Re: [PATCH (v2.47 regression)] hash.h: set NEEDS_CLONE_HELPER_UNSAFE in fallback mode
- From
Jeff King <peff@peff.net>
- Date
- Oct 3, 2024, 00:21 UTC
- Message-ID
- <20241003002140.GB3486271@coredump.intra.peff.net>
- In-Reply-To
- <Zv3huwCNuqi5C8Wm@tapette.crustytoothpaste.net>
On Thu, Oct 03, 2024 at 12:13:47AM +0000, brian m. carlson wrote:
Show 17 quoted lines
> On 2024-10-02 at 23:26:18, Jeff King wrote: > > This is a regression in v2.47.0-rc0. As mentioned above, I kind of doubt > > anybody will hit it in practice (I only did because I was trying to do > > some timing tests between the fast and dc variants). And it is almost > > tempting to leave it as a wake-up call for anybody who is still not > > using a collision-detecting sha1. ;) > > I think this is a fine fix for 2.47. I have a branch on my remote > (sha1-dc-only), which I'll send out after it passes CI (probably later > this week), that removes support for the everything but SHA-1-DC (except > for the unsafe code). > > I don't think there's a reasonable configuration where people can use > Git with other SHA-1 code except in extremely limited circumstances we > shouldn't have to maintain code for. The code is open source, so people > who really must have maximum performance with all of the vulnerabilities > can patch it back in themselves.
Yeah, I feel the same way. I only happened to try this because it was the easiest way to speed-compare different implementations using "test-tool sha1". ;)
Possibly that helper could grow an option to use the unsafe variant, though even that is probably not a high priority.
-Peff