| Message ID | 20241115113605.1504796-1-alan.maguire@oracle.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | Check DW_OP_[GNU_]entry_value for possible parameter matching | expand |
On 15/11/2024 11:36, Alan Maguire wrote: > Currently, pahole relies on DWARF to find whether a particular func > has its parameter mismatched with standard or optimized away. > In both these cases, the func will not be put in BTF and this > will prevent fentry/fexit tracing for these functions. > > The current parameter checking focuses on the first location/expression > to match intended parameter register. But in some cases, the first > location/expression does not have expected matching information, > but further location like DW_OP_[GNU_]entry_value can provide > information which matches the expected parameter register. > > Patch 1 supports this; patch 2 adds locking around dwarf_getlocation* > as it is unsafe in a multithreaded environment. > apologies, forgot to note Changes since v2: - handle multiple DW_OP_entry_value expressions by bailing if the register matches expected, otherwise save reg in return value (Eduard Yonghong, Jiri, patch 1) > Alan Maguire (1): > dwarf_loader: use libdw__lock for dwarf_getlocation(s) > > Eduard Zingerman (1): > dwarf_loader: Check DW_OP_[GNU_]entry_value for possible parameter > matching > > dwarf_loader.c | 123 +++++++++++++++++++++++++++++++++++++++---------- > 1 file changed, 98 insertions(+), 25 deletions(-) >
On Fri, Nov 15, 2024 at 11:40:16AM +0000, Alan Maguire wrote: > On 15/11/2024 11:36, Alan Maguire wrote: > > Currently, pahole relies on DWARF to find whether a particular func > > has its parameter mismatched with standard or optimized away. > > In both these cases, the func will not be put in BTF and this > > will prevent fentry/fexit tracing for these functions. > > > > The current parameter checking focuses on the first location/expression > > to match intended parameter register. But in some cases, the first > > location/expression does not have expected matching information, > > but further location like DW_OP_[GNU_]entry_value can provide > > information which matches the expected parameter register. > > > > Patch 1 supports this; patch 2 adds locking around dwarf_getlocation* > > as it is unsafe in a multithreaded environment. > > > > apologies, forgot to note > > Changes since v2: > > - handle multiple DW_OP_entry_value expressions by bailing if the > register matches expected, otherwise save reg in return value (Eduard > Yonghong, Jiri, patch 1) Thanks, applied locally, will perform tests and push publicly later today. - Arnaldo > > Alan Maguire (1): > > dwarf_loader: use libdw__lock for dwarf_getlocation(s) > > > > Eduard Zingerman (1): > > dwarf_loader: Check DW_OP_[GNU_]entry_value for possible parameter > > matching > > > > dwarf_loader.c | 123 +++++++++++++++++++++++++++++++++++++++---------- > > 1 file changed, 98 insertions(+), 25 deletions(-) > >
Currently, pahole relies on DWARF to find whether a particular func has its parameter mismatched with standard or optimized away. In both these cases, the func will not be put in BTF and this will prevent fentry/fexit tracing for these functions. The current parameter checking focuses on the first location/expression to match intended parameter register. But in some cases, the first location/expression does not have expected matching information, but further location like DW_OP_[GNU_]entry_value can provide information which matches the expected parameter register. Patch 1 supports this; patch 2 adds locking around dwarf_getlocation* as it is unsafe in a multithreaded environment. Alan Maguire (1): dwarf_loader: use libdw__lock for dwarf_getlocation(s) Eduard Zingerman (1): dwarf_loader: Check DW_OP_[GNU_]entry_value for possible parameter matching dwarf_loader.c | 123 +++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 98 insertions(+), 25 deletions(-)