From: Nicholas Piggin <npiggin@gmail•com>
To: Jordan Niethe <jniethe5@gmail•com>
Cc: Alistair Popple <alistair@popple•id.au>,
Balamuruhan S <bala24@linux•ibm.com>,
linuxppc-dev@lists•ozlabs.org, Daniel Axtens <dja@axtens•net>
Subject: Re: [PATCH v4 14/16] powerpc64: Add prefixed instructions to instruction data type
Date: Tue, 24 Mar 2020 15:40:56 +1000 [thread overview]
Message-ID: <1585028271.y7k9vfaaar.astroid@bobo.none> (raw)
In-Reply-To: <CACzsE9rbxV6HErxhwseMEJu7APezvRu4pKOx5YkepEnUWtpzqw@mail.gmail.com>
Jordan Niethe's on March 24, 2020 9:45 am:
> On Mon, Mar 23, 2020 at 6:37 PM Nicholas Piggin <npiggin@gmail•com> wrote:
>>
>> Jordan Niethe's on March 20, 2020 3:18 pm:
>> I'm a bit against using partially constructed opaque type for things
>> like this, even if it is in the code that knows about the type. We
>> could modify ppc_inst_prefixed() to assert that pad is equal to zero
>> (or some poisoned value) if it's not prefixed. Or do some validation
>> on the suffix if it is.
> Okay what about something like:
> +static inline ppc_inst ppc_inst_read(const void *ptr)
> +{
> + u32 prefix, suffix;
> + prefix = *(u32 *)ptr;
> + if (prefix >> 26 == 1)
> + suffix = *((u32 *)ptr + 1);
> + else
> + suffix = 0;
> + return PPC_INST_PREFIX(prefix, suffix);
> +}
Sure, if that's the best way to test prefix.
>> Although there's proably no real performance or atomicity issues here,
>> I'd be pleased if we could do a case for prefixed and a case for non
>> prefixed, and store the non-prefixed with "std". Just for the principle
>> of not having half-written instructions in the image.
> Do you mean store the prefixed with std?
Oops, yes.
>> > @@ -881,7 +882,6 @@ static struct bpt *new_breakpoint(unsigned long a)
>> > if (!bp->enabled && atomic_read(&bp->ref_count) == 0) {
>> > bp->address = a;
>> > bp->instr = bpt_table + ((bp - bpts) * BPT_WORDS);
>> > - patch_instruction(bp->instr + 1, PPC_INST(bpinstr));
>> > return bp;
>> > }
>> > }
>>
>> Why is this okay to remove?
> When we only had word instructions the bpt was just patched in here
> once and that was that.
> With prefixed instructions bp->instr + 1 might be the suffix. So I
> moved putting the breakpoint to insert_bpts():
> patch_instruction(bp->instr + ppc_inst_len(instr), PPC_INST(bpinstr));
Ah okay.
Thanks,
Nick
next prev parent reply other threads:[~2020-03-24 5:43 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-03-20 5:17 [PATCH v4 00/16] Initial Prefixed Instruction support Jordan Niethe
2020-03-20 5:17 ` [PATCH v4 01/16] powerpc/xmon: Remove store_inst() for patch_instruction() Jordan Niethe
2020-03-23 6:19 ` Nicholas Piggin
2020-03-20 5:17 ` [PATCH v4 02/16] xmon: Move out-of-line instructions to text section Jordan Niethe
2020-03-23 5:59 ` Balamuruhan S
2020-03-23 6:05 ` Balamuruhan S
2020-03-23 9:26 ` Jordan Niethe
2020-03-23 6:22 ` Nicholas Piggin
2020-03-20 5:17 ` [PATCH v4 03/16] powerpc: Use a datatype for instructions Jordan Niethe
2020-03-23 6:07 ` Balamuruhan S
2020-03-23 6:23 ` Nicholas Piggin
2020-03-23 9:28 ` Jordan Niethe
2020-03-23 9:51 ` Nicholas Piggin
