| Summary: | Add VG_(replaceIndexXA) | ||
|---|---|---|---|
| Product: | [Developer tools] valgrind | Reporter: | Łukasz Marek <lukasz.m.luki2> |
| Component: | general | Assignee: | Julian Seward <jseward> |
| Status: | RESOLVED FIXED | ||
| Severity: | wishlist | CC: | philippe.waroquiers |
| Priority: | NOR | ||
| Version First Reported In: | unspecified | ||
| Target Milestone: | --- | ||
| Platform: | unspecified | ||
| OS: | Linux | ||
| Latest Commit: | Version Fixed/Implemented In: | ||
| Sentry Crash Report: | |||
| Attachments: | patch | ||
Is this actually necessary? I've always done this: T* p = (T*)VG_(indexXA)(arr, index); *p = new_value; or even *(T*)VG_(indexXA)(arr, index) = new_value; which removes the possibility of the array being modified between the call to VG_(indexXA) and the assignment. (In reply to Julian Seward from comment #1) > Is this actually necessary? I've always done this: > > T* p = (T*)VG_(indexXA)(arr, index); > *p = new_value; You are right, this new function is not necessary. Just one remark. Your method doesn't mark array sorted=false. I can live with that anyway. (In reply to Łukasz Marek from comment #4) > Just one remark. Your method doesn't mark array sorted=false. I can live > with that anyway. Hmm, actually that's a good point. We do want to set sorted = false in this case. So yes, let's keep the function. Slightly modified version of the patch pushed as 081c34ea477. (I have removed some of the asserts and the call to ensureSpace as the replace operation can never make the array grow. Thanks for the patch |
Created attachment 118247 [details] patch VG_(replaceIndexXA) I found useful for myself manipulating huge array that has around 100k elements and I have to replace some part of them periodically. This allows avoiding elements shifting each time old element is removed when there is a possibility to just replace it with a new one.