Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
To:     Doug Ledford <dledford@redhat.com>,
        "Luck, Tony" <tony.luck@intel.com>, Ira Weiny <ira.weiny@intel.com>
Cc:     Jason Gunthorpe <jgg@ziepe.ca>, Leon Romanovsky <leon@kernel.org>,
        Parav Pandit <parav@mellanox.com>, linux-rdma@vger.kernel.org,
        linux-kernel@vger.kernel.org
References: <20190730202407.31046-1-tony.luck@intel.com>
 <95f5cf70-1a1d-f48c-efac-f389360f585e@embeddedor.com>
 <20190731042801.GA2179@iweiny-DESK2.sc.intel.com>
 <20190731043957.GA1600@agluck-desk2.amr.corp.intel.com>
 <09a994054e43c8bd6ee49b7d1087c9c4e793058f.camel@redhat.com>
From:   "Gustavo A. R. Silva" <gustavo@embeddedor.com>
Openpgp: preference=signencrypt
Autocrypt: addr=gustavo@embeddedor.com; keydata=
 mQINBFssHAwBEADIy3ZoPq3z5UpsUknd2v+IQud4TMJnJLTeXgTf4biSDSrXn73JQgsISBwG
 2Pm4wnOyEgYUyJd5tRWcIbsURAgei918mck3tugT7AQiTUN3/5aAzqe/4ApDUC+uWNkpNnSV
 tjOx1hBpla0ifywy4bvFobwSh5/I3qohxDx+c1obd8Bp/B/iaOtnq0inli/8rlvKO9hp6Z4e
 DXL3PlD0QsLSc27AkwzLEc/D3ZaqBq7ItvT9Pyg0z3Q+2dtLF00f9+663HVC2EUgP25J3xDd
 496SIeYDTkEgbJ7WYR0HYm9uirSET3lDqOVh1xPqoy+U9zTtuA9NQHVGk+hPcoazSqEtLGBk
 YE2mm2wzX5q2uoyptseSNceJ+HE9L+z1KlWW63HhddgtRGhbP8pj42bKaUSrrfDUsicfeJf6
 m1iJRu0SXYVlMruGUB1PvZQ3O7TsVfAGCv85pFipdgk8KQnlRFkYhUjLft0u7CL1rDGZWDDr
 NaNj54q2CX9zuSxBn9XDXvGKyzKEZ4NY1Jfw+TAMPCp4buawuOsjONi2X0DfivFY+ZsjAIcx
 qQMglPtKk/wBs7q2lvJ+pHpgvLhLZyGqzAvKM1sVtRJ5j+ARKA0w4pYs5a5ufqcfT7dN6TBk
 LXZeD9xlVic93Ju08JSUx2ozlcfxq+BVNyA+dtv7elXUZ2DrYwARAQABtCxHdXN0YXZvIEEu
 IFIuIFNpbHZhIDxndXN0YXZvQGVtYmVkZGVkb3IuY29tPokCPQQTAQgAJwUCWywcDAIbIwUJ
 CWYBgAULCQgHAgYVCAkKCwIEFgIDAQIeAQIXgAAKCRBHBbTLRwbbMZ6tEACk0hmmZ2FWL1Xi
 l/bPqDGFhzzexrdkXSfTTZjBV3a+4hIOe+jl6Rci/CvRicNW4H9yJHKBrqwwWm9fvKqOBAg9
 obq753jydVmLwlXO7xjcfyfcMWyx9QdYLERTeQfDAfRqxir3xMeOiZwgQ6dzX3JjOXs6jHBP
 cgry90aWbaMpQRRhaAKeAS14EEe9TSIly5JepaHoVdASuxklvOC0VB0OwNblVSR2S5i5hSsh
 ewbOJtwSlonsYEj4EW1noQNSxnN/vKuvUNegMe+LTtnbbocFQ7dGMsT3kbYNIyIsp42B5eCu
 JXnyKLih7rSGBtPgJ540CjoPBkw2mCfhj2p5fElRJn1tcX2McsjzLFY5jK9RYFDavez5w3lx
 JFgFkla6sQHcrxH62gTkb9sUtNfXKucAfjjCMJ0iuQIHRbMYCa9v2YEymc0k0RvYr43GkA3N
 PJYd/vf9vU7VtZXaY4a/dz1d9dwIpyQARFQpSyvt++R74S78eY/+lX8wEznQdmRQ27kq7BJS
 R20KI/8knhUNUJR3epJu2YFT/JwHbRYC4BoIqWl+uNvDf+lUlI/D1wP+lCBSGr2LTkQRoU8U
 64iK28BmjJh2K3WHmInC1hbUucWT7Swz/+6+FCuHzap/cjuzRN04Z3Fdj084oeUNpP6+b9yW
 e5YnLxF8ctRAp7K4yVlvA7kCDQRbLBwMARAAsHCE31Ffrm6uig1BQplxMV8WnRBiZqbbsVJB
 H1AAh8tq2ULl7udfQo1bsPLGGQboJSVN9rckQQNahvHAIK8ZGfU4Qj8+CER+fYPp/MDZj+t0
 DbnWSOrG7z9HIZo6PR9z4JZza3Hn/35jFggaqBtuydHwwBANZ7A6DVY+W0COEU4of7CAahQo
 5NwYiwS0lGisLTqks5R0Vh+QpvDVfuaF6I8LUgQR/cSgLkR//V1uCEQYzhsoiJ3zc1HSRyOP
 otJTApqGBq80X0aCVj1LOiOF4rrdvQnj6iIlXQssdb+WhSYHeuJj1wD0ZlC7ds5zovXh+FfF
 l5qH5RFY/qVn3mNIVxeO987WSF0jh+T5ZlvUNdhedGndRmwFTxq2Li6GNMaolgnpO/CPcFpD
 jKxY/HBUSmaE9rNdAa1fCd4RsKLlhXda+IWpJZMHlmIKY8dlUybP+2qDzP2lY7kdFgPZRU+e
 zS/pzC/YTzAvCWM3tDgwoSl17vnZCr8wn2/1rKkcLvTDgiJLPCevqpTb6KFtZosQ02EGMuHQ
 I6Zk91jbx96nrdsSdBLGH3hbvLvjZm3C+fNlVb9uvWbdznObqcJxSH3SGOZ7kCHuVmXUcqoz
 ol6ioMHMb+InrHPP16aVDTBTPEGwgxXI38f7SUEn+NpbizWdLNz2hc907DvoPm6HEGCanpcA
 EQEAAYkCJQQYAQgADwUCWywcDAIbDAUJCWYBgAAKCRBHBbTLRwbbMdsZEACUjmsJx2CAY+QS
 UMebQRFjKavwXB/xE7fTt2ahuhHT8qQ/lWuRQedg4baInw9nhoPE+VenOzhGeGlsJ0Ys52sd
 XvUjUocKgUQq6ekOHbcw919nO5L9J2ejMf/VC/quN3r3xijgRtmuuwZjmmi8ct24TpGeoBK4
 WrZGh/1hAYw4ieARvKvgjXRstcEqM5thUNkOOIheud/VpY+48QcccPKbngy//zNJWKbRbeVn
 imua0OpqRXhCrEVm/xomeOvl1WK1BVO7z8DjSdEBGzbV76sPDJb/fw+y+VWrkEiddD/9CSfg
 fBNOb1p1jVnT2mFgGneIWbU0zdDGhleI9UoQTr0e0b/7TU+Jo6TqwosP9nbk5hXw6uR5k5PF
 8ieyHVq3qatJ9K1jPkBr8YWtI5uNwJJjTKIA1jHlj8McROroxMdI6qZ/wZ1ImuylpJuJwCDC
 ORYf5kW61fcrHEDlIvGc371OOvw6ejF8ksX5+L2zwh43l/pKkSVGFpxtMV6d6J3eqwTafL86
 YJWH93PN+ZUh6i6Rd2U/i8jH5WvzR57UeWxE4P8bQc0hNGrUsHQH6bpHV2lbuhDdqo+cM9eh
 GZEO3+gCDFmKrjspZjkJbB5Gadzvts5fcWGOXEvuT8uQSvl+vEL0g6vczsyPBtqoBLa9SNrS
 VtSixD1uOgytAP7RWS474w==
Subject: Re: [PATCH V2] IB/core: Add mitigation for Spectre V1
Message-ID: <1fc90610-7189-c99b-2af1-ae516faa20b4@embeddedor.com>
Date:   Wed, 31 Jul 2019 12:52:27 -0500
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101
 Thunderbird/60.8.0
MIME-Version: 1.0
In-Reply-To: <09a994054e43c8bd6ee49b7d1087c9c4e793058f.camel@redhat.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On 7/31/19 9:52 AM, Doug Ledford wrote:

