| Title | Unclear why StackProbeDEPTH is 500 |
| Status | closed |
| Priority | nice |
| Assigned user | Gareth Rees |
| Organization | Ravenbrook |
| Description | In config.h [1] the constant StackProbeDEPTH is defined for platforms **I3MV and W3I6MV: #if defined(MPS_ARCH_I3) && defined(MPS_BUILD_MV) #define StackProbeDEPTH ((Size)500) #elif defined(MPS_PF_W3I6MV) #define StackProbeDEPTH ((Size)500) #else #define StackProbeDEPTH ((Size)0) #endif But there is no analysis explaining where the value 500 comes from, or how to ensure that it can be correctly maintained in future. |
| Analysis | It looks as though DRJ wrote the original stack probe code [2] [3] [4] [5] [6] [7] [8] and added the StackProbeDEPTH parameter [9] [10] so I asked him [11], but he didn't remember [12]. |
| How found | inspection |
| Evidence | [1] <http://www.ravenbrook.com/project/mps/master/code/config.h>[2] < https://info.ravenbrook.com/infosys/cgi/perfbrowse.cgi?%40describe+19030>[3] < https://info.ravenbrook.com/infosys/cgi/perfbrowse.cgi?%40describe+19031>[4] < https://info.ravenbrook.com/infosys/cgi/perfbrowse.cgi?%40describe+19032>[5] < https://info.ravenbrook.com/infosys/cgi/perfbrowse.cgi?%40describe+19033>[6] < https://info.ravenbrook.com/infosys/cgi/perfbrowse.cgi?%40describe+19034>[7] < https://info.ravenbrook.com/infosys/cgi/perfbrowse.cgi?%40describe+19036>[8] < https://info.ravenbrook.com/infosys/cgi/perfbrowse.cgi?%40describe+19038>[9] < https://info.ravenbrook.com/infosys/cgi/perfbrowse.cgi?%40describe+19049>[10] < https://info.ravenbrook.com/infosys/cgi/perfbrowse.cgi?%40describe+19050>[11] < https://info.ravenbrook.com/mail/2013/10/29/16-45-21/0/>[12] < https://info.ravenbrook.com/mail/2013/10/29/17-14-25/0/> |
| Introduced in | 1.110.0 |
| Created by | Gareth Rees |
| Created on | 2013-10-29 15:40:17 |
| Last modified by | Gareth Rees |
| Last modified on | 2014-10-24 10:16:39 |
| History | 2013-10-29 GDR Created. |
| Change | Effect | Date | User | Description |
|---|---|---|---|---|
| 187380 | closed | 2014-10-24 10:16:28 | Gareth Rees | Analysis justifying the value of StackProbeDEPTH. |