Messages in this thread | | | From | Cong Wang <> | Date | Thu, 14 Jun 2018 09:57:12 -0700 | Subject | Re: [PATCH] infiniband: fix a subtle race condition |
| |
On Thu, Jun 14, 2018 at 12:01 AM, Leon Romanovsky <leon@kernel.org> wrote: > On Wed, Jun 13, 2018 at 11:21:54PM -0700, Cong Wang wrote: >> On Wed, Jun 13, 2018 at 10:34 PM, Leon Romanovsky <leon@kernel.org> wrote: >> > >> > Hi Cong, >> > >> > If the compiler optimizes the first line (mutex_lock) as you wrote, >> > it will reuse "f" for the second line (mutex_unlock) too. >> >> Nope, check the assembly if you don't trust me, at least >> my compiler always fetches ctx->file without this patch. >> >> I can show you the assembly code tomorrow (too late to >> access my dev machine now). > > I trust you, so don't need to check it however wanted to emphasize > that your solution is compiler specific and not universally true.
So are you saying even with my patch compiler could still re-fetch ctx->file? I doubt...
> >> >> >> > >> > You need to ensure that ucma_modify_id() doesn't run in parallel to >> > anything that uses "ctx->file" directly and indirectly. >> > >> >> Talk is easy, show me the code. :) I knew there is probably >> some other race with this code even after my patch, possibly with >> ->close() for example, but for this specific unlock warning, this patch >> is sufficient. I can't solve all the races in one patch. > > We do prefer complete solution once the problem is fully understood. >
The unlock imbalance problem is fully understood and is clearly shown in my changelog.
My patch never intends to solve any other problem except this one.
> It looks like you are one step away from final patch. It will be conversion > of mutex to be rwlock and separating between read (almost in all places) > and write (in ucma_migrate_id) paths. >
Excuse me. How does this even solve the imbalance problem?
f = ctx->file; ucma_lock_files(f, new_file); // write sem ctx->file = new_file ucma_lock_files(f, new_file); // write sem down_read(&f->rw); // still the old file, nothing change f = ctx->file; // new file up_read(&f->rw); // still imbalance
| |