> 
> I'm not sure this is the best fix for this.  However, here is where I
> get to admit that I largely ignored the whole Spectre V1 thing, so I'm
> not sure I completely understand the vulnerability and the limits to
> that.  But, looking at the function, it seems we can do an early return
> without ever taking any of the mutexes in the function in the case of id
>> = IB_UMAD_MAX_AGENTS, so if we did that, would that separate the
> checking of id far enough from the usage of it as an array index that we
> wouldn't need the clamp to prevent speculative prefetch?  About how far,
> in code terms, does this speculative prefetch occur?
> 
> This is the patch I was thinking of:
> 

>  
> @@ -884,11 +885,18 @@ static int ib_umad_unreg_agent(struct ib_umad_file *file, u32 __user *arg)
>  
>         if (get_user(id, arg))
>                 return -EFAULT;
> +       if (id >= IB_UMAD_MAX_AGENTS)
> +               return -EINVAL;
>  
>         mutex_lock(&file->port->file_mutex);
>         mutex_lock(&file->mutex);
>  
> -       if (id >= IB_UMAD_MAX_AGENTS || !__get_agent(file, id)) {
> +       /*
> +        * Is our check of id far enough away, code wise, to prevent
> +        * speculative prefetch?
> +        */
> +       id = array_index_nospec(id, IB_UMAD_MAX_AGENTS);
> +       if (!__get_agent(file, id)) {
>                 ret = -EINVAL;
>                 goto out;
>         }
> 

This is insufficient. The speculation windows are large:

"Speculative  execution  on  modern  CPUs  can  run  several
hundred  instructions  ahead." [1]

[1] https://spectreattack.com/spectre.pdf

--
Gustavo


--
Gustavo