_ctx, " You can use --force to override this but do not post error reports if you do so"); if (user_options->quiet == false) event_log_warning (hashcat_ctx, " You can use --force to override this but do not post error reports if you do so");