2020-03-24 2:58 ` Michael Ellerman
2020-03-24 3:21 ` Jordan Niethe
2020-04-01 10:32 ` Balamuruhan S
2020-04-01 23:52 ` Jordan Niethe
2020-04-02 23:44 ` Alistair Popple
2020-04-03 0:09 ` Jordan Niethe
2020-03-20 5:17 ` [PATCH v4 04/16] powerpc: Use a macro for creating instructions from u32s Jordan Niethe
2020-03-23 6:26 ` Nicholas Piggin
2020-03-23 9:29 ` Jordan Niethe
2020-03-20 5:17 ` [PATCH v4 05/16] powerpc: Use a function for masking instructions Jordan Niethe
2020-03-23 6:37 ` Nicholas Piggin
2020-03-23 9:31 ` Jordan Niethe
2020-03-20 5:17 ` [PATCH v4 06/16] powerpc: Use a function for getting the instruction op code Jordan Niethe
2020-03-23 6:54 ` Balamuruhan S
2020-03-23 9:35 ` Jordan Niethe
2020-03-20 5:18 ` [PATCH v4 07/16] powerpc: Introduce functions for instruction nullity and equality Jordan Niethe
2020-03-23 6:43 ` Nicholas Piggin
2020-03-23 9:31 ` Jordan Niethe
2020-03-20 5:18 ` [PATCH v4 08/16] powerpc: Use an accessor for word instructions Jordan Niethe
2020-03-23 11:12 ` Balamuruhan S
2020-03-24 3:18 ` Jordan Niethe
2020-03-24 6:22 ` Balamuruhan S
2020-03-20 5:18 ` [PATCH v4 09/16] powerpc: Use a function for reading instructions Jordan Niethe
2020-03-23 8:00 ` Nicholas Piggin
2020-03-23 8:43 ` Balamuruhan S
2020-03-23 10:09 ` Jordan Niethe
2020-03-23 10:36 ` Nicholas Piggin
2020-03-20 5:18 ` [PATCH v4 10/16] powerpc: Make test_translate_branch() independent of instruction length Jordan Niethe
2020-03-20 5:18 ` [PATCH v4 11/16] powerpc: Enable Prefixed Instructions Jordan Niethe
2020-03-23 7:02 ` Nicholas Piggin
2020-03-20 5:18 ` [PATCH v4 12/16] powerpc: Define new SRR1 bits for a future ISA version Jordan Niethe
2020-03-23 7:03 ` Nicholas Piggin
2020-03-20 5:18 ` [PATCH v4 13/16] powerpc: Support prefixed instructions in alignment handler Jordan Niethe
2020-03-23 7:05 ` Nicholas Piggin
2020-03-23 9:35 ` Jordan Niethe
2020-03-20 5:18 ` [PATCH v4 14/16] powerpc64: Add prefixed instructions to instruction data type Jordan Niethe
2020-03-23 6:58 ` Nicholas Piggin
2020-03-23 7:33 ` Nicholas Piggin
2020-03-23 23:45 ` Jordan Niethe
2020-03-24 5:40 ` Nicholas Piggin [this message]
2020-03-30 9:05 ` Alistair Popple
2020-03-30 9:13 ` Jordan Niethe
2020-03-20 5:18 ` [PATCH v4 15/16] powerpc sstep: Add support for prefixed load/stores Jordan Niethe
2020-03-23 8:54 ` Balamuruhan S
2020-03-20 5:18 ` [PATCH v4 16/16] powerpc sstep: Add support for prefixed fixed-point arithmetic Jordan Niethe
2020-03-23 10:05 ` Balamuruhan S
2020-03-23 6:18 ` [PATCH v4 00/16] Initial Prefixed Instruction support Nicholas Piggin
2020-03-23 9:25 ` Jordan Niethe
2020-03-23 10:17 ` Nicholas Piggin
2020-03-24 2:54 ` Jordan Niethe
2020-03-24 2:59 ` Jordan Niethe
2020-03-24 5:44 ` Nicholas Piggin
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1585028271.y7k9vfaaar.astroid@bobo.none \
--to=npiggin@gmail$(echo .)com \
--cc=alistair@popple$(echo .)id.au \
--cc=bala24@linux$(echo .)ibm.com \
--cc=dja@axtens$(echo .)net \
--cc=jniethe5@gmail$(echo .)com \
--cc=linuxppc-dev@lists$(echo .)ozlabs.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox